Journal of Experimental Botany

Battey, Nicholas H.;Simmonds, Paul E.

doi: 10.1093/jxb/eri320pmid: 16306235

Abstract The purpose of life is its continuation: survival is the reason things live. Here we explore this ‘basic’ of biology, by reference to the extraordinary life-cycle of the aphid-like pest phylloxera, and the complexity of its relationship with its host the grapevine. The effort and ingenuity that phylloxera employs to continue itself leads to a doubt that survival alone is sufficient reason. It has frequently been suggested that the reduction of life to a catalogue of facts (by science) creates this doubt, because it robs existence of its essence (which is something other than its mechanics). The part that science is said to steal is what Robert Pirsig calls Quality—the harmonious balance of things. Pirsig seems to imply that this is something inherent in things—and independent from us. A more mundane explanation is that the difference between facts and the complete reality is us—the tendency of mind to connect freely between different kinds of information. This possibility is briefly illustrated here by a myth based on the facts of phylloxera. Evolution, grapevine, Phylloxera, plant–insect interactions Daktulosphaira vitifoliae (Fitch), also known as phylloxera (Fig. 1), devastated the French wine industry when introduced from North America in the 19th century. It has more recently threatened Californian viticulture and is monitored intensively in many parts of the world. For example, in southern Australia, where vine growing is vital and very successful, aerial infrared photography is used to assess spread of the disease The phylloxera insect is typically confined to its root-living form in commercial vineyards, but its activities can, nevertheless, lead rapidly to destruction of the vines. The only effective solution is to graft the susceptible, wine-producing species Vitis vinifera on to rootstocks of American vine species that co-evolved with phylloxera and can therefore resist or tolerate it. The discovery of this rootstock-mediated treatment for phylloxera revolutionized the viticulture industry, making grafted vines the norm (Fig. 2). Fig. 1. View largeDownload slide Daktulosphaira vitifoliae (Fitch), also known as phylloxera. Drawings of a selection of life forms by Cornu (1878), reproduced from Ordish (1972). Fig. 1. View largeDownload slide Daktulosphaira vitifoliae (Fitch), also known as phylloxera. Drawings of a selection of life forms by Cornu (1878), reproduced from Ordish (1972). Fig. 2. View largeDownload slide The benefits of grafting vines on a phylloxera-resistant rootstock. Vines on the left are grafted on to the rootstock 110R, which is resistant to phylloxera biotype B. Vines on the right are grafted on to non-resistant AxR#1. Reproduced by kind permission of Jeff Granett (UC Davis, California) from entomology.ucdavis.edu/…/granett/phy_expl.htm Fig. 2. View largeDownload slide The benefits of grafting vines on a phylloxera-resistant rootstock. Vines on the left are grafted on to the rootstock 110R, which is resistant to phylloxera biotype B. Vines on the right are grafted on to non-resistant AxR#1. Reproduced by kind permission of Jeff Granett (UC Davis, California) from entomology.ucdavis.edu/…/granett/phy_expl.htm Peritymbia is another name for phylloxera. It was coined by Westwood during the early stages of discovery and description of the insect, and alludes to the fact that the fundatrix nymph builds a gall that is both her home and her tomb, because she never leaves it. It's not life she craves, but its continuation The complete life cycle of phylloxera is passed successively on leaves and tendrils above ground, and on the roots, and its facts can be summarized as follows. A sexuparous nymph hatches in response to cooling weather in the autumn. She develops wings, emerging from below ground to lay on leaves eggs that hatch as males or females. Male and female mate, and the female lays a winter egg in splits in the bark. In spring, the egg hatches to a fundatrix nymph, a female that returns to the leaf or tendril, and there builds a gall. It's not life she craves, but its continuation The gall is created in response to secretions from the fundatrix. These secretions are said to include auxin, although this could be of plant origin, and they cause the inner cells of the leaf or tendril to expand upwards, around where the insect sits (Fig. 3). The plant hairs grow coarse and long across the entrance to the gall, barring access but allowing the fundatrix, now matured to an adult, to lay several hundred eggs. Then she dies. Fig. 3. View largeDownload slide Transverse section through a phylloxera gall on a vine tendril. Drawing by Cornu (1878), reproduced from Ordish (1972). Fig. 3. View largeDownload slide Transverse section through a phylloxera gall on a vine tendril. Drawing by Cornu (1878), reproduced from Ordish (1972). It's not life she craves, but its continuation The hatchlings can leave and establish their own galls, or remain and lay within their mother's gall. Each of three to four parthenogenetic generations lays eggs to yield by midsummer 5000 million descendants of the fundatrix. Then migrant hatchlings return below ground, where they build galls on roots, lay eggs, and multiply. Eventually, an egg hatches as a sexuparous nymph which returns to the leaves. It's not life she craves, but its continuation Why? A gall myth is presented in the box in which the gall is more than the insect's home and her tomb. The gall is a symbol, more significant than individual life: it is continuation, the potential for complete connection. The myth gives a glimpse of a Quality that we may crave but Biology suggests we should do without. A gall myth From the inside, you see clearly that the phylloxera gall is a half-world, half-plant and half-insect. Fed by gloomy green light it is moist, damp, hair-enclosed, built by the genes, proteins and carbohydrates of the plant but under instruction from the insect. So entwined are insect and vine, you imagine they were once one creature. Guttation fluid produced from vine leaves on warm, moist nights fell as fine, sticky rain. The droplets turned to insect eggs on descending to the soil. The journey back to the leaf is not just a search for light and for food, but a return to the insect's beginning. On reaching the leaf, the mother phylloxera, fundatrix, founder of each repeating cycle, buries her mouthparts in the plant. Sucking and secreting, she converses with it, relaying a blueprint of how she wants the gall built. Auxin, amino acids, perhaps some simple ions—these are shared signals, a code that drives leaf growth up and around the fundatrix, protecting and feeding her. The process suggests a plan, a complicity between insect and plant; their purpose is one day to re-unite. We are very grateful to Paul Hatcher (The University of Reading) and Richard Napier (Warwick-HRI) for helpful comments and suggestions. References Cornu M. 1878. Etudes sur la Phylloxera vastatrix. Mémoires présentés… à l'Académie des Sciences de l'Institut de FranceXXVI, 43–175. Google Scholar Crespi B, Worobey M. 1998. Comparative analysis of gall morphology in Australian gall thrips: the evolution of extended phenotypes. Evolution  52, 1686–1696. Google Scholar Granett J, Walker MA, Kocsis L, Omer AD. 2001. Biology and management of grape phylloxera. Annual Review of Entomology  46, 387–412. Google Scholar Lewin R. 1993. California's lousy vintage. New Scientist 17 April, 27-31. Google Scholar Ordish G. 1972. The great wine blight. London: JM Dent & Sons. Google Scholar Pirsig RM. 1974. Zen and the art of motorcycle maintenance: an inquiry into values. London: Vintage. Google Scholar Stone GN, Schonrogge K. 2003. The adaptive significance of insect gall morphology. Trends in Ecology and Evolution  18, 512–522. Google Scholar Wapshire AJ, Helm KF. 1987. Phylloxera and Vitis: an experimentally testable coevolutionary hypothesis. American Journal of Enology and Viticulture  38, 216–222. Google Scholar Westwood JO. 1869. New vine diseases. Gardeners' Chronicle and Agricultural Gazette 30 January, 109. Google Scholar © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]

Sheshshayee, M., S.;Bindumadhava,, H.;Ramesh,, R.;Prasad, T., G.;Lakshminarayana, M., R.;Udayakumar,, M.

doi: 10.1093/jxb/eri300pmid: 16263911

Abstract Experimental evidence is presented to show that the 18O enrichment in the leaf biomass and the mean (time-averaged) transpiration rate are positively correlated in groundnut and rice genotypes. The relationship between oxygen isotope enrichment and stomatal conductance (gs) was determined by altering gs through ABA and subsequently using contrasting genotypes of cowpea and groundnut. The Peclet model for the 18O enrichment of leaf water relative to the source water is able to predict the mean observed values well, while it cannot reproduce the full range of measured isotopic values. Further, it fails to explain the observed positive correlation between transpiration rate and 18O enrichment in leaf biomass. Transpiration rate is influenced by the prevailing environmental conditions besides the intrinsic genetic variability. As all the genotypes of both species experienced similar environmental conditions, the differences in transpiration rate could mostly be dependent on intrinsic gs. Therefore, it appears that the Δ18O of leaf biomass can be used as an effective surrogate for mean transpiration rate. Further, at a given vapour pressure difference, Δ18O can serve as a measure of stomatal conductance as well. ABA, groundnut, mean transpiration rate, 18O enrichment, rice, stomatal conductance Introduction Plant biomass production is determined by the total water used and the water use efficiency (WUE, ratio of net CO2 assimilation rate to the transpiration rate), especially under water-limited conditions (Passioura, 1976). Since water availability is the most important constraint, particularly in the semi-arid tropics, increasing WUE is regarded as one of the potential approaches to improve crop yields. Recent efforts in selecting wheat genotypes with improved WUE resulted in higher productivity under water-limited conditions (Condon et al., 2002, Rebetske et al., 2002; Richards et al., 2002). In these genotypes, the higher WUE was achieved through a reduction in stomatal conductance (gs). The gs regulates CO2 entry for photosynthesis besides controlling transpiration rate. Hence when gs is reduced, although transpiration decreases, it also hinders CO2 entry. Most often an increase in WUE is associated with reduced gs which can be counterproductive in terms of biomass accumulation (Udayakumar et al., 1998a; Sheshshayee et al., 2003). Therefore, from the agricultural point of view, it is essential to increase WUE without compromising transpiration. Such genotypes would possess superior mesophyll efficiency to assimilate CO2 and hence need to be identified so as to have a distinct advantage in improving agricultural productivity. To this end it becomes imperative simultaneously to determine the genetic variability in WUE and transpiration in plants. Carbon isotope discrimination (Δ13C) has been well established as a time-averaged surrogate for WUE (Farquhar and Richards, 1984; Farquhar et al., 1989), The application of the carbon isotope discrimination technique to assess the genetic variability in WUE has been examined and validated in container and field experiments in several crop species such as cowpea (Ashok et al., 1999; Aniya and Herzog, 2004), wheat (Condon and Hall, 1997; Richards et al., 2002; Condon et al., 2004), groundnut (Wright et al., 1988; Rao, et al., 1995; Bindumadhava et al., 2003), and rice (Sheshshayee et al., 2003; Impa et al., 2005). Similarly, the enrichment of the heavy isotope of oxygen in leaf water (or that of the biomass) relative to the source water is being adopted to assess variations in transpiration rate and stomatal conductance (gs). Although the theory explaining the phenomenon of oxygen isotopic enrichment during the evaporation of water from the ocean surface has been known for almost three decades (Craig and Gordon, 1965), the application of this theory to predict differences in gs and transpiration rate has been fairly recent (Flanagan et al., 1991b, 1994; Farquhar and Lloyd, 1993; Bindumadhava et al., 1999; Bindumadhava, 2000). However, discrepancy between the Craig–Gordon prediction and the measured δ18O of the leaf water has been reported (White et al., 1994; Buhay et al., 1996). Further, the relationship between stomatal conductance and leaf water 18O enrichment has remained equivocal (see Discussion), although increased transpiration has clearly been shown to enrich leaf water 18O (Gonfiantini et al., 1965; DeNiro and Epstein, 1979). The major objective of the investigation was to show that the 18O enrichment in the biomass can be used as a surrogate for the mean transpiration rate by a re-examination of the relationship between Δ18O and stomatal conductance. This relationship was tested both under field and growth chamber conditions. Materials and methods Plants were grown under well-watered conditions in containers (60×45×30 cm) having 20 kg of red sandy loam and farmyard manure mixed in a 3:1 (v/v) proportion. The containers were arranged randomly in open field conditions, adequate plant nutrients were supplied once a month, and prophylactic measures were taken as and when required to raise healthy plants. All experiments were conducted at the Department of Crop Physiology, GKVK Campus, University of Agricultural Sciences, Bangalore, India (12° 58′ N and 77° 35′ E). Alterations in stomatal conductance using abscisic acid (ABA) Petioles of sunflower (KBSH-1) leaves from 35-d-old plants were excised under water and immediately placed in a test tube containing 15 ml of ABA solution (cis-trans ABA; Sigma, USA). Sets of four leaves were maintained at different ABA concentrations (ranging between 10−4 M and 10−7 M and a control without ABA) and allowed to transpire in a controlled growth chamber under a vapour pressure deficit (VPD) of 15 mbar and a PPFD of 1200 μmol m−2 s−1 (400–700 nm). The leaves were initially allowed to transpire up to 10 ml of the solution to flush out the leaf water. The time taken for leaves to transpire an additional 10 ml of the solution was recorded and transpiration rate was calculated after measuring the leaf area. The leaf water was extracted immediately for determining its 18O. The 18O enrichment in leaf water (Δ18O) relative to that of the distilled water (δ18O of distilled water was −13.3). Genotypic variations in stomatal conductance among cowpea genotypes Based on the results of a previous experiment (Sheshshayee, 1998), a few cowpea genotypes with variable gs and transpiration rates were identified and raised in containers under well-watered condition in an open field. Stomatal conductance (gs) and transpiration rate of the third fully expanded leaf from the apex of 45-d-old plants were measured using a portable gas exchange system (LCA-4, ADC, Hoddesdon, EN11, ODB, UK). The temperature and RH of the leaf chamber were maintained close to those of ambient air. The mean natural light intensity was 1600 μmol m−2 s−1. All observations were recorded between 09.00 h and 11.30 h (Indian standard time). Extraction of leaf water Immediately after determining gas exchange, holes were punched, avoiding the major veins, and placed in tapering plastic tubes and immediately closed using rubber stoppers. The tubes were purged with pure, dry nitrogen gas and frozen to the liquid nitrogen temperature of −196 °C for 20 min. The tubes were transferred to a hot water bath (80 °C) and, after 10 min of thawing, the tubes were centrifuged at 5000 rpm for 5 min. The leaf water that collected at the tapering end of the tube was drawn out using a syringe and an aliquot of 200 μl was immediately introduced into a 10 ml vacutainer tube (Becton Dickinson vacutainer systems, USA). The tubes were air-tight and all the leaf water was collected instantly; care was taken to keep the fractionation that might occur during the extraction process to be minimal. Unlike the ‘classical’ method of extraction of leaf water by pumping, the present method does not cause significant isotopic fractionation due to incomplete extraction. Determination of Δ18O of leaf water The 18O composition of the leaf water was determined by CO2-equilibration technique (Scrimgeour, 1995). CO2 gas of known oxygen isotopic composition was introduced to the head-space volume of the vacutainer tube containing leaf water and equilibrated overnight at 30 °C. The CO2 was then introduced into the mass spectrometer (Tracermass, PDZ-Europa, UK) for the determination of δ18O on a continuous flow mode. The δ18O of the source water (used for irrigating the plants) was also similarly determined. The analytical uncertainty was typically ±0.15. The 18O enrichment in the leaf water over the source water (Δ18Olw) was computed as follows. \[{\Delta}^{18}\mathrm{O}_{\mathrm{lw}}{=}{\Delta}^{18}\mathrm{O}_{\mathrm{lw}}{-}\mathrm{{\delta}}^{18}\mathrm{O}_{\mathrm{iw}}\] where δ18O is the isotopic composition compared with VSMOW (Vienna-Standard Mean Ocean Water) and the subscripts lw and iw refer to leaf water and irrigation water, respectively (for the experiment with ABA solution the δ18O of the distilled water was used). Gravimetric determination of variability in mean transpiration rate (MTR) in rice and groundnut genotypes The mean transpiration rate (MTR) was determined gravimetrically (Udayakumar et al., 1998b) in selected genotypes of rice and groundnut in separate experiments. Both the experiments were carried out between January and April 2002. The experimental season was characterized with a mean temperature of 30.6 °C (maximum), 17.2 °C (minimum) and a VPD of 15 mbar. The natural photon flux density was 1600 μmol m−2 s−1. Briefly, the gravimetric method involved weighing the containers daily using a mobile weighing device for a period of 30 d. The container weight was brought back to field capacity daily by adding water. The amount of water added over the experimental period was summed to arrive at the total evapotranspiration (ET). A mobile rain out shelter was moved over the experimental area during nights and rain episodes to maintain a specific water regime in the containers. The soil surface of all the containers was covered with plastic pieces to minimize surface evaporation. Simultaneously, ‘bare’ containers (without plants) were also weighed to quantify the evaporation component (Es) of ET. Cumulative water transpired (CWT) over the experimental period was calculated as the difference between ET and Es. The MTR was computed from the ratio of CWT to the leaf area duration (LAD=(LA1+LA2)/2×30 d, where, LA1 and LA2 are leaf areas at the beginning and end of the experiment, respectively). Determination of δ18O in leaf biomass The leaves of rice and groundnut genotypes that matured during the experimental period were separately harvested and oven-dried at 70 °C for 72 h). Finely ground dry leaf powder (0.8–1.2 mg) was taken to determine δ18Olb by on-line pyrolysis using TC/EA interfaced with an IRMS (Delta Plus, ThermoFinnigan, Bremen, Germany) through a continuous flow device (Conflo-III, ThermoFinnigan, Bremen, Germany) at the National Facility for Stable isotope studies, Department of Crop Physiology, University of Agricultural Sciences, Bangalore, India. The analytical uncertainty of isotope measurements was less than 0.2. The 18O enrichment in leaf biomass (Δ18Olb) was computed as follows: Statistical analysis Analysis of variance (ANOVA) for all the experiments was computed for a completely randomized design using MSTAT-C software. The geometric mean regression (Model-II) was used to plot all relationships. Results Stomatal conductance (gs) and 18O enrichment The plant hormone abscisic acid (ABA) induces stomatal closure and, accordingly, excised leaves of sunflower (KBSH-1) fed with the highest concentration of ABA (10−4 M) recorded the lowest transpiration rate, which increased as the ABA concentration decreased. Δ18Olw also showed a very similar pattern. A strong positive correlation between transpiration rate (TR) and Δ18Olw was evident (Fig. 1). Fig. 1. Open in new tabDownload slide Relationship between Δ18Olw () and transpiration rate (mmol m−2 s−1). Transpiration rate was altered by inducing stomatal closure using ABA (filled diamonds, control, filled squares, 10−7 M, open triangles, 10−6 M, closed triangles, 10−5 M, open diamonds, 10−4 M) in excised sunflower leaves. The leaf water oxygen isotopic enrichment (Δ18Olw) was computed relative to distilled water (δ18Odw= −13.3) used for preparing ABA solutions. The set-up was kept in a growth chamber with 15 mbar VPD and 1200 μmol m−2 s−1 light intensity in PAR range. Each value is a mean of three replicates (y=2.23x+0.461; R2=0.74, P <0.001; n=15). Fig. 1. Open in new tabDownload slide Relationship between Δ18Olw () and transpiration rate (mmol m−2 s−1). Transpiration rate was altered by inducing stomatal closure using ABA (filled diamonds, control, filled squares, 10−7 M, open triangles, 10−6 M, closed triangles, 10−5 M, open diamonds, 10−4 M) in excised sunflower leaves. The leaf water oxygen isotopic enrichment (Δ18Olw) was computed relative to distilled water (δ18Odw= −13.3) used for preparing ABA solutions. The set-up was kept in a growth chamber with 15 mbar VPD and 1200 μmol m−2 s−1 light intensity in PAR range. Each value is a mean of three replicates (y=2.23x+0.461; R2=0.74, P <0.001; n=15). To examine this relationship further, the stomatal conductance of leaves of a few contrasting cowpea genotypes was determined with a gas exchange system. A significant genotypic variability in gs and Δ18Olw was noticed in this set of cowpea genotypes (Table 1). The Δ18Olw showed a significant positive relationship with gs (R2=0.90; P <0.05; n=5) reconfirming the stomatal control of 18O enrichment. Table 1. Genotypic variability in stomatal conductance, transpiration rate and Δ18Olw among cowpea genotypes Genotype . Stomatal conductance (mmol m−2 s−1)a . Transpiration rate (mmol m−2 s−1)a . Δ18Olw ()a . APC-40 GC-20 290±11 10.3±2 33.6±0.60 APC-123-V-683 260±17 8.2±1.5 31.2±0.52 V-585 227±6 7.8±2 30.3±0.22 APC-4125 230±10 7.6±1 30.5±0.26 APC-121-P-132 280±5 9.2±1 33.1±0.10 CDb (P=0.05) 11.02 0.19 0.98 Genotype . Stomatal conductance (mmol m−2 s−1)a . Transpiration rate (mmol m−2 s−1)a . Δ18Olw ()a . APC-40 GC-20 290±11 10.3±2 33.6±0.60 APC-123-V-683 260±17 8.2±1.5 31.2±0.52 V-585 227±6 7.8±2 30.3±0.22 APC-4125 230±10 7.6±1 30.5±0.26 APC-121-P-132 280±5 9.2±1 33.1±0.10 CDb (P=0.05) 11.02 0.19 0.98 Transpiration rate and stomatal conductance were recorded using a portable photosynthesis system (ADC, LCA4, UK). The leaf water 18O enrichment (Δ18Olw) over the irrigation water was determined by the CO2 equilibration method using an IRMS. Each value is a mean of three replicates. The δ18O value of irrigation water was −3.7. a Mean value ±standard deviation. b CD is the critical difference required to conclude that any two values of a parameter are significantly different from each other. Open in new tab Table 1. Genotypic variability in stomatal conductance, transpiration rate and Δ18Olw among cowpea genotypes Genotype . Stomatal conductance (mmol m−2 s−1)a . Transpiration rate (mmol m−2 s−1)a . Δ18Olw ()a . APC-40 GC-20 290±11 10.3±2 33.6±0.60 APC-123-V-683 260±17 8.2±1.5 31.2±0.52 V-585 227±6 7.8±2 30.3±0.22 APC-4125 230±10 7.6±1 30.5±0.26 APC-121-P-132 280±5 9.2±1 33.1±0.10 CDb (P=0.05) 11.02 0.19 0.98 Genotype . Stomatal conductance (mmol m−2 s−1)a . Transpiration rate (mmol m−2 s−1)a . Δ18Olw ()a . APC-40 GC-20 290±11 10.3±2 33.6±0.60 APC-123-V-683 260±17 8.2±1.5 31.2±0.52 V-585 227±6 7.8±2 30.3±0.22 APC-4125 230±10 7.6±1 30.5±0.26 APC-121-P-132 280±5 9.2±1 33.1±0.10 CDb (P=0.05) 11.02 0.19 0.98 Transpiration rate and stomatal conductance were recorded using a portable photosynthesis system (ADC, LCA4, UK). The leaf water 18O enrichment (Δ18Olw) over the irrigation water was determined by the CO2 equilibration method using an IRMS. Each value is a mean of three replicates. The δ18O value of irrigation water was −3.7. a Mean value ±standard deviation. b CD is the critical difference required to conclude that any two values of a parameter are significantly different from each other. Open in new tab Genetic variability in Δ18O, stomatal conductance and mean transpiration rate Gas exchange parameters are snap-shot measurements and do not integrate the diurnal as well as the day to-day variations in transpiration rate and gs. Similarly, the 18O composition of the leaf water also varies significantly in time. The transfer of the 18O signature from leaf water into cellulose/biomass has been well elucidated (Sternberg et al., 1986). The 18O enrichment of the leaf biomass relative to source (Δ18Olb) can serve as a measure of mean (time-integrated) transpiration rate (MTR). The latter was gravimetrically determined for rice and groundnut in separate experiments over an extended period of 30 d. Significant genotypic variability in MTR was noticed in both the crop species (Table 2). Since the biomass formation is continuous, the Δ18Olb should be a time integrated measure of Δ18Olw. Therefore, the relationship between the MTR and Δ18Olb among the genotypes was examined and a significant positive relationship was found (Fig. 2). As the leaf biomass and not the cellulose was analysed, it could be argued that the differences in organic composition among genotypes could possibly account for the observed variability in Δ18Olb especially in the long-term experiment. However, this can be safely ruled out because (i) the bulk of the dry matter of the leaf (>70%) is cellulose and (ii) the remaining components are not isotopically very different, especially in the same species of plants. It was apparent from Fig. 2 that the slopes of the regression between Δ18O and MTR were different for groundnut and rice. The difference could arise due to significant variations in the leaf structural composition and architecture of these two species. Fig. 2. Open in new tabDownload slide Relationship between MTR (mol m−2 d−1) and Δ18Olb () among selected genotypes of (A) groundnut (y=0.072x+16.33; R2=0.69; P <0.001; n=12) and (B) rice (y=0.114x+15.36; R2=0.46; P <0.05; n=11). MTR was determined gravimetrically and the δ18Olb using an IRMS interfaced with TC/EA. The analytical uncertainty of oxygen isotope measurement was less than 0.2. The leaf biomass oxygen isotopic enrichment relative to that of irrigation water (δ18Oiw= −3.7) was computed. Each value is a mean of five replicates. Fig. 2. Open in new tabDownload slide Relationship between MTR (mol m−2 d−1) and Δ18Olb () among selected genotypes of (A) groundnut (y=0.072x+16.33; R2=0.69; P <0.001; n=12) and (B) rice (y=0.114x+15.36; R2=0.46; P <0.05; n=11). MTR was determined gravimetrically and the δ18Olb using an IRMS interfaced with TC/EA. The analytical uncertainty of oxygen isotope measurement was less than 0.2. The leaf biomass oxygen isotopic enrichment relative to that of irrigation water (δ18Oiw= −3.7) was computed. Each value is a mean of five replicates. Table 2. Genotypic variability in mean transpiration rate and Δ18Olb in rice and groundnut Rice . . . Groundnut . . . Genotype . MTR (mol m−2 d−1) . Δ18Olb () . Genotype . MTR (mol m−2 d−1) . Δ18Olb () . IET15297 140.11 34.06 TNAU262 177.61 31.27 IET16348 180.56 33.26 LI-1 196.83 31.2 IET16364 138.56 33.77 JAL-18 266.67 37.66 ALM6 123.28 29.64 ATG-17 245.72 31.70 JAYA 131.44 31.44 JL-24 211.72 29.98 IET15924 175.61 33.44 TIR-17 214.67 32.69 Kirwana 159.22 33.60 VRI-4 156.56 28.87 IET16347 136.83 32.12 ALR-2 208.83 30.28 IET15963 137.78 31.20 ICGS-11 249.89 34.94 MRB-2 152.89 31.90 CO-3 296.22 29.89 MRB-1 116.39 26.20 CO-1 267.67 34.21 Sen Nghe An 231.06 32.98 Mean 144.78 31.88 218.61 32.14 CDa (P=0.05) 10.67 0.95 56.11 3.94 Rice . . . Groundnut . . . Genotype . MTR (mol m−2 d−1) . Δ18Olb () . Genotype . MTR (mol m−2 d−1) . Δ18Olb () . IET15297 140.11 34.06 TNAU262 177.61 31.27 IET16348 180.56 33.26 LI-1 196.83 31.2 IET16364 138.56 33.77 JAL-18 266.67 37.66 ALM6 123.28 29.64 ATG-17 245.72 31.70 JAYA 131.44 31.44 JL-24 211.72 29.98 IET15924 175.61 33.44 TIR-17 214.67 32.69 Kirwana 159.22 33.60 VRI-4 156.56 28.87 IET16347 136.83 32.12 ALR-2 208.83 30.28 IET15963 137.78 31.20 ICGS-11 249.89 34.94 MRB-2 152.89 31.90 CO-3 296.22 29.89 MRB-1 116.39 26.20 CO-1 267.67 34.21 Sen Nghe An 231.06 32.98 Mean 144.78 31.88 218.61 32.14 CDa (P=0.05) 10.67 0.95 56.11 3.94 Mean transpiration rate was determined by gravimetry and the leaf biomass 18O enrichment over the irrigation water (Δ18Olb) was then assessed using an IRMS. Each value is a mean of five replicates. a CD is the critical difference required to conclude that any two values of a parameter are significantly different from each other. Open in new tab Table 2. Genotypic variability in mean transpiration rate and Δ18Olb in rice and groundnut Rice . . . Groundnut . . . Genotype . MTR (mol m−2 d−1) . Δ18Olb () . Genotype . MTR (mol m−2 d−1) . Δ18Olb () . IET15297 140.11 34.06 TNAU262 177.61 31.27 IET16348 180.56 33.26 LI-1 196.83 31.2 IET16364 138.56 33.77 JAL-18 266.67 37.66 ALM6 123.28 29.64 ATG-17 245.72 31.70 JAYA 131.44 31.44 JL-24 211.72 29.98 IET15924 175.61 33.44 TIR-17 214.67 32.69 Kirwana 159.22 33.60 VRI-4 156.56 28.87 IET16347 136.83 32.12 ALR-2 208.83 30.28 IET15963 137.78 31.20 ICGS-11 249.89 34.94 MRB-2 152.89 31.90 CO-3 296.22 29.89 MRB-1 116.39 26.20 CO-1 267.67 34.21 Sen Nghe An 231.06 32.98 Mean 144.78 31.88 218.61 32.14 CDa (P=0.05) 10.67 0.95 56.11 3.94 Rice . . . Groundnut . . . Genotype . MTR (mol m−2 d−1) . Δ18Olb () . Genotype . MTR (mol m−2 d−1) . Δ18Olb () . IET15297 140.11 34.06 TNAU262 177.61 31.27 IET16348 180.56 33.26 LI-1 196.83 31.2 IET16364 138.56 33.77 JAL-18 266.67 37.66 ALM6 123.28 29.64 ATG-17 245.72 31.70 JAYA 131.44 31.44 JL-24 211.72 29.98 IET15924 175.61 33.44 TIR-17 214.67 32.69 Kirwana 159.22 33.60 VRI-4 156.56 28.87 IET16347 136.83 32.12 ALR-2 208.83 30.28 IET15963 137.78 31.20 ICGS-11 249.89 34.94 MRB-2 152.89 31.90 CO-3 296.22 29.89 MRB-1 116.39 26.20 CO-1 267.67 34.21 Sen Nghe An 231.06 32.98 Mean 144.78 31.88 218.61 32.14 CDa (P=0.05) 10.67 0.95 56.11 3.94 Mean transpiration rate was determined by gravimetry and the leaf biomass 18O enrichment over the irrigation water (Δ18Olb) was then assessed using an IRMS. Each value is a mean of five replicates. a CD is the critical difference required to conclude that any two values of a parameter are significantly different from each other. Open in new tab Discussion The first mechanistic explanation for the evaporative enrichment of 18O in large water bodies was provided by Craig and Gordon (1965). This theory was extended by several models to predict the 18O enrichment in leaf water (Flanagan et al., 1991b, 1994; Flanagan, 1993; Farquhar and Lloyd, 1993; Roden and Ehleringer, 1999). \[{\Delta}^{18}\mathrm{O}_{\mathrm{e}}{=}\mathrm{{\varepsilon}}^{{\ast}}{+}\mathrm{{\varepsilon}}_{\mathrm{k}}{+}({\Delta}^{18}\mathrm{O}_{\mathrm{v}}{-}\mathrm{{\varepsilon}}_{\mathrm{k}})e_{\mathrm{a}}/e_{\mathrm{i}}\] where, Δ18Oe is the 18O enrichment in the leaf water at the site of evaporation relative to source water, ε* and εk are the equilibrium and kinetic (oxygen isotopic) fractionation factors, respectively (expressed in per million units), Δ18Ov, the 18O enrichment of the atmospheric water vapour relative to the source water, and ea, ei, the partial pressures of water outside and inside the leaf, respectively. However, it was observed that the actual oxygen enrichment in bulk leaf water (Δ18Olw) was less than that predicted by the Craig–Gordon theory (Allison et al., 1985; Leaney et al., 1985; Flanagan et al., 1991a, b; Wang and Yakir, 1995; Barbour et al., 2002). To explain this discrepancy, several modifications have been attempted. Leaney et al. (1985) considered ‘two pools’ of water in a leaf. The first pool is an enriched fraction due to evaporation and the second, the unfractionated xylem water pool. The ‘Peclet model’ of Farquhar and Lloyd (1993) and Barbour et al. (2000) account for the progressive variation of leaf water enrichment along the leaf mesophyll tissue due to convective and diffusive mixing of the enriched water and the unfractionated xylem water. The ‘string-of-lakes’ model (Gat and Bowser, 1991; Yakir, 1992; Helliker and Ehleringer, 2000, 2002) provides for the spatial variation in δ18O across the entire leaf surface. The string-of-lakes model argues that the enriched leaf water would gradually flow into the xylem thus spatially altering the source water isotopic composition. It is to be noted that, in general, (i) all these models deal with steady-state fluxes of water from the stem to the leaf, from the leaf to the boundary layer, from the boundary layer to the atmosphere, and (ii) they do not address the effect of stomatal response to the atmospheric water vapour deficit, unlike Lindroth and Halldin (1986) who suggest an empirical relation. Recently, a non-steady-state model was proposed by Farquhar and Cernusak (2005); according to these authors, the model ‘is less important during the day and hence for determining the 18O enrichment in organic matter’. In some earlier experiments it was observed that the 18O enrichment in the leaf water was directly proportional to the transpiration rate (Walker et al., 1989; Yakir et al., 1990; Yakir, 1998). Gan et al. (2002) showed that the leaf water as well as the leaf organic matter Δ18O was significantly higher in the leaves exposed to lower RH, which indicated that the 18O enrichment linearly increased with transpiration rate. A lower stomatal conductance (gs), at a given VPD, is also known to reduce the transpiration rate and hence 18O enrichment (Fig. 1 of Wang and Yakir, 1995). Our results are consistent with these, but they contrast with the trend reported for the relationship between leaf water enrichment and transpiration among wheat genotypes (Barbour et al., 2000). A simplified Peclet model developed by Fraquhar and his coworkers (Barbour and Farquhar, 2000) was obtained from Dr Margaret Barbour and the predicted Δ18O was determined. Several input parameters such as fractionation caused during diffusion through stomata (32) and through the boundary layer (21) were considered (Cappa et al., 2003) in the model. The source water of the GKVK tube well used for irrigation was −3.7. The oxygen isotopic enrichment over the source water was estimated using the model. Figure 3 clearly demonstrates that the mean groundnut and rice values of the predicted Δ18Olb (31.7 and 31.7) and the measured Δ18Olb (32.1 and 31.9), respectively, match very well. However, the model is unable to account for the full range of values: the predicted values lie in a narrow range of 30–33, whereas the observed values vary from 26–38. A possible reason could be that several parameters used as inputs in the Peclet model, such as leaf and air temperatures, relative humidity, boundary layer and stomatal conductances, were measured at one single instance during the day. These values are known to fluctuate considerably diurnally and hence would influence the 18O enrichment in the leaf biomass in a cumulative fashion. Thus the measured Δ18O would vary more than that predicted by the model. Fig. 3. Open in new tabDownload slide Relationship between predicted and measured Δ18Olb () in groundnut (open symbols) and rice genotypes (closed symbols). A simplified Peclet model to obtain the predicted Δ18O was provided by Dr Margaret Barbour. The diffusion fractionation through stomata was considered as 32 and that through the boundary layer as 21 (Cappa et al., 2003). The equilibrium fractionation between C=O and water for carbonyl exchange (27) and for the whole leaf biomass (−8) was as per Barbour and Farquhar (2000). Other values used: δ18Ov= −4.0, δ18Osw= −3.7, RH=55%, Tair=28 °C, boundary layer conductance (gb)=1.0 mmol m−2 s−1, effective length for Peclet effect=0.018 m, proportion of exchangeable oxygen in cellulose=0.56, and proportion of xylem water in meristem=0.8. Measured leaf temperatures were used for the calculation. Fig. 3. Open in new tabDownload slide Relationship between predicted and measured Δ18Olb () in groundnut (open symbols) and rice genotypes (closed symbols). A simplified Peclet model to obtain the predicted Δ18O was provided by Dr Margaret Barbour. The diffusion fractionation through stomata was considered as 32 and that through the boundary layer as 21 (Cappa et al., 2003). The equilibrium fractionation between C=O and water for carbonyl exchange (27) and for the whole leaf biomass (−8) was as per Barbour and Farquhar (2000). Other values used: δ18Ov= −4.0, δ18Osw= −3.7, RH=55%, Tair=28 °C, boundary layer conductance (gb)=1.0 mmol m−2 s−1, effective length for Peclet effect=0.018 m, proportion of exchangeable oxygen in cellulose=0.56, and proportion of xylem water in meristem=0.8. Measured leaf temperatures were used for the calculation. Even though the mean values are predicted well by the model, the observed positive correlation between transpiration rate and 18O enrichment in the leaf biomass (Fig. 2) cannot be explained by it. Buhay et al. (1996) showed that εk and gb depend on the nature of the boundary layer, influenced by wind speed and leaf temperature, while the shape/area of the leaf is also important. In this study's experiments all plants experienced the same wind speed and air temperature. The shapes of the leaves are also comparable within species. The dependence of the leaf water isotopic composition on the leaf temperature is of a small magnitude (Majoube, 1971). Therefore, it appears necessary to modify existing models to incorporate the stomatal response to changing VPD naturally. Conclusion In this work, it has been shown experimentally that there is a positive correlation between the transpiration rate (caused by stomatal conductance) and the oxygen isotope enrichment in leaf biomass in groundnut and rice genotypes that contrast with the observations of Barbour and Farquhar (2000). Transpiration rate can increase either because of increased stomatal conductance or when the vapour pressure difference between the leaf and air is increased. When the transpiration rate was altered by inducing differential stomatal closure through ABA, the Δ18Olw closely followed the changes in transpiration rate (Fig. 1). In addition, the genotypes varying in stomatal conductance showed similar variations in mean transpiration rate. The Δ18Olb, an integrated measure of the leaf water Δ18O values, showed a significant positive relationship with mean transpiration rate (Fig. 2). These results suggest that the Δ18Olb is a good time integrated measure of stomatal conductance. Existing models (such as the Peclet model) need to be modified to explain the observed positive correlation between transpiration rate and 18O enrichment in leaf biomass. The authors thank two anonymous referees for constructive suggestions. Dr Anura V Kurpad, Head, Nutrition Division, Department of Physiology, St John's Medical College, Bangalore is thanked for his help during the initial stages of this work and Mr Shashidhar, G and Ms Impa S Muttappa for the help in conducting the groundnut and rice experiments. This work was funded by the Department of Science and Technology (DST), in the form of a Young Scientist project to MSS (HR/OY/B-03/96) and CSIR in the form of Senior Research Fellowship to HB [9/271(51)/96-EMR-1]. Financial assistance by DST and DBT (SP/IO/LF-01/98; BT/IS/06/004/98) is acknowledged. The assistance of Mr Nagabhushana and Mr A Ramesha in sample preparation and isotope analysis is acknowledged. Special thanks to Mr Ram Mhatre, of Thermo-electron for the maintenance of the IRMS. References Allison GB, Gat JR, Leaney FWJ. 1985 . The relationship between deuterium and oxygen-18 delta values in leaf water. Chemical Geology 58, 145 –156. Aniya AO, Herzog H. 2004 . Water use efficiency, leaf area and leaf gas exchange of cowpeas under midseason drought. European Journal of Agronomy 20, 327 –339. Ashok RS, Aftab Hussain IS, Prasad TG, Udayakumar M, Nageswararao RC, Wright GC. 1999 . Variations in transpiration efficiency and carbon isotope discrimination in cowpea (Vigna unguiculata (L.) Walp.) genotypes. Australian Journal of Plant Physiology 26, 503 –510. Barbour MM, Schurr U, Henry BK, Wong SC, Farquhar GD. 2000 . Variation in the oxygen isotope ratio of phloem sap sucrose from castor bean: evidence in support of the Peclet-effect. Plant Physiology 123, 671 –679. Barbour MM, Farquhar GD. 2000 . Relative humidity and ABA-induced variation in carbon and oxygen isotope ratios of cotton leaves. Plant, Cell and Environment 23, 473 –485. Barbour MM, Walcroft AS, Farquhar GD. 2002 . Seasonal variation in δ13C and δ18O of cellulose from growth rings of Pinus radiata. Plant, Cell and Environment 25, 1483 –1499. Bindumadhava H, Sheshshayee MS, Devendra R, Prasad TG, Udayakumar M. 1999 . Oxygen (18O) isotopic enrichment in the leaves as a potential surrogate for transpiration and stomatal conductance. Current Science 76, 1427 –1428. Bindumadhava H. 2000 . Oxygen (Δ18O) and carbon (Δ13C) isotope composition in plants: an approach to quantify transpiration and mesophyll factors associated with water use efficiency. Unpublished PhD thesis, University of Agricultural Science, Bangalore. India. Bindumadhava H, Sheshshayee MS, Shankar AG, Prasad TG, Udayakumar M. 2003 . Use of SPAD chlorophyll meter to assess transpiration efficiency in peanut. In: Breeding of drought-resistant peanuts. ACIAR Proceedings, 112, 3–9. Buhay WM, Edwards TWD, Aravena R. 1996 . Evaluating kinetic fractionation factors used for ecologic and paleoclimatic reconstructions from oxygen and hydrogen isotope ratios in plant water and cellulose. Geochimica et Cosmochimica Acta 60, 2209 –2218. Cappa CD, Hendricks MB, DePaulo DJ, Cohen RL. 2003 . Isotopic fractionation of water during evaporation. Journal of Geophysical Research 108, 4525 –4534. Condon AG, Richards RA, Rebetzke GT, Farquhar GD. 2002 . Improving intrinsic water use efficiency and crop yield. Crop Science 42, 128 –133. Condon AG, Hall KT. 1997 . Adaptation to adverse environment; variation in water use efficiency within crop species; In: Jackson LE, ed. Ecology in agriculture. San Diego, USA: Academic Press, 79–116. Craig L, Gordon LI. 1965 . Deuterium and oxygen-18 variations in the ocean and the marine atmosphere. In: Tongiorgi E, ed. Proceedings of a conference on stable isotopes in oceanographic studies and paleotemperatures. Spoleto, Italy, 9–130. DeNiro MJ, Epstein S. 1979 . Relationship between oxygen isotope ratios of terrestrial plant cellulose, carbon dioxide and water. Science 204, 51 –53. Farquhar GD, Cernusak LA. 2005 . On the isotopic composition of leaf water in the non-steady state. Functional Plant Biology 32, 293 –303. Farquhar GD, Ehleringer JR, Hubick KT. 1989 . Carbon isotope discrimination and photosynthesis. Annual Review of Plant Physiology and Plant Molecular Biology 40, 503 –537. Farquhar GD, Lloyd J. 1993 . Carbon and oxygen isotope effects in the exchange of carbon dioxide between terrestrial plants and the atmosphere. In: Ehleringer JR, Hall AE, Farquhar GD, eds. Stable isotopes and plant carbon-water relations. San Diego: Academic Press, 47–70. Farquhar GD, Richards RA. 1984 . Isotopic composition of plant carbon correlates with water use efficiency of wheat cultivars. Australian Journal of Plant Physiology 11, 539 –552. Flanagan LB. 1993 . Environmental and biological influence on the stable oxygen and hydrogen isotopic composition of leaf water. In: Ehleringer JR, Hall AE, Farquhar GD, eds. Stable isotopes and plant carbon-water relations. San Diego: Academic Press, 71–90. Flanagan LB, Bain JF, Ehleringer JR. 1991a. Stable oxygen and hydrogen isotope composition of leaf water in C3 and C4 plant species under field conditions. Oecologia 88, 394 –400. Flanagan LB, Comstock JP, Ehleringer JR. 1991b. Comparison of modelled and observed environmental influences on stable oxygen and hydrogen composition of leaf water in Phaseolus vulgaris L. Plant Physiology 96, 588 –596. Flanagan LB, Philips SL, Ehleringer JR, Farquhar GD. 1994 . Effect of changes in leaf water oxygen isotopic composition on discrimination against C18O16O during photosynthetic gas exchange. Australian Journal of Plant Physiology 21, 221 –234. Gan KS, Wong SC, Yong JWH, Farquhar GD. 2002 . 18O spatial patterns of vein xylem water, leaf water and dry matter in cotton leaves. Plant Physiology 130, 1008 –1021. Gat JR, Browser C. 1991 . The heavy isotope enrichment of water in coupled evaporative systems. In: Taylor HP, O'Neil JR, Kaplan IR, eds. Stable isotope geochemistry: a tribute to Samuel Epstein. Lancaster, UK: The Geochemical Society, 159–168. Gonfiantini R, Gratziu S, Tongiorgi E. 1965 . Oxygen isotopic composition in leaves. In: Use of isotopes and radiations in soil-plant nutrition studies. Technical Report Series No. 206. Isotopic Atomic Energy Commission, Vienna, 405–410. Helliker BR, Ehleringer R. 2000 . Establishing a grassland signature in veins: 18O in the leaf water of C3 and C4 grasses. Proceedings of National Academy of Sciences, USA 97, 7894 –7898. Helliker BR, Ehleringer R. 2002 . Differential 18O enrichment of leaf cellulose in C3 versus C4 grasses. Functional Plant Biology 29, 435 –442. Impa SM, Nadaradjan S, Boominathan P, Shashidhara G, Bindumadhava H, Sheshshayee MS. 2005 . Carbon isotope discrimination accurately reflects variability in WUE measured at a whole plant level in rice (Oryza sativa L.). Crop Science (in press). Leaney F, Osmond C, Allison G, Ziegler H. 1985 . Hydrogen-isotope composition of leaf water in C3 and C4 plants: its relationship to the hydrogen isotope composition of dry matter. Planta 164, 215 –220. Lindroth A, Halldin S. 1986 . Numerical analysis of pine forest evaporation and surface resistance. Agricultural and Forest Meteorology 38, 59 –79. Majoube M. 1971 . Fractionnement en oxygene 18 et en deuterium entre l'eau et sa vapeur. Journal of de Chimie Physique et de Physico-Chimie Biologique 58, 1423 –1436. Passioura JB. 1976 . Physiology of grain yield in wheat growing on stored water. Australian Journal of Plant Physiology 3, 559 –565. Rao RCN, Udayakumar M, Farquhar GD, Talwar HS, Prasad TG. 1995 . Variations in carbon isotope discrimination and its relationship with specific leaf area and Rubisco content in groundnut genotypes. Australian Journal of Plant Physiology 22, 545 –551. Rebetzke GJ, Condon G, Richards RA, Farquhar GD. 2002 . Selection for reduced carbon isotope discrimination increases aerial biomass and grain yield of rain-fed bread wheat. Crop Science 42, 122 –127. Richards RA, Rebetzke GJ, Condon AG, van Herwaarden AF. 2002 . Breeding opportunities for increasing the efficiency of water use and crop yield in temperate cereals. Crop Science 42, 111 –121. Roden JS, Ehleringer JR. 1999 . Observations of hydrogen and oxygen isotopes in leaf water confirm the Craig–Gordon model under wide-ranging environmental conditions. Plant Physiology 120, 1165 –1173. Scrimgeour CM. 1995 . Measurement of plant and soil water isotope composition by direct equilibration methods. Journal of Hydrology 172, 261 –274. Sheshshayee MS. 1998 . Dual isotope ratio (Δ13C and Δ18O): an approach to assess mesophyll factors associated with WUE. Project report presented at group monitoring workshop in March 1998. Tiruvanathapuram, India: Tropical Botanical Garden and Research Institute. Sheshshayee MS, Bindumadhava H, Shankar AG, Prasad, TG, Udayakumar M. 2003 . Breeding strategies to exploit water use efficiency for crop improvement. Journal of Plant Biology 30, 253 –268. Sternberg LSL, DeNiro MJ, Savidge RA. 1986 . Oxygen isotope exchange between metabolites and water during biochemical reactions leading to cellulose synthesis. Plant Physiology 82, 423 –427. Udayakumar M, Rao RCN, Wright GC, Ramaswamy GC, Ashok RS, Gangadhar GC, Aftab Hussain IS. 1998b. Measurement of transpiration efficiency in field conditions. Journal of Plant Physiology and Biochemistry 1, 69 –75. Udayakumar M, Sheshshayee MS, Nataraj KN, Bindumadahava H, Devendra R, Aftab Hussain IS, Prasad TG. 1998a. Why has breeding for water use efficiency not been successful? An analysis and alternate approach to exploit this trait for crop improvement. Current Science 74, 994 –1000. Walker CD, Leaney FW, Dighton JC, Allison GB. 1989 . The influence of transpiration on the equilibration of leaf water with atmospheric water vapour. Plant, Cell and Environment 12, 221 –234. Wang XF, Yakir D. 1995 . Temporal and spatial variations in the oxygen-18 content of leaf water in different plant species. Plant, Cell and Environment 18, 1377 –1385. White JWC, Lawrence JR, Broecker WS. 1994 . Modeling and interpreting D/H ratios in tree rings: A test case of white pine in the Northeastern United States. Geochimica et Cosmochimica Acta 58, 851 –862. Wright GC, Hubick KT, Farquhar GD. 1988 . Discrimination in carbon isotopes of leaves correlate with water use efficiency of field-grown peanut cultivars. Australian Journal of Plant Physiology 15, 815 –825. Yakir D. 1992 . Water compartmentation in plant tissue: isotopic evidence. In: Somero GN, Osmond CB, Bolis L, eds. Water and life. Berlin; Heidelberg, Germany: Springer-Verlag, 205–222. Yakir D. 1998 . Oxygen-18 of leaf water: a crossroads for plant associated isotopic signals: In: Griffiths H, ed. Stable isotopes: integration of biological and geochemical process. Oxford: BIOS Scientific Publishers, 147–168. Yakir D, DeNiro MJ, Gat JR. 1990 . Natural deuterium and oxygen-18 enrichment in leaf water of cotton plants grown under wet and dry conditions: evidence for water compartmentation and its dynamics. Plant, Cell and Environment 13, 49 –56. © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]

Huang, Chenghong;He, Wenliang;Guo, Jinkui;Chang, Xuexiang;Su, Peixi;Zhang, Lixin

doi: 10.1093/jxb/eri301pmid: 16263910

Abstract The Arabidopsis thaliana ascorbate-deficient vtc-1 mutant has only 30% ascorbate contents of the wild type (WT). This ascorbate-deficient mutant was used here to study the physiological roles of ascorbate under salt stress in vivo. Salt stress resulted in a more significant decrease in CO2 assimilatory capacity in the vtc-1 mutant than in the WT. Photosystem II function in the Arabidopsis vtc-1 mutant also showed an increased sensitivity to salt stress. Oxidative stress, indicated by the hydrogen peroxide content, increased more dramatically in the vtc-1 mutant than in the WT under salt stress. To clarify the reason for the increased oxidative stress in the vtc-1 mutant, the contents of small antioxidant compounds and the activities of several antioxidant enzymes in the ascorbate–glutathione cycle were measured. Despite an elevated glutathione pool in the vtc-1 mutant, the ascorbate contents and the reduced form of ascorbate decreased very rapidly under salt stress. These results showed that the activities of MDAR and DHAR were lower in the vtc-1 mutant than in the WT under salt stress. Thus, low intrinsic ascorbate and an impaired ascorbate–glutathione cycle in the vtc-1 mutant under salt stress probably induced a dramatic decrease in the reduced form of ascorbate, which resulted in both enhanced ROS contents and decreased NPQ in the vtc-1 mutant. Arabidopsis thaliana, ascorbate–glutathione cycle, oxidative stress, photosystem II, vtc-1 Introduction Ascorbate is an abundant small molecule in plants. Ascorbate is a key substance in the network of antioxidants that include ascorbate, glutathione, α-tocopherol, and a series of antioxidant enzymes. Ascorbate has also been shown to play multiple roles in plant growth, such as in cell division, cell wall expansion, and other developmental processes (Arrigoni, 1994; Smirnoff, 1996; Asada, 1999; Conklin, 2001; Pignocchi and Foyer, 2003). Many stress conditions, such as drought, salt, extreme temperatures, nutrient deprivation, UV-B radiation, and air pollutants can cause an increase of reactive oxygen species (ROS). Ascorbate, as an antioxidant, detoxifies H2O2, which is formed by the dismutation of \(\mathrm{O}_{2}^{{-}}.\) Ascorbate functions co-ordinately with glutathione and several enzymatic antioxidants to counteract \(\mathrm{O}_{2}^{{-}},\) which is produced by the Mehler reaction and photorespiration (Noctor and Foyer, 1998). Ascorbate is also believed to detoxify 1O2 and OH· (Smirnoff, 1996; Noctor and Foyer, 1998; Asada, 1999). As well as an antioxidant, ascorbate is a cofactor of violaxanthin de-epoxidase, an enzyme that converts violaxanthin to zeaxanthin under excess light, which is involved in the non-photochemical quenching of excess excited energy in photosystem II (PSII) (Demmig-Adams and Adams, 1990; Eskling et al., 1997). Therefore, ascorbate plays crucial roles in both scavenging ROS produced in photosynthesis and dissipating excess photons (Demmig-Adams and Adams, 1992; Niyogi, 1999). A mutant of Arabidopsis thaliana deficient in ascorbate (vtc-1 mutant) was isolated via its sensitivity to ozone exposure and has 30–60% ascorbate contents of the WT (Conklin et al., 1996). A point mutation occurs in vtc-1's gene of GDP-mannose pyrophosphorylase, a key enzyme in the ascorbate biosynthesis pathway. The mutation results in the considerable decrease in the activity of GDP-mannose pyrophosphorylase (Conklin et al., 1999), which controls ascorbate biosynthesis pathway flux (Keller et al., 1999). Previous work has shown that photosynthesis and the oxidative system are not perturbed in the vtc-1 mutant under normal growth conditions, although the vtc-1 mutant is smaller than the WT and shows retarded flowering and accelerated senescence (Veljovic-Jovanovic et al., 2001). Salinity stress may result in the accumulation of ROS (Hernández et al., 1995). It has also been reported that exogenous ascorbate can increase resistance to salt stress and reduce oxidative stress (Shalata and Neumann, 2001). In response to a combination of high light intensity and salt stress, ascorbate-deficient mutants of Arabidopsis thaliana showed increased sensitivity to NaCl-induced photo-oxidation (Smirnoff, 2000). Arabidopsis ascorbate-deficient vtc-1 mutants were used to study the physiological roles of ascorbate under short-term salt stress in vivo. Materials and methods Plant materials and growth conditions Arabidopsis thaliana ecotype Col-0, the background of the ascorbate-deficient mutant vtc-1, was used as the WT in this study. Seedlings of A. thaliana WT and the vtc-1 mutant were grown in a culture room at 22 °C with 12 h photoperiod under a photon flux density of 150 μmol photons m−2 s−1. Plastic pots containing 300 cm3 peat were used in the experiments. NaCl was dissolved in half-strength Hoagland nutrient solution and the plants were watered to drip with approximately 150 ml of 200 mM NaCl solution after the seedlings were grown for 5 weeks. The fully expanded leaves were collected for the experiments. For the ascorbate feeding experiment, 20 mM ascorbate was dissolved in half-strength Hoagland solution to water Arabidopsis 1 week before salt treatment and during the salt treatment experiment. Determination of H2O2 contents H2O2 contents were determined by the peroxidase-coupled assay according to Veljovic-Jovanovic et al. (2002) modified from Okuda et al. (1991). 0.1 g of Arabidopsis leaves were ground in liquid nitrogen and the powder was extracted in 2 ml 1 M HClO4 in the presence of insoluble PVP (5%). The homogenate was centrifuged at 12 000 g for 10 min and the supernatant was neutralized with 5 M K2CO3 to pH 5.6 in the presence of 100 μl 0.3 M phosphate buffer (pH 5.6). The solution was centrifuged at 12 000 g for 1 min and the sample was incubated for 10 min with 1 U ascorbate oxidase (Sigma, St Louis, USA) to oxidize ascorbate prior to the assay. The reaction mixture consisted of 0.1 M phosphate buffer (pH 6.5), 3.3 mM DMAB (Sigma, St Louis, USA), 0.07 mM MBTH (Sigma, St Louis, USA), and 0.3 U POX (Sigma, St Louis, USA). The reaction was initiated by the addition of 200 μl of the sample. The absorbance change at 590 nm was monitored at 25 °C. Ascorbate content determination Ascorbate contents were determined according to the method as described previously (Foyer et al., 1983) with some modifications. 0.1 g of Arabidopsis leaves were ground in liquid nitrogen and 1 ml of 2.5 M perchloric acid was added. The crude extract was centrifuged at 2 °C for 10 min at 10 000 g, and the supernatant was neutralized with saturated Na2CO3 using methyl orange as the indicator. The reduced ascorbate was assayed spectrophotometrically at 265 nm in 1 M NaH2PO4 buffer, pH 5.6, with 1 U ascorbate oxidase. The total ascorbate was assayed after incubation in the presence of 30 mM DTT. Glutathione content determination Total glutathione contents were measured as described by Anderson et al. (1992) except that 7% sulphosalicylic acid was used as the extraction buffer. 0.2 g of Arabidopsis leaves were ground in liquid nitrogen and 1 ml of the extraction buffer was added. The crude extract was centrifuged at 2 °C for 10 min at 10 000 g. 400 μl reagent 1 (110 mM Na2HPO4, 40 mM NaH2PO4, 15 mM EDTA, 0.3 mM 5,5′-dithiobis(2-nitrobenzoic acid), and 0.04% BSA), 320 μl reagent 2 (1 mM EDTA, 50 mM imidazole solution, and 0.02% BSA), 320 μl 5% Na2HPO4 (pH 7.5), 1.5 U glutathione reductase, 80 μl extract, and 80 μl NADPHNa4 were mixed. The reaction mixture was measured at 412 nm. Oxidized glutathione was measured as total glutathione except that 1 ml extract was incubated with 40 μl 2-vinylpyridine for 1 h at 25 °C before mixing. Ascorbate–glutathione cycle enzymes activity determination The leaves were homogenized in 50 mM phosphate buffer (pH 7.0), 0.1 mM EDTA, 0.1% phenylmethylsulphonyl fluoride, 1% polyvinylpyrrolidone, 0.1% Triton X-100, and 1 mM ascorbate. The homogenate was centrifuged at 12 000 g for 15 min at 4 °C. The extract was stored at −70 °C or used in enzyme assays immediately. Glutathione reductase (GR) and dehydroascorbate reductase (DHAR) activities were determined according to the method of Foyer and Halliwell (1976). 860 μl 1 mM oxidized glutathione, 100 μl 2 mM NADPH, and 40 μl crude enzyme were mixed and GR activity was measured at 340 nm. 700 μl phosphate buffer (pH 7.0), 700 μl 20 mM reduced glutathione, 100 μl 2 mM dehydroascorbate, and 100 μl crude enzyme were mixed and DHAR activity was measured at 265 nm. Monodehydroascorbate reductase (MDAR) was measured according to Hossain and Asada (1984). 900 μl 2 mM ascorbate, 2 U ascorbate oxidase, 30 μl 2 mM NADPH, and 30 μl crude enzyme were mixed and measured at 340 nm. All the spectrophotometrical assays were determined with DU series 640 spectrophotometer (Beckman Coulter, Inc., Fullerton, CA, USA). Chlorophyll fluorescence Plant chlorophyll fluorescence was determined by a portable fluorometer (PAM-2000, Walz, Effeltrich, Germany) connected with a leaf-clip holder (2030-B, Walz) and with a trifurcated fibre-optic (2010-F, Walz). Data acquisition software (DA-2000, Walz) was used in a notebook computer to dispose data on-line. The measurements were followed essentially according to Lu and Zhang (1998, 2000). Before measurement, the leaves were dark-adapted for 30 min. The minimal fluorescence level in the dark-adapted state (F0) was measured using the measuring light which is sufficiently low (0.8 μmol m−2 s−1) so as not to induce notable variable fluorescence. Far-red light (5 μmol m−2 s−1) was adopted to oxidize the PSII fully before measurement of the minimal fluorescence during illumination ( \(F{^\prime}_{0}\) ). Both of the maximal fluorescence levels in the dark (Fm) and under illumination ( \(F{^\prime}_{\mathrm{m}}\) ) were obtained by a saturating pulse (8000 μmol m−2 s−1). The steady-state fluorescence (Fs) was recorded after actinic light illumination for approximately 3 min. The maximum photochemical efficiency of PSII was determined from the ratio of variable (Fv) to maximum (Fm) fluorescence (Fv/Fm=(Fm−Fo)/Fm) (Kitajima and Butler, 1975). The efficiency of excitation capture by open PSII centres was calculated as \(F{^\prime}_{\mathrm{v}}/F{^\prime}_{\mathrm{m}},\) and the actual PSII efficiency (ΦPSII) was calculated from \({\Phi}_{\mathrm{PSII}}{=}(F{^\prime}_{\mathrm{m}}{-}F_{\mathrm{s}})/F{^\prime}_{\mathrm{m}}\) (Genty et al., 1989). The photochemical fluorescence quenching efficiency (qP) was calculated from \(qP{=}(F{^\prime}_{\mathrm{m}}{-}F_{\mathrm{s}})/F{^\prime}_{\mathrm{m}}{-}F{^\prime}_{0})\) (van Kooten and Snel, 1990). Non-photochemical quenching was calculated from \(NPQ{=}F_{\mathrm{m}}/F{^\prime}_{\mathrm{m}}{-}1\) (Bilger and Björkman, 1990). All the above measurements were performed in a dark room with stable ambient conditions. Photosynthetic analysis The LI-6400 portable photosynthesis system with a LI-6250 CO2 analyser (Li-Cor, Inc., Lincoln, Nebraska, USA) was used in photosynthetic analysis. The apparent photosynthetic rate (Photo.), stomatal conductance (gs), and intercellular CO2 concentration (Ci) of the fully expanded leaves were measured according to the manual of the instrument. Relative water contents (RWC) RWC was measured as described in Munné-Bosch and Alegre (2002a). Dry material was obtained after being heated at 80 °C for 48 h. RWC was calculated using the formulae: RWC (%)=(FW–DW)/FW where FW is the fresh weight and DW is the dry weight. Chlorophyll content determination Chlorophyll was extracted and measured as described previously by Porra et al. (1989). Results Effects of salt stress on relative water contents and chlorophyll contents RWC is approximately the same in both the WT and the vtc-1 mutant. Under salt stress, RWC in both the WT and the vtc-1 mutant decreased progressively (Fig. 1A). After 48 h treatment, RWC was about 76% in the WT, and 63% in the mutant. Fig. 1. View largeDownload slide Changes in relative water contents (RWC, A), chlorophyll contents (B), and the ratio of Chl a to Chl b (Chl a/b) (C) under salt stress, and chlorophyll contents without salt treatment in the WT and the ascorbate-deficiency vtc-1 mutant of Arabidopsis (D). After growing for 5 weeks, the Arabidopsis seedlings were irrigated with 200 mM NaCl for 12, 24, and 48 h, and the fully expanded leaves were collected for the measurement of RWC and chlorophyll contents. Mean values and SE were calculated from five independent experiments. Within each set of experiments, bars with different letters were significantly different at the 0.05 level. Fig. 1. View largeDownload slide Changes in relative water contents (RWC, A), chlorophyll contents (B), and the ratio of Chl a to Chl b (Chl a/b) (C) under salt stress, and chlorophyll contents without salt treatment in the WT and the ascorbate-deficiency vtc-1 mutant of Arabidopsis (D). After growing for 5 weeks, the Arabidopsis seedlings were irrigated with 200 mM NaCl for 12, 24, and 48 h, and the fully expanded leaves were collected for the measurement of RWC and chlorophyll contents. Mean values and SE were calculated from five independent experiments. Within each set of experiments, bars with different letters were significantly different at the 0.05 level. Figure 1 also shows the changes in chlorophyll contents on the basis of leaf area (B) and the ratio of chlorophyll a to b (C) in the WT and vtc-1 mutant under salt stress. Under normal conditions, chlorophyll contents and the ratio of chlorophyll a to b were slightly higher in the vtc-1 mutant than those in the WT, which is consistent with other reports (Veljovic-Jovanovic et al., 2001; Munné-Bosch and Alegre, 2002b). After NaCl treatment for 48 h, chlorophyll contents decreased by approximately 26% in the WT and by 53% in the vtc-1 mutant, and the ratio of chlorophyll a to b decreased by 8.5% in the WT and by 21.8% in the mutant. The chlorophyll contents remained almost constant in the WT and the vtc-1 mutant without salt treatment (Fig. 1D), which indicated that the vtc-1 mutant leaves used in the experiment were not at the senescent stages. Effects of salt stress on CO2 assimilation and stomatal conductance The photosynthesis rate, which was measured by the CO2 assimilatory capacity, changed from 9.1 to 7.0 μmol m−2 s−1 in the WT, and from 8.0 to 4.3 μmol m−2 s−1 in the vtc-1 mutant after NaCl treatment for 48 h (Fig. 2A). Fig. 2. View largeDownload slide Changes in photosynthetic rate (Photo., A), stomatal conductance (gs, B), and intercellular CO2 concentration (Ci, C) in the WT and vtc-1 mutant of Arabidopsis. After growing for 5 weeks, the Arabidopsis seedlings were treated as described in Fig. 1. Mean values and SE were calculated from five independent experiments. Fig. 2. View largeDownload slide Changes in photosynthetic rate (Photo., A), stomatal conductance (gs, B), and intercellular CO2 concentration (Ci, C) in the WT and vtc-1 mutant of Arabidopsis. After growing for 5 weeks, the Arabidopsis seedlings were treated as described in Fig. 1. Mean values and SE were calculated from five independent experiments. The stomatal conductance (gs) in the vtc-1 mutant was slightly higher than that in the WT under normal growth condition (Fig. 2B). After the first 12 h NaCl treatment, it decreased by approximately 58% in the WT and by 60% in the vtc-1 mutant. After 48 h treatment, gs was about 12% of the control in both the WT and vtc-1 mutant (Fig. 2B). The intercellular CO2 concentrations (Ci) were slightly higher in the vtc-1 mutant than in the WT under normal conditions and during NaCl treatment (Fig. 2C). A similar decreasing trend was found in both the WT and the vtc-1 mutant during NaCl treatment, from about 250 to 100 μmol CO2 mol−1 and from 285 to 125 μmol CO2 mol−1 in the WT and the vtc-1 mutant, respectively. Effects of salt stress on PSII photochemical activities Under normal conditions, the maximal efficiency of PSII photochemistry (Fv/Fm) is approximately similar (0.84) in the WT and the vtc-1 mutant. The maximal efficiency of PSII decreased in both the WT and the vtc-1 mutant for the duration of treatment, and the decreasing extent was greater in the vtc-1 mutant than in the WT. After 48 h treatment, Fv/Fm decreased from 0.83 to 0.79 in the WT, and from 0.84 to 0.52 in the vtc-1 mutant (Fig. 3A). Fig. 3. View largeDownload slide Photosystem II functions in the WT and vtc-1 mutant of Arabidopsis under salt stress. After growing for 5 weeks, the Arabidopsis seedlings were treated as described in Fig. 1. The maximal efficiency of PSII photochemistry (Fv/Fm, A), the efficiency of excitation capture by open PSII reaction centres ( \(F_{\mathrm{v}}{^\prime}/F_{\mathrm{m}}{^\prime},\) B), the photochemical quenching coefficient (qP, C), the actual PSII efficiency (ΦPSII, D), and the non-photochemical quenching (NPQ, E) were measured according to the Materials and methods. PAR was approximately 140 μmol photons m−2 s−1 in the measure of \(F_{\mathrm{v}}{^\prime}/F_{\mathrm{m}}{^\prime}\) and NPQ. Mean values and SE were calculated from five independent experiments. Fig. 3. View largeDownload slide Photosystem II functions in the WT and vtc-1 mutant of Arabidopsis under salt stress. After growing for 5 weeks, the Arabidopsis seedlings were treated as described in Fig. 1. The maximal efficiency of PSII photochemistry (Fv/Fm, A), the efficiency of excitation capture by open PSII reaction centres ( \(F_{\mathrm{v}}{^\prime}/F_{\mathrm{m}}{^\prime},\) B), the photochemical quenching coefficient (qP, C), the actual PSII efficiency (ΦPSII, D), and the non-photochemical quenching (NPQ, E) were measured according to the Materials and methods. PAR was approximately 140 μmol photons m−2 s−1 in the measure of \(F_{\mathrm{v}}{^\prime}/F_{\mathrm{m}}{^\prime}\) and NPQ. Mean values and SE were calculated from five independent experiments. The responses of the efficiency of excitation energy capture by open PSII reaction centres ( \(F_{\mathrm{v}}{^\prime}/F_{\mathrm{m}}{^\prime}\) ), photochemical quenching (qP), and actual quantum yield of PSII electron transport (ΦPSII) to salt stress were shown in Fig. 3B–D. Their response patterns were similar to that of Fv/Fm. Non-photochemical quenching (NPQ) was almost the same in both the WT and the vtc-1 mutant under normal conditions. It decreased greatly in the vtc-1 mutant from 0.228 to 0.103 after 48 h treatment, whereas only a slight decrease was observed in the WT during NaCl treatment (Fig. 3E). In the vtc-1 mutant, mannose metabolism may also be affected (Lukowitz et al., 2001). Conklin et al. reported that ascorbate feeding did elevate ascorbate in the vtc-1 mutant (Conklin et al., 1996). To confirm that the decreased photochemical activities in the vtc-1 mutant were due to ascorbate deficiency, ascorbate feeding experiments were performed. When the leaves were treated with ascorbate, the decreased Fv/Fm ratio in the vtc-1 mutant after treatment with NaCl for 48 h was restored to the control value (Fig. 4). Fig. 4. View largeDownload slide Effects of exogenous ascorbate on the maximal efficiency of PSII photochemistry in the WT and vtc-1 mutant of Arabidopsis under salt stress. One week before salt treatment, 20 mM ascorbate was dissolved into half-strength Hoagland solution to water Arabidopsis. During the salt treatment experiment, 20 mM ascorbate was also added to the solution used above. After salt treatment for 48 h, Fv/Fm was measured according to the Materials and methods. Mean values and SE were calculated from five independent experiments. Within each set of experiments, bars with different letters were significantly different at the 0.05 level. Fig. 4. View largeDownload slide Effects of exogenous ascorbate on the maximal efficiency of PSII photochemistry in the WT and vtc-1 mutant of Arabidopsis under salt stress. One week before salt treatment, 20 mM ascorbate was dissolved into half-strength Hoagland solution to water Arabidopsis. During the salt treatment experiment, 20 mM ascorbate was also added to the solution used above. After salt treatment for 48 h, Fv/Fm was measured according to the Materials and methods. Mean values and SE were calculated from five independent experiments. Within each set of experiments, bars with different letters were significantly different at the 0.05 level. Changes of H2O2 contents and small antioxidant molecules under salt stress The H2O2 contents were similar both in the WT and the vtc-1 mutant without salt treatment, while the H2O2 contents increased under salt stress, especially in the vtc-1 mutant. The H2O2 contents increased by approximately 18% in the WT, while it increased by about 150% in the vtc-1 mutant after 48 h salt treatment (Fig. 5A). Fig. 5. View largeDownload slide Changes in H2O2 contents (A), total ascorbate contents (B), ratio of reduced to total ascorbate contents (C), and glutatione contents (D) in the WT and vtc-1 mutant of Arabidopsis under salt stress. After growing for 5 weeks, the Arabidopsis seedlings were treated as described in Fig. 1. The contents of H2O2, ascorbate, and glutathione were expressed on the basis of dry weight. Mean values and SE were calculated from five independent experiments. Fig. 5. View largeDownload slide Changes in H2O2 contents (A), total ascorbate contents (B), ratio of reduced to total ascorbate contents (C), and glutatione contents (D) in the WT and vtc-1 mutant of Arabidopsis under salt stress. After growing for 5 weeks, the Arabidopsis seedlings were treated as described in Fig. 1. The contents of H2O2, ascorbate, and glutathione were expressed on the basis of dry weight. Mean values and SE were calculated from five independent experiments. As an ascorbate-deficiency mutant, the ascorbate contents in the vtc-1 mutant were about 30% of the WT in 5-week-old fully extended leaves under normal conditions in this study (about 17.3 μmol g−1 DW in the WT and 5.0 μmol g−1 DW in the vtc-1 mutant). After salt stress for 12 h, ascorbate contents increased by about 54% and 27% in the WT and the vtc-1 mutant, respectively. However, they decreased afterwards and after 48 h treatment they were approximately 77% and 15% of the control in the WT and the vtc-1 mutant, respectively (Fig. 5B). The ratio of reduced to total ascorbate decreased progressively under salt stress, from 95% to 45% in the WT, and from 95% to 9% in the vtc-1 mutant (Fig. 5C). The total glutathione contents were higher in the vtc-1 mutant under normal conditions, which is consistent with previous reports (Veljovic-Jovanovic et al., 2001). During the NaCl treatment, the glutathione contents increased during the 24 h salt treatment by about 54% in the WT and 73% in the vtc-1 mutant, respectively. After 48 h treatment, it showed a greater increase in the WT than in the vtc-1 mutant (Fig. 5D). The ratio of reduced to total glutathione remained almost constant, with about 95% in the WT and the vtc-1 mutant. Changes of ascorbate–glutathione cycle reducing enzymes activity under salt stress To elucidate the mechanism of redox balance maintenance, the activities of enzymes in the ascorbate–glutathione cycle were measured. GR is a thiol enzyme which uses NADPH as an electron donor and reduces oxidized glutathione (GSSG) to the reduced form. GR activity increased by 70% and 50% in the WT and the vtc-1 mutant during the 48 h salt treatment, respectively (Fig. 6A). Fig. 6. View largeDownload slide Changes in relative activities of GR (A), MDAR (B), and DHAR (C) in the ascorbate–glutathione cycle, in the WT and vtc-1 mutant of Arabidopsis under salt stress. After growing for 5 weeks, the Arabidopsis seedlings were treated as described in Fig. 1. Mean values and SE were calculated from five independent experiments. Fig. 6. View largeDownload slide Changes in relative activities of GR (A), MDAR (B), and DHAR (C) in the ascorbate–glutathione cycle, in the WT and vtc-1 mutant of Arabidopsis under salt stress. After growing for 5 weeks, the Arabidopsis seedlings were treated as described in Fig. 1. Mean values and SE were calculated from five independent experiments. Using NADPH as an electron donor, MDAR catalyses monodehydroascorbate to dehydroascorbate and ascorbate. A slight increase was observed during the first 12 h treatment in the WT, and the activity reached the control level after 48 h treatment, whereas the activity of MDAR decreased during the salt treatment in the vtc-1 mutant (Fig. 6B). Reduced glutathione and dehydroascorbate are the substrates of DHAR. It is an important enzyme to reduce ascorbate. DHAR activities were almost the same in the WT and vtc-1 mutant under normal conditions. In WT, it increased by 70% during 12 h treatment, and decreased after 24 h treatment. In the vtc-1 mutant, DHAR activity levels were almost constant during the 24 h salt treatment. After 48 h treatment, the DHAR activity decreased by 60% (Fig. 6C). Discussion The Arabidopsis vtc-1 mutant was used to study the physiological roles of ascorbate under salt stress in vivo. CO2 assimilation rate and PSII function in the Arabidopsis vtc-1 mutant showed increased sensitivity to salt stress. More oxidative stress occurred in the vtc-1 mutant under salt stress than in the WT. Under salt stress, the metabolism of the ascorbate–glutathione cycle seems to be impaired in the vtc-1 mutant, most probably due to the reduced ascorbate contents. Decrease of PSII function exacerbated under salt stress in the vtc-1 mutant Photosynthetic CO2 exchange and chlorophyll a fluorescence in vivo were measured to investigate the effects of salt stress on photosynthetic functions in the ascorbate-deficient vtc-1 mutant. The photosynthesis rate, reflected by the CO2 assimilation capacity, decreased much more in the vtc-1 mutant than in the WT under salt stress (Fig. 2A). The slightly higher values of stomatal conductance and intercellular CO2 in the vtc-1 mutant than that in WT under salt stress (Fig. 2B, C) indicated that the decrease of CO2 assimilation was not caused by CO2 limitation in plant cells. Photosynthetic CO2 assimilation is considered to be a major sink for reducing equivalents (ATP and NADPH) generated by the primary photochemical reaction. The normal recycling of NADPH is especially important for the function of NADP+ as the terminal electron acceptor and maintains the photochemical de-excitation of reaction centres in a steady-state. In addition, the proton motive force through the coupling of electron transport to ATP synthesis regulates the amplitude of the photosynthetic electron transport by feedback inhibition (Krause, 1994). This serves as a dissipation mechanism for excess excitation energy when the rate of ATP and NADPH synthesis exceeds the demand for CO2 fixation. Thus, the dramatic decrease in CO2 assimilation would result in the decrease of PSII function in the vtc-1 mutant under salt stress through a feedback system. The PSII activity, reflected by the maximum efficiency of PSII photochemistry measured as Fv/Fm, decreased much more in the vtc-1 mutant than that in WT under salt stress (Fig. 3A). In order to examine the possible changes in PSII photochemistry under normal light irradiation, the fluorescence characteristics during the steady-state of photosynthesis was investigated. Photochemical quenching (qP) decreased much more in the vtc-1 mutant than in the WT (Fig. 3C), which indicates a significant increase in the proportion of closed PSII reaction centres or the proportion of the reduced state of QA (Dietz et al., 1985; Genty et al., 1989). An increase in the fraction of QA in the reduced state estimated from the decreased qP in the vtc-1 mutant under salt stress suggests a decrease in the proportion of available excitation energy used for photochemistry (Havaux et al., 1991) and an increase in the excitation pressure on PSII under the steady-state of photosynthesis (Öquist and Huner, 1993). Such an increase in excitation pressure would result in further damage to PSII if excess excitation pressure was not dissipated safely since the excitation pressure on PSII has been shown to be a determining factor for photodamage on PSII (Demmig-Adams and Adams, 1992; Chow, 1994). In this study, NPQ decreased dramatically in the vtc-1 mutant under salt stress (Fig. 3E), whereas it remained relatively constant in the WT. This result suggested defective thermal dissipation in the vtc-1 mutant. As a result, the efficiency of excitation energy capture by open PSII reaction centres ( \(F_{\mathrm{v}}{^\prime}/F_{\mathrm{m}}{^\prime}\) ) (Fig. 3B) and actual quantum yield of PSII electron transport (ΦPSII) decreased much more in the vtc-1 mutant than in the WT under salt stress (Fig. 3D). The decrease of actual quantum yield of PSII was due to a decrease in both photochemical quenching (qP) and the efficiency of excitation pressure ( \(F_{\mathrm{v}}{^\prime}/F_{\mathrm{m}}{^\prime}\) ). The decrease of PSII function in the vtc-1 mutant under salt stress is perhaps due to photodamage caused by the over-reduction of the photosynthetic electron transport chain and decreased excitation dissipation. The ability of ascorbate to restore the Fv/Fm ratio to the control value in the vtc-1 mutant after treatment with NaCl for 48 h indicates that the decreased photosynthetic function was due to ascorbate deficiency. Therefore, the photosynthetic process in the vtc-1 mutant is more sensitive to salt stress perturbation than that in the WT. Increased oxidative stress due to impaired ascorbate–glutathione cycle in the vtc-1 mutant under salt stress Under 200 mM NaCl treatment, the decrease in chlorophyll contents and the ratio of chlorophyll a to b accelerated in the vtc-1 mutant compared with those in the WT (Fig. 1B, C). This result indicates that the vtc-1 mutant suffers more damage from salt stress than the WT. Ascorbate plays a central role in photosynthetic protection, for it functions in ROS scavenging in the Mehler peroxidase reaction as a reductant and in excess light energy dissipation in NPQ (Eskling et al., 1997; Noctor and Foyer, 1998; Asada, 1999; Müller-Moule et al., 2002, 2003, 2004). The accumulation of more H2O2 in the vtc-1 mutant than in the WT during salt treatment indicated that more oxidative stress takes place in the vtc-1 mutant under salt stress (Fig. 5A). There was about 30% ascorbate compared with the WT in 5-week-old fully extended leaves of the vtc-1 mutant under normal conditions without an increase of ROS and NPQ, which is consistent with other reports (Veljovic-Jovanovic et al., 2001; Munné-Bosch and Alegre, 2002b). This result suggests that 30% ascorbate was sufficient to maintain normal plant physiological activity under normal conditions. However, under salt stress the contents of ascorbate and the ratio of reduced to total ascorbate declined more dramatically in the vtc-1 mutant than that in the WT (Fig. 5B, C), which indicates that the regeneration of reduced ascorbate is impaired in the vtc-1 mutant. The ascorbate–glutathione cycle is very efficient in regenerating the reduced forms of ascorbate. In this cycle monodehyroascorbate radical and DHA are reduced to ascorbate by NAD(P)-dependent MDAR and glutathione-dependent DHAR, respectively. The oxidative form of glutathione (GSSG) is reduced by GR (Noctor and Foyer, 1998). Under salt stress the activity of GR increased and the glutathione contents in the vtc-1 mutant accumulated even more than in the WT, while the redox states of glutathione remained almost constant in both the WT and the vtc-1 mutant (Fig. 5D, 6A). These results are consistent with previous reports that glutathione synthesis was elevated in plants under environmental stress (Foyer et al., 1997; Ruiz and Blumwald, 2002). Thus, the changes in glutathione metabolism obviously may not account for the ascorbate decrease in the vtc-1 mutant. The activities of MDAR and DHAR, two ascorbate reducing enzymes in the ascorbate–glutathione cycle, are crucial to regenerate ascorbate. Overexpression of the DHAR gene increased the ascorbate contents in transgenic plants, which demonstrates that the ascorbate contents of plants can be elevated through enhanced ascorbate recycling (Chen et al., 2003). DHAR overproducing transgenic plants also showed enhanced oxidative resistant ability (Kwon et al., 2003). The transcripts of the MDAR and DHAR genes are induced by oxidative stress so as to meet the requirement for the regeneration of reduced ascorbate upon increased oxidative stress (Grantz et al., 1995; Chew et al., 2003). These results have shown that the activities of MDAR and DHAR were lower in the vtc-1 mutant than in the WT under salt stress. They do not meet the requirement for a higher ascorbate reducing capability in the vtc-1 mutant, which induced the inhibition of the ascorbate–glutathione cycle under salt stress. That both the total ascorbate contents and the ratio of reduced to total ascorbate decreased dramatically in the vtc-1 mutant under salt stress may be partially due to the blocked ascorbate recycling. Thus, the ascorbate-deficiency in the vtc-1 mutant seems to be the main reason for the sensitivity of the vtc-1 mutant to salt stress. Low intrinsic ascorbate and an impaired ascorbate–glutathione cycle in the vtc-1 mutant under salt stress probably induced an excessive decrease of the reduced form of ascorbate, which induced both enhanced ROS contents and decreased NPQ in the vtc-1 mutant. The dramatic decrease of NPQ in the vtc-1 mutant, a process requiring ascorbate as a cofactor of violaxanthin de-epoxidase in the xanthophyll cycle, could also provide an explanation. The aggravated oxidative stress and deficient excess energy dissipation are the considerable sources of the exacerbated decrease of PSII function in the vtc-1 mutant. The results above demonstrate the important role of ascorbate in a physiologically protective function under salt stress in vivo. This work was supported by the One Hundred Talent Project, Excellent PhD Thesis Foundation (199924), Key Natural Science Foundation of Gansu Province (ZS031-A25-034-D). We are grateful to Nottingham Stock Centre for the Arabidopsis seeds. References Anderson JV, Chevone BI, Hess JL. 1992. Seasonal variation in the antioxidant system of eastern white pine needles. Plant Physiology  98, 501–508. Google Scholar Arrigoni O. 1994. Ascorbate system in plant development. Journal of Bioenergetics and Biomembranes  26, 407–419. Google Scholar Asada K. 1999. The water–water cycle in chloroplasts: scavenging of active oxygens and dissipation of excess photons. Annual Review of Plant Physiology and Plant Molecular Biology  50, 601–639. Google Scholar Bilger W, Björkman O. 1990. Role of the xanthophyll cycle in photoprotection elucidated by measurements of light-induced absorbance changes, fluorescence and photosynthesis in Hedera canariensis. Photosynthesis Research  25, 173–185. Google Scholar Chen Z, Young TE, Chang S, Gallie DR. 2003. Increasing vitamin C content of plants through enhanced ascorbate recycling. Proceedings of the National Academy of Sciences, USA  100, 3525–3530. Google Scholar Chew O, Whelan J, Millar AH. 2003. Molecular definition of the ascorbate–glutathione cycle in Arabidopsis mitochondria reveals dual targeting of antioxidant defenses in plants. Journal of Biological Chemistry  278, 46869–46877. Google Scholar Chow WS. 1994. Photoprotection and photoinhibitory damage. Advances in Molecular and Cell Biology  10, 151–196. Google Scholar Conklin PL. 2001. Recent advances in the role and biosynthesis of ascorbic acid in plants. Plant, Cell and Environment  24, 383–394. Google Scholar Conklin PL, Norris SR, Wheeler GL, Williams EH, Smirnoff N, Last RL. 1999. Genetic evidence for the role of GDP-mannose in plant ascorbic acid (vitamin C) biosynthesis Proceedings of the National Academy of Sciences, USA  96, 4198–4203. Google Scholar Conklin PL, Williams EH, Last RL. 1996. Environmental stress sensitivity of an ascorbic acid-deficient Arabidopsis mutant. Proceedings of the National Academy of Sciences, USA  93, 9970–9974. Google Scholar Demmig-Adams B, Adams III WW. 1990. The carotenoids zeaxanthin and high-energy-state quenching of chlorophyll fluorescence. Photosynthesis Research  25, 187–198. Google Scholar Demmig-Adams B, Adams III WW. 1992. Photoprotection and other responses of plants to high light stress. Annual Review of Plant Physiology and Plant Molecular Biology  43, 599–626. Google Scholar Dietz KJ, Schreiber U, Heber U. 1985. The relationship between the redox state of QA and photosynthesis in leaves at various carbon-dioxide, oxygen and light regimes. Planta  198, 460–470. Google Scholar Eskling M, Arvidsson PO, Akerlund HE. 1997. The xanthophyll cycle, its regulation and components. Physiologia Plantarum  100, 806–816. Google Scholar Foyer CH, Halliwell B. 1976. The presence of glutathione and glutathione reductase in chloroplast: a proposed role in ascorbic acid metabolism. Planta  133, 21–25. Google Scholar Foyer CH, Lopez-Delgado H, Dat JF, Scott IM. 1997. Hydrogen peroxide- and glutathione-associated mechanisms of acclimatory stress tolerance and signaling. Physiologia Plantarum  100, 241–254. Google Scholar Foyer CH, Rowell J, Walker D. 1983. Measurements of the ascorbate content of spinach leaf protoplasts during illumination. Planta  157, 239–244. Google Scholar Genty B, Briantais JM, Baker NR. 1989. The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochimica et Biophysica Acta  990, 87–92. Google Scholar Grantz A, Brummell DA, Bennett AB. 1995. Ascorbate free-radical reductase messenger-RNA levels are induced by wounding. Plant Physiology  108, 411–418. Google Scholar Havaux M, Strasser RJ, Greppin H. 1991. A theoretical and experimental analysis of the qp and qN coefficients of chlorophyll fluorescence quenching and their relation to photochemical and non-photochemical events. Photosynthesis Research  27, 41–55. Google Scholar Hernández JA, Olmos E, Corpas FJ, Sevilla F, del Río LA. 1995. Salt-induced oxidative stress in chloroplasts of pea plants. Plant Science  105, 151–167. Google Scholar Hossain NA, Asada K. 1984. Purification of dehydroascorbate reductase from spinach and its characterization as a thiol enzyme. Plant and Cell Physiology  25, 85–92. Google Scholar Keller R, Springer F, Renz A, Kossmann J. 1999. Antisense inhibition of the GDP–mannose pyrophosphorylase reduces the ascorbate content in transgenic plants leading to developmental changes during senescence. The Plant Journal  19, 131–141. Google Scholar Kitajima M, Butler WL. 1975. Excitation spectra for photosystem I and photosystem II in chloroplasts and the spectral characteristics of the distributions of quanta between the two photosystems. Biochimica et Biophysica Acta  408, 297–305. Google Scholar Krause GH. 1994. The role of oxygen in photoinhibition of photosynthesis. In: Foyer CH, Mullineaux PM, eds. Causes of photo-oxidative stress and amelioration of defense systems in plants. Boca Raton: CRC Press, 43–76. Google Scholar Kwon SY, Choi SM, Ahn YO, Lee HS, Lee HB, Park YM, Kwak SS. 2003. Enhanced stress-tolerance of transgenic tobacco plants expressing a human dehydroascorbate reductase gene. Journal of Plant Physiology  160, 347–353. Google Scholar Lu C, Zhang J. 1998. Modifications in photosystem II photochemistry in senescent leaves of maize plants. Journal of Experimental Botany  49, 1671–1679. Google Scholar Lu C, Zhang J. 2000. Role of light in the response of PSII photochemistry to salt stress in the cyanobacterium Spirulina platensis. Journal of Experimental Botany  51, 911–917. Google Scholar Lukowitz W, Nickle TC, Meinke DW, Last RL, Conklin PL, Somerville CR. 2001. Arabidopsis cyt1 mutants are deficient in a mannose-1-phosphate guanylyltransferase and point to a requirement of N-linked glycosylation for cellulose biosynthesis. Proceedings of the National Academy of Sciences, USA  98, 2262–2267. Google Scholar Munné-Bosch S, Alegre L. 2002a. Plant aging increases oxidative stress in chloroplasts. Planta  214, 608–615. Google Scholar Munné-Bosch S, Alegre L. 2002b. Interplay between ascorbic acid and lipophilic antioxidant defences in chloroplasts of water-stressed Arabidopsis plants. FEBS Letters  524, 145–148. Google Scholar Müller-Moulé P, Conklin PL, Niyogi KK. 2002. Ascorbate deficiency can limit violaxanthin de-epoxidase activity in vivo. Plant Physiology  128, 970–977. Google Scholar Müller-Moulé P, Golan T, Niyogi KK. 2004. Ascorbate-deficient mutants of Arabidopsis grow in high light despite chronic photo-oxidative stress. Plant Physiology  134, 1163–1172. Google Scholar Müller-Moulé P, Havaux M, Niyogi KK. 2003. Zeaxanthin deficiency enhances the high light sensitivity of an ascorbate-deficient mutant of Arabidopsis. Plant Physiology  133, 748–760. Google Scholar NiyogiKK. 1999. Photoprotection revisited: genetic and molecular approaches. Annual Review of Plant Physiology and Plant Molecular Biology  50, 333–359. Google Scholar Noctor G, Foyer CH. 1998. Ascorbate and glutathione: keeping active oxygen under control. Annual Review of Plant Physiology and Plant Molecular Biology  49, 249–279. Google Scholar Okuda T, Matuda Y, Yamanaka A, Sagisaka S. 1991. Abrupt increase in the level of hydrogen peroxide in leaves of winter wheat is caused by cold treatment. Plant Physiology  97, 1265–1267. Google Scholar Öquist G, Huner NPA. 1993. Cold-hardening-induced resistance to photoinhibition of photosynthesis in winter rye is dependent upon an increased capacity for photosynthesis. Planta  189, 150–156. Google Scholar Pignocchi C, Foyer CH. 2003. Apoplastic ascorbate metabolism and its role in the regulation of cell signaling. Current Opinion in Plant Biology  6, 379–389. Google Scholar Porra RJ, Thompson WA, Kriedemann PE. 1989. Determination of accurate extinction coefficients and simultaneous equations for assaying chlorophyll a and b extracted with four different solvents: verification of the concentration of chlorophyll standards by atomic absorption spectroscope. Biochimica et Biophysica Acta  975, 384–394. Google Scholar Ruiz J, Blumwald E. 2002. Salinity-induced glutathione synthesis in Brassica napus. Planta  214, 965–969. Google Scholar Shalata A, Neumann PM. 2001. Exogenous ascorbic acid (vitamin C) increases resistance to salt stress and reduces lipid peroxidation. Journal of Experimental Botany  52, 2207–2211. Google Scholar Smirnoff N. 1996. The function and metabolism of ascorbic acid in plant. Annals of Botany  78, 661–669. Google Scholar Smirnoff N. 2000. Ascorbate biosynthesis and function in photoprotection. Philosophical Transactions of the Royal Society of London, Series B, Biological Sciences  355, 1455–1464. Google Scholar van Kooten O, Snel JFH. 1990. The use of chlorophyll fluorescence nomenclature in plant stress physiology. Photosynthesis Research  25, 147–150. Google Scholar Veljovic-Jovanovic SD, Pignocchi C, Noctor G, Foyer CH. 2001. Low ascorbic acid in the vtc-1 mutant of arabidopsis is associated with decreased growth and intracellular redistribution of the antioxidant system. Plant Physiology  127, 426–435. Google Scholar Veljovic-Jovanovic SD, Noctor G, Foyer CH. 2002. Are leaf hydrogen peroxide concentrations commonly overestimated? The potential influence of artefactual interference by tissue phenolics and ascorbate. Plant Physiology and Biochemistry  40, 501–507. Google Scholar © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]

Wang, Yu-Jun;Yu, Jia-Ning;Chen, Tao;Zhang, Zhi-Gang;Hao, Yu-Jun;Zhang, Jin-Song;Chen, Shou-Yi

doi: 10.1093/jxb/eri302pmid: 16275671

Abstract The cytosolic free-calcium concentration [Ca2+](cyt) transiently increases under abiotic stresses and the proteins that control this process are gradually disclosed. The Ca2+-permeable channel is one type of these proteins in plants. In the present study, a novel Ca2+-permeable channel gene TaTPC1 encoding a putative membrane protein was cloned from wheat. It was induced under high salinity, polyethylene glycol, low temperature (4 °C), and abscisic acid. Expression of TaTPC1 in the yeast mutant lacking CCH1 can recover its growth under lithium stress through functional complementation. TaTPC1 transgenic plants exhibited more stomatal closing in the presence of Ca2+ than the control, supporting a role for the calcium channel in regulating plant responses to environmental change. Calcium channel, stomatal closing, stress induction, transgenic plants, wheat Introduction High salinity, drought and low temperature adversely affect plant growth and crop production. Plants have common mechanisms in physiological responses to reduce the damage by abiotic stresses. One early response to these stresses in plant cells is an increase in cytosolic free-calcium concentration [Ca2+](cyt) by means of infusion from the apoplastic space and release from internal stores (Sanders et al., 1999; Xiong et al., 2002). The transient increases of cytosolic Ca2+ can be perceived by various Ca2+-binding proteins, such as CDPKs (Ca2+-dependent protein kinases) (Harmon et al., 2001), CBL1 (Cheong et al., 2003), and the SOS3 family of Ca2+ sensors, which are involved in coupling and transducing stress signals by specific protein phosphorylation cascades, and follow the plant stress response (Sheen, 1996). It is revealed that Ca2+, as a second messenger, plays an important role in stress signal transduction in plants (Sanders et al., 2002). The transport of Ca2+ is regulated by various Ca2+ channels, Ca2+-ATPases, and/or H+/Ca2+-antiporters on cell membranes. The entry of the Ca2+ into plant cells is through the function of Ca2+ channels in the plasma membrane, whereas the maintenance of an optimal Ca2+ concentration in cells is performed by Ca2+-ATPases and/or H+/Ca2+-antiporters. Therefore, calcium channels, like the Ca2+ transporters, were also involved in abiotic signal transduction by regulating the change in [Ca2+](cyt) (White, 2000; White and Broadley, 2003). Calcium channels have been reported to localize in the plasma membrane, tonoplast, endoplasmic reticulum, chloroplast, and nuclear membranes according to electrophysiological studies (Sanders et al., 2002). There exist different types of gating mechanisms of Ca2+ channels in the plant, such as ligand-, volage-, and stretch-activated (Demidchik et al., 2002). The electrophysiological and biochemical characteristics of Ca2+-permeable channels of plant cells are well known, but only a limited number of genes encoding Ca2+-permeable channels have been isolated and functionally studied (White et al., 2002). A wheat gene LCT1, encoding a low-affinity cation transporter, can complement a yeast mutant with a disruption in the MIDI gene, which encodes a stretch-activated Ca2+-permeable non-selective cation channel (Amtmann et al., 2001). The homologues of nucleotide-gated cation channels (CNGCS) have been cloned from barley, tobacco, and Arabidopsis. In Arabidopsis, CNGCSs comprise a gene family, one of which was expressed in human embryonic kindney cells, and a cyclic nucleotide-dependent increase in Ca2+ permeability was demonstrated. AtTPC1 from Arabidopsis encodes a two-pore voltage-gated channel with high affinity for Ca2+ permeation, and rescued the Ca2+ uptake activity of a yeast mutant cch1 (encoding an homologous L-type Ca2+ channel). [Ca2+](cyt) was enhanced by overexpression of AtTPC1 or suppressed by its antisense expression under sucrose stress (Furuichi et al., 2001). Most recently, the AtTPC1 protein has been found to be ubiquitous in plant vacuoles and this channel may regulate seed germination and stomatal movement (Peiter et al., 2005). Its homologue OsTPC1 from rice has also been identified and characterized (Hashimoto et al., 2004; Kurusu et al., 2004). OsTPC1-overexpressing plants showed reduced growth and abnormal greening of roots (Kurusu et al., 2004). From tobacco BY-2 cells, two homologous genes, NtTPC1A and NtTPC1B, were also identified. These two genes complemented the growth of the yeast mutant defective in CCH1, and co-suppression of them resulted in the inhibition of a rise in cytosolic free-Ca2+ concentration in response to sucrose and a fungal elicitor (Kadota et al., 2004). These results suggest that the Ca2+-permeable channels play an important role in mediating the spatial and temporal variation of Ca2+, and are involved in regulating the growth and development of plants. In the present study, a putative Ca2+-permeable channel gene TaTPC1 was cloned from wheat. Its expressions were induced in roots under abscisic acid (ABA) and various stress treatments. Expression of TaTPC1 in the yeast mutant lacking CCH1 (homologous to the 1-subunit of a voltage-gated Ca2+ channel) can recover its growth through functional complementation, and TaTPC1-overexpressing plants exhibited more stomatal closing under Ca2+ when compared with the control plants. Materials and methods Plant materials and treatments Wheat (China spring, Triticum aestivum) seeds were germinated and maintained on wet cheesecloth, supplemented with water, under continuous illumination at 25 °C. For cold treatment, the 10-d-old seedlings were kept at 4 °C for various periods and both shoots and roots were harvested for RNA isolation. For ABA, NaCl, and polyethylene glycol (PEG) treatments, the seedlings were transferred into solutions containing 100 μM ABA, 250 mM NaCl, and 25% PEG (w/v), respectively, treated for various periods, and then harvested. Construction of a wheat cDNA library and yeast one-hybrid screening A cDNA library was constructed with the mRNA from the salt- and PEG-treated wheat seedlings. According to the manufacturer's (Stratagene) protocol, the first-strand cDNAs were synthesized and followed by second-strand synthesis. The uneven termini of the double-stranded cDNA were filled in with Pfu DNA polymerase, and EcoRI adapters were ligated to the blunt ends. After XhoI digestion, the double-stranded cDNAs were ligated to the Uni-ZAP XR vector, and then packaged by Gigapack III Gold packaging extract. The library was amplified and converted to the pAD-GAL4 library by mass excision. For one-hybrid screening, the pAD-GAL4 plasmids containing wheat cDNA inserts were transformed into the Saccharomyces cerevisiae strain YRG2 harbouring the dual DRE-controlled reporter genes His3 and LacZ (Liu et al., 1998). Transformants were plated on medium lacking histidine, ura, and leucine but containing 3-AT (a competitive inhibitor of the HIS3 gene product). Approximately 1×107 clones were screened. Colonies larger than 1 mm were obtained and streaked on the same medium. After growing for 36–72 h, the colonies were further tested for the expression of β-galactosidase by filter lift assays (Stratagene). Plasmids isolated from the putative yeast colonies were transformed into Escherichia coli. The cDNA inserts in these plasmids were sequenced according to standard procedures. 5′-RACE for the 5′ sequence of TaTPC1 5′-RACE was performed because the isolated cDNA (TaTPC1) was partial in its 5′-end. Two specific primers were designed according to the partial cDNA sequence of TaTPC1 (GSP: 5′-CCAGCCAATTCTGCTACTTTCTGCAG-3′; NGSP: 5′-CTCGCGGTAGCAGCCAAGGATGGT-3′). 5′-RACE was carried out according to the SMART12 RACE cDNA Amplification Kit User Manual (Clontech). PCR reaction was performed for five cycles of 94 °C for 5 s, 72 °C for 2 min; five cycles of 94 °C for 5 s, 68 °C for 10 s, 72 °C for 2 min; 30 cycles of 94 °C for 5 s, 64 °C for 10 s, 72 °C for 2 min; and finally 72 °C for 10 min. Subsequently, the dilution of the PCR product was used in a nested PCR reaction. The nested PCR conditions were identical to the initial reaction. The products were analysed on a 1.2% agarose/EtBr gel and the corresponding DNA band was recovered and directly sequenced. Northern analysis RNA extraction was performed as described previously (Zhang et al., 1996). Total RNA (30 μg) was fractionated in a 1.2% agarose gel containing formaldehyde and blotted onto Hybond-N+ nylon membrane for northern analysis. The full-length of TaTPC1 cDNA was labelled as a probe and used in the hybridization. The membranes were washed in 2× SSC plus 0.1% SDS at 45 °C for 15 min and in 1× SSC plus 0.1% SDS at 45 °C for 5 min. The membranes were then autoradiographed using a phosphoimaging system (Amersham Pharmacia). Generation of yeast mutant for the CCH1 gene The S. cerevisiae strain used in this study was W303.1A (MATa leu2-3112 ura3-1 trp1-92 his3-11,15 ade2-1 can1-100 GAL SUC mal), which contains the yeast CCH1 gene. The CCH1 gene encodes a protein homologous to the α1-subunit of a voltage-gated L-type Ca2+ channel from mammalian cells. To knock out the CCH1 gene, a 706 bp fragment encoding a Trp marker gene was amplified from the pBD vector of the yeast two-hybrid system through PCR with 5′ primer: 5′-AGATCATCGTGGAATAGAATAGATCTGGTATCTTCTGFCAGTATTAAGCACACAAAGGCAGC-3′ and 3′ primer: 5′-CTGATTCGCCTTAAGCTTAAAATATTTCCAGTAGCCTGTTACAGTAATAACCTATTTCTTAGC-3′ (1513–2223 bp). The procedure was performed as described previously (Wang et al., 2005). The transformed yeast cells were screened on SD medium lacking Trp (SD–Trp medium). The wild-type yeast cells could not grow on SD–Trp medium, while the mutant cch1 with the mutation in the CCH1 gene could grow on it. TaTPC1 functional complementation assay The yeast plasmid pYES2 with the GAL1 promoter was used as an expression vector. The TaTPC1 coding region was amplified from the original plasmids with two primers 5′-GCAGGATCCAGAGAGATGAGCGAAGCGAG-3′ (sense primer) and 5′-AGTGAATTCTCAAGAGTTTTGAGATCCATC-3′ (antisense primer). The amplified product was digested with KpnI plus XhoI and ligated to KpnI/XhoI-digested pYES2 to construct the expression vector pYES2-TaTPC1. The plasmid pYES2-TaTPC1 or pYES2 vector were transformed into wild-type or mutant yeast cells, respectively, generated as in the section above, using the lithium acetate method. The transformant with pYES2 vector was used as a control. The transformant was cultured in SD liquid medium lacking Ura (SD–Ura medium) until OD600=1.2, and then diluted to an OD600 value of 0.01 or 0.001. Five microlitres of each dilution was dripped on basic YPGAL (1% yeast/2% peptone/2% galactose; Difco) medium or YPGAL medium containing 5 mM or 10 mM lithium chloride (LiCl), and cultured at 30 °C for 3 d. The growth status of the yeast cells was observed. Arabidopsis transformation and stomatal aperture analysis A BamH1/Kpn1 fragment encoding the full-length TaTPC1 was inserted downstream of the 35S promotor in the plant expression vector pBin438, and introduced into Arabidopsis plants by the vacuum infiltration technique. Independent homozygous transgenic lines were obtained after selection of T3 progeny on MS medium containing 50 mg l−1 kanamycin. Ten-day-old seedlings of the transgenic progenies and wild-type plants are cultured on MS medium and used for molecular analysis. Twenty-day-old seedlings of these plants grown in pots containing vermiculite were used to observe stomatal aperture. The lower epidermis was ripped off different leaves and floated in a solution consisting of 50 μM KCl, 10 mM MES-TRIS, and 100 μM, 1 mM, or 10 mM Ca2+, respectively. The peels were incubated under light at 25 °C for 1.5 h. Then the stomatal apertures were measured under an optical microscope. Results Cloning of the TaTPC1 gene from wheat The wheat cDNA library was screened for stress-inducible homologues of the CBF/DREB gene by the yeast one-hybrid method. Whereas one sequence turned out to be the homologous gene of CBF/DREB (Shen et al., 2003), another sequence from a weak positive colony exhibited similarity to voltage-gated Ca2+-permeable two-pore channel genes. Because Ca2+ channels were also involved in stress responses (White and Broadly, 2003), the current putative Ca2+ channel gene, named TaTPC1 (Triticum asetivumtwo-pore channel 1), was investigated further. Since the inserted sequence still represents the 3′ part of the TaTPC1 gene, 5′-RACE was carried out and a fragment of 1351 bp was obtained, which overlapped with the sequence of TaTPC1. The combined sequence represented a full-length cDNA of TaTPC1. The TaTPC1 was 2841 bp in length with an open reading frame of 2229 bp flanked by the 117 bp 5′-untranslated region (UTR) and 495 bp 3′-UTR plus a poly (A) tail (Fig. 1A). The open reading frame encoded a putative protein of 743 amino acids with a predicated molecular mass of 85.6 kDa. By analysis via the SMART program (Letunic et al., 2002) and comparison with the amino acid sequence from other Ca2+-permeable channel proteins, several conserved domains were identified. As shown in Fig. 1B, the TaTPC1 has two homologous domains (nos 1 and 2) similar to other TPCs (Furuichi et al., 2001; Hashimoto et al., 2004; Kurusu et al., 2004). Both domains have six transmembrane segments, S1 to S6 (Fig. 1A, B). There is a pore loop (P) between S5 and S6 in each domain. S4 contains charged residues and may function as a voltage sensor. S1, S2, S3, S5, and S6 segments of each domain have distinct hydrophobic indices while S4 and pore loop regions have smaller indices. Both N- and C-termini are hydrophilic. The connecting segment between domains II and I is two Ca2+-binding EF-hand motifs, which are also hydrophilic and may regulate its activity. The overall amino acid sequence shared 52%, 77%, and 85% identity with AtTPC1 (AB053952, Arabidopsis; Furuichi et al., 2001), OsTPC1 (AB100696, Oryza sativa; Hashimoto et al., 2004), and HvTPC1 (AY465119, Hordeum vulgare), respectively (Fig. 2). These results indicated that TaTPC1 was more closely related to the two-pore voltage-gated channel for Ca2+ from monocots. Fig. 1. View large Download slide View large Download slide Wheat TaTPC1 gene and its deduced amino acid sequence. (A) Nucleotide and deduced amino acid sequence of TaTPC1 (AY114121). (B) Kyte and Doolittle hydropathy plot. The six hydrophobic segments (S1–S6) are underlined with the solid line in (A) and indicated in (B). The pore loop segments (P) in each domain are underlined with the dashed line in (A). Two EF hands are also underlined with a dotted line in (A) and indicated with a line in (B). Fig. 1. View large Download slide View large Download slide Wheat TaTPC1 gene and its deduced amino acid sequence. (A) Nucleotide and deduced amino acid sequence of TaTPC1 (AY114121). (B) Kyte and Doolittle hydropathy plot. The six hydrophobic segments (S1–S6) are underlined with the solid line in (A) and indicated in (B). The pore loop segments (P) in each domain are underlined with the dashed line in (A). Two EF hands are also underlined with a dotted line in (A) and indicated with a line in (B). Fig. 2. View largeDownload slide Comparison of the amino acid sequence of the TaTPC1 with other homologous proteins. AtTPC1 (AB053952, Arabidopsis), OsTPC1 (AB100696, Oryza. Satia), and HvTPC1 (AY465119, Hordeum vulgare) are compared with the present wheat TaTPC1. The residues shaded in black indicate conserved amino acids. Dashes were included for maximum alignment. The consensus sequence was also presented. Numbers on the left indicate positions of the amino acid residues. Fig. 2. View largeDownload slide Comparison of the amino acid sequence of the TaTPC1 with other homologous proteins. AtTPC1 (AB053952, Arabidopsis), OsTPC1 (AB100696, Oryza. Satia), and HvTPC1 (AY465119, Hordeum vulgare) are compared with the present wheat TaTPC1. The residues shaded in black indicate conserved amino acids. Dashes were included for maximum alignment. The consensus sequence was also presented. Numbers on the left indicate positions of the amino acid residues. Expression of the TaTPC1 gene under different abiotic stresses The wheat seedlings were subjected to various treatments and expression of the TaTPC1 gene was examined. As shown in Fig. 3, an increase of TaTPC1 mRNA was detected in wheat seedling roots after treatment with high salinity, PEG, low temperature (4 °C), and ABA. However, the expression pattern was different. Under ABA treatment, the expression of TaTPC1 was induced early when compared with the patterns under NaCl and PEG treatments. When treated with low temperature, the expression of TaTPC1 was not detected until 24 h after the initiation of the treatment. TaTPC1 expression was not detected in shoots of the wheat seedlings under these treatments. These results indicate that TaTPC1 may play specific roles in roots in response to various treatments. Fig. 3. View largeDownload slide Differential expression of TaTPC1 in wheat seedlings in response to abiotic stresses. The wheat seedlings were subjected to NaCl, ABA, PEG, and low temperature (4 °C) treatments and total RNA from roots or shoots was isolated in the indicated times. Each lane was loaded with 30 μg of total RNA. The RNA blot was hybridized with the labelled full-length TaTPC1 gene. The RNA blots were also hybridized with the 18S rDNA probe to examine the RNA loading and only the result for the NaCl-treated samples was presented for simplicity. Similar equal loading was also observed for samples from other treatments. Fig. 3. View largeDownload slide Differential expression of TaTPC1 in wheat seedlings in response to abiotic stresses. The wheat seedlings were subjected to NaCl, ABA, PEG, and low temperature (4 °C) treatments and total RNA from roots or shoots was isolated in the indicated times. Each lane was loaded with 30 μg of total RNA. The RNA blot was hybridized with the labelled full-length TaTPC1 gene. The RNA blots were also hybridized with the 18S rDNA probe to examine the RNA loading and only the result for the NaCl-treated samples was presented for simplicity. Similar equal loading was also observed for samples from other treatments. Functional complementation of the yeast mutant cells by TaTPC1 To analyse the function of TaTPC1, a complementation strategy was adopted. The CCH1 gene of the yeast strain (W303.1A) (Fisher et al., 1997), which shares high homology with the α1-subunit of a voltage-gated Ca2+ channel from mammalian cells (Paidhungat and Garrett, 1997), was knocked out through the insertion of a Trp marker gene by homologous recombination (Wang et al., 2005). The yeast cch1 mutant strain was thus obtained by its survival on SD–Trp medium. The wild-type yeast cells cannot grow on the same medium. The CCH1 gene is involved in Ca2+ uptake and is instrumental in the response of a wild-type strain to ion stress (Fisher et al., 1997; Paidhungat and Garrett, 1997). Disruption of the CCH1 gene resulted in the blocking of the Ca2+ uptake in the cch1 mutant and then led to the sensitivity of the mutant to LiCl (Paidhungat and Garrett, 1997). Taking advantage of this feature of the cch1 mutant, the present TaTPC1 gene can be transformed into the mutant to test if the gene can fulfil a similar function as the CCH1 did. The open reading frame of TaTPC1 was thus ligated to the expression vector pYES2, and the recombinant plasmid pYES2-TaTPC1 was transformed into the mutant cells. The pYES2 vector itself was also transformed into the wild-type and the mutant cells as control experiments. To compare the growth status of these transformants, a series of overnight cultures grown in the SD–Ura liquid medium were diluted and dropped onto YPGal or YPGal plus 5 or 10 mM LiCl medium. On normal YPGal medium, all of the transformants showed normal growth (Fig. 4A). On 5 mM or 10 mM LiCl medium, the mutant cch1 cells harbouring the pYES2-TaTPC1 can grow normally, similar to wild-type yeast cells harbouring the pYES2 vector (Fig. 4B, C). On the contrary, the cch1 mutant with the pYES2 vector cannot grow on the same medium. These results indicate that the TaTPC1 protein can complement the function of Ca2+ channel CCH1 and may then regulate the level of [Ca2+](cyt) to maintain the normal growth of yeast mutants under lithium stress. Fig. 4. View largeDownload slide Functional complementation of the yeast mutant with TaTPC1 under LiCl stress. The fusion plasmids pYES2-TaTPC1 or pYES2 vector was transformed into wild-type (WT) or cch1 mutant (Mutant) yeast cells by using a lithium acetate method. The transformants were grown in SD–Ura liquid medium overnight until OD600=1.2, and then the cultures were diluted into 0.01 and 0.001. Five microlitres of each dilution was dripped on YPGAL medium containing 5 mM or 10 mM LiCl, respectively, and on the basic YPGAL medium as control. The plates were incubated at 30 °C for 3 d, and the growth status of yeast was observed. Fig. 4. View largeDownload slide Functional complementation of the yeast mutant with TaTPC1 under LiCl stress. The fusion plasmids pYES2-TaTPC1 or pYES2 vector was transformed into wild-type (WT) or cch1 mutant (Mutant) yeast cells by using a lithium acetate method. The transformants were grown in SD–Ura liquid medium overnight until OD600=1.2, and then the cultures were diluted into 0.01 and 0.001. Five microlitres of each dilution was dripped on YPGAL medium containing 5 mM or 10 mM LiCl, respectively, and on the basic YPGAL medium as control. The plates were incubated at 30 °C for 3 d, and the growth status of yeast was observed. Transgenic Arabidopsis plants overexpressing the TaTPC1 gene To analyse the function of TaTPC1 in the plant, the TaTPC1 gene was cloned into binary vector pBin438 under the control of the cauliflower mosaic virus (CaMV) 35S promoter (Fig. 5A), and then transformed into Arabidopsis plants using the vacuum infiltration method. Homozygous T3 lines were obtained for four lines and the integration of TaTPC1 into the Arabidopsis genome was confirmed by Southern blot (data not shown). The TaTPC1 mRNA level in transgenic plants was examined and it was found that all the four independent lines showed significantly increased mRNA levels compared with the wild-type plants (Fig. 5B). The phenotype of the transgenic plants was also examined and no significant change was found when it was compared with wild-type plants under normal or stress (salt, drought, or ABA) conditions (data not shown). Because TaTPC1 is a putative Ca2+ channel protein, it may thus facilitate the entry of Ca2+ into the plant cells in the transgenic plants overexpressing the TaTPC1 gene. However, no significant alterations in the Ca2+ contents were detected in the aerial part of the transgenic plants in comparison with the wild-type plants under both normal conditions and CaCl2 treatment (data not shown). Although no significant change in overall Ca2+ content was observed in the transgenic plants, transient variations may still exist in various tissues or cells. Fig. 5. View largeDownload slide Expression of TaTPC1 in transgenic Arabidopsis. (A) Construction of plant expression vector of pBin438-TaTPC1. (B) Northern analysis of TaTPC1 expression in transgenic plants. Nos 1–4 indicate four independent transgenic lines. CK indicates the wild-type plant. Fig. 5. View largeDownload slide Expression of TaTPC1 in transgenic Arabidopsis. (A) Construction of plant expression vector of pBin438-TaTPC1. (B) Northern analysis of TaTPC1 expression in transgenic plants. Nos 1–4 indicate four independent transgenic lines. CK indicates the wild-type plant. Stomatal aperture analysis of the TaTPC1 transgenic plants Calcium channel plays an important role in regulating the Ca2+ influx, and the transient change of [Ca2+] was involved in stomatal movement (McAinsh et al., 1990). To examine whether overexpression of TaTPC1 in the transgenic line affected stomatal movement under different concentrations of CaCl2, the stomatal aperture of wild-type and TaTPC1-overexpressing plants was analysed in vitro using isolated epidermal peels. Under a concentration of 100 μM Ca2+ that should promote stomatal opening, the distribution of open stomata in transgenic lines 3 and 4 peaked at 1.6 μM, while the control peaked at 1.9 μM (Fig. 6A), indicating that the apertures of the stomata from the transgenic lines are generally smaller than those from the wild-type plants. Under a concentration of 1 mM Ca2+, transgenic lines 3 and 4 and the control plants all showed a peak at 1.3 μM for stomatal aperture (Fig. 6B). When the plants were treated further with 10 mM CaCl2, the distribution peak of stomatal aperture for transgenic lines 3 and 4 moved further toward 1.0 μM, whereas the distribution peak for the control plants was still localized at 1.3 μM (Fig. 6C). These results indicate that overexpression of TaTPC1 most likely promoted the closing of stomata in the presence of Ca2+. Fig. 6. View largeDownload slide Stomatal aperture of wild-type and TaTPC1 transgenic Arabidopsis plants under different Ca2+. Stomatal apertures were measured on epidermal peels of the wild-type (open columns), transgenic line 3 (black columns), and transgenic line 4 (hatched columns). (A) Epidermal peels were incubated in 100 μM Ca2+ for 1.5 h in the light. (B) Epidermal peels were incubated in 1 mM Ca2+ for 1.5 h in the light. (C) Epidermal peels were incubated in 10 mM Ca2+ for 1.5 h in the light. Fig. 6. View largeDownload slide Stomatal aperture of wild-type and TaTPC1 transgenic Arabidopsis plants under different Ca2+. Stomatal apertures were measured on epidermal peels of the wild-type (open columns), transgenic line 3 (black columns), and transgenic line 4 (hatched columns). (A) Epidermal peels were incubated in 100 μM Ca2+ for 1.5 h in the light. (B) Epidermal peels were incubated in 1 mM Ca2+ for 1.5 h in the light. (C) Epidermal peels were incubated in 10 mM Ca2+ for 1.5 h in the light. Discussion Under normal growth condition, the [Ca2+](cyt) in plant cells is maintained at about a level of 100 nM through the activity of Ca2+-ATPase, Ca2+-permeable channel, and Ca2+/H+ antiporters in cell membranes (White and Broadley, 2003). Under abiotic stress conditions, [Ca2+](cyt) was enhanced. An explanation for this increase could be membrane depolarization under abiotic stress treatments, which leads to the opening of Ca2+-permeable channels (White and Broadley, 2003). A second reason could be that expression of genes encoding Ca2+-permeable channel were induced under abiotic stress. This latter case has been proved from the increase of [Ca2+](cyt) after overexpression of AtTPC1 (Furuichi et al., 2001). In the present study, a calcium channel gene TaTPC1 was cloned and characterized from wheat. The expression of TaTPC1 was induced under high salinity, PEG, low temperature (4 °C), and ABA, implying that TaTPC1 is involved in plant stress responses. The spatial and temporal distribution of Ca2+ may be changed compared with that under normal conditions. However, the level of [Ca2+](cyt) is strictly controlled because excessive Ca2+ may cause cell death. Excessive Ca2+ may be effluxed to extracellar space or stored in the vacuoles and endoplasmic reticulum, etc. through a Ca-ATPase (Chung et al., 2000) and a Ca2+/H+ antiporter (Ueoka-Nakanishi et al., 2000), whose expression was also induced by abiotic stress treatments. It should be noted that TaTPC1 was inducible in the root but not the shoot of wheat seedlings, implying that TaTPC1 may play a major role in the root by responding to stimuli from the soil. By contrast to the induction of TaTPC1 by various stresses, other TPC1 genes such as AtTPC1 from Arabidopsis and OsTPC1 from rice appeared to be ubiquitously expressed in the whole plant (Furuichi et al., 2001; Kurusu et al., 2004). The different expression patterns may result from different homologues of the TPC1 genes. In tobacco, two homologues, NtTPC1A and NtTPC1B, have been identified (Kadota et al., 2004). In wheat, two expressed sequence tags (BQ904560 and CA617077) were also identified (data not shown), which showed homology to the present TaTPC1. The two expressed sequence tags exhibited similarity to different parts of the TaTPC1 gene and are not overlapping (data not shown). This analysis indicates that the wheat genome may contain other homologous TPC1 genes. Although the TaTPC1 gene can be induced by various stresses and ABA, the TaTPC1-overexpressing plants did not show significant alterations in sensitivity to these treatments (data not shown). This observation is different from that obtained in the AtTPC1-overexpressing Arabidopsis. Peiter et al. (2005) reported that AtTPC1 overexpression or knockout in Arabidopsis regulated seed germination in response to ABA. The discrepancy may be due to the fact that the present wheat TaTPC1 gene is expressed in a heterologous Arabidopsis system. Alternatively, the wheat TaTPC1 gene may not be a real orthologue of the Arabidopsis AtTPC1 gene since their proteins shared only about 50% identity. To date, most of the results on the localization of the Ca2+-permeable channel were based on electrophysiological and biochemical methods (Sanders et al., 2002). The TaTPC1-GFP fusion protein was used to study the localization of the TaTPC1 protein in onion epidermal cells and it was found that TaTPC1 appeared to be localized in the plasma membrane (data not shown). This localization is in contrast with the result obtained by Peiter et al. (2005), who reported that the Arabidopsis AtTPC1 protein was localized in the vacuolar membrane. The reason for this difference is not known. It is possible that the discrepancy may result from the different expression systems used. Alternatively, the present TaTPC1 may represent a homologue of TPC1 in wheat, as discussed above, and different homologues may have various functions and membrane locations. Other biochemical methods should be used to demonstrate the precise location of the TaTPC1. Yeast has been proposed as a model system to study the function of plant genes in response to stress (Xiong et al., 2002). A change in [Ca2+](cyt) has been observed in yeast as well as in plants under abiotic stresses. Therefore both organisms share some of the components in the stress signal transduction pathway (Xiong et al., 2002). It has been reported that a yeast strain lacking CCH1, a gene encoding a polypeptide of 2039 residues and sharing high homology with the α1-subunit of a voltage-gated l-type Ca2+ channel from mammalian cells, exhibits a biochemical defect in calcium uptake and consequently grows slower (Paidhungat and Garrett, 1997). This delay in growth of the cch1 mutant could be explained by the necessity for extracelluar calcium as a mediator in responding to high salt concentrations (Matheos et al., 1997). In yeast, the same transport system is mediating the uptake of sodium and lithium across the plasma membrane (Borst-Pauwels, 1991). Lithium chloride used with millimolar concentrations showed the same effect as sodium chloride in molar concentrations. Since salt stress implies ionic as well as osmotic stress conditions, the osmotic stress conferred by lithium is lower because of its effectiveness at lower concentrations (Nakamura et al., 1993). In order to confer the salt stress condition, LiCl was used as an analogue of sodium chloride in this study. The sensitivity of the cch1 mutant to a high-salt condition (Paidhungat and Garrett, 1997) provided the basis for establishing a yeast system to test the function of TaTPC1 in response to salt stress in relation to calcium. Therefore, the cch1 mutant was generated in the present study and used to identify the function of the present TaTPC1 gene from wheat. The TaTPC1 protein can recover the growth of the mutant cch1 yeast cells under LiCl stress, indicating that the TaTPC1 gene has a similar function to CCH1. The Ca2+ channel gene AtTPC1 from Arabidopsis also rescues the growth rate and the Ca2+ uptake activity of the yeast cch1 mutation (Furuichi et al., 2001). The direct evidence for the Ca2+ uptake activity of the present wheat TaTPC1 protein should be investigated further. TaTPC1 transgenic plants have a higher proportion of stomata with an aperture smaller than that of the control plants. This fact suggests that the guard cells of the TaTPC1 transgenic plants may take up more Ca2+ via the TaTPC1 channel from the apoplast or the surrounding cells and make the stomata more closed. This result is consistent with the report that a rise in Ca2+ concentration in the guard cell is useful for closing stomatal apertures (Hamilton et al., 2000). When the Ca2+ concentration was increased to 1 mM, the stomatal aperture was reduced to 81% in transgenic plants and 68% in control plants. However, when the Ca2+ concentration was further increased to 10 mM, the stomatal aperture of the transgenic plants was further reduced to 62% of the original apertures, whereas the control plants maintained a reduction of 68%. This fact probably indicates that the Ca2+ effect on stomatal closing is non-linear in the transgenic and the control plants, and the guard cells of the transgenic plants are relatively less sensitive to the Ca2+ than those of the wild-type plants. The reason for this difference in sensitivity is not known. It is possible that the transgenic plants already had more TaTPC1 proteins and adsorbed the Ca2+ in guard cells, thus making the stomatal closing less sensitive at the 1 mM Ca2+ concentration. The regulation of the stomatal movement by the present TaTPC1 is consistent with the observations made by Peiter et al. (2005). They reported that an Arabidopsis tpc1 knockout mutant was defective in the response of stomata to extracellular calcium. Although TaTPC1 is responsible for Ca2+ permeation, significantly more Ca2+ was not found to have accumulated in the transgenic plants than in the control plants when using the atomic absorbance spectrometer method (data not shown). This phenomenon may result from the fact that whole aerial parts of the plants were used for the Ca2+ measurement. The uneven distribution of Ca2+ in different cells, tissues, or organs may thus be averaged in the TaTPC1 transgenic plants. Alternatively, this phenomenon may be explained by the fact that, as a second messenger, Ca2+ can only have a transient increase that may lead to the initiation of a signal transduction. Duration of this increase for a long time would be harmful for plant survival and thus must be dismissed. Therefore, a more sensitive approach to Ca2+ probing should be used to reveal the transient change in Ca2+ concentration in transgenic plants (Furuichi et al., 2001). Further research should disclose more about the function of TaTPC1 in plants. * Present address: College of Life Sciences, Shanxi Normal University, Xian 710062, China. This work was supported by the National Key Basic Research Project (2002CB111303, 2006CB100102), Chinese Academy of Sciences (KSCXZ-SW-327) and the National Science Foundation of China (30370130). References Amtmann A, Fischer M, Marsh EL, Stefanovic A, Sanders D, Schachtman DP. 2001. The wheat cDNA LCT1 generates hypersensitivity to sodium in a salt-sensitive yeast strain. Plant Physiology  126, 1061–1071. Google Scholar Borst-Pauwels GW. 1991. Ion transport in yeast. Biochemica et Biophysica Acta  650, 88–127. Google Scholar Cheong YH, Kim KN, Pandey GK, Gupta R, Grant JJ, Luan S. 2003. CBL1, a calcium sensor that differentially regulates salt, drought, and cold responses in Arabidopsis. The Plant Cell  15, 1833–1845. Google Scholar Chung WS, Lee SH, Kim JC, et al. 2000. Identification of a calmodulin-regulated soybean Ca2+-ATPase (SCA1) that is located in the plasma membrane. The Plant Cell  12, 1393–1407. Google Scholar Demidchik V, Davenport RJ, Tester M. 2002. Nonselective cation channels in plants. Annual Review of Plant Biology  53, 67–107. Google Scholar Fischer M, Schnell N, Chattaway J, Davies P, Dixon G, Sanders D. 1997. The Saccharomyces cerevisiae CCH1 gene is involved in calcium influx and mating. FEBS Letters  419, 259–262. Google Scholar Furuichi T, Cunningham KW, Muto S. 2001. A putative two pore channel AtTPC1 mediates Ca2+ flux in Arabidopsis leaf cells. Plant and Cell Physiology  42, 900–905. Google Scholar Hamilton DWA, Hills A, Kohlwer B, Blatt MR. 2000. Ca2+ channels at the plasma membrane of stomatal guard cells reactivated by hyperpolarization and abscisic acid. Proceedings of the National Academy of Sciences, USA  97, 4967–4972. Google Scholar Harmon AC, Gribskov M, Gubrium E, Harper JF. 2001. The CDPK superfamily of protein kinase. New Phytologist  151, 175–183. Google Scholar Hashimoto K, Saito M, Matsuoka H, Iida K, Iida H. 2004. Functional analysis of a rice putative voltage-dependent Ca2+ channel, OsTPC1, expressed in yeast cells lacking its homologous gene CCH1. Plant and Cell Physiology  45, 496–500. Google Scholar Kadota Y, Furuichi T, Ogasawara Y, Goh T, Higashi K, Muto S, Kuchitsu K. 2004. Identification of putative voltage-dependent Ca2+-permeable channels involved in cryptogein-induced Ca2+ transients and defense responses in tobacco BY-2 cells. Biochemical and Biophysical Research Communications  317, 823–830. Google Scholar Kurusu T, Sakurai Y, Miyao A, Hirochika H, Kuchitsu K. 2004. Identification of a putative voltage-gated Ca2+-permeable channel (OsTPC1) involved in Ca2+ influx and regulation of growth and development in rice. Plant and Cell Physiology  45, 693–702. Google Scholar Letunic I, Goodstadt L, Dickens NJ, Doerks T, Schultz J, Mott R, Ciccarelli F, Copley RR, Ponting CP, Bork P. 2002. Recent improvements to the SMART domain-based sequence annotation resource. Nucleic Acids Research  30, 242–244. Google Scholar Liu Q, Kasuga M, Sakuma Y, Abe H, Miura S, Yamaguchi-Shinozaki K, Shinozaki K. 1998. Two transcription factors, DREB1 and DREB2, with an EREBP/AP2 DNA binding domain separate two cellular signal transduction pathways in drought- and low-temperature-responsive gene expression, respectively, in Arabidopsis. The Plant Cell  10, 1391–1406. Google Scholar Matheos DP, Kingsbury TJ, Ahsan US, Cunningham KW. 1997. Tcn1p/Crz1p, a calcineurin-dependent transcription factor that differentially regulates gene expression in Saccharomyces cerevisiae. Genes and Development  11, 3445–3458. Google Scholar McAinsh MR, Brownlee C, Hetherington AM. 1990. Abscisic acid-induced elevation of guard cell cytoplasmic Ca2+ precedes stomatal closure. Nature  343, 186–188. Google Scholar Nakamura T, Liu Y, Hirata D, Namba H, Harada S, Hirokawa T, Miyakawa T. 1993. Protein phosphatase type 2B (calcineurin)-mediated, FK506-sensitive regulation of intracellular ions in yeast is an important determinant for adaptation to high salt stress conditions. EMBO Journal  12, 4063–4071. Google Scholar Paidhungat M, Garrett S. 1997. A homolog of mammalian, voltage-gated calcium channels mediates yeast pheromone-stimulated Ca2+ uptake and exacerbates the cdc1(Ts) growth defect. Molecular and Cellular Biology  17, 6339–6347. Google Scholar Peiter E, Maathuis FJM, Mills LN, Knight H, Pelloux J, Hetherington AM, Sanders D. 2005. The vacuolar Ca2+-activated channel TPC1 regulates germination and stomatal movement. Nature  434, 404–408. Google Scholar Sanders D, Brownlee C, Harper JF. 1999. Communicating with calcium. The Plant Cell  11, 691–706. Google Scholar Sanders D, Pelloux J, Brownlee C, Harper JF. 2002. Calcium at the crossroads of signaling. The Plant Cell  14 (Supplement), S401–S417. Google Scholar Sheen J. 1996. Ca2+-dependent protein kinases and stress signal transduction in plants. Science  274, 1900–1902. Google Scholar Shen YG, Zhang WK, He SJ, Zhang JS, Liu Q, Chen SY. 2003. An EREBP/AP2-type protein in Triticum aestivum was a DRE-binding transcription factor induced by cold, dehydration and ABA stress. Theoretical and Applied Genetics  106, 923–930. Google Scholar Ueoka-Nakanishi H, Tsuchiya T, Sasaki M, Nakanishi Y. 2000. Functional expression of mung bean Ca2+/H+ antiporter in yeast and its intracellular localization in the hypocotyls and tobacco cells. European Journal of Biochemistry  267, 3090–3098. Google Scholar Wang YJ, Hao YJ, Zhang ZG, Chen T, Zhang JS, Chen SY. 2005. Isolation of trehalose-6-phosphate phosphatase gene from tobacco and its functional analysis in yeast cells. Journal of Plant Physiology  162, 215–223. Google Scholar White PJ. 2000. Calcium channels in higher plants. Biochimica et Biophysica Acta  1465, 171–189. Google Scholar White PJ, Bowen HC, Demidchik V, Nichols C, Davis JM. 2002. Genes for calcium-permeable channels in the plasma membrane of plant root cells. Biochemica et Biophysica Acta  1564, 299–309. Google Scholar White PJ, Broadley M. 2003. Calcium in plants. Annals of Botany  92, 487–511. Google Scholar Xiong L, Karen SS, Zhu JK. 2002. Cell signaling during cold, drought and salt stress. The Plant Cell  14 (Supplement), S165–S183. Google Scholar Zhang JS, Zhou JM, Zhang C, Chen SY. 1996. Differential gene expression in a salt-tolerant rice mutant and its parental variety. Science in China  (Series C) 39, 310–319. Google Scholar © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]

Giuliani, Silvia;Sanguineti, Maria Corinna;Tuberosa, Roberto;Bellotti, Massimo;Salvi, Silvio;Landi, Pierangelo

doi: 10.1093/jxb/eri303pmid: 16246858

Abstract Near-isogenic hybrids (NIHs), developed from crossing maize (Zea mays L.) backcross-derived lines (BDLs) differing for the parental alleles at a major QTL for leaf ABA concentration (L-ABA), were field-tested for 2 years under well-watered and water-stressed conditions. Differences among NIHs for L-ABA and other morpho-physiological traits were not affected by water regimes. On average, the QTL allele for high L-ABA markedly reduced stomatal conductance and root lodging. To elucidate the effects of the QTL on root architecture and L-ABA, root traits of two pairs of BDLs were measured in plants grown in soil columns at three water regimes. Differences among BDLs were not affected by water regimes. Across water regimes, the QTL confirmed its effect on L-ABA and showed a concurrent effect on root angle, branching, number, diameter, and dry weight. Based on these results, it is concluded that the QTL affects root lodging through a constitutive effect on root architecture. In addition, there is speculation that the QTL effects on root traits and L-ABA are probably due to pleiotropy rather than linkage and a model is proposed in which the QTL has a direct effect on root architecture, while indirectly affecting L-ABA. Abscisic acid, root architecture, root lodging, stomatal conductance, water stress, Zea mays Introduction The application of QTL (Quantitative Trait Locus) analysis and other genomics approaches provide unprecedented opportunities with which to identify the chromosome regions controlling variation in the adaptive response to water stress and, eventually, to clone the gene/s responsible for such variation. The positional cloning of a QTL takes advantage from the availability of NILs (Near Isogenic Lines) for the target QTL (Paran and Zamir, 2003). In addition, the availability of NILs for a specific QTL allows an in-depth characterization of the QTL effects on a number of traits which, in turn, facilitates the elaboration of models and hypotheses on their cause–effect relationships (Tuberosa et al., 2002). Among the plethora of quantitative traits affecting drought tolerance, particular attention has been devoted to the concentration of abscisic acid (ABA), in view of its pivotal role in regulating other molecular and morpho-physiological processes involved in the adaptive responses of crops to an insufficient water supply (Quarrie, 1991; Tuberosa et al., 1994; Sanguineti et al., 1996; Sharp et al., 2004), particularly during the reproductive stage (Saini and Westgate, 2000; Wang et al., 2002; Boyer and Westgate, 2004). Previous studies conducted in maize (Zea mays L.) by Lebreton et al. (1995) and by Tuberosa et al. (1998) indicated that the concentration of ABA in the leaf (L-ABA) is a complex trait controlled by several QTLs. In particular, Tuberosa et al. (1998) detected 16 QTLs for L-ABA by analysing 80 F3:4 families derived by the cross between inbred lines Os420 (high L-ABA parent) and IABO78 (low L-ABA). The most important QTL was identified by the RFLP marker csu133 on chromosome 2 (bin 2.04) and accounted for 32% of the phenotypic variation for the trait. The importance of this QTL was confirmed in a subsequent work of divergent selection for L-ABA conducted in the F2 population derived from the same cross (i.e. Os420×IABO78). In fact, the allele at the RFLP locus csu133 provided by the low L-ABA parent was fixed in the eight F3:4 lines selected for low L-ABA, while it was present in only one of the eight F3:4 lines selected for high L-ABA (Landi et al., 2005). To gain more thorough information on the effects of the QTL in question on L-ABA and other traits, NILs were developed at this QTL following a marker-assisted backcross procedure. Backcross-derived lines (BDLs) were obtained for both parental inbreds Os420 and IABO78. These BDLs were then tested in the field under both water-stressed and well-watered regimes, which allowed the effect of the QTL on L-ABA to be validated (Landi et al., 2005). For a cross-pollinated species like maize, the agronomic performance of inbred lines per se is highly affected by inbreeding depression, especially for those traits (like grain yield) largely influenced by non-additive gene actions. Therefore, to achieve a more accurate evaluation of the effects of a particular QTL on heterotic traits the investigation should, preferably, be carried out using near-isogenic hybrids (NIHs), which can be obtained by crossing BDLs at the same target region introgressed in different genetic backgrounds. According to the BDLs used as parents, NIHs are either homozygous or heterozygous at the target QTL region, while being heterozygous for most of the remaining portion of the genome. In this case, the availability of BDLs for the QTL near csu133 in the Os420 and IABO78 backgrounds made it possible to obtain NIHs in order to evaluate the effects of the QTL on heterotic traits. The study presented here was conducted using two water regimes on the NIHs among such BDLs in order (i) to obtain an evaluation unbiased by inbreeding depression of the effects of the QTL on L-ABA and other physiological and agronomic traits, and (ii) to assess whether these effects are influenced by the level of water stress experienced by the plants grown in the field. Two pairs of BDLs were also tested in the greenhouse using three water regimes, in order (iii) to gather further information on the QTL effects in plants grown under controlled conditions, and (iv) to evaluate the QTL effects on root characteristics. This latter objective was suggested by a preliminary observation (Landi et al., 2005) that a pair of NIHs differed markedly for root lodging resistance. Materials and methods Plant materials The development of the BDLs for the target QTL close to csu133 is summarized here; a more detailed description has been provided by Landi et al. (2005). Two parallel backcross programmes were developed by the cross Os420×IABO78. In one programme, Os420 was the recurrent parent and IABO78 the donor of the decreasing allele (−) for L-ABA at the target QTL; in the other programme, IABO78 was the recurrent parent and Os420 the donor of the increasing allele (+) for L-ABA. Five backcross cycles were followed by two selfing cycles, so that the BDLs finally developed were BC5F3 lines. Throughout the two parallel backcross programmes, two molecular markers flanking the target QTL were used to identify the desired genotypes. Starting from BC2, two families were grown for each backcross programme; each family was developed by one ear harvested in BC1 and was identified by the suffix .1 and .2. Consequently, four couples of BDLs were finally produced, i.e. Os420.1 (+/+) and (−/−), Os420.2 (+/+) and (−/−), IABO78.1 (+/+) and (−/−), and IABO78.2 (+/+) and (−/−). The four Os420 BDLs were factorially crossed to the four IABO78 BDLs, thus producing 16 near-isogenic hybrids (NIHs). At the target QTL, four of these NIHs were homozygous (+/+), four were homozygous (−/−), and eight were heterozygous (+/−). Field evaluation of NIHs The 16 NIHs were tested at Cadriano (44° 33′ N lat., E long., Po Valley, Northern Italy) for two years (2002 and 2003) and with two water regimes. In each year, two trials were conducted in the same field; they were separated by a 4-m-wide alley and by three border rows on each side. The two trials differed only for the irrigation volumes, which corresponded either to 120% or to 40% of the estimated evapotranspiration after accounting for rainfall. The crop evapotranspiration was calculated by following the procedure described in detail by Landi et al. (1995). For each trial, the field design was a randomized complete block with four replicates; plots were 3.06 m long single rows separated by 0.80 m from adjacent rows. Trials were sown on 29 April 2002 and 28 April 2003; c. 40 d later plots were thinned leaving 11 plants per plot (corresponding to 4.5 plants m−2). Plants were grown following the field practices locally adopted. In particular, fertilizer rates were 200 kg ha−1 of N (half applied before sowing and half after thinning) and 44 kg ha−1 of P (all before sowing); K was not applied because of its high availability in the soil. Weeds were controlled mechanically and by hand. Irrigation was applied from the mid-end of stem elongation to the end of silking (i.e. from V14-15 to R1 stage, according to Ritchie et al., 1997). Total irrigation volumes corresponding to 120% of evapotranspiration were 54 mm and 90 mm in 2002 and 2003, respectively, while the irrigation volumes corresponding to 40% of evapotranspiration were 18 mm and 30 mm, respectively. Trials were hand-harvested on 3 September 2002 and 3 September 2003; then, ears were dried down in an aired storage room and shelled when a constant moisture was attained. The following traits were measured: (i) L-ABA, on leaf samples (third leaf from the top) collected at the mid-end of pollen shedding (i.e. between VT and R1 stages). Leaf samples were collected from 09.00 h to 10.00 h, immediately frozen and stored at −20 °C until ABA analysis was carried out. L-ABA was measured on duplicate samples per plot using an ABA-specific monoclonal antibody, as described in Tuberosa et al. (1998); (ii) leaf relative water content (RWC), measured on the same leaf sample used for L-ABA and following the procedure described by Sanguineti et al. (1999). RWC was computed as (fresh weight−dry weight)/(turgid weight−dry weight)×100; (iii) stomatal conductance (SC), analysed on the second leaf from the top at VT-R1 stage. Leaves were analysed from 09.00 h to 11.00 h; a steady-state porometer (model LI-1600 Li-Cor) was utilized following the procedure described by Sanguineti et al. (1999); (iv) leaf water potential (WP) measured at the same time and on the tip of the same leaf blades used for measuring SC; a pressure chamber was used following the procedure described in Landi et al. (2001). SC and WP were investigated only on two pairs of NIHs homozygous at the target QTL, i.e. Os420.1 (+/+)×IABO78.2 (+/+) and the counterpart Os420.1 (−/−)×IABO78.2 (−/−); Os420.2 (+/+)×IABO78.2 (+/+) and the counterpart Os420.2 (−/−)×IABO78.2 (−/−); (v) pollen shedding date, assessed when 50% of plants per plot had extruded anthers; (vi) anthesis-silking interval (ASI), as the difference between silking date (assessed when 50% of plants per plot had extruded silks) and the pollen shedding date; (vii) plant height, measured at the flag leaf collar; (viii) root lodging, consequent upon strong windstorms which occurred c.10 d after silking (i.e. the beginning of the R2 stage) in 2002 and at mid-stem elongation (V10-V11 stage) in 2003. Plants leaning more than 30° from the vertical were counted as lodged; (ix) grain yield, and (x) kernel weight (as a mean of 200 kernels) both adjusted to 15.5% moisture; and (xi) number of kernels per plant, calculated as the ratio between grain yield per plant and kernel weight. In each plot, leaf samples and data were collected on the central plants; in particular, the number of sampled plants was three for SC and WP, five for L-ABA, RWC, and plant height, and seven for flowering dates, root lodging, grain yield, and its components. Greenhouse evaluation of BDLs To obtain further information on the QTL effects on L-ABA concentration and morpho-physiological traits as well as to investigate the QTL effect on root characteristics in relation to the level of water stress, two pairs of BDLs [Os420.2 (+/+) and (−/−); IABO78.2 (+/+) and (−/−)] were tested in the greenhouse with three water regimes. Seeds were surface-sterilized and placed in Petri dishes at ambient temperature in the dark. After 48 h, seedlings at a similar germination stage were transferred in soil columns (diameter 44 cm, height 1 m) holding 27 kg of peat:sand sifted mixture 1:1 v/v for growing until flowering (VT-R1 stage). Twelve plants (i.e. four plants×three water regimes) were grown for each BDL under the following conditions, day: 16 h, 26–28 °C, with supplemental light 500 μE m−2 s−1 photosynthetic photon flux density; night: 16 °C. To achieve different levels of water stress, irrigation was interrupted at subsequent times (i.e. plants at the V8 stage to obtain the severe stress and at the V11 stage for the moderate stress) while irrigation was continued up to tassel appearance (i.e. a few days before VT) for the well-watered treatment. Plants were distributed according to a completely randomized design and their position was periodically changed. At flowering (VT-R1), data were taken at the single plant level for: (i) L-ABA concentration and (ii) RWC on the central part of the third leaf from the top; (iii) WP on the tip of the third leaf from the top; (iv) plant height; and (v) shoot dry weight. Traits (i) to (iv) were measured as described for the NIHs in the field. To carry out root analyses, the soil level was marked on the stem, the plants were then removed from the pots, and the soil mix was gently removed from the roots by washing. For both the first node above the soil level (A) and the first node below the soil level (B) data were taken for (vi) root number per node (for node A, only elongated and branched roots were taken into account); (vii) root diameter at 5 mm from the stem; (viii) root length; (ix) root angle of growth from the vertical (0°) using a protractor; (x) root branching, scoring for the lateral roots emerging from the whole crown roots, from 0 (no branching) to 9 (maximum branching); (xi) dry weight of the whole root system (i.e. considering roots from all the above- and below-soil nodes); and (xii) root dry weight/shoot dry weight. Root diameter, length, and angle were measured on three random roots per node. Statistical analysis For field experiment on NIHs, the analysis of variance (ANOVA) was conducted on the plot mean values of each trial and then combined across trials. Prior to ANOVA, the data of root lodging percentage were subjected to angular transformation (Steel and Torrie, 1980). The variation among the four trials was partitioned as between years, between water regimes (i.e. 120 versus 40% irrigation volumes) and water regime×year interaction. The variation among the 16 NIHs was analysed by following two alternative procedures providing complementary information. The first procedure was followed in order to obtain information on the average additive and dominance effects of the QTL in question. In particular, the variation among NIHs was partitioned into variation among the three groups of NIHs defined on the basis of their genotype at the target QTL [i.e. (+/+), (−/−) and (+/−)] and residual. The variation among the three groups was further subdivided into (+/+) versus (−/−), estimating the average additive effect of the QTL, and (+/−) versus [mean value of (+/+) and (−/−)], estimating the average dominance effect. The second procedure was followed in order to obtain information on the contribution of each BDL family to the average additive effect. In particular, the variation among the 16 NIHs was partitioned into effects of general combining ability (g.c.a.) due to the four parental BDLs derived by Os420, effects of g.c.a. due to the four parental BDLs derived by IABO78, and effects of specific combining ability (s.c.a.). It should be noted that the g.c.a. of a line is related to its mean performance across its hybrids, while the s.c.a. of a hybrid is related to the deviation between the observed and the expected performance based on the g.c.a. of the two parental lines (Falconer and Mackay, 1996). Effects of g.c.a. are mainly due to additive gene action while effects of s.c.a. are due to non-additive gene actions (mainly dominance). For both Os420 and IABO78 backgrounds, the g.c.a. effects were then partitioned into (+/+) versus (−/−) BDLs within the first family (thus estimating the additive effect of the QTL in the first family), (+/+) versus (−/−) BDLs within the second family (additive effect of the second family), and between families. All the interactions involving hybrids were partitioned following the two alternative procedures. The effects of years were considered as random and the effects due to irrigation volumes and hybrids as fixed; therefore, the F-test for irrigation volumes and hybrids was made by using, as denominator, the corresponding interactions with years in case such interactions were significant. For SC and WP, which were investigated on only two pairs of NIHs, the variation among hybrids was partitioned following the first procedure only, i.e. as between (+/+) and (−/−) groups and residual. With respect to the greenhouse experiment on BDLs, the ANOVA was conducted on the data collected on the 48 plants (i.e. on the single shoot or root data or the mean value of the three roots per node). The variation among the 12 entries (i.e. three water regimes×four BDLs) was partitioned into among water regimes, among BDLs, and water regime×BDL interaction. The among BDLs was then partitioned into between families of BDLs, i.e. Os420.2 versus IABO78.2, between the two groups of BDLs differing for the QTL [i.e. (+/+) versus (−/−)], and the interaction (Os420.2 versus IABO78.2)×[(+/+) versus (−/−)]. For the F-test a fixed model was adopted and the among plants within entry was used as the error term. Results Field evaluation of NIHs The ANOVA (not shown) indicated that the first and second order interactions involving years were significant only in a few instances; for this reason and for the sake of conciseness only mean values across years are presented and discussed. The comparison between the mean values of the two water regimes across years and hybrids was highly significant (P ≤0.01) for most traits, with the exception of RWC, pollen shedding date, and kernel weight (Table 1). As compared with the well-watered treatment, the water-stressed treatment led, as expected, to an increase in L-ABA and ASI as well as to a decrease in stomatal conductance (SC), leaf water potential (WP), plant height, root lodging, grain yield, and number of kernels per plant. Table 1. Field investigation: mean values for the two water regimes across two years (2002 and 2003) and 16 NIHs Trait   Well-watereda   Water-stresseda   b   L-ABA (ng g−1 DW)  327  481  **  Relative water content (RWC) (%)  92.4  92.5  ns  Stomatal conductance (SC) (cm s−1)c  0.53  0.35  **  Leaf water potential (WP) (bar)c  −13.0  −14.9  **  Pollen shedding dated  31.1  31.2  ns  ASI (d)  5.9  7.8  **  Plant height (cm)  226  203  **  Root lodging (%)e  31.2  26.5  **  Grain yield (mg ha−1)  7.82  4.79  **  Kernels per plant (no.)  476  311  **  Kernel weight (mg)   318   307   ns   Trait   Well-watereda   Water-stresseda   b   L-ABA (ng g−1 DW)  327  481  **  Relative water content (RWC) (%)  92.4  92.5  ns  Stomatal conductance (SC) (cm s−1)c  0.53  0.35  **  Leaf water potential (WP) (bar)c  −13.0  −14.9  **  Pollen shedding dated  31.1  31.2  ns  ASI (d)  5.9  7.8  **  Plant height (cm)  226  203  **  Root lodging (%)e  31.2  26.5  **  Grain yield (mg ha−1)  7.82  4.79  **  Kernels per plant (no.)  476  311  **  Kernel weight (mg)   318   307   ns   a Irrigation volumes corresponding to 120% (well-watered) and 40% (water-stressed) of evapotranspiration after accounting for rainfall. b Comparison between the two water regimes. **, Significant at P ≤0.01; ns, not significant. c Measured on only four NIHs. d 1=June 1st. e Data refer to actual percentages, while the ANOVA was conducted on the angular transformed data. View Large The water regime×NIH interaction was significant (P ≤0.05) only for L-ABA and due to magnitude effects rather than changes in rank; in fact, when the ANOVA was conducted on the L-ABA data previously subjected to logarithmic transformation (data not shown), the interaction was not significant. The significance of the water regime×NIH interaction of the L-ABA original data was solely due to the component water regimes × [(+/+) versus (−/−)]. Under well-watered conditions the mean values for the (+/+) and (−/−) groups of NIHs were 373 and 290 ng ABA g−1 DW, respectively, while under water-stressed conditions the corresponding values were 554 and 405 ng ABA g−1 DW, respectively (Table 2). Therefore, the average additive effects of the investigated QTL [calculated as half the difference between the means of the (+/+) and of the (−/−) groups] were 42 ng ABA g−1 DW for the well-watered and 75 ng ABA g−1 DW for the water-stressed conditions. These additive effects correspond to 12.5% (well-watered) and to 15.5% (water-stressed) as referred to the mean value of the two groups of NIHs in each water regime. Table 2. Field investigation: mean values of the three groups of NIHs [four (+/+), four (−/−) and eight (+/−) at the target QTL] across two water regimes (except L-ABA) and two years Trait   (+/+)   (−/−)   Additive effecta   Meanb   (+/−)   Dominance effectc   L-ABA, well-watered (ng g−1 DW)  373  290  42 **  331  323  −8 ns  L-ABA, water-stressed (ng g−1 DW)  554  405  75 **  480  482  2 ns  L-ABA, mean (ng g−1 DW)  464  348  58 **  406  402  −3 ns  Relative water content (RWC) (%)d  92.3  92.2  0.0 ns  92.2  92.7  0.5 ns  Stomatal conductance (SC) (cm s−1)d  0.41  0.47  −0.03 **  0.44  –  –  Leaf water potential (WP) (bar)d  −14.0  −13.9  0.0 ns  −13.9  –  –  Pollen shedding datee  31.4  31.0  0.2 ns  31.2  31.1  −0.1 ns  ASI (d)  6.9  6.7  0.1 ns  6.8  6.9  0.1 ns  Plant height (cm)  215  213  1 ns  214  214  0 ns  Root lodging (%)f  18.0  43.7  −12.9 **  30.9  26.9  −4.0 ns  Grain yield (mg ha−1)  6.27  6.13  0.07 ns  6.20  6.41  0.21 ns  Kernels per plant (no.)  402  379  15 ns  391  397  6 ns  Kernel weight (mg)   308   316   −4 *   312   313   1 ns   Trait   (+/+)   (−/−)   Additive effecta   Meanb   (+/−)   Dominance effectc   L-ABA, well-watered (ng g−1 DW)  373  290  42 **  331  323  −8 ns  L-ABA, water-stressed (ng g−1 DW)  554  405  75 **  480  482  2 ns  L-ABA, mean (ng g−1 DW)  464  348  58 **  406  402  −3 ns  Relative water content (RWC) (%)d  92.3  92.2  0.0 ns  92.2  92.7  0.5 ns  Stomatal conductance (SC) (cm s−1)d  0.41  0.47  −0.03 **  0.44  –  –  Leaf water potential (WP) (bar)d  −14.0  −13.9  0.0 ns  −13.9  –  –  Pollen shedding datee  31.4  31.0  0.2 ns  31.2  31.1  −0.1 ns  ASI (d)  6.9  6.7  0.1 ns  6.8  6.9  0.1 ns  Plant height (cm)  215  213  1 ns  214  214  0 ns  Root lodging (%)f  18.0  43.7  −12.9 **  30.9  26.9  −4.0 ns  Grain yield (mg ha−1)  6.27  6.13  0.07 ns  6.20  6.41  0.21 ns  Kernels per plant (no.)  402  379  15 ns  391  397  6 ns  Kernel weight (mg)   308   316   −4 *   312   313   1 ns   a Calculated as the halved difference between the mean values of the (+/+) and (−/−) groups. *, **, Significant at P ≤0.05 and at P ≤0.01, respectively; ns, not significant. b Mean value of the (+/+) and (−/−) NIHs. c Calculated as the difference between the mean values of the (+/−) and [(+/+) and (−/−)] groups. *, **, Significant at P ≤0.05 and at P ≤0.01, respectively; ns, not significant. d Measured on only two (+/+) NIHs and the two corresponding (−/−) NIHs. e 1=June 1st. f Data refer to actual percentages, while the ANOVA was conducted on the angular transformed data. View Large The differences among NIHs across years and water regimes were highly significant (P ≤0.01) for L-ABA, SC, and kernel weight, and significant (P ≤0.05) for root lodging; by contrast, for traits like plant height and grain yield differences were not significant. The comparison between the (+/+) and (−/−) groups largely accounted for most of the variation among NIHs for L-ABA (91.5%), for SC (87.8%), and for root lodging (73.7%), while the (+/+) and (−/−) groups had a much smaller effect on kernel weight (3.7%). The mean values of the (+/+) and (−/−) groups are presented in Table 2. The average additive effects calculated from these mean values were 58 ng g−1 DW for L-ABA [corresponding to 14.3% of the (+/+) and (−/−) mean], −0.03 cm s−1 for SC, −12.9% for root lodging, and −4 mg for kernel weight. For all traits, the mean value of the (+/−) group of NIHs did not significantly differ from the mean of the (+/+) and (−/−) groups (Table 2), thus indicating negligible dominance effects for the QTL in question. The alternative procedure followed in order to analyse the differences among the 16 NIHs, pointed out that the significance of such differences was due only to g.c.a. effects of the parental BDLs, because the s.c.a. effects were not significant for any trait. This latter finding further suggests that the target QTL does not exert important dominance effects for the investigated traits, consistent with the results seen previously. Because of the large prevalence of g.c.a. effects, the mean values of each NIH are not reported and only the mean values of the parental BDLs across their hybrids are reported in Table 3 for those traits which showed significant differences among NIHs (BDLs' mean values for SC are not reported in Table 3, despite the highly significant differences among NIHs, because only four NIHs were tested). For L-ABA, the differences between the (+/+) and (−/−) BDL mean values within each family (representing the additive effects) were always highly significant and positive, thus indicating that the four families consistently contributed to the average additive effect (58 ng ABA g−1 DW) previously reported. Table 3. Field investigation: mean values of the parental backcross-derived lines (BDLs) in hybrid combination across two water regimes and 2 years Parental BDLs   L-ABA (ng g−1 DW)   Root lodging (%)b   Kernel weight (mg)   Os420 background        .1 (+/+)  428  20.6  306  .1 (−/−)  380  36.3  301  Additive effecta  48 **  −15.7 **  5 ns  .2 (+/+)  440  22.6  325  .2 (−/−)  369  35.9  318  Additive effect  71 **  −13.3 *  7 ns  IABO78 background        .1 (+/+)  428  26.8  316  .1 (−/−)  372  40.2  323  Additive effect  56 **  −13.4 *  −7 ns  .2 (+/+)  436  19.7  294  .2 (−/−)  380  28.9  317  Additive effect   56 **   −9.2 ns   −23 **   Parental BDLs   L-ABA (ng g−1 DW)   Root lodging (%)b   Kernel weight (mg)   Os420 background        .1 (+/+)  428  20.6  306  .1 (−/−)  380  36.3  301  Additive effecta  48 **  −15.7 **  5 ns  .2 (+/+)  440  22.6  325  .2 (−/−)  369  35.9  318  Additive effect  71 **  −13.3 *  7 ns  IABO78 background        .1 (+/+)  428  26.8  316  .1 (−/−)  372  40.2  323  Additive effect  56 **  −13.4 *  −7 ns  .2 (+/+)  436  19.7  294  .2 (−/−)  380  28.9  317  Additive effect   56 **   −9.2 ns   −23 **   a Calculated as the difference between the (+/+) and (−/−) mean values. *, ** Significant at P ≤0.05 and P ≤0.01, respectively; ns not significant. b Data refer to actual percentages, while the ANOVA was conducted on the angular transformed data. View Large For root lodging, the (+/+) versus (−/−) BDLs within each family was significant or highly significant for all families except for IABO78.2; however, all the additive effects were negative and of rather similar values, indicating that, analogously to L-ABA, the four families consistently contributed to the average additive effect (−12.9%) previously reported. For kernel weight, the difference between the (+/+) and (−/−) BDLs within family was positive and statistically negligible for the two Os420 families, negative and statistically negligible for the IABO78.1 family, and negative and highly significant for the IABO78.2 family. These findings indicate a lack of consistency among such differences and that the average additive effect previously seen (i.e. −4 mg) was mainly due to the contribution of only one family (IABO78.2). For L-ABA, the relationship was investigated between the 16 NIHs tested here and the corresponding parental means of the BDLs tested per se in our previous study (Landi et al., 2005). The correlation coefficient was highly significant under well-watered (r=0.65) and water-stressed conditions (r=0.82), as well as across the two water regimes (r=0.77). These data thus indicate that the capacity to predict the NIHs' performance, based on the performance of their parental BDLs, is quite satisfactory (especially under the water-stressed conditions; r2=67.2%); this is consistent with the prevalence of additive gene action already mentioned for the target QTL. The relationship (across the two water regimes) between the L-ABA mean value of each hybrid and the corresponding root lodging mean value was also investigated. The correlation coefficient was sizeable and highly significant (r= −0.88), indicating that a large proportion (r2=76.9%) of the variability among NIHs for one trait was accounted for by its linear relationship with the other. Greenhouse evaluation of BDLs The ANOVA (not shown) indicated that the comparison between the mean values of the three water regimes was highly significant (P ≤0.01) or significant (P ≤0.05) for several traits. As expected, the water-stressed treatments in comparison with the well-watered treatment (Table 4) led to an increase in L-ABA (i.e. 228, 756, 1078 ng g−1 DW in the well-watered, moderate stress, and severe stress conditions, respectively) and to a decline in RWC, WP, and plant height. The level of water stress also affected the number of roots per node A (i.e. first node above the soil level) and the root length for node A, leading to a decline for both traits. The interaction of water regime×BDL was not significant for any trait, with the exception of L-ABA for which an interaction due to magnitude effects was observed, analogous to what was found for NIHs in the field. In fact, differences among BDLs for L-ABA were larger with the two water-stressed treatments than with the well-watered treatment (for conciseness data are not shown). Table 4. Greenhouse investigation: mean values for the three water regimes across two pairs of BDLs up to flowering Trait  Well-watereda  Water stress     b      Moderatea   Severea     L-ABA (ng g−1 DW)  228  756  1078  **  Relative water content (RWC) (%)  94  92  87  **  Leaf water potential (WP) (bar)  −4.5  −5.1  −14.8  **  Plant height (cm)  161  135  130  **  Shoot dry weight (g)  32.3  30.3  29.7  ns  Root per node Ac (no.)  7.2  6.7  5.8  *  Root per node Bc (no.)  4.5  4.5  3.9  ns  Root diameter Ac (mm)  3.9  4.4  4.3  ns  Root diameter Bc (mm)  2.8  2.8  2.6  ns  Root length Ac (cm)  43.8  38.6  36.6  *  Root length Bc (cm)  59.0  63.9  55.5  ns  Root angle Ac (°)  45.1  45.3  44.6  ns  Root angle Bc (°)  42.6  45.2  41.4  ns  Root branchd  4.4  5.3  4.8  ns  Whole root dry weight (g)  6.5  7.3  7.2  ns  Whole root dry weight/shoot dry weight   0.22   0.24   0.22   ns   Trait  Well-watereda  Water stress     b      Moderatea   Severea     L-ABA (ng g−1 DW)  228  756  1078  **  Relative water content (RWC) (%)  94  92  87  **  Leaf water potential (WP) (bar)  −4.5  −5.1  −14.8  **  Plant height (cm)  161  135  130  **  Shoot dry weight (g)  32.3  30.3  29.7  ns  Root per node Ac (no.)  7.2  6.7  5.8  *  Root per node Bc (no.)  4.5  4.5  3.9  ns  Root diameter Ac (mm)  3.9  4.4  4.3  ns  Root diameter Bc (mm)  2.8  2.8  2.6  ns  Root length Ac (cm)  43.8  38.6  36.6  *  Root length Bc (cm)  59.0  63.9  55.5  ns  Root angle Ac (°)  45.1  45.3  44.6  ns  Root angle Bc (°)  42.6  45.2  41.4  ns  Root branchd  4.4  5.3  4.8  ns  Whole root dry weight (g)  6.5  7.3  7.2  ns  Whole root dry weight/shoot dry weight   0.22   0.24   0.22   ns   a Well-watered: irrigation continued up to tassel appearance (a few days before VT); moderate and severe water stress: irrigation interrupted at the V11 and V8 stages, respectively. b Comparison between the three water regimes. *, ** Significant at P ≤0.05 and P ≤0.01, respectively; ns not significant. c A and B: first node above and below the soil level, respectively. d As score for the lateral roots emerging from the crown roots, from 0 (no branching) to 9 (maximum branching). View Large The comparison between the families of BDLs, i.e. Os420.2 versus IABO78.2, was highly significant for L-ABA (792 versus 582 ng g−1 DW, respectively), diameter of the roots for both node A (4.9 versus 3.5 mm, respectively), and B (3.3 versus 2.1 mm, respectively) and root branch (5.3 versus 4.3). For all the other traits the differences between the two families of BDLs were not significant. The comparison between the (+/+) and (−/−) groups of BDLs across the Os420 and IABO78 genetic backgrounds was significant or highly significant for L-ABA and several root traits (Table 5). As to L-ABA, the (+/+) group of BDLs showed, as expected, a higher mean value than the (−/−) group, i.e. 783 versus 591 ng g−1 DW, with a QTL additive effect of 96 ng ABA g−1 DW, or 14.0% of the overall mean. The (+/+) group also exhibited a higher mean value for the number of roots per node B (4.8 versus 3.8), root diameter for both node A (4.5 versus 3.9 mm) and B (3.0 versus 2.5 mm), root angle for node A (48.0 versus 41.9°), root branching score (5.2 versus 4.4), whole-root dry weight (8.7 versus 5.8 g), and ratio between this latter trait and shoot dry weight (0.26 versus 0.19); therefore, the additive effect for all these traits was always positive, consistently with the positive effect found for L-ABA. Although differences between the two groups of BDLs were not significant for the number of roots per node A and the root angle for node B, it is noteworthy that the (+/+) group of BDLs showed higher values for both traits, thus strengthening the positive additive effect detected for the number of roots per node B and the root angle for node A. The only root trait showing non-significant differences for both node A and B was root length; however, it should be mentioned that the coefficients of variation for root length of both nodes were close to 20%, suggesting that the lack of significant differences (especially for node A) could be due to the high level of environmental variation affecting such traits. The superiority for the whole root mass of the (+/+) group compared with the (−/−) group can also be appreciated from Fig. 1 as it shows the root system at flowering of each of the four BDLs grown with moderate water stress. Fig. 1. View largeDownload slide Photographs of the root system of the four BDLs grown at moderate water stress in the greenhouse. Representative plants were taken at flowering. The bar is equivalent to 5 cm. Fig. 1. View largeDownload slide Photographs of the root system of the four BDLs grown at moderate water stress in the greenhouse. Representative plants were taken at flowering. The bar is equivalent to 5 cm. Table 5. Greenhouse investigation: mean values across three water regimes and two genetic backgroundsaof the (+/+) and (−/−) groups of BDLs up to flowering Trait   (+/+)   (−/−)   Additive effectb     L-ABA (ng g−1 DW)  783  591  96  *  Relative water content (RWC) (%)  89.0  93.2  −2.1  ns  Leaf water potential (WP) (bar)  −10.6  −8.4  −1.1  ns  Plant height (cm)  143  141  1  ns  Shoot dry weight (g)  31.6  29.9  0.9  ns  Root per node Ac (no.)  7.1  6.1  0.5  ns  Root per node Bc (no.)  4.8  3.8  0.5  *  Root diameter Ac (mm)  4.5  3.9  0.3  *  Root diameter Bc (mm)  3.0  2.5  0.2  *  Root length Ac (cm)  37.6  41.4  −1.9  ns  Root length Bc (cm)  60.6  58.3  1.1  ns  Root angle Ac (°)  48.0  41.9  3.1  **  Root angle Bc (°)  45.3  41.4  1.9  ns  Root branchd  5.2  4.4  0.4  *  Whole root dry weight (g)  8.7  5.8  1.4  **  Whole root dry weight/shoot dry weight   0.26   0.19   0.03   **   Trait   (+/+)   (−/−)   Additive effectb     L-ABA (ng g−1 DW)  783  591  96  *  Relative water content (RWC) (%)  89.0  93.2  −2.1  ns  Leaf water potential (WP) (bar)  −10.6  −8.4  −1.1  ns  Plant height (cm)  143  141  1  ns  Shoot dry weight (g)  31.6  29.9  0.9  ns  Root per node Ac (no.)  7.1  6.1  0.5  ns  Root per node Bc (no.)  4.8  3.8  0.5  *  Root diameter Ac (mm)  4.5  3.9  0.3  *  Root diameter Bc (mm)  3.0  2.5  0.2  *  Root length Ac (cm)  37.6  41.4  −1.9  ns  Root length Bc (cm)  60.6  58.3  1.1  ns  Root angle Ac (°)  48.0  41.9  3.1  **  Root angle Bc (°)  45.3  41.4  1.9  ns  Root branchd  5.2  4.4  0.4  *  Whole root dry weight (g)  8.7  5.8  1.4  **  Whole root dry weight/shoot dry weight   0.26   0.19   0.03   **   a Os420.2 and IABO78.2 families of BDLs. b Calculated as the halved difference between the mean values of (+/+) and (−/−) groups of BDLs. *, ** Significant at P ≤0.05 and P ≤0.01, respectively; ns not significant. c A and B: first node above and below the soil level, respectively. d As score for the lateral roots emerging from the crown roots, from 0 (no branching) to 9 (maximum branching). View Large The interaction (Os420.2 versus IABO78.2)×[(+/+) versus (−/−)] was not significant for any trait, indicating that the QTL effect did not substantially change from one genetic background to the other. Discussion QTL effects on L-ABA and physiological traits Consistently with the well-known effects of water stress on maize, most of the investigated traits were significantly affected by the water regimes applied in the field and in the greenhouse, thus indicating the adequacy of such water regimes for the objectives of the present work. Interestingly, for both NIHs and BDLs the interaction water regime×genotype and its component water regime×[(+/+) versus (−/−)] were always negligible, the only exception being L-ABA, for which a change in magnitude of the differences among genotypes, or between (+/+) and (−/−) groups of genotypes, was detected. The negligible role of water stress on the QTL additive effect for L-ABA and associated traits was also pointed out when the eight BDLs were tested per se in the field (Landi et al., 2005). Altogether, these results indicate that the QTL additive effect on L-ABA and on the associated physiological traits is not much affected by the intensity of water stress, at least within the levels attained in this study. Therefore, in accordance with what was discussed by Blum (1996) concerning the distinction between constitutive and adaptive traits, the low level of the water regime×genotype interaction detected here indicates that the effects of this QTL on the investigated traits are generally constitutive rather than adaptive. On a breeding basis, a QTL with constitutive effects on a number of traits influencing the water balance and water use of the plant may contribute alleles that provide, more predictably, advantages under varying environmental conditions and rainfall patterns/water regimes, as recognized by Blum (1996, 2002). Recently, Tardieu and co-workers have applied a modelling approach in maize to investigate the constitutive versus the adaptive effects of QTLs on morphological traits in relation to environmental parameters (Reymond et al., 2003; Tardieu, 2003). In this respect, it will be valuable to follow a similar approach to ascertain, in greater detail, the influence of environmental parameters on the QTL effects for morpho-physiological traits (e.g. root and leaf elongation, pollen sterility, ovary abortion, etc) affected to varying degrees by ABA and which have been shown to influence final grain yield (Saini and Westgate, 2000; Tuberosa et al., 2002; Reymond et al., 2003). The QTL additive effect for L-ABA averaged across water regimes was always sizeable and, when referred to the genotypes' mean, was c. 14% for both the NIHs in the field and the BDLs in the greenhouse. The average additive effect of the QTL was c. 11% for the BDLs grown in the field in a previous study (Landi et al., 2005) and 12% in the study conducted on 80 random F3:4 lines obtained from the same cross (i.e. Os420×IABO78; Tuberosa et al., 1998). All these findings consistently indicate that the QTL additive effect for L-ABA is stable, irrespective of the vigour of the tested materials, i.e. of their inbreeding coefficients (which should be equal to 0 for NIHs, 0.88 for F3:4 lines, and should be close to 1 for the BDLs). Conversely, the dominance effect (as pointed out by the study on NIHs) was negligible. The much greater importance of the additive versus the dominance effect fully accounts for the marked changes of the allelic frequencies at the QTL following a divergent selection for L-ABA on the source F2 of the cross Os420×IABO78 (Landi et al., 2005). With respect to the associated effects on physiological traits, the additive effect of the QTL was negative for SC, consistent with the well-known role of ABA in reducing SC (Quarrie, 1991). Interestingly, the additive effect estimated on four NIHs (−0.03 cm s−1) was the same as that estimated on 80 random F3:4 lines investigated by Sanguineti et al. (1999), thus indicating the stability of the QTL effect across generations, not only for L-ABA but also for SC. Contrary to SC, no associated effects of the QTL were found for RWC and WP. As to RWC, the lack of associated effect is in contrast with the findings of the study on the eight BDLs in the field in which a significant and negative additive effect of the QTL was detected (Landi et al., 2005). At least in part, such a discrepancy might be due to differences among the dynamics of drought episodes experienced by the plants from one environment to the other, especially with respect to the intensity of the stress and/or to the plants' growing stage in which it occurred. Additional factors possibly involved are represented by the level of the soil water table and the intensity of evapotranspiration during the RWC measurements. For WP, the results presented here are consistent with those reported by Pekic et al. (1995) and by Landi et al. (2001), who found that materials differing in L-ABA did not significantly differ for WP. This lack of relationship between L-ABA and WP is not surprising because maize, an isohydric species (Tardieu, 1996), maintains a stable water status at varying stress conditions by regulating SC. QTL effects on agronomic traits With respect to the agronomic traits evaluated on the NIHs, the QTL showed a strong negative additive effect on root lodging, as the (+/+) group of NIHs was much less susceptible than the (−/−). This finding is consistent with what has been reported previously in a preliminary study on a pair of NIHs (Landi et al., 2005), thus suggesting that this associated effect is also quite stable across environments. It is well-known that root lodging in maize depends on both root architecture affecting the strength of root anchorage to the soil and the leverage effect exerted by the wind force on the plant. Because no significant differences between the (+/+) and (−/−) NIHs were detected in both the preliminary and the present study for plant height, it seems plausible that such differences in root lodging were mainly, if not solely, due to differences in root architecture. In contrast to root lodging, the QTL had no significant effect on grain yield. This finding is consistent with the preliminary results of Landi et al. (2005) and would suggest that grain yield is not much affected by the QTL in question. However, considerable caution should be exerted in this regard because, in all the field trials considered here and by Landi et al. (2005), root lodging occurred before flowering or just a few weeks after flowering, i.e. when grain yield potential was still not fully determined. Therefore, the (−/−) NIHs, more heavily affected by root lodging than the (+/+) NIHs, were probably more penalized in terms of grain yield because of the negative effect that can be exerted on this latter trait by root lodging (Carter and Hudelson, 1988). This implies that, in the absence of root lodging, the (−/−) NIHs could attain a higher grain yield than the (+/+). This hypothesis is supported by the findings of several studies (reviewed by Saini and Westgate, 2000) reporting a negative association between ABA levels in cereals with seed set and, hence, with grain yield. In addition, a causal role of ABA has also been suggested for ovary abortion, the most important factor determining final grain yield in maize exposed to a water deficit during the reproductive phase, i.e. when the ovary is particularly sensitive to a decline in the supply of sugars (Boyle et al., 1991; Zinselmeier et al., 1999; Boyer and Westgate, 2004). QTL effects on root traits The analysis conducted on the two pairs of BDLs in the greenhouse revealed that the investigated QTL also controls several root characteristics, with similar effects in both genetic backgrounds of Os420 and IABO78. Across the two backgrounds, the QTL positively affected root number, diameter, angle, branching, dry weight, and the ratio between root and shoot dry weight. This substantiates the hypothesis formulated on the basis of field data, i.e. that differences among NIHs for root lodging are mainly due to differences in root architecture affecting the strength of plant anchorage. In fact, several studies have shown that root strength is positively affected by characteristics such as root number, diameter, angle, and weight (Jenison et al., 1981; Ennos et al., 1993; Guingo and Hébert, 1997; Bruce et al., 2001). Moreover, the higher root branching shown in the greenhouse experiment by the (+/+) group versus the (−/−) group coupled with a wider insertion angle of the roots on the main stalk may have been paralleled by a greater root density in the shallower soil layers under field conditions, further increasing the anchorage strength. In this regard, it is worth mentioning that Bolaños et al. (1993) found that maize populations with higher root density in the shallow soil layers were also characterized by a higher root strength. The involvement of the QTL near csu133 in the control of both L-ABA and root strength was also revealed by Lebreton et al. (1995), based on the analysis of F2 plants derived from the cross Polj17×F-2. Interestingly also in that study, the additive effects of the QTL on L-ABA and anchoring strength were concurrent. On a broader scale and exploiting syntenic information, Landi et al. (2005) compared the QTL results for root traits of four maize populations with the root QTL data of seven rice (Oryza sativa, L.) populations. The highest frequency of QTLs for root characteristics in rice was observed in the region syntenic to the maize bin 2.04, further supporting the involvement of this chromosome region in the control of root characteristics. Moreover, this finding suggests that rice could be used as a model species to facilitate the positional cloning of the gene(s) responsible for this maize QTL. Hypotheses on the genetic associations among traits Until the cloning of the QTL on bin 2.04 is completed, it will not be possible to ascertain to what extent the effects on all the above-mentioned traits are due to linkage between the gene(s) for L-ABA and the gene(s) for the associated traits and/or to the pleiotropic action of one or more genes. However, the consistency of all such effects and their similarity with those detected in materials unrelated to those considered here (Lebreton et al., 1995) suggest that pleiotropy is probably involved in this respect. The assumption could be that the QTL marked by csu133 directly controls one trait only, i.e. root architecture, while affecting L-ABA and other traits through a sequence of causally related events, according to the general model described by Lebreton et al. (1995) and Tuberosa et al. (2002). In this respect, it should be mentioned that a preliminary investigation on the two pairs of BDLs tested in the greenhouse showed that the QTL had no significant effects on L-ABA before the V13 stage (S Giuliani, unpublished data). Conversely, in 2003, differences among NIHs for root lodging were already noticeable at an earlier stage (i.e. V10-V11). Therefore, based on these observations, this model hypothesizes that the (+) allele provided by Os420, as compared to the (−) allele of IABO78, may determine a larger root system with a wider insertion angle (i.e. a more horizontal root development). This, in turn, may increase root density in the superficial soil layers, thus accounting for the greater root lodging resistance provided by the Os420 allele. In addition, because the more superficial soil layers dehydrate more quickly, even under irrigated conditions, a larger and more horizontal root system implies a greater flux of xylem ABA towards the leaf, thus accounting for the higher L-ABA and lower SC of the (+/+) NIHs. Therefore, it is suggested that the QTL on bin 2.04 should be named root-ABA1, to emphasize its involvement in the control of both root architecture and L-ABA. Conclusions These results indicate that root-ABA1 exerts an important effect on L-ABA and that this effect is stable across various levels of water stress and of plant vigour of the materials tested. Moreover, this QTL exerts important effects on root architecture, root lodging, and stomatal conductance. Such associated effects are probably due to pleiotropy, with root-ABA1 directly influencing root architecture and growth, particularly in the more superficial soil layers; the effects on L-ABA and the other traits then follow as a sequence of causally related events. To elucidate the genetic basis of these associated effects, the fine mapping of the QTL will be undertaken using as base materials crosses between (+/+) and (−/−) BDLs in the Os420 background as well as in the IABO78 background. The fine mapping of root-ABA1 is an essential prerequisite to undertake its positional cloning, which would represent an important contribution towards a more effective and accurate manipulation of root architecture, a complex trait involved in crops' adaptation to drought and other abiotic stresses. Abbreviations: ASI, anthesis-silking interval; BDL, backcross-derived line; g.c.a., general combining ability; L-ABA, leaf abscisic acid concentration; NIH, near-isogenic hybrid; NIL, near-isogenic line; node A, first node above the soil level; node B, first node below the soil level; QTL, quantitative trait locus; RFLP, restriction fragment length polymorphism; RWC, relative water content; SC, stomatal conductance; s.c.a., specific combining ability; WP, leaf water potential. This work was supported by MIUR, COFIN 40%, Project ‘Analysis of the molecular and phenotypic effects associated with the variation in abscisic acid concentration in leaves of near isogenic lines’. This paper is a contribution of the Interdepartmental Centre of Biotechnology of the University of Bologna. Thanks are due to Sandra Stefanelli and Stefano Vecchi for skilful technical assistance. References Blum A. 1996. Crop responses to drought and the interpretation of adaptation. Plant Growth Regulation  20, 135–148. Google Scholar Blum A. 2002. Drought tolerance: is it a complex trait? In: Saxena NP, O'Toole JC, eds. Field screening for drought tolerance in crop plants with emphasis on rice. International Workshop on Field Screening for Drought Tolerance in Rice. ICRISAT, Patancheru, India, 17–22. Google Scholar Bolaños J, Edmeades GO, Martinez L. 1993. Eight cycles of selection for drought tolerance in lowland tropical maize. III. Responses in drought adaptive physiological and morphological traits. Field Crops Research  31, 269–286. Google Scholar Boyer JS, Westgate ME. 2004. Grain yields with limited water. Journal of Experimental Botany  55, 2385–2394. Google Scholar Boyle MG, Boyer JS, Morgan JW. 1991. Stem infusion of liquid culture medium prevents reproductive failure of maize at low water potential. Crop Science  31, 1246–1252. Google Scholar Bruce W, Desbons P, Crast O, Folkerts O. 2001. Gene expression profiling of two related maize inbred lines with contrasting root-lodging traits. Journal of Experimental Botany  52, 459–468. Google Scholar Carter PR, Hudelson KD. 1988. Influence of simulated wind lodging on corn growth and grain yield. Journal of Productive Agriculture  1, 295–299. Google Scholar Ennos AR, Crook MJ, Grimshaw C. 1993. The anchorage mechanics of maize, Zea mays. Journal of Experimental Botany  44, 147–153. Google Scholar Falconer DS, Mackay TFC. 1996. Introduction to quantitative genetics, 4th edn. London, UK: Longman. Google Scholar Guingo E, Hébert Y. 1997. Relationship between mechanical resistance of the maize root system and root morphology, and their genotypic and environmental variations. Maydica  42, 265–274. Google Scholar Jenison JR, Shank DB, Penny LH. 1981. Root characteristics of 44 maize inbreds evaluated in four environments. Crop Science  21, 233–237. Google Scholar Landi P, Conti S, Gherardi F, Sanguineti MC, Tuberosa R. 1995. Genetic analysis of leaf ABA concentration and agronomic traits in maize hybrids grown under different water regime. Maydica  40, 179–186. Google Scholar Landi P, Sanguineti MC, Conti S, Tuberosa R. 2001. Direct and correlated responses to divergent selection for leaf abscisic acid concentration in two maize populations. Crop Science  41, 335–344. Google Scholar Landi P, Sanguineti MC, Salvi S, Giuliani S, Bellotti M, Maccaferri M, Conti S, Tuberosa R. 2005. Validation and characterization of a major QTL affecting leaf ABA concentration in maize. Molecular Breeding  15, 291–303. Google Scholar Lebreton C, Lazic-Jancic V, Steed A, Pekic S, Quarrie SA. 1995. Identification of QTL for drought responses in maize and their use in testing causal relationships between traits. Journal of Experimental Botany  46, 853–865. Google Scholar Paran I, Zamir D. 2003. Quantitative traits in plants: beyond the QTL. Trends in Genetics  19, 303–306. Google Scholar Pekic S, Stikic R, Tomljanovic L, Andjelkovic V, Ivanovic M, Quarrie SA. 1995. Characterization of maize lines differing in leaf abscisic acid content in the field. 1. Abscisic acid physiology. Annals of Botany  75, 67–73. Google Scholar Quarrie SA. 1991. Implications of genetic differences in ABA accumulation for crop production. In: Davies WJ, Jones HG, eds. Abscisic acid: physiology and biochemistry. Oxford: Bios Scientific Publishers, 227–243. Google Scholar Reymond M, Muller B, Leonardi A, Charcosset A, Tardieu F. 2003. Combining quantitative trait loci analysis and an ecophysiological model to analyze the genetic variability of the responses of maize leaf growth to temperature and water deficit. Plant Physiology  131, 664–675. Google Scholar Ritchie SW, Hanway JJ, Benson GO. 1997. How a corn plant develops. Iowa State University of Science and Technology Cooperative Extension Service, Ames, Iowa. Special Report  48, 1–21 (reprinted). Google Scholar Saini HS, Westgate ME. 2000. Reproductive development in grain crops during drought. Advances in Agronomy  68, 59–96. Google Scholar Sanguineti MC, Conti S, Landi P, Tuberosa R. 1996. Abscisic acid concentration in maize leaves: genetic control and response to divergent selection in two populations. Maydica  41, 193–203. Google Scholar Sanguineti M, Tuberosa R, Landi P, Salvi S, Maccaferri M, Casarini E, Conti S. 1999. QTL analysis of drought-related traits and grain yield in relation to genetic variation for leaf abscisic acid concentration in field-grown maize. Journal of Experimental Botany  50, 1289–1297. Google Scholar Sharp RE, Poroyko V, Hejlek L, Spollen WG, Springer GK, Bohnert H J, Nguyen HT. 2004. Root growth maintenance during water deficits: physiology to functional genomics. Journal of Experimental Botany  55, 2343–2351. Google Scholar Steel RGD, Torrie JH. 1980. Principles and procedures of statistics a biometrical approach, 2nd edn. New York: McGraw-Hill Book Company. Google Scholar Tardieu F. 1996. Drought perception by plants. Do cells of droughted plants experience water stress? Plant Growth Regulation  20, 93–104. Google Scholar Tardieu F. 2003. Virtual plants: modelling as a tool for the genomics of tolerance to water deficit. Trends in Plant Science  8, 9–14. Google Scholar Tuberosa R, Sanguineti MC, Landi P. 1994. Abscisic acid concentration in the leaf and xylem sap, leaf water potential, and stomatal conductance in drought-stressed maize. Crop Science  34, 1557–1563. Google Scholar Tuberosa R, Sanguineti MC, Landi P, Salvi S, Casarini E, Conti S. 1998. RFLP mapping of quantitative trait loci controlling abscisic acid concentration in leaves of drought-stressed maize (Zea mays L.). Theoretical and Applied Genetics  97, 744–755. Google Scholar Tuberosa R, Salvi S, Sanguineti MC, Landi P, Maccaferri M, Conti S. 2002. Mapping QTLs regulating morpho-physiological traits and yield in drought-stressed maize: case studies, shortcomings and perspectives. Annals of Botany  89, 941–963. Google Scholar Wang Z, Mambelli S, Setter TL. 2002. Abscisic acid catabolism in maize kernels in response to water deficit at early endosperm development. Annals of Botany  90, 623–630. Google Scholar Zinselmeier C, Jeong BR, Boyer JS. 1999. Starch and the control of kernel number in maize at low water potentials. Plant Physiology  121, 25–36. Google Scholar Published by Oxford University Press [2005] on behalf of the Society for Experimental Biology.

Quilot, B.;Génard, M.;Lescourret, F.;Kervella, J.

doi: 10.1093/jxb/eri304pmid: 16234284

Abstract Ecophysiological models are increasingly expected to describe genotypic variation within breeding populations. Accordingly, the ability of an ecophysiological model of peach to explain variation in fruit quality among 100 genotypes of a second backcross progeny derived from a clone of wild peach (Prunus davidiana) crossed with two commercial nectarine (Prunus persica) varieties was explored. Experimental measurements were carried out to calibrate the model for each genotype. The predictive quality of the model was tested on several independent datasets. The genotypic variation in dry and fresh growth of the fruit and the stone were effectively described by the model. Prediction of the amount of total sugar in flesh at maturity was accurate, whereas prediction of flesh dry matter content and total sugar concentration was suitable but less accurate. This approach and the results have allowed physiological processes to be ranked according to their contribution to the variation in fruit quality between genotypes. Fruit growth demand and the hydraulic conductance in the fruit were the main processes that explained the fruit quality variation. Shortcomings and further potential uses of the model are discussed. Fruit, genotypic variation, growth, modelling, Prunus persica, sugars Introduction Ecophysiological models have generally been developed and calibrated on a few varieties of a given species. Increasingly, it is required that they also describe the genetic variation between varieties or within a breeding population. The ecophysiological models, which are able to represent genotypic variation accurately, require genetic coefficients that are specific to each genotype and constant under a wide range of environmental conditions (Boote et al., 2001; Tardieu, 2003). It is not clear whether current ecophysiological models are efficient for such applications. Adequate models should be mechanistic enough to give a representative description of physiological processes. However, very complex models are not suitable when a large number of genotypes needs characterization, because they require too many inputs, expensive or time-consuming measurements, and large amounts of plant material. Studies are thus still needed to test the potential uses of current ecophysiological models for plant breeding, to identify their limitations, and to specify the necessary modifications for such applications. The present study aims to test the efficiency of an ecophysiological model developed to simulate genotypic variation in fruit quality traits. Quality traits were chosen because they are under the control of many processes and environmental factors. Peach was selected for study because ecophysiological models (Lescourret et al., 1998; Fishman and Génard, 1998; Quilot, 2003) that reproduce important quality traits (fruit and stone sizes, flesh dry matter content, and total sugar amount in the flesh) are available for this species. This study used a population of genotypes derived from a clone of wild peach (Prunus davidiana) with three generations of crosses with commercial varieties of nectarine (Prunus persica). The ecophysiological model used was a combination of three sub-models dealing with carbon balance, water balance, and sugar accumulation. Adaptations of the model were necessary to take into account the specific behaviours of the genotypes studied. Initially, the results are presented of a sensitivity analysis to reveal the parameters to which the model is most sensitive. These parameters are expected to explain most of the variations within outputs. Secondly, the values of most of the model parameters were estimated for each genotype. Their variations between genotypes were analysed to identify genotypic key parameters. A third step examined the genotypic ability of the model to describe fruit and stone growth and flesh sugar concentration. The predictive quality of the model was also evaluated. Finally, the relative contribution of the genotypic key parameters to variations in fruit quality was analysed. The model The model integrates three sub-models developed independently, all concerning the stage of fruit enlargement at the end of cell division. The modelled system is the ‘shoot bearing fruit’ which is represented by three interconnected compartments: fruits including flesh and stone, 1-year-old stems, and leafy shoots. The focus here is on the processes that particularly relate to the fruit quality issue, giving emphasis to equations that were added or modified to describe the behaviour of wild genotypes. The main variables predicted by the integrated model are dry and fresh masses of fruit and stone, flesh dry matter content, and total sugar amount and concentration in the flesh. The parameters are presented in Table 1. (Details about the processes described by the sub-models and the equations are given as supplementary data in the Appendix which is available at JXB online.) Table 1. Symbols, definitions and units of the model parameters Parameter   Definition   Unit   Origin   Leaf assimilation            r1  Leaf structural mass/leafy shoot structural mass  Dimensionless  Experiments      SLA  Specific Leaf Area  m2 g−1  Experiments      p1  Light-saturated maximal leaf photosynthesis  μmol CO2 m−2 s−1  Experiments      p  Concern leaf photosynthesis regulation by reserves  Dimensionless  Quilot et al. (2004a)      k    Dimensionless  Quilot et al. (2004a)      r2  Leaf reserve mass/leafy shoot reserve mass  Dimensionless  Ben Mimoun (1997)      p3  Concern the calculation of leaf photosynthesis from radiation and light-saturated photosynthesis  μmol CO2 m−2 s−1  Higgins et al. (1992)      p4    μmol CO2 μmol photon−1  Higgins et al. (1992)  Radiation in the shade            p7  Concern the calculation of radiation received by shaded leaves  μmol photon m−2 s−1  Lescourret et al. (1998)      p8    m2 s μmol photon−1  Lescourret et al. (1998)      r3    Dimensionless  Lescourret et al. (1998)  Reserve mobilization            r4  Leafy shoot mobile fraction of reserves  Dimensionless  Ben Mimoun (1997)      r5  1-year-old stem mobile fraction of reserves  Dimensionless  Ben Mimoun (1997)      r6  Threshold ratio of reserves in the leaves  Dimensionless  Ben Mimoun (1997)  Maintenance respiration demand            MRROst  Maintenance respiration rate of current-year stem (Ost), 1-year-old-stem (st), leaf and fruit compartments at the reference temperature  s−1  Grossman and DeJong (1994a)      MRRst    s−1  Grossman and DeJong (1994a)      MRRleaf    s−1  Grossman and DeJong (1994a)      MRRfruit    s−1  DeJong and Goudriaan (1989)       \(Q_{10}^{Ost}\)     Dimensionless  Grossman and DeJong (1994b)       \(Q_{10}^{\mathrm{st}}\)   Q10 value for current-year stem (Ost), 1-year-old-stem (st), leaf and fruit compartments  Dimensionless  Grossman and DeJong (1994b)       \(Q_{10}^{\mathrm{leaf}}\)     Dimensionless  Grossman and DeJong (1994b)       \(Q_{10}^{\mathrm{fruit}}\)     Dimensionless  DeJong et al. (1987)  Fruit growth demand            GRCflesh  Growth respiration coefficient of fruit  Dimensionless  DeJong and Goudriaan (1989)       \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   Initial relative flesh growth rate  Degree-days−1  Experiments       \(m_{\mathrm{flesh}}^{\mathrm{max}}\)   Concerns the limitation of flesh growth  g  Experiments  Stone elaboration             \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   Potential stone dry mass at maturity  g  Experiments      kstone  Empirical coefficient relating stone dry mass to fruit dry mass Empirical coefficients relating stone fresh mass to stone dry mass  Dimensionless  Experiments       \(df_{\mathrm{stone}}^{1}\)     Dimensionless  Experiments       \(df_{\mathrm{stone}}^{2}\)     Degree-days−1  Experiments       \(df_{\mathrm{stone}}^{3}\)     Degree-days  Experiments  Fruit fresh mass elaboration            aL  Hydraulic conductance per unit of fruit surface  g cm−2 bar−1 h−1  Experiments      Y  Threshold value of hydrostatic pressure needed for growth  MPa  Fishman and Génard (1998)      ϕ  Cell wall extensibility coefficient in Lockhart's equation  MPa−1 h−1  Fishman and Génard (1998)      πoc  Osmotic pressure in fruit due to other compounds than soluble sugars  MPa  Fishman and Génard (1998)      rsu  Concerns the calculation throughout growth of the part of sucrose in the total amount of sugars in the flesh  Dimensionless  Experiments      s1  Empirical parameters relating fruit surface to fruit mass  cm2  Experiments      s2    g  Experiments      ρ  Permeation coefficient of fruit surface to water vapour  cm h−1  Experiments  Total sugar accumulation            ksugar   Coefficient of the transfer function between sugars and other compounds   d−1   Experiments   Parameter   Definition   Unit   Origin   Leaf assimilation            r1  Leaf structural mass/leafy shoot structural mass  Dimensionless  Experiments      SLA  Specific Leaf Area  m2 g−1  Experiments      p1  Light-saturated maximal leaf photosynthesis  μmol CO2 m−2 s−1  Experiments      p  Concern leaf photosynthesis regulation by reserves  Dimensionless  Quilot et al. (2004a)      k    Dimensionless  Quilot et al. (2004a)      r2  Leaf reserve mass/leafy shoot reserve mass  Dimensionless  Ben Mimoun (1997)      p3  Concern the calculation of leaf photosynthesis from radiation and light-saturated photosynthesis  μmol CO2 m−2 s−1  Higgins et al. (1992)      p4    μmol CO2 μmol photon−1  Higgins et al. (1992)  Radiation in the shade            p7  Concern the calculation of radiation received by shaded leaves  μmol photon m−2 s−1  Lescourret et al. (1998)      p8    m2 s μmol photon−1  Lescourret et al. (1998)      r3    Dimensionless  Lescourret et al. (1998)  Reserve mobilization            r4  Leafy shoot mobile fraction of reserves  Dimensionless  Ben Mimoun (1997)      r5  1-year-old stem mobile fraction of reserves  Dimensionless  Ben Mimoun (1997)      r6  Threshold ratio of reserves in the leaves  Dimensionless  Ben Mimoun (1997)  Maintenance respiration demand            MRROst  Maintenance respiration rate of current-year stem (Ost), 1-year-old-stem (st), leaf and fruit compartments at the reference temperature  s−1  Grossman and DeJong (1994a)      MRRst    s−1  Grossman and DeJong (1994a)      MRRleaf    s−1  Grossman and DeJong (1994a)      MRRfruit    s−1  DeJong and Goudriaan (1989)       \(Q_{10}^{Ost}\)     Dimensionless  Grossman and DeJong (1994b)       \(Q_{10}^{\mathrm{st}}\)   Q10 value for current-year stem (Ost), 1-year-old-stem (st), leaf and fruit compartments  Dimensionless  Grossman and DeJong (1994b)       \(Q_{10}^{\mathrm{leaf}}\)     Dimensionless  Grossman and DeJong (1994b)       \(Q_{10}^{\mathrm{fruit}}\)     Dimensionless  DeJong et al. (1987)  Fruit growth demand            GRCflesh  Growth respiration coefficient of fruit  Dimensionless  DeJong and Goudriaan (1989)       \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   Initial relative flesh growth rate  Degree-days−1  Experiments       \(m_{\mathrm{flesh}}^{\mathrm{max}}\)   Concerns the limitation of flesh growth  g  Experiments  Stone elaboration             \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   Potential stone dry mass at maturity  g  Experiments      kstone  Empirical coefficient relating stone dry mass to fruit dry mass Empirical coefficients relating stone fresh mass to stone dry mass  Dimensionless  Experiments       \(df_{\mathrm{stone}}^{1}\)     Dimensionless  Experiments       \(df_{\mathrm{stone}}^{2}\)     Degree-days−1  Experiments       \(df_{\mathrm{stone}}^{3}\)     Degree-days  Experiments  Fruit fresh mass elaboration            aL  Hydraulic conductance per unit of fruit surface  g cm−2 bar−1 h−1  Experiments      Y  Threshold value of hydrostatic pressure needed for growth  MPa  Fishman and Génard (1998)      ϕ  Cell wall extensibility coefficient in Lockhart's equation  MPa−1 h−1  Fishman and Génard (1998)      πoc  Osmotic pressure in fruit due to other compounds than soluble sugars  MPa  Fishman and Génard (1998)      rsu  Concerns the calculation throughout growth of the part of sucrose in the total amount of sugars in the flesh  Dimensionless  Experiments      s1  Empirical parameters relating fruit surface to fruit mass  cm2  Experiments      s2    g  Experiments      ρ  Permeation coefficient of fruit surface to water vapour  cm h−1  Experiments  Total sugar accumulation            ksugar   Coefficient of the transfer function between sugars and other compounds   d−1   Experiments   The origin of the values used in these simulations is mentioned. View Large The carbon assimilation and allocation sub-model (Lescourret et al., 1998; Génard et al., 1998; Quilot et al., 2002) simulates carbon partitioning based on organ demand and priority rules. Flesh growth potential demand was described in terms of degree-days by a logistic equation (see Equation 9 of the Appendix in the supplementary data at JXB online) which enabled both logistic and exponential growth to be described by modifying the value of the parameter \(m_{\mathrm{flesh}}^{\mathrm{max}},\) which contributes to the slowing down of growth with the maturation process. Since the population studied showed high variation in stone versus flesh partitioning, an equation relating stone growth to fruit growth (see Equation 10 of the Appendix in the supplementary data at JXB online) was added to describe the potential stone growth. Sugar accumulation is simulated by a simple sub-model that predicts the increase of total sugar concentration in the flesh during fruit growth (Quilot et al., 2004b). A function depending on the thermal time describes the partitioning of total sugar between sucrose and other sugars for each genotype (see Equation 17 of the Appendix in the supplementary data at JXB online). The sub-model of water flux simulates fruit growth in fresh mass (Fishman and Génard, 1998). The rate of change of the amount of water in the flesh is computed daily from the water flux through xylem and phloem and the water loss due to fruit transpiration. Materials and methods Plant material The breeding population was derived from clone P1908 of Prunus davidiana and two cultivars of nectarine as follows (Pascal et al., 1998). Firstly, P1908 was crossed with Prunus persica ‘Summergrand’ (S) and an F1 progeny was obtained. Then, one F1 hybrid resistant to powdery mildew was back-crossed to S to produce a BC1 progeny. Finally, BC1 individuals were used to pollinate P. persica ‘Zéphir’ (Z) to derive the breeding population (BC2). S and Z are, respectively, yellow and white nectarine cultivars with large fruit. The study was conducted in three orchards, in the St Paul and Garrigues experimental sites of the INRA Research Centre of Avignon (France) and in the orchard of Gotheron near Valence (120 km north from Avignon). BC2 genotypes and the three parents were planted in the orchards of St Paul and Gotheron in a completely randomized design with one tree per genotype. One tree of these genotypes were also available in the collection orchard at the Garrigues site. In the three sites all genotypes were grafted on GF305 seedling rootstocks and grown under optimal conditions of irrigation, fertilization, and pest control. Trees were 3-years-old in 2001. Data were also used from an experiment carried out in 2000 in Avignon (Quilot et al., 2002) on trees of P1908 and S, grown in 50 l pots. Experiments Experiments were carried out to measure the parameter values for each of the BC2 genotypes and the three parents. Moreover, specific experiments were performed in order to evaluate the predictive quality of the newly parameterized model. Table 2 presents a summary of the characteristics of each experiment. Table 2. Characteristics of the five experiments and summary of the corresponding measurements   Parameterization and analysis of goodness-of-fit   Parameterization   Tests of predictive quality       Dataset  StPBC202  GarBC2  StPBC201  StP00  GothBC202  Year  2002  2001 and 2002  2001  2000  2002  Site  St Paul  Garrigues  St Paul  St Paul  Gotheron  Genotype  P1908, S, Z, 139 BC2    P1908, S, Z, 87 BC2  P1908, S        50 BC2      6 BC2  Fruit load  Very low  Medium  Low  Medium and high  High  Shoot treatments        • Girdling  • Girdling          • Leafy shoot growth stopped    Stem and shading characterization  Stem length and diameter    Stem length and diameter  • Stem length and diameter  • Stem length and diameter          • Hemispherical photographs  • Hemispherical photographs  Weekly monitoring  Fruit diameter    Fruit diameter  Fruit diameter  • Fruit diameter            • Leafy shoot length  Destructive measurements during fruit growth    • Leaf and shoot mass            • Leaf area            • Fruit transpiration            • Fruit diameter and area            • Fruit and stone mass        Measurements on harvested fruits at maturity  • Diameter  • Diameter  • Diameter  • Diameter  • Diameter    • Flesh and stone fresh and dry masses  • Flesh and stone fresh and dry masses  • Flesh and stone fresh and dry masses  • Flesh and stone fresh and dry masses  • Flesh and stone fresh and dry masses    • Fruit transpiration  • Fruit transpiration        Biochemical analysis   Sugar concentration of individual fruits     Sugar concentration of individual fruits     Mean sugar concentration of grouped fruits     Parameterization and analysis of goodness-of-fit   Parameterization   Tests of predictive quality       Dataset  StPBC202  GarBC2  StPBC201  StP00  GothBC202  Year  2002  2001 and 2002  2001  2000  2002  Site  St Paul  Garrigues  St Paul  St Paul  Gotheron  Genotype  P1908, S, Z, 139 BC2    P1908, S, Z, 87 BC2  P1908, S        50 BC2      6 BC2  Fruit load  Very low  Medium  Low  Medium and high  High  Shoot treatments        • Girdling  • Girdling          • Leafy shoot growth stopped    Stem and shading characterization  Stem length and diameter    Stem length and diameter  • Stem length and diameter  • Stem length and diameter          • Hemispherical photographs  • Hemispherical photographs  Weekly monitoring  Fruit diameter    Fruit diameter  Fruit diameter  • Fruit diameter            • Leafy shoot length  Destructive measurements during fruit growth    • Leaf and shoot mass            • Leaf area            • Fruit transpiration            • Fruit diameter and area            • Fruit and stone mass        Measurements on harvested fruits at maturity  • Diameter  • Diameter  • Diameter  • Diameter  • Diameter    • Flesh and stone fresh and dry masses  • Flesh and stone fresh and dry masses  • Flesh and stone fresh and dry masses  • Flesh and stone fresh and dry masses  • Flesh and stone fresh and dry masses    • Fruit transpiration  • Fruit transpiration        Biochemical analysis   Sugar concentration of individual fruits     Sugar concentration of individual fruits     Mean sugar concentration of grouped fruits   View Large Parameterization and analysis of goodness-of-fit Experiments were performed in St Paul in 2002 on 139 genotypes of the BC2 population, S, Z, and P1908 (StPBC202). It was necessary to ensure that all fruits were under non-limiting source conditions (i.e. under maximum growth conditions). For this purpose fruits were harvested at an early stage, leaving only a very light fruit load. These experiments were used to estimate some parameters from non-destructive measurements. Experiments carried out at the Garrigues site in 2001 and 2002 (GarBC2) only consisted of the destructive sampling of fruits and leafy-shoots throughout growth and at maturity in order to estimate other parameters. The StPBC202 dataset was then used to check the goodness-of-fit of the model. Test of the predictive quality of the model Three other experiments were carried out in order to test the predictive quality of the different parts of the model parameterized with the StPBC202 and GarBC2 datasets. Experiments were performed in 2001 in St Paul on 87 out of the 139 BC2 genotypes of the StPBC202 dataset, as well as on S and Z (StPBC201). Again a light fruit load was applied to each tree. Dry mass of fruit and stone was measured and values of the other outputs of the model, i.e. fresh mass of fruit and stone, flesh dry matter content, and sugar variables, were predicted. Experiments on P1908 and S in St Paul in 2000 (StP00) were performed using two leaf-to-fruit ratio treatments referred to as ‘heavy’ and ‘medium’ (5 and 30 leaves per fruit, respectively). There were applied to chosen shoot-bearing-fruits (i.e. 1-year-old woody stems (‘shoots’), bearing fruits and leafy shoots) isolated from tree by girdling. The leafy shoot vegetative growth was stopped during fruit growth by removing the new terminal and lateral apices. The sugar accumulation part of the model was not tested here since no sugar concentration data were available. In the experiments performed on six BC2 genotypes in Gotheron in 2002 (GotBC202), a heavy fruit load (one growing leafy shoot per fruit) was left on the shoot-bearing-fruits again isolated from the tree by girdling. Field measurements Non-destructive measurements: Fruit cheek diameter was measured once a week from the end of May (about 85 d or 590 degree-days after full bloom) to fruit maturity (from mid June to September depending on the genotype). For StPBC201 and StPBC202 experiments, three to five fruits per tree and per genotype were recorded. For GotBC202 experiments, three fruits per shoot and two to five shoots were monitored per genotype. Associated changes in the length of the leafy shoots were also monitored weekly during the same period, to evaluate shoot growth demand. For StP00 experiments, 23 shoots per genotype bearing from one to eight fruits were monitored. The photosynthetic response to radiation intensity and leaf conductance were studied in trees of StPBC201 and StPBC202 experiments. Measurements were taken with the ADC–LCA 4 portable photosynthesis system. They were made on several dates for each genotype on well-expanded sunlit leaves, between 07.00 h and 10.00 h standard time, in order to avoid stomatal closure because of high temperatures and water stress. Light saturation occurs around 1000 μmol m−2 s−1 under spring conditions in the South of France, so measurements made below 1000 μmol m−2 s−1 were not considered further. From these data, p1, corresponding to maximum light-saturated photosynthesis, was estimated. Destructive measurements: Monitored fruits were considered ripe when they stopped growing, softened, and were easily picked. The flesh fresh mass (Wfresh) was determined immediately after harvest. Fruit flesh was cut into small pieces. The flesh dry mass (Wdry) was determined after drying for 72 h at 70 °C to constant weight. For three monitored fruits from StPBC201 and StPBC202, some flesh was immediately frozen (–80 °C) until sugar analysis. For fruits from GotBC202, flesh pieces from the same shoot were bulk frozen as an average sample. For fruits from StP00, no sugar analysis was performed. To compute dry (Wdry) and fresh (Wfresh) fruit, flesh and stone masses for each monitored fruit, several allometric relationships between fruit diameter and dry and fresh fruit masses, stone dry mass and fruit dry mass, stone fresh mass and stone dry mass, were determined for each genotype. To establish these relationships, fruit diameter and dry and fresh masses of fruit and stone were recorded. These measurements were carried out in 2001 and 2002 (i), at maturity, on five fruits per tree, (ii) at thinning, on the fruits removed, and (iii) throughout fruit growth on fruits sampled from trees at the Garrigues site. The permeation coefficient of water vapour through the fruit surface, ρ, was estimated by monitoring fruit mass loss, which is assumed to be proportional to the fruit surface area and to be driven by the difference in relative humidity between the air-filled space within the fruit (100% RH) and the ambient atmosphere. Freshly harvested fruits were placed in a controlled environment room (temperature, RH, and air speed) and periodically weighed. The fruit surface of each fruit was approximated as an ellipsoidal surface computed from the three diameters of the fruit. Measurements were performed (i) throughout growth on fruits at the Garrigues site and (ii) at maturity on monitored fruits not frozen for sugar analysis, from StPBC201 and StPBC202 experiments. Shoots were characterized by two parameters: leaf area relative to the structural part of the leaf (SLA, m2 g−1), estimated from the measurements of surface and mass of 20 leaves for each genotype, and the leaf mass to leafy shoot mass ratio (r1), estimated from the measurement of 3–10 leafy shoots for each genotype. All measurements were made in the morning in May, so that reserves in the leaves were limited. Environmental inputs and initial status Hourly total radiation and daily temperature values were recorded at Avignon in 2001 and 2002 and at Gotheron in 2002. Degree-days were calculated from daily minimum and maximum temperatures with upper and lower temperature thresholds at 35 °C and 7 °C, respectively. Degree-days were summed from full-bloom to maturity for each genotype. To take into account assimilation reduction due to shade, the model requires two series of hourly coefficients (Lescourret et al., 1998). The first one characterizes the mutual shading of leaves occurring within a shoot and the second one the mean light environment of a shoot. Both coefficients were calculated for GotBC202 and StP00 experiments only, using gap fractions derived from digitized hemispherical photographs (Génard and Baret, 1994). Initial dry masses of the monitored 1-year-old stems, at the beginning of the simulations, were estimated from their volumes, calculated from the length and diameter of each stem considered to be cone-shaped and converted into dry mass on the basis of a mean peach wood specific dry weight (0.575 g cm−3). A sensitivity study conducted by Lescourret et al. (1998) showed that errors in assessing the initial reserves of leafy shoots and 1-year-old stems were not critical to the model response, so these initial reserves were set to the value taken by Lescourret et al. (1998), i.e. 10% of the initial dry masses of leafy shoots and 1-year-old stems. Initial total sugar concentration was approximated from early measurements performed by Quilot et al. (2004b). Biochemical analysis Frozen fruit flesh samples were immersed in liquid nitrogen and ground for 2 min to powder (Dangoumeau 300 ball-crusher, Prolabo). Five grams of the powder were mixed with 20 ml of ultra pure water. The mixture was centrifuged at 15 000 g for 15 min at 4 °C. The supernatant was immediately filtered through a Waters C18 cartridge (Waters) to eliminate any interfering apolar residues and through a 0.45 μm Sep-Pak filter (Jasco France) to eliminate large particles. The extract was stored at −80 °C (sealed tube), prior to sugar measurement by HPLC (see Gomez et al., 2002, for details). Statistical analysis Sensitivity analysis: To select the parameters in the integrated model to be measured, the sensitivity of the model to parameter variation was tested. The model outputs (fresh fruit and stone masses, flesh dry matter content, and total sugar concentration) at maturity were compared for high and low values of each parameter and for two contrasting fruit loads corresponding to source and sink limiting conditions. The high and low parameter values were set to plus or minus 50% of the reference parameter values estimated for the ‘Summergrand’ cultivar. A default value taken from the literature (Lescourret et al., 1998) was used when no value was available for ‘Summergrand’. For each parameter and fruit load level, the sensitivity criterion was the difference between the output value for high (OH) and low (OL) values of the parameter, expressed as a percentage of the output value at maturity for the default parameter value (Oo): 100×(OH−OL)/(Oo). Parameters were selected for further study when the absolute value of the sensitivity criterion exceeded 5% for at least one of the four outputs and one of the two fruit loads considered. Parameter estimation and comparison of parameter values between genotypes: Parameter values were estimated for each genotype studied. It was then tested whether there was significant variation between genotypes. Some parameters could be directly computed for each genotype as the mean of the observed values. For these parameters, between- and within-genotype variances were compared using a test of ‘comparison of means of various independent samples’. The result was compared to the critical value derived from the distribution of Fisher–Snedecor. Other parameters could be estimated by fitting a non-linear simple function to the observed data. Lastly, one parameter could only be estimated by calibrating the model for each of the genotypes, by comparing fresh fruit growth predictions and observations. In these two cases, the ‘nls’ procedure of Splus (Splus software, MathSoft Inc., Cambridge, MA) was used. This procedure is described by Chambers and Hastie (1992). To test whether the values of these parameters were significantly different between genotypes, different models were compared. A simple model corresponds to a unique adjustment curve whatever the genotype, i.e. the parameter values are equal for all genotypes (Quilot et al., 2002). In a complex model, adjustment curves are different between genotypes so that the values of all the parameters are specific to each genotype. Lastly, in intermediate models some parameters are constants and others are specific to each genotype. The null hypothesis of no difference in the parameter values between the genotypes was tested by performing a χ2 test. In all cases, a threshold level of probability (α) of 0.05 was used. Determination of the relative importance of the genotypic key parameters: The relative importance of the genotypic key parameters with regard to the between-genotype variation was compared according to three criteria: the sensitivity of the model to the parameter, the variation in the parameter value observed in the population, and the mean error of estimation of the parameter. It is worth taking into account parameter variation from one genotype to another only for those which show high sensitivity. Moreover, the larger the variation in a parameter value within a population, the more likely the parameter is to explain large output variation observed between genotypes. Comparison between observed and predicted data: Two criteria were used to evaluate the model for each quality trait and for each genotype. First, the goodness-of-fit of the model was evaluated on the basis of data used for the parameterization, i.e. the StPBC202 dataset. Second, the predictive quality of the model was evaluated with independent data sets (StPBC201, GotBC201, and StP00 datasets). The adopted criterion was the root mean squared error (RMSE), a common criterion to quantify the mean difference between simulation and measurement in the case of non-linear models (Kobayashia and Us Salam, 2000). The global goodness-of-fit of the model was computed by averaging the relative RMSE (RRMSE) values of all genotypes (see Quilot et al., 2004a, for details). Usually, RRMSE values greater than 0.5 are considered not to be relevant and values lower than 0.25 as suitable. Spearman's rank correlation coefficients were also calculated with the ‘cor.test’ procedure of Splus. These coefficients compare the ranking of genotypes on the basis of observed and predicted values at maturity. Indeed, for use in breeding programmes, the ability of the model to rank the genotypes correctly is particularly important. Results Sensitivity analysis The relative variation in the model outputs caused by the variation in the values of each parameter were analysed (see Table SP1 in the supplementary data at JXB online). Thirteen of the 39 parameters for fruit load and an additional 12 parameters in the case of heavy fruit load were considered as important for further study. The 13 parameters were related to fruit growth demand, stone build-up, flesh mass increase, and total sugar accumulation. Among them, two were not estimated: Y, the threshold value of hydrostatic pressure needed for growth and πoc, the fruit osmotic pressure due to compounds other than soluble sugars. The 12 additional parameters (associated with heavy fruit load) were related to light interception, C assimilation, and fruit growth demand. Nine of them, p and k involved in the regulation of leaf photosynthesis by reserves, the ratio of leaf reserve mass to leafy shoot reserve mass, the growth respiration coefficient of fruit, and the parameters associated with leaf assimilation, radiation intensity in shade, and reserve mobilization (r2, GRCfruit, p4, p7, r3, r4, r6,) could not easily be measured for a large number of genotypes. Thus, 14 parameters of the 25 parameters selected based on the sensitivity analysis were measured for many or all genotypes studied. The values for the 11 remaining parameters that were not measured were fixed to values taken from the literature or from related experiments (Quilot et al., 2004a). Variation in parameter values between genotypes and parameterization A total of 23 parameters of the model were not measured and their values were taken from the literature or from related experiments (Quilot et al., 2004a). The origin of the parameter values used in the simulations is given in Table 1. As regards the 14 parameters selected based on the sensitivity analysis, the variation in their values between genotypes was analysed (Table 3). Table 3. Parameter values observed in the population Parameter  Value         Constanta  Number of genotypes  Experiments    Min   Max   Mean           r1          0.672  101  GarBC2, StPBC201 and StPBC202  SLAb  0.0143  0.0197  0.0169  (7.7)    49  GarBC2  p1          19.47  117  StPBC201 and StPBC202  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   0.0010  0.0050  0.0025  (36)    142  StPBC202  \(m_{\mathrm{flesh}}^{\mathrm{max}}\) c  1.5  52.9  13.1  (87)    18  StPBC202  \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   2.28  12.27  5.50  (32.5)    142  GarBC2, StPBC201 and StPBC202  kstone          0.583  142  GarBC2, StPBC201 and StPBC202  \(df_{\mathrm{stone}}^{1}\)   1.26  1.90  1.53  (7.8)    142  GarBC2, StPBC201 and StPBC202  \(df_{\mathrm{stone}}^{2}\)           0.0028  142  GarBC2, StPBC201 and StPBC202  \(df_{\mathrm{stone}}^{3}\)           354  142  GarBC2, StPBC201 and StPBC202  s1  4.77  5.51  5.09  (2.7)    142  GarBC2, StPBC201 and StPBC202  s2          0.6419  142  GarBC2, StPBC201 and StPBC202  aL  0.0021  0.0182  0.0071  (39.4)    142  StPBC202  ρ  197.16  752.4  349.6  (97.5)    41  GarBC2, StPBC202  rsu  −0.26  0.51  0.29  (27.9)    142  StPBC202  ksugar   0.0032   0.1210   0.0335   (49.5)     142   StPBC202   Parameter  Value         Constanta  Number of genotypes  Experiments    Min   Max   Mean           r1          0.672  101  GarBC2, StPBC201 and StPBC202  SLAb  0.0143  0.0197  0.0169  (7.7)    49  GarBC2  p1          19.47  117  StPBC201 and StPBC202  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   0.0010  0.0050  0.0025  (36)    142  StPBC202  \(m_{\mathrm{flesh}}^{\mathrm{max}}\) c  1.5  52.9  13.1  (87)    18  StPBC202  \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   2.28  12.27  5.50  (32.5)    142  GarBC2, StPBC201 and StPBC202  kstone          0.583  142  GarBC2, StPBC201 and StPBC202  \(df_{\mathrm{stone}}^{1}\)   1.26  1.90  1.53  (7.8)    142  GarBC2, StPBC201 and StPBC202  \(df_{\mathrm{stone}}^{2}\)           0.0028  142  GarBC2, StPBC201 and StPBC202  \(df_{\mathrm{stone}}^{3}\)           354  142  GarBC2, StPBC201 and StPBC202  s1  4.77  5.51  5.09  (2.7)    142  GarBC2, StPBC201 and StPBC202  s2          0.6419  142  GarBC2, StPBC201 and StPBC202  aL  0.0021  0.0182  0.0071  (39.4)    142  StPBC202  ρ  197.16  752.4  349.6  (97.5)    41  GarBC2, StPBC202  rsu  −0.26  0.51  0.29  (27.9)    142  StPBC202  ksugar   0.0032   0.1210   0.0335   (49.5)     142   StPBC202   A constant value is given when no difference was found between genotypes otherwise minimal, maximal, and mean values are given. The coefficient of variations of the parameter in the population is denoted between brackets. The number of genotypes used for the parameter estimation is indicated as well as the corresponding experimental designs. a When no significant difference was found between genotypes, a single value was estimated for all genotypes. b SLA was set to the mean value in case of light fruit load simulations (StPBC201 and StPBC202) and to the specific genotypic value in the case of heavy fruit load simulations (StP00, GothBC202). c Values of \(m_{\mathrm{flesh}}^{\mathrm{max}}\) only for the 18 genotypes with logistic flesh growth. For genotypes with exponential flesh growth, it was set to the arbitrary high value of 3000 since they do not display a plateau at maturity. View Large For two of them, r1 and p1, the variation between genotypes was not significant. Accordingly, for the leaf structural mass to leafy shoot structural mass ratio (r1), the general average (0.672) was used. The light-saturated leaf photosynthesis (p1) was set to the single value (19.47 μmol m−2 s−1) estimated for all genotypes. For the four parameters involved in allometric relationships (see Equations 10 and 14 of the Appendix in the supplementary data at JXB online), statistical models were compared. \(w_{\mathrm{stone}}^{\mathrm{matu}}\) and s1 were found to be genotype-dependent and kstone and s2 were constants. The specific leaf area (SLA) was measured for 49 genotypes and significant differences were found among these genotypes. Since SLA appeared to have no effect on the model outputs in the case of light fruit load (see Table SP1 of the Appendix in the supplementary data at JXB online), the averaged value (0.0169 m2 g−1) was used. For heavy fruit load simulations, the specific genotypic values were used (StP00 and GotBC202). For the other six parameters, \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) \(m_{\mathrm{flesh}}^{\mathrm{max}},\) ksugar, rsu, ρ, and aL, large differences were observed between genotypes and the range of the observed values was higher than the range considered in the sensitivity analysis. For the coefficient of the transfer function between sugars and other compounds, ksugar, the values estimated by Quilot et al. (2004b) were used on the same StPBC202 dataset. The estimation of the values of ρ, the permeation coefficient of fruit surface to water vapour, was very time-consuming and required many fruits. This parameter was estimated for 41 genotypes. A mean value was used for the others, although the 41 genotypes displayed high variation in ρ values. The model parameterized as described above was then used to estimate values of aL, the hydraulic conductance per unit of fruit surface, by calibration. Great variations in the values of aL, higher than those tested through the sensitivity analysis, were observed in the population. Since ρ, as well as aL, are involved in the computation of water accumulation in the fruit, the aL estimation probably contains the genotypic variation that was not included in the ρ values. Therefore, for the 41 genotypes with individual estimates of ρ, the effect on aL values of setting a constant value of ρ, instead of the specific ρ value, was computed. This accounted for 10% of variation in aL. This is not negligible, but it is small in comparison with genotypic variation in aL. Finally, four out of the 14 parameters measured or estimated were considered constant, whereas the other ten ( \(w_{\mathrm{stone}}^{\mathrm{matu}},\) \(df_{\mathrm{stone}}^{1},\) s1, \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) \(m_{\mathrm{flesh}}^{\mathrm{max}},\) ksugar, rsu, ρ, and aL for both fruit loads, and SLA for heavy fruit loads only) were considered as genotypic key parameters. Goodness-of-fit of the model on the basis of data used for parameterization The observed genotypic variation in dry and fresh fruit masses was well reproduced by the model (Fig. 1). The equation of fruit demand for growth appeared very robust as it reproduced the large range of growth patterns displayed by the population. The goodness-of-fit criteria (RRMSE) ranged from 0.030 to 0.376 for fruit dry matter growth and from 0.041 to 0.311 for fruit fresh growth depending on genotype (see Table SP2 of the Appendix in the supplementary data at JXB online). The observed data at maturity were also reproduced well for all the output variables (Fig. 2, line 1). For dry and fresh masses of fruit and stone, and flesh dry matter content, individual RRMSE ranged from 0 to 0.4 and mean RRMSE over the population from 0.06 to 0.1 (Table 4). The mean RRMSE of the two variables relating to total sugar were nearly equal to 0.1 and were satisfactory. However, individual RRMSE were highly variable between genotypes since they ranged from 0.002 to 0.6. Hence, for some genotypes total sugar amount and concentration were not well reproduced by the model. Fig. 1. View largeDownload slide Dry (A) and fresh (B) masses of fruits throughout fruit growth (days after bloom) for 9 BC2 contrasted genotypes. Filled circles denote the observed values of various fruits of one genotype (StPBC202 dataset). The line shows the mean value predicted by the model for each genotype. Fig. 1. View largeDownload slide Dry (A) and fresh (B) masses of fruits throughout fruit growth (days after bloom) for 9 BC2 contrasted genotypes. Filled circles denote the observed values of various fruits of one genotype (StPBC202 dataset). The line shows the mean value predicted by the model for each genotype. Fig. 2. View largeDownload slide Predicted values at maturity plotted against corresponding observed values for dry and fresh fruit masses, stone fresh mass, flesh dry matter content, total sugar amount as carbon in flesh, and total sugar concentration. First line of graphs (1) correspond to data from StPBC202, the second (2) to the StPBC201 dataset, the third (3) to the GothBC202, and last (4) to StP00. Accordingly, the first line of graphs was used to check the goodness-of-fit of the model and the other three lines to test its predictive quality. Each point of the graphs from lines 1 to 3 represent an averaged value for one genotype. On the graphs of line 4, open diamonds stand for S replications and filled diamonds for P1908 replications, both fruit loads mixed. The Spearman correlation coefficient is indicated on the upper left-hand corner of the corresponding plot. Fig. 2. View largeDownload slide Predicted values at maturity plotted against corresponding observed values for dry and fresh fruit masses, stone fresh mass, flesh dry matter content, total sugar amount as carbon in flesh, and total sugar concentration. First line of graphs (1) correspond to data from StPBC202, the second (2) to the StPBC201 dataset, the third (3) to the GothBC202, and last (4) to StP00. Accordingly, the first line of graphs was used to check the goodness-of-fit of the model and the other three lines to test its predictive quality. Each point of the graphs from lines 1 to 3 represent an averaged value for one genotype. On the graphs of line 4, open diamonds stand for S replications and filled diamonds for P1908 replications, both fruit loads mixed. The Spearman correlation coefficient is indicated on the upper left-hand corner of the corresponding plot. Table 4. Estimated values of the relative mean squared error (RRMSE) for evaluating the adjustment quality of the model and its predictive quality at maturity   Traits at maturity     Fruit     Stone     Flesh         Dataset     Dry mass (g)   Fresh mass (g)   Dry mass (g)   Fresh mass (g)   Dry matter content (g g−1)   Total sugar amount as carbon (g)   Total sugar concentration (g (100 g−1 FM))   Adjustment quality  StPBC202  Min  0.001  0  0  0  0  0.002  0.002      Max  0.300  0.242  0.308  0.400  0.215  0.602  0.356      Mean  0.059  0.063  0.090  0.099  0.060  0.098  0.114  Predictive quality  StPBC201  Min  –  0.005  –  0  0  0.003  0.004      Max  –  0.623  –  0.190  0.417  0.486  1.435      Mean  –  0.168  –  0.063  0.135  0.269  0.231    GotBC202  Min  0.159  0.153  0.102  0.074  0.074  0.313  0.190      Max  0.265  0.270  0.172  0.157  0.161  0.654  0.484      Mean  0.199  0.203  0.147  0.108  0.110  0.512  0.385    StP00  P1908  0.088  0.146  –  0.061  0.132  –  –      S  0.363  0.295  –  0.187  0.211  –  –      Mean   0.225   0.220   –   0.124   0.171   –   –     Traits at maturity     Fruit     Stone     Flesh         Dataset     Dry mass (g)   Fresh mass (g)   Dry mass (g)   Fresh mass (g)   Dry matter content (g g−1)   Total sugar amount as carbon (g)   Total sugar concentration (g (100 g−1 FM))   Adjustment quality  StPBC202  Min  0.001  0  0  0  0  0.002  0.002      Max  0.300  0.242  0.308  0.400  0.215  0.602  0.356      Mean  0.059  0.063  0.090  0.099  0.060  0.098  0.114  Predictive quality  StPBC201  Min  –  0.005  –  0  0  0.003  0.004      Max  –  0.623  –  0.190  0.417  0.486  1.435      Mean  –  0.168  –  0.063  0.135  0.269  0.231    GotBC202  Min  0.159  0.153  0.102  0.074  0.074  0.313  0.190      Max  0.265  0.270  0.172  0.157  0.161  0.654  0.484      Mean  0.199  0.203  0.147  0.108  0.110  0.512  0.385    StP00  P1908  0.088  0.146  –  0.061  0.132  –  –      S  0.363  0.295  –  0.187  0.211  –  –      Mean   0.225   0.220   –   0.124   0.171   –   –   Minimal, maximal and mean values of RRMSE are presented for each experiment and each output variable. View Large The model performed very well in ranking genotypes for dry and fresh fruit masses, fresh stone mass, and total sugar amount in the flesh, with the Spearman correlation coefficient ranging from 0.91 to 0.98 (Fig. 2). For flesh dry matter content and total flesh sugar concentration, the Spearman correlation coefficient reached 0.8 and 0.73, respectively, still a good performance. Predictive quality of the model for independent data The model was used to predict fruit observations from independent data under various fruit load conditions. Observations from 87 BC2 genotypes grown under light fruit load (StPBC201 dataset), from seven BC2 genotypes grown under heavy fruit load (GotBC202 dataset) and from S and P1908 for two contrasted fruit loads (StP00 dataset) were used to test the ability of the model to predict fruit quality. The model correctly predicted growth kinetics of dry and fresh (see Figures SP1 and SP2 of the Appendix in the supplementary data at JXB online) fruit masses for different environmental and growing conditions (year, site, light and heavy fruit loads). Mean RRMSE for growth of dry and fresh fruit varied between 0.17 and 0.3 depending on the experiment (see Table SP2 of the Appendix in the supplementary data at JXB online). The ability of the model to rank the genotypes at maturity appeared accurate for both dry and fresh fruit masses, with the Spearman correlation coefficients from 0.62 to 0.94 (Fig. 2, lines 2, 3, and 4). However, it was less accurate when leafy shoot growth was to be considered (Fig. 2, line 3). In addition to its ability to reproduce variation between genotypes, the model was able to reproduce variations between fruits of the same tree (see Fig. SP1 of the Appendix in the supplementary data at JXB online). The model predictions appeared especially good for dry and fresh stone masses. Mean RRMSE at maturity ranged from 0.063 to 0.147 for these two variables, and individual RRMSE values never exceeded 0.19 (Table 4). Similarly, the Spearman correlation coefficient varied between 0.75 and 0.98 depending on the experiment (Fig. 2, lines 2, 3, and 4). Predictions of flesh dry matter content were good as indicated by the RRMSE criteria, but according to the Spearman correlation coefficient were moderately accurate. Mean RRMSE did not exceed 0.17 (Table 4), whereas the Spearman correlation coefficient varied between 0.49 and 0.83 (Fig. 2, lines 2, 3, and 4). Considering total sugar variables, predictions were globally less accurate, but the accuracy depended on the experiment and the genotype. These variables appeared to be predicted better in the case of light fruit load. Mean RRMSE varied between 0.23 and 0.51 (Table 4) and the Spearman correlation coefficient ranged from 0.22 to 0.78 (Fig. 2, lines 2 and 3). Determination of the relative importance of the ten genotypic key parameters The relative importance of the 10 genotypic key parameters was compared with respect to the three criteria: the sensitivity of the model to the parameter, the variation in the parameter value observed in the population, and the mean error of estimation of the parameter (Table 5). Table 5. Key genotypic parameters of the model are ordered from the least to the most noteworthy, according to three criteria: the sensitivity of the model to the parameter, the variation of the parameter value observed in the population, and the error of estimation of the parameter Parameter  Sensitivitya  Averaged value of the parameter in the population  Variation of the parameter value in the population       Averaged value of the estimation error in the population     Global scoree        Standard deviation   Coefficient of variationb           SLA  *  0.0169  0.0013  7.69  *c  0.0005  **d  4  \(df_{\mathrm{stone}}^{1}\)   *  1.53  0.12  7.84  *  0.058  **  4  s1  ****  5.09  0.14  2.75  *  0.051  **  7  ρ  ***  349.6  97.5  27.89  **  40.3  **  7  rsu  **  0.29  0.14  48.28  ***  0.053  **  7  \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   **  5.50  1.79  32.55  ***  0.391  ***  8  ksugar  ***  0.0335  0.0166  49.55  ***  0.00693  **  8  \(m_{\mathrm{flesh}}^{\mathrm{max}}\) f  **  13.1  11.45  87.4  ****  5.45  **  8  aL  ****  0.0071  0.0028  39.44  ***  0.00048  ***  10  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   ****   0.0025   0.0009   36   ***   0.000087   ****   11   Parameter  Sensitivitya  Averaged value of the parameter in the population  Variation of the parameter value in the population       Averaged value of the estimation error in the population     Global scoree        Standard deviation   Coefficient of variationb           SLA  *  0.0169  0.0013  7.69  *c  0.0005  **d  4  \(df_{\mathrm{stone}}^{1}\)   *  1.53  0.12  7.84  *  0.058  **  4  s1  ****  5.09  0.14  2.75  *  0.051  **  7  ρ  ***  349.6  97.5  27.89  **  40.3  **  7  rsu  **  0.29  0.14  48.28  ***  0.053  **  7  \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   **  5.50  1.79  32.55  ***  0.391  ***  8  ksugar  ***  0.0335  0.0166  49.55  ***  0.00693  **  8  \(m_{\mathrm{flesh}}^{\mathrm{max}}\) f  **  13.1  11.45  87.4  ****  5.45  **  8  aL  ****  0.0071  0.0028  39.44  ***  0.00048  ***  10  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   ****   0.0025   0.0009   36   ***   0.000087   ****   11   a Qualitative score (deduced from Table SP1of the Appendix in the supplementary data at JXB online) in the case of heavy fruit load. The more sensitive the model is to the parameter, the more this parameter gets stars. b Ratio between the standard deviation and the averaged value of the parameter in the population in percent. c Qualitative score deduced from the coefficient of variation. The smaller, the more stars. d Qualitative score deduced from the comparison between the averaged value of the estimation error of a parameter and the standard deviation of this parameter in the population. The smaller the former compared with the latter, the more stars. e Global score of the parameters considering the qualitative scores for the three criteria analysed. The score is the total number of stars. f Only the values of \(m_{\mathrm{flesh}}^{\mathrm{max}}\) for the 18 genotypes with logistic flesh growth were considered. View Large Thus, rules were established taking into account these aspects in order to define the score of each parameter and to rank them. A qualitative score was defined for each criterion attributing different numbers of stars to different levels of the criteria (Table 5). A global score was then defined as the total number of stars. The parameters SLA and \(df_{\mathrm{stone}}^{1}\) had only four stars, and were considered the least important among the ten. aL and \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) were considered the most important genotypic parameters of the model. Discussion Simulation of genotypic variation in fruit quality An ecophysiological model was used to reproduce variation between genotypes, between fruit loads, and among fruits of the same tree for different years (2001 and 2002), sites (St Paul and Gotheron), and growing conditions (orchards or pots). This study confirmed that the ecophysiological model used was efficient in simulating genotypic variation under various growing conditions. However, it was less accurate when fruit load was heavy. Importance of the 11 parameters not measured Eleven of the parameters that largely influenced the outputs were not measured for a large number of genotypes. Two parameters, involved in the regulation of leaf photosynthesis through leaf reserves (p, k), were investigated separately in a study of six highly contrasted genotypes (Quilot et al., 2004a). Their values appeared to be equal for all genotypes. Two others, p4 and GRCfruit, involved in leaf assimilation and fruit growth respiration, were unlikely to influence the model outputs. Indeed, Quilot et al. (2002) suggested, on the basis of data from the literature, that variation in these parameters within closely related species were small. As far as is known, no data are available on variation in r2, p7, r3, r4, and r6 within a species or between two related species. However, it is assumed that these five parameters barely affect model outputs, since they influence sources that only influence growth in dry mass and for which predictions were accurate even under heavy fruit load conditions. Variation in Y and πoc was more likely to create output variation. Indeed, the wall yielding threshold pressure (Y), was shown to range from 0.3 MPa to 0.8 MPa for various fruit cells belonging to distant species (Green and Cummins, 1974). The sensitivity analysis tested this range and showed high variation in all outputs at both fruit loads. Although variation between closely related species and their hybrids may be lower than interspecific variations, further information is needed on the variation of this parameter. Similarly, variation in the osmotic pressure induced by compounds other than sugars (πoc), may induce high variation in the output values. As far as is known, no data are available on the variation in πoc. Nevertheless, πoc is likely to display large variation between genotypes because of qualitative and quantitative variation in soluble cellular compounds (organic acids, amino acids, potassium …). Peach genotypes are known to vary widely in organic acid concentrations (Wu et al., 2003). However, it seems that organic acids account for less than half of the osmotic pressure induced by osmotically active compounds other than sugars. Little is known on the influence of amino acids and minerals on the osmotic pressure in peach fruit (Moing et al., 1998, 2003; Lobit et al., 2002). Finally, variation in Y and πoc between genotypes are probably not negligible. Ignoring this variation may have resulted in a poor estimation of aL. The major physiological processes responsible for variation in fruit quality The use of an ecophysiological model and the identification of genotypic key parameters highlighted the major physiological processes responsible for genotypic variation in different fruit quality criteria. The description of fruit quality under non-limiting fruit growth conditions revealed large variation between genotypes, suggesting that fruit growth demand was one of the main processes responsible for fruit mass variation in particular. Moreover, this process also influenced total sugar concentrations. Accordingly, \(m_{\mathrm{flesh}}^{\mathrm{max}}\) and especially \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) were identified as the major parameters according to different criteria. aL, associated with water flux in the fruit, also appeared to be a major parameter. It is not only involved in fruit fresh matter growth, but also in the dilution of dry matter and sugars. Therefore, it influences fruit fresh mass, dry matter content, and total sugar concentration. Another major parameter, ksugar, was identified which is linked to sugar metabolism and appears to be of particular importance in describing the variation in dry matter content and total sugar concentration. The partitioning between sucrose and hexose sugars studied through rsu as well as the fruit transpiration (influenced by ρ) were shown to contribute to genotypic variation in fruit quality as well, although to a lower extent than the four previous processes. Lastly, the stone mass, described through \(w_{\mathrm{stone}}^{\mathrm{matu}},\) was highly variable between genotypes and not negligible: an increase in \(w_{\mathrm{stone}}^{\mathrm{matu}}\) resulted in a decrease in fruit fresh mass. Potential uses of this approach Different prospects can be envisaged for this approach. Indeed, the identification of both genotypic key parameters and main physiological processes involved in quality variation should be useful from a genetic point of view. These results point out the parameters that deserve further study. They could guide research towards these key processes and orientate breeding programmes, in the way suggested by Jackson et al. (1996). Further analysis of the genotypic parameters should be performed in order to determine their genetic control. Such a study has been performed by Quilot et al. (2005) on the genotypic key parameters identified in this paper. Model simulations may also be helpful in understanding fruit biology. The model may be used to study the links between the different processes, particularly when processes have opposite effects on a quality trait. The model can also be used to describe the genotypic variation in fruit quality under different climatic or fruit load conditions. We gratefully acknowledge J Hostalery for her assistance in the field experiments, J Besset for the experiments performed at the Gotheron site, and T Pascal for his advice on tree management. We thank E Rubio and L Gomez for sugar analyses. We are grateful to C Borel for critically revising the manuscript. We thank A Lacombe and Dr O Savolainen for improving the English. This research was funded in part by grants from the Ministère de la Recherche (PhD grant), from Région Provence-Alpes-Côte d'Azur (projects DEB 02-252 and DEB 03-543) and from the Institut National de la Recherche Agronomique, France (A.I.P. PFI P00232 and A.I.P. REA P00251). References Ben Mimoun M. 1997. Vers la maîtrise de la variabilité des fruits au sein de l'arbre: Etude et modélization de la croissance des pêches (Prunus persica) à l'échelle du rameau. PhD thesis, INAPG, France. Google Scholar Boote KJ, Kropff MJ, Bindraban PS. 2001. Physiology and modelling of traits in crop plants: implications for genetic improvement. Agricultural Systems  70, 395–420. Google Scholar Chambers JM, Hastie TJ. 1992. Statistical models. In: Brook-Cole W, ed. Computer science series. California: Pacific Grove. Google Scholar DeJong TM, Doyle JF, Day KR. 1987. Seasonal patterns of reproductive and vegetative sink activity in early and late maturing peach (Prunus persica) cultivars. Physiologia Plantarum  71, 83–88. Google Scholar DeJong TM, Goudriaan J. 1989. Modeling peach fruit growth and carbohydrate requirements: re-evaluation of the double-sigmoid growth pattern. Journal of the American Society for Horticultural Science  114, 800–804. Google Scholar Fishman S, Génard M. 1998. A biophysical model of fruit growth: simulation of seasonal and diurnal dynamics of mass. Plant, Cell and Environment  21, 739–752. Google Scholar Génard M, Baret F. 1994. Spatial and temporal variation of light inside peach trees. Journal of the American Society for Horticultural Science  119, 669–677. Google Scholar Gomez L, Rubio E, Augé M. 2002. A new procedure for extraction and measurement of soluble sugars in ligneous plants. Journal of the Science of Food and Agriculture  82, 360–369. Google Scholar Green PB, Cummins WR. 1974. Growth rate and turgor pressure: auxin effect studied with an automated apparatus for single coleoptiles. Plant Physiology  54, 863–869. Google Scholar Grossman YL, DeJong TM. 1994 a. PEACH: a simulation model of reproductive and vegetative growth in peach trees. Tree Physiology  14, 329–345. Google Scholar Grossman YL, DeJong TM. 1994 b. Carbohydrate requirements for dark respiration by peach vegetative organs. Tree Physiology  14, 37–48. Google Scholar Higgins SS, Larsen FE, Bendel RB, Radamaker GK, Bassman JH, Bidlake WR, Al Wir A. 1992. Comparative gas exchange characteristics of potted, glasshouse-grown almond, apple, fig, grape, olive, peach, and Asian pear. Scientia Horticulturae  52, 313–329. Google Scholar Jackson P, Robertson M, Cooper M, Hammer GL. 1996. The role of physiological understanding in plant breeding; from a breeding perspective. Field Crops Research  49, 11–37. Google Scholar Kobayashia K, Us Salam M. 2000. Comparing simulated and measured values using mean squared deviation and its components. Agronomy Journal  92, 345–352. Google Scholar Lescourret F, Ben Mimoun M, Génard M. 1998. A simulation model of growth at the shoot-bearing-fruit level. I Description and parameterization for peach. European Journal of Agronomy  9, 173–188. Google Scholar Lobit P, Soing P, Génard M, Habib R. 2002. Theoretical analysis of relationships between composition, pH, and titratable acidity of peach fruit. Journal of Plant Nutrition  25, 2775–2792. Google Scholar Moing A, Poëssel JL, Svanella-Dumas L, Loonis M, Kervella J. 2003. Biochemical basis of low fruit quality of Prunus davidiana, a pest and disease resistance donor for peach breeding. Journal of the American Society for Horticultural Science  128, 55–62. Google Scholar Moing A, Svanella L, Rolin D, Gaudillère M, Gaudillère JP, Monet R. 1998. Compositional changes during the fruit development of two peach cultivars differing in juice acidity. Journal of the American Society for Horticultural Science  123, 770–775. Google Scholar Pascal T, Kervella J, Pfeiffer F, Sauge MH, Esmenjaud D. 1998. Evaluation of the interspecific progeny Prunus persica cv. Summergrand. Acta Horticulturae  465, 185–192. Google Scholar Quilot B. 2003. Vers une démarche de sélection variétale assistée par des modèles écophysiologiques et génétiques. Cas de la qualité organoleptique chez la pêche (Prunus persica (L.) Batsch). PhD thesis, INAPG, France. Google Scholar Quilot B, Génard M, Kervella J. 2004 a. Leaf light saturated photosynthesis for wild and cultivated peach genotypes and their hybrids: a simple mathematical modelling analysis. Journal of Horticultural Science and Biotechnology  79, 546–553. Google Scholar Quilot B, Génard M, Kervella J, Lescourret F. 2002. Ecophysiological analysis of genotype variation in peach fruit growth. Journal of Experimental Botany  53, 1613–1625. Google Scholar Quilot B, Génard M, Kervella J, Lescourret F. 2004 b. Analysis of genotypic variations in flesh total sugar content via a modelling approach applied to peach fruit. Theoretical and Applied Genetics  109, 440–449. Google Scholar Quilot B, Kervella J, Génard M, Lescourret F. 2005. Analysing the genetic control of peach fruit quality through an ecophysiological model combined with a QTL approach. Journal of Experimental Botany  56, (in press). Google Scholar Tardieu F. 2003. Virtual plants: modelling as a tool for the genomics of tolerance to water deficit. Trends in Plant Science  8, 9–14. Google Scholar Wu B, Quilot B, Kervella J, Génard M, Li S. 2003. Analysis of genotypic variation of sugar and acid contents in peaches and nectarines through the Principle Component Analysis. Euphytica  132, 375–384. Google Scholar © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]

Quilot, B.;Kervella, J.;Génard, M.;Lescourret, F.

doi: 10.1093/jxb/eri305pmid: 16234283

Abstract Ecophysiological models are increasingly expected to include genetic information via genotype-dependent parameters. These parameters could be considered as quantitative traits and submitted to analysis. A pre-existing ecophysiological model of fruit quality was used and the distribution of the genotypic parameters in a second backcross population derived from a clone of a wild peach (Prunus davidiana) and commercial nectarine varieties (P. persica (L.) Batsch) was analysed. The correlations between the two years of experimentation were higher for the genotypic parameters than for the quality traits commonly studied by breeders. The correlations between the genotypic parameters and the quality traits were low. Quantitative trait loci (QTLs) for the genotypic key parameters of the ecophysiological model were detected by linear regression. Co-locations of QTLs for parameters were observed as well as co-locations of QTLs for parameters and quality traits. The ecophysiological model and the results of the QTL analysis were combined by substituting each parameter in the model by the sum of QTL effects. This combined model can simulate the behaviour of genotypes carrying diverse combinations of alleles. The quality of this combined model was moderately suitable, but had some shortcomings. Improvements are suggested and further use of this combined model as a tool for breeders is discussed. Ecophysiology, fruit quality, genotypic variation, modelling, peach, QTL Introduction Fruit breeders must satisfy two requests concurrently: the production of high quality fruits and the use of sustainable practices. Wild germplasm is commonly used as a source of resistance to pests and diseases, but its use is limited because it is of low agronomic value. First, it is difficult to achieve the required agronomic improvement because selection is on quantitative traits, such as fruit mass or flesh sugar concentration, which result from several linked processes, such as carbon assimilation or fruit sink strength. Second, it is difficult to select for traits that are sensitive to environmental factors. QTLs controlling these traits often show low stability (Veldboom and Lee, 1996). To overcome these difficulties, an interdisciplinary approach has been developed by ecophysiological modellers and geneticists (Shorter et al., 1991; Boote et al., 1996; Hammer et al., 1996). Molecular markers make it possible to carry out QTL analyses, which study the genetic variation of a character, locate the genes responsible for this variation, and quantify their effects and interactions. It is then possible to predict the behaviour of genotypes with any given combination of alleles, but only under environmental conditions similar to those where the QTLs were detected. Conversely, an ecophysiological model predicts the behaviour of one genotype in many environments. It decomposes the development of a trait into various processes subjected to environmental factors, with model parameters independent of the environment. An interdisciplinary approach consists of including genetic information in ecophysiological models via genotype-dependent parameters. These parameters could be considered as quantitative traits and characterize a genotype. Such an approach was applied to peach (Prunus persica) fruit quality because it results from many controlled processes and because it is highly sensitive to environment. Indeed, few QTLs associated with organoleptic fruit quality have been mapped (Abbott et al., 1998; Quarta et al., 1998) and genes controlling organoleptic fruit quality often remain unknown (Saliba-Colombani et al., 2001; Etienne et al., 2002). Microclimatic gradients (Corelli-Grappadelli and Coston, 1991; Marini et al., 1991), leaf area near the fruit (Kliewer and Weaver, 1971; Génard, 1992) and vegetative vigour of shoots bearing fruit (Génard and Bruchou, 1992) may cause within-plant variation in quality. The present study was carried out on a population of genotypes derived from a clone of a wild peach (P. davidiana) by three generations of crosses with commercial nectarine varieties. The ecophysiological model used was described by Quilot et al. (2005) who identified genotypic key parameters of the model. These parameters can be analysed with QTL methods. First, they are estimated for numerous genotypes of the population. Second, they are highly variable from one genotype to another, and mostly independent of the environment. Lastly, they appeared to explain much of the variation in fruit quality in the population. The distribution of the genotypic key parameters in the population and their stability through two years of experimentation was analysed. The correlations between these parameters and the quality traits commonly studied by breeders was also studied. A QTL analysis of the genotypic key parameters (QTL model) was then performed. An attempt has been made to explain the co-locations of QTLs for parameters and quality traits in order to interpret the functions of the QTLs detected. The QTL model was used to predict, for each genotype of the studied population, the values of each genotypic key parameter of the ecophysiological model. The goodness-of-fit of this combination of models was tested. Finally, the importance of such an approach for selection and for biological understanding was discussed. Materials and methods Description of the ecophysiological model Our ecophysiological model simulates carbon assimilation, its partitioning at the ‘shoot-bearing fruit’ level, water flux, and sugar accumulation in the flesh during fruit growth, under the influence of environmental factors. Its mathematical formulation and the definition of its parameters have been described previously (Quilot et al., 2005). The outputs relevant for this study are dry and fresh fruit masses, stone fresh mass, dry matter content, and total sugar concentration in the flesh. In addition to this ecophysiological model that is only concerned with fruit growth after the end of the stage of active cell division, the early growth of fruit, during which cells divide, was considered in an empirical way. The fruit size at the end of cell division is an indicator of fruit sink size and, consequently, of its potential expansion (Scorza et al., 1991). Cell division was reported to stop around 50–80 d after bloom (DAB), (Ognjanov et al., 1995; Yamaguchi et al., 2002), depending on the variety. Accordingly, it was assumed that cell division was fully completed at 590 degree-days (dd), which closely corresponds to 80 DAB. Early fruit growth was only considered after 321 dd, as it is not possible to measure diameters without causing fruit damage. Early fruit dry matter growth between 321 and 590 dd was roughly described by a linear function of degree-days after bloom (dd):  \[W_{\mathrm{fruit}}^{\mathrm{early}}(\mathrm{dd}){=}W_{\mathrm{fruit}}^{321}{+}GR_{\mathrm{fruit}}^{\mathrm{early}}{\times}(\mathrm{dd}{-}321)\] where \(W_{\mathrm{fruit}}^{321}\) corresponds to fruit dry mass at 321 dd and \(GR_{\mathrm{fruit}}^{\mathrm{early}}\) is the fruit early growth rate (g dd−1). The initial fruit dry mass, \(W_{\mathrm{fruit}}^{\mathrm{ini}}{=}W_{\mathrm{fruit}}^{\mathrm{early}}(590),\) input for the ecophysiological model, is computed from the early growth model with dd=590. The genotypic key parameters of the ecophysiological model When fruit loads were light, nine of the 40 parameters of the model were identified as genotypic key parameters by Quilot et al. (2005). These parameters satisfied three main conditions: the model was sensitive to their variation with respect to potential fruit growth; they varied widely in the population, and their value was accurately estimated. However, the parameter involved in fruit growth limitation close to maturity ( \(m_{\mathrm{flesh}}^{\mathrm{max}}\) ) was estimated for only 18 genotypes. Since to analyse genetic variation of a trait is not reliable on so few genotypes, this parameter was not considered in the following study. Consequently, eight genotypic key parameters were studied further. In addition to these eight parameters, the initial fruit dry mass at 590 dd ( \(W_{\mathrm{fruit}}^{\mathrm{ini}},\) an initial state value of the model) and growth duration from full bloom to maturity (ddmax), were important in this study based on model sensitivity and variability in the population. By extension, they were dealt with as parameters. Two parameters of the early growth model, \(W_{\mathrm{fruit}}^{321}\) and \(GR_{\mathrm{fruit}}^{\mathrm{early}},\) were also considered as possible genotypic key parameters. A description of these 12 parameters is given in Table 1. Table 1. Symbols, definitions and units of the parameters in the QTL analysis Parameter  Definition  Unit  Number of genotypes observeda             2001   2001/2002   2002   ddmax  Growth duration from full bloom to maturity  Degree-days  87    136  \(W_{\mathrm{fruit}}^{321}\)   Fruit dry mass at 321 dd  g  87    136  \(GR_{\mathrm{fruit}}^{\mathrm{early}}\)   Dry fruit mass growth rate between 321 and 590 dd  g degree-days−1  87    136  \(W_{\mathrm{fruit}}^{\mathrm{ini}}\)   Initial fruit dry mass at 590 dd  g  87    136  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   Initial relative dry flesh mass growth rate  Degree-days−1  87    136  \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   Potential maximal stone dry mass at maturity  g    149    \(df_{\mathrm{stone}}^{1}\)   Concerns the allometric equation relating stone fresh mass to stone dry mass  Dimensionless    155    ksugar  Coefficient of the transfer function between sugars and compounds other than sugars  Day−1  87    134  rsu  Concerns the calculation along growth of the proportion of sucrose in the total amount of sugar in the flesh  Dimensionless    154    ρ  Permeation coefficient of the fruit surface to water vapour  cm h−1    41    s1  Concerns the allometric equation relating fruit area to fruit fresh mass  Dimensionless    149    aL   Hydraulic conductance per unit of fruit surface   g cm−2 bar−1 h−1   87     134   Parameter  Definition  Unit  Number of genotypes observeda             2001   2001/2002   2002   ddmax  Growth duration from full bloom to maturity  Degree-days  87    136  \(W_{\mathrm{fruit}}^{321}\)   Fruit dry mass at 321 dd  g  87    136  \(GR_{\mathrm{fruit}}^{\mathrm{early}}\)   Dry fruit mass growth rate between 321 and 590 dd  g degree-days−1  87    136  \(W_{\mathrm{fruit}}^{\mathrm{ini}}\)   Initial fruit dry mass at 590 dd  g  87    136  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   Initial relative dry flesh mass growth rate  Degree-days−1  87    136  \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   Potential maximal stone dry mass at maturity  g    149    \(df_{\mathrm{stone}}^{1}\)   Concerns the allometric equation relating stone fresh mass to stone dry mass  Dimensionless    155    ksugar  Coefficient of the transfer function between sugars and compounds other than sugars  Day−1  87    134  rsu  Concerns the calculation along growth of the proportion of sucrose in the total amount of sugar in the flesh  Dimensionless    154    ρ  Permeation coefficient of the fruit surface to water vapour  cm h−1    41    s1  Concerns the allometric equation relating fruit area to fruit fresh mass  Dimensionless    149    aL   Hydraulic conductance per unit of fruit surface   g cm−2 bar−1 h−1   87     134   The parameter values were estimated either separately in 2001 and 2002 or jointly, depending on the parameters. a Number of BC2 genotypes for which the parameter values were estimated in 2001 and in 2002 or jointly in both years. View Large Plant material The breeding population is a second backross progeny derived from clone P1908 of Prunus davidiana as follows (Pascal et al., 1998). Initially, P1908 with small green fruit, was crossed with P. persica ‘Summergrand’ (S) and an F1 progeny was obtained. One F1 hybrid resistant to powdery mildew was then back-crossed to S to produce a BC1 progeny. Finally, BC1 individuals were used to pollinate P. persica ‘Zéphir’ (Z) to derive the breeding population (BC2). S and Z are, respectively, yellow and white nectarine cultivars with large tasty fruits. The study was conducted at the INRA Research Centre of Avignon (France). BC2 genotypes and the three parents were planted in a completely randomized design with one tree per genotype. Trees were 3 years old in 2001. All genotypes were grafted on GF305 seedling rootstocks and were grown under optimal conditions of irrigation, fertilization, and pest control. Experiments Experimental observations were carried out in 2002 on 139 genotypes of BC2 and on S, Z, and P1908 (BC202 dataset) and in 2001 on 87 genotypes of the BC2 population common to both years and S and Z (BC201 dataset). A very light fruit load was left on each tree (only five fruits per tree) to ensure that all fruits were under non-limiting source conditions (i.e. under maximum growth conditions). However, these non-limiting source conditions appeared to be hardly met for numerous genotypes in 2001. Diametric fruit growth was monitored from fruitlet thinning to maturity. At maturity, dry and fresh fruit and stone masses were measured. The total amount of sugar (gC) and total flesh sugar concentration were also determined. Details on these measurements have been described by Quilot et al. (2005). These data were used by Quilot et al. (2004) to detect QTLs for quality traits commonly studied by breeders. These data (BC202 dataset) were also used to estimate the values of the ecophysiological model parameters (Quilot et al., 2005) and the values of the two parameters, \(W_{\mathrm{fruit}}^{321}\) and \(GR_{\mathrm{fruit}}^{\mathrm{early}},\) of the early growth model. QTL analysis The interspecific map for BC2 progenies developed by Foulongne et al. (2003) and complemented by Quilot et al. (2004) was used. QTL detection was performed using a forward multiple linear regression of the phenotypic values of the genotype at each of the molecular markers, with Splus (Splus software, MathSoft Inc., Cambridge, MA). The most likely QTL position corresponded to the locus with the strongest association with the trait. A threshold of significance of 5% was chosen to declare a putative QTL. This method was described by Quilot et al. (2004) to detect QTLs for quality traits. QTL detection was carried out for the 12 parameters and for fruit dry mass. Combination of ecophysiological and QTL models The approach consists of introducing, in the ecophysiological model, the values estimated from the QTL model instead of the measured values of the parameters. The QTL model takes into account both the origin of the allele at a detected locus and the effect of the alleles at this locus on the parameter value. With a marker from P1908, the effect of the P1908 allele presence (scored 1) is determined by comparison with the presence of an allele coming from S (scored 0). With a marker of S (Z) genome, the effect of a S (Z) allele (scored 1) is determined by comparison with the presence of the other S (Z) allele (scored 0). The effects of the allele scored 0 are set to 0. The parameter value is estimated as the sum of the allele effects, either positive, negative or null, added to an intercept, μ. The intercept corresponds to the parameter value when the genotype only possesses the alleles set to 0. The epistatic effects between two loci were added in the same way to the QTL model. Accordingly, the value of a parameter X for which N QTLs and M epistatic interactions were detected is estimated for an individual i by:  \[X_{\mathrm{i}}{=}\mathrm{{\mu}}{+}{{\sum}_{\mathrm{n}{=}1}^{\mathrm{N}}}a_{\mathrm{n}}{\times}G_{\mathrm{i},{\,}\mathrm{n}}{+}{{\sum}_{\mathrm{m}{=}1}^{\mathrm{M}}}e_{\mathrm{m}}{\times}G_{\mathrm{i},{\,}\mathrm{m}}\] where an corresponds to the additive effect of the QTL, n and em to the effect of the epistatic interaction m. Gi,n and Gi,m are genetic QTL scores of the individual i that take the value 0 or 1 depending, respectively, on the allele of the corresponding QTL n and on the combination of alleles of the loci involved in the epistatic interaction m. Statistical analysis Most of the parameter values have been estimated by Quilot et al. (2005) on the BC202 dataset. However, the hydraulic conductance per unit of fruit surface (aL) was estimated again, setting the value of the permeation coefficient of the fruit surface to water vapour (ρ) constant to the mean observed value for all genotypes, in order to avoid distortions between genotypes. Indeed, ρ was estimated for only 41 genotypes of the population and the estimated aL value may depend on the ρ value. The parameter values from BC201 dataset were estimated as described by Quilot et al. (2005). Goodness-of-fit of the combined model for each genotype was evaluated using the root mean squared error (RMSE), a common criterion to quantify the mean difference between simulation and measurement in the case of non-linear models (Kobayashi and Us Salam, 2000). The global goodness-of-fit of the model was computed by averaging the relative RMSE (RRMSE) values of all genotypes (see Quilot et al., 2004a, for details). All data analyses were performed with the Splus software. Results Distribution of the key parameter values estimated on the BC202 dataset The distributions of the 12 key parameter values were very similar for the two years. The parameter values of S (‘Summergrand’) and Z (‘Zéphir’) were nearly identical for seven parameters and only slightly different for five parameters including the growth duration ddmax, the coefficient of the transfer function between sugars and other compounds ksugar, and the hydraulic conductance per unit of fruit surface aL (Fig. 1). By contrast, values of P1908 were clearly different from those of S and Z for five parameters. They were greater for ksugar and s1, and lower for the three fruit growth parameters \(GR_{\mathrm{fruit}}^{\mathrm{early}},W_{\mathrm{fruit}}^{\mathrm{ini}},\) and \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}.\) Fig. 1. View largeDownload slide Distribution of the 12 key parameters of the ecophysiological model estimated on the 2002 dataset. The values of the parents ‘Summergrand’ (S), ‘Zéphir’ (Z), and P. davidiana (D) are indicated by arrows. Fig. 1. View largeDownload slide Distribution of the 12 key parameters of the ecophysiological model estimated on the 2002 dataset. The values of the parents ‘Summergrand’ (S), ‘Zéphir’ (Z), and P. davidiana (D) are indicated by arrows. The population exhibited considerable genotypic variation in parameters. Most of the parameters were nearly normally distributed, apart from ddmax for which the distribution was bimodal (Fig. 1). Transgressive segregants were observed for high and/or low levels of all parameters. For example, transgressive segregants were very frequent for high levels of growth duration, ddmax, since most of the genotypes showed a value higher than the values of the three parents. Trangressions for high levels were also observed for fruit growth parameters ( \(W_{\mathrm{fruit}}^{321},GR_{\mathrm{fruit}}^{\mathrm{early}},W_{\mathrm{fruit}}^{\mathrm{ini}},\) and \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) ), the parameter concerning the calculation of the sucrose to total sugar ratio, rsu, and aL and \(df_{\mathrm{stone}}^{1}.\) For \(w_{\mathrm{stone}}^{\mathrm{matu}},\) transgressive segregants towards low values were observed. Conversely, for ksugar and s1 none of the genotypes in the population showed higher values than the parents. Stability of the trait and the key parameter values between 2001 and 2002 Seven ( \(W_{\mathrm{fruit}}^{321},\) \(GR_{\mathrm{fruit}}^{\mathrm{early}},\) \(W_{\mathrm{fruit}}^{\mathrm{ini}},\) ddmax, \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) ksugar, and aL) of the 12 key parameters were estimated separately from 2001 and 2002 data. The correlations between 2001 and 2002 values were highly significant for all the key parameters (Table 2) and were higher overall than for quality traits. The stone fresh mass was the most stable trait over years. The highest correlations between years for the parameters were observed for growth duration (ddmax) and the two parameters of dry matter growth rate, \(GR_{\mathrm{fruit}}^{\mathrm{early}}\) and \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}.\) The sugar concentration in the flesh and the parameter related to sugar metabolism, ksugar, showed least stability. Table 2. Correlation coefficients between 2001 and 2002 fruit traits at maturity and parameter values for the 87 genotypes common to the two years Fruit trait     Parameter     Fruit dry mass  0.52  ddmax  0.96  Fruit fresh mass  0.47  \(W_{\mathrm{fruit}}^{321}\)   0.52  Stone fresh mass  0.60  \(GR_{\mathrm{fruit}}^{\mathrm{early}}\)   0.81  Flesh dry matter content  0.49  \(W_{\mathrm{fruit}}^{\mathrm{ini}}\)   0.74  Total flesh sugar concentration  0.35  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   0.90      Ksugar  0.37      aL   0.66   Fruit trait     Parameter     Fruit dry mass  0.52  ddmax  0.96  Fruit fresh mass  0.47  \(W_{\mathrm{fruit}}^{321}\)   0.52  Stone fresh mass  0.60  \(GR_{\mathrm{fruit}}^{\mathrm{early}}\)   0.81  Flesh dry matter content  0.49  \(W_{\mathrm{fruit}}^{\mathrm{ini}}\)   0.74  Total flesh sugar concentration  0.35  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   0.90      Ksugar  0.37      aL   0.66   All correlations appeared highly significant (P <0.001). View Large Correlations between traits and key parameters Among the correlations between the 12 parameters and five traits of interest at maturity (Table 3), the strongest were between stone fresh mass and the three early growth parameters, \(W_{\mathrm{fruit}}^{321},\) \(GR_{\mathrm{fruit}}^{\mathrm{early}},\) and \(W_{\mathrm{fruit}}^{\mathrm{ini}}\) and, as expected, \(w_{\mathrm{stone}}^{\mathrm{matu}},\) the potential maximal stone dry mass at maturity. Other correlations were significant but not strong: dry and fresh fruit masses appeared correlated to the three early growth parameters and to \(w_{\mathrm{stone}}^{\mathrm{matu}}.\) Fruit fresh mass correlated with the parameter \(df_{\mathrm{stone}}^{1},\) which was also correlated with stone mass. Surprisingly, the correlation between fruit dry mass and the initial relative flesh growth rate \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) was low and only just significant. The low but significant negative correlation between aL and the fruit dry mass was not expected since the water flux submodel does not influence the carbon submodel. Flesh dry matter content was negatively correlated with the parameter aL, a water uptake parameter, but no correlation was found with the parameter ρ which also interacts in the water fluxes of the fruit. As expected, total sugar concentration was significantly correlated with the two parameters ksugar and aL. Lastly, no correlation was found between the maturity date and any of the traits. Table 3. Correlation coefficients between fruit traits and parameter values measured for the BC2 progeny in 2002   Fruit dry massa   Fruit fresh massa   Stone fresh massa   Flesh dry matter contenta   Total flesh sugar concentrationa   ddmax  −0.02  −0.008  −0.01  −0.14  0.12  \(W_{\mathrm{fruit}}^{321}\)   0.33***  0.30***  0.64***  0.10  −0.04  \(GR_{\mathrm{fruit}}^{\mathrm{early}}\)   0.47***  0.50***  0.58***  0.09  0.02  \(W_{\mathrm{fruit}}^{\mathrm{ini}}\)   0.52***  0.53***  0.70***  0.10  0.01  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   0.20*  0.14*  0.10  0.11  0.09  \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   0.36***  0.34***  0.81***  0.19*  −0.2  \(df_{\mathrm{stone}}^{1}\)   0.21*  0.30***  0.47***  −0.22*  −0.12  ksugar  −0.10  −0.08  0.09  −0.06  −0.41***  rsu  0.04  0.02  0.09  0.05  0.02  ρ  −0.10  −0.17  0.14  0.09  −0.03  s1  −0.19*  −0.11  −0.07  −0.22*  −0.08  aL   −0.26**   −0.13   0.04   −0.41***   −0.59***     Fruit dry massa   Fruit fresh massa   Stone fresh massa   Flesh dry matter contenta   Total flesh sugar concentrationa   ddmax  −0.02  −0.008  −0.01  −0.14  0.12  \(W_{\mathrm{fruit}}^{321}\)   0.33***  0.30***  0.64***  0.10  −0.04  \(GR_{\mathrm{fruit}}^{\mathrm{early}}\)   0.47***  0.50***  0.58***  0.09  0.02  \(W_{\mathrm{fruit}}^{\mathrm{ini}}\)   0.52***  0.53***  0.70***  0.10  0.01  \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\)   0.20*  0.14*  0.10  0.11  0.09  \(w_{\mathrm{stone}}^{\mathrm{matu}}\)   0.36***  0.34***  0.81***  0.19*  −0.2  \(df_{\mathrm{stone}}^{1}\)   0.21*  0.30***  0.47***  −0.22*  −0.12  ksugar  −0.10  −0.08  0.09  −0.06  −0.41***  rsu  0.04  0.02  0.09  0.05  0.02  ρ  −0.10  −0.17  0.14  0.09  −0.03  s1  −0.19*  −0.11  −0.07  −0.22*  −0.08  aL   −0.26**   −0.13   0.04   −0.41***   −0.59***   a * P <0.05; ** P <0.01; *** P <0.001; others not significant. View Large Relationships between the key parameters Pairwise correlations between \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar, and rsu, were strong (Fig. 2). Parameters aL and ksugar displayed a particularly tight linear relationship (correlation=0.81). These four parameters were also highly negatively correlated to growth duration ddmax (correlation coefficient ranging from −0.58 to −0.83). A non-linear and three linear equations described the relationships between ddmax and \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar and rsu, respectively (Fig. 2). Fig. 2. View largeDownload slide Relationships between the values of the five parameters \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar, rsu, and ddmax for the BC2 population. The lines represent the global adjustments for the relationships between ddmax and the four other parameters. For the relationship between ddmax and \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) a curve was adjusted, whereas for the others a linear adjustment was done. Correlation is indicated. All correlations appeared highly significant (P <0.001). Fig. 2. View largeDownload slide Relationships between the values of the five parameters \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar, rsu, and ddmax for the BC2 population. The lines represent the global adjustments for the relationships between ddmax and the four other parameters. For the relationship between ddmax and \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) a curve was adjusted, whereas for the others a linear adjustment was done. Correlation is indicated. All correlations appeared highly significant (P <0.001). Detection of QTLs for traits of interest and key parameters QTLs were detected for the 12 parameters and for fruit dry mass (see Table SP in the supplementary data available at JXB online). QTLs accounted for between 7% and 67% of the observed variation. Main QTLs were detected for both years, but the fraction of total variation of each trait explained by the QTL was generally lower in 2001 than in 2002. The location of the QTLs on the linkage map is presented in Fig. 3 together with the QTLs detected by Quilot et al. (2004) for the traits of interest: fruit fresh mass, stone cheek diameter and fresh mass, total flesh sugar concentration, and flesh soluble solid content. Fig. 3. View largeDownload slide Location of putative QTLs controlling genotypic key parameters of the ecophysiological model and fruit quality traits analysed for two successive years: \(W_{\mathrm{fruit}}^{321},GR_{\mathrm{fruit}}^{\mathrm{early}},W_{\mathrm{fruit}}^{\mathrm{ini}},w_{\mathrm{stone}}^{\mathrm{matu}},RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) ksugar, ρ, s1, rsu, aL, \(df_{\mathrm{stone}}^{\mathrm{{^\prime}}},\) ddmax, fruit dry mass (DMass), fruit fresh mass (FMass), flesh dry matter content (FDMC), fruit polar diameter (FPolarD), stone cheek diameter (SCheekD), stone fresh mass (SMass), soluble solid content (SSC), total sugar (TSugar) concentrations, and of putative QTLs for the residuals ( \(res.RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) res.aL, res.ksugar, and res.rsu) of the relationships described in Fig. 2, linking ddmax and the four parameters \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) ksugar, aL, and rsu. Markers are listed on the right of each linkage group and genetic distances on the left. QTLs associated with markers of S or Z (superscript) genomes that could be assigned to linkage groups are listed on the left of each linkage group. QTLs associated with markers of the P1908 genome are listed in italics on the right of each linkage group. Underlined QTLs are those for which the P1908 allele confers a positive effect for a horticulture perspective. Year of observation is denoted by 1 and 2 for 2001 and 2002, respectively. When co-located, QTLs are ordered by decreasing individual contribution from left to right on each side of the linkage group. Fig. 3. View largeDownload slide Location of putative QTLs controlling genotypic key parameters of the ecophysiological model and fruit quality traits analysed for two successive years: \(W_{\mathrm{fruit}}^{321},GR_{\mathrm{fruit}}^{\mathrm{early}},W_{\mathrm{fruit}}^{\mathrm{ini}},w_{\mathrm{stone}}^{\mathrm{matu}},RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) ksugar, ρ, s1, rsu, aL, \(df_{\mathrm{stone}}^{\mathrm{{^\prime}}},\) ddmax, fruit dry mass (DMass), fruit fresh mass (FMass), flesh dry matter content (FDMC), fruit polar diameter (FPolarD), stone cheek diameter (SCheekD), stone fresh mass (SMass), soluble solid content (SSC), total sugar (TSugar) concentrations, and of putative QTLs for the residuals ( \(res.RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) res.aL, res.ksugar, and res.rsu) of the relationships described in Fig. 2, linking ddmax and the four parameters \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) ksugar, aL, and rsu. Markers are listed on the right of each linkage group and genetic distances on the left. QTLs associated with markers of S or Z (superscript) genomes that could be assigned to linkage groups are listed on the left of each linkage group. QTLs associated with markers of the P1908 genome are listed in italics on the right of each linkage group. Underlined QTLs are those for which the P1908 allele confers a positive effect for a horticulture perspective. Year of observation is denoted by 1 and 2 for 2001 and 2002, respectively. When co-located, QTLs are ordered by decreasing individual contribution from left to right on each side of the linkage group. QTLs with the highest individual contribution were detected for ddmax (38% in 2002). For both years, QTLs were detected for ddmax and associated with SSR marker UDP96-003 on LG4, with differences between both S alleles and Z alleles. However, the global R2 only reached 0.39 and 0.54, respectively, in 2001 and 2002. Most QTLs for the four parameters \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar, and rsu were also detected at the same loci as those for ddmax. Indeed, for both years, the same three QTLs were detected for the dry flesh growth rate, \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) at the markers UDP96-003 (LG4, S, and Z) and CFF13 (LG3). Three of the four QTLs detected in 2002 for the parameter related to sugar metabolism, ksugar, also co-located with QTLs for ddmax. Considering the tight links between ddmax and the four parameters \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar, and rsu, a QTL analysis was performed on the residuals of the relationships ( \(res.RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) res.aL, res.ksugar, and res.rsu) linking ddmax and the parameters \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar, and rsu (Fig. 2). Some QTLs for these residuals were co-located with QTLs for the associated parameters. However, no QTL was detected for the residuals at the same markers as those for ddmax, except for a QTL detected by marker UDP96-003 (LG4) for res.ksugar in 2002 (R2=0.05). QTLs with high individual contribution were detected for the early growth parameter \(W_{\mathrm{fruit}}^{321}\) and the potential maximal stone dry mass \(w_{\mathrm{stone}}^{\mathrm{matu}}\) at the PC60 marker, on LG6. QTLs for \(W_{\mathrm{fruit}}^{321}\) and for \(W_{\mathrm{fruit}}^{\mathrm{ini}}\) and \(GR_{\mathrm{fruit}}^{\mathrm{early}}\) were co-located on LG1. These two regions of LG1 and LG6 and the regions of LG4 and LG8, where QTLs were detected for \(W_{\mathrm{fruit}}^{\mathrm{ini}}\) and \(GR_{\mathrm{fruit}}^{\mathrm{early}}\) , respectively, also harboured QTLs for stone mass ( \(w_{\mathrm{stone}}^{\mathrm{matu}}\) and SMass). Alleles coming from P1908 enhanced the values of these parameters at the QTL on LG6 and decreased them at the QTL on LG1 and 4. QTLs for res.aL, res.ksugar, and res.rsu were detected on LG1, each co-located with QTLs for aL, ksugar, and rsu. Three QTLs (LG4, 6, and 7) for the permeation coefficient of fruit surface to water vapour, ρ, were detected; however, this parameter was observed for 36 genotypes only. QTLs detected for the parameters and residuals were often co-located with QTLs for quality traits. Most QTLs for fresh and dry fruit mass appeared co-located with QTLs for the fruit dry growth parameters, \(W_{\mathrm{fruit}}^{321},GR_{\mathrm{fruit}}^{\mathrm{early}},W_{\mathrm{fruit}}^{\mathrm{ini}}\) (LG1), \(RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) (LG4 and 7). They were also co-located with QTLs for res.rsu and rsu (LG1), aL (LG2), \(res.RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) (LG4), ρ (LG4), res.aL, \(res.RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) and ρ (LG7). QTLs for total sugar concentration were detected in the same region of LG1 as QTLs for res.ksugar and res.aL and in the same region of LG6 as QTL for \(res.RGR_{\mathrm{flesh}}^{\mathrm{ini}}.\) Last, QTLs for flesh dry matter content and res.ksugar were co-located on LG3. Combination of the ecophysiological and genetic models Parameters of the ecophysiological model \(W_{\mathrm{fruit}}^{\mathrm{ini}},w_{\mathrm{stone}}^{\mathrm{matu}},RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) ksugar, s1, rsu, aL, and \(df_{\mathrm{stone}}^{1}\) were estimated using the QTL results (see Table SP in the supplementary data that can be found at JXB online), concerning 2002 data only. The observed value of ddmax for each genotype was used since the model is highly sensitive to this parameter and QTLs detected for ddmax only explained a small fraction of the total variation observed, despite a high correlation between the 2001 and 2002 values. For the four parameters for which QTLs were detected on the residuals of the relationship with ddmax ( \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar, and rsu), the effects of the QTL were added to the equation of this relationship. For example, the estimated value of aL for an individual i was computed as follows:  \begin{eqnarray*}&&aL_{\mathrm{i}}{=}f(dd_{\mathrm{max}}){+}\mathrm{{\mu}}{+}a_{\mathrm{UDP}96{\,}{-}008}{\times}G_{\mathrm{i,UDP}96{\,}{-}008}{+}a_{\mathrm{CFF}11}\\&&{\times}\mathrm{G}_{\mathrm{i,CFF}11}{+}a_{\mathrm{CFM}8}{\times}\mathrm{G}_{\mathrm{i,CFM}8}{+}\mathrm{e}_{\mathrm{PPCT}025_\mathrm{CFF}1}{\times}G_{\mathrm{i,PPCT}025_CFF1}\end{eqnarray*} Consequently:  \begin{eqnarray*}&&aL_{\mathrm{i}}{=}(0.01478{-}4.498{\times}10^{{-}06}{\times}dd_{\mathrm{max}}){-}0.0014\\&&{+}0.0010{\times}G_{\mathrm{i,UDP}96{-}008}{+}0.0014{\times}\mathrm{G}_{\mathrm{i,CFF}11}\\&&{+}0.0011{\times}\mathrm{G}_{\mathrm{i,CFM}8}{+}0.003{\times}\mathrm{G}_{\mathrm{i,PPCT}025_\mathrm{CFF}11}\end{eqnarray*} where the genetic QTL scores Gi,n took the values 0 or 1 depending on the allele of i at the corresponding loci. The combined model remained accurate for most of the output variables. Goodness-of-fit of the combined model was high for flesh dry matter content, total sugar concentration, and stone fresh mass, since mean RRMSE values over the population were low (Table 4). For dry and fresh fruit masses, mean RRMSE were higher, but remained satisfactory. Evaluating that the model efficiently ranked the genotypes for fruit and stone masses, predictions of the combined model were well correlated with the observations. By contrast, predictions were less reliable for dry matter content and total sugar concentration of flesh, although it is worth noting that a few genotypes were badly represented by the combined model. Table 4. Evaluation of the combined model (QTL and ecophysiological models combined) at maturity   Fruit dry mass (g)   Fruit fresh mass (g)   Stone fresh mass (g)   Flesh dry matter content (g g−1)   Total flesh sugar concentration g (100 gFM)−1   RRMSE  0.31  0.33  0.18  0.11  0.17  COR   0.55   0.51   0.67   0.16   0.27     Fruit dry mass (g)   Fruit fresh mass (g)   Stone fresh mass (g)   Flesh dry matter content (g g−1)   Total flesh sugar concentration g (100 gFM)−1   RRMSE  0.31  0.33  0.18  0.11  0.17  COR   0.55   0.51   0.67   0.16   0.27   Mean values of relative mean squared error (RRMSE) over the population and Spearman correlation coefficients (COR) between observed and predicted values are presented for each output variable. View Large Discussion Contributions of the approach An innovative approach has been applied consisting of analysing the parameters involved in the development of traits, instead of considering these traits directly. The analysis of the stability between years of the parameter and quality trait values revealed better correlations between 2001 and 2002 values for the genotypic parameters than for the quality traits. Consequently, the detection of QTLs for such parameters was expected to be more successful than for quality traits (Yin et al., 1999). QTLs were detected for all the genotypic parameters and a number of them were common to both years of experimentation. The sum of QTL effects for each genotypic key parameter was included in the ecophysiological model. Thus parameter values could be predicted for each genotype. Finally, the quality of the combined model turned out to be moderately suitable. Following a similar approach to that presented here, Yin et al. (2000) encountered difficulties with the initial accuracy of the ecophysiological model they used. Reymond et al. (2003) applied this method with success to a simple ecophysiological model, with only three parameters, restricted to the description of leaf elongation rate of maize. Such a method was also tested by Buck-Sorlin and Bachmann (2000) integrating additive gene effects into a morphological model. In this context, this study represents a further step towards the inclusion of genetic information into a complex ecophysiological model. The approach used here led to promising results and various potential uses of the combined model are attractive. Perspectives of improvement of the approach The relevance of the approach depends on the characteristics of the genotypic parameters that influence the level of the QTLs effect and the stability of the QTLs over years. Different ways lead to identifying such parameters. Reymond et al. (2003) have considered the parameters involved in the response curves of leaf elongation rate to environmental conditions. Response curves were based on experimental relationships valid over a large range of environmental conditions for a given genotype. Therefore parameters were considered as a stable characteristic of a genotype. In this study, some parameters ( \(w_{\mathrm{stone}}^{\mathrm{matu}},df_{\mathrm{stone}}^{1},\) rsu, ρ, s1) were likewise estimated from response curves of a phenotypic trait to a measured plant signal in different environmental conditions. Other parameters ( \(W_{\mathrm{fruit}}^{321},\) \(W_{\mathrm{fruit}}^{\mathrm{ini}},\) \(GR_{\mathrm{fruit}}^{\mathrm{early}},RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) ksugar, and ddmax) were estimated under potential growth conditions. In this case, parameter values should reflect the intrinsic value of the genotype. However, some QTLs detected for the parameters were not common to both years and the fraction of total variation of each trait explained by the QTLs was generally low. The fraction of total variation of each trait explained by the QTLs was generally lower in 2001 than in 2002. This may be due to the fact that non-limiting fruit growth conditions were hardly met in 2001 for all genotypes. Trees were young and fruit growth may have undergone competition with vegetative and root system growths. A further experiment under maximum fruit growth conditions is required to overcome insufficient year of testing and to check the QTL stability. Besides the characteristics of the genotypic parameters, the detection of QTLs also depends on the saturation of the genetic map. Correlation between years for a trait provides an order of magnitude of its heritability. Accordingly, if most QTLs for a trait were detected, the total variation explained by these QTLs was expected to be approximatively equal to the corresponding correlation between 2001 and 2002 observations for this trait. In most cases, it was much lower. The most obvious case was the growth duration (ddmax) for which the R2 was 0.39 and 0.54 in 2001 and 2002, respectively, whereas the correlation between years was much higher (0.96). For this reason, it was hypothesized that not all the polymorphism arising from the S and Z genomes with respect to the growth duration, and, perhaps, other parameters had been detected; this, in turn, may have reduced the power of detecting P1908 alleles affecting those traits. To cope with these limitations, it is necessary to integrate new markers for the S and Z genomes. Further understanding of quality build-up This approach can provide a basis for the understanding of physiological and genetical phenomena, via the dissection of the quality traits into elementary processes. Indeed, since each parameter is involved in a few identified processes, this approach helps to highlight the main processes responsible for the variations in a complex trait. Through the study of the co-locations between QTLs of parameters and traits, a physiological hypothesis could be proposed for connections between processes. Physiological mechanisms that influence a quality trait at each co-located QTL could be deduced from the function in which the parameters intervene. For example, on LG4 and 7, QTLs for fruit fresh mass are co-located with QTLs for \(res.RGR_{\mathrm{flesh}}^{\mathrm{ini}}\) : the pulp demand for dry matter growth influences the fruit fresh growth. In LG1, QTLs for fruit fresh mass are located in the same region as QTLs for parameters involved in sugar metabolism and early fruit growth. Lastly, on LG2, 4, and 7, they are co-located with QTLs for parameters involved in water fluxes in the fruit (aL, res.aL and ρ). In addition, a parameter could influence different quality traits. For instance, QTLs for res.ksugar were located in the same region as a QTL for total sugar concentration (LG1) and a QTL for flesh dry matter content (LG3). Indeed, when ksugar increases, carbon is further used for the synthesis of compounds other than sugars and total sugar concentration decreases. As a result, the osmotic potential decreases and less water enters the fruit so that flesh dry matter content increases. Growth duration (ddmax) was highly correlated with four genotypic parameters ( \(RGR_{\mathrm{flesh}}^{\mathrm{ini}},\) aL, ksugar, and rsu) and QTLs for ddmax were co-located with QTLs for these parameters. The sensitivity analysis of the model to the parameter variations revealed that quality traits were influenced by variations of ddmax (Quilot et al., 2005). However, no correlation was found between ddmax and quality traits. Further studies are necessary to understand these observations and the low correlations generally observed between parameters and quality traits. This approach also highlighted the lack of knowledge regarding fruit quality development and the need for ecophysiological models dealing with genotypic variation in quality traits. Indeed, the ecophysiological model used only considered fruit growth during the phase of cell enlargement. Effects of early fruit growth and harvest time were taken into account through \(W_{\mathrm{fruit}}^{\mathrm{ini}}\) and ddmax. These two parameters appeared highly variable between genotypes and highly influential concerning quality traits at maturity. Describing the early growth stage via a model of cell division, taking into account limitations of assimilate supply, should make it possible to predict better the fruit mass at the end of the cell division stage and the sink potential of the fruit. Since maturity date appeared to be influenced by tree fruit load (Johnson and Handley, 1989), ecophysiological models should describe the underlying mechanisms involved in the maturation stage before harvest in order to predict maturity date whatever the year and the fruit load. Potential contributions to crop improvement The combined models may be used for practical purposes, such as predicting the genotypic variations of a plant response to environmental conditions. Yin et al. (2003) supported the idea that such models may help to solve genotype×environment interactions. Tardieu (2003) stated that they theoretically make it possible to predict the behaviour of plants with any combination of alleles under any climatic scenario. The interactions between processes underlined here result in difficulties to improve some traits, since the enhancement of some processes appeared to be favourable to some traits of interest but undesirable to others. In a context of multi-criteria objectives, this combined model may also provide a potential tool for rationalizing the contradictions between the effects of the processes, enhancing some traits without diminishing the others too much. Integrating the knowledge and potentialities of physiology, genetics, and modelling to enhance the understanding of plant functioning has been considered a major challenge over the past few years. Besides the implications for genetic improvement, it is essential to note that all disciplines will benefit from this multidisciplinary approach. Indeed, modellers need to integrate the latest insight into biological mechanisms and may also incorporate the action of genes in their models. In return, models can help to test hypotheses on likely mechanisms, guide research, accelerate scientific understanding, and lead to practical applications of quantitative genetics. We gratefully acknowledge K Moreau for genotyping. We thank C Borel for critically revising the manuscript. We thank A Lacombe and Dr O Savolainen for improving the English. This research was funded in part by grants from the Ministère de la Recherche, from Région Provence-Alpes-Côte d'Azur (projects DEB 02-252 and DEB 03-543) and from the Institut National de la Recherche Agronomique, France (A.I.P. PFI P00232 and A.I.P. REA P00251). References Abbott AG, Rajapakse S, Sosinski B, Lu ZX, Sossey-Alaoui K, Scorza R, Callahan A. 1998. Construction of saturated linkage maps of peach crosses segregating for characters controlling fruit quality, tree architecture and pest disease. Acta Horticulturae  465, 41–50. Google Scholar Boote KJ, Jones JW, Pickering NB. 1996. Potential uses and limitations of crop models. Agronomy Journal  88, 704–716. Google Scholar Buck-Sorlin G, Bachmann K. 2000. Simulating the morphology of barley spike phenotypes using genotype information. Agronomie  20, 691–702. Google Scholar Corelli-Grappadelli L, Coston DC. 1991. Thinning pattern and light environment in peach tree canopies influence fruit quality. HortScience  26, 1464–1466. Google Scholar Etienne C, Rothan C, Moing A, Plomion C, Bodenes C, Svanella-Dumas L, Cosson P, Pronier V, Monet R, Dirlewanger E. 2002. Candidate genes and QTLs for sugar and organic acid content in peach [Prunus persica (L.) Batsch]. Theoretical and Applied Genetics  105, 145–159. Google Scholar Foulongne M, Pascal T, Pfeiffer F, Kervella J. 2003. QTLs for powdery mildew resistance in peach×Prunus davidiana crosses: consistency across generations and environments. Molecular Breeding  12, 33–50. Google Scholar Génard M. 1992. Influence du nombre de feuilles et de la répartition des fruits sur la production et la qualité des pêches. Canadian Journal of Plant Science  72, 517–525. Google Scholar Génard M, Bruchou C. 1992. Multivariate analysis of within-tree factors accounting for the variation of peach fruit quality. Scientia Horticulturae  52, 37–51. Google Scholar Hammer GL, Butler DG, Muchow RC, Meinke H. 1996. Integrating physiological understanding and plant breeding via crop modelling and optimization. In: Cooper M, Hammer GL, eds. Plant adaptation and crop improvement. Wallingford, UK: CAB International, 419–441. Google Scholar Johnson RS, Handley DF. 1989. Thinning response of early-, mid-, and late-season peaches. Journal of the American Society for Horticultural Science  114, 852–855. Google Scholar Kliewer WM, Weaver RJ. 1971. Effect of crop level and leaf area on growth, composition and coloration of Tokay grapes. American Journal of Enology and Viticulture  22, 172–177. Google Scholar Kobayashi K, Us Salam M. 2000. Comparing simulated and measured values using mean squared deviation and its components. Agronomy Journal  92, 345–352. Google Scholar Marini RP, Sowers DL, Marini MC. 1991. Peach fruit quality is affected by shade during final swell of fruit growth. Journal of the American Society for Horticultural Science  116, 383–389. Google Scholar Ognjanov V, Vujanic-Varga D, Misic PD, Veresbaranji I, Macet K, Tesovic Z, Krstic M, Petrovic N. 1995. Anatomical and biochemical studies of fruit development in peach. Scientia Horticulturae  64, 33–48. Google Scholar Pascal T, Kervella J, Pfeiffer F, Sauge MH, Esmenjaud D. 1998. Evaluation of the interspecific progeny Prunus persica cv. Summergrand×Prunus davidiana for disease resistance and some agronomic features. Acta Horticulturae  465, 185–192. Google Scholar Quarta R, Dettori MT, Verde I, Gentile A, Broda Z. 1998. Genetic analysis of agronomic traits and genetic linkage mapping in a BC1 peach population using RFLPs and RAPDs. Acta Horticulturae  465, 51–60. Google Scholar Quilot B, Génard M, Lescourret F, Kervella J. 2005. Simulating genotypic variations of fruit quality in an advanced peach×Prunus davidiana cross. Journal of Experimental Botany  56, (in press). Google Scholar Quilot B, Wu B, Kervella J, Génard M, Foulongne M, Moreau K. 2004. QTL analysis of quality traits in an advanced backcross between Prunus persica cultivars and the wild relative species P. davidiana. Theoretical and Applied Genetics  109, 884–897. Google Scholar Reymond M, Muller B, Leonardi A, Charcosset A, Tardieu F. 2003. Combining quantitative trait loci. Analysis and an ecophysiological model to analyze the genetic variability of the responses of maize leaf growth to temperature and water deficit. Plant Physiology  131, 664–675. Google Scholar Saliba-Colombani V, Causse M, Langlois D, Philouze J, Buret M. 2001. Genetic analysis of organoleptic quality in fresh market tomato. I. Mapping QTLs for physical and chemical traits. Theoretical and Applied Genetics  102, 259–272. Google Scholar Scorza R, May LG, Purnell B, Upchurch B. 1991. Differences in number and area of mesocarp cells between small- and large-fruited peach cultivars. Journal of the American Society for Horticultural Science  116, 861–864. Google Scholar Shorter R, Lawn RJ, Hammer GL. 1991. Improving genotypic adaptation in crops: a role for breeders, physiologists and modellers. Experimental Agriculture  27, 155–175. Google Scholar Tardieu F. 2003. Virtual plants: modelling as a tool for the genomics of tolerance to water deficit. Trends in Plant Science  8, 9–14. Google Scholar Veldboom LR, Lee M. 1996. Genetic mapping of quantitative trait loci in maize in stress and non-stress environments. I. Grain yield and yield components. Crop Science  36, 1310–1319. Google Scholar Yamaguchi M, Haji T, Miyake M, Yaegaki H. 2002. Studies on the varietal differences and yearly deviation of mesocarp cell numbers and lengths and fruit weight among commercial peach [Prunus persica (L.) Batsch] cultivars and selections, wild types, and their hybrids. Journal of the Japanese Society for Horticultural Science  71, 459–466. Google Scholar Yin X, Chasalow SC, Dourleijn CJ, Stam P, Kropff MJ. 2000. Coupling estimated effects of QTLs for physiological traits to a crop growth model: predicting yield variation among recombinant inbred lines in barley. Heredity  85, 539–549. Google Scholar Yin X, Kropff MJ, Stam P. 1999. The role of ecophysiological models in QTL analysis: the example of specific leaf area in barley. Heredity  82, 415–421. Google Scholar Yin X, Stam P, Kropff MJ, Schapendonk HCM. 2003. Crop modelling, QTL mapping, and their complementary role in plant breeding. Agronomy Journal  95, 90–98. Google Scholar © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]

Nardini, Andrea;Salleo, Sebastiano

doi: 10.1093/jxb/eri306pmid: 16246857

Abstract The hydraulic architecture, water relationships, and gas exchange of leaves of sunflower plants, grown under different levels of water stress, were measured. Plants were either irrigated with tap water (controls) or with PEG600 solutions with osmotic potential of −0.4 and −0.8 MPa (PEG04 and PEG08 plants, respectively). Mature leaves were measured for hydraulic resistance (Rleaf) before and after making several cuts across minor veins, thus getting the hydraulic resistance of the venation system (Rvenation). Rleaf was nearly the same in controls and PEG04 plants but it was reduced by about 30% in PEG08 plants. On the contrary, Rvenation was lowest in controls and increased in PEG04 and PEG08 plants as a likely result of reduction in the diameter of the veins' conduits. As a consequence, the contribution of Rvenation to the overall Rleaf markedly increased from controls to PEG08 plants. Leaf conductance to water vapour (gL) was highest in controls and significantly lower in PEG04 and PEG08 plants. Moreover, gL was correlated to Rvenation and to leaf water potential (Ψleaf) with highly significant linear relationships. It is concluded that water stress has an important effect on the hydraulic construction of leaves. This, in turn, might prove to be a crucial factor in plant–water relationships and gas exchange under water stress conditions. Gas exchange, leaf hydraulic architecture, sunflower, water relations, water stress Introduction Several attempts at measuring the hydraulics of plant organs have been reported in the past, but only over the last 25 years has the concept of ‘plant hydraulic architecture’, as introduced by Zimmermann (1978), been widely adopted, with the purpose of interpreting the water balance of plants as the integrated result of the hydraulic properties of their organs (Tyree and Zimmermann, 2002). Over this time, changes in the hydraulics of roots, stems, and leaves of several species have been reported to occur in response to changing environmental factors. The discovery of aquaporins in plant cells (Maurel, 1997; Luu and Maurel, 2005) has made a powerful contribution to the explanation of why cell membranes have a high permeability to water and has elucidated some of its sources of variation (Henzler et al., 1999; Nardini et al., 2005). In particular, studies of the water pathways within the root (Steudle et al., 1987; Steudle and Peterson, 1998) have made it possible to discriminate between the contribution of the symplastic and that of the apoplastic hydraulic resistance along the soil-to-stele xylem path and have provided a solid biophysical and molecular basis for analogous studies at the leaf level. The hydraulic properties of the leaf, however, are poorly understood at present, although this organ has been (and still is) widely studied for gas exchange, water status and, of course, photosynthesis. Leaf hydraulics are intrinsically difficult to study because: (i) the extreme morpho-anatomical heterogeneity of this organ, even within one individual, generates analogous heterogeneity in the data and makes them difficult to generalize; (ii) liquid and gaseous water flow in a leaf are hard to discriminate from each other using the techniques presently available for measuring hydraulic variables, like the vacuum chamber (Kolb et al., 1996; Nardini et al., 2001) or the high pressure flow meter (Tyree et al., 1995; Sack et al., 2002); (iii) a typical leaf of an angiosperm consists of a highly branched xylem system connected to the photosynthetic tissues through the vein living cells (the bundle sheath) about whose functional features very little is known at present. In addition, liquid water flows within the leaf lamina both in the vascular compartment and through the mesophyll living cells (the extra-vascular compartment). Each of these leaf compartments has its own hydraulic properties, the former mainly depending on the geometry of the xylem conduits (Canny, 1990; Cochard et al., 2004) and water permeability of the pits (Sperry et al., 2005), whereas the latter are closely dependent on water permeability of cell membranes and, ultimately, on cell metabolism (Morillon and Chrispeels, 2001). The present understanding of leaf hydraulic construction is still limited and data from the literature are sometimes contradictory (Zwieniecki et al., 2002; Cochard et al., 2004), especially when referring to the partitioning of leaf hydraulic resistance into venation and extra-vascular components. Among the environmental factors that may influence leaf hydraulics, water stress can be safely expected to be a major one. Water stress is well known to impair the conductive efficiency of the plant vascular system through xylem embolism (Tyree and Sperry, 1989) and it is now known that the leaf xylem also undergoes cavitation-induced hydraulic failure (Salleo et al., 2001). As an example, Kikuta et al. (1997) measured the threshold values of leaf water potential (Ψleaf) triggering vein cavitation in leaves of several deciduous and evergreen trees, on the basis of ultrasound acoustic emissions. Vein cavitation was found to occur at Ψleaf values between −0.5 and −2.0 MPa, i.e. well within the range of minimum Ψleaf experienced by plants in the field. More recently, hydraulic measurements of leaf blades of different species have revealed the potential impact of vein cavitation on leaf hydraulics (Nardini et al., 2001, 2003; Brodribb and Holbrook, 2003) and gas exchange (Lo Gullo et al., 2003; Brodribb and Holbrook, 2004). However, knowledge about the influence of middle- to long-term reduction of water availability on the hydraulic construction of leaves is still limited, although it is known that other environmental factors influencing leaf growth (e.g. light) have important effects on their hydraulic features (Sack et al., 2003, 2005). The present study investigates the effects of moderate water stress developing during plant growth and leaf maturation on leaf hydraulic architecture, in terms of the contribution of the hydraulic resistance of the leaf vasculature Rvenation to Rleaf, as well as of the impact of eventual changes in Rvenation on leaf gas exchange. Studies were conducted on sunflower because plants of the modern cultivars of this species are fairly stable genetically and they have a sufficiently uniform leaf structure. Materials and methods Plant material and growth conditions All experiments were conducted on 21 sunflower (Helianthus annuus L. cv. Margot) plants 7–9 weeks old. Seeds (provided by Maisadour Semences Italy srl) were planted in greenhouse trays and, after cotyledons were fully expanded, seedlings were transferred to 1.5 l pots filled with a mixture (1:1) of peat and sand (one seedling per pot). Plants were grown in a room where air temperature was adjusted to vary between 23 °C and 16 °C (day/night), relative humidity was set at 50±5% and light was provided by lamps (HQI-T 1000 W/D; Osram GmbH, München, Germany) with a photosynthetically active radiation (PAR) of 400±50 μmol m−2 s−1. The photoperiod was set at 12 h. Plants were irrigated daily for 20 d with 200 ml of tap water. After this time (when plants were bearing two pairs of leaves, excluding cotyledons), they were randomly divided into three groups of seven plants each. One group continued to be irrigated with tap water (controls). Increasing water stress levels were imposed on the other two groups of plants by irrigating them daily with 200 ml of polyethylene glycol (PEG600; Sigma-Aldrich) at a concentration of 0.18 M and 0.36 M, resulting in solutions with osmotic potential (π) of −0.4 and −0.8 MPa, respectively, as measured using a dew-point hygrometer (WP4; Decagon Devices, Pullman, USA). Although PEG600 is thought to enter cell membranes with time, use of higher-size glycols is possibly not the most convenient way to depress aquaporin activity and decrease cell hydraulic conductivity (Ye et al., 2004). The above PEG concentrations were also used because previous studies (Lo Gullo et al., 2004) had shown that leaf water potential at the turgor loss point (Ψtlp) of sunflower plants grown under similar environmental conditions was about −1.0 MPa. Therefore, the PEG solutions used in the present study were planned to correspond approximately to 40% and 80% of the expected Ψtlp, i.e. to a mild and a severe water stress level, respectively. Hereafter, the two groups of plants irrigated with the 0.18 M and 0.36 M PEG600 solutions are referred to as ‘PEG04’ and ‘PEG08’ plants, respectively. Every irrigation with PEG solution was always preceded by irrigation with 200 ml of tap water. This procedure was aimed at preventing a build-up of PEG600 concentration in the soil and, hence, soil water potential dropping below the desired, pre-set value. To avoid eventual differences in the availability of nutrients as caused by different watering treatments, each plant received 2.5 g of fertilizer (Nitrophoska Top, BASF Italia SpA; 15% N, 10% P2O5, 15% K2O, 2% MgO, 12% SO3, 0.02% B, 0.01% Zn) at 10 d intervals. Soil water potential (Ψsoil) was measured at regular intervals using a dew-point hygrometer (see above) on soil samples collected from one pot per group. In particular, Ψsoil was measured 2, 4, 8, and 16 d after the beginning of the water stress treatments. In each case, soil samples were collected 2 h after the last irrigation, i.e. after excessive water had drained out. Plants growing in the pots from which soil samples were taken were not used for subsequent experiments. All other measurements started 21 d after the beginning of the water stress treatment and were completed within the following 15 d. Measurements were performed on mature leaves sampled from the two most apical nodes. These leaves were chosen because a previous study (Lo Gullo et al., 2004) had shown that apical leaves of mature sunflower plants have maximum hydraulic conductance. Measurements of leaf gas exchange, water potential, and pressure–volume curves Leaf conductance to water vapour (gL) was measured using a steady-state porometer (LI-1600; Li-Cor Inc., Lincoln, NE, USA). Measurements were taken at the middle of the light period, i.e. between 5 and 7 h after lights had been turned on. A leaf chamber, 2 cm2 in surface area, was used, and gL was measured at the central portion of the leaf blade, close to the middle of the major vein. Immediately after gL measurements, leaves were cut off and their hydraulic architecture was measured as described below. Six leaves per group (one leaf per plant) were measured for gL and hydraulic architecture. This procedure was adopted to get information about eventual relationships existing between leaf gas exchange rate and the corresponding hydraulic characteristics. To obtain information about relationships between gL and the leaf water potential (Ψleaf), another set of leaves (six leaves per group, one leaf per plant) was measured for gL as described above. In this case, however, immediately after gL measurements, leaves were cut off, immediately enclosed in dark plastic bags, and their Ψleaf was measured using a pressure chamber (model 3005; Soilmoisture Equipment Corp., Santa Barbara, CA, USA). To check changes in leaf osmotic potential at full turgor (πo) and water potential at the turgor loss point (Ψtlp), as induced by water stress, four leaves per group from different plants were collected and rehydrated overnight by placing the cut end of the petiole in contact with distilled water. Leaf water potential isotherms (Tyree and Hammel, 1972) were measured using the procedure described by Salleo (1983) and Salleo et al. (1997). From pressure–volume curves, π0 and Ψtlp were calculated. Measurements of leaf hydraulic architecture Leaf hydraulic resistance (Rleaf) was measured using a high pressure flow meter (HPFM) (Tyree et al., 1995). The instrument has proved to yield values of Rleaf consistent with those obtained using independent methods (Sack et al., 2002; Salleo et al., 2003). Leaves were cut off immediately after gL measurements had been completed (see above) leaving about 30 mm of petiole for connection to the HPFM. Leaves were then connected to the instrument using compression fittings, within 5 min from cutting. Degassed water filtered at 0.1 μm, was forced into the petiole at a pressure of 0.15 MPa and R was measured as the pressure-to-flow ratio at 16 s intervals until values became stable (i.e. the coefficient of variation of the last 20 readings was <3%), which usually took 10–20 min. During measurements, leaves were maintained under usual laboratory irradiance (PAR <10 μmol m−2 s−1). Although Rleaf is known to be reduced upon illumination at PAR ≥1200 μmol m−2 s−1 in several species (Sack et al., 2002; Tyree et al., 2005), this is not the case of sunflower leaves when collected during their light period, as recently reported in a study by Nardini et al. (2005). The hydraulic resistance of the leaf vasculature (Rvenation) was estimated using the technique first proposed by Sack et al. (2004) consisting of serial measurements of leaf hydraulic resistance after increasing numbers of minor veins of the fourth order or higher had been cut open in order to by-pass the extra-vascular leaf compartment. Minor veins were cut at random locations throughout the lamina by making 1.5–2 mm incisions with a scalpel. This procedure was basically similar to that originally proposed by Sack et al. (2004) and repeated by Gascò et al. (2004) and Nardini et al. (2005). In the study by Sack et al. (2004), 120–150 cuts were reported to be sufficient to yield stable R values. In the case of sunflower, however, higher numbers of cuts (up to 500 cuts per leaf, corresponding to approximately 3–5 cuts cm−2) were necessary to achieve stable and relatively invariant R values. At the end of measurements, the leaf surface area (Aleaf) was measured using a leaf area meter (LI-3000A; Li-Cor Inc.), and both Rleaf and Rvenation were normalized by Aleaf (one side only). Anatomical measurements Because Rvenation is likely to be related to anatomical features of the leaf veins as eventually modified by water stress, the diameter of xylem conduits of the midrib was measured for each leaf used for HPFM measurements. Cross-sections of the midrib (one section per leaf) were taken at the proximal third of the leaf length using fresh razor blades. Sections were immediately observed under a light microscope and the total number of conduits per midrib was recorded. Then, the diameter of the 10–15 widest conduits per section was measured. The diameter of conduits that were elliptical in shape was computed by averaging the major and minor axes. The authors are aware that dimensions of the xylem conduits in the midrib may not necessarily give information of the hydraulic efficiency of the whole leaf venation system but they at least provide an idea of the likely reduction of conduit dimensions in higher order veins, because hydraulic traits of midrib, major veins, and minor veins are known to be correlated with each other (Sack et al., 2005). Statistics Data were analysed with the SigmaStat 2.0 (SPSS, Chicago, IL, USA) statistics package. One-way ANOVA was used to test differences between experimental groups. The statistical significance of correlations between parameters was tested using the Pearson Product Moment Correlation. Results Repeated plant irrigations with PEG aqueous solutions caused soil water potential (Ψsoil) to decrease from −0.22 MPa (controls) to −0.58 MPa for plants irrigated with PEG solution at π=−0.4 MPa and further to −0.94 MPa for plants irrigated with PEG at π=−0.8 MPa (Table 1). The main morphological/anatomical effects of the two levels of water stress applied to growing sunflower plants are summarized in Table 2, and an example of the general appearance of plants is given in Fig. 1. PEG04 plants grew less and had thinner stems than controls, i.e. their height above the ground was 0.57 m versus 0.73 m recorded for controls and their base stem diameter was about 8.0 mm versus 9.7 mm for controls. PEG08 plants showed an even larger reduction in height and stem diameters which were about 34% and 28% less than controls, respectively. Also the total leaf surface area was consistently reduced as a consequence of the water stress applied, in terms of both number of leaves per plant and mean leaf surface area (Aleaf). In particular, Aleaf of mature, fully expanded leaves was about 113 and 105 cm2 in PEG04 and PEG08 plants, respectively, versus about 140 cm2 as recorded for control leaves, i.e. leaves from water-stressed plants were 19–25% smaller than leaves from well-hydrated plants. In addition, the widest xylem conduits in the midrib were significantly narrower in PEG04 and PEG08 plants (37.6 and 35.1 μm in diameter, respectively) than those of controls (43.2 μm in diameter). Fig. 1. View largeDownload slide General appearance of controls and water-stressed sunflower plants (PEG04 and PEG08; see text for details). The picture was taken 40 d after seeding and 20 d after the beginning of the water stress treatments. Fig. 1. View largeDownload slide General appearance of controls and water-stressed sunflower plants (PEG04 and PEG08; see text for details). The picture was taken 40 d after seeding and 20 d after the beginning of the water stress treatments. Table 1. Soil water potential (Ψsoil), leaf water potential (Ψleaf), Ψleaf at the turgor loss point (Ψtlp) and leaf osmotic potential at full turgor (π0) as measured in control plants (controls) and in water-stressed plants (PEG04 and PEG08)   Controls   PEG04   PEG08   Ψsoil (MPa) (n=4)  −0.22±0.10 a  −0.58±0.12 b  −0.94±0.19 c  Ψleaf (MPa) (n=6)  −0.65±0.12 a  −1.26±0.10 b  −1.59±0.08 c  Ψtlp (MPa) (n=4)  −1.02±0.11 a  −1.57±0.20 b  −1.71±0.18 b  Ψ0 (MPa) (n=4)   −0.83±0.07 a   −1.17±0.18 b   −1.24±0.03 b     Controls   PEG04   PEG08   Ψsoil (MPa) (n=4)  −0.22±0.10 a  −0.58±0.12 b  −0.94±0.19 c  Ψleaf (MPa) (n=6)  −0.65±0.12 a  −1.26±0.10 b  −1.59±0.08 c  Ψtlp (MPa) (n=4)  −1.02±0.11 a  −1.57±0.20 b  −1.71±0.18 b  Ψ0 (MPa) (n=4)   −0.83±0.07 a   −1.17±0.18 b   −1.24±0.03 b   The number of samples (n) is reported. Different letters (a, b, c) indicate significant differences for Tukey pairwise comparisons (P <0.05). View Large Table 2. Plant height (h), stem diameter (Φ) measured 5 cm above the soil, number of leaves per plant, mean leaf surface area (AL) and mean diameter of the midrib's widest conduits as measured in control plants (controls) and in water-stressed plants (PEG04 and PEG08)   Controls   PEG04   PEG08   h (m) (n=7)  0.73±0.12 a  0.57±0.04 b  0.48±0.03 c  Φ, mm (n=7)  9.7±0.6 a  8.0±0.3 b  7.0±0.2 c  No. of leaves/plant (n=7)  12.1±1.6 a  10.4±1.0 b  9.4±0.5 b  AL (cm2) (n=6)  140±27 a  113±11 b  105±4 b  Conduit diameter (μm) (n=60)   43.2±6.7 a   37.6±6.4 b   35.1±7.0 b     Controls   PEG04   PEG08   h (m) (n=7)  0.73±0.12 a  0.57±0.04 b  0.48±0.03 c  Φ, mm (n=7)  9.7±0.6 a  8.0±0.3 b  7.0±0.2 c  No. of leaves/plant (n=7)  12.1±1.6 a  10.4±1.0 b  9.4±0.5 b  AL (cm2) (n=6)  140±27 a  113±11 b  105±4 b  Conduit diameter (μm) (n=60)   43.2±6.7 a   37.6±6.4 b   35.1±7.0 b   The number of samples (n) is reported. Different letters (a, b, c) indicate significant differences for Tukey pairwise comparisons (P <0.05). View Large In response to the osmotic stress applied to roots, plants showed increasingly larger drops in Ψleaf (Table 1), from −0.65 MPa as recorded for controls at the middle of the light period, to −1.26 MPa for PEG04 plants and to −1.59 MPa for PEG08 plants. The driving force for liquid water flow through the plant body corresponding to (Ψsoil−Ψleaf), turned out to be higher for water-stressed plants (0.68 and 0.65 MPa for PEG04 and PEG08 plants, respectively) than for controls (0.43 MPa). It is of interest to note that the lower level of water stress applied (PEG04) allowed leaves to retain most of their turgor pressure. Isotherms of leaf water potential gave information of Ψleaf at the turgor loss point (Ψtlp) from which the average leaf turgor pressure (Pt=Ψleaf−Ψtlp) can be computed. This was of the order of 0.37 MPa in controls and still as high as 0.31 MPa in leaves from PEG04 plants. Plants growing at Ψsoil =−0.94 MPa (PEG08) still retained some turgor, i.e. Pt was estimated to be about 0.12 MPa. Really, Ψleaf was found not to be significantly different from Ψtlp in this case, but the statistical power of this comparison was likely to be affected by the difference in the numbers of replications (n=4 for Ψtlp and n=6 for Ψleaf). Therefore, the lack of statistical significance of the difference between Ψleaf and Ψtlp for PEG08 plants remained dubious. The general appearance of leaves of these plants, in fact, indicated that their turgor was not zero (Fig. 1). Data from Table 1 also show that water-stressed plants had adjusted leaf osmotic potential because this variable, when estimated at full turgor, was found to be significantly more negative in PEG04 plants than in controls (πo was about −1.2 MPa in PEG04 plants versus −0.83 MPa in controls). In turn, PEG08 plants showed leaf osmoregulation basically similar to that measured for PEG04 ones. Plants growing in soils with experimentally decreased Ψ showed a significant reduction in their maximum stomatal aperture, as indicated by gL that was found to be reduced by about 28% in PEG04 plants with respect to controls, i.e. from 250 to 180 mmol m−2 s−1 and further to 150 mmol m−2 s−1 in PEG08 plants (Fig. 2). It has to be noted, however, that PEG08 plants still maintained significant gas exchange, although gL of these plants was 40–50% less than that of controls. Fig. 2. View largeDownload slide Leaf conductance to water vapour (gL) as measured in the middle of the light period in sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentration (PEG04 and PEG08 plants). Mean values are reported ±standard deviation (n=12). Different letters indicate significant differences for Tukey pairwise comparisons. Fig. 2. View largeDownload slide Leaf conductance to water vapour (gL) as measured in the middle of the light period in sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentration (PEG04 and PEG08 plants). Mean values are reported ±standard deviation (n=12). Different letters indicate significant differences for Tukey pairwise comparisons. Several serial cuttings of leaf veins of the fourth or higher order (Fig. 3), followed by repeated measurements of leaf hydraulic resistance (Rleaf), allowed the resistance of the leaf venation system (Rvenation) to be estimated in terms of the residual R once this variable had become approximately constant (Sack et al., 2004). This condition was achieved after about 500 cuts had been made across minor veins. All plants subjected to water stress had Rvenation values higher than those of controls with no difference between plants growing at the two Ψsoil levels tested. The initial Rleaf values of Fig. 3, corresponding to the R of leaves with intact lamina, are reported in Fig. 4A. It can be noted that the initial Rleaf of controls and that of PEG04 plants were very similar to each other, while Rleaf recorded in PEG08 plants was one-third less than that of controls (2.4 e+3 versus about 3.2 e+3 MPa s m2 kg−1). By contrast, Rvenation was lower in controls (about 0.6 e+3 MPa s m2 kg−1) than in water-stressed plants (about 0.8 e+3 MPa s m2 kg−1) with no difference between PEG04 and PEG08 plants in this regard (Fig. 4B). The fractional amount of Rvenation with respect to total Rleaf was found to increase as a consequence of the water stress applied (Fig. 4C). The Rvenation:Rleaf ratio, in fact, was about 0.2 for leaves from plants irrigated with water and as high as 0.4 in plants irrigated with PEG solutions at π=−0.8 MPa. Controls and PEG04 plants did not differ from each other for the above ratio, statistically, although a possible tendency to higher Rvenation:Rleaf ratios might exist in the latter. When Rvenation values from all the plants studied were plotted against the diameters of the 10–15 widest conduits in the midrib, a linear, negative relationship appeared to exist between the two variables (Fig. 5) with r=−0.778 and high significance (P <0.001). This strongly suggests that the higher Rvenation recorded in water-stressed plants, with respect to controls, was at least partly due to the narrower xylem conduits in the veins of the former. Fig. 3. View largeDownload slide Leaf hydraulic resistance (Rleaf) as measured in sunflower leaves before and after an increasing number of cuts were made across minor veins. Different symbols refer to sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentration (PEG04 and PEG08). Means are reported ±standard deviation (n=6). Fig. 3. View largeDownload slide Leaf hydraulic resistance (Rleaf) as measured in sunflower leaves before and after an increasing number of cuts were made across minor veins. Different symbols refer to sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentration (PEG04 and PEG08). Means are reported ±standard deviation (n=6). Fig. 4. View largeDownload slide Hydraulic resistance of the whole leaf (A, Rleaf) and of the leaf venation system (B, Rvenation) as well as the Rvenation:Rleaf ratio (C) as measured in sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentration (PEG04 and PEG08 plants). Means are reported ±standard deviation (n=6). Different letters indicate significant differences for Tukey pairwise comparisons. Fig. 4. View largeDownload slide Hydraulic resistance of the whole leaf (A, Rleaf) and of the leaf venation system (B, Rvenation) as well as the Rvenation:Rleaf ratio (C) as measured in sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentration (PEG04 and PEG08 plants). Means are reported ±standard deviation (n=6). Different letters indicate significant differences for Tukey pairwise comparisons. Fig. 5. View largeDownload slide Relationship between the hydraulic resistance of the leaf venation system (Rvenation) and mean diameter of the midrib's widest conduits. Different symbols refer to sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentration (PEG04 and PEG08 plants). The correlation coefficient (r) and the P value are reported (Pearson Product Moment Correlation). Fig. 5. View largeDownload slide Relationship between the hydraulic resistance of the leaf venation system (Rvenation) and mean diameter of the midrib's widest conduits. Different symbols refer to sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentration (PEG04 and PEG08 plants). The correlation coefficient (r) and the P value are reported (Pearson Product Moment Correlation). Leaf conductance to water vapour (gL) turned out to be related to Rvenation with a closely linear, negative correlation between the two variables (Fig. 6) and high significance (P <0.001). The highest gL values recorded (250–300 mmol m−2 s−1) corresponded to Rvenation values of about 0.6 e+3 MPa s m2 kg−1, whereas the lowest gL values recorded (about 125 mmol m−2 s−1) corresponded to Rvenation values of about 0.97 e+3 MPa s m2 kg−1. In the inset of Fig. 6, the relationship is reported between Ψleaf and Rvenation. Because Ψleaf was not recorded on the same leaves where Rvenation was measured (see above), the correlation between these two variables was calculated on the basis of mean values recorded for each experimental group. As a consequence, the statistical power of the correlation is low. Nonetheless, the observed Ψleaf-to-Rvenation relationship (with r=−0.986; inset of Fig. 6) suggests that Ψleaf was closely related to Rvenation whose changes were apparently due to the water stress applied (Table 2; Fig. 5). In agreement with many similar reports (see below), gL also resulted to be a positive, linear function of leaf bulk water potential (Fig. 7). Fig. 6. View largeDownload slide Relationship between leaf conductance to water vapour (gL) and hydraulic resistance of the leaf venation system (Rvenation). Different symbols refer to sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentrations (PEG04 and PEG08 plants). The correlation coefficient (r) and the P value are reported (Pearson Product Moment Correlation). The inset reports the relationship between leaf water potential (Ψleaf) and Rvenation as calculated on the basis of mean values of the two parameters for the three different experimental groups. Fig. 6. View largeDownload slide Relationship between leaf conductance to water vapour (gL) and hydraulic resistance of the leaf venation system (Rvenation). Different symbols refer to sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentrations (PEG04 and PEG08 plants). The correlation coefficient (r) and the P value are reported (Pearson Product Moment Correlation). The inset reports the relationship between leaf water potential (Ψleaf) and Rvenation as calculated on the basis of mean values of the two parameters for the three different experimental groups. Fig. 7. View largeDownload slide Relationship between leaf conductance to water vapour (gL) and leaf water potential (Ψleaf) as measured in the middle of the light period in sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentrations (PEG04 and PEG08 plants). The correlation coefficient (r) and the P value are reported (Pearson Product Moment Correlation). Fig. 7. View largeDownload slide Relationship between leaf conductance to water vapour (gL) and leaf water potential (Ψleaf) as measured in the middle of the light period in sunflower plants either irrigated with tap water (controls) or water stressed by irrigating them with PEG600 solutions at progressively higher concentrations (PEG04 and PEG08 plants). The correlation coefficient (r) and the P value are reported (Pearson Product Moment Correlation). Discussion The aim of the present study was to check whether moderate water stress had an impact on the partitioning of Rleaf into its vascular and non-vascular components and how changes in Rvenation influenced leaf hydraulics and gas exchange. The results of this study provide evidence that: (i) water stress developing during plant active growth increased Rvenation consistently; (ii) although Rvenation represented a minor fraction of Rleaf, both in well-hydrated and water-stressed plants of sunflower, Rleaf was the result of parallel (and opposite) changes in the R of the two leaf compartments occurring in response to water stress; (iii) changes in the hydraulic efficiency of the leaf venation system were co-ordinated with leaf gas exchange rates. Several studies have appeared in classical, as well as recent literature, reporting the effects of experimentally induced water stress on plant hydraulic architecture and gas exchange (Sperry and Pockman, 1993; Salleo et al., 2000; Trifilò et al., 2004). In most studies, water stress was applied by depriving plants of irrigation until Ψleaf or Ψsoil reached pre-established values (Trifilò et al., 2004). In the present case, sunflower plants were grown under favourable conditions from seeding to the production of the first two leaf pairs, which took about 20 d. Then, Ψsoil was decreased by repeatedly irrigating plants with PEG solutions for 20 more days. As a consequence, the decrease in Ψsoil occurred while plants were actively growing and producing most of their leaves and the inflorescence. Because sunflower typically takes about 50–60 d to complete its life cycle from seed germination to full anthesis, the water stress period applied covered about 40–60% of the plants' life and apparently affected the differentiation and maturation of most leaves, as well as the general structure of the plant. Osmotic stress applied to roots induced the typical symptoms reported for plants growing under arid conditions, i.e. reduced plant size as due to reduced cell expansion and smaller photosynthetic surface area. Plants growing under imposed experimental water stress, showed apparent leaf osmoregulation (more negative πo with respect to controls; Table 1). This allowed PEG04 plants to increase (Ψsoil−Ψleaf), i.e. the driving force for water uptake and vertical conduction to leaves by >50% with respect to controls. This probably helped leaves from PEG04 plants to retain approximately the same turgor as control leaves did (Ψleaf−Ψtlp; Table 1). In spite of nearly equal average leaf turgor pressure in PEG04 plants with respect to controls and similar Rleaf (Fig. 4A), as well as a similar Rvenation:Rleaf ratio (Fig. 4C), recorded in the two plant groups, leaf gas exchange was significantly less in the former than in the latter, as indicated by gL of PEG04 plants that was 28% lower than that of controls (Fig. 2). This seemingly paradoxical stomatal behaviour can be explained if the increase by 20% in Rvenation of PEG04 plants with respect to controls is taken into account. This was likely to be due to the smaller diameter of xylem conduits in their leaf veins. Even if only the widest conduits in the midrib were measured for their diameter in the present study, it can be assumed that changes observed in the conduit lumen at the midrib level reflect analogous changes in the diameter of conduits of higher-order veins. In the present study, the hydraulic resistance of roots (Rroot) and stem (Rstem) or of the whole plant were not measured. Any increase of Rroot (e.g. through impairment of aquaporins) or of Rstem (due to xylem cavitation) can be expected to contribute to a reduction in leaf gas exchange rates (Sperry, 2000; Meinzer, 2002). Therefore, the possibility that the observed gL-to-Rvenation relationship might reflect the general drop in plant hydraulic conductance (Kplant), as the result of the water stress applied, cannot be ruled out. Some insight can be gained by a rough estimate of Kplant based on the ratio of leaf-level transpiration rate (data not shown) and the water potential drop between soil and leaves as calculated on the basis of data reported in Table 1. This simple exercise suggests that Kplant was in the order of 12 mmol m−2 s−1 MPa−1 in controls and dropped to about 5 and 4.5 mmol m−2 s−1 MPa−1 in PEG04 and PEG08 plants, respectively. Although this calculation does not take into proper account the whole plant transpiration rate, it nonetheless suggests that the overall Kplant was probably affected by the water stress treatment. Much to the authors' surprise, Rleaf of PEG08 plants was found to be significantly lower (by 25%) than that recorded for leaves of PEG04 and control plants, while their gL was about 40% less than that of controls (although only 17% less than that of leaves from PEG04 plants). Rvenation was nearly the same for leaves from PEG04 and PEG08 plants but the fractional amount of Rvenation was as high as 39% of the total in leaves from PEG08 plants versus only about 20% in controls. The substantially lower Rleaf recorded in PEG08 leaves might at first sight be interpreted as the result of some cellular death (Gascò et al., 2004), but this was not the case, as indicated not only by the general appearance of these leaves but also by some electrolyte leakage tests (not reported in this study). On the contrary, the lower Rleaf of PEG08 plants strongly suggests that some mechanisms were adopted by plants to favour leaf hydraulics when under severe water stress developing over the middle term. Because Rvenation was nearly the same for all water-stressed plants, it appears that the general reduction of Rleaf recorded in PEG08 leaves was entirely due to the strong reduction in the resistance of the non-vascular leaf compartment (if the vascular and the non-vascular compartments in a leaf are assumed to be arranged in series with each other; Cochard et al., 2004). The present understanding of leaf hydraulics is that the hydraulic resistance of the extra-vascular water pathway is about of the same order of magnitude as that of the vascular compartment, the former or the latter prevailing according to species-specific features (Salleo et al., 2003; Sack et al., 2005) and environmental factors (Gascò et al., 2004; Nardini et al., 2005). At present, it is not possible to discriminate between apoplastic and symplastic water pathways within the leaf extra-vascular compartment. Many years ago, Cruiziat et al. (1980) tried to get such information by comparing the time-course of rehydration of sunflower leaves previously dehydrated in a pressure chamber, with an ideal apoplastic or a symplastic hydraulic model. Experimental data fitted equally well with the two models indicating that both water pathways could be followed by water in the leaf extra-vascular compartment. It is also known that up-regulation or new expression of aquaporins can decrease leaf hydraulic resistance of sunflower during the day by about 30% (Nardini et al., 2005). Therefore, it is conceivable that the reduced Rleaf recorded in PEG08 plants was at least partly due to reduced cell membrane resistance. The possible involvement of aquaporins in the observed reduction of Rleaf in plants growing under severe water stress conditions deserves, in our opinion, further detailed studies. Leaf conductance to water vapour resulted in being a linear negative function of both Rvenation (Fig. 6) and Ψleaf (Fig. 7). In turn, Ψleaf was related to Rvenation (inset of Fig. 6). These correlations do not necessarily indicate causal relationships between the above variables. According to Tyree and Hammel (1972) and Tyree and Zimmermann (2002), Ψleaf, when measured using the pressure chamber, is a measure of the xylem pressure (Px) inside leaf veins. If this is the case, it can be concluded that stomata responded to changes in Px. In turn, Px is a function of vein hydraulic resistance according to the Ohm's law analogue. If stomata respond to Px and Px is partly determined by Rvenation, then gL can be expected to be closely related to Rvenation, as shown in Fig. 6. In conclusion, vein hydraulics and leaf gas exchange would be co-ordinated through changes in Px inside leaf veins. This interpretation of the present data would also provide an explanation for the responsiveness of stomata to Ψleaf values approaching the vein cavitation threshold (Salleo et al., 2001; Lo Gullo et al., 2003; Trifilò et al., 2003). Clearly, the relationships between leaf hydraulic construction, plant water status, and gas exchange capacity still await better elucidation and deserve more extensive studies. We are grateful to Antonio Gascò and Emmanuelle Gortan for technical assistance. References Brodribb TJ, Holbrook NM. 2003. Stomatal closure during leaf dehydration: correlation with other physiological traits. Plant Physiology  132, 2166–2173. Google Scholar Brodribb TJ, Holbrook NM. 2004. Diurnal depression of leaf hydraulic conductance in a tropical tree species. Plant, Cell and Environment  27, 820–827. Google Scholar Canny MJ. 1990. What becomes of the transpiration stream? New Phytologist  114, 341–368. Google Scholar Cochard H, Nardini A, Coll L. 2004. Hydraulic architecture of leaf blades: where is the main resistance? Plant, Cell and Environment  27, 1257–1267. Google Scholar Cruiziat P, Tyree MT, Bodet C, Lo Gullo MA. 1980. Kinetics of rehydration of detached sunflower leaves following substantial water loss. New Phytologist  84, 293–306. Google Scholar Gascò A, Nardini A, Salleo S. 2004. Resistance to water flow through leaves of Coffea arabica is dominated by extra-vascular tissues. Functional Plant Biology  31, 1161–1168. Google Scholar Henzler T, Waterhouse RN, Smyth AJ, Carvajal M, Cooke DT, Schaffner AR, Steudle E, Clarkson DT. 1999. Diurnal variations in hydraulic conductivity and root pressure can be correlated with the expression of putative aquaporins in the roots of Lotus japonicus. Planta  210, 50–60. Google Scholar Kikuta SB, Lo Gullo MA, Nardini A, Richter H, Salleo S. 1997. Ultrasound acoustic emissions from dehydrating leaves of deciduous and evergreen trees. Plant, Cell and Environment  20, 1381–1390. Google Scholar Kolb KJ, Sperry JS, Lamont BB. 1996. A method for measuring xylem hydraulic conductance and embolism in entire root and shoot systems. Journal of Experimental Botany  47, 1805–1810. Google Scholar Lo Gullo MA, Castro Noval L, Salleo S, Nardini A. 2004. Hydraulic architecture of plants of Helianthus annuus L. cv. Margot: evidence for plant segmentation in herbs. Journal of Experimental Botany  55, 1549–1556. Google Scholar Lo Gullo MA, Nardini A, Trifilò P, Salleo S. 2003. Changes in leaf hydraulics and stomatal conductance following drought stress and irrigation in Ceratonia siliqua (carob tree). Physiologia Plantarum  117, 186–194. Google Scholar Luu DT, Maurel C. 2005. Aquaporins in a challenging environment: molecular gears for adjusting plant water status. Plant, Cell and Environment  28, 85–96. Google Scholar Maurel C. 1997. Aquaporins and water permeability of plant membranes. Annual Review of Plant Physiology and Plant Molecular Biology  48, 399–429. Google Scholar Meinzer FC. 2002. Co-ordination of vapour and liquid phase water transport properties in plants. Plant, Cell and Environment  25, 265–274. Google Scholar Morillon R, Chrispeels MJ. 2001. The role of ABA and the transpiration stream in the regulation of the osmotic water permeability of leaf cells. Proceedings of the National Academy of Sciences, USA  98, 14138–14143. Google Scholar Nardini A, Salleo S, Andri S. 2005. Circadian regulation of leaf hydraulic conductance in sunflower (Helianthus annuus L. cv Margot). Plant, Cell and Environment  28, 750–759. Google Scholar Nardini A, Salleo S, Raimondo F. 2003. Changes in leaf hydraulic conductance correlate with leaf vein embolism in Cercis siliquastrum L. Trees – Structure and Function  17, 529–534. Google Scholar Nardini A, Tyree MT, Salleo S. 2001. Xylem cavitation in the leaf of Prunus laurocerasus and its impact on leaf hydraulics. Plant Physiology  125, 1700–1709. Google Scholar Sack L, Cowan PD, Jaikumar N, Holbrook NM. 2003. The ‘hydrology’ of leaves: co-ordination of structure and function in temperate woody species. Plant, Cell and Environment  26, 1343–1356. Google Scholar Sack L, Melcher PJ, Zwieniecki MA, Holbrook NM. 2002. The hydraulic conductance of the angiosperm leaf lamina: a comparison of three measurement methods. Journal of Experimental Botany  53, 2177–2184. Google Scholar Sack L, Streeter CM, Holbrook NM. 2004. Hydraulic analysis of water flow through leaves of sugar maple and red oak. Plant Physiology  134, 1824–1833. Google Scholar Sack L, Tyree MT, Holbrook NM. 2005. Leaf hydraulic architecture correlates with regeneration irradiance in tropical rainforest trees. New Phytologist  167, 403–413. Google Scholar Salleo S. 1983. Water relations parameters of two Sicilian species of Senecio (groundsel) measured by the pressure-bomb technique. New Phytologist  95, 179–188. Google Scholar Salleo S, Lo Gullo MA, Raimondo F, Nardini A. 2001. Vulnerability to cavitation of leaf minor veins: any impact on leaf gas exchange? Plant, Cell and Environment  24, 851–859. Google Scholar Salleo S, Nardini A, Lo Gullo MA. 1997. Is sclerophylly of Mediterranean evergreens an adaptation to drought? New Phytologist  135, 603–612. Google Scholar Salleo S, Nardini A, Pitt F, Lo Gullo MA. 2000. Xylem cavitation and hydraulic control of stomatal conductance in laurel (Laurus nobilis L.). Plant, Cell and Environment  23, 71–79. Google Scholar Salleo S, Raimondo F, Trifilò P, Nardini A. 2003. Axial-to-radial water permeability of leaf major veins: a possible determinant of the impact of vein embolism on leaf hydraulics? Plant, Cell and Environment  26, 1749–1758. Google Scholar Sperry JS. 2000. Hydraulic constraints on plant gas exchange. Agricultural and Forest Meteorology  104, 13–23. Google Scholar Sperry JS, Hacke UG, Wheeler JK. 2005. Comparative analysis of end wall resistivity in xylem conduits. Plant, Cell and Environment  28, 456–465. Google Scholar Sperry JS, Pockman WT. 1993. Limitations of transpiration by hydraulic conductance and xylem cavitation in Betula occidentalis. Plant, Cell and Environment  16, 279–287. Google Scholar Steudle E, Oren R, Schulze ED. 1987. Water transport in maize roots. Plant Physiology  84, 1220–1232. Google Scholar Steudle E, Peterson CA. 1998. How does water get through roots? Journal of Experimental Botany  49, 775–788. Google Scholar Trifilò P, Nardini A, Lo Gullo MA, Salleo S. 2003. Vein cavitation and stomatal behaviour of sunflower (Helianthus annuus) leaves under water limitation. Physiologia Plantarum  119, 409–417. Google Scholar Trifilò P, Raimondo F, Nardini A, Lo Gullo MA, Salleo S. 2004. Drought resistance of Ailanthus altissima: root hydraulics and water relations. Tree Physiology  24, 107–114. Google Scholar Tyree MT, Hammel HT. 1972. The measurement of the turgor pressure and water relations of plants by the pressure bomb technique. Journal of Experimental Botany  23, 267–282. Google Scholar Tyree MT, Nardini A, Salleo S, Sack L, ElOmari B. 2005. The dependence of leaf hydraulic conductance on irradiance during HPFM measurements: any role for stomatal response? Journal of Experimental Botany  56, 737–744. Google Scholar Tyree MT, Patiño S, Bennink J, Alexander J. 1995. Dynamic measurements of root hydraulic conductance using a high-pressure flowmeter in the laboratory and field. Journal of Experimental Botany  46, 83–94. Google Scholar Tyree MT, Sperry JS. 1989. The vulnerability of xylem to cavitation and embolism. Annual Review of Plant Physiology and Plant Molecular Biology  40, 19–38. Google Scholar Tyree MT, Zimmermann MH. 2002. Xylem structure and the ascent of sap, 2nd edn. Berlin: Springer Verlag. Google Scholar Ye Q, Wiera B, Steudle E. 2004. A cohesion/tension mechanism explains the gating of water channels (aquaporins) in Chara internodes by high concentration. Journal of Experimental Botany  55, 449–461. Google Scholar Zimmermann MH. 1978. Hydraulic architecture of some diffuse-porous trees. Canadian Journal of Botany  56, 2286–2295. Google Scholar Zwieniecki MA, Melcher PJ, Boyce CK, Sack L, Holbrook NM. 2002. Hydraulic architecture of leaf venation in Laurus nobilis L. Plant, Cell and Environment  25, 1445–1450. Google Scholar © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]

Hertel, Brigitte;Horváth, Ferenc;Wodala, Barnabás;Hurst, Annette;Moroni, Anna;Thiel, Gerhard

doi: 10.1093/jxb/eri307pmid: 16263909

Abstract Structural analysis of K+ channel pores suggests that the selectivity filter of the pore is an inherent sensor for extracellular K+ \((\mathrm{K}_{\mathrm{o}}^{{+}});\) channels seem to be inactivated at low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) because of a destabilization of the conducting state and a collapse of the pore. In the present study, the effect of depleting \(\mathrm{K}_{\mathrm{o}}^{{+}}\) on the activity of a plant K+ channel, KAT1, from Arabidopsis thaliana was investigated. This channel is thought to be insensitive to \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) The channel was therefore expressed in mammalian HEK293 cells and measured with patch clamp technology in the whole cell configuration. The effect of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) depletion on channel activity was monitored from the tail currents before, during, and after washing \(\mathrm{K}_{\mathrm{o}}^{{+}}\) from the medium. The data show that a depletion of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) results in a decrease in channel conductance, irrespective of whether K+ is simply removed or replaced by either Na+ or Li+. Quantitative analysis suggests that the channel has two binding sites for K+ with the dissociation constant in the order of 20 μM. This high sensitivity of the channel to \(\mathrm{K}_{\mathrm{o}}^{{+}}\) could serve as a safety mechanism, which inactivates the channel at low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) and, in this way, prevents leakage of K+ from the cells via this type of channel. Cation sensitive gating, HEK293 cells, KAT1, potassium affinity This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details) Introduction The activity of voltage-gated K+ (Kv) channels in animal cells is modulated by external \((\mathrm{K}_{\mathrm{o}}^{{+}})\) and internal K+ ions (Almers and Armstrong, 1980; Pardo et al., 1992; Baukrowitz and Yellen, 1995; Eghbali et al., 2002). When \(\mathrm{K}_{\mathrm{o}}^{{+}}\) is lowered in the external medium to millimolar concentrations, the channels reversibly inactivate. In other classes of K+ channels, such as the MaxiK channel, the proteins interact with external \(\mathrm{K}_{\mathrm{o}}^{{+}}\) with such a high affinity that its concentration has to be decreased well below 10 μM to achieve channel inactivation (Vergara et al., 1999). There are good reasons to believe that the sensitivity of K+ channels to the removal of potassium is related to a collapse of the channel pore. A crystallographic X-ray structure of the KcsA channel determined at low K+ concentration shows a distortion of the selectivity filter with respect to the reference structure obtained in high K+ (Zhou et al., 2001). A similar distorted selectivity filter is also seen in molecular dynamic (MD) simulations of the KcsA K+ carried out in the absence of ions (Bernèche and Roux, 2005). These findings led to the conclusion that a depletion of K+ results in a non-conducting state of the channel because of a partial collapse of the channel pore (Bernèche and Roux, 2005; Zhou et al., 2001). The general architecture, and particularly the pore of plant inward rectifiers, is very similar to that of animal Kv channels (Latorre et al., 2003). This similarity implies that animal and plant K+ channels share fundamental properties which are inherent in their common structure. The plant inward rectifiers, ZmK1 and ZmK2.1 from Zea mays were found to close down, very much like the animal K+ channels, at low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (Philippar et al., 2003; Su et al., 2005). By contrast, different studies report that KAT1, another Shaker-like K+ channel from Arabidopsis thaliana, is insensitive to a depletion of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (Véry et al., 1995; Brüggemann et al., 1999; Su et al., 2005). This unusual behaviour of KAT1 implies either that the architecture of the KAT1 channel is completely different from that of all the other K+ channels or that the hypothesis of a pore collapse at low K+ is not correct. In the present work, the sensitivity of KAT1 to a depletion of extracellular \(\mathrm{K}_{\mathrm{o}}^{{+}}\) is therefore readdressed. KAT1 was expressed in mammalian HEK293 cells and monitored before and during depletion of \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) A combination of an extensive washing with \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -free solution and measuring the remaining K+ contamination in the bath solution revealed that the conductance decreased steeply in a voltage-independent manner when \(\mathrm{K}_{\mathrm{o}}^{{+}}\) was decreased below millimolar concentrations. Quantitative analysis of channel conductance as a function of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) reveals that the channel has probably two binding sites with a binding constant K0.5 in the order of 20 μM. Materials and methods Transfection of mammalian cell lines For functional expression, KAT1 was transfected with 23 μg ml−1 DNA (pCB6-KAT1) into modified human HEK293 cells. HEK293 cells were transiently transfected using a standard calcium phosphate protocol. Electrophysiology Experiments were performed on cells incubated after transfection at 37 °C in 5% CO2 for 2–3 d. On the day before the experiment, cells were dispersed by trypsin, plated at a low density on 35 mm culture dishes and allowed to settle overnight. Dishes were then placed on the stage of an inverted microscope and single cells patch-clamped in the whole-cell configuration according to standard methods (Hamill et al., 1981) using an EPC-9 patch clamp amplifier (HEKA, Lambrecht, Germany). Data acquisition and analysis were performed using PULSE software (HEKA). Cells were perfused (c. 1 chamber volume min−1) at room temperature with a control solution containing 1.8 mM CaCl2, 1 mM MgCl2, 5 mM HEPES (pH 7.4), and either 20 mM KCl, LiCl, or NaCl. Choline-Cl was used to adjust the osmolarity to 300 mOsM. To remove K+ from the bath medium the incubation chamber was exchanged with nominally \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -free medium. Measurements of this solution revealed that different batches of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -free medium contained K+ contaminations ranging from 7–200 μM. Only after taking specific care (new plastic containers for storing, not using pH electrodes which had been stored in KCl) it was possible routinely to obtain solutions with a remaining contamination of c. 7 μM. This remaining contamination could not be eliminated. Patch pipettes contained 130 mM D-potassium-gluconic acid, 10 mM NaCl, 5 mM HEPES, 5 mM EGTA, 0.1 mM GTP (Na salt), 0.1 mM CaCl2 (free Ca2+c. 100 nM), 2 mM MgCl2, 5 mM phosphocreatin, and 2 mM ATP (Na salt) (pH 7.4). K+ measurements K+ concentrations in the bath solution samples were determined as described previously (Bérczi et al., 1982) using a Hitachi Z-8200 atomic absorption spectrophotometer. Fitting Data were fitted using a non-linear Marquardt–Levenburg algorithm. The goodness of fits was judged from the χ2 value. Results To examine the dependency of KAT1 activity on extracellular potassium, the channel protein was expressed in mammalian HEK293 cells. These cells are suitable for heterologous expression of an inward rectifier, because they exhibit, at voltages more negative than about 0 mV, only a very low endogenous conductance (Fig. 1A, C). The mean current at −140 mV of non- or mock-transfected cells (n=22) was only −98±35 pA. With this low background conductance the expression of recombinant KAT1 is easily detectable. Figure 1B shows the current response of a HEK293 cell transfected with kat1 DNA. When clamped from the holding voltage of −10 mV to a series of test voltages between +60 mV and −140 mV, the cell exhibits a large inward current. The mean current at −140 mV of kat1 transfected cells was 3.4±0.2 nA (n=32). This inward current exhibits the typical steady-state I/V relationship and kinetic features of KAT1 (Fig. 1B, C). The KAT1 conductance activates slowly at voltages more negative than about −60 mV and deactivates fast at a post-voltage of −10 mV. Fig. 1. View largeDownload slide Current/voltage relations of HEK293 cells after a mock transfection (A) or transfection with KAT1 DNA (B). Current responses of cells in the control bath medium with 20 mM KCl to the standard voltage protocol (top: −10 mV holding voltage, test voltages between +60 and −140 mV in steps of 20 mV, −10 mV post-voltage) were recorded in the whole cell configuration. Steady-state Iss/V relations of currents collected at the end of the test pulse (indicated by arrows) as a function of clamp voltage are shown for both cells in (C). Symbols in (C) cross reference to symbols in (A) and (B). Fig. 1. View largeDownload slide Current/voltage relations of HEK293 cells after a mock transfection (A) or transfection with KAT1 DNA (B). Current responses of cells in the control bath medium with 20 mM KCl to the standard voltage protocol (top: −10 mV holding voltage, test voltages between +60 and −140 mV in steps of 20 mV, −10 mV post-voltage) were recorded in the whole cell configuration. Steady-state Iss/V relations of currents collected at the end of the test pulse (indicated by arrows) as a function of clamp voltage are shown for both cells in (C). Symbols in (C) cross reference to symbols in (A) and (B). To examine the effect of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) on KAT1 conductance, the bath medium with 20 mM K+ was exchanged for a nominally K+-free solution. Before, during, and after removal of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (Fig. 2A) KAT1 activity was determined by analysis of the respective activation curves. Therefore, cells were clamped as in Fig. 1 from the holding voltage of −10 mV to a series of test voltages in order to activate KAT1 fully. From these voltages the membrane was stepped back to the common test voltage of −10 mV and the amplitude of the tail currents plotted against the conditioning voltage (Fig. 2B). For a quantitative comparison of the activation curves in different \(\mathrm{K}_{\mathrm{o}}^{{+}}\) the plot was expressed as cord conductance (GK) according to the equation  \[G_{\mathrm{K}}{=}I_{\mathrm{t}}/(V{-}V_{\mathrm{r}})\] (1)where It is the amplitude of the tail current, V the test voltage at which It is collected, and Vr the reversal voltage of KAT1. Figure 2B shows that the resulting GK/V relation obtained in 20 mM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) was well fitted by a Boltzmann function of the form  \[G_{\mathrm{K}}{=}1/(1{+}\mathrm{exp}(zF(V_{0.5}{-}V)/RT))\] (2)where z is the voltage-sensitive coefficient, V0.5 the voltage for half-maximal activation, and where R, T, and F have their usual thermodynamic meaning. Fitting the data in Fig. 2B with equation 2 yielded the voltage-dependent coefficient z=1.4 and a voltage for half-maximal activity V0.5= −109 mV. The maximal conductance, GK-max, obtained from the fits for experiments with 20 mM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) was set to unity. From 12 similar experiments in 20 mM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) mean values for V0.5 and z of −107±6 mV and 1.4±0.3, respectively, were obtained (Table 1). These values for the voltage-dependent coefficient are similar to those reported for KAT1 expressed in Xenopus oocytes (Becker et al., 1996; Moroni et al., 1998). However, the voltage dependency of KAT1 in HEK293 cells is shifted positive compared with that in oocytes; the estimated value for V0.5 in HEK293 cells is 40–60 mV more positive than that reported for KAT1 in oocytes (Véry et al., 1995; Moroni et al., 1998). Fig. 2. View largeDownload slide Removal of external \(\mathrm{K}_{\mathrm{o}}^{{+}}\) decreases KAT1 conductance. (A) Current responses of HEK293 cells expressing KAT1 to the standard voltage protocol (see Fig. 1) in control medium (top, 20 mM K+) and at different times after washing with K+-free medium. The last I/V scan was recorded with an external concentration of 20 μM. The tail currents at −10 mV (boxed) are magnified on the right of the current traces. (B) GK/V relation obtained from tail currents in 20 mM (solid symbols) and 20 μM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (open symbols). Data are fitted with Boltzmann equation (equation 2). (C) Development of tail current amplitudes in three cells during the removal of K+ from the bath medium. Fig. 2. View largeDownload slide Removal of external \(\mathrm{K}_{\mathrm{o}}^{{+}}\) decreases KAT1 conductance. (A) Current responses of HEK293 cells expressing KAT1 to the standard voltage protocol (see Fig. 1) in control medium (top, 20 mM K+) and at different times after washing with K+-free medium. The last I/V scan was recorded with an external concentration of 20 μM. The tail currents at −10 mV (boxed) are magnified on the right of the current traces. (B) GK/V relation obtained from tail currents in 20 mM (solid symbols) and 20 μM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (open symbols). Data are fitted with Boltzmann equation (equation 2). (C) Development of tail current amplitudes in three cells during the removal of K+ from the bath medium. Table 1. \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -sensitivity, voltage dependency, and tail current kinetics of KAT1 in control solution (20 mM K+) and after removal of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) to concentrations <100 μM without or with replacing K+ by Na+ Bath medium   K0.5/n   V0.5   z   τ   20 mM K+    −107±6 (12)  1.4±0.3 (12)  57±13 (12)  <100 μM K+, nominally          0 mM Na+  0.024/2  −106±10 (9)  1.4±0.2 (9)  46±10 (9)  <100 μM K+, 20 mM Na+   0.011/2   −114±4 (15)   1.6±0.3 (15)   45±8 (15)   Bath medium   K0.5/n   V0.5   z   τ   20 mM K+    −107±6 (12)  1.4±0.3 (12)  57±13 (12)  <100 μM K+, nominally          0 mM Na+  0.024/2  −106±10 (9)  1.4±0.2 (9)  46±10 (9)  <100 μM K+, 20 mM Na+   0.011/2   −114±4 (15)   1.6±0.3 (15)   45±8 (15)   The \(\mathrm{K}_{\mathrm{o}}^{{+}}\) concentration for half-maximal inhibition of GK-max (K0.5 in mM and the Hill coefficient n) were obtained from fits in Fig. 3. The voltage for half-maximal conductance (V0.5) and the voltage-dependent coefficient z were obtained by fitting equation 2 to GK/V data as in Fig. 2. The relaxation kinetics (τ) of the tail currents were obtained by fitting a single exponential to the currents. View Large Perfusing the bath medium with the nominally K+-free solution resulted in a progressive dilution of the external K+ concentration. Initially this caused an increase in the tail current amplitude (Fig. 2C). This increase was expected simply because of an increased thermodynamic driving force for K+. Further perfusion of the bath chamber with \(\mathrm{K}_{\mathrm{o}}^{{+}},\) however, resulted in a progressive decrease in tail currents (Fig. 2C). This decrease, which was observed in all experiments during removal of K+ from the bath medium, clearly demonstrates a change in channel activity. Continuous lowering of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) and a constant intracellular concentration should, because of the increase in the driving force for K+ efflux, have resulted in a further increase in tail current amplitude and not in a decrease. To quantify the effect of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) on KAT1 conductance, collections were taken immediately after the last I/V scan samples from the bath medium. The actual K+ concentration of this solution was determined with an atomic absorption spectrometer. In the case of the experiment shown in Fig. 2A and B, a \(\mathrm{K}_{\mathrm{o}}^{{+}}\) concentration, corresponding to the time of the last I/V scan, of 20 μM was measured. With this value and the known intracellular K+ activity, the K+ equilibrium voltage and the corresponding GK/V relationship were calculated and fitted by equation 2. To compare the two GK/V relations in low and high \(\mathrm{K}_{\mathrm{o}}^{{+}},\) Gmax obtained in 20 μM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) was expressed as a fraction of that measured in 20 mM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (Fig. 2B). Fitting the GK/V relation in low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) with equation 2 shows that the maximal chord conductance in 20 μM K+ is about 2.5 times lower than in 20 mM \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) The values for V0.5 (−109 mV) and z (1.5) in low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) are not appreciably different from those obtained in 20 mM \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Also in nine other experiments in which the extracellular K+ was reduced below 100 μM, the values for V0.5 and z were not significantly different from the reference values in 20 mM K+ (Table 1). Figure 3 shows the results of similar experiments with a plot of the relative GK-max values as a function of the corresponding \(\mathrm{K}_{\mathrm{o}}^{{+}}\) concentration. The data show that lowering of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) below about 100 μM results in a drastic decrease in GK-max. To determine the K+ concentration for half-maximal inhibition the data were fitted by a Hill equation of the form:  \[G_{\mathrm{K}{-}\mathrm{max}}{=}1/(1{+}(K_{0.5}/K_{\mathrm{o}}^{{+}})^{\mathrm{n}})\] (3)where K0.5 is the concentration for half-maximal inhibition and n denotes the Hill factor. When fitted with integer numbers for n of 1 to 3 the best results were obtained with a Hill coefficient of 2 and a K0.5 of 22 μM. Fig. 3. View largeDownload slide KAT1 cord conductance decreases in a K+-specific manner at micromolar \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Relative maximal cord conductance of KAT1 (GK-max in 20 mM K+/GK-max in low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) ) as a function of \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Data were obtained by simply removing K+ from the bath (closed symbols) or by replacing K+ by an equimolar concentration of Na+ (open symbols). Data were fitted by equation (3) yielding a \(\mathrm{K}_{\mathrm{o}}^{{+}}\) concentration for half-maximal concentration (K0.5) of 14 and 22 μM, respectively, for the presence (dashed line) or absence of Na+ (solid line) in the medium. Fig. 3. View largeDownload slide KAT1 cord conductance decreases in a K+-specific manner at micromolar \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Relative maximal cord conductance of KAT1 (GK-max in 20 mM K+/GK-max in low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) ) as a function of \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Data were obtained by simply removing K+ from the bath (closed symbols) or by replacing K+ by an equimolar concentration of Na+ (open symbols). Data were fitted by equation (3) yielding a \(\mathrm{K}_{\mathrm{o}}^{{+}}\) concentration for half-maximal concentration (K0.5) of 14 and 22 μM, respectively, for the presence (dashed line) or absence of Na+ (solid line) in the medium. To examine the effect of depleting K+ from the external medium on the kinetics of KAT1, the time-course of the tail currents was estimated in high and low \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) The exemplary data from tail currents in Fig. 4 (conditioning pulse at −140 mV, test voltage for tail current −10 mV) measured in one cell at high (20 mM) and low (20 μM) \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (same as in Fig. 2) show that depletion of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) has a minute accelerating effect on the kinetics of channel deactivation. This small effect, however, was not significant (Table 1). Fig. 4. View largeDownload slide Removal of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) has no effect on tail current kinetics. Two tail currents (top) from an HEK293 cell expressing KAT1 in the bath medium with 20 mM (black line) and 20 μM K+ (grey line). Tail currents were recorded during a 250 ms depolarization step from −140 mV to −10 mV. Scaling of both tail currents to the same ordinate (bottom) reveals that they have the same kinetics. Fig. 4. View largeDownload slide Removal of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) has no effect on tail current kinetics. Two tail currents (top) from an HEK293 cell expressing KAT1 in the bath medium with 20 mM (black line) and 20 μM K+ (grey line). Tail currents were recorded during a 250 ms depolarization step from −140 mV to −10 mV. Scaling of both tail currents to the same ordinate (bottom) reveals that they have the same kinetics. Channel inhibition is K+ selective In further experiments, the cation specificity of this decrease in GK-max was determined following \(\mathrm{K}_{\mathrm{o}}^{{+}}\) depletion The entire K+ in the bath medium (20 mM) was therefore replaced by an equimolar concentration of either Na+ or Li+. Experiments were performed as in Fig. 2 by monitoring the tail currents before and at different times after replacing K+ by either Na+ or Li+ (Fig. 5A, B). The results of these experiments show that the amplitude of tail currents also decreased in this type of experiment with time of washout (Fig. 2); the constant background of millimolar Na+ or Li+ in the bath did not, apparently, prevent this decrease in KAT1 conductance. Fig. 5. View largeDownload slide The presence of Na+ or Li+ does not prevent the decrease of KAT1 conductance upon removal of \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Development of tail current amplitudes in cells during the replacement of K+ in the bath medium by Na+ (A) or Li+ (B). The exchange of the bath medium started at time zero. Different symbols represent different cells. (C) GK/V relation obtained from tail currents in the bath medium with 20 mM K+, 0 mM Na+ (solid symbols) and 6 μM K+, 20 mM Na+ (open symbols) \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Data are fitted with the Boltzmann equation (equation 2). Fig. 5. View largeDownload slide The presence of Na+ or Li+ does not prevent the decrease of KAT1 conductance upon removal of \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Development of tail current amplitudes in cells during the replacement of K+ in the bath medium by Na+ (A) or Li+ (B). The exchange of the bath medium started at time zero. Different symbols represent different cells. (C) GK/V relation obtained from tail currents in the bath medium with 20 mM K+, 0 mM Na+ (solid symbols) and 6 μM K+, 20 mM Na+ (open symbols) \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Data are fitted with the Boltzmann equation (equation 2). For a quantitative assessment of the impact of Na+ on KAT1 gating, bath solution was again collected during and at the end of the wash-out process in order to estimate the actual K+ concentration. With this parameter the respective cord conductance of KAT1 could be estimated at high and low \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) The relevant reversal voltages were calculated from the Goldmann equation assuming a selectivity of KAT1 for K+ over Na+ of 50 (Uozumi et al., 1995). An example of a GK/V relation from one cell at high (20 mM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) /nominally 0 mM Na+) and low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (6 μM \(\mathrm{K}_{\mathrm{o}}^{{+}}\) /20 mM Na+) is illustrated in Fig. 5C. Both data sets are well fitted by equation 2 with a common V0.5 of −110 mV and a z of 1.5. A similar independency of the voltage-dependent parameters, including the kinetics of tail current deactivation on the replacement of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) by Na+, was found in all other cells tested (Table. 1). The main difference between the GK/V plots in Fig. 5 is again a drop of GK-max following the replacement of K+ by Na+. In the present case the relative GK-max (GK-max 20 mM K+/6 μM) is reduced by 4.7-fold upon lowering \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Figure 3 shows the relative GK-max of KAT1 from 15 measurements as a function of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) in a solution with a background of 20 mM Na+. The plot reveals that GK-max also decreases drastically in these experiments at concentrations below about 100 μM K+. A fit of the data with equation 3 yields a dependency of GK-max on \(\mathrm{K}_{\mathrm{o}}^{{+}},\) which is not much different from the value obtained in the absence of Na+; the fit yields a Hill coefficient of 2 and a K0.5 value of 14 μM (Table 1). This means that Na+ has no appreciable effect on the \(\mathrm{K}_{\mathrm{o}}^{{+}}\) dependent conductance of KAT1. Discussion Experimental studies on a number of animal K+ channels have revealed that a removal of extracellular K+ results in a cation-selective inactivation of channel activity; (Pardo et al., 1992; Eghbali et al., 2002; Philippar et al., 2003, Su et al., 2005). On an atomic scale this can now be explained in the context of the finding that K+ ions, but not Na+ ions, are able to stabilize the conformation of the selectivity filter in the model K+ channel pore KcsA (Zhou et al., 2001; Bernèche and Roux, 2005). The main finding in the present study is that the plant K+ inward rectifier KAT1 makes, other than suggested previously (Véry et al., 1995; Brüggemann et al., 1999; Su et al., 2005), no exception from this rule. Upon depletion of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) to μM concentrations, the channel inactivates in a voltage-independent manner. The KAT1 channel has the same overall pore architecture as other K+ channels (Latorre et al., 2003). Hence the present results imply that the stability of the pore of this plant channel is also, in solutions of very low K+, due to a binding of K+ ions to the channel; the analysis further predicts two binding sites for this action. The only remarkable feature of KAT1 compared with many other animal and plant K+ channels is the apparent high affinity of this binding site. While other channels already inactivate when \(\mathrm{K}_{\mathrm{o}}^{{+}}\) is decreased to concentrations in the millimolar range (Pardo et al., 1992; Eghbali et al., 2002; Su et al., 2005), KAT1 reaches half-maximal inhibition only at 20 μM \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) In this respect KAT1 behaves like the Ca2+-activated K+ (BKCa) channel (MaxiK-channel) in which a \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -sensitive inactivation occurred at μM concentration of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (Vergara et al., 1999). Previous investigations have detected a cation-sensitive gating mechanism in KAT1 (Moroni et al., 2000). According to this gating scheme the channel is released in a voltage and cation-dependent manner from an inhibited state into the open state. The same scheme implies that the deactivation of the channel follows different kinetics depending on whether the channel is active in high and low \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) The analysis of the present data, however, shows that the kinetics of deactivation of the channel is not affected by high and low concentrations of \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) Hence the mechanism underlying the decrease in open probability at micromolar K+ concentration must depend on a different molecular mechanism. The reduction in tail current amplitude without any kinetic changes rather suggests that it is due to a diminished number of functional channels with unchanged gating properties. These results are in accordance with the criteria of a typical C-type inactivation (Eghbali et al., 2002). Hence a removal of external K+ also seems to destabilize the conducting state in KAT1. Overall, these results suggest that in KAT1, as well as in other voltage-gated K+ channels (Yellen, 2002), the selectivity filter can act as a gate and close with low \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) The present results are in qualitative agreement with previous reports, which suggested a down-regulation in the activity of native plant K+ inward rectifiers at low K+ concentrations. However, the present straightforward estimate of the binding constant for external \(\mathrm{K}_{\mathrm{o}}^{{+}}\) to KAT1 is more than one order of magnitude lower than that obtained indirectly from previous studies. In native K+ inward rectifiers from Vicia faba guard cells, a binding constant in the order of 400 μM was estimated from competition between α-dendrotoxin (DTX) and \(\mathrm{K}_{\mathrm{o}}^{{+}}\) in intact guard cells (Obermeyer et al., 1994). A similar value was obtained from recordings in V. faba guard cell protoplasts. Depletion of the extracelluar K+ concentration to submillimolar concentrations resulted in a negative shift of the activation voltage; this resulted in a decline in channel activity at moderate membrane voltages (Schroeder and Fang, 1991). The \(\mathrm{K}_{\mathrm{o}}^{{+}}\) sensitive inhibition mechanism of KAT1 could be physiologically relevant. KAT1 seems to be predominantly expressed in guard cells where it conducts K+ uptake for stomatal opening (Thiel and Wolf, 1997). The concentration of K+ in the apoplast around guard cells occurs generally in the range of some mM (Mühling and Läuchli, 1999; Felle et al., 2000). However, this is also true for the apoplast of stomatal guard cells \(\mathrm{K}_{\mathrm{o}}^{{+}}\) activities below 50 μM as reported by Blatt (1985). Hence under conditions of severe K+ starvation and low apoplastic \(\mathrm{K}_{\mathrm{o}}^{{+}},\) a low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) poses a substantial problem to plant cells, because the driving force for K+ can, in this case, be directed in favour of K+ efflux (Maathuis and Sanders, 1993). If the K+ inward rectifier were active under these conditions K+ would leak from the cell through this channel. To judge the effect of a \(\mathrm{K}_{\mathrm{o}}^{{+}}\) controlled conductance of a K+ inward rectifier, the K+ fluxes at different \(\mathrm{K}_{\mathrm{o}}^{{+}}\) conditions were simulated in the context of the relevant transporters in a plant plasma membrane. Therefore, a modified skeleton model was used for plasma membrane transport (Gradmann et al., 1993; Thiel and Gradmann, 1994), which comprises for the present purpose a H+-ATPase, a KAT1-like K+ inward rectifier, a Cl− channel, and a 2H+/Cl− symporter with the rate constants for activation and inactivation listed in Table 2. Figure 6 illustrates the results of a simulation for the free-running membrane voltage, the ionic currents, and the changes in K+ concentration inside a cell under conditions of a low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) of 5 μM. Independent of whether the K+ inward rectifier is inhibited by low \(\mathrm{K}_{\mathrm{o}}^{{+}}\) or not, the free-running membrane voltage settles at a value positive to the K+ reversal voltage (Fig. 6A). This essentially resembles the experimentally recorded values in A. thaliana root cells with micromolar \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (Maathuis and Sanders, 1993). The simulations illustrate that, in this case, K+ leaks from the cell. However, when the experimentally determined sensitivity of the K+ inward rectifier to \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (Fig. 3) is considered, the rate of K+ leakage is drastically reduced. Based on the same simulation the rate of K+ loss from a cell was estimated over a period of 10 s of simulation (as in Fig. 6A) over a wide range of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) (Fig. 6B). It seems that under the model conditions used for the simulation, K+ starts to leak from the cells at \(\mathrm{K}_{\mathrm{o}}^{{+}}\) below c. 50 μM. In the absence of any regulation of the inward rectifier by external K+, the leakage increases steadily with lower \(\mathrm{K}_{\mathrm{o}}^{{+}}.\) On the other hand, if the experimentally determined sensitivity of the inward rectifier to \(\mathrm{K}_{\mathrm{o}}^{{+}}\) is considered in the simulation, potassium leakage is greatly reduced. Hence, a high affinity inhibition mechanism in KAT1, which selectively measures \(\mathrm{K}_{\mathrm{o}}^{{+}}\) in the apoplast and inactivates the channel if \(\mathrm{K}_{\mathrm{o}}^{{+}}\) drops to micromolar concentration, can prevent or at least reduce this unwanted leakage. Fig. 6. View largeDownload slide Simulation of electrical parameters and of changes in total cell K+ concentration \((\mathrm{K}_{\mathrm{i}}^{{+}}).\) For simulation a cell was virtually clamped to −140 mV; at time 0 the clamp was released giving rise to the free-running dynamics of the electrical parameters. (A) Simulations of dynamic behaviour of voltage, transporters (1, ATPase; 2, K+ inward rectifier; 3, Cl− channel; 4, 2H+/Cl− symporter) and \({\Delta}\mathrm{K}_{\mathrm{i}}^{{+}}\) were calculated for \(\mathrm{K}_{\mathrm{o}}^{{+}}\) of 10 μM by algorithm reported in Gradmann et al. (1993) using the parameters listed in Table 2. Data were computed considering no (left panel) or a \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -sensitive inhibition of the K+ inward rectifier (right panel) as predicted by data in Fig. 3. (B) Estimated changes in total cellular K+ concentration over 10 s of simulation as a function of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) with (open symbols) or without (closed symbols) \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -sensitive inactivation of the inward rectifier. The \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -sensitive decrease in K+ conductance was estimated from Fig. 3 using a K0.5 of 20 μM. Fig. 6. View largeDownload slide Simulation of electrical parameters and of changes in total cell K+ concentration \((\mathrm{K}_{\mathrm{i}}^{{+}}).\) For simulation a cell was virtually clamped to −140 mV; at time 0 the clamp was released giving rise to the free-running dynamics of the electrical parameters. (A) Simulations of dynamic behaviour of voltage, transporters (1, ATPase; 2, K+ inward rectifier; 3, Cl− channel; 4, 2H+/Cl− symporter) and \({\Delta}\mathrm{K}_{\mathrm{i}}^{{+}}\) were calculated for \(\mathrm{K}_{\mathrm{o}}^{{+}}\) of 10 μM by algorithm reported in Gradmann et al. (1993) using the parameters listed in Table 2. Data were computed considering no (left panel) or a \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -sensitive inhibition of the K+ inward rectifier (right panel) as predicted by data in Fig. 3. (B) Estimated changes in total cellular K+ concentration over 10 s of simulation as a function of \(\mathrm{K}_{\mathrm{o}}^{{+}}\) with (open symbols) or without (closed symbols) \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -sensitive inactivation of the inward rectifier. The \(\mathrm{K}_{\mathrm{o}}^{{+}}\) -sensitive decrease in K+ conductance was estimated from Fig. 3 using a K0.5 of 20 μM. Table 2. Parameter set for simulation according toGradmann et al. (1993)and Gradmann and Hoffstadt (1998) Transporter   E (mV)   kai (s−1)   kia (s−1)   Gmax (S m−2)   H+-ATPase  −400  10  1000  1  (2H+/Cl−)+-symporter  0  1  0.08  1  Cl− channel              Fast    0.5  25      10      0.4      Slow    1  0.1    K+ inward rectifier   \(E_{K^{{+}}}\)   100   1   0.002 to 1   Transporter   E (mV)   kai (s−1)   kia (s−1)   Gmax (S m−2)   H+-ATPase  −400  10  1000  1  (2H+/Cl−)+-symporter  0  1  0.08  1  Cl− channel              Fast    0.5  25      10      0.4      Slow    1  0.1    K+ inward rectifier   \(E_{K^{{+}}}\)   100   1   0.002 to 1   kia and kai are rate constants for activation and inactivation at zero voltage. The K+ reversal voltage \((E_{K^{{+}}})\) is calculated for the respective \(\mathrm{K}_{\mathrm{o}}^{{+}}\) and assuming a cytoplasmic activity of 100 mM. View Large * These authors contributed equally to this work. We thank Professor U Lüttge (Darmstadt) for helpful comments on the manuscript. We thank Ágnes Vashegyi for carrying out the AAS measurements. The project was supported by a traveling grant to FH form BMBF at DLR (Project Nr. UNG-030-99). References Almers W, Armstrong CM. 1980. Survival of K+ permeability and gating currents in squid axons perfused with K+ free media. Journal of General Physiology  75, 61–78. Google Scholar Baukrowitz T, Yellen G. 1995. Modulation of K+ currents by frequency and external [K+]: a tale of two inactivation mechanisms. Neurone  15, 951–960. Google Scholar Becker D, Dreyer I, Hoth S, Reid JD, Busch H, Lehnen M, Palme K, Hedrich R. 1996. Changes in voltage activation, Cs+ sensitivity, and ion permeability in H5 mutants of the plant K+ channel KAT1. Proceedings of the National Academy of Sciences, USA  93, 8123–8128. Google Scholar Bérczi A, Oláh Z, Erdei L. 1982. Potassium transport in wheat seedlings grown with different potassium supply. I. Ion contents and potassium influx. Physiologia Plantarum  55, 371–376. Google Scholar Bernèche S, Roux B. 2005. A gate in the selectivity filter of potassium channels. Structure  13, 591–600. Google Scholar Blatt MR. 1985. Extracellular potassium activity in attached leaves and its relation to stomatal function. Journal of Experimental Botany  36, 240–251. Google Scholar Brüggemann L, Dietrich P, Becker D, Dreyer I, Palme K, Hedrich R. 1999. Channel-mediated high-affinity K+ uptake into guard cells from Arabidopsis. Proceedings of the National Academy of Sciences, USA  96, 3298–3302. Google Scholar Eghbali M, Olcese R, Zarei MM, Toro L, Stefani E. 2002. External pore collapse as an inactivation mechanism for Kv4.3 K+ channels. Journal of Membrane Biology  188, 73–86. Google Scholar Felle HH, Hanstein S, Steinmeyer R, Hedrich R. 2000. Dynamics of ionic activities in the apoplast of the sub-stomatal cavity of intact Vicia faba leaves during stomatal closure evoked by ABA and darkness. The Plant Journal  24, 297–304. Google Scholar Gradmann D, Blatt MR, Thiel G. 1993. Electrocoupling of ion transporters in plants. Journal of Membrane Biology  136, 327–332. Google Scholar Gradmann D, Hoffstadt J. 1998. Electrocoupling of ion transporters in plants: interaction with internal ion concentrations. Journal of Membrane Biology  166, 51–59. Google Scholar Hamill OP, Marty A, Neher E, Sakmann B, Sigworth FJ. 1981. Improved patch-clamp technique for high-resolution recording from cells and cell-free membrane patches. Pflügers Archiv  391, 85–100. Google Scholar Latorre R, Munoz F, Gonzalez C, Cosmelli D. 2003. Structure and function of potassium channels in plants: some inferences about the molecular origin of inward rectification in KAT1 channels. Molecular and Membrane Biology  20, 19–25. Google Scholar Maathuis JM, Sanders D. 1993. Energization of potassium uptake in Arabidopsis thaliana. Planta  191, 302–307. Google Scholar Moroni A, Bardella L, Thiel G. 1998. The impermeant ion methylammonium blocks K+ and \(\mathrm{NH}_{4}^{{+}}\) currents through KAT1 channel differently: evidence for ion interaction in channel permeation. Journal of Membrane Biology  16, 25–35. Google Scholar Moroni A, Gazzarrini S, Cerana R, Colombo R, Sutter JU, DiFrancesco D, Gradmann D, Thiel G. 2000. Mutation in pore domain uncovers cation- and voltage-sensitive recovery from inactivation in KAT1 channel. Biophysics Journal  78, 1862–1871. Google Scholar Mühling KH, Läuchli A. 1999. Effect of K+ nutrition, leaf age and light intensity on apoplastic K+ in leaves of Vicia faba. Journal of Plant Nutrition and Soil Science  162, 571–576. Google Scholar Obermeyer G, Armstrong F, Blatt MR. 1994. Selective block by α-dendrotoxin of the K+ inward rectifier at the Vicia guard cell plasma membrane. Journal of Membrane Biology  137, 249–259. Google Scholar Pardo LA, Heinemann SH, Terlau H, Ludewig U, Lorra C, Pongs O, Stühmer W. 1992. Extracellular K+ specifically modulates a rat brain K+ channel. Proceedings of the National Academy of Sciences, USA  89, 2466–2470. Google Scholar Philippar K, Büchsenschütz K, Abshagen M, Fuchs I, Geiger D, Lacombe B, Hedrich R. 2003. The K+ channel KZM1 mediates potassium uptake into the phloem and guard cells of the C4 grass Zea mays. Journal of Biological Chemistry  278, 16973–16981. Google Scholar Schroeder J, Fang H. 1991. Inward-rectifying K+ channels in guard cells provide a mechanism for low-affinity K+ uptake. Proceedings of the National Academy of Sciences, USA  88, 11583–11587. Google Scholar Su Y-H, North H, Grignon C, Thibaud J-B, Sentenac H, Véry AA. 2005. Regulation by external K+ in a maize inward shaker channel targets transport activity in the high concentration range. The Plant Cell  17, 1532–1548. Google Scholar Thiel G, Gradmann D. 1994. Electrophysiology of stomata. In: Behnke H-D et al. eds. Progress in botany. Heidelberg: Springer, 59–78. Google Scholar Thiel G, Wolf A. 1997. Operation of K+ channels in stomatal movement. Trends in Plant Science  2, 339–345. Google Scholar Uozumi N, Gassmann W, Cao Y, Schroeder JI. 1995. Identification of strong modifications in cation selectivity in Arabidopsis inward rectifying potassium channel by mutant selection in yeas. Journal of Biological Chemistry  270, 24276–24281. Google Scholar Vergara C, Alvarez O, Latorre R. 1999. Localization of the K+ lock-In and the Ba2+ binding sites in a volatge gated calcium modulated channel. Implications for survival of K+ permeability. Journal of General Physiology  114, 365–376. Google Scholar Véry AA, Gaymard F, Bosseux C, Sentenac H, Thibaud J-B. 1995. Expression of a cloned plant K+ channel in Xenopus oocytes: analysis of macroscopic currents. The Plant Journal  7, 321–329. Google Scholar Yellen G. 2002. The voltage-gated potassium channels and their relatives. Nature  419, 35–42. Google Scholar Zhou Y, Morais Cabral JH, Kaufman A, MacKinnon R. 2001. Chemistry of ion hydration and coordination revealed by a K+ channel-Fab complex at 2.0 Å resolution. Nature  414, 43–48. Google Scholar © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. The online version of this article has been published under an Open Access model. Users are entitled to use, reproduce, disseminate, or display the Open Access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and the Society for Experimental Biology are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact: [email protected]

Petersen, Mette la Cour;Hejgaard, Jørn;Thompson, Gary A.;Schulz, Alexander

doi: 10.1093/jxb/eri308pmid: 16246856

Abstract Serpins are unique inhibitors of serine proteases that are located in various plant tissues and organs. An orthologue of the pumpkin (Cucurbita maxima) phloem serpin CmPS-1 was amplified from cucumber (Cucumis sativus) RNA by RT-PCR, cloned, and designated as CsPS-1 (GenBank accession no. AJ866989). Alternative amino acid sequences in the reactive centre loop suggest distinct inhibitory specificity between CmPS-1 and CsPS-1. A difference in the electrophoretic mobility of these serpins was used in heterografts to establish that serpins are phloem-mobile. Immuno light microscopy revealed that the phloem serpins are localized exclusively to sieve elements (SE), while the phloem filament protein CmPP1, used as a reference, is localized to both SEs and companion cells (CCs). Similar to CmPS-1, CsPS-1 accumulates over time in phloem exudates, indicating that serpins differ from other phloem-mobile proteins whose concentrations appear to be stable in phloem exudates. These differences could reflect alternative mechanisms regulating protein turnover and/or inaccessibility of protein degradation. The functionality of the pore/plasmodesma units connecting SEs and CCs was tested with graft-transmitted CmPP1 as a transport marker. The occurrence of CmPP1 in the CCs of the Cucumis graft partner shows that translocated 88 kDa phloem filament protein monomers can symplasmically exit the SE and accumulate in the CC. By contrast, serial sections probed with the serpin antibody demonstrate that the 43 kDa serpin does not enter CCs. Collectively, these data indicate that CCs play a decisive role in homeostasis of exudate proteins; proteins not accessing the CCs accumulate in SEs and display a time-dependent increase in concentration. Cucumis sativus, Cucurbita maxima, long-distance transport, phloem exudate, phloem protein, proteinase inhibitor, serpin Introduction The phloem of higher plants distributes carbohydrates, amino acids, and other nutrients from source to sink tissues. Recent studies have shown that hormones, mRNAs, and proteins are also transported within the phloem and that some of these transported molecules play pivotal roles in the communication between organs to co-ordinate plant development and physiology (Thompson and Schulz, 1999; Citovsky and Zambryski, 2000; Lucas et al., 2001; van Bel et al., 2002). The sieve elements (SEs) within the phloem are specialized for long-distance transport and, at functional maturity, lack the cellular machinery required for protein synthesis and processing. The closely associated companion cells (CCs) are thought to maintain the enucleated SE by symplasmically trafficking macromolecules and other compounds accessing the SEs through sieve pore/plasmodesma units (PPUs) (Oparka and Turgeon, 1999; van Bel and Knoblauch, 2000). Thus, proteins found in mature SEs can either be synthesized in the immature (nucleate) SEs or imported via PPUs from the transcriptionally and translationally active CCs. Since the early 1990s, a variety of analytical and collection techniques has been used to identify soluble proteins from vascular exudates. The phloem sap seems to be particularly rich in proteinase inhibitors, even though target proteinases have yet to be identified. The complement of proteinase inhibitors detected in phloem exudate includes a number of low molecular weight (<15 kDa) reversible proteinase inhibitors belonging to different mechanistic classes (Murray and Christeller, 1995; Xu et al., 2001; Christeller et al., 1998; Kehr et al., 1999; Schobert et al., 1998; Dannenhoffer et al., 2001). In addition to the low molecular weight inhibitors, a 43 kDa serpin (serine proteinase inhibitor) has been identified in the phloem exudates from Cucurbita maxima (CmPS-1; Yoo et al., 2000). Interestingly, the amount of CmPS-1 present in phloem exudates appears in the plants between 10 d and 14 d after germination and increases thereafter (Yoo et al., 2000), when levels of other phloem exudate proteins have stabilized (Dannenhoffer et al., 1997; Golecki et al., 1998). Serpins employ a unique irreversible mechanism of suicide-substrate inhibition. Serpins contain an exposed reactive centre loop (RCL) acting as bait for the target proteinase. During inhibition, a peptide bond at the reactive centre (P1-P1′) is cleaved and the serpin is bound covalently to the active site of the proteinase through an ester bond. Only after many hours or days is the intact proteinase released from the now inactivated serpin (for details about the serpin inhibitory mechanism, including the conformational changes during inhibition, see Gettins, 2002). Serpins are widespread in the plant kingdom, but only the properties of cereal serpins, abundant in the grains, have been characterized in detail (Dahl et al., 1996a, b; Østergaard et al., 2000; Hejgaard, 2001; Hejgaard and Hauge, 2002). Roles in defence against pathogens and/or pests have been suggested, but the physiological functions of these serpins are unknown. The serpin identified in the phloem exudate of Cucurbita maxima (CmPS-1) was shown to be an inhibitor of pancreas elastase, a non-physiological target enzyme. In vivo feeding experiments suggested a role in defence against piercing-sucking aphids, but in vitro feeding with the purified serpin failed to demonstrate a direct effect on aphid survival (Yoo et al., 2000). Serpins in mammals have evolved to act as specific regulatory inhibitors of many complex intra- as well as extracellular proteolytic systems, including complement activation and blood coagulation (Gettins, 2002). Thus, it is reasonable to hypothesize that serpins could have regulatory functions in phloem transport. An accurate analysis of phloem-mobile compounds is dependent upon the collection technique. Exudates collected from cut plant segments contain contaminants from other vascular and non-vascular cells due to the sudden pressure release that dislocates cellular contents (Lehmann, 1981). Even the elegant aphid stylectomy technique leads to small pressure changes and, thereby, potentially to local contamination. Therefore, the occurrence of a protein or mRNA in the phloem exudate neither establishes its cellular location in the intact phloem system nor its participation in long-distance transport. The cellular location can be derived from immunolocalization and in situ hybridization, although fixation-induced dislocations have in any case to be considered. Long-distance transport of macromolecules in the phloem can be tested by grafting experiments, where the formation of phloem bridges allows transport of macromolecules from one graft partner to the other (Tiedemann and Carstens-Behrens, 1994; Golecki et al., 1998). Heterografts between pumpkin and cucumber have been used to demonstrate the long-distance transport of several phloem proteins (PPI, PP2, and PFTI; Golecki et al., 1998, 1999; Dannenhoffer et al., 2001) and mRNAs (CmNACP, CmGAIP, and CmPP16; Ruiz-Medrano et al., 1999; Xoconostle-Cázares et al., 1999). A serpin has been immunolocalized to the phloem of barley, but the exact cellular localization was not determined (Roberts et al., 2003). Similarly, the cellular localization of the pumpkin phloem serpin has not been determined. The goals of the present study were to evaluate the divergence between orthologous phloem serpins by cloning a cucumber serpin cDNA, to establish the cellular localization of serpins in the phloem by immunocytochemistry, and to determine whether serpins are phloem-mobile in pumpkin–cucumber heterografts. The results indicate that CCs play a key role in phloem protein homeostasis and assigns a significant role to the PPUs that allow selected compounds of the phloem sap access to the CCs. Materials and methods Plant material Seeds from pumpkin (Cucurbita maxima Duch., cv. Gelber Zentner) and cucumber (Cucumis sativus L., cv. Hoffmanns Produkta) were germinated on wet filter-paper and transferred to wet vermiculite in 12 cm pots and then fertilized weekly with conventional soluble greenhouse fertilizer. The plants were grown in walk-in growth chambers at 25 °C at a light intensity of 150 μE with a day–night cycle of 12/12 h. Seven to ten days after germination grafts between Cucurbita maxima and Cucumis sativus were made using the approach graft technique (Golecki et al., 1999). In the figures, the heterografts are identified by letter codes, for example, Cs/Cm designates that Cucumis sativus is the scion and Cucurbita maxima the rootstock and bold indicates the sampled grafting partner (Cm). RT-PCR and sequencing Petioles from the last-formed leaf of 22-d-old Cucumis sativus plants were ground in liquid nitrogen and RNA extracted from 600 mg of tissue using the Trizol method (Invitrogen, Taastrup, Denmark; www.powow.com/akatlab/tigr.html). The reverse primer 5′-CCTCGAAGTGTTAGCTAG-3′, designed against the 3′UTR of CmPS-1, was used for first-strand cDNA synthesis from 1 μg of total RNA using the Invitrogen Superscript™ First-strand Synthesis System. From this reaction, 1 μl was PCR-amplified from the N-terminal primer 5′-GCTGCCGGAAATGGACATC-3′ (containing the start codon) and the C-terminal primer 5′-ATCCACAAGAGGGTTTAACAC-3′ (without the stop codon). PCR conditions were 98 °C for 30 s, followed by 35 cycles at 98 °C for 10 s, 55 °C for 30 s, and 72 °C for 20 s, and, finally, 72 °C for 7 min using HF phusion DNA polymerase (Medinova, Glostrup, Denmark). An 1177 bp amplicon was cloned into the pCR 2.1-TOPO vector (Invitrogen, Taastrup, Denmark) and sequenced (MWG Biotech, Eberswalde, Germany). A control PCR with these primers and total RNA from Cucurbita maxima petioles resulted in an amplicon of 1177 bp identical to CmPS-1. SDS-PAGE and immunoblotting Twenty-five days after grafting, SE exudate was collected in 2× Laemmli buffer (Laemmli, 1970). Samples from ungrafted plants were collected just below the last-formed leaf (∼1 cm in length). Samples from grafting partners were collected just above the graft union. Samples were heated to 95 °C for 5 min and stored at −20 °C. The protein concentration was measured by the Lowry method with BSA as a standard, and 30 μg of protein per sample were separated by SDS–PAGE (15% gel) and then transferred to a nitrocellulose membrane by semi-dry electroblotting. Serpins were immunodetected with a protein A-purified, polyclonal antibody (R360) raised in rabbits against recombinant barley serpin BSZx (dilution 1:5000) as the primary antibody and alkaline phosphatase conjugated goat-anti-rabbit Ig (DAKO, Glostrup, Denmark) as the secondary antibody (dilution 1:5000). Blocking, washing, and antibody incubations were performed in PBS, pH 7.4, containing 0.2% casein and 0.1% Tween-20. Goat normal serum (10%) was added to the blocking and primary antibody solutions. Proteins were visualized by chemiluminescence (CDP-star, Tropix, Applied Biosystems) or with NBT/BCIP (Sigma). Immunohistochemistry Tissue was collected 25 d after grafting. Sections (less than 2 mm on the thinnest side) from the root neck, root tips, and adventitious roots were fixed in 3.7% paraformaldehyde, 0.1% glutaraldehyde, 0.05 M cacodylate buffer for 2–2.5 h, washed 3×20 min each in 0.05 M cacodylate buffer, and dehydrated in a graded isopropanol series. Tissues from ungrafted plants were sampled at 42 DAG and identically fixed. Samples were infiltrated with paraplast and 7 μm sections were mounted on superfrost+ slides (Menzel-Gläser, Braunschweig, Germany) then allowed to dry overnight at 43 °C. Sections were de-paraffinized with histoclear, rehydrated in a graded IPA series, and transferred to TBS (10 mM TRIS, 150 mM NaCl, pH 7.5). To reduce non-specific labelling, sections were blocked with 1% BSA and 10% goat normal serum in TBS for 1 h. After blocking, the sections were briefly rinsed and protein A-purified R360 or protein A-purified non-immune serum (controls) was applied (dilution 1:600 in TBS buffer). Sections were incubated overnight at room temperature in a humid chamber. Slides were washed 3×10 min each in TBS, incubated for 1 h in the secondary antibody diluted in TBS, washed as described above, and rinsed in ddH2O for 3×5 min each. Silver-enhancement (IntenSE M, Amersham Biosciences, Hillerød, Denmark) was applied for 12–15 min according to the manufacturer's instructions. Sections were subsequently rinsed in ddH2O, 2×10 min each and observed with a Zeiss Photomikroscop II (Oberkochen, Germany) equipped with a CCD camera. For comparison of the CmPP1 and CmPS-1 labelling patterns, four serial sections were incubated with the serpin antibody on sections 1 and 3 and with the CmPP1 antibody on sections 2 and 4. Antibody concentrations were as described above. Control sections were incubated with protein A-purified non-immune serum (1:400) and ungrafted Cucumis sativus served as a negative control for CmPP1 labelling. Results To determine if an orthologue of the Cucurbita maxima phloem serpin was expressed in Cucumis sativus, RT-PCR was performed on total RNA extracted from cucumber petioles using primers designed against the phloem serpin CmPS-1 from Cucurbita maxima (Genbank accession no. AF284038; Yoo et al., 2000). A 1177 bp amplicon (CsPS-1; GenBank accession no. AJ866989) obtained from Cucumis sativus petiole tissue matched CmPS-1 in size. The nucleotide sequence translated into a sequence of 389 amino acids with a calculated molecular mass of 42601.86 Da. The amino acid sequence was compared with those of previously characterized inhibitory plant serpins and of putative serpins for which full-length sequences have been obtained (not shown). The two related Cucurbitaceae serpins, CsPS-1 and CmPS-1, have 70% identical amino acid sequences, but CsPS-1 was also similar to serpins from other dicot plant families, including putative serpins from Citrus paradisi, Malus domestica, Glycine max, Gossypium hirsutum, and Arabidopsis thaliana (61–69% amino acid identity). Sequence identity with serpins from monocots, including Triticum aestivum, Hordeum vulgare, Zea mays, and Oryza sativa was in the range of 51–55%. Most plants for which one or more serpin RCL sequences have been established, mainly based on EST analyses (not shown), contain a serpin expressed in various vegetative tissues with a conserved reactive centre sequence P2– \(\mathrm{P}{^\prime}_{1}\) Leu-Arg-Ser/Gly (Hejgaard et al., 2005). This sequence, which is the major determinant of the inhibitory specificity, is also present in CsPS-1 (Fig. 1). In addition, the RCL sequence immediately preceding the reactive centre is also highly conserved in these serpins, as illustrated by the close similarity of the P8-P1′ sequences in serpins from Cucumis, Arabidopsis, Malus, Nicotiana, and Hordeum representing five different angiosperm orders. A related serpin mRNA was detected in a phloem-specific library from Picea glauca (Fig. 1). These observations suggest that the RCL structures and, therefore, the inhibitory specificity of these plant serpins, has been highly conserved during plant evolution. Interestingly, the related and previously characterized pumpkin serpin CmPS-1 has conserved the neutral, mainly hydrophobic P7-P2 sequence, but not the basic P1 Arg (Fig. 1). Fig. 1. View largeDownload slide Reactive centre loop (RCL) sequences of CsPS-1 and selected plant serpins. Sequences between the highly conserved residues P8 Ser or Thr and P9′-P10′ Asp-Phe (Cucumis numbering). A gap with an arrow indicates the reactive centre P1-P1′ bond. Dots show identity with the highlighted CsPS-1 sequence and dashes indicate gaps introduced to maximize similarity. The RCL sequences of Cucumis sativus phloem serpin (CsPS-1) and Cucurbita maxima phloem serpin (CmPS-1) are compared with RCL sequences of serpins and putative serpins with conserved P2-P1 Leu-Arg from: Arabidopsis thaliana (At1g47710), Malus domestica (MdZ2a), Nicotiana tabacum (NtZ1), Hordeum vulgare (BSZx), and Picea glauca phloem (PgZ1). GenBank numbers are given in brackets to the right. Fig. 1. View largeDownload slide Reactive centre loop (RCL) sequences of CsPS-1 and selected plant serpins. Sequences between the highly conserved residues P8 Ser or Thr and P9′-P10′ Asp-Phe (Cucumis numbering). A gap with an arrow indicates the reactive centre P1-P1′ bond. Dots show identity with the highlighted CsPS-1 sequence and dashes indicate gaps introduced to maximize similarity. The RCL sequences of Cucumis sativus phloem serpin (CsPS-1) and Cucurbita maxima phloem serpin (CmPS-1) are compared with RCL sequences of serpins and putative serpins with conserved P2-P1 Leu-Arg from: Arabidopsis thaliana (At1g47710), Malus domestica (MdZ2a), Nicotiana tabacum (NtZ1), Hordeum vulgare (BSZx), and Picea glauca phloem (PgZ1). GenBank numbers are given in brackets to the right. Immunoblot analysis of phloem exudate proteins collected at 42 d after germination (42 DAG) using the barley BSZx antibody confirmed the presence of cross-reacting 43 kDa serpins in Cucurbita maxima and Cucumis sativus (Fig. 2, lanes 2 and 3). Similar to CmPS-1, CsPS-1 accumulated over time in the phloem exudate (Fig. 3A, B), but at different time intervals reflecting the difference in the growth rates of the two plants. The serpins could already be detected in the exudate between 10 and 14 DAG in Cucurbita maxima, whereas in CS, the proteinase inhibitor appeared between 14 and 21 DAG. In relation to the total protein content of the exudate, serpin is less concentrated in Cucumis sativus than in Cucurbita maxima, provided that the antibody binds to both serpins with the same affinity (compare Fig. 3A and B). Fig. 2. View largeDownload slide Identification of serpins in phloem exudates from Cucurbita maxima (Cm) and Cucumis sativus (Cs). Proteins in the exudate from Cucurbita maxima (lanes 1 and 2) or Cucumis sativus (lanes 3 and 4) were separated by SDS–PAGE and stained with Serva G Blue (lanes 1 and 4) or by immunoblotting with barley serpin BSZx antibody (lanes 2 and 3). Open arrowheads indicate the specific immunostaining of the Cucurbita maxima serpin CmPS-1 (lane 2) and the homologous Cucumis sativus serpin CsPS-1 (lane 3). Arrows indicate the two major exudate proteins, phloem filament protein PP1 and phloem lectin PP2, used as reference proteins in the present study. kDa, molecular weight standards on the left. Fig. 2. View largeDownload slide Identification of serpins in phloem exudates from Cucurbita maxima (Cm) and Cucumis sativus (Cs). Proteins in the exudate from Cucurbita maxima (lanes 1 and 2) or Cucumis sativus (lanes 3 and 4) were separated by SDS–PAGE and stained with Serva G Blue (lanes 1 and 4) or by immunoblotting with barley serpin BSZx antibody (lanes 2 and 3). Open arrowheads indicate the specific immunostaining of the Cucurbita maxima serpin CmPS-1 (lane 2) and the homologous Cucumis sativus serpin CsPS-1 (lane 3). Arrows indicate the two major exudate proteins, phloem filament protein PP1 and phloem lectin PP2, used as reference proteins in the present study. kDa, molecular weight standards on the left. Fig. 3. View largeDownload slide Time-dependent increase and long-distance transport of phloem serpins shown by western blotting. Exudate proteins were separated by SDS–PAGE and immunostained with anti-BSZx (A, B, C) or with anti-CmPP2 (D). (A, B) Increasing amounts of serpin were detected in Cucurbita maxima (A) or Cucumis sativus (B) from 10–42 d after germination (DAG). (C, D) Graft transmission of CmPS-1 (C) and CmPP2 (D) from Cucurbita maxima to Cucumis sativus where Cucumis sativus was the scion and Cucurbita maxima the rootstock in two independent experiments (graft a and graft b). Exudate was collected from the stock (Cs/Cm) or the scion (Cs/Cm). Control was exudate from ungrafted Cucurbita maxima and Cucumis sativus 42 DAG. Anti-CmPP2 did not cross-react with any exudate protein in the ungrafted Cucumis sativus control. Open arrowheads, presence of CmPS-1 in the Cucumis sativus grafting partner; filled arrowheads, CmPP2 in the Cucumis sativus grafting partner, kDa, molecular weight standards on the left. Fig. 3. View largeDownload slide Time-dependent increase and long-distance transport of phloem serpins shown by western blotting. Exudate proteins were separated by SDS–PAGE and immunostained with anti-BSZx (A, B, C) or with anti-CmPP2 (D). (A, B) Increasing amounts of serpin were detected in Cucurbita maxima (A) or Cucumis sativus (B) from 10–42 d after germination (DAG). (C, D) Graft transmission of CmPS-1 (C) and CmPP2 (D) from Cucurbita maxima to Cucumis sativus where Cucumis sativus was the scion and Cucurbita maxima the rootstock in two independent experiments (graft a and graft b). Exudate was collected from the stock (Cs/Cm) or the scion (Cs/Cm). Control was exudate from ungrafted Cucurbita maxima and Cucumis sativus 42 DAG. Anti-CmPP2 did not cross-react with any exudate protein in the ungrafted Cucumis sativus control. Open arrowheads, presence of CmPS-1 in the Cucumis sativus grafting partner; filled arrowheads, CmPP2 in the Cucumis sativus grafting partner, kDa, molecular weight standards on the left. Phloem transport of CmPS-1 In order to test for long-distance mobility of CmPS-1 in the phloem, Cucumis sativus was grafted upon Cucurbita maxima and 25 d later, the presence of the two serpins was tested by immunoblotting of SE exudates collected separately from stock and scion. A small difference in electrophoretic mobility between CmPS-1 and CsPS-1 (Fig. 2, lanes 2, 3) allowed for the specific detection of the two serpins (Fig. 3C). As a positive control for protein transport across the graft union, a replicate protein blot was incubated with an antibody against CmPP2, known to be phloem-mobile (Golecki et al., 1999). In both grafts (Fig. 3C: graft a and graft b), a protein with immunoreactivity and the electrophoretic mobility of CmPS-1 appeared in the Cucumis sativus grafting partner. The phloem transport capacity varied considerably between heterografts, but the ratio between translocated CmPS-1 and CmPP2 was fairly constant, as confirmed by densitometry (compare graft a and b in Fig. 3C, D). Grafts that were negative for translocation of CmPP2 were also negative for CmPS-1 (results not shown). These results confirm that CmPS-1 belongs to the group of phloem-mobile proteins (Tiedeman and Carstens-Behrens, 1994; Golecki et al., 1998). Tissue localization of serpin In order to determine whether the serpin is restricted to the phloem tissue, serpins in different regions of Cucurbita maxima or Cucumis sativus plants, including the leaf petiole, tendril (Cucumis sativus only), stem, root–shoot transition region, primary root, and adventitious root (Cucurbita maxima only), were immunolocalized. Serpin was only found in the phloem independent of the origin of the tissue. Strong labelling appeared in cells close to the cambium and in the bundle-associated phloem in the root–shoot transition region in both Cucurbita maxima and Cucumis sativus (Fig. 4A, B). Using serial sections, the localization of serpin was compared with that of another phloem exudate protein, the phloem filament protein CmPP1 (Fig. 4C). Interestingly, the localization patterns were quite different. While serpin was present in similar concentrations in all regions of the phloem, i.e. primary and secondary phloem, CmPP1 was prominent in bundle-associated phloem, present in primary bundle phloem, and least abundant in young secondary phloem (i.e. the phloem close to the cambium; compare Fig. 4A and C, arrows). As expected, controls with non-immune serum did not show any silver labelling, and the species-specific antibody against CmPP1 did not cross-react with any phloem protein in the ungrafted cucumber control in serial sections (Fig. 4D and E, respectively). Fig. 4. View largeDownload slide Immunolocalization of PS-1 and PP1 in cross and longitudinal sections through the root neck of 42-d-old, ungrafted Cucurbita maxima (A, C, F, G) and Cucumis sativus (B, D, E) plants. Serial sections were immunostained with anti-BSZx (A, B, F) and anti-CmPP1 (C, E, G), respectively. (A, B) Phloem elements all over the bundle including those close to the cambium (arrows) and bundle-associated phloem (b) were positive for serpin. (C) CmPP1 localized to phloem elements in the outer phloem region and to bundle-associated phloem (b), but rarely to the phloem elements close to the cambium (arrows). (D) Non-immune controls did not show unspecific labelling. (E) The CmPP1 antibody did not show cross-reactivity with any Cucumis sativus proteins. (F) Sieve elements close to the cambium were identified by aniline blue fluorescence of their sieve plates (open arrowheads) and were positive for serpins. (G) The sieve elements close to the cambium, identified by aniline blue fluorescence (open arrowheads), only contained minor amounts of CmPP1. b, Bundle associated phloem; c, cambium; X, xylem. Fig. 4. View largeDownload slide Immunolocalization of PS-1 and PP1 in cross and longitudinal sections through the root neck of 42-d-old, ungrafted Cucurbita maxima (A, C, F, G) and Cucumis sativus (B, D, E) plants. Serial sections were immunostained with anti-BSZx (A, B, F) and anti-CmPP1 (C, E, G), respectively. (A, B) Phloem elements all over the bundle including those close to the cambium (arrows) and bundle-associated phloem (b) were positive for serpin. (C) CmPP1 localized to phloem elements in the outer phloem region and to bundle-associated phloem (b), but rarely to the phloem elements close to the cambium (arrows). (D) Non-immune controls did not show unspecific labelling. (E) The CmPP1 antibody did not show cross-reactivity with any Cucumis sativus proteins. (F) Sieve elements close to the cambium were identified by aniline blue fluorescence of their sieve plates (open arrowheads) and were positive for serpins. (G) The sieve elements close to the cambium, identified by aniline blue fluorescence (open arrowheads), only contained minor amounts of CmPP1. b, Bundle associated phloem; c, cambium; X, xylem. This localization pattern suggests that serpin is primarily present in the young secondary phloem and CmPP1 in older parts of the vascular bundle. This result was confirmed by immunolocalization of longitudinal sections from the root–shoot transition region, again showing that serpin was localized to cells close to the cambium in the area where SEs are differentiating (Fig. 4F). By contrast, CmPP1 was localized to cells present at the periphery of the vascular bundle (Fig. 4G). The late detection of serpins in the phloem exudate led to the investigation in which seedling stage serpins start to appear. CmPS-1 transcript was detected in seedlings as early as 1 d after germination, and the protein was immunolocalized to the phloem of very young vascular bundles 2 d after germination of both Cucurbita maxima and Cucumis sativus seedlings (data not shown). This shows that serpin is already present in young sieve tubes and increases in concentration, reaching the detection limit in the exudate about 14 DAG. Serpins immunolocalize to SEs The different phloem localization patterns of serpins and CmPP1 led to the identification of the specific cell-types that accumulate serpin within the phloem. A cross-section of the external phloem in a vascular bundle within the root–shoot transition region (Fig. 5A, B) depicts a high number of serpin-positive cells in both Cucurbita maxima and Cucumis sativus (Fig. 5A, B). Under higher magnification and with aniline blue counterstaining, the serpin was shown to be SE-specific (Fig. 5C, D, arrows). The strong labelling at the sieve plates that are identified by aniline blue staining (Fig. 5E, F, arrowheads) does not necessarily reflect the in vivo distribution of serpin, but merely suggests that CmPS-1 accumulates on sieve plates during sample preparation. Fig. 5. View largeDownload slide Sieve-element specific localization of serpin in cross-sections through the root–shoot transition region of 42-d-old Cucurbita maxima (A, C, E) and Cucumis sativus (B, D, F) plants probed with anti-BSZx. (C, E) Details from (A) showing sieve-element-specific staining. (D, F) Detail from the framed area in (B) showing serpin in contiguous sieve elements, of which three show strong labelling (arrows). Aniline blue-callose fluorescence (arrowheads) identifies the sieve pores connecting the sieve elements. Arrows: sieve elements, arrowheads: sieve pores. Fig. 5. View largeDownload slide Sieve-element specific localization of serpin in cross-sections through the root–shoot transition region of 42-d-old Cucurbita maxima (A, C, E) and Cucumis sativus (B, D, F) plants probed with anti-BSZx. (C, E) Details from (A) showing sieve-element-specific staining. (D, F) Detail from the framed area in (B) showing serpin in contiguous sieve elements, of which three show strong labelling (arrows). Aniline blue-callose fluorescence (arrowheads) identifies the sieve pores connecting the sieve elements. Arrows: sieve elements, arrowheads: sieve pores. The accumulation in the exudate over development characteristic of serpins (Fig. 3) raised the question whether serpin is detectable in functional SEs. In order to test this, reciprocal graft combinations between Cucumis sativus and Cucurbita maxima were used (Cm/Cs; Fig. 6E) and the Cucurbita-specific protein CmPP1 was chosen as a molecular marker for functional SEs. The presence of CmPP1 will identify those SEs in Cucumis sativus that participate in long-distance transport of sugars and mobile proteins from scion to stock. Serial cross-sections of the root neck of the stock Cucumis sativus were incubated with antibodies against serpin or CmPP1 (Fig. 6A, B, respectively). Comparison of details (Fig. 6C, D) reveals that most SEs positive for serpin also contained CmPP1, confirming that the serpin occurs in functional SEs. Fig. 6. View largeDownload slide Immunolocalization of serpin (A, C) and CmPP1 (B, D) in serial cross-sections through the root–shoot transition region of a Cucurbita maxima/Cucumis sativus heterograft. (A, B) Most of the serpin-positive sieve elements in the Cucumis sativus stock were also positive for CmPP1, which was imported from the Cucurbita maxima scion. (C, D) Boxed regions from (A) and (B), respectively. X, xylem, PF, phloem fibres. Fig. 6. View largeDownload slide Immunolocalization of serpin (A, C) and CmPP1 (B, D) in serial cross-sections through the root–shoot transition region of a Cucurbita maxima/Cucumis sativus heterograft. (A, B) Most of the serpin-positive sieve elements in the Cucumis sativus stock were also positive for CmPP1, which was imported from the Cucurbita maxima scion. (C, D) Boxed regions from (A) and (B), respectively. X, xylem, PF, phloem fibres. Thorough analysis of different plant organs demonstrated that serpin exclusively occurs in SEs in both control and grafted plants, whereas, the adjacent CCs were unlabelled (Fig. 7A, C). By contrast, the graft-transmitted CmPP1 was present both within the SEs and the CCs, the latter showing very intense labelling (Fig. 7B, C). This implies that CmPP1 was able to pass the sieve PPUs and accumulate in the CCs. The CmPP1 present within the SEs was shown to be associated with the P-protein bodies (Fig. 7D, arrows) indicating that it had become a part of the structural P-proteins present in the SEs of the Cucumis sativus grafting partner. Fig. 7. View largeDownload slide Immunolocalization of PS-1 (A, C) and CmPP1 (B, D) in serial longitudinal sections from the root–shoot transition region of Cucurbita maxima/Cucumis sativus (Cm/Cs) heterografts. (A) Serpin localization in the Cucumis stock. (B) CmPP1 imported from the Cucurbita scion appeared in sieve elements and companion cells of the Cucumis stock. (C) Aniline-blue staining of the callose in sieve plates. (D) Detail from (B). CmPP1 occurred in the P-protein bodies (arrows) and densely in companion cells. Open arrowheads, sieve plates; CC, companion cell; N, nucleus. Fig. 7. View largeDownload slide Immunolocalization of PS-1 (A, C) and CmPP1 (B, D) in serial longitudinal sections from the root–shoot transition region of Cucurbita maxima/Cucumis sativus (Cm/Cs) heterografts. (A) Serpin localization in the Cucumis stock. (B) CmPP1 imported from the Cucurbita scion appeared in sieve elements and companion cells of the Cucumis stock. (C) Aniline-blue staining of the callose in sieve plates. (D) Detail from (B). CmPP1 occurred in the P-protein bodies (arrows) and densely in companion cells. Open arrowheads, sieve plates; CC, companion cell; N, nucleus. Discussion Serpins belong to the emerging group of phloem-localized proteinase inhibitors in cucurbits that include low molecular weight serine proteinase inhibitors of several families, cystatins, and aspartic proteinase inhibitors (Murray and Christeller, 1995; Christeller et al., 1998; Walz et al., 2004). The overall similarity between the serpins in phloem exudates of cucumber (CsPS-1) and pumpkin (CmPS-1) suggests a common evolutionary origin, but their divergence within the RCL strongly suggests that they have evolved in different directions with respect to their target proteinases. Closely related proteinase inhibitors often exhibit different substrate specificity. For example, members of the highly conserved group of PFTIs show individual specificity for either trypsin or chymotrypsin (Murray and Christeller, 1995). Collectively, proteinase inhibitors within the phloem provide a broad spectrum of substrate specificity as well as alternative mechanisms to inactivate proteinases within the assimilate transport stream or to serve as a defence against phloem-associated pathogens or insects. Long-distance transport of serpins in sieve elements The concept of ‘group transfer’ of phloem mobile proteins between graft partners (Tiedemann and Carstens-Behrens, 1994; Golecki et al., 1998) implies that the transfer of one soluble protein within the assimilate stream is indicative of the transfer of all other soluble proteins. The movement of CmPS-1 from the pumpkin to the cucumber grafting partner concomitant with the phloem mobile marker proteins CmPP1 and CmPP2, as well as the conserved ratio between the amounts of CmPS-1 and CmPP2 that are transported across the graft border (Fig. 3) agree with this concept. As expected from their movement over long distances in the phloem, serpins were shown to occur in the conducting SEs in Cucurbita maxima and Cucumis sativus. This localization pattern is similar that of CmPP16, which predominantly localizes to sieve elements (Xoconostle-Cázares et al., 1999). By contrast, most other phloem-mobile proteins accumulate in both SEs and CCs. The site of synthesis for the phloem serpins is unknown, but the proteinases could be synthesized in either immature SEs or in the CCs at undetectable levels and rapidly trafficked into the SEs through PPUs. In general, differences in the occurrence of phloem proteins in SEs and CCs might indicate a difference in the release of proteins synthesized in the CCs towards the SEs, or a difference in the rate of protein degradation in the CCs wherever recycling involves the return of the protein in question into the CC (Fisher et al., 1992). Serpin synthesis is not restricted to secondary phloem as the late appearance in the phloem exudate might suggest. The serpin was immunolocalized to the primary and secondary phloem (Fig. 4) and was detected in the protophloem as early as 2 d after germination; this is temporally similar to the initial detection of the pumpkin fruit trypsin inhibitor PFTI in the protophloem of pumpkin hypocotyls (Dannenhoffer et al., 2001). In phloem exudates, PFTIs could be detected at these early developmental stages and rapidly reached steady-state levels that remained stable throughout the subsequent development, as did the other phloem-mobile proteins except for the serpins (Clark et al., 1997; Dannenhoffer et al., 1997). Serpins seem to accumulate gradually over an extended period of time reaching a detectable level in the exudate considerably later than the other phloem-mobile proteins mentioned. The results of grafting studies appear to shed some light on a possible explanation for the differential accumulation of proteinases in the phloem. The standard marker proteins for phloem mobility, PP1 and PP2, are synthesized in CCs and then transported through PPUs into the SEs (Bostwick et al., 1992; Clark et al., 1997). Grafting studies have shown that both proteins enter the assimilate stream where they are transported across the graft union and can subsequently be detected in both SEs and CCs (Golecki et al., 1999). Although it was unclear from the initial studies performed by Golecki and coworkers, the transport experiments reported here clearly document that the 88 kDa PP1 (Leineweber et al., 2000) exits the assimilate stream in SEs and enters the CCs of the transport phloem. Thus, PPUs between SEs and CCs appear to regulate the transport of high molecular weight macromolecules in either direction. The consistent accumulation of serpins in SEs and their conspicuous absence from CCs in the transport phloem of both ungrafted and grafted plants indicate, that unlike the marker proteins, serpins cannot exit the SEs by trafficking through the PPUs linking SEs with their CCs. In general, phloem protein homeostasis in the transport phloem is most easily explained by the ability of CCs both to synthesize and to degrade translocated proteins (Fisher et al., 1992; Thompson and Schulz, 1999). Exclusion from transfer into CCs can explain the time-dependent accumulation observed for serpins (data presented here and in Yoo et al., 2000). The lack of degradation eventually leads to an accumulation of this protein within the phloem of a plant. The search for a CC targeting sequence for any long-distance transported phloem protein has yet to prove successful. The plasmodesma trafficking machinery could require a specific structural motif in combination with plasmodesma-gating regulatory proteins (Lee et al., 2003). Therefore, the absence of serpin from the CCs could be due to the lack of either the motif or a specific regulatory protein, gating PPUs to traffic serpins from the SEs into the CCs. Function of serpin in the long-distance transport system The phloem mobility of serpin and its presence within the SEs suggests that it could function in long-distance signalling and/or be involved in the protection of the long-distance transport pathway. In animals, serpins have been shown to possess distinct functions in different tissues (Silvermann et al., 2001). The presence of serpins in both phloem tissue and seed tissues of the same plant suggests that this could also be the case in plants (Roberts et al., 2003). One serpin with RCL-properties similar to those of CsPS-1, the barley serpin BSZx, has been characterized in detail (Dahl et al., 1996a, b). BSZx is an efficient irreversible inhibitor at P1 Arg of trypsin and several proteinases of the blood coagulation system with trypsin-like specificity. However, it is also an inhibitor of chymotrypsin and cathepsin G at the overlapping site P2 (Leu), and thus inhibits proteinases with specificity for basic as well as large hydrophobic residues at P1 in the protein substrates cleaved. By contrast, CmPS-1 (Yoo et al., 2000) has specificity for elastase-like proteinases with valine at the P1 position resulting in a preference for small hydrophobic residues at P1 in their substrates (Yoo et al., 2000). The role of the phloem serpins remains unclear as long as exogenous or endogenous target proteases are not identified. The gut of the phloem-feeding insect Nilaparvata lugens (rice brown plant hopper) has both cathepsin b-like and trypsin-like protease activity that appear to contribute to digestive proteolysis of the phloem sap (Foissac et al., 2002). This and other observations have led investigators to hypothesize that the primary function of phloem proteinase inhibitors is defensive. It was suggested that CmPS-1 functions as a defence protein directed against the proteases from insects (Yoo et al., 2000), but no direct interaction with proteases from Myzus persicae was demonstrated. Plants do have endogenous serine proteases (Vierstra, 1996; Beer et al., 2004), but none have been identified that can be inhibited by serpins. BLAST searches of the A. thaliana database using the trypsin sequence failed to identify homologous proteases (Silvermann et al., 2001), but such a search does not exclude the presence of target proteases in plants. These results indicate that the active form of serpin is enriched in SEs during development and that it is not subjected to protein degradation in CCs. The rapid accumulation of low molecular weight proteinase inhibitors like PFT1 and possibly also of other phloem protease inhibitors (Habu et al., 1996) protects the developing phloem, whereas serpins take over and protect against proteolytic activity at later developmental stages. This will inhibit the damaging activity of exogenous pest proteases and down-regulate the activity of endogenous target proteases. Future investigations into the role of phloem serpins will involve the search for target proteases and the silencing of serpin in selected cucurbits. Abbreviations: CC, companion cell; CmPP1, Cucurbita maxima phloem filament protein; CmPP2, Cucurbita maxima phloem lectin; DAG, days after germination; SE, sieve element; PPU, pore/plasmodesma unit; RCL, reactive centre loop. This work was supported by the VELUX Visiting Professor Programme, The Villum Kann Rasmussen Fund, Denmark to GAT, and by the Danish Research Agency to AS. References Beer EP, Jones AM, Dickerman AW. 2004. The S8 serine, C1A cystein and A1 aspartic protease families in Arabidopsis. Phytochemistry  65, 43–58. Google Scholar Bostwick DE, Dannenhoffer JM, Skaggs MI, Lister RM, Larkins BA, Thompson GA. 1992. Pumpkin phloem lectin genes are specifically expressed in companion cells. The Plant Cell  4, 1539–1548. Google Scholar Clark AM, Jacobsen KR, Bostwick DE, Dannenhoffer JM, Skaggs MI, Thompson GA. 1997. Molecular characterization of a phloem-specific gene encoding the filament protein, phloem protein (PP1) from Cucurbita maxima. The Plant Journal  12, 49–61. Google Scholar Christeller JT, Farley PC, Ramsay RJ, Sullivan PA, Laing WA. 1998. Purification, characterization and cloning of an aspartic proteinase inhibitor from squash phloem exudate. European Journal of Biochemistry  254, 160–167. Google Scholar Citovsky VV, Zambryski PC. 2000. Systemic transport of RNA in plants. Trends in Plant Science  5, 52–54. Google Scholar Dahl S, Rasmussen SK, Hejgaard J. 1996b. Heterologous expression of three plant serpins with distinct inhibitory specificities. Journal of Biological Chemistry  271, 25083–25088. Google Scholar Dahl S, Rasmussen SK, Petersen LC, Hejgaard J. 1996a. Inhibition of coagulation factors by recombinant barley serpin BSZx. FEBS Letters  394, 165–168. Google Scholar Dannenhoffer JM, Schulz A, Skaggs MI, Bostwick DE, Thompson GA. 1997. Expression of the phloem lectin is developmentally linked to vascular differentiation in cucurbits. Planta  201, 405–414. Google Scholar Dannenhoffer JM, Suhr RC, Thompson GA. 2001. Phloem-specific expression of the pumpkin fruit trypsin inhibitor. Planta  212, 155–162. Google Scholar Fisher DB, Wu Y, Ku MSB. 1992. Turnover of soluble proteins in the wheat sieve tube. Plant Physiology  100, 1433–1441. Google Scholar Foissac X, Edwards MG, Du JP, Gatehouse AMR, Gatehouse JA. 2002. Putative protein digestion in a sap-sucking homopteran plant pest (rice brown plant hopper; Nilaparvata lugens: Delphacidae): identification of trypsin-like and cathepsin B-like proteases. Insect Biochemistry and Molecular Biology  32, 967–978. Google Scholar Gettins PGW. 2002. Serpin structure, mechanism and function. Chemical Reviews  102, 4751–4804. Google Scholar Golecki B, Schulz A, Carstens-Behrens U, Kollmann R. 1998. Evidence for graft transmission of structural phloem proteins or their precursors in heterografts of Cucurbitaceae. Planta  206, 630–640. Google Scholar Golecki B, Schulz A, Thompson GA. 1999. Translocation of structural P-protein in the phloem. The Plant Cell  11, 127–140. Google Scholar Habu Y, Fokushima H, Sakata Y, Abe H, Funada R. 1996. A gene encoding a major Kunitz proteinase inhibitor of storage organs of winged beans is also expressed in the phloem of stems. Plant Molecular Biology  32, 1209–1213. Google Scholar Hejgaard J. 2001. Inhibitory serpins from rye grain with glutamine as P1 and P2 residues in the reactive center. FEBS Letters  488, 149–153. Google Scholar Hejgaard J, Laing WA, Marttila S, Gleave AP, Roberts TH. 2005. Serpins in fruit and vegetative tissues of apple (Malus domestica): expression of four serpins with distinct reactive centres and characterization of a major inhibitory seed form, MdZ1b. Functional Plant Biology  32, 517–527. Google Scholar Hejgaard J, Hauge S. 2002. Serpins of oat (Avena sativa) grain with distinct reactive centers and inhibitory specificity. Physiologia Plantarum  116, 155–163. Google Scholar Kehr J, Haebel S, Blechschmidt-Schneider S, Willmitzer L, Steup M, Fisahn J. 1999. Analysis of phloem protein patterns from different organs of Cucurbita maxima Duch. by matrix-assisted laser disorbtion/ionization time of flight mass spectrometry combined with sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Planta  207, 612–619. Google Scholar Laemmli UK. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature  227, 680–685. Google Scholar Lee JY, Yoo BC, Rojas MR, Gomez-Ospina N, Staehelin LA, Lucas WJ. 2003. Selective trafficking of non-cell-autonomous proteins mediated by NtNCAPP1. Science  17, 392–396. Google Scholar Lehmann J. 1981. Versuch zur Bestimmung der Enzymverteilung zwischen Geleitzellen und Phloemparenchymzellen bei Cucurbita pepo. Zeitschrift für Pflanzenphysiologie  103, 323–333. Google Scholar Leineweber K, Schulz A, Thompson GA. 2000. Dynamic transitions in the translocated phloem filament protein. Australian Journal of Plant Physiology  27, 733–741. Google Scholar Lucas WJ, Yoo BC, Kragler F. 2001. RNA as a long-distance information macromolecule in plants. Nature Reviews Molecular Cell Biology  2, 849–857. Google Scholar Murray C, Christeller JT. 1995. Purification of a trypsin inhibitor (PFT) from pumpkin fruit phloem exudate and isolation of a putative trypsin and chymotrypsin inhibitor cDNA clones. Biological Chemistry Hoppe-Seyler  387, 281–287. Google Scholar Oparka KJ, Turgeon R. 1999. Sieve-elements and companion cells: traffic control centers of the phloem. The Plant Cell  11, 739–750. Google Scholar Østergaard H, Rasmussen SK, Roberts TH, Hejgaard J. 2000. Inhibitory serpins form wheat grain with reactive centers resembling glutamine-rich repeats of prolamin storage proteins. Journal of Biological Chemistry  275, 33272–33279. Google Scholar Roberts TH, Marttilla S, Rasmussen SK, Hejgaard J. 2003. Differential gene expression of suicide-substrate serpin proteinase inhibitors (serpins) in vegetative and grain tissues of barley. Journal of Experimental Botany  54, 2251–2263. Google Scholar Ruiz-Medrano R, Xoconostle-Cazares B, Lucas WJ. 1999. Phloem long-distance transport of CmNACP mRNA: implications for supracellular regulation in plants. Development  126, 4405–4419. Google Scholar Schobert C, Baker L, Szederkényi J, Grossmann P, Komor E, Hayashi H, Chino M, Lucas WJ. 1998. Identification of immunologically related proteins in sieve-tube exudate collected from monocotyledonous and dicotyledonous plants. Planta  206, 245–252. Google Scholar Silvermann GA, Bird PI, Carrell RW, et al. 2001. The serpins are an expanding superfamily of structurally similar but functionally diverse proteins. Journal of Biological Chemistry  276, 33293–33296. Google Scholar Thompson GA, Schulz A. 1999. Macromolecular trafficking in the phloem. Trends in Plant Science  4, 354–360. Google Scholar Tiedemann R, Carstens-Behrens U. 1994. Influence of grafting on the phloem protein patterns in Cucurbitaceae. I. Additional phloem exudate proteins in Cucumis sativus grafted onto two Cucurbita species. Journal of Plant Physiology  143, 189–194. Google Scholar Van Bel A, Ehlers K, Knoblauch M. 2002. Sieve elements caught in the act. Trends in Plant Science  7, 126–132. Google Scholar Van Bel A, Knoblauch M. 2000. Sieve element and companion cell: the story of the comatose patient and the hyperactive nurse. Australian Journal of Plant Physiology  27, 477–487. Google Scholar Vierstra RD. 1996. Proteolysis in plants: mechanisms and functions. Plant Molecular Biology  32, 275–302. Google Scholar Walz C, Giavalisco P, Schad M, Juenger M, Klose J, Kehr J. 2004. Proteomics of curcurbit phloem exudate reveals a network of defence proteins. Phytochemistry  65, 1795–1804. Google Scholar Xoconostle-Cázares B, Xiang Y, Ruiz-Medrano R, Wang HL, Monzer J, Yoo BC, McFarland KC, Francheschi VR, Lucas WJ. 1999. Plant Paralog to viral movement protein that potentiates transport of mRNA into the phloem. Science  283, 94–98. Google Scholar Xu ZF, Qi WQ, Ouyyang XZ, Yeung E, Chye ML. 2001. A proteinase inhibitor II of Solanum americanum is expressed in phloem. Plant Molecular Biology  47, 727–738. Google Scholar Yoo BC, Aoki K, Xiang Y, Campbell LR, Hull RJ, Xoconostle-Cázares B, Monzer J, Lee JY, Ullmann DE, Lucas WJ. 2000. Characterization of Cucurbita maxima phloem serpin-1 (CmPS-1). Journal of Biological Chemistry  275, 35122–35128. Google Scholar © The Author [2005]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved. For Permissions, please e-mail: [email protected]