Expression of a bacterial salicylate hydroxylase gene <i>NahG</i> impedes the expression of <scp>PAMP</scp>‐triggered immunity‐related genes in plants

The plant hormone salicylic acid (SA) plays a vital role in plant development and response to biotic and abiotic stresses. There are two characterized pathways for SA biosynthesis in plants, catalysed by phenylalanine ammonia-lyase (PAL) and isochorismate synthase (ICS), respectively. Pathogen-induced rapid accumulation of SA in Arabidopsis thaliana (hereafter ‘Arabidopsis’) leaves is primarily dependent on the ICS pathway. Perturbation of specific immune signalling components can sometimes result in the up-regulation of immune responses, termed autoimmunity, which is typically accompanied by the overaccumulation of SA. To evaluate the requirement of SA in those autoimmunity studies and to dissect the SA-independent phenotypes in Arabidopsis and crop species, the introduction of an exogenous salicylate hydroxylase gene, NahG, from Pseudomonas putida or mutation of the isochorismate synthase1 gene (ICS1) by crossing sid2 (SA INDUCTION-DEFICIENT2) mutant plants is usually widely used in plant research. The sid2 mutant or NahG transgenic plant produces approximately 10% of the SA level observed in the wild type (WT) under normal growth conditions (Heck et al., 2003). Interestingly, only about 300 differentially expressed genes (DEGs) are detected in sid2 plants under normal growth conditions (Mine et al., 2018), whereas the exogenous application of SA activates around 2000 DEGs in different Arabidopsis accessions (van Leeuwen et al., 2007). The NahG transgene is widely used as an alternative to the loss-of-function approach for ICS1; however, the exact role of NahG introduction at the molecular level is not well understood. We profiled the overall gene expression in 4-week-old NahG transgenic Arabidopsis using RNA sequencing (RNA-Seq). The introduction of NahG leads to both up-and down-regulated expression of about 2000 DEGs (Figure S1a–c), and these numbers are significantly different from the published RNA-Seq data for sid2 plants from similar growth conditions and developmental stage (Mine et al., 2018), which has 358 down-regulated and 85 up-regulated DEGs (Figures 1a and S1a). We then compared DEGs from NahG and sid2 plants (Mine et al., 2018), and found that 60% of the down-regulated DEGs in sid2 plants are also down-regulated in NahG plants (Figure 1b). These overlapped genes are enriched in different GO terms but related to plant immunity, including ‘defence response’, ‘response to biotic stimulus’ and ‘protein kinase activity’ (Figure S1d,e). Interestingly, we found that multiple receptor-like kinases (RLKs) and receptor-like proteins (RLPs) are among the representative down-regulated DEGs (Figure S1f). The observations of down-regulated expression of RLP/RLKs and key signalling components of the plant immune system led us to examine the immune responses of sid2 and NahG plants. We first treated Arabidopsis leaves with different elicitors, including flg22, nlp20 and chitin, which could be perceived by leucine-rich repeat receptor-like kinase (LRR-RK) FLS2, leucine-rich repeat receptor-like protein (LRR-RLP) RLP23 and lysin motif receptor-like kinase (LysM-RLK) CERK1, respectively. We then measured elicitor-triggered ROS burst, MAPK activation, seedling growth inhibition, and bacterial growth inhibition. The flg22-triggered ROS burst was not affected in sid2 plants, while there was no ROS burst in NahG plants (Figures 1c and S2a). After nlp20 treatment, sid2 and NahG plants showed significantly less ROS accumulation (Figures 1d and S2b). Interestingly, however, after chitin treatment, NahG plants accumulated more ROS compared to Col-0 and sid2 (Figures 1e and S2c). Flg22- and nlp20-triggered MAPK activation was significantly reduced in NahG plants but unaltered in sid2 plants (Figure 1f). Interestingly, chitin-triggered MAPK activation was unaffected in both NahG and sid2 plants (Figure 1f). In PAMP-triggered plant growth inhibition, both flg22- and nlp20-triggered growth inhibition were alleviated in NahG and sid2 plants (Figure 1g,h). Interestingly, we found that chitin cannot inhibit seedling growth in our experimental settings, and NahG plants can grow even to a larger biomass after chitin treatment (Figure 1i). In the PAMP-triggered bacterial growth restriction assay, we found that flg22-triggered bacterial resistance was abolished in NahG plants but only slightly affected in sid2 plants (Figure 1j), while nlp20 can trigger antibacterial activity in NahG and sid2 plants (Figure 1k). In our condition, chitin treatment did not lead to enhanced bacterial resistance (Figure 1l). We measured the expression of RBOHD, which encodes the NADPH/respiratory burst oxidase protein D, and observed no difference between NahG and sid2 (Figure S2d). We then measured the expression levels for BAK1 and SOBIR1, two co-receptors of PAMP recognition receptors (PRRs). The results showed that the expression of BAK1 is unaffected in either NahG or sid2 plants (Figure S2e), but the expression of SOBIR1 in NahG is much lower compared to WT and sid2 (Figure S2f). We profiled the expression of PRRs, and qPCR results showed that the expression of, including FLS2, RLP23 and CERK1, is lower in NahG compared to WT and sid2 (Figure S2g–i). We then challenged Arabidopsis plants with different bacterial strains, including Pseudomonas syringae pv. tomato DC3000 (Pto DC3000), Pto ΔhrcC and Pto avrRpt2, to evaluate the plant basal, PTI (PAMP-triggered immunity) and ETI (effector-triggered immunity) resistance in NahG and sid2 plants. NahG plants accumulated much larger numbers of different bacterial strains than WT, while sid2 plants supported more bacterial growth as well, but much less than NahG plants (Figure 1m–o). Global profiling of gene expression in NahG plants reveals that NahG reprogrammes the plant transcriptome extensively (Figures 1a and S1a–c), and the over-represented down-regulated genes, shared by NahG and sid2, are enriched in RLK/RLP and other immune receptors and signalling components (Figure S1d–f). This is reminiscent of a recent study that overexpressing the SA-inactivating hydroxylases DOWNY MILDEW-RESISTANT6 (DMR6) or DMR6-LIKE OXYGENASE1 strongly reduces the expression of immune receptors and signalling genes (van Butselaar et al., 2024), suggesting the essential role of the basal level of SA in maintaining the expression of immunity-related genes in unchallenged conditions. It seems that the expression of either a NahG or DMR6 to hydrolyse SA has a much more profound effect on transcriptomic reprogramming than only the inactivation of ICS1. Plants have both PAL and ICS pathways for SA biosynthesis; disabling either would promote the other to compensate for SA accumulation, which somehow explains that sid2 has a mild immunity-related phenotype compared to NahG (Figures 1 and S1 and S2); however, SA hydroxylases degrade overall SA synthesized from different biosynthesis pathways. NahG catabolizes SA into catechol, which regulates cellular redox homeostasis and might affect plant immunity (Van Wees and Glazebrook, 2003). In the root of Nicotiana attenuata, catechol treatment alters plant redox homeostasis; still, it does not affect H2O2 content (Wang et al., 2017), which is the major form of ROS measured using the luminol-based chemiluminescence assay (Figure 1c–e and S2a–c). Moreover, chitin treatment activating more ROS production in NahG plants suggests that NahG expression or catechol does not affect cellular redox homeostasis (Figure 1e and S2c). The observations of chitin-treated immune responses are somehow unexpected (Figure 1e,i,l), as chitin activates ROS production but cannot inhibit seedling growth or restrict bacterial growth in WT; even it promotes ROS production and seedling growth in NahG plants, suggesting a difference in downstream PTI responses. The authors thank Alberto P. Macho for helpful discussions and for sharing biological materials. This work is supported by the National Natural Science Foundation of China (32270284 to GY). YW is partially supported by the China Postdoctoral Science Foundation (2024M751987). GY designed experiments. YW, YO and LJ performed experiments and analysed data. JX analysed the RNA-Seq data. GY wrote the manuscript with inputs from all the authors. The authors declare no competing interests. Figure S1 Differentially expressed gene (DEG) analysis with NahG-transgenic and sid2-2 mutant Arabidopsis plants. Figure S2 Reduced immune responses in NahG-transgenic and sid2-1 mutant Arabidopsis plants. Figure S3 PCR primers for qRT-PCR. Data S1 Plant materials and growth. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.