Tomato (Solanum lycopersicum) is one of the most important vegetable crops worldwide. Similarly to other plant species, tomato is very sensitive to high ambient temperatures, a globally increasing phenomenon due to climatic changes. For tomato plants, temperatures exceeding 32°C during the day, and 21°C during the night create conditions of heat stress, leading to various physiological and developmental effects that drive dramatic reduction in fruit set and total yield, thus necessitating the identification of novel resources for tomato breeding (Hazra et al., 2007; Liu et al., 2023). The negative effect of heat stress on yield is the consequence of multiple molecular, physiological and developmental processes that are hindered by elevated temperatures, underlining tomato thermotolerance a highly complex trait.

The reproductive phase of development is more thermosensitive compared with the vegetative phase. Starting at early stages of flowering, the number of flowers produced under high temperature conditions may be reduced (Charles and Harris, 1972; Abdul-Baki, 1991; Xu et al., 2017). Additionally, flower abnormalities are very abundant under heat stress conditions, therefore these traits are indicative of heat stress sensing and damage. High temperatures may also lead to protrusion of the style over the anther cone thus hindering self-pollination (Rick and Dempsey, 1969; Lohar and Peat, 1998). Excessive heat stress may also result in anther deformations, as the cone-shaped anthers burst open, hindering pollination as well (Bita and Gerats, 2013; Müller et al., 2016; Ayenan et al., 2019). The most dramatic effect of heat stress on tomato flowers is bud abscission (Sato et al., 2001). Nevertheless, the most sensitive tissue to elevated temperatures is the male gametophyte in the period of 8–13 days prior to anthesis (Sato et al., 2002). High temperatures lead to reduced pollen viability and germination, vital functions for successful fertilization (Charles and Harris, 1972; Peet et al., 1998; Pressman et al., 2002). Consequently, fruit set is severely damaged. For example, under mean daily temperature of 29°C compared with control conditions at 25°C, tomato fruit set was reduced by 77% (Peet et al., 1997). Later studies presented similar results under different heat stress regimes (Sato et al., 2000; Firon et al., 2006; Sato et al., 2006). Due to the reduced ability of the pollen grains to initiate fertilization under heat stress conditions, the number of seeds per fruit is significantly reduced, occasionally accompanied by a reduction in fruit size (Sato et al., 2001).

For most crop species, the largest reservoir of genetic variation exists not within modern cultivars, but in their undomesticated wild relatives. In tomato, it was estimated that less than 5% of the available genetic variation exists in cultivars and the other 95% is found in wild relatives (Miller and Tanksley, 1990). For this reason, interrogating wild relatives for abiotic stress resistance traits is of great value. To-date, 17 species of wild tomato relatives are recognized, originating from various environments, from high mountains to dry coastal regions, and they exhibit great differences in morphological characters, disease susceptibility, and stress resistance traits (Kimura and Sinha, 2008; Peralta et al., 2008; Conesa et al., 2017). The use of wild relatives may be sometimes hindered due to cross incompatibility, F1 hybrid sterility, infertility of the segregating generations, reduced recombination between the chromosomes of the two species, or linkage-drag (genes of negative effect being tightly linked to the trait of interest) (Zamir, 2001). Despite these obstacles, backcross breeding was successfully used in tomato to improve quantitative traits by the introgression of genes from wild germplasm into elite cultivars while maintaining the favorable horticultural characteristics of the elite materials (Hartman and St Clair, 1999; Doganlar et al., 2002; Kabelka et al., 2004).

The wild species Solanum pimpinellifolium, considered as the ancestor of the cultivated tomato, is originated in the coastal deserts of Peru and Ecuador (Peralta and Spooner, 2000; Pease et al., 2016). The S. lycopersicum and S. pimpinellifolium species are both self-compatible, red-fruited and they are known to hybridize naturally (Rick, 1958). Despite their close relationship, the two species differ in many morphological aspects, especially in fruit size and growth habits and several other economically important traits, many of which are polygenic. For this reason, S. pimpinellifolium is considered an attractive source for tomato breeding. Several studies have already utilized S. pimpinellifolium for the identification of important quantitative trait loci (QTLs). Among these are QTLs for increased yield, soluble solids content, and improved fruit color (Grandillo and Tanksley, 1996). S. pimpinellifolium was shown to be a suitable source for improving nitrogen use efficiency. A collection of 29 Introgression Lines (ILs) resulting from a cross between the To-937 accession of S. pimpinellifolium and the S. lycopersicum cv Moneymaker (MM) was used to identify specific regions in the S. pimpinellifolium genome involved in the responses to N inputs of fruit production and fruit quality. Interestingly, the identified region contains genes involved in C and N metabolism (Renau-Morata et al., 2024).

In terms of abiotic stress tolerance, it was reported that seeds of S. pimpinellifolium exhibit better germination under drought stress compared with cultivated tomato (Lin et al., 2002; Villalta et al., 2008; Rao et al., 2013), while five QTLs were found to be linked with salt tolerance during vegetative growth (Foolad et al., 2001). Additionally, several S. pimpinellifolium accessions were characterized as having high salinity tolerance (Lin et al., 2002; Villalta et al., 2008; Rao et al., 2013). In regard to heat stress, pollen number, pollen viability and style protrusion, representing key reproductive heat stress-related traits, were evaluated in various wild species, among them eight accessions of S. pimpinellifolium (LA1670, LA1630, LA1645, LA1629, LA0114, LA1579, LA1237 and LA1547). Although no overall thermotolerant species was identified, several S. pimpinellifolium individuals of the LA1630 accession outperformed the best performing cultivars in terms of pollen viability under heat stress conditions (Driedonks et al., 2018). In another study, 22 QTLs involved in reproductive traits at different temperatures were identified, using an S. pimpinellifolium recombinant inbred lines (RIL) population. All 168 lines were subjected to increased temperature conditions and individual lines presented improved performance under these conditions (Gonzalo et al., 2020), demonstrating the high value and potential of S. pimpinellifolium as a genetic source for heat stress tolerance in tomato.

Climatic models in tomato-growing locations around the globe predict that temperatures will continue to rise and the severity and frequency of above-optimal temperature episodes will increase (Bell et al., 2000). These conditions strongly and negatively affect reproductive success and thus yield, implying that breeding for thermotolerant cultivars is critical for food security. The sub-optimal growth conditions in which S. pimpinellifolium was evolved in, its proven resistance to various abiotic stresses and its genetic relatedness to the cultivated species S. lycopersicum, place it as a good source for the identification of heat stress tolerance traits. Here we present a detailed phenotypic characterization of heat stress related traits of a new population based on S. pimpinellifolium. Using this new BILs (Backcross Introgression Lines) population originated from a cross between S. pimpinellifolium and S. lycopersicum, we found high variation in heat stress related traits, and identified specific lines with better performance under heat stress conditions, making this BILs population an additional and highly relevant source for thermotolerance QTL identification and breeding of thermotolerant tomato cultivars.

The increased ambient temperatures in many agricultural regions around the world have detrimental influence on crop yield, therefore the development of heat stress resistant cultivars is of upmost importance. Utilizing the broad genetic variation that exists in wild species is a prevalent approach for the identification of genes and alleles that improve crops’ performance under harsh environmental conditions, including heat stress (Zamir, 2001). In order to evaluate a new BILs population generated from a cross between M82 (S. lycopersicum) and the S. pimpinellifolium LA1589 accession for its relevance as a source for the identification of heat stress tolerance genes, we screened the population under non-controlled heat stress and under controlled, moderate chronic heat stress (MCHS) conditions. Since thermotolerance is a highly complex trait, we examined several vegetative and reproductive parameters that include plant stature and flowering time, inflorescence length, flower size and color, anther structure and apparent damage, fruit color, size and seeds content. Additionally, we analyzed the number of flowers and fruits, the rate of fruit set, single fruit weight, pollen germination rate, and the number of seeds per fruit. We observed a wide range of responses for all traits tested ( Figure 3 ).

Owing to the controlled experimental set up, which allowed the comparison between MCHS and control conditions, we could identify the traits that are directly affected by MCHS. These include number of fruits and fruit set rate, pollen germination rate, and the number of seeds per fruit, all are well known heat stress related traits (Peet et al., 1998; Pressman et al., 2007), thus supporting our hypothesis that this BIL population is a relevant germplasm for exploring heat stress response in tomato.

In the controlled experiments, the number of flowers and single fruit weight were generally not affected by the MCHS conditions employed. Several studies reported a reduction in flower production under high temperature conditions, but under moderate heat stress conditions, flower number seemed to be unaffected (Abdul-Baki, 1991; Xu et al., 2017). Others show that there is no significant decrease in the number of flowers under heat treatment compared to the control (Peet et al., 1997; Sato et al., 2006; Gonzalo et al., 2020). Fruit weight was also reported to be reduced up to 83% under heat stress conditions (Peet et al., 1998). These differences may be attributed to the heat stress conditions employed, i.e., acute stress versus moderate chronic stress. Moreover, the response to the different stress regimes is genotype-dependent. Style protrusion is highly variable between lines and cultivars as the frequency of this trait may range from 12.5% to 80% (Levy et al., 1978). Notably, this phenomenon is naturally occurring in wild relatives of tomato (Chen et al., 2007) and therefore cannot be considered as a pure heat stress related trait in the S. pimpinellifolium population. Nevertheless, our phenotypic screen revealed that about half of the lines (54%) did not present style protrusion at all, and 25% of the lines presented moderate (9%), high (9%), or extreme (7%) level of style protrusion. Levy et al. (1978) reported up to 66% flower buds drop rate in a tomato heat sensitive cultivar. Our population screen results show that although most lines did not present buds or flowers abortion, 16 BILs presented varying abortion rates, ranging between 5% and 70%. Since pollen grain is the most thermosensitive organ (Zinn et al., 2010), special attention is given to pollen quality and performance under heat stress conditions. Pressman et al. (2002) reported a reduction of 62% and 92%, in pollen viability and germination, respectively. Similarly, Firon et al. (2006) showed a 29–78% decrease in pollen viability and 80–90% decrease in pollen germination as a result of heat stress. Our results indicate as well a very strong reduction in pollen viability and germination under MCHS conditions. While some BILs completely lost germination capability, the best performing lines presented a decrease of 32% in pollen germination compared with reductions of 80% in the M82 parental line.

It was very clear throughout our experiments, that LA1589, the parental S. pimpinellifolium wild-species itself, was not more tolerant than the cultivated tomato lines, in any of the traits tested. In fact, it was the most sensitive. This can partially explained by the unnatural environment in the greenhouse which includes detached soil growth media, and consistent watering and fertilization for homogenous treatment to all plants across all experiments, but are actually stressful for the wild-adapted S. pimpinellifolium. Nonetheless, the hypersensitivity of LA1589 to heat stress in our hands, together with the wide variation in heat-stress related traits among the BILs, and the identification of three tolerant lines, suggest that epistatic effects contribute to the observed heat stress tolerance.

Overall, our results highlight the relevance and importance of this BIL population as a new resource for the identification of heat stress related tolerance genes and QTLs. This population was never tested for heat stress related traits. Specifically, the genetic structure of this population may facilitate the identification of candidate genes or QTLs by aligning genetic maps of phenotypically similar lines, as demonstrated by the phenotypic and genetic comparison of pi-058, pi-117 and pi-147 ( Figures 7 – 9 ; Supplementary Figure S3 ). In the analysis presented here, we identified a 0.8 Mbp region on chromosome 9 found exclusively in lines pi-058, pi-117 and pi-147 and absent in the other, non-tolerant lines ( Supplementary Figure S3 ), which was enriched for genes related to heat stress response, thus these may be involved in the observed heat stress tolerance of pi-058, pi-117 and pi-147. This introgression includes seven genes related to heat stress response ( Table 3 ), some of them are known to contribute to tolerance under heat stress conditions. For example, the chaperonin 60 (Cpn60, Solyc09g091180) was found to be overexpressed in thermotolerant tomato that presents high pollen viability under heat stress (Mazzeo et al., 2018) and to support photosystem II under conditions of heat stress (Preczewski et al., 2000). Two additional genes found in the common introgression, Solyc09g091660 and Solyc09g091670, encode ABC transporters responsible for heat stress tolerance by maintaining cell membrane integrity under heat stress conditions. These genes also regulate the transport, distribution and accumulation of ABA and secondary metabolites in different organs of the plant, which in turn modulate thermotolerance (Dahuja et al., 2021). Solyc09g090870 and Solyc09g090890 are involved in DNA mismatch repair which plays an important role in DNA repair following heat stress (Kantidze et al., 2016). Therefore, although Cpn60 is an immediate candidate for supporting the heat stress tolerance of pi-058, pi-117 and pi-147, it is possible that a joint activity of all or some of these seven genes ( Table 3 ) are responsible for the observed effect. Moreover, this introgression may not be the sole effector that promotes thermotolerance, as epistatic effects may also take place, as mentioned here before.

It is interesting to note that among the three thermotolerant lines identified in this study, pi-147 is, for some traits, not performing as well as pi-058 and pi-117 which are highly similar ( Figures 7A , 8 ). Since pi-058 and pi-117 contain another common introgression on chromosome 9 which is absent in pi-147 ( Figure 9 ), genes found in this region may be driving this difference, an hypothesis that requires further studies. Since in our experimental setup, LA1589 was not performing well under MCHS conditions, having evolved in harsh conditions, it is unlikely that the heat stress tolerance of specific BILs will be attributed to a single introgression. It is possible though, that a specific interaction between a single introgression and the cultivated genome will generate a causal network for tolerance. This should be further studied considering the common introgression on chromosome 9 ( Figure 9 ).

Taking into consideration other publications demonstrating a beneficial contribution of S. pimpinellifolium accessions to heat stress tolerance, we suggest that this BIL population may be useful for the identification of heat stress related QTLs. Introgression Lines (IL) population originated from a cross between a cultivated tomato and the S. pimpinellifolium TO-937 accession was used in a study that successfully identified 22 QTLs involved in reproductive traits under high temperatures. For example, fruit-set was reduced by less than 50% in several RILs while in the Moneymaker background, reduction was up to 87% (Gonzalo et al., 2020). Moreover, the specific S. pimpinellifolium accession LA1589 was also suggested to contribute to heat stress tolerance as two clusters of QTLs involved in the responses of reproductive traits to heat stress were identified in a LA1589-derived population (Gonzalo et al., 2022). For each of the heat stress related traits that were tested, we could identify specific lines that were performing significantly better than other lines, and of both parental lines. These lines will serve as a starting genetic material in future studies aiming to identify QTLs and causal genes for each of the tolerance traits. Moreover, since the controlled MCHS experiments were performed on selected lines, not including the entire population (due to space limitations), our results suggest that additional tolerant lines may be identified in future studies that will complement the MCHS controlled screen for the entire population, in a future perspective of developing heat stress tolerant tomato cultivars.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

NB: Conceptualization, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. GM: Formal analysis, Investigation, Methodology, Project administration, Supervision, Writing – review & editing. TD-M: Formal analysis, Investigation, Writing – review & editing. AB: Formal analysis, Investigation, Writing – review & editing. BO: Writing – review & editing, Funding acquisition. DZ: Writing – review & editing, Resources. ML-L: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.