Acclimation to high daily thermal amplitude converts a defense response regulator into susceptibility factor

Plant Immunity, fungal pathogen, acclimation, priming, transcription factor 10 11

12 Acclimation enables plants to adapt to immediate environmental fluctuations, supporting biodiversity 13 and ecosystem services. However, global changes are altering conditions for plant disease outbreaks, 14 increasing the risk of infections by pathogenic fungi and oomycetes, and often undermining plant 15 immune responses. Understanding the molecular basis of plant acclimation is crucial for predicting 16 climate change impacts on ecosystems and improving crop resilience. Here, we investigated how 17 Arabidopsis thaliana quantitative immune responses acclimates to daily temperature fluctuations. We 18 analyzed responses to the fungal pathogen Sclerotinia sclerotiorum following three acclimation 19 regimes that reflect the distribution areas of both species. Mediterranean acclimation, characterized 20 by broad diurnal temperature amplitudes, resulted in a loss of disease resistance in three natural A. 21 thaliana accessions. Global gene expression analyses revealed that acclimation altered nearly half of 22 the pathogen-responsive genes, many of which were down-regulated by inoculation and associated 23 with disease susceptibility. Phenotypic analysis of A. thaliana mutants identified novel components of 24 quantitative disease resistance following temperate acclimation. Several of these mutants were 25 however more resistant than wild type following Mediterranean acclimation. Notably, mutant lines in 26 the NAC42-like transcription factor did not show a loss of resistance under Mediterranean acclimation. 27 This resistance was linked to an acclimation-mediated switch in the repertoire of NAC42-like targets 28 differentially regulated by inoculation. These findings reveal the rewiring of immune gene regulatory 29 networks by acclimation and suggest new strategies to maintain plant immune function in a warming 30 climate. 31

32 The ability of species to cope with rising temperatures is a crucial factor influencing range shifts 33 and local extinctions, as their distribution and range boundaries closely align with temperature 34 gradients. Evidence shows that plant species adapt to local environmental conditions through genetic 35 variation (Fournier-Level et al., 2011b; Katz et al., 2021; Clauw et al., 2022) but they also exhibit 36 phenotypic plasticity allowing individual plants to adjust rapidly their physiology to environmental 37 variations (Valladares et al., 2014; Brancalion et al., 2018). The short-term, reversible process that 38 allows plants to cope with immediate environmental fluctuations is often referred to as acclimation 39 (Kleine et al., 2021). Plant acclimation help maintain the balance of natural systems, supporting 40 biodiversity and the services that ecosystems provide, such as carbon sequestration and water 41 regulation. With climate change modifying the distribution area of plants (Sloat et al., 2020) and 42 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 2 causing more frequent and severe weather events (Newman and Noy, 2023), knowledge of how plants 43 acclimate can inform strategies to manage ecosystems and agriculture. In this context, crops that can 44 acclimate effectively are more likely to maintain high yields despite stressors. A better understanding 45 of the genetic underpinnings of plant acclimation is therefore crucial for predicting the impact of 46 climate change on ecosystems and for improving crop resilience. 47 Acclimation distinguishes from adaptation for involving changes to the expression of the 48 genome instead of heritable changes to genome sequences (Kleine et al., 2021). Epigenetic and 49 transcriptional regulation mechanisms mediating somatic stress memory are important players in 50 plant acclimation (Charng et al., 2023; Zuo et al., 2023; Hadj-Amor et al., 2024). Cold acclimation, by 51 which decreasing temperatures enhance freezing tolerance in plants involves alterations in membrane 52 composition, the production of cryoprotective polypeptides and solutes, the activation of cold-53 responsive (COR) genes regulated by C-repeat binding transcription factors (CBFs/DREB1) (Liu et al., 54 2019). The accumulation of heat shock proteins (HSPs) regulated by heat shock transcription factors 55 (HSFs) and histone 3 K4 methylation play a key role in heat acclimation (Kappel et al., 2023; Nishad and 56 Nandi, 2021). Besides transcription factors and epigenetic marks, the hormone abscisic acid (ABA) is a 57 central mediator of the accumulation of LEA-like protective proteins, stomatal closure and 58 downregulation of photosynthesis under drought acclimation (Sadhukhan et al., 2022). Despite recent 59 efforts, the interplay between regulatory mechanisms, molecular and phenotypic responses to plant 60 acclimation is elusive. 61 With changes to the climate, not only the distribution range of plants changes, but also that of 62 their enemies. Suitable conditions for plant disease outbreaks are expected to shift in time and space 63 leading to a global poleward movement of plant pathogen geographic niches (Bebber et al., 2013) and 64 an increased risk of infection by pathogenic fungi and oomycetes (Chaloner et al., 2021). Fungi, 65 especially generalists with a broad range of plant hosts, are the most widespread and most rapidly 66 spreading pathogens, so that if current rates persist, several major food producing countries would 67 have fully saturated pathogen distributions by 2050 (Bebber et al., 2014). A paradigmatic example of 68 such broad host range pathogen is the white and stem mold fungus Sclerotinia sclerotiorum, which 69 infects hundreds of plant species and causes significant losses to vegetable and oil crops worldwide 70 (Navaud et al., 2018; Peltier et al., 2012; Cohen, 2023). Although climate change may alter the overlap 71 between crops cultivation area and S. sclerotiorum distribution range (Mehrabi et al., 2019), pathogen 72 strains adapted to warm temperatures have been reported (Uloth et al., 2015) and extreme 73 temperature may promote fungal development (Lane et al., 2019; Shahoveisi et al., 2022), raising 74 concern about Sclerotinia disease incidence in the future (Singh et al., 2023). 75 Plant respond to S. sclerotiorum by activating quantitative disease resistance (QDR), an 76 immune response involving multiple genes of weak to moderate phenotypic effect (Roux et al., 2014; 77 Sucher et al., 2020). Molecular players involved in QDR against S. sclerotiorum include immune 78 receptors, reactive oxygen species, phytohormones such as ABA, jasmonic acid and ethylene, 79 transcription factors and phytoalexins (Perchepied et al., 2010; Mbengue et al., 2016; Derbyshire and 80 Raffaele, 2023). Several of these determinants contribute to multiple biological processes such as plant 81 development and response to the abiotic environment (Corwin et al., 2016; Badet et al., 2019; Léger 82 et al., 2022). The genetic architecture of QDR suggests that the expression of many genes involved in 83 QDR could be modulated by environmental conditions (Hadj-Amor et al., 2024), and that climate 84 change may alter plant QDR response to S. sclerotiorum at the phenotypic and molecular level. 85 Temperature increase notably is known to frequently impair plant immune responses, including QDR 86 (Desaint et al., 2021; Aoun et al., 2017). 87 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 3 Analyses of plant immune responses under abiotic constraints generally focus on pathogen 88 inoculation under prolonged and stable abiotic conditions. In A. thaliana , immunity against the 89 bacterial pathogen Pseudomonas syringae pv. tomato at elevated temperature can be restored by the 90 constitutive expression of CBP60g, a major transcriptional regulator of plant immunity genes and 91 salicylic acid (SA) defense hormone production, downregulated by temperature (Kim et al., 2022). This 92 finding indicates that engineering plant transcriptional circuits can mitigate the negative effect of 93 climate change on some plant immune responses. Whether this strategy would restore resistance 94 against necrotrophic pathogens such as S. sclerotiorum, only weakly sensitive to SA-mediated defense, 95 remains to be determined. Another promising target is the disordered protein TWA1, a temperature 96 sensor proposed to orchestrate acclimation by integrating temperature with ABA and JA signaling 97 (Bohn et al., 2024), which play important roles in plant defense against necrotrophs. When applied 98 sequentially, prior abiotic signals may alter the transcriptional and metabolic response to a subsequent 99 pathogen inoculation (Coolen et al., 2016; Garcia-Molina et al., 2020; Garcia-Molina and Pastor, 2024). 100 In addition to mean temperature increase, climate change drives an expansion of diurnal temperature 101 range (Zhong et al., 2023). Daily fluctuations of the environment may alter plant metabolism, growth 102 and flowering (Burghardt et al., 2016; Deng et al., 2021; Matsubara, 2018) as well as gene regulation 103 and invasive growth of fungal pathogens (Jallet et al., 2020; Bernard et al., 2022). Yet, how plant 104 immunity acclimates to daily temperature fluctuations remains largely unexplored. 105 To fill this gap, we analyzed A. thaliana immune responses upon S. sclerotiorum inoculation 106 following three acclimation regimes representing the distribution area of these two species. 107 Mediterranean acclimation, characterized by a broad diurnal temperature amplitude, caused a loss of 108 disease resistance in the three natural accessions we tested. Using global gene expression analyses, 109 we show that acclimation alters the expression of nearly a half of pathogen-responsive genes, many 110 of which are down-regulated by inoculation and associated with disease susceptibility. The phenotypic 111 analysis of A. thaliana mutants identified novel components of QDR following temperate acclimation. 112 Several of these mutants were however more resistant than wild type following Mediterranean 113 acclimation. In particular, contrary to wild type, two mutant lines in the NAC42-like transcription factor 114 showed no loss of resistance upon Mediterranean acclimation. These phenotypes associated with a 115 switch in the repertoire of NAC42-like targets differentially regulated by inoculation according to 116 acclimation. These findings reveal the rewiring of immune gene regulatory networks by acclimation 117 and open new perspectives to safeguard the functioning of the plant immune system in a warming 118 climate. 119

120 Mediterranean-like acclimation impairs the resistance of several A. thaliana accessions to S. 121 sclerotiorum 122 To determine the effect of acclimation on A. thaliana quantitative disease resistance (QDR), we 123 analyzed phenotypic variation of three A. thaliana accessions after growth in three simulated climates. 124 We selected accessions Col-0, Rld-2 and Shahdara (Sha) as representatives of genetic and geographical 125 diversity of A. thaliana species. For acclimation, plants were grown under day length, day and night 126 temperatures corresponding to the 30-year average for the month of April in areas with a temperate 127 (Cfa), continental (Dfb) and Mediterranean (Csa) climates ( Fig. 1A, Fig. S1 ). These correspond to 128 climates in the distribution range of A. thaliana with major projected area variation by the end of this 129 century (Alonso-Blanco et al., 2016; Peel et al., 2007; Cui et al., 2021). Plants were grown for 35 days 130 under temperate and Mediterranean climate, corresponding to 13,405 and 13,930 °C.days, and for 70 131 days under continental climate corresponding to 11,480 °C.days before inoculation with S. 132 sclerotiorum under infection-conducive conditions. Plant susceptibility was assessed using time-133 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 4 resolved automated phenotyping (Barbacci et al., 2020). After temperate acclimation, all accessions 134 appeared similarly susceptible with only a slightly lower susceptibility (-9% average) for Rld-2 and a 135 slightly higher susceptibility for Sha (+18% average) compared to Col-0 ( Fig. 1B, Table S1 ). These 136 phenotypes were not significantly altered upon continental acclimation. Mediterranean acclimation 137 rendered all accession significantly more susceptible, with an average increase by 29% for Sha, 37% 138 for Col-0 and 86% for Rld-2 as compared to temperate acclimation (Fig 1B, C). These results show that 139 both genotype and acclimation affect the susceptibility of A. thaliana to S. sclerotiorum and that, 140 regardless of genotype, Mediterranean acclimation caused the most significant loss of resistance. 141 142 Fig 1. Effect of three distinct pre-infection climate conditions (acclimation) on A. thaliana quantitative disease resistance 143 to S. sclerotiorum . (A) Experimental design showing acclimation and inoculation phases. Daylength, day and night 144 temperatures typical of temperate, continental and Mediterranean climate conditions define the three acclimation 145 conditions used in this work. (B) Susceptibility phenotype in response to S. sclerotiorum infection as a function of acclimation 146 and genotype. Each experiment was repeated at least 3 times and the significance of the results was assessed by an ANOVA 147 followed by a Tukey HSD test, with significance groups labelled by letters. Boxplots show first and third quartiles (box), median 148 (thick line), and the most dispersed values within 1.5 times the interquartile range (whiskers). (C) Representative symptoms 149 of Col-0 plants between 10 and 50 hours post-inoculation by S. sclerotiorum on leaves harvested on plants acclimated in 150 temperate and Mediterranean (Mediterran.) conditions. 151 Acclimation primarily alters the expression of infection-downregulated genes 152 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 5 To study the molecular bases of quantitative disease resistance acclimation, we performed a global 153 transcriptome analysis of A. thaliana accessions Col-0, Rld-2 and Sha grown in temperate, continental 154 and Mediterranean climates, followed or not by S. sclerotiorum inoculation. To identify genes 155 responsive to infection we performed a differential expression analysis using non-inoculated plants as 156

in each of nine conditions (three climate priming, times three plant genotypes). We found 157 17,137 nuclear-encoded genes with sufficient coverage ( Table S2), and identified 9,580 differentially 158 expressed genes (DEGs) upon inoculation at |Log2 Fold Change|≥2 and Bonferroni-adjusted p-159 val<0.0001 ( Fig 2A, Fig S2, Table S3 ). The number of upregulated genes ranged from 1,744 (Rld-2 160 Mediterranean acclimation) to 3,084 (Rld-2 continental acclimation), the number of downregulated 161 genes ranged from 442 (Rld-2 Mediterranean acclimation) to 3,387 (Sha temperate acclimation). The 162 three accessions showed a reduced number of DEGs when acclimated in conditions very divergent to 163 the climate at their area of origin. Downregulated genes showed a relatively high degree of specificity 164 with 1,212 genes (23%) unique to one genotype-acclimation pair and only 89 (1.7%) genes differential 165 in all nine genotype-acclimation conditions tested ( Fig 2B ). Upregulated genes showed higher 166 robustness with 1,105 genes (26.8%) differential in all nine genotype-acclimation conditions. 167 Next, we performed an analysis of variance on the 9,580 DEGs to determine which of the plant 168 genotype, infection status, acclimation, and their interactions, contributed the most to expression 169 variation for each gene. As expected, infection contributed significantly (Benjamini-Hochberg 170 corrected p-val <1E-3) to the expression variance for 8,523 genes (89%). Genotype and acclimation 171 contributed significantly to the expression variance for 3,111 and 3,716 genes (32.5% and 38.8%) 172 respectively ( Fig 2C, Table S4 ). Considering genes the expression variance of which is significantly 173 altered by either acclimation alone or interaction between acclimation and any other factor, 174 acclimation had an impact on the expression of 4,430 genes responsive to infection (46.2% of DEGs, 175 Fig S3, Table S5). 176 To document the relationship between transcriptional response to S. sclerotiorum inoculation and 177 acclimation, we built a gene co-expression hierarchical network with genes modulated by inoculation 178 both in the differential and ANOVA analyses. For this, we used normalized read counts to calculate 179 Spearman rank correlation coefficient for all pairwise gene comparisons across our 54 RNA-seq 180 samples. Highly co-expressed gene pairs were grouped into hierarchical gene communities using the 181 HiDeF algorithm (Zheng et al., 2021). 6,620 genes were included into communities of at least four 182 genes (Fig 2D, Data S1 ). Four major top-level communities (labeled α to δ in Fig 2D) encompassed 183 5,933 genes (89.6% of the network). In average, communities α and β included genes with expression 184 anticorrelated with resistance (putative susceptibility factors, Table S6 ), frequently acclimation 185 dependent and downregulated upon S. sclerotiorum inoculation and upon heat stress. By contrast, 186 genes from communities and γ and δ had expression correlated with resistance, frequently 187 acclimation-independent, up-regulated upon S. sclerotiorum inoculation and heat stress ( Fig 2D, Fig 188 S4). Accordingly, the median LFC upon inoculation was -1.56 in acclimation-dependent DEGs but 0.70 189 in acclimation independent DEGs, the median correlation between LFC and susceptibility phenotype 190 was 0.09 in acclimation-dependent DEGs but -0.01 in acclimation independent DEGs ( Fig 2E). Genes 191 downregulated by infection were 64.3% and 46.7% among genes acclimation-dependent and 192 independent respectively (1.37-fold enrichment). Reciprocally, genes upregulated by infection were 193 35.5% and 52.6% among genes acclimation-dependent and independent respectively (1.48-fold 194 depletion). Gene the expression of which is correlated (Pearson >0.5) with the susceptibility phenotype 195 were 12.3% and 19.2% among genes acclimation-dependent and independent respectively (1.56-fold 196 enrichment). We conclude that acclimation primarily alters the expression of genes down-regulated 197 by infection and genes associated with disease susceptibility. 198 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 6 199 Fig 2. Global gene expression profiling of A. thaliana plants inoculated by the fungal pathogen S. sclerotiorum following 200 temperate, continental and Mediterranean acclimation. (A) Number of differentially expressed genes (DEGs) upregulated 201 (yellow) and down-regulated (blue) 48 hours post inoculation by S. sclerotiorum in each of three plant genotypes and three 202 acclimation conditions. Box plots show the distribution of DEG number per genotype (columns) and acclimation (rows), first 203 and third quartiles (box), median (thick line), and the most dispersed values within 1.5 times the interquartile range (whiskers) 204 are shown. (B) Number of up- and down-regulated DEGs according to the number of differential assignments, out of 9 tested 205 conditions. (C) Distribution of acclimation-dependent DEGs identified by ANOVA according to the factors explaining gene 206 expression variance. (D) A hierarchical network of genes mis-regulated by S. sclerotiorum infection identified through 207 differential and variance analyses. Nodes represent gene communities sized according to the number of DEGs they contain, 208 fill color corresponding average correlation between gene expression and plant susceptibility, border color correspond to 209 average LFC upon inoculation. Four major communities are labelled and their number of genes indicated. (E) Distribution of 210 infection LFC and correlation between LFC and susceptibility phenotype for acclimation dependent and independent DEGs. 211 Violin plots show a gaussian kernel, median (dot) and standard deviation (dotted lines). 212 Mediterranean acclimation turns some pathogen-responsive genes into susceptibility factors 213 To get insights into the role of DEGs in A. thaliana QDR against S. sclerotiorum, we first analyzed Gene 214 Ontologies (GO) enriched in each of the four major top-level gene communities from our hierarchical 215 network, relative to the rest of the network ( Fig 3A, Table S7 ). Community α was enriched in 96 216 biological process (BP) and 9 molecular function (MF) GO, with ‘Starch metabolism’, ‘Photosynthesis’, 217 ‘Translation’, ‘Primary metabolism’, ‘Chlorophyll binding’, ‘Exopeptidase activity’ and ‘Constituent of 218 ribosome’ among the most enriched, reflecting a general downregulation of energetic functions of the 219 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 7 plant cell during infection. Community β was enriched in 18 BP and 4 MF GOs with ‘Regulation of gene 220 expression’, ‘Regulation of metabolic process’, ‘Regulation of developmental process’ and 221 ‘Transcription factor activity’ among the most enriched. Community γ was enriched in 68 BP and 22 222 MF GOs, with ‘Phytoalexin metabolic process’, ‘Response to chitin’, ‘Chorismate metabolic process’, 223 ‘Immune response’, ‘Glutathione binding’, ‘Oxidoreductase activity’, ‘Carbohydrate binding’ and ‘Ion 224 binding” among the most enriched, reflecting the probable involvement of genes from this community 225 in disease resistance. Finally, community δ was enriched in 40 BP and 10 MF GOs, with ‘Vesicle-226 mediated transport’, ‘Protein catabolic process’, ‘Response to osmotic stress’ and ‘Signal transduction’ 227 among the most enriched, consistent with a role in stress response. 228 To study the role of DEGs in disease resistance against S. sclerotiorum, we analyzed the phenotype of 229 14 mutant lines in the Col-0 background corresponding to 11 distinct genes, with a focus on genes from 230 community γ that were not previously associated with plant immunity ( Table S8). For comparison 231 purposes, we included mutants in one gene from community β ( AT1G12290), one from community δ 232 (AT5G64990) and three genes not differentially expressed in our RNA-seq experiment ( AT1G34190, 233 AT2G43790 and AT5G60600). The natural accessions Col-0, Rld-2 and Sha were used as references. 234 After temperate acclimation ( Fig 3B), four mutant lines were significantly more susceptible than the 235 Col-0 wild type, affecting genes AT5G06230, AT3G12910 and AT5G37840 from community γ. After 236 Mediterranean acclimation (Fig 3C, Table S9), all three natural accessions were more susceptible than 237 after temperate acclimation, consistent with our previous set of experiments (Fig 1B). Rld-2 was more 238 strongly affected by Mediterranean acclimation and became significantly more susceptible than Col-0 239 in these conditions. To our surprise, mpk6-1 was the only mutant significantly more susceptible than 240 Col-0 after Mediterranean acclimation. Nine mutants were more resistant than wild type after 241 Mediterranean acclimation, covering genes AT1G76600, AT1G07135, AT5G24600, AT5G06230, 242 AT3G12910 and AT5G37840 from community γ, AT5G64990 and AT5G06230 from community δ. While 243 natural accessions had their resistance phenotype reduced by ~37% in average after Mediterranean 244 compared to temperate acclimation, only four mutant lines showed >25% resistance reduction, 245 including three in genes not differentially expressed upon inoculation ( AT1G34190, AT2G43790 and 246 AT5G60600) and one from community G ( AT1G76600). By contrast, six mutants showed increased 247 resistance after Mediterranean compared to temperate acclimation. 248 Together these results confirm that community γ includes several genes contributing to resistance 249 against S. sclerotiorum after temperate acclimation. Mutations in several genes from community γ 250 render plants more resistant than wild type after Mediterranean acclimation, indicating that they act 251 as susceptibility factors in these conditions. Remarkably, AT5G06230, AT3G12910 and AT5G37840 252 would classify as resistance factors in temperate-acclimated plants but as susceptibility factors in 253 Mediterranean-acclimated plants. 254 255 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 8 , 256 Fig 3. Functional analysis of pathogen-induced genes. (A) Gene ontology (GO) enrichment in the four major gene 257 communities identified based on a co-expression network of genes mis-regulated by S. sclerotiorum infection. Communities 258 α, β, γ, δ are labelled on the hierarchical network shown with the same layout as in Fig2D. A selection of the most enriched 259 biological process (BP, black) and molecular function (MF, blue) GOs are labelled, with enrichment fold and adjusted p-value 260 relative to A. thaliana genome indicated. Genes analyzed through mutant phenotyping are labeled according to their position 261 in the network. (B, C) Disease resistance phenotype of natural accessions and mutant plants following temperate acclimation 262 (B) or Mediterranean acclimation (C) and inoculated by S. sclerotiorum . Comm. Major community of the co-expression 263 network to which the gene belongs. ND, gene not differentially expressed upon S. sclerotiorum inoculation (not part of the 264 co-expression network). Pie chart in (C) indicate mean % variation of disease resistance relative to infection following 265 temperate acclimation. Boxplots show first and third quartiles (box), median (thick line), and the most dispersed values within 266 1.5 times the interquartile range (whiskers). Colors of the data points indicate independent inoculation experiments. Leaves 267 from n=29 to 315 plants were tested for each genotype. Significance of the difference from Col-0 wild type was assessed by 268 a Student’s t test followed by Benjamini-Hochberg correction for multiple testing (*** p<0.01, ** p<0.05., * p<0.1). 269 270 Acclimation shifts the repertoire of NAC42-L target genes upon S. sclerotiorum inoculation 271 AT3G12910 encodes a member of the NAC family of transcription factors that includes several 272 regulators of pathogen and abiotic stress response (Nuruzzaman et al., 2013). Its closest homolog in 273 A. thaliana genome is NAC42/JUNGBRUNNEN1 (AT2G43000) (Ooka et al., 2003), we will thus refer to 274 AT3G12910 as NAC42-Like ( NAC42-L) hereafter. NAC42-L is strongly induced upon S. sclerotiorum 275 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 9 inoculation both in plants temperate- (LFC 7.8 p-adj. 3E-08 in Col-0) and Mediterranean-acclimated 276 (LFC 7.5 p-adj. 8E-22 in Col-0). Yet two mutant alleles of NAC42-L resulted in lower disease resistance 277 in temperate acclimated plants but enhanced disease resistance in Mediterranean acclimated plants 278 (Fig 3B). To study how acclimation alters the activity of NAC42-L at the molecular level, we analyzed 279 the expression of its target genes upon infection in temperate- and Mediterranean-acclimated plants. 280 For this, we first identified targets presumably regulated by NAC42-L by searching for NAC42-L DNA 281 binding motif determined by (O’Malley et al., 2016) in the promoter of A. thaliana genes. This 282 identified 2276 potential NAC-L binding sites in 1795 different gene promoters, with a maximum of 5 283 binding sites per promoter ( Table S10 ). Among NAC42-L targets, 394 genes were DEGs upon S. 284 sclerotiorum inoculation in temperate- or Mediterranean-acclimated Col-0 plants (Fig 4B). There were 285 227 NAC-L targets (57.6% of NAC42-L target DEGs) uniquely differential following growth under one of 286 the two climates, indicating a significant switch in the regulation of NAC42-L target genes upon 287 infection according to acclimation. The effect of acclimation on the regulation of NAC42-L target genes 288 upon S. sclerotiorum inoculation was clearly detectable in the three accessions we analyzed (Fig 4C). 289 To test whether the acclimation-mediated switch in NAC42-L targets was dependent on NAC42-L, we 290 measured by quantitative RT-PCR the expression of eight of these targets in two nac42-L mutant lines 291 (42-L1 and 42-L2) following temperate and Mediterranean acclimation (Fig 4D, Table S11, Fig S5). Six 292 of these genes had an expression significantly altered by the inactivation of NAC42-L, supporting their 293 position as targets of NAC42-L regulation. Yet for all of them, the impact of NAC42-L inactivation on 294 their expression was only detected after one particular acclimation regime. Indeed, AT1G10040, 295 AT3G09010, AT3G26200 and AT3G19615 were upregulated upon S. sclerotiorum inoculation following 296 temperate acclimation in wild-type plants but significantly less in 42-L1 and 42-L2 plants, while the 297 expression of these genes was similar in all three genotypes following Mediterranean acclimation. 298 Conversely, AT5G65510 showed a similar expression in wild type and nac42-L mutant lines upon 299 inoculation following temperate acclimation, but it was significantly mis-regulated in nac42-L mutants 300 following Mediterranean acclimation. Together, these results suggest that acclimation alters the 301 contribution of NAC42-L to quantitative disease resistance by switching the repertoire of genes 302 regulated by this transcription factor (Fig 4E). 303 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 10 304 Fig 4. Effect of temperate and Mediterranean acclimation on the regulation of gene expression by the transcription factor 305 NAC42-L. (A) Sequence logo of the promoter motif bound by NAC42-L according to DAP-seq data. (B) Distribution of genes 306 harboring NAC42-L motifs in their promoter between up- and down- regulated genes upon S. sclerotiorum inoculation in Col-307 0 plants temperate- and Mediterranean-acclimated. (C) Relative induction of the 394 genes differentially expressed upon 308 inoculation harboring NAC42-L motifs in their promoter in Col-0 and Sha accessions following temperate (Temp.) and 309 Mediterranean (Med.) acclimation. (D) Expression of six NAC42-L predicted targets in wild type (WT) and nac42-L mutants 310 (42-L1, 42-L2) in mock-treated and S. sclerotiorum-inoculated plants following temperate acclimation and S. sclerotiorum-311 inoculated plants following Mediterranean acclimation. Boxplots show expression independent measurements for 3-9 plants 312 (dots) with first and third quartiles (box), median (thick line), and the most dispersed values within 1.5 times the interquartile 313 range (whiskers). Letters indicate groups of significance determined by a Tuckey HSD test following one-way ANOVA. (E) 314 Schematic representation of the proposed mechanism through which acclimation switches NAC42-L from a positive regulator 315 of disease resistance (temperate acclimation) to a negative regulator (Mediterranean acclimation). S. sclerotiorum 316 inoculation triggers the expression of NAC42-L (brown arrow) and accumulation of NAC42-L protein (brown circle) which 317 regulates positively (red arrow) or negatively (blue blocked arrow) target genes. Upon temperate acclimation, NAC42-L 318 targets (green arrow) may positively contribute to disease resistance, while upon Mediterranean acclimation, NAC42-L “off-319 targets” (orange arrows) mostly promote susceptibility to pathogens. 320 321

322 Phenotypic plasticity, a component of acclimation, allows plant species to adjust to environmental 323 conditions, together with adaptation through natural selection or migration to follow conditions to 324 which they are adapted. Understanding the molecular mechanisms of acclimation is crucial for 325 predicting changes in species distributions, community composition and crop productivity under 326 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 11 climate change. In this work we show that A. thaliana Mediterranean acclimation is detrimental for 327 disease resistance to the fungus S. sclerotiorum and converts several genes that contribute positively 328 to quantitative immunity following temperate acclimation into susceptibility factors. Mediterranean 329 acclimation involves a shift in the repertoire of targets of the pathogen-induced transcription factor 330 NAC42-like that may impair the regulation of quantitative immune responses. 331 Experiments in controlled conditions have been instrumental in unraveling complex stressor 332 interactions through tightly controlled factorial experiments. These studies emphasized that combined 333 effects of various environmental stressors resulted in unique transcriptional changes distinct from 334 individual stress responses (Sewelam et al., 2014; Zandalinas and Mittler, 2022). These interactions 335 can be synergistic, where stressors amplify each other's negative effects, or antagonistic, where they 336 dampen each other's impacts (Zarattini et al., 2021). Research on plant-pathogen interactions under 337 abiotic constraints often relies on long-lasting stable temperature shifts, overlooking the complex 338 acclimation processes plants undergo in response to gradual climatic shifts (Aoun et al., 2017; Desaint 339 et al., 2021). Several studies investigated the effect of temperature acclimation by applying a stable 340 temperature shift over a few days prior to a second stress application. For instance, growth of A. 341 thaliana for 7 days at 4°C enhanced survival to freezing in a NPR1-dependent manner (Olate et al., 342 2018), and two-days growth at 30°C rendered plant more susceptible to the bacterial pathogen 343 Pseudomonas syringae pv. tomato DC3000 when inoculation is performed either at 23°C or 30°C (Huot 344 et al., 2017). Nevertheless, the impact of day-night temperature cycles on subsequent stress response 345 is rarely considered. We have chosen to approximate realistic climate change scenarios by simulating 346 30-year day and night average temperatures and photoperiods representing three climates of the 347 Köppen-Geiger classification (Peel et al., 2007). Since the 1980s, Mediterranean climates with dry 348 summer (Cs) have gradually replaced areas with temperate climate (Cf) (Cui et al., 2021). Predictions 349 suggest that the Mediterranean (Csa) climate may replace a portion of the continental (Df, Dw, Ds) 350 climates by the end of the century (Beck et al., 2018; Cui et al., 2021). Significant poleward shifts were 351 observed for temperate (C), continental (D), and polar (E) climates with averages of 35.4, 16.2, and 352 12.6 km.decennia -1 (0.32, 0.15, and 0.11° latitude.decennia -1 respectively), and are expected to 353 accelerate in the coming decades (Chan and Wu, 2015; Cui et al., 2021). The three selected climates 354 therefore cover a significant part of A. thaliana distribution area and reflect the poleward shift of 355 climate zones and associated changes in temperature and day length. Our work revealed a significant 356 loss of quantitative disease resistance upon Mediterranean acclimation in multiple A. thaliana 357 accessions, although the daily average temperature was only 0.7°C higher under Mediterranean 358 acclimation (average 15.67°C) than under temperate acclimation (average 14.96°C). Given the current 359 data, we cannot determine whether the observed phenotypic differences are attributable to daytime 360 temperatures, nighttime temperatures, the photoperiod, or a combination of these factors. 361 Nevertheless, our findings indicate that the typical April conditions of the Mediterranean climate zone, 362 expected to expand by the end of the century due to global warming, are detrimental to resistance 363 against Sclerotinia diseases. Combined with episodes of high humidity conducive to infection, global 364 warming may therefore increase the incidence of these plant diseases. 365 Our global transcriptome analyses indicated that the three A. thaliana accessions tended to show a 366 higher number of DEGs when acclimated in conditions close to the climate at their area of origin. Col-367 0, originating from temperate (Cfb, Fig. S1) climate area (Somssich, 2019), and Rld-2, originating from 368 continental (Dfb) climate area (Alonso-Blanco et al., 2016) has more upregulated and downregulated 369 genes when acclimated under temperate and continental conditions respectively. Sha, originating 370 from Mediterranean (Csa) climate (Alonso-Blanco et al., 2016) showed more down-regulated following 371 temperate acclimation but more upregulated genes following Mediterranean acclimation. Although 372 this did not reflect at the phenotype level, this transcriptome pattern suggests that A. thaliana 373 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 12 accessions adapted to their climate of origin to acclimate more efficiently, producing a stronger 374 immune response at the molecular level. It also suggests that mapping to the Col-0 reference genome 375 did not introduce major bias in gene expression quantification in other accessions. Adaptation to 376 environmental change involves variations in allele frequencies within a population's gene pool over 377 several generations, while acclimation occurs reversibly within an organism’s lifestyle. Genetic 378 variation is crucial for both plastic and adaptive potential (Fox et al., 2019). Reduced genetic variation 379 from positive selection or limited migration can lower phenotypic plasticity. Conversely, plastic traits 380 may become fixed or constitutively expressed through genetic assimilation (Wood et al., 2023). 381 High genetic variation in natural populations enhances their ability to withstand and adapt to new 382 biotic and abiotic environmental changes, including climate change (Van Kleunen and Fischer, 2005; 383 Nicotra et al., 2010). This genetic variation partly determines the capacity of plants to sense 384 environmental changes and generate plastic responses. For instance, cis-regulatory and epigenetic 385 variation at the FLOWERING LOCUS C floral repressor regulating vernalization can aid plant populations 386 in adapting to temperature fluctuations (Hepworth et al., 2020). Yet, the role of selection and whether 387 gene expression plasticity facilitates or hinders adaptation remains a matter of debate (Levis and 388 Pfennig, 2016). Comparative analysis of gene expression in forest and urban populations of Anolis 389 lizards showed that rapid parallel regulatory adaptation to urban heat islands primarily resulted from 390 selection for reduced and/or reversed heat-induced plasticity, which is maladaptive in urban thermal 391 conditions (Campbell-Staton et al., 2021). A meta-analysis of reciprocal transplant experiments 392 indicated that adaptation to new environments only leads to genes losing their expression plasticity 393 by genetic assimilation in rare cases (Chen and Zhang, 2024). In agreement, our results suggest that 394 adaptation to their climate of origin maintained high expression plasticity of immunity genes in A. 395 thaliana accessions. These insights will be valuable for assessing the adaptive potential of populations 396 in the face of ongoing global climate change. 397 We identified three genes the inactivation of which in the Col-0 background lead to increased pathogen 398 susceptibility following temperate acclimation but increased resistance following Mediterranean 399 acclimation. Mutation in six other genes resulted in increased resistance following Mediterranean 400 acclimation but no significant phenotype change following temperate acclimation. Conditionally 401 beneficial or neutral mutations, that are deleterious in some environments but beneficial or neutral in 402 others, have been reported in a wide range of organisms including plants (Elena and de Visser, 2003; 403 Anderson et al., 2013). Recombinant inbred lines of the Brassicaceae plant Boechera stricta of diverse 404 origin revealed that selection favored local alleles in contrasted environments, and 8.1% of the 405 assessed markers showed evidence for conditional neutrality for the probability of flowering 406 (Anderson et al., 2013). In the perennial grass Panicum hallii, an allele of the FLOWERING LOCUS T-like 407 9 locus from coastal ecotypes conferred a fitness advantage only in its local habitat but not at the 408 inland site (Weng et al., 2022). 409 Loss of function alleles contribute to species adaptation (Olson, 1999; Xu and Guo, 2020) and have 410 played an important role in crop domestication (Monroe et al., 2020). Naturally occurring loss of 411 function variants are relatively rare, with an average 57 per genome in A. thaliana (Xu et al., 2019) and 412 18 per genome in soybean (Torkamaneh et al., 2019) but they are found in 19% of soybean genes and 413 66% of A. thaliana genes. Conditionally neutral mutations are sufficient to drive patterns of local 414 adaptation in simulations (Mee and Yeaman, 2019) and can emerge as a compensation to deleterious 415 mutations (Steinberg and Ostermeier, 2024; Farkas et al., 2022). Simulations of long-term evolution in 416 changing environments produced complex gene regulatory networks with an increased rate of 417 beneficial mutations, while a majority of mutations remain neutral (Crombach and Hogeweg, 2008). 418 Patterns of local adaptation in A. thaliana (Fournier-Level et al., 2011a), the complexity of quantitative 419 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 13 immunity networks (Delplace et al., 2020) and our focus on inoculation up-regulated genes may 420 explain the high proportion of conditionally beneficial loss-of-function we have identified. This finding 421 suggests that targeted gene knockouts may be a promising strategy to improve climate resilience of 422 plant immunity. 423 We identified NAC42-L ( AT3G12910) as a resistance factor following temperate acclimation but a 424 susceptibility factor following Mediterranean acclimation. Its closest homolog, ANAC042/ 425 JUNGBRUNNEN1 (AT2G43000), was identified as a regulator of camalexin biosynthesis and positive 426 regulator of resistance against the fungus Alternaria brassicicola (Saga et al., 2012), longevity (Wu et 427 al., 2012) and tolerance to heat and drought (Ebrahimian-Motlagh et al., 2017; Shahnejat-Bushehri et 428 al., 2012). In addition, exposure to 90min at 37°C enhanced survival of JUB1 overexpressors to a 429 subsequent treatment at 45°C, compared with WT and jub1–1 knock-down seedlings (Shahnejat-430 Bushehri et al., 2012). Molecular changes induced in plants by heat and other environmental signals 431 persist longer than the signals themselves and modifies subsequent responses, phenomenon referred 432 to as somatic environment memory (SEM). In this work, pathogen inoculations were done in standard 433 conditions, indicating that some form of SEM of previous growth conditions had influenced plant 434 immunity. The molecular mechanisms by which SEM mediates the priming of plant-microbe 435 interactions remain largely unknown. Our results implicated a switch in the transcriptional targets of 436 NAC42-L in this process. The underlying molecular bases may include variation in trans, through 437 changes to the composition, stoichiometry and post-transcriptional regulation of protein complexes 438 including NAC42-L, or variation in cis affecting the conformation and accessibility of target gene 439 promoter regions. Recent studies have identified chromatin state modifications as crucial components 440 in the memory of repeated stress events in plants, particularly in response to heat, cold, and drought 441 priming (Balazadeh, 2022; Crisp et al., 2016; Liu et al., 2022). Future investigations will aim at 442 deciphering which molecular mechanisms mediate NAC-L target switch upon acclimation and what 443 controls the duration and breadth of this switch. 444 Our results highlight rewiring of quantitative immunity gene networks as a key process in acclimation, 445 with adverse consequence to disease resistance under warm Mediterranean-like climates. We show 446 that acclimation can reverse the contribution of genes upregulated by pathogen inoculation to the 447 disease resistance phenotype. We identified several mutations mitigating the negative impact of 448 Mediterranean acclimation on disease resistance, opening perspectives for the preservation of plant 449 immunity functions in a warming climate context. 450 451 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 14

452 Plant material and growth conditions 453 A. thaliana Natural accessions and mutant lines were obtained from the Nottingham Arabidopsis Stock 454 Center. We selected Col-0 (CS76778, 6909), Rld-2 (CS78349, 7457) and Shahdara (Sha, CS78397, 6962) 455 as three natural accessions of A. thaliana originating from areas with contrasted climate conditions. 456 Plants were grown in jiffy pots for 35 or 70 days in Percival E41-L3 and E41-L2PLT growth cabinets 457 equipped with ultra-sonic humidifier, far-red LED clusters, closed-loop light dimming, Intellus and 458 WeatherEZE controllers. We set day and night temperatures for each climate according to ERA5T 459 models based on 30-year average of hourly weather simulations for daily maximum and minimum 460 temperatures for the month of April at GPS coordinates 56.25°N, 34.19°E (Continental climate, origin 461 of Rld-2 accession); 38.35°N, 68.48°E (Mediterranean climate, origin of Sha accession) and 38.30°N, 462 92.30°O (Temperate climate) according to https:/www.meteoblue.com consulted on April 2017 ( Fig 463 S1). Day temperatures were 11°C, 20°C and 23°C and night temperatures were 1°C, 9°C and 7°C for 464 continental, temperate and Mediterranean climates respectively. Plants were grown in long day under 465 190 µmol/m²/s light, with photoperiod variation between climates to represent photoperiod variability 466 during April in the northern hemisphere, water was kept not limiting for the whole experiment at 80% 467 relative humidity. These growth conditions were classified into Continental, Mediterranean and 468 Temperate according to Köppen-Geiger classification (Cui et al., 2021) of climate at the corresponding 469 GPS coordinates. Inoculations were performed on detached leaves at a constant 23°C under constant 470 40 µmol/m²/s light and high humidity following the procedure described in (Barbacci et al., 2020). 471 Fungal strains and disease resistance phenotyping 472 0.5-cm-wide plugs of PDA agar medium containing S. sclerotiorum strain 1980, grown for 72 hours at 473 20°C on 14 cm Petri dishes, were placed on the adaxial surface of detached leaves. These leaves were 474 positioned in a Navautron system (Barbacci et al., 2020), and records were made using high-definition 475 (HD) cameras “3MP M12 HD 2.8-12mm 1/2.5 IR 1:1.4 CCTV Lens” every 10 minutes. For each genotype, 476 a minimum of 28 leaves were imaged from a minimum of two independent acclimation and inoculation 477 experiments. Kinetics of S. sclerotiorum disease lesions were analyzed using INFEST script v1.0 478 (https://github.com/A02l01/INFEST). Statistical analyses of disease phenotypes were conducted using 479 the Tukey test or the Student t test followed by Benjamini-Hochberg correction for multiple testing in 480 R 4.2.1. Disease susceptibility ( Fig. 1) corresponded to the slope of disease lesion growth over time. 481 Resistance (Fig. 3) corresponded to -log2 of the slope of disease lesion growth over time. 482 RNA collection and sequencing 483 Total RNA was extracted from a 3mm-wide ring of leaf tissue at the edge of ~1.5cm wide disease lesions 484 collected at 30 hours post inoculation (hpi) for temperate and mediterranean acclimation and 48 hpi 485 for continental acclimation. The samples were harvested with a scalpel on a cool glass slide and 486 immediately frozen in liquid nitrogen. Samples were ground with metal beads (2.5 mm) in a Retschmill 487 apparatus (24hertz for 2x1min). RNA was extracted using the RNAplus kit (Macherey Nagel) following 488 the manufacturer’s instructions. A Turbo DNAse treatment (Ambion) was applied to remove genomic 489 DNA. The quality and concentrations of RNAs preparations were assessed with an Agilent Bioanalyzer 490 using the Agilent RNA 6000 Nano kit. For the analysis of gene expression in natural accessions, libraries 491 synthesis and sequencing was outsourced to Fasteris SA (Plan-les-Ouates, Switzerland). Libraries were 492 sequenced as paired-end reads on an Illumina HiSeq 2500 instrument in High Output v4 mode with 493 2x125+8 cycles on 7 lanes of HiSeq Flow Cells v4 with the HiSeq SBS Kit v4. Basecalling was performed 494 with the HiSeq Control Software 2.2.58, RTA 1.18.64.0 and CASAVA-1.8.2. Reads QC was performed 495 using spiked-PhiX in-lane controls yielding Q30 error rate <0.4% for all lanes. Paired-end reads were 496 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 15 trimmed and mapped to the TAIR10.0 reference genome using the RNA-seq analysis tool of the CLC 497 Genomics Workbench 11.0.1 software (Qiagen). The following mapping parameters were used: 498 mismatch cost 2, insertion cost 3, deletion cost 3, length fraction 0.8, similarity fraction 0.8, both 499 strands mapping, and 10 hits maximum per read, with expression value given as total read count per 500 gene. 501 Differential expression and expression variance analyses 502 Differential gene expression analysis was performed with the DESeq2 Bioconductor package version 503 1.8.2 (Love et al., 2014) in R 3.4.0 in a pairwise manner using expression in uninfected plants as a 504

with ~replicates + inoculation as the design formula. Genes with baseMean 0 in all 505 differential comparisons and non-nuclear genes were discarded from further analyses. Genes with 506 |Log2 Fold Change|≥2 and Bonferroni-adjusted p-val<0.001 in DESeq2 Wald test were considered 507 significant for differential expression. For ANOVA, read counts were mean-normalized to homogenize 508 the total number of mapped reads per sample. The ANOVA was performed on each gene using the 509 dplyr package in R with ReadCount ~ Genotype * Infection * Climate as the model formula. P-values 510 associated with each factor were corrected for multiple testing using the Benjamini-Hochberg 511 procedure. 512 Gene network reconstruction and analyses 513 For gene network reconstruction, we focused on the 7,279 genes whose expression was pathogen 514 inoculation-dependent both in the differential analysis (|Log2 Fold Change|≥2 and Bonferroni-515 adjusted p-val<0.001) and in the ANOVA analysis (Benjamini-Hochberg corrected p-val <1E-7). Pairwise 516 Spearman rank correlation was calculated for the expression of these genes using the rcorr function 517 from the R package Hmisc, using the raw counts per gene from all 54 RNA-seq samples. The top 25 518 correlated genes (Spearman ρ>0.85) was extracted for each gene. The top 25 expression correlations 519 were used as edges for network reconstruction with weight ρ. Reconstruction of the hierarchical gene 520 cluster network was performed using the Community Detection 1.12.0 plugin in Cytoscape 3.10.0, 521 using the HiDeF algorithm with maximum resolution 45.0, consensus threshold 65, persistent 522 threshold 6 and the Louvain algorithm. Correlation with susceptibility was the Spearman rank 523 correlation coefficient between normalized read counts (averaged over three replicates) for each gene 524 and the slope of disease lesion growth. Average LFC was the mean log2 fold change over all nine 525 genotype-acclimation modalities tested. Values for gene clusters are the mean of values for all genes 526 in a cluster. Gene ontology enrichment were analyzed with the BinGO plugin in Cytoscape 3.10.0 using 527 a hypergeometric test with Benjamini and Hochberg false discovery rate correction, at significance 528 level 0.05 with A. thaliana whole annotation as a reference set. 529 Characterization of A. thaliana mutant lines 530 T-DNA insertion lines in AT1G76600 (SALK_052389), AT1G34190 (SALK_044777), AT1G07135 531 (SALK_133656), AT2G43790 (mpk6-1), AT5G24600 (SALK_201248C), AT1G12290 (SALK_125493), 532 AT5G60600 (SALK_059118), AT5G23160 (SALK_041095C), AT5G64990 (SALK_088173), AT5G06230 533 (SALK_008492C), AT3G12910 (SALK_016619C and SALK_078841) and AT5G37840 (SALK_002404) in 534 the Col-0 background were obtained from the Nottingham Arabidopsis Stock Centre. To identify 535 homozygous insertion lines, the lines were genotyped by PCR and, if needed, self-crossed and the 536 progeny genotyped by PCR. The disease resistance phenotype of mutant lines was analyzed as 537 previously described in a total of 23 independent inoculation experiments each including the Col-0 538 reference, with a minimum of 2 independent experiments for each mutant line. Primers used in all 539 experiments are shown in Table S12. 540 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 16 Identification of NAC42-L predicted target genes 541 The HMM model NAC_tnt.AT3G12910_col_a_m1 for the unique DAP-seq motif bound by AT3G12910 542 was obtained from the Plant Cistrome Database ( http://neomorph.salk.edu/dap_web/) (O’Malley et 543 al., 2016). A.thaliana genes harboring the corresponding motif were identified using FIMO v.5.5.5 544 (Grant et al., 2011) using the sequence 1Kbp upstream of the translation start site from the Araport11 545 annotation with 1E-4 as p-value threshold, both strands scanning and NRDB frequencies as a 546

model 547 Quantitative RT-PCR Analyses 548 RNA for qRT-PCR analysis was extracted from plants 24 hours post-inoculation with S. sclerotiorum as 549 for RNA-sequencing. For cDNA synthesis, 1 µg of RNA and 0.5 µL of Transcriptor reverse transcriptase 550 (Roche) were used in a 20 µL reaction volume according to the manufacturer’s protocol. The resulting 551 cDNA diluted 1:10 served as the template for quantitative RT-PCR. qRT-PCR reactions were carried out 552 with 5 pmol of specific oligonucleotides (Table S12), 2 µL of cDNA, and 3.5 µL of SYBR GREEN I in a total 553 volume of 7 µL. Amplification reactions were performed using a LightCycler 480 (Roche Diagnostics) 554 with the following protocol: 9 minutes at 95°C, followed by 45 cycles of 5 seconds at 95°C, 10 seconds 555 at 65°C, and 20 seconds at 72°C. Relative gene expression was calculated as the ratio of target gene 556 expression to the reference gene AT2G28390 and expressed as the difference between target and 557

crossing times (ΔCt). 558 559

560 This work was supported by the French Laboratory of Excellence project 'TULIP' (ANR-10-LABX-41; 561 ANR-11-IDEX-0002-02), a Starting grant from the European Research Council (ERC-StG-336808), 562 l’Agence Nationale pour la Recherche (ANR-19-CE20-15, ANR-21-CE20-10, ANR-21-CE20-30), and the 563 INRAE. M.D. benefited from a phD grant of the INRAE SPE division. We are grateful to the LIPME 564 Bioinformatics team for invaluable assistance with data storage and analysis. We thank Mehdi Khafif 565 for excellent technical help. ChatGPT-4 and Perplexity AI beta v0 were used to polish some sections of 566 the introduction and discussion of this manuscript. 567 568 DATA AVAILABILITY 569 Raw RNA-seq reads data and processed gene expression files generated in this work are available 570 under NCBI GEO accession number GSE272240. 571 572 AUTHOR CONTRIBUTIONS 573 A.B. and S.R. designed research; Natural accessions phenotyping: M.D., J.S.; RNA-seq sampling: J.S.; 574 RNA-seq data analysis: J.S., M.D., S.R.; Analysis of mutant lines genotype and phenotype: M.D; NAC42-575 L targets identification: M.D; Sampling for qRT-PCR: M.D.; Performed qRT-PCR: M.D., P.C-S, M.Z; 576 Analyzed qRT-PCR results: M.D, P.C-S, SR; Supervised and coordinated research: A.B, S.R; Wrote the 577 paper: M.D and S.R. 578 579 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 17

580 Alonso-Blanco, C. et al. (2016). 1,135 Genomes Reveal the Global Pattern of Polymorphism in 581 Arabidopsis thaliana. Cell 166: 481–491. 582 Anderson, J.T., Lee, C.-R., Rushworth, C.A., Colautti, R.I., and Mitchell-Olds, T. (2013). Genetic trade-583 offs and conditional neutrality contribute to local adaptation. Mol. Ecol. 22: 699–708. 584 Aoun, N., Tauleigne, L., Lonjon, F., Deslandes, L., Vailleau, F., Roux, F., and Berthomé, R. (2017). 585 Quantitative disease resistance under elevated temperature: Genetic basis of new resistance 586 mechanisms to ralstonia solanacearum. Front. Plant Sci. 8: 1–16. 587 Badet, T., Léger, O., Barascud, M., Voisin, D., Sadon, P., Vincent, R., Le Ru, A., Balagué, C., Roby, D., 588 and Raffaele, S. (2019). Expression polymorphism at the ARPC 4 locus links the actin 589 cytoskeleton with quantitative disease resistance to Sclerotinia sclerotiorum in Arabidopsis 590 thaliana. New Phytol. 222: 480–496. 591 Balazadeh, S. (2022). A ‘hot’ cocktail: The multiple layers of thermomemory in plants. Curr. Opin. Plant 592 Biol. 65: 102147. 593 Barbacci, A., Navaud, O., Mbengue, M., Barascud, M., Godiard, L., Khafif, M., Lacaze, A., and Raffaele, 594 S. (2020). Rapid identification of an Arabidopsis NLR gene as a candidate conferring susceptibility 595 to Sclerotinia sclerotiorum using time-resolved automated phenotyping. Plant J. 103: 903–917. 596 Bebber, D.P., Holmes, T., and Gurr, S.J. (2014). The global spread of crop pests and pathogens. Glob. 597 Ecol. Biogeogr. 23: 1398–1407. 598 Bebber, D.P., Ramotowski, M.A.T., and Gurr, S.J. (2013). Crop pests and pathogens move polewards 599 in a warming world. Nat. Clim. Chang. 3: 985–988. 600 Beck, H.E., Zimmermann, N.E., McVicar, T.R., Vergopolan, N., Berg, A., and Wood, E.F. (2018). Present 601 and future Köppen-Geiger climate classification maps at 1-km resolution. Sci. Data 5: 180214. 602 Bernard, F., Chelle, M., Fortineau, A., Riahi El Kamel, O., Pincebourde, S., Sache, I., and Suffert, F. 603 (2022). Daily fluctuations in leaf temperature modulate the development of a foliar pathogen. 604 Agric. For. Meteorol. 322: 109031. 605 Bohn, L., Huang, J., Weidig, S., Yang, Z., Heidersberger, C., Genty, B., Falter-Braun, P., Christmann, 606 A., and Grill, E. (2024). The temperature sensor TWA1 is required for thermotolerance in 607 Arabidopsis. Nature. 608 Brancalion, P.H.S., Oliveira, G.C.X., Zucchi, M.I., Novello, M., van Melis, J., Zocchi, S.S., Chazdon, R.L., 609 and Rodrigues, R.R. (2018). Phenotypic plasticity and local adaptation favor range expansion of 610 a Neotropical palm. Ecol. Evol. 8: 7462–7475. 611 Burghardt, L.T., Runcie, D.E., Wilczek, A.M., Cooper, M.D., Roe, J.L., Welch, S.M., and Schmitt, J. 612 (2016). Fluctuating, warm temperatures decrease the effect of a key floral repressor on flowering 613 time in Arabidopsis thaliana. New Phytol. 210: 564–576. 614 Campbell-Staton, S.C., Velotta, J.P., and Winchell, K.M. (2021). Selection on adaptive and maladaptive 615 gene expression plasticity during thermal adaptation to urban heat islands. Nat. Commun. 12: 616 6195. 617 Chaloner, T.M., Gurr, S.J., and Bebber, D.P. (2021). Plant pathogen infection risk tracks global crop 618 yields under climate change. Nat. Clim. Chang. 11: 710–715. 619 Chan, D. and Wu, Q. (2015). Significant anthropogenic-induced changes of climate classes since 1950. 620 Sci. Rep. 5: 13487. 621 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 18 Charng, Y.Y., Mitra, S., and Yu, S.J. (2023). Maintenance of abiotic stress memory in plants: Lessons 622 learned from heat acclimation. Plant Cell 35: 187–200. 623 Chen, P. and Zhang, J. (2024). Transcriptomic analysis reveals the rareness of genetic assimilation of 624 gene expression in environmental adaptations. Sci. Adv. 9: eadi3053. 625 Clauw, P., Kerdaffrec, E., Gunis, J., Reichardt-Gomez, I., Nizhynska, V., Koemeda, S., Jez, J., and 626 Nordborg, M. (2022). Locally adaptive temperature response of vegetative growth in Arabidopsis 627 thaliana. Elife 11: 1–17. 628 Cohen, S.D. (2023). Estimating the Climate Niche of Sclerotinia sclerotiorum Using Maximum Entropy 629 Modeling. J. Fungi 9. 630 Coolen, S. et al. (2016). Transcriptome dynamics of Arabidopsis during sequential biotic and abiotic 631 stresses. Plant J. 86: 249–67. 632 Corwin, J.A., Copeland, D., Feusier, J., Subedy, A., Eshbaugh, R., Palmer, C., Maloof, J., and 633 Kliebenstein, D.J. (2016). The Quantitative Basis of the Arabidopsis Innate Immune System to 634 Endemic Pathogens Depends on Pathogen Genetics. PLoS Genet. 12: 1–29. 635 Crisp, P.A., Ganguly, D., Eichten, S.R., Borevitz, J.O., and Pogson, B.J. (2016). Reconsidering plant 636 memory: Intersections between stress recovery, RNA turnover, and epigenetics. Sci. Adv. 2. 637 Crombach, A. and Hogeweg, P. (2008). Evolution of Evolvability in Gene Regulatory Networks. PLOS 638 Comput. Biol. 4: e1000112. 639 Cui, D., Liang, S., and Wang, D. (2021). Observed and projected changes in global climate zones based 640 on Köppen climate classification. WIREs Clim. Chang. 12: e701. 641 Delplace, F., Huard-Chauveau, C., Dubiella, U., Khafif, M., Alvarez, E., Langin, G., Roux, F., Peyraud, 642 R., and Roby, D. (2020). Robustness of plant quantitative disease resistance is provided by a 643 decentralized immune network. Proc. Natl. Acad. Sci. 117: 18099–18109. 644 Deng, Y., Bossdorf, O., and Scheepens, J.F. (2021). Transgenerational effects of temperature 645 fluctuations in Arabidopsis thaliana. AoB Plants 13: 1–11. 646 Derbyshire, M.C. and Raffaele, S. (2023). Till death do us pair: Co-evolution of plant–necrotroph 647 interactions. Curr. Opin. Plant Biol. 76: 102457. 648 Desaint, H., Aoun, N., Deslandes, L., Vailleau, F., Roux, F., and Berthomé, R. (2021). Fight hard or die 649 trying: when plants face pathogens under heat stress. New Phytol. 229: 712–734. 650 Ebrahimian-Motlagh, S., Ribone, P.A., Thirumalaikumar, V.P., Allu, A.D., Chan, R.L., Mueller-Roeber, 651 B., and Balazadeh, S. (2017). JUNGBRUNNEN1 Confers Drought Tolerance Downstream of the 652 HD-Zip I Transcription Factor AtHB13. Front. Plant Sci. 8: 2118. 653 Elena, S.F. and de Visser, J.A.G.M. (2003). Environmental stress and the effects of mutation. J. Biol. 2: 654 12. 655 Farkas, Z., Kovács, K., Sarkadi, Z., Kalapis, D., Fekete, G., Birtyik, F., Ayaydin, F., Molnár, C., Horváth, 656 P., Pál, C., and Papp, B. (2022). Gene loss and compensatory evolution promotes the emergence 657 of morphological novelties in budding yeast. Nat. Ecol. Evol. 6: 763–773. 658 Fournier-Level, A., Korte, A., Cooper, M.D., Nordborg, M., Schmitt, J., and Wilczek, A.M. (2011a). A 659 map of local adaptation in Arabidopsis thaliana. Science 334: 86–9. 660 Fournier-Level, A., Korte, A., Cooper, M.D., Nordborg, M., Schmitt, J., and Wilczek, A.M. (2011b). A 661 map of local adaptation in Arabidopsis thaliana. Science (80-. ). 334: 86–89. 662 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 19 Fox, R.J., Donelson, J.M., Schunter, C., Ravasi, T., and Gaitán-Espitia, J.D. (2019). Beyond buying time: 663 the role of plasticity in phenotypic adaptation to rapid environmental change. Philos. Trans. R. 664 Soc. B Biol. Sci. 374: 20180174. 665 Garcia-Molina, A., Kleine, T., Schneider, K., Mühlhaus, T., Lehmann, M., and Leister, D. (2020). 666 Translational Components Contribute to Acclimation Responses to High Light, Heat, and Cold in 667 Arabidopsis. iScience 23. 668 Garcia-Molina, A. and Pastor, V. (2024). Systemic analysis of metabolome reconfiguration in 669 Arabidopsis after abiotic stressors uncovers metabolites that modulate defense against 670 pathogens. Plant Commun. 5: 100645. 671 Grant, C.E., Bailey, T.L., and Noble, W.S. (2011). FIMO: scanning for occurrences of a given motif. 672 Bioinformatics 27: 1017–1018. 673 Hadj-Amor, K., Léger, O., Raffaele, S., Kovacic, M., Duclos, A., Carvalho-Silva, P., Garcia, F., and 674 Barbacci, A. (2024). “Out of sight, out of mind”. The antagonistic role of the transcriptional 675 memory associated with repeated acoustic stimuli on the priming of plant defense. bioRxiv: 676 2024.05.03.592392. 677 Hepworth, J. et al. (2020). Natural variation in autumn expression is the major adaptive determinant 678 distinguishing Arabidopsis FLC haplotypes. Elife 9: e57671. 679 Huot, B., Castroverde, C.D.M., Velásquez, A.C., Hubbard, E., Pulman, J.A., Yao, J., Childs, K.L., Tsuda, 680 K., Montgomery, B.L., and He, S.Y. (2017). Dual impact of elevated temperature on plant defence 681 and bacterial virulence in Arabidopsis. Nat. Commun. 8: 1–11. 682 Jallet, A.J., Le Rouzic, A., and Genissel, A. (2020). Evolution and Plasticity of the Transcriptome Under 683 Temperature Fluctuations in the Fungal Plant Pathogen Zymoseptoria tritici. Front. Microbiol. 11. 684 Kappel, C., Friedrich, T., Oberkofler, V., Jiang, L., Crawford, T., Lenhard, M., and Bäurle, I. (2023). 685 Genomic and epigenomic determinants of heat stress-induced transcriptional memory in 686 Arabidopsis. Genome Biol. 24: 1–23. 687 Katz, E., Li, J.J., Jaegle, B., Ashkenazy, H., Abrahams, S.R., Bagaza, C., Holden, S., Pires, C.J., Angelovici, 688 R., and Kliebenstein, D.J. (2021). Genetic variation, environment and demography intersect to 689 shape Arabidopsis defense metabolite variation across Europe. Elife 10: 1–25. 690 Kim, J.H. et al. (2022). Increasing the resilience of plant immunity to a warming climate. Nature 607: 691 339–344. 692 Kleine, T. et al. (2021). Acclimation in plants – the Green Hub consortium. Plant J. 106: 23–40. 693 Van Kleunen, M. and Fischer, M. (2005). Constraints on the evolution of adaptive phenotypic plasticity 694 in plants. New Phytol. 166: 49–60. 695 Lane, D., Denton-Giles, M., Derbyshire, M., and Kamphuis, L.G. (2019). Abiotic conditions governing 696 the myceliogenic germination of Sclerotinia sclerotiorum allowing the basal infection of Brassica 697 napus. Australas. Plant Pathol. 48: 85–91. 698 Léger, O., Garcia, F., Khafif, M., Carrere, S., Leblanc-Fournier, N., Duclos, A., Tournat, V., Badel, E., 699 Didelon, M., Le Ru, A., Raffaele, S., and Barbacci, A. (2022). Pathogen-derived mechanical cues 700 potentiate the spatio-temporal implementation of plant defense. BMC Biol. 20: 292. 701 Levis, N.A. and Pfennig, D.W. (2016). Evaluating ‘Plasticity-First’ Evolution in Nature: Key Criteria and 702 Empirical Approaches. Trends Ecol. Evol. 31: 563–574. 703 Liu, H., Able, A.J., and Able, J.A. (2022). Priming crops for the future: rewiring stress memory. Trends 704 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 20 Plant Sci. 27: 699–716. 705 Liu, Y., Dang, P., Liu, L., and He, C. (2019). Cold acclimation by the CBF–COR pathway in a changing 706 climate: Lessons from Arabidopsis thaliana. Plant Cell Rep. 38: 511–519. 707 Love, M.I., Huber, W., and Anders, S. (2014). Moderated estimation of fold change and dispersion for 708 RNA-seq data with DESeq2. Genome Biol. 15: 550. 709 Matsubara, S. (2018). Growing plants in fluctuating environments: why bother? J. Exp. Bot. 69: 4651–710 4654. 711 Mbengue, M., Navaud, O., Peyraud, R., Barascud, M., Badet, T., Vincent, R., Barbacci, A., and 712 Raffaele, S. (2016). Emerging Trends in Molecular Interactions between Plants and the Broad 713 Host Range Fungal Pathogens Botrytis cinerea and Sclerotinia sclerotiorum. Front. Plant Sci. 7: 714 422. 715 Mee, J.A. and Yeaman, S. (2019). Unpacking Conditional Neutrality: Genomic Signatures of Selection 716 on Conditionally Beneficial and Conditionally Deleterious Mutations. Am. Nat. 194: 529–540. 717 Mehrabi, Z., Pironon, S., Kantar, M., Ramankutty, N., and Rieseberg, L. (2019). Shifts in the abiotic 718 and biotic environment of cultivated sunflower under future climate change. OCL - Oilseeds fats, 719 Crop. Lipids 26. 720 Monroe, J.G., Arciniegas, J.P., Moreno, J.L., Sánchez, F., Sierra, S., Valdes, S., Torkamaneh, D., and 721 Chavarriaga, P. (2020). The lowest hanging fruit: Beneficial gene knockouts in past, present, and 722 future crop evolution. Curr. Plant Biol. 24: 100185. 723 Navaud, O., Barbacci, A., Taylor, A., Clarkson, J.P., and Raffaele, S. (2018). Shifts in diversification 724 rates and host jump frequencies shaped the diversity of host range among Sclerotiniaceae fungal 725 plant pathogens. Mol. Ecol. 726 Newman, R. and Noy, I. (2023). The global costs of extreme weather that are attributable to climate 727 change. Nat. Commun. 14. 728 Nicotra, A.B., Atkin, O.K., Bonser, S.P., Davidson, A.M., Finnegan, E.J., Mathesius, U., Poot, P., 729 Purugganan, M.D., Richards, C.L., Valladares, F., and van Kleunen, M. (2010). Plant phenotypic 730 plasticity in a changing climate. Trends Plant Sci. 15: 684–692. 731 Nishad, A. and Nandi, A.K. (2021). Recent advances in plant thermomemory. Plant Cell Rep. 40: 19–732 27. 733 Nuruzzaman, M., Sharoni, A.M., and Kikuchi, S. (2013). Roles of NAC transcription factors in the 734 regulation of biotic and abiotic stress responses in plants. Front. Microbiol. 4: 248. 735 O’Malley, R.C., Huang, S.C., Song, L., Lewsey, M.G., Bartlett, A., Nery, J.R., Galli, M., Gallavotti, A., 736 and Ecker, J.R. (2016). Cistrome and Epicistrome Features Shape the Regulatory DNA Landscape. 737 Cell 165: 1280–1292. 738 Olate, E., Jiménez-Gómez, J.M., Holuigue, L., and Salinas, J. (2018). NPR1 mediates a novel regulatory 739 pathway in cold acclimation by interacting with HSFA1 factors. Nat. Plants 4: 811–823. 740 Olson, M. V (1999). When Less Is More: Gene Loss as an Engine of Evolutionary Change. Am. J. Hum. 741 Genet. 64: 18–23. 742 Ooka, H. et al. (2003). Comprehensive Analysis of NAC Family Genes in Oryza sativa and Arabidopsis 743 thaliana. DNA Res. 10: 239–247. 744 Peel, M.C., Finlayson, B.L., and McMahon, T.A. (2007). Updated world map of the Köppen-Geiger 745 climate classification. Hydrol. Earth Syst. Sci. 11: 1633–1644. 746 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 21 Peltier, A.J., Bradley, C.A., Chilvers, M.I., Malvick, D.K., Mueller, D.S., Wise, K.A., and Esker, P.D. 747 (2012). Biology, Yield loss and Control of Sclerotinia Stem Rot of Soybean. J. Integr. Pest Manag. 748 3: 1–7. 749 Perchepied, L., Balagué, C., Riou, C., Claudel-Renard, C., Rivière, N., Grezes-Besset, B., and Roby, D. 750 (2010). Nitric oxide participates in the complex interplay of defense-related signaling pathways 751 controlling disease resistance to Sclerotinia sclerotiorum in Arabidopsis thaliana. Mol. Plant-752 Microbe Interact. 23: 846–860. 753 Roux, F., Voisin, D., Badet, T., Balagué, C., Barlet, X., Huard-Chauveau, C., Roby, D., and Raffaele, S. 754 (2014). Resistance to phytopathogens e tutti quanti: Placing plant quantitative disease resistance 755 on the map. Mol. Plant Pathol. 15: 427–432. 756 Sadhukhan, A., Prasad, S.S., Mitra, J., Siddiqui, N., Sahoo, L., Kobayashi, Y., and Koyama, H. (2022). 757 How do plants remember drought? Planta 256: 1–15. 758 Saga, H., Ogawa, T., Kai, K., Suzuki, H., Ogata, Y., Sakurai, N., Shibata, D., and Ohta, D. (2012). 759 Identification and Characterization of ANAC042, a Transcription Factor Family Gene Involved in 760 the Regulation of Camalexin Biosynthesis in Arabidopsis. Mol. Plant-Microbe Interact. 25: 684–761 696. 762 Sewelam, N., Oshima, Y., Mitsuda, N., and Ohme-Takagi, M. (2014). A step towards understanding 763 plant responses to multiple environmental stresses: a genome-wide study. Plant. Cell Environ. 764 37: 2024–2035. 765 Shahnejat-Bushehri, S., Mueller-Roeber, B., and Balazadeh, S. (2012). Arabidopsis NAC transcription 766 factor JUNGBRUNNEN1 affects thermomemory-associated genes and enhances heat stress 767 tolerance in primed and unprimed conditions. Plant Signal. Behav. 7: 1518–1521. 768 Shahoveisi, F., Riahi Manesh, M., and del Río Mendoza, L.E. (2022). Modeling risk of Sclerotinia 769 sclerotiorum-induced disease development on canola and dry bean using machine learning 770 algorithms. Sci. Rep. 12: 1–10. 771 Singh, B.K., Delgado-Baquerizo, M., Egidi, E., Guirado, E., Leach, J.E., Liu, H., and Trivedi, P. (2023). 772 Climate change impacts on plant pathogens, food security and paths forward. Nat. Rev. 773 Microbiol.: 1–17. 774 Sloat, L.L., Davis, S.J., Gerber, J.S., Moore, F.C., Ray, D.K., West, P.C., and Mueller, N.D. (2020). 775 Climate adaptation by crop migration. Nat. Commun. 11: 1–9. 776 Somssich, M. (2019). A short history of Arabidopsis thaliana (L.) Heynh. Columbia-0. PeerJ Prepr. 7: 777 e26931v5. 778 Steinberg, B. and Ostermeier, M. (2024). Environmental changes bridge evolutionary valleys. Sci. Adv. 779 2: e1500921. 780 Sucher, J., Mbengue, M., Dresen, A., Barascud, M., Didelon, M., Barbacci, A., and Raffaele, S. (2020). 781 Phylotranscriptomics of the pentapetalae reveals frequent regulatory variation in plant local 782 responses to the fungal pathogen sclerotinia sclerotiorum. Plant Cell 32: 1820–1844. 783 Torkamaneh, D., Laroche, J., Valliyodan, B., O’Donoughue, L., Cober, E., Rajcan, I., Abdelnoor, R.V., 784 Sreedasyam, A., Schmutz, J., Nguyen, H.T., and Belzile, F. (2019). Soybean Haplotype Map 785 (GmHapMap): A Universal Resource for Soybean Translational and Functional Genomics. bioRxiv: 786 534578. 787 Uloth, M.B., You, M.P., Cawthray, G., and Barbetti, M.J. (2015). Temperature adaptation in isolates 788 of Sclerotinia sclerotiorum affects their ability to infect Brassica carinata. Plant Pathol. 64: 1140–789 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 22 1148. 790 Valladares, F. et al. (2014). The effects of phenotypic plasticity and local adaptation on forecasts of 791 species range shifts under climate change. Ecol. Lett. 17: 1351–1364. 792 Weng, X., Haque, T., Zhang, L., Razzaque, S., Lovell, J.T., Palacio-Mejía, J.D., Duberney, P., Lloyd-793 Reilley, J., Bonnette, J., and Juenger, T.E. (2022). A Pleiotropic Flowering Time QTL Exhibits Gene-794 by-Environment Interaction for Fitness in a Perennial Grass. Mol. Biol. Evol. 39: msac203. 795 Wood, D.P., Holmberg, J.A., Osborne, O.G., Helmstetter, A.J., Dunning, L.T., Ellison, A.R., Smith, R.J., 796 Lighten, J., and Papadopulos, A.S.T. (2023). Genetic assimilation of ancestral plasticity during 797 parallel adaptation to zinc contamination in Silene uniflora. Nat. Ecol. Evol. 7: 414–423. 798 Wu, A. et al. (2012). JUNGBRUNNEN1, a reactive oxygen species-responsive NAC transcription factor, 799 regulates longevity in Arabidopsis. Plant Cell 24: 482–506. 800 Xu, Y.-C. and Guo, Y.-L. (2020). Less Is More, Natural Loss-of-Function Mutation Is a Strategy for 801 Adaptation. Plant Commun. 1: 100103. 802 Xu, Y.-C., Niu, X.-M., Li, X.-X., He, W., Chen, J.-F., Zou, Y.-P., Wu, Q., Zhang, Y.E., Busch, W., and Guo, 803 Y.-L. (2019). Adaptation and Phenotypic Diversification in Arabidopsis through Loss-of-Function 804 Mutations in Protein-Coding Genes. Plant Cell 31: 1012–1025. 805 Zandalinas, S.I. and Mittler, R. (2022). Plant responses to multifactorial stress combination. New 806 Phytol. 234: 1161–1167. 807 Zarattini, M., Farjad, M., Launay, A., Cannella, D., Soulié, M.-C., Bernacchia, G., and Fagard, M. 808 (2021). Every cloud has a silver lining: how abiotic stresses affect gene expression in plant-809 pathogen interactions. J. Exp. Bot. 72: 1020–1033. 810 Zheng, F., Zhang, S., Churas, C., Pratt, D., Bahar, I., and Ideker, T. (2021). HiDeF: identifying persistent 811 structures in multiscale ‘omics data. Genome Biol. 22. 812 Zhong, Z. et al. (2023). Reversed asymmetric warming of sub-diurnal temperature over land during 813 recent decades. Nat. Commun. 14: 7189. 814 Zuo, D.D., Ahammed, G.J., and Guo, D.L. (2023). Plant transcriptional memory and associated 815 mechanism of abiotic stress tolerance. Plant Physiol. Biochem. 201: 107917. 816 817 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 23 SUPPLEMENTARY MATERIAL 818 819 Supplementary Figure S1. Climate data at sites in the distribution range of A. thaliana and S. 820 sclerotiorum used for simulated climates in our experiments (A) and at the site of Col-0 accession 821 origin (B). Data come from an ERA5T model of 30-year average of hourly simulations collected from 822 meteoblue.com. Satellite views of the designated coordinates were obtained from Google Maps. Grey 823 bars show monthly precipitations in mm, red lines are mean daily maximum (plain) and hot days 824 maximum (dotted), blue lines are mean daily minimum (plain) and cold nights minimum (dotted). 825 Values for April are labelled. The corresponding Köppen-Geiger climate was obtained from climate-826 data.org and coded as follows: Csa, hot summer Mediterranean climate; Cfa, humid subtropical 827 climate; Cfb, Temperate oceanic climate or subtropical highland climate; Dfa, Hot-summer humid 828 continental climate. Alt., altitude; Temp., Temperature. 829 830 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 24 831 Supplementary Figure S2. Arabidopsis genes differentially expressed upon S. sclerotiorum 832 inoculation analyzed with relaxed thresholds. Analysis of genes differentially expressed upon 833 inoculation at |Log2 Fold Change|≥1.5, adjusted p-val<0.1, using non-inoculated plants as reference 834 in each of nine conditions (three climate priming, times three plant genotypes). (A) Identification of 835 13 370 differentially expressed genes (DEGs) upon inoculation. In conditions where they are 836 differential, 5 460 DEGs were upregulated only, 7 464 were down-regulated only, and 446 were either 837 up or down-regulated. We detected 1 887 DEGs upregulated in all nine conditions and 1 232 DEGs 838 down-regulated in all nine conditions, representing 10.9% and 9.2% of the expressed genes 839 respectively. These 3 119 genes are differentially expressed in a consistent manner regardless of plant 840 genotype and acclimation, they can therefore be regarded as a core transcriptome responsive to S. 841 sclerotiorum in A. thaliana. (B) Heatmap of Log2 fold change for 200 genes forming major functional 842 groups including DEGs always up and always down (numbers indicated between brackets). Major 843 functional groups in the core transcriptome included the PRR-associated BOTRYTIS-INDUCED KINASE 844 1 (BIK1), BONZAI1-ASSOCIATED PROTEIN (BAP) 1 and 2, members of the Calmodulin (CaM) and CaM-845 binding, the pathogen and abiotic stress response, cadmium tolerance, disordered region-containing 846 (PADRE), the jasmonate-zim-domain proteins (JAZ), and the NAC-domain transcription factor families, 847 with all core DEGs being upregulated by S. sclerotiorum inoculation. Ferric reduction oxidases (FRO), 848 Mitochondrial transcription termination factors (MTTF), Photosystems I and II, phototropin and 849 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 25 phototropic-responsive NPH3 genes showed all core DEGs downregulated. The pleiotropic drug 850 resistance (PDR), cysteine-rich receptor-like protein kinase (CRK), WRKY and GRAS transcription factor, 851 cytochrome P450, Leucine-rich repeat (LRR), receptor like protein (RLP) families included several core 852 genes responsive to S. sclerotiorum either consistently up- or down-regulated. (C) Distance tree and 853 correlation matrix showing the similarity in the 13 370 DEGs regulation across conditions. Tree based 854 on Manhattan distance between samples and Ward clustering, correlation values shown by bubbles 855 are Spearman rank correlations calculated using LFC for 13 370 DEGs in each condition. Transcriptomes 856 measured after priming under temperate acclimation clustered together and with the transcriptome 857 of Rld-2 primed under continental acclimation and that of Sha under Mediterranean acclimation. The 858 remaining four conditions formed a second cluster. There was no clear clustering based on genotype 859 or climates alone, suggesting a significant interaction between these factors. Cont, continental; Temp, 860 temperate; Med, Mediterranean. 861 862 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 26 863 Supplementary Figure S3. A. thaliana genes differentially expressed in pairwise acclimation 864 comparisons. (A) Differentially expressed genes in pairwise comparisons between acclimation regimes 865 for S. sclerotiorum-inoculated samples (|LFC|≥1.5, adjusted p-val<0.1). The circos tracks show number 866 of expressed genes (x 1,000) with connectors representing genes differentially expressed between two 867 acclimation regimes (down-regulated in blue, upregulated in yellow). Labels show the sum of up- and 868 down-regulated genes for each acclimation comparison. The number of DEGs ranged from 127 in Sha 869 when comparing temperate and Mediterranean acclimation to 2,453 in Col-0 when comparing 870 continental and Mediterranean acclimation, representing 0.71% to 13% of expressed genes. (B) 871 Overall, 6,653 genes were differential in at least one comparison (35.2% of expressed genes). Of those, 872 4,895 (73.6%) were differentially expressed in one or two acclimation comparisons only, consistent 873 with a specific effect of acclimation on A. thaliana transcriptome, and supporting a significant 874 interaction between genotype and acclimation. (C) The genes most sensitive to acclimation were four 875 genes differentially regulated in eight out of nine comparisons. These genes encoded plant defensins 876 PDF1.3 (AT2G26010) and PDF1.2 (AT5G44420), the MYB transcription factor LHY (LATE ELONGATED 877 HYPOCOTYL, AT1G01060) and the chloroplastic lipocalin gene CHL (AT3G47860). Histograms show 878 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 27 Log2 Fold change of expression for AT5G44420 and AT3G47860 in 3 acclimation comparisons for 3 879 genotypes. Error bars show estimated standard errors for the estimated coefficients on the log2 scale 880 from DESeq2. (D) Venn diagram showing the distribution of DEGs for all three accessions according to 881 the acclimation regimes compared. The full upward triangle indicates up-regulated genes, the empty 882 downward triangle indicates down-regulated genes. For each acclimation comparison the total 883 number of DEGs is indicated between parenthesis. The percentage of shared genes is given relative to 884 the total number of DEGs in acclimation comparisons. Cont., continental acclimation; Med., 885 Mediterranean acclimation; Temp., temperate acclimation. 886 887 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 28 888 Supplementary Figure S4. Properties of gene communities in a network of DEGs upon S. sclerotiorum 889 inoculation. (A) We mapped the average gene LFC upon heat treatment reported in a recent meta-890 analysis (Guo et al. 2021). Genes from communities γ and δ showed a trend for up-regulation upon 891 heat treatment (average LFC 0.26 and 0.17 respectively) while genes from communities α and β were 892 rather down-regulated (average LFC -0.52 and -0.45 respectively). (B) We mapped the percentage of 893 gene from each community considered acclimation dependent based on our ANOVA analysis. 894 Communities γ and δ had a low proportion of acclimation-dependent genes while community α had a 895 majority of acclimation-dependent genes. (C) Considering the clear phenotypic effect of 896 Mediterranean acclimation, which had the highest day temperature and highest daily thermal 897 amplitude, we mapped the average LFC variation between plants acclimated under temperate and 898 Mediterranean climates. Average LFC variation was >1.0 for communities α and γ but <-0.7 for 899 community δ. (D) To summarize these analyses, we calculated the correlation between properties of 900 the six largest gene communities. We observed a clear correlation between association with 901 susceptibility phenotype and LFC variation upon temperate and Mediterranean acclimation (0.97), and 902 anti-correlation with average LFC upon S. sclerotiorum inoculation (-0.81 and -0.86). This suggested 903 that in our experiments, differential gene expression mostly associated with a decrease in plant 904 susceptibility which is strongly altered upon Mediterranean acclimation. 905

Guo, M., Liu, X., Wang, J., Jiang, Y., Yu, J., & Gao, J. (2021). Transcriptome profiling 906 revealed heat stress-responsive genes in Arabidopsis through integrated bioinformatics analysis. 907 Journal of Plant Interactions, 17(1), 85–95. https://doi.org/10.1080/17429145.2021.2014580 908 909 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 29 910 Supplementary Figure S5. Analysis of gene expression by quantitative RT-PCR in wild type and 911 NAC42-L mutant lines. (A) Relationship between gene expression in Col-0 determined by RNA-912 sequencing (X axis) and quantitative RT-PCR (Y axis). Error bars show standard error of the mean from 913 3 to 11 independent replicates. Dots represent expression of AT1G10040, AT1G56130, AT3G09010, 914 AT3G12910, AT3G19210, AT3G19615, AT3G26200, AT4G39950, AT5G65510 in mock-treated and S. 915 sclerotiorum inoculated plants grown under temperate and Mediterranean acclimation. (B) Relative 916 expression of NAC42-L (AT3G12910) in Col-0, nac42-L1 and nac42-L2 mutant lines determined by 917 quantitative RT-PCR. Boxplots show expression independent measurements for 3-9 plants (dots) with 918 first and third quartiles (box), median (thick line), and the most dispersed values within 1.5 times the 919 interquartile range (whiskers). P-values were determined by a Student t test with Benjamini-Hochberg 920 correction for multiple testing. 921 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint 30 Supplementary tables online 922 Table Related to Description S1 Fig 1B Raw data for A. thaliana susceptibility phenotype in response to S. sclerotiorum infection as a function of acclimation and genotype S2 Fig 2A Normalized read counts for the 54 RNA-seq samples and identification of expressed and differentially expressed genes ('1' = YES; '0'=no) S3 Fig 2A, 2B Log2 fold change and p-values for genes differentially expressed (inoculated samples versus mock treated samples as a reference) S4 Fig 2C Analysis of variance to determine the contribution of genotype (Geno), inoculation (Inoc) and acclimation (Clim) to expression variance S5 Fig S3 Genes differentially expressed in inoculated samples based on pairwise acclimation comparisons S6 Fig 2D, S4 Content and representative properties of gene communities in the co-expression network of A. thaliana genes differentially expressed upon S. sclerotiorum inoculation S7 Fig 3A Gene ontologies enriched in major gene communities in the co-expression network of A. thaliana genes differentially expressed upon S. sclerotiorum inoculation S8 Fig 3B, 3C Raw data for A. thaliana mutant lines susceptibility phenotype in response to S. sclerotiorum inoculation S9 Fig 3B, 3C Summary statistics for disease susceptibility of natural accessions and mutant lines inoculated by S. sclerotiorum following temperate and Mediterran acclimation S10 Fig 4B List of target sequences and genes harboring the DAP-seq motif bound by AT3G12910 identified by FIMO S11 Fig 4D Raw quantitative RT-PCR data for NAC42-L target genes in wild type and NAC42-L1 (SALK_016619C) and L2 (SALK_078841) mutant lines S12 Fig3, 4 List of oligonucleotide primers used in this work 923 924 Supplementary Data online 925 Data S1. Cytoscape session file containing the network of A. thaliana genes differentially expressed 926 upon S. sclerotiorum inoculation, associated metadata and hierarchical clustering network. 927 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted August 22, 2024. ; https://doi.org/10.1101/2024.08.22.609129doi: bioRxiv preprint