Euphytica (2018) 214:67 https://doi.org/10.1007/s10681-018-2150-2 Exploring the natural variation for reproductive thermotolerance in wild tomato species . . . . Nicky Driedonks Mieke Wolters-Arts Heidrun Huber Gert-Jan de Boer . . Wim Vriezen Celestina Mariani Ivo Rieu Received: 22 December 2017 / Accepted: 7 March 2018 / Published online: 13 March 2018 The Author(s) 2018 Abstract Climate change has become a serious germplasm screenings for thermotolerance have often threat for crop productivity worldwide. The increased used yield as the main measured trait. However, due to frequency of heat waves strongly affects reproductive the complex nature of yield and the relatively narrow success and thus yield for many crop species, implying genetic variation present in the cultivated germplasm that breeding for thermotolerant cultivars is critical for screened, there has been limited progress in under- food security. Insight into the genetic architecture of standing the genetic basis of reproductive heat toler- reproductive heat tolerance contributes to our funda- ance. Extending the screening to wild accessions of mental understanding of the stress sensitivity of this related species that cover a range of climatic condi- process and at the same time may have applied value. tions might be an effective approach to ﬁnd novel, In the case of tomato (Solanum lycopersicum), more tolerant genetic resources. The purpose of this study was to provide insight into the sensitivity of individual reproductive key traits (i.e. the number of Electronic supplementary material The online version of pollen per ﬂower, pollen viability and style protrusion) this article (https://doi.org/10.1007/s10681-018-2150-2) con- to heat-wave like long-term mild heat (LTMH), and tains supplementary material, which is available to authorized users. determine the extent to which genetic variation exists for these traits among wild tomato species. We found N. Driedonks M. Wolters-Arts C. Mariani that these traits were highly variable among the I. Rieu (&) screened accessions. Although no overall thermotol- Department of Molecular Plant Physiology, Institute for Water and Wetland Research, Radboud University, erant species were identiﬁed, several S. pimpinelli- 6525 AJ Nijmegen, The Netherlands folium individuals outperformed the best performing e-mail: email@example.com cultivar in terms of pollen viability under LTMH. Furthermore, we reveal that there has been local H. Huber Department of Experimental Plant Ecology, Institute for adaptation of reproductive heat tolerance, as acces- Water and Wetland Research, Radboud University, sions from lower elevations and higher annual 6525 AJ Nijmegen, The Netherlands temperature are more likely to show high pollen viability under LTMH. G.-J. de Boer Enza Zaden Research and Development B.V, 1600 AA Enkhuizen, The Netherlands Keywords Heat tolerance Pollen Male fertility Wild tomato W. Vriezen Bayer Vegetable Seeds, 6080 AA Haelen, The Netherlands 123 67 Page 2 of 12 Euphytica (2018) 214:67 Introduction of reproductive thermotolerance under long-term mild heat (LTMH). Ambient temperatures are rising as part of the current Secondly, germplasm used in thermotolerance global climate change, threatening agricultural output screening has mainly consisted of S. lycopersicum (IPCC 2007, 2013). High temperatures cause mor- cultivars (Abdul-Baki and Stommel 1995; Dane et al. phological, physiological, biochemical and molecular 1991; Grilli et al. 2007; Kugblenu et al. 2013a; Opena changes in plants that affect growth, and is particularly et al. 1992). However, as a result of domestication and detrimental during the reproductive stages (Wahid intensive breeding, the cultivated tomato germplasm et al. 2007). This leads to reduced yields in crop has a rather narrow genetic base (Bergougnoux 2014), species, and thus has a large impact on global food meaning that only a subset of the genes and alleles production (Barnabas et al. 2008; Hedhly et al. 2009). available in the wild progenitor gene pool are still For example, during cultivation, tomato is often present among crop cultivars (Godfray et al. 2010; exposed to high temperature either in the greenhouse Ladizinsky 1985; Olsen and Wendel 2013). Espe- or in the ﬁeld, and consequently, fruit set is reduced in cially, as breeding efforts have mainly targeted yield at many S. lycopersicum cultivars. more or less optimal cultivation conditions, it seems Exploration of natural variation may offer insight likely that abiotic stress tolerance traits have been lost into the genetics of stress tolerance, and can provide (Ladizinsky 1985; Paran and Van Der Knaap 2007). genetic diversity useful for breeding (Grandillo et al. This implies that the potential gain in heat tolerance 2011). This also applies to reproductive heat tolerance, level from cultivated germplasm is likely to be limited. but so far, screening for variation in tomato heat A broader genetic diversity can be found in species sensitivity has yielded only a few genotypes consid- related to tomato, and could serve as an alternative ered to be thermotolerant (Opena et al. 1992), and source of plant thermotolerance traits (Vı´quez- these have limited applicability for pre-breeding Zamora et al. 2013). Wild tomato species are found (Grilli et al. 2007). There seem to be at least two in a variety of habitats ranging from sea level to above major reasons for this. Firstly, in previous germplasm 3000 m in altitude and from temperate deserts to wet screenings fruit set was the main trait of interest. tropical rainforests, and thus face a range of environ- However, fruit set is a complex trait, i.e. it represents mental challenges. As a result of natural selection, the sum of multiple sub-traits (yield components). these wild species vary broadly in terms of morphol- Thus, there may be a relatively small chance that ogy, physiology, biochemistry and stress tolerance optimal sub-traits combine to generate a strongly levels (Dolferus 2014; Grandillo et al. 2011; Madu- outperforming genotype. Furthermore, the complexity raimuthu and Prasad 2014). There is also variation in of the relation among different traits involved in fruit mating systems among wild accessions, i.e. self- set complicates genetic analysis. As an alternative compatible (SC) versus self-incompatible (SI), which approach, it might be more effective to analyse the is likely to affect reproductive traits and putatively various contributing sub-traits individually and com- their performance under LTMH (Arroyo 1973; Baker bine them afterwards in a breeding context. For 1955; Cruden 1977; Georgiady and Lord 2002; Peralta example, decreases in tomato fruit set under long-term et al. 2008). mildly elevated temperatures has been shown to Here, we hypothesised that higher levels of repro- correlate with a decrease in pollen viability (Dane ductive thermotolerance are present in wild relatives et al. 1991; Firon et al. 2006; Kinet and Peet 1997; of tomato than in the cultivated tomato germplasm. Levy et al. 1978; Peet et al. 1998; Pressman 2002; We investigated the performance of 64 accessions Pressman et al. 2006; Sato et al. 2000, 2006; Xu et al. across 13 wild species and 7 S. lycopersicum cultivars, 2017b). Also, style protrusion may affect reproductive including a subset known for relatively good repro- success under high temperature (Charles and Harris ductive thermotolerance under control temperature 1972; Dane et al. 1991; Rick and Dempsey 1969; and long-term mild heat. We focused on reproductive Rudich et al. 1977; Saeed et al. 2007; Xu et al. 2017b). traits generally assumed to contribute to overall Investigation of these traits separately might provide a fertility, i.e. the number of pollen per ﬂower, pollen more effective strategy to determine the genetic basis viability and the distance between the top of the anther and the stigma (style protrusion). In addition, we 123 Euphytica (2018) 214:67 Page 3 of 12 67 tested whether the mating system inﬂuenced these from the vegetative to the generative phase occurred, traits under LTMH, and determined whether local ﬂower buds were removed and the cuttings were adaption to thermotolerance had occurred. treated similarly as the mother plants. Phenotypic assessment Materials and methods To determine pollen quality, anthers of the three most Plant material and screening procedure recently opened ﬂowers were cut into 4 equal trans- verse sections. After addition of 200 lL peroxidase Sixty-four accessions belonging to 13 wild species (S. indicator (Rodriguez-Riano and Dafni 2000) consist- arcanum, S. cheesmaniae, S. chilense, S. chmielewskii, ing of 1 vial peroxidase indicator (Sigma 3901-10VL) S. corneliomulleri, S. galapagense, S. habrochaites, S. in 0.012% (v/v) H O and 10% (v/v) Trizmal buffer 2 2 huaylasense, S. lycopersicum, S. neorickii, S. pennel- (903C; Sigma–Aldrich, St Louis, MO, USA). Pollen lii, S. peruvianum and S. pimpinellifolium) and 7 S. were considered viable when roundly shaped and lycopersicum cultivars (‘‘Hotset’’, ‘‘Malintka101’’, stained dark. In order to determine the pollen viability ‘‘Moneyberg’’, ‘‘Nagcarlang’’, ‘‘NCHS-1’’, ‘‘Salad- (PV, in %), 100 pollen were assessed per ﬂower. To ette’’ and ‘‘Tof Hamlet’’) were obtained from various determine the number of pollen per ﬂower (PN) the sources (Table S1). Seeds were incubated in 2.5% number of pollen was counted in 25 chambers hypochlorite for 30 min at room temperature to (0.04 mm ) of a haemocytometer. In addition, style improve germination and reduce pathogen load (Rick protrusion (SP in mm) was measured. For PN, PV and and Borgino, TGRC, http://tgrc.ucdavis.edu/seed_ SP, three ﬂowers were analysed per plant. germ.aspx), followed by germination on potting soil (Horticoop, Lentse Potgrond, Slingerland Potgrond) Climate data covered with vermiculite (Agra-Vermiculite) under standard greenhouse conditions. Seedlings were Climatic data sets for the earth land surface area were transferred to 0.5 L pots after 2 weeks and, after downloaded from CHELSA (Karger et al. 2017). 1 month, placed in 12 L pots, containing potting soil Using the R package ‘‘raster’’ version 2.5–8 (Hijmans -1 and 4 g L Osmocote Exact Standard 3–4 M and van Etten 2012), all 19 bioclimatic variable data (Everris). When the transition from the vegetative to (BIO1 to BIO19) were extracted for the period the generative phase occurred, ﬂower buds were 1979-2013 for each accession according to the GPS removed and the plants were transferred to a climate coordinates of the original collection site (Table S1). chamber maintaining a 14/10 h day/night photoperiod -1 -2 (* 300 lmol s m at plant height; Philips Statistical analysis D-Papillon daylight spectrum 340 W lamps and Phi- lips MastergreenPower TLD58 W/840 ﬂuorescent All statistical analyses were performed using trans- tubes) and humidity of 70–80% at either control formed data, value’ = Log(value?1), except for temperature of 25/19 C (CT) or long-term mild heat PV, to which a logit transformation was applied, of 32/26 C (LTMH) for at least 14 days. Plants were value’ = LN((value?1)/(101-value)). The relation grown and analysed in a staggered manner over a time between traits was determined by a Pearson correla- course of 4 months in batches of 15 individuals, with tion analysis. To assess the heritability of the traits, a complete randomisation of accessions. To determine Pearson correlation analysis of the means of clones the inﬂuence of the genotype on the studied traits, (cuttings) and their corresponding mother plant were cuttings were taken from several plants. In order to set performed using a paired sample correlation analysis. roots, the cuttings were put in potting soil (Horticoop, Broad-sense heritability was calculated for PN and PV Lentse Potgrond, Slingerland Potgrond) and kept in by dividing the variance among clones by the total the greenhouse in a plastic container to maintain a high variance among and within clones (i.e. variance humidity. After 2 weeks, cuttings were placed in 12 L among clones/total variance). To test for variation in -1 pots, containing potting soil and 4 g L Osmocote heat tolerance among species, a two-way ANOVA Exact Standard 3–4 M (Everris). When the transition was performed at species level (using mean values of 123 67 Page 4 of 12 Euphytica (2018) 214:67 accessions and temperature treatment). Differences in clones of PN and PV measurements of the cuttings performance under LTMH between species (using revealed a broad-sense heritability of 0.78 and 0.85, mean values of accessions) were assessed by a one- respectively (Table 1). way ANOVA followed by Tukey’s HSD as post hoc To evaluate relationships between the traits of test. At accession level (using mean values of plants), interest, Pearson correlation analyses between the trait differences between accessions within species were means per species were performed. In addition, to assessed by a one-way ANOVA using Tukey’s HSD correct for putative species-speciﬁc effects, a as post hoc test. For PN and PV, the wild accessions between-species Pearson correlation analysis was were compared to the best tomato cultivar by a one- performed. In both cases, no signiﬁcant correlations way ANOVA with LSD post hoc test. To test whether were detected (Table 2; data not shown). In addition, the best performing genotypes from the best wild when considering all accessions as independent accession outperformed Nagcarlang, a one-way (n = 71), no signiﬁcant correlations were detected ANOVA with LSD post hoc test was performed, between any of the traits under CT and LTMH (data using clones as replicates. To analyse the effect of not shown). Together, this suggests that the three traits temperature treatment and mating system, a two-way under study are largely independently inherited in ANOVA with temperature treatment and mating these species and accessions. system class variables was performed. The different Variation in the number of pollen per ﬂower geographical characteristics were correlated with the physiological plant traits under LTMH by Pearson under LTMH correlation analysis. All statistical analyses were performed using IBM SPSS Statistics version 21. Temperature treatment and species both had a signif- icant effect on the number of pollen per ﬂower (PN), and no interaction was found between them (Table 3). Results Exposure to LTMH reduced PN by 76.3% on average. PN of S. corneliomulleri was signiﬁcantly higher than To assess reproductive performance under long-term that of S. chmielewskii, which had the lowest PN under mild heat (LTMH) in the wild tomato germplasm, 64 LTMH (Table S2). For the S. chilense, S. chmielews- wild accessions belonging to 13 wild species and 7 S. kii, S. galapagense, S. huaylasense, S. pennellii, S. lycopersicum cultivars were screened. Accessions peruvianum and S. lycopersicum cultivars, signiﬁcant were selected to roughly encompass the spatial and differences in PN among accessions within the species elevation distribution of accessions of each species, were detected under LTMH (Table S3). However, and where possible, including accessions with anno- none of the wild accessions performed better, i.e. had a tations related to high temperature or other abiotic signiﬁcantly higher PN, than the best performing stresses (Table S1). None of the wild accessions cultivar, NCHS-1 (Fig. 2a; Table S3). screened in this study were previously determined to be heat tolerant. In total, 201 and 317 plants were Variation in pollen viability under LTMH exposed to control (CT) or LTMH conditions, respec- tively, and three reproductive traits, the number of In response to LTMH, PV was reduced by 85.6% on pollen per ﬂower (PN), pollen viability (PV) and style average, at the species level. No signiﬁcant differences were detected between any of the screened species, protrusion (SP), were analysed. nor was there a signiﬁcant interactive effect between Trait heritability and inter-trait relations species and treatment (Table 3). Within S. corne- liomulleri, S. neorickii, S. pimpinellifolium and the S. To determine the inﬂuence of genotype on the selected lycopersicum cultivars, signiﬁcant differences were traits, cuttings from several individuals were grown detected among accessions (Table S3). However, also and exposed to LTMH. Signiﬁcant correlations for this trait, none of the wild accession performed between mother plants and cuttings were detected better under LTMH, i.e. had a signiﬁcant higher PV, for all traits (Fig. 1). The division of the variance than the best performing cultivar, Nagcarlang among clones by the total variance among and within (Fig. 2b; Table S3). As wild accessions may exhibit 123 Euphytica (2018) 214:67 Page 5 of 12 67 Fig. 1 Correlations of traits between mother plant and cuttings LTMH for a The number of pollen per ﬂower (*1000), b pollen under long-term mild heat. Mother plants of 4 accessions (from viability (%), and c style protrusion (mm). Trait values for the S. corneliomulleri, S. peruvianum, S. pimpinellifolium and S. cuttings represent the average (n = 2–15). The Pearson’s pennellii) and two tomato cultivars (Moneyberg and Nagcar- correlation coefﬁcient is given in each graph (r). Signiﬁcance lang) were selected based on their difference in pollen viability level (two-tailed): *P \ 0.05; **P \ 0.01; ***P \ 0.001 under LTMH. Cuttings were generated and phenotyped under Table 1 Broad-sense heritability of the number of pollen per Table 3 P-values of a two-way ANOVA testing for the effects ﬂower and pollen viability of temperature treatment and species on key reproductive sub- traits PN PV Trait Treatment Species Treatment * Species Genotype 0.65 0.79 PN \ 0.001 \ 0.001 0.060 Cutting 0.18 0.14 PV \ 0.001 0.214 0.272 Error 0.17 0.06 SP \ 0.001 \ 0.001 0.165 Broad-sense heritability 0.78 0.85 PN pollen number, PV pollen viability, SP style protrusion. For The values given for genotype and cutting represent the details per species see Table S2 variance among and within cuttings, respectively. Error represents the unexplained variance PN number of pollen per ﬂowers, PV pollen viability Variation in style protrusion under LTMH SP was signiﬁcantly increased by LTMH, and differed Table 2 Pearson’s correlation coefﬁcients among traits of species under long-term mild heat signiﬁcantly among species, but no interactive effect between temperature treatment and species was Trait PN PV SP detected, indicating that the different species respond PN 1 similarly to LTMH with respect to SP (Table 3 and PV 0.08 1 Table S2). Within species, signiﬁcant differences SP 0.17 0.15 1 under LTMH were observed only in the cases of S. pimpinellifolium and the tomato cultivars (Table S3). PN pollen number per ﬂower, PV pollen viability, SP style protrusion. None of the correlations (two-tailed) were None of the wild accessions outperformed the culti- signiﬁcant vars under LTMH (i.e. had lower SP), as cultivar Saladette did not show any protrusion (Table S3). Many wild accessions had higher SP then the cultivars, genotypic diversity, we tested whether the best already under control temperature. performing genotypes from the best wild accession (S. pimpinellifolium LA1630) outperformed Nagcar- lang, using multiple clones per genotype. Indeed, four of the ﬁve tested genotypes had signiﬁcantly higher PV under LMTH than Nagcarlang (Fig. 3). 123 67 Page 6 of 12 Euphytica (2018) 214:67 Fig. 2 Cultivars and the three best performing wild accessions interquartile range (IQR), with indication of the median. Lower with respect to pollen number per ﬂower and pollen viability and upper whiskers represent the smallest and largest observa- under long-term mild heat. a Pollen number per ﬂower (PN), and tions smaller than or equal to lower and upper hinge ± 1.5 * b pollen viability (PV). Box of boxplot represents the IQR, respectively. Each dot represents an individual plant Comparison between self-compatible and self- incompatible accessions To test the effect of an accession’s mating system on reproductive traits, self-compatible (SC) and self- incompatible (SI) accessions were compared. SI accessions had signiﬁcantly higher PN and SP than SC accessions under both temperature treatments (Fig. 4). None of the traits showed signiﬁcant inter- action between the two factors. Relationship between trait performance and climatic parameters at site of origin Due to the wide variation in geographical origin of the accessions, ranging from a latitude of - 24.211 to 0.867, longitude of - 91.417 to - 43.083, and elevation from 0 up to 3450 meters above sea level (Fig. 5), accessions habitats were also diverse and Fig. 3 Comparison between Nagcarlang and best performing varied from very dry to wet locations and from sandy genotypes of S. pimpinellifolium LA1630 concerning pollen viability (PV) under long-term mild heat coastal areas to high up in the mountains. To assess whether adaptation to local conditions had occurred, a Pearson correlation analysis between the phenotypic data and various geographical and climatic parameters 123 Euphytica (2018) 214:67 Page 7 of 12 67 Fig. 4 Trait values in self-compatible and self-incompatible ***P \ 0.001, n.s., not signiﬁcant. a Number of pollen number accession under control and long-term mild heat. To compare per ﬂower (PN), b pollen viability (PV), and c style protrusion genotypes, multiple cuttings per genotype were evaluated. (SP). SC self-compatible, SI self-incompatible, CT control Values represent the mean ± standard deviation (n = 5–20 temperature, LTMH long-term mild heat. Values represent the plants). Differences were assessed by a one-way ANOVA mean ± standard deviation (n = 33 and 37 for SI and SC followed by LSD post hoc test. Asterisks above the bars indicate accessions, respectively). Signiﬁcance of treatment/mating a signiﬁcant difference between the respective genotype and the system/interaction was determined by a two-way ANOVA: cultivar Nagcarlang. Signiﬁcance level: *P \ 0.05; ***P \ 0.001; n.s., not signiﬁcant Fig. 5 Geographical origin of accessions screened. Elevation is in meters above sea level was performed. A signiﬁcant negative correlation was temperature parameters (Table S4). Thus, accessions found between the accessions’ PV and the elevation at derived from lower elevations and warmer climates the site of origin (Table 4). A signiﬁcant positive are more likely to be tolerant to LTMH with respect to correlation was found with the annual mean temper- PV. ature, and the same trend was visible for related 123 67 Page 8 of 12 Euphytica (2018) 214:67 Table 4 Pearson’s correlation coefﬁcients among geographi- Villareal et al. 1978). However, it is a complex trait, cal and physiological traits under long-term mild heat and higher heritability may be found by separating the underlying individual key traits affecting fruit set. Trait EL TEMP PREC Indeed, we have recently reported two major QTLs for PN 0.048 - 0.080 - 0.102 PN and PV in an S. lycopersicum mapping population ** * PV - 0.372 0.279 - 0.102 (Xu et al. 2017a). In the current study, screening of SP 0.103 - 0.106 - 0.281 clones of individuals for PN and PV under LTMH in climate chambers indicated that a large fraction of the Signiﬁcance level (two-tailed): *P \ 0.05; **P \ 0.01. Additional correlations with bioclimatic variables are total phenotypic variance was explained by the genetic presented in Table S4 variance. Whether these sub-traits also express in PN number of pollen per ﬂower, PV pollen viability, SP style other genetic backgrounds and environmental condi- protrusion; EL elevation, TEMP mean annual temperature in tions, such as ﬁeld conditions, remains to be period 1979–2013, PREC mean annual precipitation in period determined. 1979–2013. n = 59–69 accessions Our study did not detect overall thermotolerance of yield-contributing sub-traits in wild species compared to the performance of cultivars, but we show that Discussion several genotypes from the accession LA1630 outper- form the best performing cultivar in terms of PV under The reduction in tomato yield under long-term mild LTMH. Given the previously reported correlation heat (LTMH) may be attributed to the plant’s vulner- between PV and fruit set under LTMH (e.g. Dane et al. ability during reproductive development, resulting in a 1991; Sato et al. 2000; Xu et al. 2017b), we conclude lower number of pollen per ﬂower (PN) and pollen that wild germplasm might indeed be a valuable viability (PV) (Dane et al. 1991; Firon et al. 2006; resource to enrich domesticated germplasm for repro- Kinet and Peet 1997; Levy et al. 1978; Peet et al. 1998; ductive thermotolerance. Pressman 2002; Pressman et al. 2006; Sato et al. 2000, 2006; Xu et al. 2017b). In this study, we Mating system advantages under LTMH analysed the natural variation of reproductive thermo- tolerance in wild tomato species, which may serve as In the tomato clade, the mating system ranges between gene sources for cultivated tomato. self-incompatible (SI) to self-compatible (SC) cross- ers (Miller and Tanksley 1990; Rick et al. 1977). In Superior heat tolerant wild genotypes with regard general, ﬂowers of SI plant species produce more to pollen viability pollen than closely related SC species, probably because a much smaller fraction of the pollen will Yield screenings of cultivated S. lycopersicum under reach a compatible stigma in the case of SI (Arroyo high temperatures have shown phenotypic variation, 1973; Baker 1955; Cruden 1977; Georgiady and Lord but only a few cultivars, including Nagcarlang, Hotset 2002). Indeed, this study indicated that PN was and Saladette seem to perform relatively well under signiﬁcantly higher in SI compared to SC accessions. such conditions (Abdul-Baki and Stommel 1995; Importantly, no interaction with temperature treatment Dane et al. 1991; Kugblenu et al. 2013b; Levy et al. was found. Thus, SI accessions seem to be a good 1978; Rudich et al. 1977; Villareal et al. 1978; Xu et al. source for a high PN under LTMH and could be 2017b). Indeed, the Asian Vegetable Research and interesting for thermotolerance breeding purposes. In Development Center (AVRDC; now World contrast to PN, PV was not signiﬁcantly different Vegetable Center) concluded from screenings of [ between the mating types in either temperature 4000 wild and cultivated accessions under hot treatment. conditions that less than 1% could be considered Several studies indicated that protrusion of the style highly heat tolerant for fruit set (Opena et al. 1992; from the antheridial cone of [ 1 mm prevents fruit set Villareal et al. 1978). Fruit set under high temperature from self-fertilisation (Dane et al. 1991; Rudich et al. has been reported to have low narrow-sense heritabil- 1977; Saeed et al. 2007). As reported previously ity (El Ahmadi and Stevens 1979; Hanson et al. 2002; (Grandillo et al. 2011; Peralta et al. 2008), SP was 123 Euphytica (2018) 214:67 Page 9 of 12 67 reduced in SC accessions and almost absent in some temperature proﬁle (Li et al. 2015). Such local cultivars, probably due to strong trait selection. SP was adaptation may also be seen at the molecular level, enhanced under high temperature, and although SC as the heat stress response of Arabidopsis and accessions still showed less protrusion of the style in Chenopodium album accessions, as measured by LTMH, in many accessions the distance between induction of heat shock proteins, was more strongly anther and style was likely too large to allow self- induced in accessions originating from cooler rather pollination. The tomato cultivars performed relatively than warmer environments (Barua et al. 2008; Zhang well for SP, suggesting wild relatives are less useful et al. 2015). for improving this trait. By enhancing SP under high temperature, SC plants seem to mimic the constitutive SI phenotype. Stimu- Conclusion lating cross-pollination under LTMH via increased SP, and lower PN and PV, might increase the chance We conclude that PN and PV are variable among wild that SC individuals are fertilised by another plant. This and cultivated tomato accessions, and that this vari- ﬁts with the idea that genetic recombination may be ation is adaptive to the local environment in the case of beneﬁcial under stress conditions, allowing the cre- PV. The absence of overall thermotolerant accessions ation of more adapted genotypes (Hedhly et al. 2004; with regards to PV suggests that selective pressure is ¨ not very strong, or that there is a trade-off with an Muller and Rieu 2016). unknown, beneﬁcial trait. Although the best perform- Local adaptation ing wild accessions were equally thermotolerant to the best performing cultivars in terms of PN and PV, the Wild tomatoes occur over a wide range of ecological genetic background of these traits in the wild acces- and climatic conditions, but individual species and sions may be novel and could thus be valuable for accessions are often adapted to particular microcli- thermotolerance breeding of tomato, especially if the mates (Bauchet and Causse 2012; Zuriaga et al. 2009). traits show additivity. In the case of PV, several The diversity of conditions is expressed at the outperforming individuals were identiﬁed. Interspeci- morphological, physiological, sexual and molecular ﬁc QTL analysis with S. lycopersicum would be a levels (Bauchet and Causse 2012; Peralta and Spooner logical step towards characterisation and application 2005). We explored whether variation in environment of the traits. Phenotypic improvement from QTLs at the sites of origin of accessions has resulted in depend on unpredictable interactions with the genetic variation in thermotolerance and found that the mean background, probably because variation often PV of accessions correlated negatively with elevation. involves additional, undetected small-effect loci In line with our results, chilling tolerance in tomato (Mackay et al. 2009). Moreover, for successful has also been shown to correlate with elevation: for application, it will be important to consider the geographical populations of L. hirsutum (S. habro- environmental context dependency of the expression chaites), chilling tolerance, including traits such as of QTLs (Collins et al. 2008). In the end, reproductive seedling survival rate and pollen tube growth, was success depends on multiple traits and we hypothesize greatest in those derived from the higher elevations that combining optimal variants of all these traits will (Patterson et al. 1978; Zamir et al. 1981). It seems be needed to signiﬁcantly improve tolerance of likely that the effects of elevation are mainly due to reproduction to LTMH. Traits may be transferred local temperature proﬁles. Indeed, we found that and stacked through marker assisted breeding or by temperature variables such as mean annual tempera- application of newly developed genetic modiﬁcation ture correlated positively to PV under LTMH. Simi- methods (Sander and Joung 2014; Woo et al. 2015). larly, seedling survival and root growth at high Acknowledgement This work was supported by the temperature for natural populations of Arabidopsis Technological Top Institute Green Genetics (grant thaliana correlated to temperature parameters at the number 4CFD047RP, to IR) and the Dutch Topsector site of origin (Zhang et al. 2015). In rice, the presence Horticulture and Starting Materials (grant number 2013-H320, of a major quantitative trait locus (QTL) for thermo- to IR). We are thankful to Jacob Monash for his kind help in proofreading the manuscript. tolerance, TT1, has also been linked to the local 123 67 Page 10 of 12 Euphytica (2018) 214:67 Open Access This article is distributed under the terms of the Sci Hortic 109:212–217. https://doi.org/10.1016/j.scienta. Creative Commons Attribution 4.0 International License (http:// 2006.03.007 creativecommons.org/licenses/by/4.0/), which permits unre- Georgiady MS, Lord EM (2002) Evolution of the Inbred Flower stricted use, distribution, and reproduction in any medium, Form in the Currant Tomato Lycopersicon pimpinelli- provided you give appropriate credit to the original folium. Int J Plant Sci 163:531–541. https://doi.org/10. author(s) and the source, provide a link to the Creative Com- 1086/340542 mons license, and indicate if changes were made. Godfray C, Beddington J, Crute I, Haddad L, Lawrence D, Muir JF et al (2010) Food security: the challenge of feeding 9 billion people. Science 327(5967):812–818 Grandillo S, Chetelat R, Knapp S, Spooner D, Peralta I, Cam- mareri M et al. (2011) ‘‘Solanum sect. Lycopersicon’’. In: References C Kole (ed) Wild crop relatives: genomic and breeding resources vegetables, Springer, Heidelberg, pp 129–216 https://doi.org/10.1007/978-3-642-20450-09 Abdul-Baki AA, Stommel JR (1995) Pollen viability and fruit Grilli GVG, Braz LT, Gertrudes E, Lemos M (2007) QTL set of tomato genotypes under optimum- and high-tem- identiﬁcation for tolerance to fruit set in tomato by fAFLP perature regimes. HortScience 30:115–117 markers. Crop Breed Appl Biotechnol 7:234–241 Arroyo MTK (1973) Chiasma frequency evidence on the evo- Hanson PM, Chen J, Kuo G (2002) Gene action and heritability lution of autogamy in Limnanthes ﬂoccosa. Evolution of high-temperature fruit set in tomato line CL5915. 27(4):679–688 HortScience 37:172–175 Baker HG (1955) Self-compatibility and establishment after Hedhly A, Hormaza JI, Herrero M (2004) Effect of temperature ‘‘long-distance’’ dispersal. Evolution 9(3):347–349 ´ ¨ ´ on pollen tube kinetics and dynamics in sweet cherry, BarnabasB,Jager K, Feher A (2008) The effect of drought and heat stress on reproductive processes in cereals. Plant, Cell Prunus avium (Rosaceae). Am J Bot 91:558–564. https:// Environ 31:11–38. https://doi.org/10.1111/j.1365-3040. doi.org/10.3732/ajb.91.4.558 2007.01727.x Hedhly A, Hormaza JI, Herrero M (2009) Global warming and Barua D, Heckathorn SA, Coleman JS (2008) Variation in heat- sexual plant reproduction. Trends Plant Sci 14:30–36. shock proteins and photosynthetic thermotolerance among https://doi.org/10.1016/j.tplants.2008.11.001 Hijmans RJ, van Etten J (2012) Raster: geographic analysis and natural populations of Chenopodium album L. from con- modeling with raster data. http://cran.r-project.org/ trasting thermal environments: implications for plant package=raster responses to global warming. J Integr Plant Biol IPCC (2007) Climate change 2007: the physical science basis. 50:1440–1451. https://doi.org/10.1111/j.1744-7909.2008. Contribution of Working Group I to the Fourth Assessment 00756.x Report of the Intergovernmental Panel on Climate Change. Bauchet G, Causse M (2012) ‘‘Genetic Diversity in Tomato Solomon S, Qin D, Manning M, Alley RB, Berntsen T, (Solanum lycopersicum) and its Wild Relatives,’’. In: M Bindoff NL, Chen ZC. Cambridge University Press, C ¸ alis¸kan (ed) Genetic diversity in plants, pp 133–162 Cambridge. http://www.ipcc.ch/publications_and_data/ Bergougnoux V (2014) The history of tomato: from domesti- publications_ipcc_fourth_assessment_report_wg1_report_ cation to biopharming. Biotechnol Adv 32:170–189. the_physical_science_basis.htm https://doi.org/10.1016/j.biotechadv.2013.11.003 IPCC (2013) Summary for policymakers climate change 2013: Charles WB, Harris RE (1972) Tomato fruit-set at high and low the physical science basis contribution of working group I temperatures. Can J Plant Sci 111:461–476 to the ﬁfth assessment report of the intergovernmental Collins NC, Tardieu F, Tuberosa R (2008) Quantitative trait loci panel on climate change, Stocker TF, Qin D, Plattner GK, and crop performance under abiotic stress: where do we Tignor M, Allen SK, Boschung J, Naue A. Cambridge and stand? Plant Physiol 147:469–486. https://doi.org/10.1104/ New York. https://www.ipcc.ch/report/ar5/ pp.108.118117 Karger DN, Conrad O, Bo¨hner J, Kawohl T, Kreft H, Soria- Cruden RW (1977) Pollen-ovule ratios: a conservative indicator Auza RW et al (2017) Climatologies at high resolution for of breeding systems in ﬂowering plants. Evolution the earth’s land surface areas. Sci. Data 4:170122. https:// 31(1):32–46 doi.org/10.1038/sdata.2017.122 Dane F, Hunter AG, Chambliss OL (1991) Fruit set, pollen fertility, and combining ability of selected tomato geno- Kinet JM, Peet MM (1997) Tomato. The physiology of veg- types under high temperature ﬁeld conditions. J Am Soc etable crops (Commonwealth Agricultural Bureau (CAB) Hortic Sci 116:906–910 international. Wallingford, UK), pp 207–258 Dolferus R (2014) To grow or not to grow: a stressful decision Kugblenu YO, Oppong Danso E, Ofori K, Andersen MN, for plants. Plant Sci 229:247–261. https://doi.org/10.1016/ Abenney-Mickson S, Sabi E et al (2013a) Screening tomato j.plantsci.2014.10.002 genotypes in Ghana for adaptation to high temperature. Acta Agric Scand Sect B—Soil Plant Sci 63:516–522. https://doi. El Ahmadi A, Stevens M (1979) Reproductive responses of org/10.1080/09064710.2013.813062 heat-tolerant tomatoes to high temperatures. J Am Soc Kugblenu YO, Oppong Danso E, Ofori K, Andersen MN, Hortic Sci 104:686–691 Abenney-Mickson S, Sabi EB et al (2013b) Screening Firon N, Shaked R, Peet MM, Pharr DM, Zamski E (2006) tomato genotypes for adaptation to high temperature in Pollen grains of heat tolerant tomato cultivars retain higher West Africa. Acta Agric Scand Sect B—Soil Plant Sci carbohydrate concentration under heat stress conditions. 123 Euphytica (2018) 214:67 Page 11 of 12 67 63:516–522. https://doi.org/10.1080/09064710.2013. Pressman E, Harel D, Zamski E, Shaked R, Althan L, Rosenfeld 813062 K et al (2006) The effect of high temperatures on the Ladizinsky G (1985) Founder effect in crop-plant evolution. expression and activity of sucrose-cleaving enzymes dur- Ecomonic Bot 39:191–199 ing tomato (Lycopersicon esculentum) anther develop- Levy A, Rabinowitch HD, Kedar N (1978) Morphological and ment. J Hortic Sci Biotechnol 81:341–348. https://doi.org/ physiological characters affecting ﬂower drop and fruit set 10.1080/14620316.2006.11512071 of tomatoes at high temperatures. Euphytica 27:211–218. Rick CM, Dempsey WH (1969) Position of the stigma in rela- https://doi.org/10.1007/BF00039137 tion to fruit setting of the tomato. Bot Gaz 130:180–186. Li X-M, Chao D-Y, Wu Y, Huang X, Chen K, Cui L-G et al http://www.jstor.org/stable/2474143. (2015) Natural alleles of a proteasome a2 subunit gene Rick C, Fobes J, Holle M (1977) Genetic variation in Lycop- contribute to thermotolerance and adaptation of African ersicon pimpinellifolium: evidence of evolutionary change rice. Nat Genet 47:827–833. https://doi.org/10.1038/ng. in mating systems. Plant Syst Evol 127:139–170. papers:// 3305 cc1b5c7f-a7db-41a0-b721-8bdfaf6a062e/Paper/p57 Mackay TFC, Stone EA, Ayroles JF (2009) The genetics of Rodriguez-Riano T, Dafni A (2000) A new procedure to asses quantitative traits: challenges and prospects. Nat Rev pollen viability. Sex Plant Reprod 12:241–244. https://doi. Genet 10:565–577. https://doi.org/10.1038/nrg2612 org/10.1007/s004970050008 Maduraimuthu D, Prasad PVV (2014) ‘‘High Temperature Rudich J, Zamski E, Regev Y (1977) Genotypic variation for Stress,’’. In: Jackson M, Ford-Lloyd BV, Perry ML (eds) sensitivity to high temperature in the tomato: pollination Plant genetic resources and climate change, (CABI), and fruit set. Botanical Gazette 138:448–452 pp 201–220 Saeed A, Hayat K, Khan AA, Iqbal S (2007) Heat tolerance Miller JC, Tanksley SD (1990) RFLP analysis of phylogenetic studies in tomato (Lycopersicon esculentum Mill.). Int J relationships and genetic variation in the genus Lycoper- Agric Biol 9:649–652 http://www.fspublishers.org sicon. Theor Appl Genet 80:437–448 Sander JD, Joung JK (2014) CRISPR-Cas systems for editing, Mu¨ller F, Rieu I (2016) Acclimation to high temperature during regulating and targeting genomes. Nat Biotechnol pollen development. Plant Reprod 29:107–118. https://doi. 32:347–355. https://doi.org/10.1038/nbt.2842 org/10.1007/s00497-016-0282-x Sato S, Peet MM, Thomas JF (2000) Physiological factors limit Olsen KM, Wendel JF (2013) A bountiful harvest: genomic fruit set of tomato (Lycopersicon esculentum Mill.) under insights into crop domestication phenotypes. Annu Rev chronic, mild heat stress. Plant, Cell Environ 23:719–726. Plant Biol 64:47–70. https://doi.org/10.1146/annurev- https://doi.org/10.1046/j.1365-3040.2000.00589.x arplant-050312-120048 Sato S, Kamiyama M, Iwata T, Makita N, Furukawa H, Ikeda H Opena RT, Chen JT, Kuo GT, Chen HM (1992) Genetic and (2006) Moderate increase of mean daily temperature Physiological Aspects of Tropical Adaptation in Tomato. adversely affects fruit set of Lycopersicon esculentum by Adaptation of food crops to temperature and water stress. disrupting speciﬁc physiological processes in male repro- Asian Vegetable Research and Development Center, ductive development. Ann Bot 97:731–738. https://doi. Shanua, pp 257–270 org/10.1093/aob/mcl037 Paran I, Van Der Knaap E (2007) Genetic and molecular regu- Villareal RL, Lai SH, Wong SH (1978) Screening for heat tol- lation of fruit and plant domestication traits in tomato and erance in the genus Lycopersicon. HortScience pepper. J Exp Bot 58:3841–3852. https://doi.org/10.1093/ 13:479–481 jxb/erm257 Vı´quez-Zamora M, Vosman B, van de Geest H, Bovy A, Visser Patterson B, Paull R, Smillie R (1978) Chilling Resistance in RGF, Finkers R et al (2013) Tomato breeding in the Lycopersicon hirsutum humb. & bonpl., a wild tomato with genomics era: insights from a SNP array. BMC Genomics a wild altitudinal distribution. Aust J Plant Physiol 14:354. https://doi.org/10.1186/1471-2164-14-354 5:609–617 Wahid A, Gelani S, Ashraf M, Foolad M (2007) Heat tolerance Peet MM, Sato S, Gardner RG (1998) Comparing heat stress in plants: an overview. Environ Exp Bot 61:199–223. effects on male-fertile and male-sterile tomatoes. Plant, https://doi.org/10.1016/j.envexpbot.2007.05.011 Cell Environ 21:225–231 Woo JW, Kim J, Kwon S Il, Corvalan C, Cho SW, Kim H et al Peralta IE, Spooner DM (2005) Morphological characterization (2015) DNA-free genome editing in plants with pre- and relationships of wild tomatoes (Solanum L. sect. Ly- assembled CRISPR-Cas9 ribonucleoproteins. Nat copersicon). A Festschrift William G. D’Arcy Biotechnol 33:1162–1164. https://doi.org/10.1038/nbt. 104:227–257 3389 Peralta IE, Spooner DM, Knapp S (2008) ‘‘Taxonomy of wild Xu J, Driedonks N, Rutten MJM, Vriezen WH, de Boer G-J, tomatoes and their relatives (Solanum sect. Lycopersi- Rieu I (2017a) Mapping quantitative trait loci for heat coides, sect. Juglandifolia, sect. Lycopersicon; Solana- tolerance of reproductive traits in tomato (Solanum ceae),’’. In: Anderson C (ed) Systematic botany lycopersicum). Mol Breed 37:37–58. https://doi.org/10. monographs, (The American Society of Plant Tax- 1007/s11032-017-0664-2 onomists), 1–186 https://doi.org/10.1126/science.113. Xu J, Wolters-Arts M, Mariani C, Heidrun H, Rieu I (2017b) 2935.3. Heat stress affects vegetative and reproductive perfor- Pressman E (2002) The effect of heat stress on tomato pollen mance and trait correlations in tomato (Solanum lycoper- characteristics is associated with changes in carbohydrate sicum). Euphytica 213:156. https://doi.org/10.1007/ concentration in the developing anthers. Ann Bot s10681-017-1949-6 90:631–636. https://doi.org/10.1093/aob/mcf240 123 67 Page 12 of 12 Euphytica (2018) 214:67 Zamir D, Tanksley SD, Jones RA (1981) Low temperature responses to high-temperature stress. AoB Plants. https:// effect on selective fertilization by pollen mixtures of wild doi.org/10.1093/aobpla/plv101 and cultivated tomato species. Theor Appl Genet Zuriaga E, Blanca JM, Cordero L, Sifres A, Blas-Cerda´n WG, 59:235–238. https://doi.org/10.1007/BF00265501 Morales R et al (2009) Genetic and bioclimatic variation in Zhang N, Belsterling B, Raszewski J, Tonsor SJ (2015) Natural Solanum pimpinellifolium. Genet Resour Crop Evol populations of Arabidopsis thaliana differ in seedling 56:39–51. https://doi.org/10.1007/s10722-008-9340-z
Euphytica – Springer Journals
Published: Mar 13, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera