Enhancement of Arabidopsis growth by Enterobacter sp. SA187 under elevated CO2 is dependent on ethylene signalling activation and primary metabolism reprogramming

13 As atmospheric CO2 levels continue to increase, optimizing the CO2 fertilization effect 14 which often falls short of its potential due to the physiological and metabolic limitations of 15 plants becomes crucial. This study investigates the role of Enterobacter sp. SA187 (SA187), a 16 plant growth -promoting bacterium, in enhancing growth and development of Arabidopsis 17 thaliana under elevated atmospheric CO2 (eCO2) conditions. While SA187 inoculation did not 18 have major effect s under ambient CO 2, it was found to significantly enhance root and shoot 19 biomass, and to increase N- and reduce C-contents under eCO2. Moreover, transcriptomics and 20 metabolomics suggested that SA187 modulated phytohormonal homeostasis, with activation of 21 the salicylic acid, jasmonic acid and ethylene signal ling pathways, and increased primary 22 metabolism including the TCA cycle, N and carbohydrate metabolisms. Finally, the growth-23 promoting effects of SA187 were shown to be mediated through ethylene-dependent pathways, 24 as evidenced with the ethylene-insensitive mutant ein2-1 which did not show similar benefits 25 in plant fresh weight and altered gene expression. Th is beneficial plant-microbe interaction 26 under eCO2 in a non-leguminous plant highlights a novel aspect of microbial influence on plant 27 physiology in the context of climate change. These insights underscore the potential of utilizing 28 SA187 to enhance plant performance and adaptability in future high CO 2 environments, 29 providing a sustainable approach to agricultural productivity as global CO2 levels increase. 30

Elevated CO 2 (eCO2), Plant growth -promoting bacteria (PGPB) , 31 Enterobacter sp. SA187 , Arabidopsis thaliana , Ethylene signal ling, Transcriptomic 32 reprogramming, Carbon and nitrogen metabolism 33 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint I. Introduction 34 The hallmark of the Anthropocene is the rise in atmospheric CO 2 levels, which have 35 increased from approximately 280 ppm in the 17th century to over 430 ppm today (NOAA, 36 2025). Projections suggest that without significant mitigation efforts, these levels could reach 37 between 800 to 1000 ppm by the end of the century (Breecker et al., 2010; Meinshausen et al., 38 2011). Moreover, t he accumulation of greenhouse gases is driving climate change, altering 39 global weather patterns, leading to unpredictable seasons and shifts in temperature and 40 precipitation (Abbass et al., 2022; Mikhaylov et al., 2020). 41 Elevated atmospheric CO 2 concentration (eCO 2) has a direct positive effect on plant 42 growth, development and productivity, by enhancing photosynthetic CO 2 assimilation and 43 lowering wasteful photorespiration. However, this eCO2 fertilization effect is often lower than 44 predicted and it can depend on plant species and environmental factors (Rho et al. , 2020; 45 Ainsworth and Long, 2021). The lower -than-expected increase is the consequence of plant 46 photosynthetic acclimation during long -term eCO 2 exposures, which is attributed to various 47 physiological and metabolic factors (Long et al., 2004; Ainsworth and Long, 2005; Leakey et 48 al, 2009). Plants growing in an eCO2 atmosphere also exhibit a reduction in plant tissue mineral 49 nutrient concentrations that adversely affect crop quality (Gojon et al. , 2022). An average 50 decrease in N-content of 10-15 % is often observed in eCO2 grown plants although this is highly 51 variable (Taub and Wang, 2008). These e CO2 acclimations are associated with a complex 52 reprogramming of gene expression (Moore et al., 1999; Cheng et al., 1998), leading to lower 53 amounts of ribulose -1,5-bisphosphate carboxylase/oxygenase (Rubisco) protein, and a 54 reduction in nitrate uptake and assimilation (Rubio-Asensio et al. , 2017). Photosynthetic 55 acclimation to eCO2 is associated with sink limitations, leading to an accumulation of non-56 structural carbohydrates in photosynthetic tissues (Taub and Wang, 2008; Aranjuelo et al. , 57 2011; Rogers and Ainsworth, 2006 ) and hexokinase-dependent sugar signal ling (Cheng et al 58 1998; Moore et al 1999; Peyo et al 2000; Long et al 2004) that negatively impact, among others, 59 photosynthetic gene expression ( Krapp et al., 1993; Moore et al., 1999) and thus photosynthetic 60 activity (Pérez et al., 2011; Aranjuelo et al., 2013). Improved sink strength has been associated 61 to a reduction in eCO2 acclimation as reported in tobacco (Ruiz-Vera et al 2017), rice (Zhu et 62 al 2014) and barley (Torralbo et al 2019). 63 Several factors have been associated with the reduction of N content under eCO2. Much 64 of the decrease can be accounted for by the reduction in Rubisco protein content, a finding 65 supported by Free Air Concentration Enrichment ( FACE) studies (Ainsworth & Long, 2005). 66 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint It is also believed that eCO2 leads to the dilution of N in plant tissues due to an increased and 67 faster growth (Taub et al., 2008; Taub & Wang, 2008; Gifford et al. 2000, Gojon et al., 2023). 68 A l owering of leaf transpiration from eCO2-related changes in both stomatal density and 69 aperture size can reduce N-uptake and alter N-partitioning (McDonald et al. 2002; Long et al., 70 2004; Ainsworth and Rogers, 2007) . It has also been proposed that nitrate uptake and 71 subsequent shoot nitrate assimilation are both negatively impacted by eCO2-associated 72 inhibition of photorespiration (Rachmilevitch et al. , 2004). Therefore, it appears that 73 maintaining a balance between photosynthetic CO2 assimilation and N uptake and assimilation 74 is crucial to sustain the eCO2 fertilization effect and this is reflected in the plant C/N ratio (Ruiz-75 Vera et al., 2017, Kant et al., 2012). 76 A strategy to enhance plant resilience to reduce the negative effects of eCO2 77 acclimations could be the exploitation of plant beneficial microbes. Plant growth-promoting 78 bacteria (PGPB) have received particular attention due to their diversity, abundance, and ease 79 of manipulation (Glick, 2012 ). PGPB have been shown to improve plant growth and 80 development under normal and various abiotic stress conditions, such as drought, salinity, and 81 heat, by modulating plant hormonal balance including auxin s, ethylene, cytokinins and 82 gibberellins (Persello-Cartieaux et al., 2003; Vessey, 2003; Hardoin et al., 2008), improving 83 root architecture, and enhancing mineral acquisition by N-fixation, phosphate and zinc 84 solubilisation, and iron sequestration (Compant et al. , 2010). PGPB can also produce 85 antimicrobial agents an d induce systemic resistance against plant pathogens (Glick, 2012; 86 Pieterse et al., 2008). 87 In the context of e CO2, it has been shown that the symbiotic relationship between 88 leguminous plant species and N2-fixing rhizobia located in root nodules appear to be less 89 affected by eCO2 acclimations when compared to many non-leguminous plants (Cotrufo et al. 90 1998; Jablonski et al. 2002; Taub et al. 2008). This is believed to be due to rhizobia consuming 91 large quantities of plant photosynthates in exchange for fixed N, thus providing a strong carbon-92 sink and reducing N-limitations (Irigoyen et al., 2014; Singer et al., 2020). However, there is 93 no report of PGPB maintaining an eCO2.fertilization effect in non-leguminous C3 plants. 94 A well-studied PGPB shown to improve plant resilience to climate change-associated 95 abiotic stresses such as salt or heat is Enterobacter sp. SA187 (hereafter named SA187) (de 96 Zélicourt et al., 2018; Shekhawat et al., 2021). This endophytic bacterium was isolated from 97 root nodules of Indigofera argentea, a leguminous plant found in the Saudi Arabia n desert 98 (Andres Barrao, et al 2017). In Arabidopsis thaliana it was shown to colonize the surface and 99 inner tissues of both roots and shoots (de Zélicourt et al., 2018). The functional analysis of the 100 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint SA187 genome indicated genes involved in nutrient uptake/exchange, chemotaxis, plant 101 colonization, and oxidative stress. Moreover, bacterial metabolic pathways were identified that 102 could potentially contribute to plant growth promotion (Andres Barrao, et al 2017). 103 SA187 improved the abiotic stress tolerance of in vitro grown Arabidopsis thaliana and 104 field-grown Medicago sativa (alfalfa) and Triticum durum (wheat) (de Zélicourt et al., 2018; 105 Shekhawat et al. , 2021 ). This was in part due to the maintenance of primary metabolism 106 including photosynthesis (de Zélicourt et al., 2018). Ethylene sensing was found also to play 107 an important role since Arabidop sis mutants impaired in ethylene perception no longer 108 exhibited a beneficial response to SA187 while this was maintained in ethylene synthesis 109 mutants (de Zélicourt 2018, Shekhawat et al., 2021). 110 This study aims to explore whether SA187 can enhance plant growth under eCO 2 111 conditions. Based on transcriptomic and metabolomic analyses, initial findings indicate that 112 SA187 stimulates growth under eCO 2 through the activation of several phytohormonal 113 signalling pathways while modulating plant N and carbohydrate metabolisms including the 114 TCA cycle. This research highlights how plant-microbial interactions can be harnessed to 115 support C3 plant acclimations to rapidly changing global climates, offering new perspectives 116 to help sustainable agricultural practices. 117 II. Materials and methods 118 A. Bacterial and plant materials 119 Enterobacter sp. SA187 was originally isolated from root nodules of Indigofera 120 argentea in Jizan, Saudi Arabia (Andrés -Barrao et al., 2017). Arabidopsis thaliana ecotype 121 Columbia-0 ( Col-0, wild -type) and the ethylene -insensitive mutant ein2-1 (Guzman et al. , 122 1990) were cultivated under controlled greenhouse conditions (16 h photoperiod, 18/20°C 123 night/day, 50 ± 10% relative humidity) for seed propagation. 124 B. Plant growth and bacterial colonization 125 Seed sterilization and bacterial inoculation were performed as described previously 126 (Saad et al., 2018). Briefly, sterilized Arabidopsis seeds were sown on half-strength Murashige 127 and Skoog (½ MS) medium ( Duchefa basal salts only, 0.5 g/L MES, pH 5.8, 9 g/L agar) 128 containing SA187 (+SA187, concentration = 2.105 cells/mL) or not (mock) and stratified for 24 129 h in the dark at 4°C. These plates were then placed vertically in growth chambers for seed 130 germination (16 h photoperiod, 18 /20°C night/day, 50±10% relative humidity ), either at 450 131 ppm CO2 for aCO2 conditions or 1000 ppm CO2 for e CO2 conditions, resulting in 4 132 experimental groups: (i) Mock_aCO2 (ii) SA187_ aCO2, (iii) Mock_eCO2, (iv) SA187_eCO2. 133 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint At 5 days post-germination (dpg), seedlings were transferred to fresh ½ MS agar plates 134 and grown vertically. Primary root length (PRL) was measured using ImageJ software after 135 scanning of the plates 12 dpg. Lateral root density (LRD) was evaluated as detectable number 136 of lateral roots under a stereo microscope divided by the PRL at 12 dpg. Fresh weight (FW) of 137 shoots and roots was measured 1 5 dpg of seedlings. Dry weight (DW) was measured after 138 drying shoot and roots for 2 days at 70˚C. Bacterial colonization was quantified as described 139 previously (Saad et al., 2018). Samples were immediately flash-frozen in liquid-N2 and stored 140 at –80°C for subsequent analyses. 141 C. Transcriptomics 142 1. RNA-sequencing and data analyses 143 Shoot and root samples from 15 dpg plants from all experimental groups at a 1.03 144 developmental growth stage were collected for RNAseq experiments to obtain 3 biological 145 replicates (Boyes, 2001). Each replicate was composed of either rosettes or roots from 4 to 6 146 plants per condition. Total RNA was extracted using Nucleospin RNAplus kit (Macherey 147 Nagel), according to the supplier’s instructions and further purified using the RNA Clean & 148 Concentrator Kits (Zymo Research®, California, USA). RNA -seq libraries were constructed 149 using the TruSeq Stranded mRNA library prep kit (Illumina®, California, USA) according to 150 the supplier’s instructions. Libraries were sequenced in single-end (SE) mode with 75 bases for 151 each read on a NextSeq500 to generate between 16 and 45 millions of reads per sample. 152 Adapter sequences and bases with a Q -Score below 20 were trimmed out from reads 153 using Trimmomatic (v0.36, Bolger et al. 2014) and reads shorter than 30 bases after trimming 154 were discarded. Reads corresponding to rRNA sequences were removed using sortMeRNA 155 (v2.1, Kopylova E. et al. 2012) against the silva -bac-16s-id90, silva-bac-23s-id98, silva-euk-156 18s-id95 and silva-euk-28s-id98 databases. 157 Filtered reads were then mapped and counted using STAR (v2.7.3a, Dobin et al. 2013) 158 with the following parameters --alignIntronMin 5 --alignIntronMax 60000 --159 outSAMprimaryFlag AllBestScore --outFilterMultimapScoreRange 0 --160 outFilterMultimapNmax 20 on the Arabidopsis thaliana genome (ARAPORT version 11) and 161 its associated GTF annotation file. At least 97% of the reads were associated to annotated genes. 162 Statistical analyses were carried out separately on shoots and roots with R v3.6.2 (R 163 Core Team, 2020) using the Bioconductor package edgeR (v 3.28.0, Robinson et al., 2010; 164 McCarthy et al. , 2012). For both analyses, low counts genes were filtered using the 165 “filterByExpr” function of the R package edgeR with a minimum count threshold equal to 15. 166 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Raw counts were normalized using the trimmed mean of M values (TMM) method. Differential 167 analyses were based on a negative binomial generalized linear model. For each analysis, the 168 log2 of the average normalized gene expression is an additive function of a CO2 condition factor 169 (2 modalities), a treatment factor (2 modalities), a replicate factor and an interaction between 170 the CO2 condition factor and the treatment factor. In each analysis, by using a likelihood ratio 171 test, the difference between the two treatments at each CO 2 condition, the difference between 172 the two CO 2 conditions given a treatment and the interaction effect defined as the difference 173 between the two treatments at eCO 2 minus the difference between the same two treatments at 174 ambient CO2 were evaluated. The distribution of the raw p-values were checked following the 175 quality criterion described by Rigaill et al., 2018 and a gene was declared differentially 176 expressed (DEG) if its adjusted p-value was lower than 0.05. Hierarchical clustering was done 177 using MeV software and GO -term enrichment analysis was performed with g:Profiler 178 (Raudvere et al., 2019). 179 180 2. qRT-PCR analyses 181 Extracted mRNA (see above) was reverse-transcribed using the Promega ImProm-II™ Reverse 182 Transcription System with oligo -dT primers, following the manufacturer protocol. The 183 resulting cDNA was used for qRT -PCR. Reactions were performed and carried out with a 184 LightCycler® 480 SYBR Green I Master mix (Roche) and the following thermal cycling 185 conditions: A pre-incubation at 95°C for 10 min; 40 cycles of amplification (95°C for 10 s, 60 186 °C for 10 s, and 72 °C for 10 s); and a dissociation step (melting curve) to validate the PCR 187 products. Gene expression was analyzed using the ΔΔCt method (Rao et al., 2013), normalized 188 against two constitutive reference genes ( Actin and YSL8). Primer sequences are provided in 189 Supplementary Material (Sup Table 1). 190 191 D. Metabolomics 192 1. Extraction 193 Freeze-ground samples (50 mg FW) were resuspended in water/acetonitrile/isopropanol (2:3:3) 194 with 4 µg/mL ribitol (internal standard), shaken (4°C, 10 min), centrifuged (16000 g, 15 min), 195 and dried (30°C, 4 h) in a Speed-Vac and stored at -80°C. 196 2. GC-MS analyses 197 All steps were carried out as described previously (Fiehn, 2006; Fiehn et al., 2008). Dry aliquot 198 of 150 µl of the Extraction solution were taken and dr ied for a second time in a Speed -Vac 199 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint evaporator for 2 h at 30°C at 14000 rpm before adding 10 µL of 20 mg.mL-1 methoxyamine in 200 pyridine to the samples. The first step of derivatization was performed for 90 min at 30°C under 201 continuous shaking in an Eppendorf thermomixer. Then 90 µL N -methyl-N-trimethylsilyl- 202 trifluoroacetamide (MSTFA) (Regis Technologies, Morton Grove, IL, USA) were added and 203 the reaction was continued for 30 min at 37°C. After cooling, all samples were transferred to 204 an Agilent vial for injection. 205 One microliter of each sample was injected into an Agilent 7890B gas chromatograph 206 coupled to an Agilent 5977A mass spectrometer. The column was an Rxi-5SilMS from Restek 207 (30 m with 10 m Integra -Guard column). An injection in split mode with a ratio of 1:30 was 208 systematically performed for saturated compound quantification. Oven temperature ramp was 209 60°C for 1 min then 10°C min-1 to 325°C for 10 min. Helium constant flow was 1.1 mL.min-1. 210 Temperatures were the following: injector: 250°C, transfer line: 290°C, source: 230°C and 211 quadrupole: 150°C. The quadrupole mass spectrometer was switched on after a 5.90 min 212 solvent delay time, scanning from 50 to 600 m/z. Absolute retention times were locked to the 213 internal standard Ribitol using the RTL system provided by Agilent’s Masshunter software. 214 The Agilent Fiehn GC/MS Metabolomics RTL Library (version June 2008) was employed for 215 metabolite identifications. Peak areas were determined using the Masshunter Quantitative 216 Analysis (Agilent Technologies, Santa Clara, CA, USA) in splitless and split 30 modes. 217 Resulting areas were compiled into one single MS Excel file for comparisons. Peak areas were 218 normalized to Ribitol and DW. A total of 114 metabolites were identified and expressed in 219 arbitrary units (semi-quantitative determination). Pathway enrichment analysis was done using 220 MetaboAnalyst 6.0 (Pang et al., 2024). 221 3. Total C and N contents 222 Lyophilized shoot and root samples (1 mg) were combusted in a Pyrocube Elemental 223 Analyzer (Elementar, France), and N2/CO2 levels were quantified using a thermal conductivity 224 detector (TCD) against standards (ammonium sulfate 21.2% N, benzoic acid 68.85% C, 225 glutamic acid 9.51% N / 40.82% C and glutamine 19.17% N / 41.09% C). Elemental C and N 226 contents are given in % (mass fraction). 227 III. Results 228 A. SA187 increases plant growth and development under eCO2 229 To evaluate the impact of SA187 on the growth and development of Arabidopsis 230 thaliana, plants were cultivated under aCO2 (450 ppm) or eCO2 (1000 ppm) conditions, with 231 or without SA187 inoculation (+SA187 or mock), following previously established protocols 232 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint (de Zélicourt et al., 2018; Saad et al., 2018). Growth parameters, including FW, DW, PRL, and 233 LRD were assessed at 10 dpg (Figure 1). 234 In uninoculated plants , as expected, eCO 2 significantly increased plant biomass, with 235 total FW and DW increasing by 59% and 71%, respectively, compared to plants grown under 236 aCO2 (Figure 1A) . This growth enhancement affected equally root s and shoot s with FW 237 increases of 61% and 51% respectively (Figure 1A). These results align with past studies on 238 CO2 fertilization in C3 plants, which report enhanced biomass accumulation under eCO 2 239 (Ainsworth & Long, 2005, Ainsworth and Long, 2020). 240 Under aCO 2 conditions, SA187 did not significantly impact overall plant biomass 241 (Figure 1 A and D), except for a 39% increase in root FW and a 23% increase in PRL (Figure 242 1 C and E), indicating an effect on root development (Figure 1C). However, under eCO2, SA187 243 markedly improved the total FW (+51%), shoot FW (+43%), root FW (+97%), total DW 244 (+26%) compared to mock plants grown under eCO2 conditions (Figure 1, A-D). These results 245 demonstrated that SA187 could significantly enhance the growth and development of 246 Arabidopsis thaliana under eCO2 conditions. In addition, SA187 increased PRL by a similar 247 magnitude under both aCO 2 (+23.2%) and eCO 2 (+16.9%), suggesting that SA187 promotes 248 primary root elongation independently of atmospheric CO2 levels. In contrast, SA187 did not 249 influence LRD under either CO2 condition, implying that its beneficial effect is specific to 250 primary root growth rather than lateral root formation (Figure 1E-F). 251 As expected, the well-documented fertilization effect of eCO 2 was observed . 252 Interestingly, while SA187 did not exhibit any significant beneficial effect on plant biomass 253 under aCO2 conditions, with the exception of root FW , it noticeably increased various growth 254 parameters under eCO2. The significant increase in plant FW under eCO2 conditions, which 255 was augmented further in the presence of SA187 , suggests a synergistic interaction between 256 SA187 and eCO 2 in promoting biomass accumulation . Such a synergy aligns with previous 257 reports demonstrating that SA187 enhances plant biomass and stress tolerance under various 258 abiotic conditions, including salinity and heat stress (de Zélicourt et al. , 2018). Moreover, 259 SA187 increased PRL under both aCO2 and eCO2 conditions, but did not influence LRD. In the 260 literature, SA187 was found to increase LRD under salt stress conditions, indicating that the 261 effect of SA187 on root architecture may be context-dependent (de Zélicourt et al., 2018). This 262 suggests that SA187 can modulate different regulatory pathways according to environmental 263 context. 264 Overall, these results highlight SA187’s ability to enhance further plant growth under 265 eCO2 when compared to the non-inoculated plants perhaps by reducing acclimation processes 266 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint that limit the eCO 2 fertilization effect . The observed synergy between SA187 and eCO 2 267 revealed the bacterium's capability to modulate plant physiological processes to optimize root 268 and shoot growth, reinforcing its value as a biostimulant for enhancing plant development in 269 the face of escalating atmospheric CO2 levels. 270 271 B. eCO2 does not benefit SA187 proliferation and plant colonization 272 To investigate whether the enhanced plant growth effects in the presence of SA187 273 under eCO2 were associated with changes in bacterial proliferation, SA187 was cultured in 274 liquid medium under both aCO2 and eCO2 conditions. Optical density measurements taken over 275 10 h showed no significant differences in bacterial proliferation between the two conditions 276 (Figure 2A), indicating that eCO2 does not directly influence SA187 proliferation in vitro. 277 Further, bacterial colonization of Arabidopsis was quantified by measuring colony -278 forming units (CFU) per mg of plant tissue FW under both CO 2 conditions. Consistent with 279 prior studies (de Zélicourt et al. , 2018), under aCO 2, SA187 was found predominantly 280 colonizing the roots rather than the shoots, with respective densities of 10^6.1 CFU/mg FW and 281 10^4.5 CFU/mg FW. Similar colonization densities were observed under eCO 2, demonstrating 282 that eCO2 did not affect SA187 abundance in/on plant tissues (Figure 2B). 283 These results confirm that the plant growth-promoting effects of SA187 in eCO 2 284 conditions were not associated with an increased bacterial proliferation or colonization under 285 eCO2 when compared to aCO2. 286 287 C. SA187 inoculation alters phytohormone signaling pathway and primary 288 metabolism under eCO2 289 To elucidate the molecular basis of the impact of SA187 on plant growth and 290 development under eCO2, RNA sequencing (RNA-seq) was performed on Arabidopsis shoots 291 and roots. The analysis revealed differential expression of 4048 genes in shoots and 1256 genes 292 in roots compared to control (mock) plants under eCO 2 (Table S1). To gain a comprehensive 293 overview, transcriptomics data were organized using hierarchical clustering (Fig ure 3 and 294 Figure S1) and analyzed for Gene Ontology (GO) enrichment. 295 1. Shoot Transcriptome 296 Hierarchical clustering grouped shoot DEGs into 11 clusters (Figure 3 and Sup Table 297 2). Clusters 2, 3, 5 and 7 appeared to be the most interesting, as they comprised DEGs regulated 298 by SA187 under eCO 2 (Figure 3) . Overall, SA187 inoculation seemed to modify stress 299 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint phytohormone homeostasis as GO terms associated with response to salicylic acid (SA), 300 jasmonic acid (JA), and ethylene were induced in cluster s 2, 5 and 7. Indeed, cluster 2 (772 301 genes) was comprised of genes that seem ed to be unaffected by eCO 2 alone but exhibit ed 302 significant induction in response to SA187 under eCO2 conditions. This cluster was enriched in 303 genes responsive to stress phytohormones SA, JA, and ethylene. Genes in Cluster 7 (580 genes) 304 were similarly unresponsive to eCO2 alone but robustly induced upon SA187 inoculation under 305 eCO2. Notably, in addition to SA responses, this cluster was also enriched in genes associated 306 with N-metabolism. Genes in Cluster 5 (756 genes) exhibited a synergistic effect, being 307 upregulated by eCO 2 and further induced by SA187. Cluster 5 genes participate in primary 308 metabolism, including the TCA cycle and C-metabolism, alongside defense responses and 309 ethylene signalling (Figure 3B). In contrast to the previous clusters, genes in cluster 3 (136 310 genes) were highly induced by eCO 2 alone but repressed by SA187 under eCO 2, particularly 311 those involved in carbohydrate metabolism (including starch and sucrose pathways). 312 2. Root Transcriptome 313 In root tissues, five clusters were identified (Figure S1 and Sup Table 3). Cluster 5 was 314 the largest cluster identified from the root RNA-seq data and contained 468 genes, that were 315 highly induced by eCO2 and down-regulated by SA187 under both aCO2 and eCO2 conditions. 316 This cluster was enriched in genes related to carbohydrate metabolic process es, but also in 317 flavonoid biosynthesis and sulfur compound related processes (Figure S1). Conversely, cluster 318 2 included genes consistently upregulated by SA187, and independent of CO2 level. The genes 319 in this cluster were enriched in defence responses and hypoxia, which is a typical SA187 320 response. 321 Taken together, the observed differential responses highlight ed the role of SA187 in 322 regulating shoot and root transcriptomes under eCO 2. They indicated that the beneficial effect 323 of SA187 was possibly mediated by the activation of stress phytohormonal signalling pathways. 324 This was confirmed by RT -qPCR analyses of the ethylene marker ERF104, the SA marker 325 WRKY30, and the stress and defense markers MYB51 and CRK4 (Figure 3 C-F). This induction 326 was shown previously and ethylene was considered as the major signalling pathway to mediate 327 SA187-dependent plant beneficial effect s (de Zélicourt 2018, Shekhawat 2021). The 328 transcriptomic analyses also revealed possible modifications of C (carbohydrates and starch, in 329 cluster 3 and 5 ) and N metabolisms (cluster 7) that might help explain the improved plant 330 growth under eCO2 when inoculated with SA187. 331 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint D. SA187 Modulates TCA Related Metabolites and C/N Content Under eCO2 332 To complement the transcriptomic analyses, GC -MS-based metabolic profiling was 333 carried out on extracts from Arabidopsis seedlings grown in the same 4 conditions in order to 334 determine the metabolic changes occurring in plants when inoculated with SA187 in air and in 335 eCO2. When compared to mock plants, out of the 114 metabolites identified and quantified, 20 336 metabolites were found to be differentially accumulated metabolites (DAM) in shoot s in 337 response to SA187 inoculation (Figure 4A and Sup Table 4), of which 9 were more abundant: 338 beta-alanine, sinapic acid, tryptophan, trehalose, arbutin, nicotinic acid, oxoglutaric acid , 339 pyruvic acid and asparagine. These metabolites are associated with diverse metabolic pathways 340 including the TCA cycle (pyruvate and oxoglutarate), and amino acid metabolism (beta-alanine, 341 asparagine and tryptophan) as well as defense responses (e.g. sinapic acid, arbutin). Regarding 342 the reduced metabolites in response to SA187 inoculation, 11 of them were considered as less 343 abundant, some being also associated to the TCA cycle such as citric acid, or amino acid with 344 proline, or sugar with proline. This modulation of TCA related metabolites is partially in 345 agreement with the transcriptome results, for instance the GO term TCA cycle was found to be 346 significantly enriched in cluster 5, representing genes upregulated by SA187. In roots, 18 of 20 347 DAMs were significantly reduced including TCA cycle intermediates (succinic acid, fumaric 348 acid) and amino acids such as beta-alanine, proline and glycine (Figure S2 and Sup Table 5). 349 Since the transcriptomic studies suggested that both C and N metabolisms could be 350 deregulated in SA187 inoculated plants, elemental analyses were performed to quantify total C 351 and N contents. Interestingly, no changes were observed for plant C, N contents and C/N ratio 352 in response to SA187 inoculat ion in aCO 2 conditions (Figure 4 B-D). However, under eCO 2 353 conditions whereas the C content was reduced upon SA187 inoculation , N content increased, 354 thus leading to a significant decrease in plant C/N ratio to a value similar to that observed in 355 aCO2. 356 Therefore, the metabolite profiling confirmed that SA187 inoculation modifies plant 357 primary metabolisms including the TCA cycle and amino acid metabolism in agreement with 358 the conclusions obtained from the transcriptomic analyses. 359 360 E. SA187-induced Growth Promotion Under eCO2 Requires Ethylene Signalling 361 The transcriptomic analyses indicated a possible role for several phytohormones, among 362 them ethylene, in the SA187 -induced growth enhancement observed under eCO 2 (Figure 3, 363 cluster 2 a nd 5). To test whether ethylene signa lling was involved , the SA187 response was 364 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint compared between the ethylene-insensitive mutant ein2-1 and wild-type Col-0 plants under 365 eCO2 conditions. 366 As expected, SA187 significantly enhanced the FW of wild-type seedling under eCO2 367 (Figure 5A), consistent with previous observations (Figure 1). However, while an increase of 368 PRL was still observed for ein2-1 after SA187 inoculation (Sup Figure 3 ), seedling FW 369 significantly decreased in SA187 -inoculated ein2-1 plants when compared to mock -treated 370 ein2-1 plants (-24% compared to mock conditions) . This indicated that in the absence of a 371 functional ethylene signa lling pathway, SA187 shifted from promoting to inhibiting plant 372 growth, and underscored that ethylene signalling was indispensable for the growth-promoting 373 action of SA187 under eCO2, acting as a pivotal regulatory switch determining the bacterium's 374 influence on plant growth. This aligned with existing literature emphasizing ethylene's key role 375 in mediating plant responses to SA187 interactions and environmental stresses (de Zélicourt et 376 al., 2018; Shekhawat et al., 2021). 377 Given the transcriptomic insights into N et C metabolisms and ethylene signal ling as 378 key processes influenced by SA187 under eCO2, several genes representative of the GO terms 379 significantly enriched in the transcriptomic datasets were selected to determine whether their 380 regulation upon SA187 inoculation and/or eCO2 was still occurring in the absence of ethylene 381 signalling: the ethylene marker gene ethylene response factor 104 (ERF104, Cluster 7), nitrate 382 transporter 2.5 (NRT2.5, Cluster 7), starch branching enzyme 2.1 (SBE2.1, cluster 3), fructose-383 1,6-bisphosphate aldolase 6 (FBA6, Cluster 5 ), wound responsive 3/nitrate transporter 3.1 384 (WR3/NRT3.1, Cluster 2). As expected ERF104, but also NRT2.5, and WR3/NRT3.1 expression 385 were no longer induced in response to SA187 in ein2-1 when compared to wild type Col -0 386 (Figure 5B, D, F). Moreover, it appeared that FBA6 and SBE2.1, associated to carbohydrate 387 metabolism were unresponsive to both SA187 and to eCO2 conditions, as their expression were 388 not induced (or even repressed ) when compared to aCO 2 conditions even in non -inoculated 389 plants. 390 IV. Discussion 391 Based on different models, the level of atmospheric CO 2 is estimated to rise to 700- 392 1000 ppm by the next century if no action is taken. Although eCO 2 enhances photosynthetic 393 CO2 assimilation and lower photorespiration rates, it has been shown that many C 3 plants 394 acclimate to long term eCO2, leading to a lower than predicted yield increase. This study aimed 395 to determine whether inoculation with SA187 could increase plant growth and development 396 under eCO₂ and to explore the underlying signalling and metabolic pathways. 397 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint A. SA187 inoculation increases plant growth under eCO2 398 While SA187 inoculation had almost no effect on plant growth and development in 399 aCO2 conditions, its significant effect on Arabidopsis thaliana under eCO 2 suggested a 400 synergistic interaction between bacterial inoculation and eCO2 levels. The observed increase in 401 plant biomass under eCO 2 conditions aligns with previous studies documenting the CO 2 402 fertilization effect of C3 plants. Historically, this is primarily attributed to enhanced 403 photosynthetic rates and a reduction of photorespiratory losses, together increasing net CO 2 404 assimilation (Ainsworth & Long, 2005; Ainsworth and Long, 2020) . Interestingly, SA187 led 405 to an additional growth enhancement of both roots and shoots under eCO 2 conditions when 406 compared to mock -treated plants, suggesting that SA187 could modulate plant processes to 407 optimize growth in high CO 2 environments. Previous observations of a bacterial beneficial 408 effect on plant growth under eCO 2 conditions mainly concerned the symbiotic interaction 409 between legume plants and N2-fixing rhizobacteria (Palit et al. 2020). That said, there are a few 410 examples where a PGPR led to increased plant growth under eCO2. A good example concerns 411 Rahnella sp. WP5 , isolated from poplar, that maintained the photosynthetic activity of rice 412 under eCO2 (Rho et al., 2019). 413 It should be noted that the growth-promoting effects of SA187 were independent of 414 bacterial proliferation (Figure 2), suggesting that the observed phenotypic enhancements were 415 due to intrinsic modifications within the plant in an eCO2 context, altering both metabolic and 416 physiological processes. 417 Surprisingly, in our conditions, SA187 increased PRL, with no effect on LRD 418 independently of atmospheric CO2 level. This feature contradicts previous findings indicating 419 that SA187 did not affect PRL and only increased LRD under salt stress conditions (de Zélicourt 420 et al., 2018). Although growth conditions were similar ( culture medium, long-day conditions 421 and temperature), differences could be due to site-specific growing conditions. 422 B. SA187 Reprograms Plant Transcriptome and Modifies C & N metabolism 423 To obtain more insights into the molecular mechanisms affected by SA187 inoculation, 424 transcriptomics and metabolic profiling were performed on separated roots and shoots of mock 425 and SA187 inoculated plants under aCO 2 and eCO2. Firstly, transcriptomic analysis revealed 426 significant reprogramming of gene expression in both shoots and roots of Arabidopsis plants 427 inoculated with SA187 under eCO 2 conditions. GO term enrichment analyses identified 428 significant enrichments for stress phytohormone signalling including genes responsive to SA, 429 JA, and ethylene in shoots (Figure 3B). This is in agreement with previous reports showing that 430 SA187 inoculation triggers such signalling pathways under normal conditions, but also in 431 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint response to salt stress (de Zélicourt et al., 2018; Rolli et al., 2021). Induction of biotic stress 432 responsive genes with no penalty on plant growth suggests that SA187 may enhance plant 433 resistance to pathogen s under eCO 2, as SA and JA are well -known for their roles in plant 434 defence responses, while ethylene is involved in various stress responses and developmental 435 processes (Thilakarathne et al., 2025). This may be of a major importance in future climate 436 change scenarios that are expected to facilitate plant disease outbreaks (Lahlali et al., 2024). 437 In addition to the induction of stress related phytohormones, it appeared that SA187 was 438 also altering plant primary metabolism s in both shoot s and root s. Bacterial colonization 439 appeared to increase the expression of shoot TCA and carbohydrate metabolism related genes 440 (Figure 3A, cluster 5) as exemplified by the induction of the glycolytic/gluconeogenesis gene 441 FBA6 encoding a fructose-1,6-bisphosphate aldolase (Figure 5). This was also supported by the 442 accumulation of the final product of glycolysis and TCA precursor pyruvic acid, but also 443 oxoglutaric acid (Figure 4A). On the other hand, starch metabolism gene expression appeared 444 to be reduced, as observed in Shoot cluster 3, and SBE2.1 gene expression analyses. Such 445 observations suggest that inoculated plants invest more photosynthetically assimilated C into 446 energy metabolism instead of C-reserves. Indeed, the TCA cycle is a central metabolic pathway 447 that plays a crucial role in energy production and the biosynthesis of various organic acids 448 associated with amino acid metabolism . Indeed, a stimulated TCA cycle activity could be 449 related to the enhanced growth of the inoculated plants as overexpression of several TCA 450 related enzymes have been shown to increase plant development and performance (Zhand & 451 Fernie, 2023). 452 Additionally, N-metabolism also appeared to be stimulated in SA187 inoculated plants 453 under eCO 2 conditions as shown by the significant enrichment of genes associated with N-454 metabolism (cluster 7) and the induced expression of the high affinity nitrate transporter NRT2.5 455 gene. It is possible that such changes in gene expression gave rise to an enhanced N-assimilation 456 and metabolism that led to the observed increase in N-content and the altered C/N ratio under 457 eCO2 conditions upon SA187 inoculation. This is particularly important as eCO2 is believed to 458 lead to a N-dilution effect in plant tissues due to increased growth (Taub & Wang, 2008). By 459 improving N-metabolism, SA187 may help maintain a balanced C/N ratio, which is an essential 460 parameter to maintain optimal plant growth and development ( Kant et al., 2012). This further 461 supports the idea that SA187 could improve N-use-efficiency, a critical factor for sustaining 462 the CO2 fertilization effect (Ruiz-Vera et al., 2017; Kant et al., 2012). Indeed, N is a key limiting 463 nutrient for plant growth, and its efficient assimilation and metabolism would be essential for 464 maximizing the benefits of eCO2 (Gojon et al., 2022; Rubio-Ascension and Bloom, 2017). 465 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Concerning the root system, a lower number of genes were deregulated in the tested 466 conditions resulting in a reduced number of clusters regrouping genes influenced by SA187 467 inoculation. One cluster (cluster 2) regrouped the genes induced by SA187 independently of 468 CO2 levels, and their functions were associated with, among others, plant defence processes. 469 This was not surprising as SA187 produces Microbe-Associated Molecular Patterns (MAMPs) 470 that are recognized by the plant, which in return will activate defence mechanisms. This has 471 been documented before, as the SA187 genome contains fliC genes encoding flagella proteins 472 that have N-terminus sequences 60 to 80% similar to flg22 peptides, that will be recognized by 473 the flg22 receptor FLS2 (Rolli et al., 2021). On the other hand, cluster 5 represented genes that 474 were induced by eCO 2 in mock plants compared to aCO 2 mock plants and that SA187 475 inoculation appeared to alleviate. It regrouped genes associated to carbohydrate and sulphur 476 compound metabolic processes and this is coherent with previously obtained results showing 477 that SA187 inoculation reduced the expression level of sulphur related genes when grown under 478 salt stress (Andres Barrao et al. 2021). The SA187-related reduction of carbohydrate metabolic 479 processes deduced from transcriptomics appeared to be mirrored in SA187 inoculated root s 480 including the reduced levels of TCA related intermediates such as succinic acid and fumaric 481 acid. This and the reduced content in total C per mg DW might reflect an additional SA187 C-482 sink associated with the roots thus increasing sink strength from the shoot to the root . This 483 scenario might lead to a reduced induction of starch related genes in the shoot of SA187 plants 484 under eCO 2. Indeed, both parameters have been shown to be important to explain plant 485 acclimation to eCO2 (Jauregui et al., 2018; Ruiz-Vera et al., 2017, Krapp et al., 1999). 486 C. Role of Ethylene Signalling 487 As genes involved in ethylene response were found to be significantly enriched in 488 clusters 2 and 5 of the shoot transcriptomics (Figure 3) , the requirement of this signalling 489 pathway for SA187 -induced growth promotion under eCO 2 was tested using the ethylene -490 insensitive mutant ein2-1. The absence of growth enhancement in SA187 inoculated ein2-1 491 underscores the major role of ethylene in mediating the beneficial effects of SA187. The lack 492 of induction of SA187 -responsive genes ( ERF104, NRT2.5, WR3/NRT3.1) in ein2-1 further 493 supports the involvement of ethylene signalling in SA187 -mediated growth promotion. 494 Interestingly, the ethylene signalling pathway dependence of the increased expression of 495 NRT2.5 and NRT3.1 further illustrates the complex interconnection between ethylene and N-496 response (Ma et al., 2022) and it is consistent with previous studies that have emphasised the 497 pivotal role of ethylene in plant responses to microbial interactions and environmental stresses 498 (de Zélicourt et al., 2018). Transcript analyses by qRT-PCR also showed the repression of genes 499 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint associated with C and starch metabolisms (FBA6, SBE2.1) in ein2-1 in response to eCO2 thus 500 suggesting that ethylene signalling may also regulate primary metabolic pathways in response 501 to eCO2 independently of SA187 inoculation. Our observations appear to agree with a previous 502 report sho wing the importance of ethylene in plant adaptation to eCO 2, indeed the ein2-5 503 insensitive to ethylene mutant lacked the eCO2 fertilization effect and this correlated with a lack 504 of activation of the starch associated genes (de Smet et al., 2020). 505 V. Conclusions 506 This study supports SA187 as a potential biostimulant for improving crop performance 507 under anticipated future atmospheric eCO2 levels. Interestingly the growth -promoting 508 phenotypes of SA187 were independent of bacterial proliferation, as colonization levels did not 509 vary significantly between aCO2 and eCO2 conditions. This suggested that the observed 510 enhancements were due to intrinsic modifications within the plant in an eCO2 context, altering 511 both metabolic and physiological processes. Transcriptomic, metabolic and physiological 512 studies highlighted an extensive reprogramming of gene expression and altered metabolic 513 processes in SA187-inoculated plants in eCO 2 conditions. Major components included stress 514 phytohormones like ethylene, and primary metabolisms such as the TCA cycle, amino acid and 515 carbohydrate metabolisms. Such reprogramming appeared essential for the beneficial plant 516 growth and development observed in an eCO₂ atmosphere, that may also enhance resilience to 517 other environmental stresses . The contrasting responses observed in the ein2-1 ethylene-518 insensitive mutant compared to wild -type plants emphasised the significance of ethylene 519 signalling in mediating many of the observed SA187 effects. 520 Further work to better understand the molecular mechanisms by which SA187 521 influences plant responses under eCO 2 would be useful . This could involve genetic, 522 transcriptomic, and metabolomic studies using specific ethylene and N-uptake and signalling 523 mutants. Metabolic flux analyses could offer deeper insights into how plant metabolism is 524 rewired and C and N are dynamically allocated within SA187-inoculated plants under eCO 2. 525 Such studies would be paramount to help decipher the complex interactions between beneficial 526 microbes and plants in adapting to climate change and pave the way to develop bio-based 527 solutions to mitigate the negative impact of climate change on food security. Integrating 528 microbial treatments with genetic improvements could provide a holistic approach to 529 sustainable agriculture, promoting resilience and productivity in crops as global CO 2 levels 530 continue to rise. 531 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Author Contributions 532 A.I., M.H., and A.d.Z. designed the experiments, analyzed the data, and interpreted the 533 results. A.I. performed all experiments except RNA-seq and metabolomics. C.M. conducted the 534 metabolomics and elemental analyses and contributed to data interpretation. S.P. and C.P.-L.R. 535 performed RNA-seq experiments and carried out initial analyses. J.B. genotyped the plant lines 536 and provided technical support. A.d.Z. supervised the project and secured funding. A.I. wrote 537 the first draft of the manuscript and A.d.Z. prepared the figures. A.d.Z. and M.H. revised the 538 manuscript. 539 Author Approvals 540 All authors have seen and approved the final version of the manuscript. The work described 541 has not been published previously and is not accepted for publication elsewhere. 542 Data Availability 543 NGS2020_09_SA187 RNA-Seq project was deposited to the Gene Expression Omnibus 544 of the National Center of Biotechnology Information (Edgard R. et al. 2002): submission to 545 GEO/NCBI in progress). All steps of the experiment, from growth conditions to bioinformatic 546 analyses, were detailed in CATdb database (Gagnot S. et al. 2007): http://tools.ips2.u -547 psud.fr.fr/CATdb/; registered as NGS2020_09_SA187 according to the MINSEQE ‘minimum 548 information about a high-throughput sequencing experiment’. 549

550 Amina Ilyas was supported by a PhD scholarship from the French Ministry of Higher 551 Education and Research (Ministère de l'Enseignement supérieur et de la Recherche). This work 552 has benefited from a French State grant (Saclay Plant Sciences, reference n° ANR -17-EUR-553 0007, EUR SPS-GSR) under a France 2030 program (reference n° ANR-11-IDEX-0003). The 554 authors would like to thank all students, colleagues and collaborators who contributed to 555 discussions and technical support throughout the course of this study. 556 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint

557 Abbass, K., Qasim, M. Z., Song, H., Murshed, M., Mahmood, H., and Younis, I. (2022). A 558 review of the global climate change impacts, adaptation, and sustainable mitigation 559 measures. Environmental Science and Pollution Research 29, 42539 –42559. 560 10.1007/s11356-022-19718-6. 561 Ainsworth, E.A., and Long, S.P. (2005). What have we learned from 15 years of free -air CO2 562 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy 563 properties and plant production to rising CO2. New Phytologist 165, 351–372. 564 Ainsworth, E.A., and Rogers, A. (2007). The response of photosynthesis and stomatal 565 conductance to rising [CO2]: mechanisms and environmental interactions. Plant Cell 566 Environ. 30, 258–270. 10.1111/j.1365-3040.2007.01641.x. 567 Ainsworth, E.A., and Long, S.P. (2021). 30 years of free-air carbon dioxide enrichment (FACE): 568 What have we learned about future crop productivity and its potential for adaptation? 569 Global Change Biology 27, 27–49. 570 Andrés-Barrao, C., Lafi, F.F., Alam, I., de Zélicourt, A.A., Eida, A.A., Bokhari, A., Alzubaidy, 571 H., Bajic, V.B., Hirt, H., and Saad, M.M. (2017). Complete genome sequence analysis 572 of Enterobacter sp. SA187, a plant multi -stress tolerance promoting endophytic 573 bacterium. Frontiers in Microbiology 8, 2023. 574 Andrés-Barrao, C., Alzubaidy, H., Jalal, R., Mariappan, K.G., de Zélicourt, A., Bokhari, A., 575 Artyukh, O., Alwutayd, K., Rawat, A., Shekhawat, K., Almeida-Trapp, M., Saad, M.M., 576 and Hirt, H. (2021). Coordinated bacterial and plant sulfur metabolism in Enterobacter 577 sp. SA187 -induced plant salt stress tolerance. Proc Natl Acad Sci U S A 118, 578 e2107417118. 10.1073/pnas.2107417118. 579 Aranjuelo, I., Sanz -Sáez, Á., Jáuregui, I., Irigoyen, J.J., Araus, J.L., Sánchez -Díaz, M., and 580 Erice, G. (2013). Harvest index, a parameter conditioning responsiveness of wheat plants 581 to elevated CO2. Journal of Experimental Botany 64, 1879–1892. 582 Aranjuelo, I., Cabrera -Bosquet, L., Morcuende, R., Avice, J.C., Nogués, S., Araus, J.L., 583 Martínez-Carrasco, R., and Pérez, P. (2011). Does ear C sink strength contribute to 584 overcoming photosynthetic acclimation of wheat plants exposed to elevated CO2? 585 Journal of Experimental Botany 62, 3957. 586 Bolger, A.M., Lohse, M., and Usadel, B. (2014). Trimmomatic: a flexible trimmer for Illumina 587 sequence data. Bioinformatics 30, 2114–2120. 10.1093/bioinformatics/btu170. 588 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Boyes, D.C., Zayed, A.M., Ascenzi, R., McCaskill, A.J., Hoffman, N.E., Davis, K.R., and 589 Görlach, J. (2001). Growth stage-based phenotypic analysis of Arabidopsis: a model for 590 high throughput functional genomics in plants. Plant Cell 13, 1499 –1510. 591 10.1105/tpc.010011. 592 Breecker, D.O., Sharp, Z.D., and McFadden, L.D. (2010). Atmospheric CO2 concentrations 593 during ancient greenhouse climates were similar to those predicted for A.D. 2100. 594 Proceedings of the National Academy of Sciences 107, 576–580. 595 Compant, S., Clément, C., and Sessitsch, A. (2010). Plant growth -promoting bacteria in the 596 rhizo- and endosphere of plants: Their role, colonization, mechanisms involved and 597 prospects for utilization. Soil Biology and Biochemistry 42, 669–678. 598 Cotrufo, M.F., Ineson, P., and Scott, A. (1998). Elevated CO2 reduces the nitrogen concentration 599 of plant tissues. Glob. Change Biol. 4, 43–54. 600 Dobin, A., Davis, C.A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P., Chaisson, 601 M., and Gingeras, T.R. (2013). STAR: ultrafast universal RNA -seq aligner. 602 Bioinformatics 29, 15–21. 10.1093/bioinformatics/bts635. 603 Gifford, R.M., Barrett, D.J., and Lutze, J.L. (2000). The effects of elevated [CO2] on the C:N 604 and C:P mass ratios of plant tissues. Plant Soil 224, 1–14. 605 Glick, B.R. (2012). Plant Growth -Promoting Bacteria: Mechanisms and Applications. 606 Scientifica (Cairo) 2012, 1–15. 607 Gojon, A., Cassan, O., Bach, L., Lejay, L., and Martin, A. (2023). The decline of plant mineral 608 nutrition under rising CO2: physiological and molecular aspects of a bad deal. Trends 609 Plant Sci. 28, 185–198. 10.1016/j.tplants.2022.09.002. 610 Guo, H., et al. (2013). Elevated CO2 modifies N acquisition of Medicago truncatula by 611 enhancing N fixation and reducing nitrate uptake from soil. PLoS One 8, e81373. 612 Guzmán, P., and Ecker, J.R. (1990). Exploiting the triple response of Arabidopsis to identify 613 ethylene-related mutants. Plant Cell 2, 513–523. 614 Hardoim, P.R., van Overbeek, L.S., and van Elsas, J.D. (2008). Properties of bacterial 615 endophytes and their proposed role in plant growth. Trends Microbiol. 16, 463–471. 616 Irigoyen, J.J., Goicoechea, N., Antolín, M.C., Pascual, I., Sánchez-Díaz, M., Aguirreolea, J., and 617 Morales, F. (2014). Growth, photosynthetic acclimation and yield quality in legumes 618 under climate change simulations: An updated survey. Plant Science 226, 22–29. 619 Jablonski, L.M., Wang, X., and Curtis, P.S. (2002). Plant reproduction under elevated CO2 620 conditions: a meta-analysis of reports on 79 crop and wild species. New Phytol. 156, 9–621 26. 622 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Kant, S., Seneweera, S., Rodin, J., Materne, M., Burch, D., Rothstein, S.J., and Spangenberg, G. 623 (2012). Improving yield potential in crops under elevated CO2: integrating the 624 photosynthetic and nitrogen utilization efficiencies. Frontiers in Plant Science 3, 162. 625 Kopylova, E., Noé, L., and Touzet, H. (2012). SortMeRNA: fast and accurate filtering of 626 ribosomal RNAs in metatranscriptomic data. Bioinformatics 28, 3211 –3217. 627 10.1093/bioinformatics/bts611. 628 Krapp, A., et al. (1993). Regulation of the expression of rbcS and other photosynthetic genes by 629 carbohydrates: a mechanism for the sink regulation of photosynthesis. Plant J 3, 817 –630 826. 631 Lahlali, M.T., Laasli, S.E., Gachara, G., Ezzouggari, R., Belabess, Z., Aberkani, K., Assouguem, 632 A., Meddich, A., El Jarroudi, M., and Ait Barka, E. (2024). Effects of climate change on 633 plant pathogens and host-pathogen interactions. Crop and Environment 3. 634 Leakey, A.D.B., Ainsworth, E.A., Bernacchi, C.J., Rogers, A., Long, S.P., and Ort, D.R. (2009). 635 Elevated CO2 effects on plant carbon, nitrogen, and water relations: Six important 636 lessons from FACE. Journal of Experimental Botany 60, 2859–2876. 637 Long, S.P., Ainsworth, E.A., Rogers, A., and Ort, D.R. (2004). Rising Atmospheric Carbon 638 Dioxide: Plants FACE the future. Annual Review of Plant Biology 55, 591–628. 639 Ma, Z. (2021). Effects of elevated CO2 on plant root form and function: a review. Plant Root 640 15. 641 Ma, B., Ma, T.L., Xian, W., Hu, B., and Chu, C. (2022). Interplay between ethylene and nitrogen 642 nutrition: How ethylene orchestrates nitrogen responses in plants. Journal of Integrative 643 Plant Biology 64, 1904–1921. 10.1111/jipb.13355. 644 McCarthy, D.J., Chen, Y., and Smyth, G.K. (2012). Differential expression analysis of 645 multifactor RNA-Seq experiments with respect to biological variation. Nucleic Acids 646 Res 40, 4288–4297. 647 McDonald, E.P., Erickson, J.E., and Kruger, E.L. (2002). Can decreased transpiration limit plant 648 nitrogen acquisition in elevated CO2? Funct. Plant Biol. 29, 1115–1120. 649 Meinshausen, M., Smith, S.J., Calvin, K., Daniel, J.S., Kainuma, M.L.T., Lamarque, J., 650 Matsumoto, K., Montzka, S.A., Raper, S.C.B., Riahi, K., Thomson, A., Velders, G.J.M., 651 and van Vuuren, D.P.P. (2011). The RCP greenhouse gas concentrations and their 652 extensions from 1765 to 2300. Climatic Change 109, 213–241. 653 Mikhaylov, A., Moiseev, N., Aleshin, K., and Burkhardt, T. (2020). Global climate change and 654 greenhouse effect. Entrepreneurship and Sustainability Issues 7, 2897 –2913. 655 10.9770/jesi.2020.7.4(21). 656 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Moore, B.D., Cheng, S.H., Sims, D., and Seemann, J.R. (1999). The biochemical and molecular 657 basis for photosynthetic acclimation to elevated atmospheric CO2. Plant, Cell & 658 Environment 22, 567–582. 659 NOAA (2025). Trends in Atmospheric Carbon Dioxide. NOAA. 660 Palit, P., Kudapa, H., Zougmore, R., Kholova, J., Whitbread, A., Sharma, M., and Varshney, 661 R.K. (2020). An integrated research framework combining genomics, systems biology, 662 physiology, modelling and breeding for legume improvement in response to elevated 663 CO2 under climate change scenario. Curr Plant Biol. 22, 100149. 664 10.1016/j.cpb.2020.100149. 665 Pérez, P., Alonso, A., Zita, G., Morcuende, R., and Martínez -Carrasco, R. (2011). 666 Downregulation of Rubisco activity under combined increases of CO2 and temperature 667 minimized by changes in Rubisco kcat in wheat. Plant Growth Regulation 65, 439–447. 668 Persello-Cartieaux, F., Nussaume, L., and Robaglia, C. (2003). Tales from the underground: 669 molecular plant-rhizobacteria interactions. Plant, Cell Environ. 26, 189–199. 670 Pieterse, C.M.J., Zamioudis, C., Berendsen, R.L., Weller, D.M., Van Wees, S.C.M., and Bakker, 671 P.A.H.M. (2014). Induced Systemic Resistance by Beneficial Microbes. Annu Rev 672 Phytopathol. 52, 347–375. 673 Rachmilevitch, S., Cousins, A.B., and Bloom, A.J. (2004). Nitrate assimilation in plant shoots 674 depends on photorespiration. Proc Natl Acad Sci U S A 101, 11506 –11510. 675 10.1073/pnas.0404388101. 676 Raudvere, U., Kolberg, L., Kuzmin, I., Arak, T., Adler, P., Peterson, H., and Vilo, J. (2019). 677 g:Profiler: a web server for functional enrichment analysis and conversions of gene lists 678 (2019 update). Nucleic Acids Res. 47, W191-W198. 10.1093/nar/gkz369. 679 Rho, H., Doty, S.L., and Kim, S.H. (2020). Endophytes alleviate the elevated CO2 -dependent 680 decrease in photosynthesis in rice, particularly under nitrogen limitation. Journal of 681 Experimental Botany 71, 707–718. 682 Rao, X., Huang, X., Zhou, Z., and Lin, X. (2013). An improvement of the 2^( -delta delta CT) 683

for quantitative real -time polymerase chain reaction data analysis. Biostat 684 Bioinforma Biomath 3, 71. 685 Rigaill, G., Balzergue, S., Brunaud, V., Blondet, E., Rau, A., Rogier, O., Caius, J., Maugis -686 Rabusseau, C., Soubigou-Taconnat, L., Aubourg, S., Lurin, C., Martin-Magniette, M.L., 687 and Delannoy, E. (2018). Synthetic data sets for the identification of key ingredients for 688 RNA-seq differential analysis. Briefings in Bioinformatics 19, 65 –76. 689 10.1093/bib/bbw092. 690 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Robinson, M.D., McCarthy, D.J., and Smyth, G.K. (2010). edgeR: a Bioconductor package for 691 differential expression analysis of digital gene expression data. Bioinformatics 26, 139–692 140. 693 Rogers, A., and Ainsworth, E.A. (2006). The Response of Foliar Carbohydrates to Elevated 694 [CO2]. Managed Ecosystems and CO2, 293–308. 695 Rogers, A., Ainsworth, E.A., and Leakey, A.D.B. (2009). Will elevated carbon dioxide 696 concentration amplify the benefits of nitrogen fixation in legumes? Plant Physiol 151, 697 1009–1016. 698 Rolli, E., de Zélicourt, A., Alzubaidy, H., Karampelias, M., Parween, S., Rayapuram, N., Han, 699 B., Froehlich, K., Abulfaraj, A.A., Alhoraibi, H., Mariappan, K., Andrés -Barrao, C., 700 Colcombet, J., and Hirt, H. (2022). The Lys-motif receptor LYK4 mediates Enterobacter 701 sp. SA187 triggered salt tolerance in Arabidopsis thaliana. Environ Microbiol. 24, 223–702 239. 10.1111/1462-2920.15839. 703 Rubio-Asensio, J.S., and Bloom, A.J. (2017). Inorganic nitrogen form: a major player in wheat 704 and Arabidopsis responses to elevated CO2. J Exp Bot 68, 2611–2625. 705 Ruiz-Vera, U.M., de Souza, A.P., Long, S.P., and Ort, D.R. (2017). The role of sink strength 706 and nitrogen availability in the down -regulation of photosynthetic capacity in field -707 grown Nicotiana tabacum L. at elevated CO2 concentration. Frontiers in Plant Science 708 8, 998. 709 Saad, M., de Zelicourt, A., Rolli, E., Synek, L., and Hirt, H. (2018). Quantification of root 710 colonizing bacteria. Bio-protocol 8, e2927. 10.21769/BioProtoc.2927. 711 Shekhawat, K., Saad, M.M., Sheikh, A., Mariappan, K., Al-Mahmoudi, H., Abdulhakim, F., and 712 Hirt, H. (2021). Root endophyte induced plant thermotolerance by constitutive chromatin 713 modification at heat stress memory gene loci. EMBO reports 22, e51049. 714 Singer, S.D., Chatterton, S., Soolanayakanahally, R.Y., Subedi, U., Chen, G., and Acharya, S.N. 715 (2020). Potential effects of a high CO2 future on leguminous species. Plant-Environment 716 Interactions 1, 67–94. 717 Smet, D., Depaepe, T., Vandenbussche, F., Callebert, P., Nijs, I., Ceulemans, R., and Van Der 718 Straeten, D. (2020). The involvement of the phytohormone ethylene in the adaptation of 719 Arabidopsis rosettes to enhanced atmospheric carbon dioxide concentrations. 720 Environmental and Experimental Botany 177, 104128. 721 Taub, D., Miller, B., and Allen, H. (2008). Effects of elevated CO2 on the protein concentration 722 of food crops: a meta-analysis. Glob. Change Biol. 14, 565–575. 723 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Taub, D.R., and Wang, X. (2008). Why are nitrogen concentrations in plant tissues lower under 724 elevated CO2? A critical examination of the hypotheses. Journal of Integrated Plant 725 Biology 50, 1365–1374. 726 Thilakarathne, A.S., Liu, F., and Zou, Z. (2025). Plant Signaling Hormones and Transcription 727 Factors: Key Regulators of Plant Responses to Growth, Development, and Stress. Plants 728 14, 1070. 10.3390/plants14071070. 729 Torralbo, F., Vicente, R., Morcuende, R., González-Murua, C., and Aranjuelo, I. (2019). C and 730 N metabolism in barley leaves and peduncles modulates responsiveness to changing 731 CO2. J Exp Bot. 70, 599–611. 10.1093/jxb/ery380. 732 Vessey, J.K. (2003). Plant growth promoting rhizobacteria as biofertilizers. Plant Soil 255, 571–733 586. 734 de Zélicourt, A., Synek, L., Saad, M.M., Alzubaidy, H., Jalal, R., Xie, Y., and Hirt, H. (2018). 735 Ethylene induced plant stress tolerance by Enterobacter sp. SA187 is mediated by 2-keto-736 4-methylthiobutyric acid production. PLoS genetics 14, e1007273. 737 Zhang, Y., and Fernie, A.R. (2023). The Role of TCA Cycle Enzymes in Plants. Adv Biol 738 (Weinh) 7, e2200238. 10.1002/adbi.202200238. 739 Zhu, C., Zhu, J., Cao, J., Jiang, Q., Liu, G., and Ziska, L.H. (2014). Biochemical and molecular 740 characteristics of leaf photosynthesis and relative seed yield of two contrasting rice 741 cultivars in response to elevated [CO2]. J Exp Bot. 65, 6049–6056. 10.1093/jxb/eru344. 742 743 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Figure legends 744 Figure 1: Growth phenotyping of Arabidopsis Col-0 grown under aCO 2 and elevated 745 eCO2 conditions 746 (A) Plant fresh weight, (B) Shoot fresh weight, (C) Root fresh weight and (D) Total dry weight 747 of plant 1 5 dpg on ½ MS medium under ambient (aCO 2) and elevated CO 2 (eCO2) without 748 (Mock) and with (SA187) SA187. (E) Primary root length and (F) Lateral root density of plants 749 12 dpg under ambient (aCO 2) and elevated CO 2 (eCO2) without (Mock) and with (SA187) 750 SA187. Light green indicate mock plants and dark green SA187 inoculated plants. For each 751 measured parameter and every condition, data represent means (n > 30, 3 independent 752 experiments) with standard deviations. Asterisks indicate a statistical difference based on two-753 sided, unpaired Student’s t-test: *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant. 754 Figure 2: SA187 growth under aCO2 and eCO2 conditions 755 (A) SA187 growth in liquid LB culture under agitation at 30°C under aCO 2 and eCO 2 756 conditions, data represent the mean of 3 independent experiments (B) CFU counting of SA187 757 bacteria retrieved from root or shoot organs of Arabidopsis thaliana plants grown under 758 ambient (aCO2) and elevated CO2 (eCO2) conditions. Blue bars indicate values from aCO2 and 759 orange from eCO 2 conditions. Data represent means (n =3 independent experiments) with 760 standard deviations. 761 Figure 3: Arabidopsis shoot transcriptome analysis in response to SA187 inoculation and 762 eCO2 763 (A) Heat map of DEGs in response to SA187, eCO 2 both compared to mock plants grown in 764 ambient conditions. Original mean counts were subjected to data adjustment by normalizing 765 genes across all samples. Hierarchical clustering is displayed by average linkage under Pearson 766 Correlation (MeV version 4). The colour scale indicates high and low expression levels. ( B) 767 Four selected clusters showing contrasting expression patterns between eCO 2 and SA187 + 768 eCO2 conditions using functional profiling with G:Profiler. qPCR expression analysis of (C) 769 ERF104, (D) WRKY30, (E) MYB51, (F) CRK4, in shoots, in the four different tested conditions. 770 Normalized expression indicates the linear fold change compared to mock -treated plants in 771 aCO2. Data represent mean (n = 3 independent experiments) with standard errors. Asterisks 772 indicate a statistical difference based on the Mann & Whitney test: *p < 0.05; ***p < 0.001. 773 Figure 4: Arabidopsis shoot metabolic analysis in response to SA187 inoculation and eCO2 774 (A) Heat map of differentially accumulated metabolites in response to SA187 in either aCO2 or 775 eCO2 conditions. Original mean counts were subjected to data adjustment by normalizing 776 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint metabolites across all samples. Hierarchical clustering is displayed by average linkage under 777 Pearson Correlation (MeV version 4). The colour scale indicates high and low accumulation 778 levels. (B) Shoot C/N ratio, (C) shoot C content (% DW), (D) shoot N content (% DW) of mock 779 (light green) and SA187 inoculated seedlings (dark green) under aCO2 or eCO2 conditions. Data 780 represent means (n = 3 independent experiments) with standard deviations. Asterisks indicate 781 a statistical difference based on the Mann & Whitney test: *p < 0.05. 782 Figure 5: ein2 response to SA187 inoculation under eCO2 conditions 783 (A) Plant fresh weight of mock and SA187 inoculated Col -0 and ein2-1 plants in aCO 2 and 784 eCO2 conditions. qPCR expression analysis of (B) ERF104, (C) SBE2.1, (D) NRT2.5, (E) FBA6 785 and (F) WR3/NRT3.1 in Col-0 and ein2-1 shoots under Mock (light green and light orange, 786 respectively) or SA187 inoculation (light green and light orange, respectively) in eCO 2 787 conditions. Expression values are indicated as linear fold changes compared to mock aCO 2 788 conditions of a given genotype. Data represent means (n = 3 independent experiments) with 789 standard errors. Asterisks indicate a statistical difference based on the Mann & Whitney test: 790 *p < 0.05; ***p < 0.001. 791 Supplemental Figure 1: Arabidopsis root transcriptome analysis in response to SA187 792 and eCO2. 793 (A) Heat map of DEGs in response to SA187 in either aCO2 or eCO2 conditions. Original mean 794 counts were subjected to data adjustment by normalizing genes across all samples. Hierarchical 795 clustering is displayed by average linkage under Pearson Correlation (MeV version 4). The 796 colour scale indicates high and low expression levels. B Two selected clusters showing 797 contrasting expression patterns between eCO 2 and SA187 + eCO2 conditions using functional 798 profiling with G:Profiler. 799 Supplemental Figure 2: Arabidopsis root metabolic analysis in response to SA187 800 inoculation 801 (A) Heat map of differentially accumulated metabolites in roots in response to SA187 802 inoculation in aCO 2 or eCO 2 conditions. Original mean counts were subjected to data 803 adjustment by normalizing metabolites across all samples. Hierarchical clustering is displayed 804 by average linkage under Pearson Correlation (MeV version 4). The colour scale indicates high 805 and low accumulation levels. 806 Supplemental Figure 3: Growth phenotyping of Arabidopsis Col-0 and ein2 grown under 807 elevated eCO2 conditions 808 (A) Root Length, (B) Lateral root density, of eCO 2 grown Col -0 and ein2-1 plants 12 dpg 809 without (Mock) and with (SA187) SA187. Light and dark green indicate Col -0 mock and 810 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint SA187 plants respectively and light and dark orange indicate ein-1 mock and SA187 plants 811 respectively. For each measured parameter and every condition, data represent means (n > 30, 812 3 independent experiments) with standard deviations. Asterisks indicate a statistical difference 813 based on two-sided, unpaired Student’s t-test: **p < 0.01. 814 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint List of figures 815 Figure 1: Growth phenotyping of Arabidopsis Col-0 grown under aCO 2 and elevated eCO 2 816 conditions 817 Figure 2: SA187 growth under aCO2 and eCO2 conditions 818 Figure 3: Arabidopsis shoot transcriptome analysis in response to SA187 inoculation and eCO2 819 Figure 4: Arabidopsis shoot metabolic analysis in response to SA187 inoculation and eCO2 820 Figure 5: ein2 response to SA187 inoculation under eCO2 conditions 821 822 List of supplemental information 823 Supplemental Figure 1: Arabidopsis root transcriptome analysis in response to SA187 and 824 eCO2. 825 Supplemental Figure 2: Arabidopsis root metabolic analysis in response to SA187 inoculation 826 Supplemental Figure 3: Growth phenotyping of Arabidopsis Col-0 and ein2 grown under 827 elevated eCO2 conditions 828 Supplemental Table 1: List qRT-PCR Primer sets used in this study. 829 Supplemental Table 2: List of genes deregulated in shoots, at least, one condition compared 830 to Mock aCO2 condition. 831 Supplemental Table 3: List of genes deregulated in roots, at least, one condition compared to 832 Mock aCO2 condition. 833 Supplemental Table 4: List of metabolites identified and semi-quantified in shoot by GC-MS 834 analysis. 835 Supplemental Table 5: List of metabolites identified and semi -quantified in root by GC -MS 836 analysis. 837 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint +24% ns 0 10 20 30 40 50 60 aCO2 eCO2 Shoot fresh weight (mg) 0 5 10 15 20 25 aCO2 eCO2 Root fresh weight (mg) 0 10 20 30 40 50 60 70 80 aCO2 eCO2 Plant fresh weight (mg) ** +51% *** +59% +20% ns ** +43% ** +51% +39% * ** +97% *** +61% 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 aCO2 eCO2 Planrt dry weight (mg) +24% ns *** +26% * +71% 0 1 2 3 4 5 6 7 8 9 10 aCO2 eCO2 Root length (cm) +23% *** +16% ** 0 0.5 1 1.5 2 2.5 aCO2 eCO2 Lateral root density (LR/cm) Figure 1: Growth phenotyping of Arabidopsis Col-0 grown under aCO2 and elevated eCO2 conditions (A) Plant fresh weight, (B) Shoot fresh weight, (C) Root fresh weight and (D) Total dry weight of plant 15 dpg on ½ MS medium under ambient (aCO2) and elevated CO 2 (eCO2) without (Mock) and with (SA187) SA187. (E) Primary root length and (F) Lateral root density of plants 12 dpg under ambient (aCO2)a n de l e v a t e dC O2 (eCO2) without (Mock) and with (SA187) SA187. Light green indicate mock plants and dark green SA187 inoculated plants. For each measured parameter and every condition, data represent means (n > 30, 3 independent experiments) with standard deviations. Asterisks indicate a statistical difference based on two-sided, unpaired Student’s t-test: * p <0 . 0 5 ;* *p < 0.01; *** p < 0.001; ns, not significant. A D B E C F aCO2 eCO2 aCO2 eCO2 aCO2 eCO2 aCO2 eCO2 aCO2 eCO2 aCO2 eCO2 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 368 1 0 Bacterial cell density OD600 Time (hours) aCO2 eCO2 0 1 2 3 4 5 6 7 Root Shoot Bacterial density (log10CFU / mg FW) Figure 2: SA187 growth under aCO2 and eCO2 conditions (A) SA187 growth in liquid LB culture under agitation at 30°C under aCO 2 and eCO2 conditions, data represent the mean of 3 independent experiments (B) CFU counting of SA187 bacteria retrieved from root or shoot organs of Arabidopsis thaliana plants grown under ambient (aCO2)a n de l e v a t e dC O2 (eCO2) conditions. Blue bars indicate values from aCO 2 and orange from eCO2 conditions. Data represent means (n =3 independent experiments) with standard deviations. A B aCO2 eCO2 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint -1.5 1.50 Normalized expression -2 -1 0 1 2 Normalized expression Mock SA187 eCO 2 SA187 + eCO2 Response to SA, JA, ethylene Response to hypoxia Response to bacterium Indole glucosinolate metablic process Defense response by cell wall thickening -2 -1 0 1 2 Normalized expression Tricarboxylic acid cycle Response to hypoxia Response to ethylene Defense response Carbon metabolism -2 -1 0 1 2 Normalized expression Response to hypoxia Response to salycilic acid Response to nitrogen Response to cold, salt stress Figure 3: Arabidopsis shoot transcriptome analysis in response to SA187 inoculation and eCO2 (A) Heat map of DEGs in response to SA187, eCO 2 both compared to mock plants grown in ambient conditions. Original mean counts were subjected to data adjustment by normalizing genes across all samples. Hierarchical clustering is displayed by average linkage under Pearson Correlation (MeV version 4). The colour scale indicates high and low expression levels. ( B) Four selected clusters showing contrasting expression patterns between eCO 2 and SA187 + eCO2 conditions using functional profiling with G:Profiler. qPCR expression analysis of (C) ERF104, (D) WRKY30, (E) MYB51, (F) CRK4, in shoots, in the four different tested conditions. Normalized expression indicates the linear fold change compared to mock- treated plants in aCO2. Data represent mean (n = 3 independent experiments ) with standard errors. Asterisks indicate a statistical difference based o nt h eM a n n&W h i t n e yt e s t :*p < 0.05; ***p < 0.001. AB Starch catabolism Flavonoid metabolic process Carbohydrate metabolic process Starch and sucrose metabolism S F C S -2 -1 0 1 2 Normalized expression C2 C3 C5 C7 C2 C3 C5 C7 C DEF Mock SA187 eCO 2 SA187 + eCO2 Mock SA187 eCO 2 SA187 + eCO2 Mock SA187 eCO 2 SA187 + eCO2 Mock SA187 eCO 2 SA187 + eCO2 WRKY30 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint Figure 4: Arabidopsis shoot metabolic analysis in response to SA187 inoculation and eCO2 (A) Heat map of differentially accumulated metabolites in response to SA187 in either aCO 2 or eCO2 conditions. Original mean counts were subjected to data adjustment by normalizing metabolites across all samples. Hierarchical clustering is displayed by average linkage under Pearson Correlation (MeV version 4). The colour scale indicates high and low accumulation levels. ( B) Shoot C/N ratio, (C) shoot C content (% DW), (D) shoot N content (% DW) of mock (light green) and SA187 inoculated seedlings (dark green) under aCO2 or eCO 2 conditions. Data represent means (n = 3 independent experiments) with standard deviations. Asterisks indicate a statistical difference based on the Mann & Whitney test: *p <0 . 0 5 . AB C D * * 4.0 4.5 5.0 5.5 aCO2 eCO2 Shoot C/N ratio D 7.0 7.2 7.4 7.6 7.8 8.0 8.2 8.4 8.6 aCO2 eCO2 N content (% DW) 37.0 38.0 39.0 40.0 41.0 42.0 aCO2 eCO2 C content (% DW) * -1.5 1.50 Normalized accumulation Mock SA187 eCO 2 SA187 + eCO2 aCO2 eCO2 aCO2 eCO2aCO2 eCO2 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint 0 1 2 3 4 5 Mock_eCO2 SA187_eCO2 ERF104 Expression change compared to mock_aCO2 0 50 100 150 200 Mock_eCO2 SA187_eCO2 NRT2.5 Expression change compared to mock_aCO2 0 10 20 30 40 50 Mock_eCO2 SA187_eCO2 WR3/NRT3.1 Expression change compared to mock_aCO2 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Mock_eCO2 SA187_eCO2 FBA6 Expression change compared to mock_aCO2 0.0 1.0 2.0 3.0 4.0 Mock eCO2 SA187 eCO2 SBE2.1 Expression change compared to mock_aCO2 0 10 20 30 40 50 Col-0 ein2-1 +21% ** -24% ** Plant fresh weight (mg) AB CD EF Figure 5: ein2 response to SA187 inoculation under eCO2 conditions (A) Plant fresh weight of mock and SA187 inoculated Col-0 and ein2-1 plants in aCO2 and eCO2 conditions. qPCR expression analysis of (B) ERF104, (C) SBE2.1, (D) NRT2.5, (E) FBA6 and (F) WR3/NRT3.1 in Col-0 and ein2-1 shoots under Mock (light green and light orange, respectively) or SA187 inoculation (light green and light orange, respectively) in eCO2 conditions. Expression values are indicated as linear fold changes compared to mock aCO2 conditions of a given genotype. Data represent means (n = 3 independent experiments) with standard errors. Asterisks indicate a statistical difference based on the Mann & Whitney test: *p < 0.05; ***p < 0.001. Mock eCO2 SA187 eCO2 Mock eCO2 SA187 eCO2 Mock eCO2 SA187 eCO2 Mock eCO2 SA187 eCO2 Mock eCO2 SA187 eCO2 Col-0 ein2-1 (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprintthis version posted July 9, 2025. ; https://doi.org/10.1101/2025.07.08.663752doi: bioRxiv preprint