TY - JOUR
AU - Deighton, N.
AB - Abstract Infection of leaves of Arabidopsis thaliana with conidial suspensions of the necrotrophic pathogen Botrytis cinerea resulted in a large decrease in the level of ascorbic acid and increases in intensity of a single‐peak free radical and Fe(III) (g=4.27) signals in electron paramagnetic resonance (EPR) spectra. These changes were not confined to the spreading lesions or associated areas of chlorosis, but extended to other apparently healthy tissues in the infected leaves. They are, therefore, consistent with the existence of high levels of oxidative stress being generated as a result of the infection process. The expected accompanying increases in levels of the aldehydic products of lipid peroxidation, malondialdehyde (MDA) and 4‐hydroxy‐2‐nonenal (4‐HNE), were not observed, and in the case of MDA the levels in tissue from infected plants were appreciably lower than in the healthy controls. These last findings are surprising and demonstrate a difference in the response of A. thaliana to infection with B. cinerea compared with tissues from other plant families studied previously. Arabidopsis thaliana, ascorbate, lipid peroxidation, necrotroph, oxidative stress. Introduction A wide range of environmental stresses of biotic and abiotic origin can damage crop plants and result in substantial yield losses. The enhanced production of active oxygen species (AOS), is a common phenomenon of oxidative stress. It is also considered to be an integral part of the host defence responses (Levine et al., 1994; Daub et al., 1998), although there have been few studies of plant responses to attack by necrotrophic fungi that acquire nutrients from tissues they kill. AOS are formed in several organelles and have the potential to damage DNA, lipids and proteins (Kramer et al., 1991). Under conditions of normal healthy growth, plants possess a number of enzymatic and non‐enzymatic detoxification mechanisms to efficiently scavenge either the AOS themselves or their secondary reaction products (Bartling et al., 1993). Major enzymes of the defence system, which eliminate AOS directly, are superoxide dismutases (Kliebenstein et al., 1998), catalases (Loprasert et al., 1996) and ascorbate peroxidases (Kvaratskhelia et al., 1999), whereas antioxidants such as ascorbic acid and glutathione (GSH) are involved in the neutralization of secondary products of AOS reactions (Noctor and Foyer, 1998; Conklin et al., 2000). During prolonged periods of oxidative stress, however, these detoxification systems become overwhelmed and tissue damage results. Arabidopsis thaliana, a member of the Brassicaceae, is a widely used model plant. Botrytis cinerea is a pathogen with a wide host range, which attacks foliage, stems, flowers, and fruits, and results in serious economic damage in agriculture and horticulture (Williamson et al., 1995; Elad, 1997; Coley‐Smith et al., 1980). In this paper the roles of oxidative stress in the infection of A. thaliana by B. cinerea have been investigated. The approach that has been adopted follows closely that used previously to investigate Botrytis infection processes in fruits of Capsicum annuum (Deighton et al., 1999) and leaves of Phaseolus vulgaris (Muckenschnabel et al., 2001). This involved measuring (a) the generation of free radicals and changes in the chemical forms of oxidizable transition metal ions by electron paramagnetic resonance (EPR) spectroscopy, (b) the levels of the aldehydic products of lipid peroxidation, malondialdehyde (MDA) and 4‐hydroxy‐2‐nonenal (4‐HNE) by a combined HPLC‐mass spectrometric technique (LC‐MS), and (c) the major antioxidant, ascorbic acid, also using the LC‐MS technique. This paper highlights surprising differences in response to infection by B. cinerea between A. thaliana and tissues from other plant families studied previously. Materials and methods Plant, material and cultural conditions Plants of Arabidopsis thaliana (CS2360 Wassilewskija, Ohio State University, Columbus, USA) were grown from seeds in standardized and quality controlled compost/sand, containing 1200 l sphagnum moss peat pH 3.8–4.4 (Bordnamonia, Newbridge, Republic of Ireland), 400 l sand, 2.5 kg dolomitic limestone, 2.5 kg ground limestone, 100 l Perlite® (LBS Horticulture, Lancashire, UK), 1.5 kg immediately available fertilizers (N:P:K:MgO 12:14:24:3 plus trace elements) and 1.5 kg Celcote® (LBS Horticulture, Lancashire, UK). Seeds (c. 50) were sown in a flat 25×40 cm pot and grown in a glasshouse at 20 °C with supplementary lighting of 150 μE m−2 s−1 provided by Hg‐discharge lamps (HPLR 400 W, Philips, Belgium) to give a minimum photoperiod of 14 h light. They were mounted on a 240 V/50 Hz gear unit (HD4000, Philips, Croydon, UK) and positioned 1 m above the foliage. At an age of 15 d, the plants were transferred into individual 7 cm diameter pots and kept under the same conditions as described above. For inoculation with B. cinerea, the plants were covered by transparent plastic propagator boxes (50×40×30 cm, Fyba Pot Co. Ltd., Yorkshire, UK) to provide c. 100% humidity air, and moved to a specialist glasshouse at 18 °C with shading, air conditioning and reduced light intensity (50 μE m−2 s−1). The precise specification of light quality (Thiel et al., 1996) during the growth period, as well as humidity, temperature and light intensity during the formation of soft rots (Islam and Honda, 1998; Islam et al., 1998) was found to be essential to achieve spreading lesions. Mycological techniques: preparation of inoculum Botrytiscinerea Pers.: Fr strain B.05.10 (Quidde et al., 1999) was used throughout this work. Prior to use a conidial stock suspension was kept in 75% glycerol at −80 °C. The conidia were plated out on potato dextrose agar containing 250 g l−1 homogenate of leaves from Phaseolus vulgaris L. grown in the glasshouse. The plates were kept for 6–8 d in darkness and then placed underneath near‐UV light (Philips TL/65/80W/05, maximum emission 450 nm) for sporulation. Conidia were harvested from 14‐d‐old cultures by irrigation with inoculation medium (Gamborg's medium B5, Sigma, UK), supplemented with 10 mM sucrose, 10 mM KH2PO4 and 0.05% Tween 80. They were then filtered through cheesecloth, centrifuged at 100 g for 5 min, washed once and resuspended in the inoculation medium. For all experiments the inoculum concentration was adjusted to c. 1×105 conidia ml−1. The conidial suspension was kept for 24 h at room temperature (c. 20 °C) to allow initial formation of germ tubes. For inoculation a 5 μl droplet was placed on the upper surface of attached leaves. It was important for the formation of the soft rot lesions that the inoculation droplets did not dry out within the first 2 d. Sampling The following samples were taken 6 d after inoculation: (A) leaves of healthy control plants grown on the open glasshouse bench, (B) leaves of control plants grown in propagator boxes similar to those used for the inoculated plants, (C) green proximal areas of infected leaves, (D) chlorotic areas adjacent to the soft‐rots, with a tendency to extend to the distal regions of the leaves, and (E) rotted tissue. Samples were cut with a scalpel from the different areas of the leaves and samples from 10 leaves were pooled. Three samples of pooled material of each group (A–E) were analysed by EPR‐spectroscopy. Leaf tissue was inserted (within 30 s of sampling) into an open‐ended 3 mm diameter EPR tube and stored at −196 °C until analysis. For MS‐analyses 6–12 samples were taken per group. Fresh tissue samples were transferred immediately into 2 ml tubes containing 1 ml ice‐cold methanol and 0.2 mM butylated hydroxytoluene (BHT), then homogenized with a MiniBead Beater (Biospec Products, Bartlesville, OK, USA) using a mixture of 0.5 g of 0.1 mm diameter and 0.5 g of 0.5 mm diameter glass beads. EPR spectroscopy All results were obtained with a Bruker ESP300E computer‐controlled spectrometer (Bruker UK Ltd., Coventry, UK) operating at X‐band frequency (9.5 GHz) with an ER4103TM cylindrical cavity. Microwave generation was by means of a klystron (ER041MR) and the frequency was measured with a built‐in frequency counter. Samples were held at −196 °C in liquid nitrogen in a quartz ‘finger dewar’ (Wilmad WG‐816‐B, Fluorochem Ltd., Old Glossop, Derbyshire, UK) and spectra were obtained in 1024 points using a modulation frequency of 100 kHz. As is conventional in EPR spectroscopy, 1st derivative spectra were recorded of the microwave absorption as a function of magnetic field. Key parameters used for recording the EPR spectra are shown in Table 1, the values of the microwave power and modulation amplitude being chosen to give non‐saturating and non‐broadening conditions for the individual spectral components. The receiver gain was adjusted to optimize the signal‐to‐noise ratios for individual spectra. Table 1. EPR parameters for the detection of stable free radicals, manganese (II), and iron (III) in leaves from A. thaliana Experimental parameters   Free radical   Mn(II)   Fe(III)   Power (mW)    0.016    5   10  Amplitude (mT)    0.4    1    1.8  Centre field (mT)  336  336  157  Scan range (mT)   10  120   80  Experimental parameters   Free radical   Mn(II)   Fe(III)   Power (mW)    0.016    5   10  Amplitude (mT)    0.4    1    1.8  Centre field (mT)  336  336  157  Scan range (mT)   10  120   80  View Large Liquid chromatography‐mass spectrometry (LC‐MS) Analyses were carried out on a Finnigan MAT SSQ710C single quadrupole mass spectrometer (Finnigan Mat, San José, CA, USA) with an APCI interface (ThermoQuest, Hemel Hempstead, UK). Quantification of the dinitrophenylhydrazine (DNPH) derivatives of MDA and 4‐HNE was performed using a modification of the method described previously (Deighton et al., 1997). Separation was achieved on a Hypersil‐C18 column (Phenomenex, Macclesfield, UK) with a gradient of acetonitrile in water at a flow of 0.3 ml min−1, both containing 40 mM ammonium acetate. Detection was achieved in negative ionization mode at m/z=234, m/z=303 and m/z=335 (M−H)− for the DNPH‐derivatives of MDA, 4‐fluorobenzaldehyde (internal standard) and 4‐HNE, respectively. Quantification of ascorbic acid was performed using the method described earlier (Muckenschnabel et al., 2001). 20 μl of 400 mM dithiothreitol (DTT) solution were added to a 100 μl fraction of the supernatant obtained after homogenization of the leaf tissue in methanol/BHT as described above. DTT was added to stabilize the antioxidant in its reduced form and to minimize degradation during sample preparation. Samples were stored at −80 °C prior to analysis (not longer than 1 week). LC‐separations were carried out on a 250×4.6 mm Sphereclone‐5 μm SAX column (Phenomenex, Macclesfield, UK) using 0.1% trifluoroacetic acid as mobile phase and a flow rate of 0.25 ml min−1. Detection was achieved in positive ionization mode at m/z=177 (M+H)+. All analytical values were calculated relative to the fresh weight of the tissues. Numerical treatment of data EPR intensities were estimated from measurements of heights and widths between peak maxima and minima and then adjusted to take account of the receiver gain and the (fresh) weights of the samples. For the Fe(III) (g=4.27) signal, the results (Fig. 4) are expressed as the means and standard deviations of six replicate specimens. Other EPR signals are treated qualitatively because of difficulties in accurate estimation of intensities. Analyses for ascorbic acid, MDA and 4‐HNE (Fig. 5) are all expressed as means and standard deviations of 16 replicate samples. Results Free radical and transition metal EPR signals Examples of typical wide scan EPR spectra from healthy and rotted tissue are shown in Fig. 1. In both cases, they consist of components that can be assigned to Fe(III), Mn(II) and free radicals. Spectra of the free radical region obtained with the narrow scan range (Table 1) are shown in Fig. 2. Intense single peak free radical signals with a g‐value of 2.0035 were formed in the rotted tissue (E). The spectral characteristics of these radicals are different from those observed in non‐infected control samples (A and B), where the radicals are primarily those associated with photosynthetic processes (Kohl et al., 1965). The g‐value of the main peak in these non‐infected samples is 2.0031, which is similar to that reported for the spectrum of a cyanobacterium mutant, in which the D1 and D2 polypeptides of photosystem II did not contain the tyrosine residues responsible for the dark‐stable multiplet signal (Debus et al., 1988) (seen as the high and low field satellites in Fig. 2A–C). The spectra from the green areas of infected leaves (C) are complex, with satellite peaks equivalent to those seen with photosynthesis‐related radicals, but with a central peak that is relatively more intense than in the control tissues. In chlorotic areas adjacent to rotted tissue (D), the satellite free radical signals were no longer detectable. These spectra consisted of a single peak free radical signal with peak‐to‐peak width intermediate between that of the central peak in green tissue of infected leaves and that of the signal observed in rotted tissue (Table 2). Quantitative analysis of the individual free radical components is difficult in many of the tissues, because of overlap between the various signals. Although these spectra were recorded with a low microwave power (0.016 mW, Table 1), in order to avoid distortions in relative intensities of the different spectral components as a result of differences in their saturation characteristics, accurate deconvolution of the spectra into their individual components was impractical. The healthy tissue, for example, contains at least two signals from radicals associated with the photosynthetic pathway (Kohl et al., 1965), and the free radical associated with the rotted tissue is clearly different from either of them. An attempt has been made, however, partially to deconvolute the spectra from the C samples (i.e. those from green areas of infected leaves) using the following procedure, which assumes that part of the signal corresponds with that seen with control samples. An appropriate multiple of the spectrum from control sample A (Fig. 3ii) was subtracted from spectrum C (Fig. 3i) to produce zero intensity in the position of the low‐field resonance. This produced a residual signal (Fig. 3iii), which is believed to represent the spectrum of an additional radical in this tissue. The parameters for this spectrum (g=2.0035, peak‐to‐peak linewidth=0.58 mT) are virtually identical to those of the signal obtained with rotted tissue (Fig. 3iv), which suggests that the free radical that is associated with the rotted tissue is also formed in the green (apparently healthy) regions of infected leaves. In parallel with the generation of the single peak free radical signal in tissue from infected leaves, the EPR spectra showed a marked increase in intensity of the Fe(III) g=4.27 signal (Figs 1, 4), which corresponds to mononuclear Fe(III) complexes with (near) rhombic symmetry (Griffith, 1964). This signal was 12‐fold higher in soft rot lesions (E) than in tissue from the control samples (A, B). In chlorotic tissue (D) this signal was of similar intensity to that in rotted tissue, and it was also 4‐fold higher in green areas of infected leaves (C) than in both sets of controls. A second Fe(III) component with g=2.0, indicated by the dashed lines in Fig. 1, was also seen in the EPR spectra. This is normally assigned to magnetically‐interacting Fe(III) ions, such as occur in polymeric materials, including Fe(III) oxyhydroxides (Riemsdijk et al., 1996). Since such species make up the core of iron storage proteins, such as ferritin (Briat and Lobréaux, 1997), this component probably corresponds to iron in such forms. It has not been considered further in the present paper. The intensity of the Mn(II) EPR signal, which was seen in healthy tissue, did not change appreciably as a result of infection (Fig. 1). There were, however, qualitative differences between different samples of rotted tissues. Some of the rotted tissue samples (e.g. Fig. 1(i)) showed the presence of a second sextet component, in addition to that which is a common feature of healthy plant tissue (Lisowski et al., 1993; Zimmermann et al., 1993), whereas others did not. This second Mn(II) feature was found in c. 50% of the rotted samples in the present measurements, whereas it was always present with P. vulgaris leaves rotted with this strain of B. cinerea (B.05.10) (Muckenschnabel et al., 2001). Also, this second Mn(II) sextet was found only in rotted tissue and not in chlorotic and green tissue of infected plants or in the controls. Fig. 1. View largeDownload slide Typical wide scan EPR spectra from (i) rotted (sample E) and (ii) control leaves (samples A and B) of A. thaliana. Peaks shown correspond to Fe(III) and free radical signals, respectively, and the sextets to Mn(II). An additional signal, indicated approximately by the dashed lines, also arises from Fe(III). Note that the spectral intensities have been normalized relative to sample weight. Fig. 1. View largeDownload slide Typical wide scan EPR spectra from (i) rotted (sample E) and (ii) control leaves (samples A and B) of A. thaliana. Peaks shown correspond to Fe(III) and free radical signals, respectively, and the sextets to Mn(II). An additional signal, indicated approximately by the dashed lines, also arises from Fe(III). Note that the spectral intensities have been normalized relative to sample weight. Fig. 2. View largeDownload slide Typical free radical EPR signals from leaves of A. thaliana from (A) non‐infected plants grown under optimum conditions, (B) non‐infected plants kept under inoculation conditions, (C) green areas of infected leaves, (D) chlorotic areas adjacent to rotted tissue, and (E) rotted tissue. Fig. 2. View largeDownload slide Typical free radical EPR signals from leaves of A. thaliana from (A) non‐infected plants grown under optimum conditions, (B) non‐infected plants kept under inoculation conditions, (C) green areas of infected leaves, (D) chlorotic areas adjacent to rotted tissue, and (E) rotted tissue. Fig. 3. View largeDownload slide Partial deconvolution of the EPR spectrum of tissue from the green areas of infected leaves (i) (sample C). In this example, a multiple of 0.2 of the EPR spectrum from healthy tissue (ii) (sample A) was subtracted to yield a residual (iii). The spectrum from rotted tissue (sample E) is shown in (iv) for comparison. Note: Spectra (i) and (ii) were recorded with different receiver gains and this is taken into account in the multiplication factor. Fig. 3. View largeDownload slide Partial deconvolution of the EPR spectrum of tissue from the green areas of infected leaves (i) (sample C). In this example, a multiple of 0.2 of the EPR spectrum from healthy tissue (ii) (sample A) was subtracted to yield a residual (iii). The spectrum from rotted tissue (sample E) is shown in (iv) for comparison. Note: Spectra (i) and (ii) were recorded with different receiver gains and this is taken into account in the multiplication factor. Fig. 4. View largeDownload slide Typical Fe(III) (g=4.27) EPR spectra from leaves of A. thaliana and their mean relative intensities. (A, B) Non‐infected control plants, (C) green areas of infected leaves, (D) chlorotic areas adjacent to rotted tissue, and (E) rotted tissue (n=6; bar=standard deviation). Fig. 4. View largeDownload slide Typical Fe(III) (g=4.27) EPR spectra from leaves of A. thaliana and their mean relative intensities. (A, B) Non‐infected control plants, (C) green areas of infected leaves, (D) chlorotic areas adjacent to rotted tissue, and (E) rotted tissue (n=6; bar=standard deviation). Table 2. Peak‐to‐peak width of the EPR spectrum from tissues of A. thaliana (n=9) Sample   Width (mT)   A: Healthy  0.956±0.005a  B: Control  0.945±0.006a  C: Green  0.877±0.007  D: Chlorotic  0.667±0.007  E: Rotted  0.556±0.005  Sample   Width (mT)   A: Healthy  0.956±0.005a  B: Control  0.945±0.006a  C: Green  0.877±0.007  D: Chlorotic  0.667±0.007  E: Rotted  0.556±0.005  aPeak‐to‐peak width of the main peak. View Large Ascorbic acid A 30‐fold decrease in ascorbic acid concentration was observed between the leaves of control plants (A, B) and those that were infected with B. cinerea (C–E) (Fig. 5). Levels of ascorbic acid in green areas of infected leaves, chlorotic and rotted tissues were, however, all similar. Dehydroascorbate (m/z=175 in positive ionization mode) was not detected in any of these assays, as expected if the DTT and BHT in the extraction solutions convert the free antioxidant to its reduced form. Fig. 5. View largeDownload slide Mean ascorbic acid, MDA and 4‐HNE contents of leaves of A. thaliana. (A) Non‐infected plants grown under optimum conditions, (B) non‐infected plants kept under inoculation conditions, (C) green areas of infected leaves, (D) chlorotic areas adjacent to rotted tissue, and (E) rotted tissue (n=16; bar=standard deviation). Fig. 5. View largeDownload slide Mean ascorbic acid, MDA and 4‐HNE contents of leaves of A. thaliana. (A) Non‐infected plants grown under optimum conditions, (B) non‐infected plants kept under inoculation conditions, (C) green areas of infected leaves, (D) chlorotic areas adjacent to rotted tissue, and (E) rotted tissue (n=16; bar=standard deviation). MDA and 4‐HNE The concentrations of MDA and 4‐HNE were lower in rotted tissue than in non‐infected controls (Fig. 5). The MDA concentrations in control B samples (i.e. those non‐infected plants under the same experimental conditions as the infected specimens) were somewhat higher than in the control A samples (from plants on the open bench in the glasshouse). In the latter samples the MDA concentrations were similar to those in green tissue (C) from infected leaves and c. 2‐fold higher than in chlorotic (D) and rotted (E) tissue. The trends in 4‐HNE concentrations were comparable to those seen with MDA, with the lowest values in chlorotic tissue, slightly higher values in rotted tissue and in green areas of infected leaves and the highest values in control samples. Discussion Decreases in the levels of antioxidants, such as ascorbic acid, are expected consequences of oxidative stress in biological tissues (Dai et al., 1997; Noctor and Foyer, 1998). In addition, the fungus may utilize ascorbic acid, which leads to the production of oxalate and H2O2 (Loewus, 1999). If iron(II) is present, H2O2 is broken down to yield ·OH radicals. These radicals are highly reactive and, in the presence of O2, oxalate is broken down by a radical chain reaction (Park et al., 1997). Many other molecules are also attacked by ·OH and the net result is indiscriminate tissue damage in the host plant. Conklin et al. have shown that an ascorbic acid‐deficient Arabidopsis mutant was very sensitive to a range of environmental stresses, an observation which demonstrates the key protective role for this molecule in Arabidopsis foliar tissues (Conklin et al., 1996). A direct protective role for ascorbic acid has also been demonstrated in rice, Oryza sativa, where partial protection against damage caused by a release from flooding conditions was provided by the prior addition of ascorbic acid (Thongbai and Goodman, 2000). In the present work, tissue levels of ascorbic acid showed a 30‐fold decrease upon infection, indicating that a massive failure of the antioxidative defence system had occurred, even in the green regions of infected leaves. The stressed control plants (B), however, showed only minor decreases in ascorbic acid levels compared to those in the control A plants. This observation is in contrast to that seen with leaves of P. vulgaris, where large reductions in ascorbic acid were seen under similar stress conditions (Muckenschnabel et al., 2001). In previous work on leaves of P. vulgaris, it has been shown that losses of ascorbic acid as a result of infection by B. cinerea are accompanied by increases in the Fe(III) (g=4.27) and single peak stable free radical EPR signals (Muckenschnabel et al., 2001). Similar increases in the Fe(III) and free radical signals have been reported for Botrytis‐infected fruit of C. annuum (Deighton et al., 1999). In A. thaliana, the iron(III) (g=4.27) signal showed a progressive increase from healthy non‐infected leaves to green areas of infected leaves and to chlorotic areas of infected leaves. This signal was, however, of similar intensity in rotted and chlorotic areas of the leaves. By contrast, the single peak free radical signal was much higher in rotted tissue than in the adjacent chlorotic areas. The signal from the photosynthesis‐derived radicals decreased as the single peak resonance increased from the control to the chlorotic tissues, and was undetectable in these latter samples. This observation is in contrast to the results from Botrytis‐infected leaves of P. vulgaris (Muckenschnabel et al., 2001) where no chlorotic halos developed adjacent to the rotted tissue and free radicals associated with photosynthesis were detectable in tissues adjacent to the soft‐rot. The data in the present work suggest a definite sequence in the dynamic events that occur as a result of the infection process of A. thaliana by B. cinerea. A decrease in ascorbic acid levels occurs in advance of the increases in the Fe(III) (g=4.27) and single peak free radical EPR signals, showing that these EPR‐detectable products only accumulate after this antioxidant has been severely depleted. Nevertheless, the observation of elevated Fe(III) and free radical EPR signals in the green areas of infected leaves, relative to the levels in control tissues, indicates considerable oxidative damage in the apparently healthy tissue of infected leaves. However, whereas depletion of ascorbic acid was essentially complete in the green areas of infected leaves, the EPR signals showed further increases in the chlorotic tissue adjacent to the soft‐rot lesions. Whilst the signal intensity for Fe(III) was similar in rotted tissue and in adjacent chlorotic areas, the free radical signal intensity was much higher in the soft rots. This may represent further oxidative damage to the plant tissue, but it is also possible that it is related to fungal growth and the onset of sporulation. In cultures grown on agar, a dramatic increase in the free radical signal intensity was seen for B. cinerea when it started sporulation (unpublished results). The pattern for the decrease in the photosynthesis‐related radicals was difficult to determine accurately, but it appeared to be similar to that for the increase in the Fe(III) (g=4.27) signal. It was certainly slower than the decrease in ascorbic acid, since appreciable quantities of these radicals were detected in the green areas of infected leaves. It seems likely, therefore, that damage to the photosynthetic apparatus occurs after the depletion of endogenous antioxidants. Although a cluster of four Mn(II) ions represents a fundamental component of photosystem II, it is not possible in the present experiments to say whether or not its breakdown is the source of the 2nd Mn(II) which occurs in rotted tissue. What evidence there is suggests that it is not, because the appearance of the Mn(II) component follows a different temporal path to the disappearance of the photosynthesis‐derived free radical signals in the EPR spectra, but further studies need to be carried out before a conclusive statement can be made. The most surprising observation in the present work is the decrease in the levels of MDA and 4‐HNE in the tissues from the infected plants. This is in complete contrast to the observations made with tissues from other plant families inoculated with this pathogen (Deighton et al., 1999; Muckenschnabel et al., 2001), and the general acceptance of these molecules as markers of oxidative stress (Liu et al., 1997; Laurent et al., 1999; Petersen et al., 1999). It is also in contrast to the results of Conklin et al., who showed that increased MDA levels (as measured by TBARS) were formed in an ascorbic acid deficient Arabidopsis mutant when subjected to oxidative stress (by ozone) (Conklin et al., 1996). Nevertheless, the occurrence of oxidative stress in the B. cinerea‐infected leaves of A. thaliana in the present work was clearly established by the marked decrease in ascorbic acid levels and increases in the single peak free radical and Fe(III) (g=4.27) EPR signals relative to the controls. It must be emphasized that the levels of MDA and HNE in the healthy Arabidopsis tissue were not exceptional and were similar to those observed with healthy leaves of P. vulgaris (Muckenschnabel et al., 2001). The MDA values were, however, much lower than those reported previously (Conklin et al., 1996), even after allowing for the fact that their analyses are expressed relative to the dry weight of the sample and those in the present work are relative to the wet weight. (Interestingly, the ascorbic acid levels determined by the two groups for their control samples are of comparable magnitude.) The different MDA values probably reflect fundamental differences in the methodologies used. In the present work only the free forms of MDA and 4‐HNE are measured with the LC‐MS method, and adducts with other molecules are not detected. With the TBARS method, there may be the possibility of releasing weakly bound adducts during sample preparation, but the method is also known to be non‐specific for MDA, with non‐lipid related materials as well as lipid‐derived materials other than MDA being known to provide a positive response (Janero, 1990). In the present experiment, the observed decrease in MDA and HNE between healthy and rotted tissue may be a consequence of the limited stability of the aldehydes in biological samples (Esterbauer et al., 1976). This could then lead to a progressive decrease in their levels with time after the tissue has been fully rotted. It is also possible that A. thaliana leaf tissue may be more effective at removing them from stressed/rotted tissue than the other plant families that have been investigated. Conclusions Infection of leaves of A. thaliana by B. cinerea resulted in oxidative damage as observed by decreases in levels of ascorbic acid and increases in the EPR signals from Fe(III) and stable free radicals. Massive depletion of ascorbic acid levels occurred before visible infection and the generation of the EPR signals associated with the plant–pathogen interaction, indicating that damage to the antioxidant mechanisms (i.e. redox status) represents an early event in the infection process. There were a number of differences in the behaviours of A. thaliana and P. vulgaris (Muckenschnabel et al., 2001) in response to infection by B. cinerea. Most notably, oxidative damage in A. thaliana was not accompanied by increases in the levels of MDA and 4‐HNE, which are generally considered to be good markers of oxidative stress in biological systems. Caution should, therefore, be exercized in extrapolating results from A. thaliana to the behaviour of other plant species. 1 To whom correspondence should be addressed. Fax: +44 (0)1382 56246. E‐mail: B.Goodman@scri.sari.ac.uk This project was jointly funded by the European Union (FAIR project no. PL97‐3351) and the Scottish Executive Rural Affairs Department (SERAD). We also thank Professor Paul Tudzynski for providing a culture of the B. cinerea, Professor Alex Levine for supplying A. thaliana seeds, and Dr Paul Wood for helpful discussions. References Bartling D, Radzio R, Steiner U, Weiler EW. 1993. A glutathione‐S‐transferase with glutathione‐peroxidase‐activity from Arabidopsis thaliana—molecular cloning and functional characterization. European Journal of Biochemistry  216, 579–586. CrossRef Search ADS PubMed  Google Scholar Briat JF, Lobréaux S. 1997. Iron transport and storage in plants. Trends in Plant Science  2, 187–193. CrossRef Search ADS   Google Scholar Coley‐Smith JR, Verhoeff K, Jarvis WR. 1980. The biology of  Botrytis. London: Academic Press. Google Scholar Conklin PL, Saracco SA, Norris SR, Last RL. 2000. Identification of ascorbic acid‐deficient Arabidopsis thaliana mutants. Genetics  154, 847–856. PubMed  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. CrossRef Search ADS   Google Scholar Dai QJ, Yan B, Huang SB, Liu XZ, Peng SB, Miranda MLL, Chavez AQ, Vergara BS, Olszyk DM. 1997. Response of oxidative stress defence systems in rice (Oryza sativa) leaves with supplemental UV‐B radiation. Physiologia Plantarium  101, 301–308. CrossRef Search ADS   Google Scholar Daub ME, Ehrenshaft M, Jenns AE, Chung KR. 1998. Active oxygen in fungal pathogenesis of plants. In: Romeo et al., eds. Phytochemical signals and plant–microbe interactions . New York: Plenum Press. Google Scholar Debus RJ, Barry BA, Babcock GT, McIntosh L. 1988. Site‐directed mutagenesis identifies a tyrosine radical involved in the photosynthetic oxygen‐evolving system. Proceedings of the National Academy of Sciences, USA  85, 427–430. CrossRef Search ADS   Google Scholar Deighton N, Magill WJ, Bremner DH, Benson EE. 1997. Malondialdehyde and 4‐hydroxy‐2‐nonenal in plant tissue cultures: LC‐MS determination of 2,4‐dinitrophenylhydrazone derivatives. Free Radical Research  27, 255–265. CrossRef Search ADS PubMed  Google Scholar Deighton N, Muckenschnabel I, Goodman BA, Williamson B. 1999. Lipid peroxidation and the oxidative burst associated with infection of Capsicum annuum by Botrytis cinerea. The Plant Journal  20, 485–492. CrossRef Search ADS PubMed  Google Scholar Elad Y. 1997. Responses of plants to infection by Botrytis cinerea and novel means involved in reducing their susceptibility to infection. Biological Reviews of the Cambridge Philosophical Society  72, 381–422. CrossRef Search ADS   Google Scholar Esterbauer H, Ertl A, Scholz N. 1976. The reaction of cysteine with α,β‐unsaturated aldehydes. Tetrahedron  32, 285–289. CrossRef Search ADS   Google Scholar Griffith JS. 1964. Theory of the isotropic g value of 4.27 found for some high‐spin ferric ions. Molecular Physiology  8, 213–217. CrossRef Search ADS   Google Scholar Islam SZ, Honda Y. 1998. Phototropism of conidial germ tubes of Botrytis cinerea and its implications in plant infection processes. Plant Disease  82, 850–856. CrossRef Search ADS   Google Scholar Islam SZ, Honda Y, Arase S. 1998. Light‐induced resistance of broad bean against Botrytis cinerea. Journal of Phytopathology  146, 479–485. CrossRef Search ADS   Google Scholar Janero DR. 1990. Malondialdehyde and thiobarbituric acid reactivity as diagnostic indices of lipid peroxidation and peroxidative tissue injury. Free Radical Biology and Medicine  9, 515–540. CrossRef Search ADS PubMed  Google Scholar Kliebenstein DJ, Monde RA, Last RL. 1998. Superoxide dismutase in Arabidopsis: an eclectic enzyme family with disparate regulation and protein localization. Plant Physiology  118, 637–650. CrossRef Search ADS PubMed  Google Scholar Kohl DH, Townsend J, Commoner B, Crespi HL, Dougherty RC, Katz JJ. 1965. Effects of isotopic substitution on electron spin resonance signals in photosynthetic organisms. Nature  206, 1105–1110. CrossRef Search ADS PubMed  Google Scholar Kramer GF, Norman HA, Krizek DT, Mirecki RM. 1991. Influence of UV‐B radiation on polyamines, lipid‐peroxidation and membrane‐lipids in cucumber. Phytochemistry  30, 2101–2108. CrossRef Search ADS   Google Scholar Kvaratskhelia M, Winkel C, Naldrett MT, Thorneley RNF. 1999. A novel high activity cationic ascorbate peroxidase from tea (Camellia sinensis)—a class III peroxidase with unusual substrate specificity. Journal of Plant Physiology  154, 273–282. CrossRef Search ADS   Google Scholar Laurent A, Alary J, Debrauwer L, Cravedi JP. 1999. Analysis in the rat of 4‐hydroxynonenal metabolites excreted in bile: evidence of enterohepatic circulation of these byproducts of lipid peroxidation. Chemical Research and Toxicology  12, 887–894. CrossRef Search ADS   Google Scholar Levine A, Tenhaken R, Dixon RA, Lamb C. 1994. H2O2 from the oxidative burst orchestrates the plant hypersensitivity disease resistance response. Cell  79, 583–593. CrossRef Search ADS PubMed  Google Scholar Lisowski J, Jezierski A, Bylinska E. 1993. EPR investigation of Mn(II), Fe(III) and free‐radical centers in green parts of living plants—effect of environmental pollution. Applied Magnetic Resonance  5, 15–23. CrossRef Search ADS   Google Scholar Liu J, Yeo HC, Doniger SJ, Ames BN. 1997. Assay of aldehydes from lipid peroxidation: gas chromatography–mass spectrometry compared to thiobarbituric acid. Analytical Biochemistry  245, 161–166. CrossRef Search ADS PubMed  Google Scholar Loewus FA. 1999. Biosynthesis and metabolism of ascorbic acid in plants and of analogs of ascorbic acid in fungi. Phytochemistry  52, 193–210. CrossRef Search ADS   Google Scholar Loprasert S, Vattanaviboon P, Praituan W, Chamnongpol S, Mongkolsuk S. 1996. Regulation of the oxidative stress protective enzymes, catalase and superoxide dismutase in Xanthomonas—a review. Gene  179, 33–37. CrossRef Search ADS PubMed  Google Scholar Muckenschnabel I, Williamson B, Goodman BA, Lyon GD, Stewart D, Deighton N. 2001. Markers for oxidative stress associated with soft rots in French beans (Phaseolus vulgaris) infected by Botrytis cinerea. Planta  212, 376–381. CrossRef Search ADS PubMed  Google Scholar Noctor G, Foyer CH. 1998. Ascorbate and glutathione: keeping active oxygen under control. Annual Reviews in Plant Physiology and Plant Molecular Biology  49, 249–279. CrossRef Search ADS   Google Scholar Park JSB, Wood PM, Davies MJ, Gilbert BC, Whitwood AC. 1997. A kinetic and ESR investigation of iron(II) oxalate oxidation by hydrogen peroxide and dioxygen as a source of hydroxyl radicals. Free Radical Research  27, 447–458. CrossRef Search ADS PubMed  Google Scholar Petersen DR, Reichard J, Kolaja KL, Hartley DP. 1999. 4‐Hydroxynonenal and malondialdehyde hepatic protein adducts in rats treated with carbon tetrachloride: immunochemical detection and lobular localization. Toxicology and Applied Pharmacology  161, 23–33. CrossRef Search ADS PubMed  Google Scholar Quidde T, Büttner P, Tudzynski P. 1999. Evidence for three different specific saponin‐detoxifying activities in Botrytis cinerea and cloning and functional analysis of a gene coding for a putative avenacinase. European Journal of Plant Pathology  105, 273–283. CrossRef Search ADS   Google Scholar Riemsdijk WHV, Venema P, Hiemstra T. 1996. Intrinsic proton affinity of reactive surface groups of metal (hydr)oxides: the bond valence principle. Journal of Colloidal Interface Science  184, 680–692. CrossRef Search ADS   Google Scholar Thiel S, Dohring T, Kofferlein M, Kosak A, Martin P, Seidlitz HK. 1996. A phytotron for plant stress research: how far can artificial lighting compare to natural sunlight? Journal of Plant Physiology  148, 456–463. CrossRef Search ADS   Google Scholar Thongbai P, Goodman BA. 2000. Free radical generation and post‐anoxic injury in rice grown in an iron‐toxic soil. Journal of Plant Nutrition  23, 1887–1900. CrossRef Search ADS   Google Scholar Williamson B, Duncan GH, Harrison JG, Harding LA, Elad Y, Zimand G. 1995. Effect of humidity on infection of rose petals by dry‐inoculated conidia of Botrytis cinerea. Mycological Research  99, 1303–1310. CrossRef Search ADS   Google Scholar Zimmermann JL, Boussac A, Rutherford AW. 1993. The manganese center of oxygen‐evolving and Ca2+‐depleted photosystem II. A pulsed EPR spectroscopy study. Biochemistry  32, 4831–4841. CrossRef Search ADS PubMed  Google Scholar © Society for Experimental Biology
TI - Infection of leaves of Arabidopsis thaliana by Botrytis cinerea: changes in ascorbic acid, free radicals and lipid peroxidation products
JF - Journal of Experimental Botany
DO - 10.1093/jexbot/53.367.207
DA - 2002-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/infection-of-leaves-of-arabidopsis-thaliana-by-botrytis-cinerea-gxLWRSPxSk
SP - 207
EP - 214
VL - 53
IS - 367
DP - DeepDyve
ER -