Surprisingly, F1 progeny from parents exposed to the highest degrees of soil salinity showed increased resistance to both biotrophic Hpa and necrotrophic Pc (Fig. 6e,f and Fig. S7a,b). This finding argues against the concept of TAR specificity, since the benefits, rather than the costs, were apparent in mismatched environments. However, the evolutionary significance of this non-specific TAR by salt stress must be considered against the lack of adaptive effects in the matched environment (Fig. 5) and the severe fitness costs arising from reduced plant growth, seed production and seed viability (Fig. 2).
 

Stress intensity acts as a weighted indicator for TAR investment.

Since TAR is associated with costs in mismatched environments, we considered the possibility that plants can adjust TAR investment in accordance to the reliability of the environmental stress signal. We hypothesised that severe stress is perceived as a more reliable predictor of the progeny environment, resulting in stronger TAR investment. In matched environments, Pst- and Pc-elicited TAR was strongest in F1 progeny from parents exposed to the highest stress levels, although the difference in TAR intensity between the lowest and highest parental stress intensities was not statistically significant (Fig. 3a,b, Fig. S2a,b, Fig. 4a and Fig. S3a). Similarly, in mismatched environments, F1 progeny from parents exposed to the highest level of Pst disease showed increased Pc susceptibility compared to progeny from parents exposed to the lowest level of Pst disease; however, this difference was not statistically significant (Fig. 6a and Fig. S5a). By contrast, the increased sensitivity of F1 progeny from Pst-exposed parents to salt showed statistically significant differences that were proportional to the levels of parental disease severity (Fig. 6b and Fig. S5c). Similarly, Hpa susceptibility in F1 progeny from Pc-exposed parents (Fig. 6c and Fig. S6a), as well as non-specific TAR in F1 progeny from salt-exposed parents, showed statistically significant differences that were proportional to the level of parental stress (Fig. 6e,f, Fig. S7). Thus, although the intensity of pathogen-elicited TAR in matched environments of F1 progeny does not show a statistically significant dose effect, the response is associated with dose-dependent costs that become evident in mismatched environments. Finally, we investigated whether the transgenerational stability of TAR into the F2 generation is proportional to the level of parental stress in matched environments. To this end, we determined pathogen-elicited TAR in F2 progeny after one stress-free F1 generation. In contrast to F1 progeny (Fig. 3b and Fig. S2b), F2 progeny from parents exposed to the lowest levels of Pst disease no longer showed TAR against Hpa (Fig. 3c and Fig. S2c). Furthermore, only one F2 population from parents exposed to intermediate levels of Pst disease had maintained a statistically significant TAR response, whereas all F2 populations from parents exposed to the highest levels of Pst disease had maintained a statistically significant TAR response (Fig. 3c and Fig. S2c). F2 populations from parents exposed to low and intermediate levels of Pc disease all failed to show TAR (Fig. 4b and Fig. S3b). However, ANOVA of pooled F2 populations from similarly treated parental plants, as well as nested ANOVA with F2 population as a random variable, revealed a statistically significant effect of parental stress treatment (Fig. S3b), indicating a residual amount of TAR in F2 populations from Pc-exposed parents. This is further supported by the observation that F2 populations from parents exposed to the highest degree of Pc disease also showed the highest level of Pc resistance (Fig. 4b and Fig. S3b). Hence, TAR in response to relatively high levels of disease by Pst or Pc can persist into the F2 generation, whereas TAR elicited by low or intermediate disease levels is reverted or weakened after one stress-free F1 generation. Together, these results demonstrate that the intensity, costs and/or transgenerational stability of TAR have a dose-dependent relationship with parental stress intensity, which supports our hypothesis that plants use stress intensity as a weighted indicator of TAR investment.
  

Discussion

Evolutionary models predict that parental effects on specific traits act as an adaptive mechanism to increase fitness in changeable environments (Leimar & McNamara, 2015; Pigeault et al., 2016; Proulx & Teotonio, 2017). However, despite numerous reports of transgenerational effects of stress in plants, it has remained uncertain whether these responses are adaptive (Uller et al., 2013; Burggren, 2015; Crisp et al., 2016). Although the importance of full factorial designs to address this question has been stressed (Marshall & Uller, 2007; Bonduriansky et al., 2012; Uller et al., 2013; Burgess & Marshall, 2014; Tetreau et al., 2019),
Most previous reports about TAR are based on experiments in which resistance was tested in matched environments and therefore do not address the specificity of the response. The reciprocal design of our study (Fig. 1) allowed us to examine progeny resistance phenotypes in both matched and mismatched environments. In the case of disease stress, we found strong evidence that TAR is specific. Disease by biotrophic Pst bacteria elicits TAR against taxonomically unrelated Hpa with a similar biotrophic lifestyle (Fig. 3b,c) but fails to protect against necrotrophic Pc or abiotic salt stress (Fig. 6a,b,c). Similarly, parental disease by Pc elicited TAR against the same necrotrophic fungus (Fig. 4) but not against biotrophic Hpa and abiotic salt stress (Fig. 6c,d). This specificity supports the notion that pathogen-elicited TAR is an adaptive response. By contrast, we found no evidence for increased salt tolerance in F1 or F2 progeny from salt-exposed parents (Fig. 5), even though F1 progeny from salt-exposed parents showed non-specific resistance against biotrophic and necrotrophic pathogens (Fig. 6e,f). Previous studies have shown that abiotic stress causes non-specific transgenerational tolerance to other abiotic stresses (Boyko et al., 2010; Rahavi et al., 2011), but not biotic stresses. The lack of specific TAR against salt stress (Fig. 4), along with the associated reduction in reproductive fitness (Fig. 2b,c), suggests that salt-elicited TAR is non-adaptive and unlikely provides a selective advantage in natural environment. The discrepancy between salt- and pathogen-elicited TAR can be explained by differences in parental response to these stresses (Fig. 2a,b,c). While all three stresses caused an immediate reduction in plant growth, Pst and Pc had no, or even stimulatory, effects on reproductive fitness (Fig. 2a,b,c), indicating that Arabidopsis uses induced resistance to mitigate Pst and Pc stress, compensating potential fitness loss from reduced growth with increased seed production at the end of its life cycle. This ability to recover from stress constitutes a reliable cue that TAR will improve fitness of progeny in the same environment, thereby justifying TAR investment. By contrast, the progressive loss of seed production and viability upon increasing levels of stress from soil salinity (Fig. 2b,c) indicates that the parental plants do not recover well from this stress. Investment in a transgenerational response is therefore not beneficial, even in a matched environment. This hypothesis is supported by modelling which shows that TAR in invertebrates occurs at intermediate levels of disease stress, but not when there are more severe impacts on mortality (Pigeault et al., 2016).
            Costs are central to the evolution of adaptive transgenerational responses. The fact that pathogen-elicited TAR is inducible and reversible in the absence of stress implies that the response is associated with costs (Fig. 3,4) (Stassen et al., 2018). Previous work has identified transgenerational impacts of parental stress on vegetative and reproductive development (e.g. Rahavi et al., 2011; Suter & Widmer, 2013; Groot et al., 2016), but it remains unclear how far these changes influence fitness. While we did not observe consistent effects on plant growth or seed set in F1 and F2 progeny from disease-exposed plants (data not shown), the reciprocal design of our experiments strongly indicates ecological costs arising from increased susceptibility to other stresses (Fig. 6a,b,c,d). Antagonism between plant defence pathways against biotrophic pathogens, necrotrophic pathogens and abiotic stress is well-documented (Koornneef & Pieterse, 2008; Pieterse et al., 2012), and transgenerational persistence of these effects have been reported previously (Luna et al., 2012; Singh et al., 2017). Accordingly, we propose that negative cross-talk between defence pathways imposes a major cost on adaptive TAR responses to pathogens.
Although examples of transgenerational phenotypic plasticity are now widespread, there are instances where researchers have failed to identify such effects in Arabidopsis (Pecinka et al., 2009; Suter & Widmer, 2013). Indeed, evolutionary theory predicts that transgenerational plasticity is not a universal trait and that its occurrence depends highly on genotype, mode of reproduction, ecological niche and life history traits, as well as the nature and consistency of the eliciting stress (Crisp et al., 2016; Groot et al., 2016). Compared to fixed genetic adaptation, it can be expected that transgenerational phenotypic plasticity offers a suitable adaptation strategy under variable environments. However, it is unlikely that either adaptation strategy will be under positive selection in highly variable, unpredictable environments, since the frequency of incurred costs would outweigh the specific benefits. Adaptive parental effects would therefore more likely emerge when the same type of environmental stress occurs regularly (Tricker, 2015). Under such conditions, stress-exposed plants can optimise fitness either by maximising their own immediate performance to the detriment of their progeny (‘selfish parental effects’), or by modifying progeny traits to provide enhanced performance in the altered environment (Marshall & Uller, 2007). The latter strategy can take form in either a diversified bet-hedging strategy, or a more deterministic provision of specific adaptive traits, such as pathogen-elicited TAR, which is tailored to the parental environment (Marshall & Uller, 2007; Crean & Marshall, 2009; Proulx & Teotonio, 2017). Not only do evolutionary models predict that transgenerational phenotypic plasticity is likely to evolve in fluctuating environments (Leimar & McNamara, 2015; Pigeault et al., 2016; Proulx & Teotonio, 2017), the model developed by Proulx and Teotonio (2017) suggests that deterministic (adaptive) parental effects like pathogen-elicited TAR provide increased fitness over a wider range of environmental parameters than a randomising bet-hedging strategy.
Central to the provision of adaptive transgenerational traits is the ability to make accurate and reliable predictions about future progeny environments. While this aspect has been emphasised in both evolutionary theory and modelling of parental effects (Burgess & Marshall, 2014; Leimar & McNamara, 2015), it has rarely been addressed experimentally. The few studies to have included this concept applied the same stress repeatedly over multiple generations rather than applying different stress intensities within the same generation. In one of the most comprehensive studies of this type, Groot et al. (2016) found complex interactions between parental (P), grandparental (GP) and great-grandparental (GGP) salt stress in Arabidopsis. When the stress was applied to only one generation, P effects were typically stronger than GP and GGP effects. For treatments over multiple generations, the impacts of GP and GGP stress were additive to P treatments for some traits, but antagonistic for others. Furthermore, the transgenerational effects in the study by Groot et al. (2016) varied between controlled environments and field conditions, making it difficult to conclude whether the effects were adaptive. In our study, varying levels of three different stresses were applied within one generation, providing a straightforward design to assess whether parents can distinguish stress severities and adjust the transgenerational response accordingly. Our pathogen treatments resulted in dose-dependent impacts on relative growth rate during the treatment period (Fig. 2a), indicating that Arabidopsis perceives these stresses in a dose-dependent manner. Furthermore, analysis of the transgenerational stability of TAR provided evidence for a dose-dependent relationship with parental disease severity. Although F1 populations from both Pst- and Pc-exposed parents expressed TAR to statistically similar levels across stress levels (Fig. 3a,b and Fig. 4a), TAR only persisted over a stress-free generation when elicited by the highest stress levels (Fig. 3c and Fig. 4b). Furthermore, in mismatched environments, there was a dose-dependent effect on the costs of pathogen-elicited TAR: both salt sensitivity of F1 progeny from Pst-infected parents and Hpa susceptibility of F1 progeny from Pc-infected parents correlated with the severity of parental disease stress treatment (Fig. 6b,c). Overall, these results support our hypothesis that plants perceive disease severity as a predictive cue to adjust TAR investment.
Collectively, our study demonstrates that parental investment in pathogen-elicited TAR provides fitness benefits in matched environments and costs in mismatched environments. This stress-specific TAR is dependent on the intensity of the stress experienced by the parents, which holds predictive value for future progeny environments. Accordingly, our findings are consistent with the evolutionary prediction that pathogen-elicited TAR is a genuine adaptive trait in Arabidopsis. In one of the most convincing cases of adaptive parental effects in plants, Galloway and Etterson (2007) used field-based studies to demonstrate adaptive transgenerational plasticity in response to the light environment. It will now be of interest to undertake ecological field studies and verify our laboratory experiments in support of TAR as an adaptive transgenerational effect in nature.
  

Materials and Methods

Plant material and growth conditions.

All Arabidopsis thaliana lines described in this study are in the genetic background of accession Col-0 (NCBI, Tax ID 3702). To exclude confounding effects of TAR from stress in previous generations, all lines originated from a common ancestor of a population that had maintained under stress-free conditions (mock-inoculated) in two previous generations (Luna et al., 2012). Except for the stress treatments, all plants were grown under similar conditions (see Supplementary Methods in Supporting Information for details). To generate F1 populations, 6-8 parental plants of 4.5-weeks-old were subjected to mock/stress treatments over a duration of 3 weeks, after which 4 parental plants with representative symptoms were moved to long-day conditions (16 h light/8 h darkness) to set seed and generate F1 populations (Fig. 1). Three individual plants from each F1 population were kept apart under stress-free conditions to set seed, resulting in 3 F2 populations from each F1 population and a total of 12 F2 populations per parental treatment (Fig. 1). Details of all F1 and F2 populations are presented in Table S1 in the Supporting Information.
 

Stress treatments of parental plants.

Inoculation with biotrophic Pseudomonas syringae pv. tomato (Pst) was performed at 3-4 day intervals over a total period of 3 weeks, as detailed in the Supplementary Methods. Plants were subjected to different Pst disease pressures: no disease (Mock; 6 subsequent inoculations with the mock suspension), low disease (Pst-I; 2 inoculations with Pst followed by 4 mock inoculations), medium disease (Pst-II; 4 inoculations with Pst followed by 2 mock inoculations solution), and high disease (Pst-III; 6 subsequent inoculations with Pst). To ensure necrotrophic infection by Plectophaerella cucumerina (Pc), inoculation was performed by placing 6 µl-droplets (106 spores ml-1) onto fully expanded leaves of approximate similar age (Petriacq et al., 2016), as detailed in the Supplementary Methods. Plants were subjected to different Pc disease pressures: no disease (Mock; 6 leaves were mock-inoculated), low disease (Pc-I; 2 leaves Pc-inoculated and 4 leaves mock-inoculated), medium disease (Pc-II; 4 leaves Pc-inoculated and 2 leaves mock-inoculated), and high disease (Pc-III; 6 leaves Pc-inoculated). After inoculation, plants were kept at 100% RH for 2 weeks until visible disease symptoms appeared in >80% of the leaf surface (necrosis and chlorosis). To prevent sporulation and ongoing disease progression, plants were returned to 60% RH before moving to long-day conditions 1 week later. Salt stress was applied by soil-drenching with 100 mM NaCl solution. Plants were subjected to different stress levels over the 3-week period: mock treatment (S-I; drenched 6x with water), low stress (S-II; drenched 2x with NaCl and 4x with water), medium stress (S-III; drenched 4x with NaCl and 2x with water), high stress (S-III; plants drenched 6x with NaCl). Plants returned to a normal watering regime when moved to long-day conditions.

 

Quantification of fitness parameters.

Relative growth rate (RGR) was determined non-destructively by quantification of green leaf area (GLA) before and after stress treatments, as detailed in the Supplementary Methods. Reproductive fitness was estimated by seed production and seed viability, as described in the Supplementary Methods.

 

Quantification of transgenerational resistance phenotypes.

To quantify resistance against biotrophic Pst, the inoculum was prepared and adjusted to 2x105 CFU mL-1 (see Supplementary Methods). Inoculation was performed by syringe infiltration of 4 leaves/plant of approximate similar age. Bacterial growth was quantified at 3 days post inoculation (dpi) by dilution plating on selective agar plates (see Supplementary Methods). Inoculation with biotrophic Hpa and quantification of Hpa resistance was performed as described previously (Lopez Sanchez et al., 2016; see also Supplementary Methods). Quantification of salt tolerance was based on root growth analysis on agar plates containing 0mM, 50 mM and 100 mM NaCl. Assays were conducted as described previously (Verslues et al., 2006; Claeys et al., 2014) with minor modifications (see Supplementary Methods).

 

Statistical analysis.

Analytical statistics was performed using R studio (v 1.1.456, https://rstudio.com/), supporting R software (v 3.5.1, https://www.r-project.org/). Statistical significance of treatment effects on continuous variables was analysed by linear models; statistical significance of treatment effects on categorical variables (class frequencies) was analysed by Fisher’s exact tests. Details about data transformations, statistical models, and R software packages are described in the Supplementary Methods.
 

Acknowledgements

The work was supported by ERC (no. 309944 “Prime-A-Plant” and no. 824985 “ChemPrime”) to JT, a Research Leadership Award from the Leverhulme Trust (no. RL-2012-042) to JT, a BBSRC-IPA grant to JT (BB/P006698/1) and a BBSRC responsive mode grant to MR and JT (BB/L008939/1). Seeds of Arabidopsis F1 and F2 populations can be made available upon request with JT (Table S1).
  
Competing interests statement
There are no competing interests of a financial or non-financial nature.

  

Author Contributions

JT and MR conceived the project; JT and ALS designed and supervised the experiments; ALS, DP and LF performed bioassays; JT performed statistical analyses; MR, ALS and JT wrote the manuscript. All authors reviewed and approved the final manuscript. All authors declare no competing interests.
  

References

Bonduriansky R, Crean AJ, Day T. 2012. The implications of nongenetic inheritance for evolution in changing environments. Evolutionary Applications 5(2): 192-201.
Bošković A, Rando OJ. 2018. Transgenerational epigenetic inheritance. Annual Review of Genetics 52(1): 21-41.
Boyko A, Blevins T, Yao Y, Golubov A, Bilichak A, Ilnytskyy Y, Hollander J, Meins F, Kovalchuk I. 2010. Transgenerational adaptation of Arabidopsis to stress requires DNA methylation and the function of Dicer-Like proteins. PLoS ONE 5(3): e9514.
Burgess SC, Marshall DJ. 2014. Adaptive parental effects: the importance of estimating environmental predictability and offspring fitness appropriately. Oikos 123(7): 769-776.
Burggren WW. 2015. Dynamics of epigenetic phenomena: intergenerational and intragenerational phenotype ‘washout’. The Journal of Experimental Biology 218(1): 80-87.
Claeys H, Van Landeghem S, Dubois M, Maleux K, Inze D. 2014. What is stress? Dose-response effects in commonly used in vitro stress assays. Plant Physiol 165(2): 519-527.
Crean AJ, Marshall DJ. 2009. Coping with environmental uncertainty: dynamic bet hedging as a maternal effect. Philosophical Transactions of the Royal Society B: Biological Sciences 364(1520): 1087-1096.
Crisp PA, Ganguly D, Eichten SR, Borevitz JO, Pogson BJ. 2016. Reconsidering plant memory: Intersections between stress recovery, RNA turnover, and epigenetics. Sci Adv 2(2): e1501340.
Furci L, Jain R, Stassen J, Berkowitz O, Whelan J, Roquis D, Baillet V, Colot V, Johannes F, Ton J. 2019. Identification and characterisation of hypomethylated DNA loci controlling quantitative resistance in Arabidopsis. eLife 8: e40655.
Galloway LF, Etterson JR. 2007. Transgenerational plasticity is adaptive in the wild. Science 318(5853): 1134-1136.
Gehring M. 2019. Epigenetic dynamics during flowering plant reproduction: evidence for reprogramming? New Phytologist 224(1): 91-96.
Groot MP, Kooke R, Knoben N, Vergeer P, Keurentjes JJB, Ouborg NJ, Verhoeven KJF. 2016. Effects of multi-generational stress exposure and offspring environment on the expression and persistence of transgenerational effects in Arabidopsis thaliana. PLoS ONE 11(3): e0151566.
Herman JJ, Sultan SE. 2011. Adaptive transgenerational plasticity in plants: case studies, mechanisms, and implications for natural populations. Front Plant Sci 2: 102.
Herman JJ, Sultan SE, Horgan-Kobelski T, Riggs C. 2012. Adaptive transgenerational plasticity in an annual plant: grandparental and parental drought stress enhance performance of seedlings in dry soil. Integr Comp Biol 52(1): 77-88.
Holeski LM, Jander G, Agrawal AA. 2012. Transgenerational defense induction and epigenetic inheritance in plants. Trends Ecol Evol 27(11): 618-626.
Iwasaki M, Paszkowski J. 2014. Identification of genes preventing transgenerational transmission of stress-induced epigenetic states. Proc Natl Acad Sci U S A 111(23): 8547-8552.
Koornneef A, Pieterse CM. 2008. Cross talk in defense signaling. Plant Physiol 146(3): 839-844.
Lämke J, Bäurle I. 2017. Epigenetic and chromatin-based mechanisms in environmental stress adaptation and stress memory in plants. Genome Biology 18(1): 124.
Leimar O, McNamara JM. 2015. The evolution of transgenerational integration of information in heterogeneous environments. The American Naturalist 185(3): E55-E69.
Lopez Sanchez A, Stassen JHM, Furci L, Smith LM, Ton J. 2016. The role of DNA (de)methylation in immune responsiveness of Arabidopsis. Plant Journal 88(3): 361-374.
Luna E, Bruce TJA, Roberts MR, Flors V, Ton J. 2012. Next-generation systemic acquired resistance. Plant Physiology 158(2): 844-853.
Luna E, Ton J. 2012. The epigenetic machinery controlling transgenerational systemic acquired resistance. Plant Signal Behav 7(6): 615-618.
Marshall DJ, Uller T. 2007. When is a maternal effect adaptive? Oikos 116(12): 1957-1963.
Pecinka A, Rosa M, Schikora A, Berlinger M, Hirt H, Luschnig C, Mittelsten Scheid O. 2009. Transgenerational stress memory is not a general response in Arabidopsis. PLoS ONE 4(4): e5202.
Perez MF, Lehner B. 2019. Intergenerational and transgenerational epigenetic inheritance in animals. Nature Cell Biology 21(2): 143-151.
Petriacq P, Stassen JH, Ton J. 2016. Spore Density Determines Infection Strategy by the Plant Pathogenic Fungus Plectosphaerella cucumerina. Plant Physiol 170(4): 2325-2339.
Pieterse CM, Van der Does D, Zamioudis C, Leon-Reyes A, Van Wees SC. 2012. Hormonal modulation of plant immunity. Annu Rev Cell Dev Biol 28: 489-521.
Pigeault R, Garnier R, Rivero A, Gandon S. 2016. Evolution of transgenerational immunity in invertebrates. Proceedings of the Royal Society B: Biological Sciences 283(1839).
Proulx SR, Teotonio H. 2017. What kind of maternal effects can be selected for in fluctuating environments? Am Nat 189(6): E118-e137.
Rahavi MR, Migicovsky Z, Titov V, Kovalchuk I. 2011. Transgenerational adaptation to heavy metal salts in Arabidopsis. Frontiers in Plant Science 2(91).
Rasmann S, De Vos M, Casteel CL, Tian D, Halitschke R, Sun JY, Agrawal AA, Felton GW, Jander G. 2012. Herbivory in the previous generation primes plants for enhanced insect resistance. Plant Physiol 158(2): 854-863.
Schlichting CD. 1986. The evolution of phenotypic plasticity in plants. Annual Review of Ecology and Systematics 17(1): 667-693.
Singh P, Dave A, Vaistij FE, Worrall D, Holroyd GH, Wells JG, Kaminski F, Graham IA, Roberts MR. 2017. Jasmonic acid-dependent regulation of seed dormancy following maternal herbivory in Arabidopsis. New Phytologist 214(4): 1702-1711.
Stassen JHM, Lopez A, Jain R, Pascual-Pardo D, Luna E, Smith LM, Ton J. 2018. The relationship between transgenerational acquired resistance and global DNA methylation in Arabidopsis. Sci Rep 8(1): 14761.
Suter L, Widmer A. 2013. Phenotypic effects of salt and heat stress over three generations in Arabidopsis thaliana. PLoS ONE 8(11): e80819.
Tetreau G, Dhinaut  J, Gourbal B, Moret Y. 2019. Trans-generational immune priming in invertebrates: Current knowledge and future prospects. Frontiers in Immunology 10(1938).
Tricker PJ. 2015. Transgenerational inheritance or resetting of stress-induced epigenetic modifications: two sides of the same coin. Front Plant Sci 6: 699.
Uller T, Nakagawa S, English S. 2013. Weak evidence for anticipatory parental effects in plants and animals. J Evol Biol 26(10): 2161-2170.
Verslues PE, Agarwal M, Katiyar-Agarwal S, Zhu J, Zhu JK. 2006. Methods and concepts in quantifying resistance to drought, salt and freezing, abiotic stresses that affect plant water status. Plant J 45(4): 523-539.
Wilkinson SW, Magerøy MH, López Sánchez A, Smith LM, Furci L, Cotton TEA, Krokene P, Ton J. 2019. Surviving in a hostile world: Plant strategies to resist pests and diseases. Annual Review of Phytopathology 57(1): 505-529.
 
 

Figure legends

Fig. 1. Full factorial experimental design of the study. Arabidopsis thaliana plants (accession Col-0) from a common ancestor were exposed to increasing stress intensities (Mock, Low, Medium and High) by the (hemi)biotrophic bacterial pathogen Pseudomonas syringae pv. tomato (green), the necrotrophic fungal pathogen Plectosphaerella cucumerina (blue), or soil salinity (NaCl; orange). Plants in this parental generation (P) were evaluated for impacts on fitness parameters. Four plants per stress level were selected to generate F1 populations, which were analysed for transgenerational changes in resistance against all 3 stresses, in order to determine the specificity of transgenerational acquired resistance (TAR), potential costs arising from increased susceptibility, and dose-dependency of TAR intensity on parental stress. Four individual plants from 3 independent F1 populations were randomly selected to set seed in the absence of stress. The resulting F2 populations were analysed for resistance against the parental stress to examine dose-dependent effects on TAR stability. Circles indicate individual plants; small (thin-lined) boxes indicate F1/F2 populations derived from a common ancestor in the previous generation; big (bold-lined) boxes indicate pooled F1/F2 populations from a common ancestor 2 generations earlier.
 
Fig. 2. Differential impacts of three (a)biotic stresses on parental fitness parameters. Plants in the parental generation (4.5-weeks-old) were exposed to varying stress intensities by P. syringae pv tomato (Mock, Pst-I, Pst-II, Pst-III), P. cucumerina (Mock, Pc-I, Pc-II, Pc-III) or soil salinity (Mock, S-I, S-II, S-III) over a 3-week period before transferring to long-day conditions to trigger flowering and set seed. Boxplots show the interquartile range (IQR; box) ± 1.5xIQR (whiskers), including median (horizonal line) and replication units (single dots). (a) Impacts on relative growth rate (RGR) during the period of stress exposure. Data represent RGR values of single plants (n= 5-6) normalised to the average RGR of Mock-treated plants (100%). Different letters indicate statistically significant differences (ANOVA + Tukey post-hoc test, α=0.05). (b) Impacts on seed production. Data represent seed numbers per plant (n= 5-6) normalised to average value of Mock-treated plants (100%). Different letters indicate statistically significant differences (Pst: Welch ANOVA + Games-Howell posthoc test, α=0.05; Pc and salt: ANOVA + Tukey post-hoc test, α=0.05). (c) Impacts on seed viability. Seed viability was determined 5 days after planting of surface-sterilised and stratified seeds onto 0.2x Murashige and Skoog (MS) agar plates. Data represent mean germination percentages per plate (25 seeds/plate) of seed batches from 4 similarly treated parents (n=15-60). Different letters indicate statistically significant differences (Welch ANOVA + Games-Howell post-hoc test; α=0.05). Viability data for seed batches from individual plants are presented in Fig. S1a-c.
 
Fig. 3. Parental plants had been exposed to different disease severities by the biotrophic bacterium Pseudomonas syringae pv. tomato (Pst; Mock, Pst-I, Pst-II, Pst-III).
F1 and F2 plants were analysed for resistance against the same pathogen (Pst) and/or the biotrophic Oomycete Hyaloperonospora arabidopsidis (Hpa). (a) TAR against Pst in F1 progeny at 3 days post inoculation (dpi). Boxplots show the interquartile range (IQR; box) ± 1.5xIQR (whiskers), including median (horizonal line) and replication units (dots). Data represent 10Log-transformed bacterial titres (Log cfu cm-2) in leaves of single plants within F1 populations from similarly treated parents (n=42). Different letters indicate statistically significant differences (Welch ANOVA + Games-Howell test, α=0.05). Data for individual F1 populations are shown in Fig. S2a. (b) TAR against Hpa in F1 progeny. Hpa colonisation was quantified at 6 dpi by assigning trypan-blue stained leaves to 4 Hpa resistance classes (I: healthy; II: hyphal colonisation only; III hyphal colonization with conidiospores; IV hyphal colonisation with conidiospores and oospores). Stacked bars show leaf frequency distributions within F1 populations from similarly treated parental plants (n=600-1000). Different letters indicate statistically significant differences (Pairwise Fisher’s exact tests + Bonferroni FDR, α=0.05). Data for individual F1 populations are shown in Fig. S2b. (c) TAR against Hpa in F2 progeny at 6 dpi after one stress-free F1 generation. Stacked bars show leaf frequency distributions across Hpa resistance classes within F2 populations that share a common parental ancestor (n=300-350). Different letters indicate statistically significant differences (Pairwise Fisher’s exact tests + Bonferroni FDR; α=0.05). Data for individual F2 populations are shown in Fig. S2c.
 
Fig. 4. Intensity and transgenerational stability of Plectosphaerella cucumerina-elicited TAR in matched environments.  Parental plants had been exposed to different disease severities by necrotrophic Plectosphaerella cucumerina (Pc; Mock, Pc-I, Pc-II, Pc-III). F1 and F2 plants were analysed for resistance against the same pathogen. Lesion diameters were determined in 4 leaves/plant at 15 days post inoculation (dpi) and the average lesion diameter per plant was used as statistical unit of replication. Boxplots show the interquartile range (IQR; box) ± 1.5xIQR (whiskers), including median (horizonal line) and replication units (dots). (a) TAR against Pc in F1 progeny. Data represent lesion diameters (mm) of plants within F1 populations from similarly treated parents (n=40). Different letters indicate statistically significant differences (Welch ANOVA + Games-Howell test, α=0.05). Data for individual F1 populations are shown in Fig. S3a. (b) TAR against Pc in F2 progeny after a stress-free F1 generation. Data represent lesion diameters of plants within F2 populations that share a common parental ancestor (n=20). Different letters indicate statistically significant differences (ANOVA + Tukey post-hoc test; α=0.05). Data for individual F2 populations are shown in Fig. S3b.