XLG2 mediates plant defence at elevated temperature
Heterotrimeric G proteins mediate resistance against a diverse range of pathogens in different experimental conditions however, the relative performance of the G protein-mediated defence at different temperatures and their possible involvement in acclimation-induced priming has not been studied 20,22,32,39,48−50. We therefore designed a series of experiments to test the effect of temperature on the G protein α subunit XLG2-mediated defence. When Arabidopsis wild-type (WT) and xlg2-1 mutant plants grown at 23°C were inoculated with F. oxysporum and subsequently incubated at either 23°C or 29°C we observed that xlg2-1 mutants were significantly more susceptible than WT plants at 29°C (Fig. 1a, b), indicating that XLG2 contributes to the defence against F. oxysporum at elevated temperature. Next, we tested whether acclimatising plants to 29°C prior to pathogen inoculation affects resistance to F. oxysporum, and if so, whether this response involves XLG2. WT and xlg2-1 plants grown at 23°C were split into two groups; one group was maintained at 23°C while the second group was transferred to 29°C for 24 hours. Both groups of plants were subsequently challenged with F. oxysporum and further incubated at 29°C for disease evaluation. We observed that pre-treatment at 29°C resulted in increased resistance in WT plants (Fig. 1c) while xlg2-1 plants did not display any significant difference between pre-treated and not pre-treated groups (Fig. 1c). Together, these observations suggest that elevated temperature can be a priming factor enhancing the defence response against F. oxysporum in an XLG2-dependent manner.
The interaction between Arabidopsis and the bacterial pathogen Pst DC3000 is commonly studied in laboratory conditions at temperatures ranging from 20°C to 23°C. However, this pathogen has been found to be more infectious at elevated temperatures, partly due to inhibition of SA biosynthesis 18,51. Our evaluations revealed that the xlg2-1 mutants exhibited increased susceptibility to Pst DC3000 compared to WT plants at both 23°C and 29°C (Fig. 1d, e). We then evaluated the resistance of WT and xlg2-1 plants to Pst DC3000, after acclimatisation to 29°C for 48 h. Acclimatisation significantly increased Arabidopsis susceptibility to the pathogen (Fig. 1f). Notably, the susceptibility increased similarly in WT and xlg2-1 plants pointing to a common mechanism affected by the acclimatization, perhaps the suppression of SA biosynthesis 18. We hypothesised that, similar to our observations for F. oxysporum, elevated temperature had an XLG2-dependent priming effect against Pst DC3000 infection, but it was masked by the strong negative impact of SA deprivation. To test this hypothesis, we first measured the effects of elevated temperatures on the isochorismate synthase 1 (ICS1) transcript levels, which is a key enzyme in SA biosynthesis, and the XLG2 gene. Prolonged exposure to 29°C resulted in a significant reduction of ICS1 transcript levels after 48 hours (Fig. 1g), while XLG2 levels were significantly increased as early as 24 hours of exposure (Fig. 1h), suggesting that XLG2 may play a role in plant responses to elevated temperatures. To investigate a potential XLG2-mediated priming against Pst DC3000, we conducted experiments on plants expressing the SA degrading enzyme, NahG. As expected, the NahG plants displayed high susceptibility to the pathogen, even at 23°C (Fig. 1i). However, in contrast with WT plants, NahG plants acclimatised to 29°C for 48 hours, showed increased resistance compared to those maintained at 23°C (Fig. 1i), indicating that acclimatisation to elevated temperatures had a positive priming effect on the immunity of NahG plants. Importantly, analysis of the double mutant xlg2-1 NahG confirmed that the priming effect observed after exposure of the NahG plants to 29°C was mediated by XLG2, as previously observed for F. oxysporum (Fig. 1i).
To gain further insights into the underlying molecular mechanisms, we analysed the expression patterns of four marker genes indicative of bacterial (PR1 & PR2) or fungal (PR4 & PDF1.2) infections. As expected, all four genes were strongly induced in response to the corresponding pathogens at 23°C (Fig. 1j-m). When the plants were incubated at 29°C after pathogen inoculation, genes were upregulated to a lesser degree (Fig. 1j-m). Importantly, while the relative expression levels for PR1, PR2 and PR4 was similar for both WT and xlg2-1 mutants at 23°C, all four genes consistently showed lower expression levels in xlg2-1 mutants at 29°C, again confirming a positive role for XLG2 in the defence response at elevated temperatures.
Elevated Ambient Temperature Promotes Accumulation Of Xlg2 Protein In The Nucleus
In order to further understand the mechanism by which XLG2 mediates temperature-induced priming of immunity, we sought to investigate the dynamics of XLG2 localization in response to temperature elevation in transgenic Arabidopsis plants expressing GFP-XLG2 fusion proteins. Under normal growth conditions (23°C) GFP-XLG2 was primarily observed at the plasma membrane, with only faint fluorescence present in nuclei (Fig. 2a), as previously reported 20,45. Interestingly, when plants were incubated at 29°C, nuclear GFP fluorescence notably increased, while fluorescence intensity at the plasma membrane decreased (Fig. 2a). The fluorescence patterns of GFP-XLG1 and GFP-XLG3 did not change in response to elevated temperature, with GFP-XLG1 detected at the plasma membrane and GFP-XLG3 at the plasma membrane and nuclei (Fig. 2a). Quantification of the fluorescence intensity revealed that GFP-XLG2 protein levels rapidly accumulated in nuclei 10 minutes after exposure to 29°C (P < 0.0001) and maintained high levels for the duration of the 29°C treatment (Fig. 2b). When plants were transferred back to 23°C, nuclear GFP-XLG2 levels remained high for several hours, with a small but statistically significant decrease in fluorescence intensity observed after six hours (P = 0.0132), progressively decreasing to the original low levels after 12 hours (Fig. 2b). At the plasma membrane, a significant decrease in GFP-XLG2 was observable half an hour after exposure to 29°C (P < 0.007) with the lowest value reached after 6 hours (Fig. 2c). When plants were returned to 23°C, fluorescence intensity at the plasma membrane was restored to initial values after 12 hours (Fig. 2c). Prolonged exposure to 29°C (up to one week) showed continued strong GFP-XLG2 fluorescence in the nuclei and weak signal at the plasma membrane (Extended Data Fig. 1). Many plant developmental responses to elevated temperature are mediated by auxins 52. To test whether nuclear accumulation of GFP-XLG2 was auxin dependent, GFP-XLG2 expressing plants were treated with the synthetic auxin, 1-naphthaleneacetic acid (NAA), or the auxin transport inhibitor, N-1-naphthylphthalamidic acid (NPA) at either 23°C or 29°C. Neither treatment altered the fluorescence at either temperature (Extended Data Fig. 2). As FLS2 interacts with and phosphorylates XLG2 36, we investigated whether flg22 treatment or mutations of XLG2 phosphorylated amino acids affects GFP-XLG2 localization. Our results indicate that in transgenic Arabidopsis plants expressing GFP-XLG2, regardless of flg22 treatment, the localization of GFP fluorescence primarily occurred at the plasma membrane at 23°C, while at 29°C it is observed in the nuclei (Extended Data Fig. 3a). To study the effect of phosphorylation on XLG2 localization we expressed either a phosphomimetic GFP-XLG2-6D or a non-phosphorylatable GFP-XLG2-6A in N. benthamiana leaves. Confocal microscopy revealed that the phosphomimetic and non-phosphorylatable versions of XLG2 had the same localization patterns as the non-mutagenized GFP-XLG2 protein, suggesting that the phosphorylation state of XLG2 does not affect its cellular localization (Extended Data Fig. 3b). Given the important role of XLG2 in defence against Pst DC3000 and F. oxysporum, it was tempting to speculate that infection by either of these pathogens might affect the localization patterns of XLG2. However, neither pathogen had any effect on the localization of GFP-XLG2 in transgenic Arabidopsis plants at either 23°C or 29°C (Extended Data Fig. 4). Finally, we observed that exposure to salt stress had no effect on XLG2 localization (Extended Data Fig. 4).
The simultaneous decrease of GFP-XLG2 intensity at the membrane and increase in the nucleus suggest that there is active translocation of XLG2 protein from the plasma membrane to the nucleus. However, it is also plausible that elevated temperatures induce degradation of XLG2 at the plasma membrane, while de-novo synthetized XLG2 is directly imported to the nucleus. To test these two possible scenarios, we treated GFP-XLG2 expressing plants with the protein synthesis inhibitor cycloheximide (CHX) before exposing them to 29°C. No difference was observed between CHX-treated and control plants (Fig. 2d), supporting the translocation hypothesis.
Given the fast dynamic nature of temperature-driven XLG2 translocation, we hypothesised that it may be regulated by circadian temperature fluctuations, potentially promoting priming of immunity after short exposure to elevated temperatures. The Arabidopsis ecotype Col-0 was originally collected from the Gorzów Wielkopolski region of Poland (formerly known as Landsberg an der Warthe), and as such, it is adapted to the relatively moderate temperatures of Central Europe 53. To determine the duration of exposure of Col-0 ancestors to elevated temperatures in their native environment, we analysed temperature data recorded in Gorzów Wielkopolski in June during the last two decades. We found that even during the hottest days, when maximum daytime temperatures surpassed 29°C, the periods with temperatures above 29°C did not exceed ten hours, with an average duration of six hours (Fig. 2e). Therefore we predicted that, opposite to the detrimental effect observed by long exposures to high temperature (Fig. 1e) 18, short (~ 6 hours) exposure to elevated temperature could prime immunity against Pst DC3000 through XLG2 without significantly affecting SA-mediated defence mechanisms, thereby avoiding negative effects associated with SA deprivation. Consistent with our prediction, WT plants pre-treated at 29°C for six hours displayed enhanced resistance to Pst DC3000 compared to control plants grown at 23°C, while the priming effect was not observed in xlg2-1 mutant plants (Fig. 2f).
Nuclear Localization Is Required For Xlg2-mediated Defence Against Pathogens
To investigate the relationship between the defence roles of XLG2 and its subcellular localization, we used a complementation strategy on xlg2 xlg3 double mutants. We aimed to produce plants with XLG2 localization confined to either the plasma membrane or the nucleus by tagging XLG2 with an established Nuclear Export Signal (NES) 54 or a Nuclear Localization Signal (NLS) 55, respectively. To verify the localization of the tagged proteins, transgenic Arabidopsis lines expressing GFP fusions with either XLG2(NES), or XLG2(NLS) were generated. Fluorescence microscopy analysis confirmed that GFP-tagged XLG2(NES) was localized at the plasma membrane or in the peripheral cytosol at 23°C and 29°C, while GFP-tagged XLG2(NLS) was predominantly found in the nucleus at 23°C and 29°C with weak fluorescence observed at the plasma membrane (Extended Data Fig. 5).
For complementation assays, we generated transgenic xlg2 xlg3 Arabidopsis lines expressing either wild-type XLG2 (WT), XLG2(NES), or XLG2(NLS) under the control of the native XLG2 promoter. Analysis by qRT-PCR and western blot using XLG2-specific antibodies confirmed that the transgenic lines had similar transcript and protein levels to WT plants and two independent lines per construct were selected for further analysis (Extended Data Fig. 6a, b). Disease assays showed that xlg2 xlg3 mutants were hypersensitive to F. oxysporum and Pst DC3000 compared to WT plants, as previously reported 20. Analysis of the complementation lines showed that control plants, XLG2(WT), and the two independent XLG2(NLS) lines displayed restoration of resistance to wild-type levels for both pathogens (Fig. 3a, b), while the two XLG2(NES) lines showed susceptibility levels similar to the xlg2 xlg3 mutants (Fig. 3a, b), indicating that nuclear localization is important for disease resistance against the tested pathogens. Previous studies have implicated XLG2 in the response to bacterial-derived flagellin mediated by the FLS2 receptor 20,36. Consistent with those studies, we observed that the production of ROS elicited by flg22 in the xlg2 xlg3 mutant was significantly reduced compared to wild-type plants (Fig. 3c). Similar to the observations in the disease susceptibility assays, expression of XLG2(WT) and XLG2(NLS) restored the phenotype to wild-type levels, while XLG2(NES) failed to do so (Fig. 3c), supporting the role of nuclear-localized XLG2 in flg22-induced ROS accumulation. In addition to defence responses, XLG2 has roles in plant development with xlg2 xlg3 mutants displaying higher stomatal density compared to wild-type plants 45. We found that the increased stomatal density in xlg2 xlg3 mutants was restored to wild-type levels in plants expressing XLG2(WT) and XLG2(NLS), whereas expression of XLG2(NES) did not rescue the mutant phenotype (Fig. 3d). These results suggest that the nuclear localization of XLG2 is crucial for innate immune responses and stomatal development.
XLG2 interacts with several cell-surface RLKs including BIR1, a master regulator of programmed cell death 20,32,56. We previously demonstrated that loss of XLG2 function partially suppresses the seedling lethal phenotype of the bir1 mutant 20. To determine whether this response is dependent on XLG2 subcellular localization, we crossed the XLG2(NES) and XLG2(NLS) transgenic lines with the bir1 xlg2 double mutant to obtain bir1 xlg2 XLG2(NES) and bir1 xlg2 XLG2(NLS) plants. Morphology analyses revealed that expression of either XLG2(NES) or XLG2(NLS) only partially restored the bir1 phenotype (Fig. 3e, f), suggesting that BIR1-mediated programmed cell death involves both plasma membrane-localized and nuclear-localized XLG2.
Xlg2 Interacts With Transcription Factors And Linker Histones
The functional importance of XLG2's nuclear localization established in this work, as well as previous reports on XLG2's interactions with transcription factors and histones 44,56 prompted us to investigate the potential interactions between XLG2 and several well-known defence-related transcription factors. Two in planta approaches were chosen: bimolecular fluorescent complementation (BiFC) and split-firefly luciferase complementation (SFLC) assays, while the established interaction between XLG2 and RTV1 44 was used as positive control. Interaction assays showed that XLG2 interacts with the transcription factors TGA2, TGA5, WRKY33, WRKY50, MYC2 and HSFA1D as well as the linker histones H1.1 and H1.3 (Fig. 4a, b).
TGA2 is a well-known transcriptional regulator of genes involved in plant defence, including several PR genes 57,58. In the absence of pathogen attack, TGA2 homodimers bind to specific DNA sequences repressing the expression of PR genes, however upon interaction with SA-activated NPR1, TGA2 functions as a transcriptional activator of PR genes 57,59. We used the TGA2/NPR1 complex to investigate the potential mechanism of XLG2 in the regulation of PR genes. First, we confirmed the interaction between TGA2 and NPR1 and the TGA2 dimer formation (as interaction between TGA2-CmVenus and TGA2-NmVenus) (Fig. 4c). We also established that XLG2 interacts with NPR1 (Fig. 4c). Since XLG2 interacts with both TGA2 and NPR1 we tested whether XLG2 can affect the TGA2 dimer formation or the interaction strength between NPR1 and TGA2. Quantitative SFLC assays in Nicotiana benthamiana demonstrated that co-expression of XLG2 significantly increased interaction strength for both NPR1/TGA2 and TGA2/TGA2 protein pairs (Fig. 4d, e), suggesting that XLG2 may regulate NPR1/TGA2 transcriptional activity by modulating their interaction strength. Subsequent experiments showed that the XLG2(NLS) variant enhanced the interaction between NPR1 and TGA2, while the XLG2(NES) variant had no effect on the interaction (Fig. 4f). To investigate the role of XLG2 in regulating the transcriptional activity of the NPR1/TGA2 complex, we evaluated the expression of the PR1 gene, a marker of plant defence, in response to combination of elevated temperature and SA treatment. SA was chosen as an elicitor instead of flg22 or a pathogen because it directly binds to NPR1, which then activates the expression of PR1 60. In non-inducing conditions, incubation at 29°C for 6 hours increased PR1 levels (fivefold) in Col-0, while no increase was observed in xlg2-1 plants (Fig. 4g). As expected, application of SA resulted in a significant increase in PR1 transcription in both genotypes, with an average increase of over two orders of magnitude (Fig. 4g). Most importantly, SA induction of PR1 was almost doubled in WT plants by priming at 29°C for 6 hours (Fig. 4g), but no difference was observed in xlg2-1 plants (Fig. 4g), suggesting that the presence of XLG2 enhances the induction of PR1 by SA, perhaps due to its stabilising effect in the NPR1/TGA2 interaction.