Roles of microbiota in autoimmunity in Arabidopsis leaves

Cheng, Yu Ti; Thireault, Caitlin A.; Zhang, Li; Paasch, Bradley C.; Sohrabi, Reza; He, Sheng Yang

doi:10.1038/s41477-024-01779-9

Download PDF

Article
Open access
Published: 06 September 2024

Roles of microbiota in autoimmunity in Arabidopsis leaves

Nature Plants volume 10, pages 1363–1376 (2024)Cite this article

3024 Accesses
84 Altmetric
Metrics details

Subjects

Abstract

Over the past three decades, researchers have isolated plant mutants that show constitutively activated defence responses in the absence of pathogen infection. These mutants are called autoimmune mutants and are typically dwarf and/or bearing chlorotic/necrotic lesions. Here, from a genetic screen for Arabidopsis genes involved in maintaining a normal leaf microbiota, we identified TIP GROWTH DEFECTIVE 1 (TIP1), which encodes an S-acyltransferase, as a key player in guarding leaves against abnormal microbiota level and composition under high-humidity conditions. The tip1 mutant has several characteristic phenotypes of classical autoimmune mutants, including a dwarf stature, showing lesions, and having a high basal level of defence gene expression. Gnotobiotic experiments revealed that the autoimmune phenotypes of the tip1 mutant are largely dependent on the presence of microbiota as axenic tip1 plants have markedly reduced autoimmune phenotypes. We found that the microbiota dependency of autoimmune phenotypes is shared by several ‘lesion mimic’-type autoimmune mutants in Arabidopsis. It is worth noting that autoimmune phenotypes caused by mutations in two Nucleotide-Binding, Leucine-Rich Repeat (NLR) genes do not require the presence of microbiota and can even be partially alleviated by microbiota. Our results therefore suggest the existence of at least two classes of autoimmunity (microbiota-dependent versus microbiota-independent) in plants. The observed interplay between autoimmunity and microbiota in the lesion mimic class of autoimmunity is reminiscent of the interactions between autoimmunity and dysbiosis in the animal kingdom. These parallels highlight the intricate relationship between host immunity and microbial communities across various biological systems.

A critical role of a eubiotic microbiota in gating proper immunocompetence in Arabidopsis

Article Open access 17 August 2023

The plant NADPH oxidase RBOHD is required for microbiota homeostasis in leaves

Article 30 June 2021

PSKR1 balances the plant growth–defence trade-off in the rhizosphere microbiome

Article 16 November 2023

Main

In the past 40 years, tremendous progress has been made in the understanding of plant immune responses against pathogens^1,2,3. The plant innate immune system comprises both constitutive physical barriers and inducible immune responses. Inducible immunity can be initiated by plasma membrane-residing receptor kinases that recognize conserved microbe-associated molecular patterns, resulting in a broad-spectrum basal immunity called pattern-triggered immunity (PTI). Successful pathogens have evolved virulence-associated effector molecules to defeat the plant immune system and/or to create a conducive microenvironment within the host tissue as two major pathogenic mechanisms^4,5,6,7. In response, plants have evolved an array of intracellular nucleotide-binding leucine-rich repeat (NLR) immune receptors that recognize the presence of specific pathogen effectors, leading to effector-triggered immunity that often involves hypersensitive cell death^2,8. Activation and mutual potentiation of PTI and effector-triggered immunity lead to accumulation of defence hormones, including salicylic acid, activation of defence gene expression and other cellular responses^8,9. Activation of plant immune response is often accompanied by growth inhibition, a phenomenon known as growth–defence trade-offs^10,11.

In nature, plants spend most of their life in an environment that is occupied by enormously diverse, mostly non-pathogenic (commensal) microorganisms^12,13. Plant-associated commensal microbial community plays a vital role in influencing host growth, development and stress responses^14,15. However, compared with plant–pathogenic microbe interactions, less is known about how plants recognize and communicate with their surrounding non-pathogenic microbial communities and how plants fine tune their immune system to achieve a long-term, harmonious state in the context of complex microbial communities. Only in the past decade, thanks to the advent of high-throughput sequencing technologies and availability of genetic mutants in model systems, such as Arabidopsis and rice, increasing efforts are being devoted to the study of the interplay between plant host genetics and associated microbial communities^12,14,16. Nevertheless, the mechanisms of plant–microbiome interactions in terms of (1) how plants recognize, configure and maintain a homeostatic composition of their associated microbiota and (2) how the host immune system distinguishes non-self signals derived from commensal versus pathogenic microbes are still largely unclear.

The lifestyle of commensal bacterial microbiota in plant leaves resembles those of non-pathogenic strains of phyllosphere bacterial phytopathogens: both are adapted to live in plant tissues but are unable to multiply to a high population level in planta. Recent studies have shown that the transcriptomes of commensal bacteria in planta share a high degree of similarity to that of a non-pathogenic mutant of the phyllosphere-adapted bacterial pathogen Pseudomonas syringae pv. tomato (Pst) DC3000^17,18. Furthermore, two Arabidopsis quadruple mutants, min7 fls2 efr cerk1 (mfec) and min7 bak1 bkk1 cerk1 (mbbc), which are defective in PTI and MIN7-associated intracellular vesicle trafficking, not only fail to control the proliferation of non-pathogenic mutants of Pst DC3000 but also are unable to maintain a typical endophytic leaf microbiota^19,20. In addition, immunity-associated reactive oxygen species (ROS) is required for maintaining a homeostatic leaf microbiota as the Arabidopsis rbohD mutant, which is defective in the generation of immunity-associated ROS, has altered leaf microbiota composition²¹. Taken together, these initial studies begin to identify plant genes/pathways that are required for maintaining a normal bacterial microbiota in Arabidopsis leaves and provide a strong link between plant immune regulation and microbiota homeostasis¹⁵.

To identify additional plant genes involved in regulating plant–microbiome interactions and microbiota homeostasis, we conducted a forward genetic screen to isolate Arabidopsis mutants that show an altered response to non-pathogenic mutants of Pst DC3000 and endophytic leaf microbiota. Characterization of the resulting mutants led to an unexpected broad connection between microbiota and autoimmunity in plants.

Results

Identifying Arabidopsis mutants with altered leaf microbiota

We previously found that Arabidopsis mutant plants compromised in three pattern-recognition co-receptor genes, BAK1, BKK1 and CERK1, are still capable of preventing over-proliferation of non-pathogenic strain Pst D28E¹⁹ and endogenous leaf microbiota²⁰, suggesting that genes whose functions are either independent of and/or additive to these three pattern-recognition co-receptors are involved in plant interactions with non-pathogenic microbes. We therefore conducted a forward genetic screen in the bak1-5 bkk1-1 cerk1-2 triple mutant background^22,23 (bbc hereafter), intending to identify such genes (see Supplementary Fig. 1 for the setup of the genetic screen).

We carried out the primary screen by flood-inoculating 3-week-old plate-grown M2 seedlings with Pst D28E²⁴, a non-pathogenic mutant strain which was constructed by deleting 28 effector genes of Pst DC3000. Individual plants that showed disease-like symptoms (for example, necrosis or chlorosis) were transplanted to soil and grown for seeds. Secondary screen of putative mutants was conducted by monitoring disease-like symptoms after syringe infiltration of non-pathogenic Pst ΔhrcC mutant strain²⁵ (Pst DC3000 defective in type III secretion of all effectors, including the 28 effectors absent in the Pst D28E mutant strain) into leaves of 4-week-old soil-grown M3 plants. Next, we monitored disease-like symptoms induced by natural soil-derived microbiota under holoxenic conditions in a peat-based gnotobiotic system²⁶. In total, we identified ten mutants with various degrees of enhanced disease-like symptoms in response to non-pathogenic Pst DC3000 strains and/or natural soil-derived microbiota; we named them guardian of normal microbiota (grm) mutants (see Extended Data Fig. 1 for the morphology of soil-grown grm mutants).

Characterization of the grm1 mutant

Next, we conducted detailed characterization of one of the identified grm mutants, grm1. When grown in soil, the grm1 mutant had a smaller stature compared with its progenitor, the bbc triple mutant. Notably, lower leaves that were in contact or in close proximity to the soil showed mild chlorosis (Fig. 1a). Previously, we found that mfec and mbbc mutant plants showed dysbiotic endophytic leaf microbiota and leaf tissue damage when grown under high humidity, a common environmental condition associated with plant disease outbreaks in nature^19,20. The chlorotic lower leaves of the grm1 mutant resembled those of mfec and mbbc mutant plants and prompted us to investigate whether grm1 also shows dysbiotic endophytic leaf microbiota under high humidity. As expected in wild-type Arabidopsis (accession Col-0, which is the progenitor of the bbc triple mutant), characteristic hyponastic changes in leaf morphology were induced after 5 days of high-humidity treatment (~95% relative humidity (RH); Fig. 1a, bottom panel); however, high humidity did not cause over-proliferation of endophytic leaf microbiota, and no leaf chlorosis nor necrosis was observed (Fig. 1b). Like Col-0 plants, the bbc mutant also maintained a low level of culturable endophytic leaf microbiota, similar to these plants under ambient humidity (~50% RH). By contrast, after 5 days of high-humidity treatment, most of the grm1 true leaves showed strong chlorosis (Fig. 1a, bottom panel). Quantification of culturable endophytic microbiota loads by colony counts on R2A agar plates revealed that, compared with Col-0 and the bbc mutant, the endophytic microbiota within leaves of grm1 plants grown under high humidity was more than three orders of magnitude higher (Fig. 1b). In addition to a drastic increase in leaf endophytic microbiota, the relative abundance of leaf microbiota members in the grm1 mutant also shifted overwhelmingly to Proteobacteria (Fig. 1c, ~97% in grm1 leaves compared with ~45–55% in Col-0 and bbc leaves). In the grm1 mutant, amplicon sequence variants (ASVs) in Proteobacteria belong predominantly to the genus Pseudomonas of the class Gammaproteobacteria, while Bacillus and Paenibacillus belonging to Firmicutes became nearly undetectable (Fig. 1c and Supplementary Data 1). Shannon index that measures richness and evenness of a microbial community composition also decreased in the grm1 mutant, reflecting the overwhelming presence of Proteobacteria (Fig. 1d).

**Fig. 1: The appearance and leaf microbiota phenotypes of the *grm1* mutant.**

Identification of the causal mutation in the grm1 mutant

To identify the causative mutation in the grm1 mutant, grm1 plants were backcrossed with its progenitor, the bbc triple mutant, to generate a segregated F2 population. Analysing the mapping-by-sequencing data from grm1 co-segregates revealed that the mutation was located on chromosome 5 between 5 Mb and 8 Mb with the allele frequency of the grm1-like pool peaking around 7 Mb (Extended Data Fig. 2a; see Supplementary Table 1 for a list of mutated loci in this region). Of all candidates, a G-to-A mutation on chromosome 5 at position 6,877,509 was tightly associated with the grm1 phenotype and has the strongest effect. This G-to-A mutation occurs at the splicing junction of the 5′ end of the third intron of the TIP GROWTH DEFECTIVE 1 (TIP1) gene and is expected to disrupt the splicing pattern, leading to a premature stop codon instead of tryptophan at position 171 (Fig. 2a and Extended Data Fig. 3a). PCR using grm1 complementary DNA as template and primers flanking the mutation locus followed by Sanger sequencing revealed that, indeed, the mutation altered the splicing pattern of the TIP1 transcript (Extended Data Fig. 2b). Transgene complementation with the full-length TIP1 gene driven by its native promoter could complement grm1 mutant phenotypes (that is, reversion of the dwarf stature and humidity-dependent dysbiotic phenotypes to those in bbc and wild-type Col-0), confirming the causative mutation in grm1 is in the TIP1 gene (Extended Data Figs. 2c,d).

**Fig. 2: Characterization of *tip1* single mutant plants.**

To determine whether the dysbiotic phenotypes of the grm1 mutant (that is, the bbc tip1 quadruple mutant) are dependent on the background bbc mutations, we segregated the tip1 mutation from the bbc triple mutations by outcrossing the grm1 mutant with wild-type Col-0 plants and genotyping the resulting F2 population (see Supplementary Table 2 for primes used for genotyping). From The Arabidopsis Biological Resource Center²⁷, we also obtained two independent tip1 single mutant alleles (SALK_020996 and SALK_052842) carrying transfer DNA (T-DNA) insertions in the TIP1 gene (Fig. 2a). All three tip1 single mutants were larger than the original grm1 (bbc tip1) mutant but still smaller than wild-type Col-0 plants (Fig. 2b). It is worth noting that the humidity-dependent dysbiosis phenotypes (for example, leaf chlorosis) (Extended Data Fig. 3b) and over-proliferation of endophytic leaf microbiota were still observed in all three alleles of tip1 single mutant plants, suggesting dysbiotic phenotypes observed in grm1 do not require bbc triple mutations in the background (Fig. 2c). As all three mutant alleles of tip1 behaved similarly, we used tip1^W171STOP (tip1 hereafter), the single mutant allele isolated from this screen, for all subsequent experiments.

The tip1 mutant has features of autoimmune mutants

We noticed that the morphological phenotypes of grm1 and tip1 (that is, small statures and chlorotic leaves) were reminiscent of typical autoimmune mutants, which have been isolated in the past few decades. One hallmark of autoimmunity is constitutive high basal expression of immune-related marker genes in the absence of pathogen attacks. We therefore analysed two immune-related molecular markers, Pathogenesis-Related Gene 1 (PR1) and flg22-Induced Receptor-like Kinase 1 (FRK1). Indeed, both grm1 and tip1 plants have heightened PR1 and FRK1 expression under basal condition (that is, in the absence of pathogen inoculation) (Figs. 2d,e). It is worth noting that TIP1 gene expression itself can be induced by flg22²⁸, a flagellin-derived peptide that induces many PTI-associated genes^29,30 (Extended Data Fig. 4). This suggests that TIP1 is a flg22 responsive gene that is involved in maintaining leaf microbiota homeostasis.

tip1- and snc1-mediated autoimmunity are distinct

We were intrigued by the morphological phenotypes and heightened immune-related marker gene expression of the tip1 mutant as they point to a connection between dysbiosis and autoimmunity. However, it is not known (1) whether all autoimmune mutants possess a defect in maintaining microbiota homeostasis and/or (2) whether their autoimmune phenotype is dependent on microbiota. To investigate on a possible connection between altered microbiome and autoimmunity, we first examined microbiota-related phenotypes of a widely studied Arabidopsis autoimmune mutant, snc1, which contains a gain-of-function E552K mutation that results in an elevated level of the SNC1^E552K protein^31,32.

As expected, snc1 mutant plants had a dwarf morphology (Fig. 4a, top panel) and constitutively elevated immune-related marker gene expression (Fig. 3a,b). Both snc1 and tip1 showed enhanced resistance to the virulent pathogen Pst DC3000 (Fig. 3c). In addition, tip1 leaves showed significantly higher levels of PTI-associated ROS production, as induced by flg22, elf18³³ (an EF-Tu-derived peptide) and AtPep1³⁴ (an endogenous damage-associated peptide) compared with snc1 and wild-type Col-0 leaves (Extended Data Fig. 5). However, snc1 and tip1 mutant plants behaved differently in response to non-pathogenic bacteria. As shown in Fig. 3d, tip1 single mutant plants were more susceptible to the non-pathogenic Pst ΔhrcC mutant strain, whereas snc1 plants were marginally more resistant to the Pst ΔhrcC mutant compared with wild-type Col-0 plants.

**Fig. 3: The autoimmune phenotypes of *tip1* and *snc1* mutants.**

Furthermore, while most tip1 leaves showed severe chlorotic lesions when shifted to high-humidity condition for 5 days, snc1 mutant plants were morphologically similar under either ambient or high-humidity conditions (Fig. 4a). In addition, enumeration of leaf endophytic microbiota showed that snc1 plants carried similar levels of culturable bacteria inside their leaves as that of Col-0 plants, whereas tip1 plants had more than 1,000-fold increase of endophytic bacterial load compared with Col-0 after higher humidity shift (Fig. 4b). In addition to Arabidopsis potting soil mix, different leaf endophytic microbiota levels between tip1 and snc1 mutants were observed across diverse soil types from Michigan and North Carolina (Extended Data Fig. 6), suggesting that the dysbiotic phenotypes associated with the tip1 mutant plants were robust and irrespective of soil sources and/or the microbiota associated with these soils. Along with variances in the microbiota level, the leaf microbiota compositions in tip1 and snc1 plants were also different. Profiling bacterial communities inside Col-0, tip1 and snc1 leaves using 16S ribosomal RNA gene amplicon sequencing revealed that, compared with Col-0, tip1 plants had substantially reduced leaf endophytic microbiota diversity (Fig. 4c) with overwhelming high relative abundance of Gammaproteobacteria (Fig. 4d and Supplementary Data 2). By contrast, snc1 plants had a diverse leaf endophytic microbiota composition, similar to that of Col-0 plants (Figs. 4c,d). Finally, when grown in aseptic agar plates, snc1 plants continued to show heightened immune-related marker gene expression in the absence of microbes, whereas PR1 and FRK1 expression in tip1 mutant plants greatly subsided to close to the low levels observed in wild-type Col-0 plants (Figs. 4e,f).

**Fig. 4: The distinct autoimmune phenotypes of *tip1* and *snc1* mutants.**

Next, we retrieved bulk culturable endophytic bacteria from tip1 leaves following a previous study³⁵. These tip1-associated endophytic leaf bacteria, in bulk, when infiltrated into leaves caused disease-like chlorosis in wild-type Col-0 plants (Extended Data Fig. 7), suggesting that microbiota assembled in the tip1 mutant can cause dysbiotic plant phenotypes.

Role of microbiota in Arabidopsis lesion-mimic autoimmunity

The interesting contrast in microbiota dependency for autoimmune phenotypes between the tip1 and snc1 mutants prompted us to further investigate whether there is a broad connection between microbiota dependency and autoimmunity in other reported Arabidopsis autoimmune mutants^36,37 (see Supplementary Table 3 for list of mutants and their stock numbers). Based on whether they possessed tissue lesions when grown in potting soil under standard growth chamber conditions, we found that the tip1-like autoimmune mutant category includes (1) the aca4 aca11 double mutant, which harbours mutations on two vacuolar calcium ion pumps ACA4 and ACA11 (ref. ³⁸), (2) the acd5 mutant, which carries mutation in a ceramide kinase^39,40 and (3) the lsd1 mutant, which has a defective zinc-finger protein in Arabidopsis⁴¹. The snc1-like autoimmune mutant category consists of (1) the chs3 mutant, which carries a gain-of-function mutation in a Toll/interleukin-1 receptor (TIR)-type NLR immune receptor⁴², (2) dnd1 (ref. ⁴³) and (3) dnd2 (ref. ⁴⁴) mutants, which carry mutations in two cyclic nucleotide-gated cation channels. Mutant plants in the tip1-like category showed severe leaf lesions (Fig. 5a, top panel) and harboured high levels of endophytic leaf microbiota under high humidity (Fig. 5b). Principal coordinates analysis of weighted UniFrac distances from 16S rRNA gene sequencing of endophytic leaf microbiota revealed that bacterial communities associated with the tip1-like mutants mostly clustered together and apart from those associated with wild-type Col-0 plants (Extended Data Fig. 8a). In addition, bacterial communities associated with most tip1-like mutants showed a significant reduction in alpha diversity, as evidenced by lower richness compared with Col-0 (Extended Data Fig. 8b). Albeit to different degrees, mutant plants of the tip1-like category carried higher levels of Gammaproteobacteria, most notably two genera, Pseudomonas and Stenotrophomonas, than Col-0 plants (Extended Data Fig. 8c and Supplementary Data 3). Mutant plants in the snc1-like category had no visible lesions (Fig. 5a, bottom panel) and carried low levels of endophytic leaf microbiota similar to wild-type Col-0 plants (Fig. 5c).

**Fig. 5: The appearance and leaf microbiota phenotypes of *Arabidopsis* autoimmune mutants.**

It is worth noting that leaf endophytic microbiota profiling showed that tip1 mutant plant samples collected at Michigan State University (MSU) carried one dominant ASV (an unspecified species in Pseudomonas sp; Supplementary Data 2), whereas tip1 plant samples collected at Duke University showed three dominant ASVs (Pseudomonas nitroreducens, an unspecified species in Pseudomonas and an unspecified species Stenotrophomonas; Supplementary Data 3). We observed dysbiosis at both locations, suggesting that the dysbiotic phenotypes in tip1 mutant plants are robust, albeit with somewhat different enriched ASVs.

To further characterize a possible microbiota dependency of autoimmune phenotypes in these two categories of mutants, we grew mutant plants in the absence (axenic) or presence (holoxenic) of a natural ‘MSU’ soil-derived microbiota using GnotoPots, a peat-based gnotobiotic system as described²⁶. Growing under the holoxenic condition, lesion-mimic mutants showed various degrees of chlorosis and lesions (Fig. 6a, top panel, and Supplementary Fig. 2a) and heightened immune-related marker gene expression (Fig. 6c, right panel). However, in the absence of microbiota, these lesion-mimic autoimmune mutants showed neither chlorosis nor lesions (Fig. 6a, bottom panel), and their immune marker gene expression also subsided to a low basal level, with the exception of lsd1 (Fig. 6c, left panel). By contrast, mutants in the snc1-like category have high basal PR1 expression even in the axenic condition. For example, the chs3 mutant showed heightened PR1 expression regardless of the presence or absence of microbiota (Fig. 6d), behaving similarly to the snc1 mutant, which shows microbiota-independent autoimmunity. Furthermore, compared with microbiota-induced lesions and immune-related gene expression in lesion mimic mutants, the autoimmune dwarf phenotype of chs3 mutants was noticeably alleviated in the presence of microbiota (Fig. 6b), again similar to the snc1 mutant. In dnd1 and dnd2 mutants, PR1 expression was elevated to a higher level when they were grown in the presence of microbiota compared with when grown in the axenic condition (Fig. 6d), behaving intermediately between lesion mimic type and snc1 type. Together, these results suggest that there are at least two types of autoimmunity in plants: one depends on microbiota for autoimmune phenotypes, as exemplified by the tip1 mutant, and the other is independent of microbiota, as exemplified by the snc1 mutant. dnd1 and dnd2 mutants share with snc1-type in that they do not show lesions in the presence of microbiota and have a high basal defence gene expression in the absence of microbiota, although defence gene expression is further enhanced in the presence of microbiota.

**Fig. 6: Microbiota dependency for autoimmunity in *Arabidopsis* autoimmune mutants.**

To capture genome-wide gene expression in tip1 and snc1 plants beyond the PR1 and FRK1 marker genes and to further characterize gene expression patterns associated with the distinct microbiota dependency of autoimmune phenotypes in these two mutants, we conducted transcriptomic profiling of these two mutants grown in the absence (axenic) or presence (holoxenic) of a natural ‘MSU’ soil-derived microbiota in GnotoPots. Heat-killed ‘MSU’ soil-derived microbial communities were included to investigate whether autoimmune phenotypes could be activated by microbe-associated molecular patterns associated with heat-killed microbes. Principal component analysis of gene expression data revealed distinct expression patterns in Col-0, tip1 and snc1 plants (Extended Data Fig. 9a). The respective gene expression patterns in Col-0 and tip1 plants under axenic and heat-killed conditions clustered together, suggesting that in the absence of live microbes, tip1 mutant plants behaved similarly as wild-type Col-0 and that heat-killed microbes were not sufficient to alter gene expression patterns in Col-0 and tip1 plants (Extended Data Fig. 9a). Under the holoxenic condition, gene expression pattern in tip1 plants is distinct from that under the axenic condition or Col-0 plants under holoxenic condition, which is consistent with the dramatic microbiota-dependent autoimmunity phenotype. By contrast, gene expression patterns in snc1 plants under all three growth conditions clustered closely and away from Col-0 and tip1 gene expression patterns, which is consistent with the microbiota-independent autoimmune phenotype in the snc1 mutant.

Furthermore, in accordance with increased PR1 and FRK1 marker gene expression (Figs. 3a,b), PTI/salicylic acid-associated immune genes were induced, to various degrees, in tip1 as well as snc1 plants under the holoxenic condition (Extended Data Fig. 9b). It is worth noting that jasmonate-associated genes appear to be uniquely enriched in tip1 plants but not in snc1 plants under the same condition probably related to tissue damage in the tip1 mutant (Extended Data Fig. 9c).

Role of microbiota in autoimmunity in wild accessions

Autoimmunity has been observed in natural Arabidopsis populations/accessions⁴⁵. For example, Arabidopsis thaliana accessions Est-1 and C24 have constitutively elevated defence gene expression and enhanced disease resistance toward the virulent pathogen Pst DC3000 when grown in potting soil^46,47. We were therefore interested in knowing whether autoimmune phenotypes of natural accessions are dependent on microbiota. When grown in potting soil under ambient humidity, Est-1 showed chlorosis and lesions on older leaves, whereas C24 had curly leaves and small stature but did not show chlorosis or lesions (Supplementary Fig. 3). However, like the tip1 mutant, Est-1 leaves showed stronger leaf lesions under high humidity (Fig. 7a) and harboured a higher level of endophytic bacterial microbiota compared with Col-0 plants (Fig. 7b). By contrast, C24 plants did not show chlorosis nor necrosis under ambient (Supplementary Fig. 3) or high humidity (Fig. 7a); in addition, similarly to the snc1 mutant, C24 plants maintained similar levels of endophytic leaf bacterial microbiota compared with Col-0 (Fig. 7b). The phenotypic resemblance between tip1 and Est-1 and between snc1 and C24 prompted us to investigate whether microbiota is required for the autoimmune phenotypes in Est-1 and C24. As shown in Fig. 7c, under holoxenic condition, Est-1 shows lesions on leaves, albeit to a lesser extent compared with Est-1 grown under the conventional potting soil growth condition (Supplementary Fig. 3 and Supplementary Fig. 2b). However, like tip1, Est-1 plants did not show leaf lesions under the axenic condition. Furthermore, like the tip1 mutant, the heightened PR1 expression in Est-1 subsided to a low level when grown in the axenic condition (Fig. 7d, left plot). Conversely, like the snc1 mutant, C24 plants had elevated PR1 expression regardless of growth in the presence or absence of a microbial community (Fig. 7d). Another similarity between the snc1 mutant and C24, which is in contrast to the tip1 mutant, is the alleviation of their stunted growth morphology in the presence of microbiota (Figs. 6b and 7c).

**Fig. 7: Microbiota dependency for autoimmunity in *A. thaliana* natural accessions.**

Discussion

A healthy microbiome can play a vital role in initiating, training and maintaining host immune homeostasis. In return, the host immune system can fine tune its immune strength to accommodate commensal/symbiotic microbes and to prepare for a robust immune response against pathogenic microbe invasion. As plants spend most of their life interacting with a vast number of commensal microbes and occasionally encountering pathogens, understanding the intricate interplays between plant immunity and the endophytic commensal microbiota is important for explaining how plants dial their plant immune system to maximize the effectiveness of plant immune responses to nurture beneficial microbes and/or fight against pathogens. In this study, we conducted a forward genetic screen aimed at identifying Arabidopsis mutants that cannot maintain a normal leaf microbiota. Among putative mutants isolated was grm1, which we characterized in detail.

The grm1 mutant contains a missense mutation in the TIP1 gene that encodes S-acyltransferase. The first mutant allele of TIP1 was isolated in a genetic screen for mutants that had defects in root hair development⁴⁸. The causal mutation was later mapped to At5g20350 (ref. ⁴⁹). At5g20350/TIP1 encodes one of the 24 DHHC (Asp-His-His-Cys)-containing S-acyltransferases and it is also one of the two ankyrin repeats containing DHHC S-acyltransferases in Arabidopsis. Ankyrin repeats containing DHHC S-acyltransferases are conserved across eukaryotes (Supplementary Fig. 4 and Supplementary Table 4). Both the Arabidopsis TIP1 protein and the human HIP14 (huntingtin-interacting protein 14; zinc finger DHHC-type palmitoyltransferase 17 (zDHHC17)) have been shown to be functional orthologues of yeast Akr1p^49,50. Akr1p and zDHHC17 are involved in vesicle trafficking. For example, acylation of yeast Yck2 protein by Akr1p is required for proper localization of Yck2 to the plasma membrane via secretory vesicles, and Yck2p’s membrane association is essential for its biological function in yeast morphogenesis⁵¹.

Before this study, however, the connection between TIP1 and leaf microbiota homeostasis was not known. A comprehensive Arabidopsis acylome using multiple tissue types identified close to 1,100 putative S-acylated proteins⁵². Thirty-seven per cent of identified proteins overlapped with those identified in a previous study which used Arabidopsis root cell culture⁵³. Notably, many of the identified proteins have been demonstrated to be associated with microbe perception and plant immune responses⁵³, including FLS2 and EFR receptors^54,55,56. While abolishing predicted acylation sites on these receptor-like kinases attenuated their immune responses, it is not known whether they are substrates of TIP1. Our data showed that ROS response activated by flg22, elf18 and AtPep1, which are perceived FLS2 (ref. ⁵⁷), EFR⁵⁸ and AtPEPR1 (ref. ³⁴), respectively, were not compromised in tip1 plants (Extended Data Fig. 5). This suggested that FLS2, EFR and AtPEPR1 are not likely the direct substrates of TIP1 S-acyltransferase. It is worth noting that TIP1 gene expression itself can be induced by flg22 regardless of the presence of the microbiota (Extended Data Fig. 4). In light of our findings on a genetic connection between TIP1 and maintenance of a normal leaf microbiota, future research is needed to identify specific TIP1 substrate(s) that is required for microbiota homeostasis in Arabidopsis leaves.

A key finding in this study is that the tip1 mutants not only are unable to control the proliferation or maintain a normal composition of a leaf microbiota but also show dysbiosis-associated tissue damages and autoimmunity in the presence of microbiota (Fig. 4). The microbiota-dependent autoimmune phenotypes of the tip1 mutant led us to broadly examine a potential connection between microbiota and previously reported ‘autoimmune’ mutants in Arabidopsis. Based on how they respond to the existence of microbiota, it appears that autoimmune mutants in Arabidopsis can be divided into at least two classes (Supplementary Fig. 5). One class, exemplified by the tip1 mutant, shows microbiome-dependent autoimmunity in leaves. The autoimmune phenotypes in this class largely disappeared when grown in the axenic conditions. Given that these mutants also have an increased level and altered composition of endophytic leaf microbiota, this result suggests that the autoimmune phenotypes in this class of mutants are a consequence of harbouring an overabundant and/or aberrant microbial community. The other class of autoimmunity in our study is independent of microbiota and is represented by the snc1 mutant. The autoimmune phenotypes of this class do not require the presence of microbiota. That is, they have small statures and high PR1 expression regardless of the presence or absence of microbial communities. It is worth noting that the presence of microbiota alleviates the stunted growth morphology of snc1 and chs3 (Fig. 6b), which is in striking contrast to those of the tip1 class (Fig. 6a). TIP1 and other autoimmune-causing genes are themselves differentially expressed in Col-0 leaves under axenic, heat-killed, holoxenic conditions, albeit with no consistent patterns (Supplementary Fig. 6).

While the autoimmune phenotypes of snc1 and chs3 mutants are independent of microbiota and their stunted growth morphology can be partially alleviated by the presence of microbiota, not all NLR autoimmune mutants may behave in a similar fashion. For example, in Arabidopsis No-0 ecotype, which carries mutations in SLH1 (Sensitive to Low Humidity 1), a TIR-type NLR with a WRKY domain at its C-terminus showed normal growth on agar plate but showed growth arrest and development of necrotic lesions when transferred to a microbe-rich soil environment⁵⁹. However, in this case, it remains to be determined whether the autoimmune phenotype differences of No-0 plants between growth on agar plate and growth in soil is indeed caused by microbiota. In addition, observations of autoimmune phenotypes attributed to NLR proteins were made in hybrid necrosis⁶⁰, a term to describe the phenomenon of incompatibility in the F1 progeny from interspecific or intraspecific crosses. In at least one case, this hybrid necrosis/hybrid autoimmunity caused by NLR genes is independent of microbiota⁶¹. Finally, a recent study in root-associated microbiota showed that mutations in the Arabidopsis PSKR1 (Phytosulfokin Receptor 1) gene led to autoimmunity. It is worth noting that the autoimmunity phenotype in this case is associated with reduced Pseudomonads in the rhizosphere⁶². It will be of interest in the future to determine whether there are microbiota changes in the leaves of the pskr1 mutant.

We find it interesting that microbiota-dependent and -independent autoimmunity can also be observed in Arabidopsis natural accessions. Although autoimmunity is often associated with fitness trade-offs at the individual plant level^10,11, the presence of heightened basal immunity in natural accessions suggests a fitness advantage at the population level⁴⁷. For example, if a devastating disease spreads through a largely susceptible Arabidopsis population, accessions such as C24 and Est-1 may be able to survive and reproduce to avoid extinction of the entire population. The microbiota-dependent and -independent expression of autoimmunity in Est-1 versus C24, as observed in our study (Fig. 7), may reflect different paths by which the two types of autoimmunity have convergently evolved in natural populations under different abiotic and biotic pressures. Future research is needed to determine whether the TIP1-dependent pathway plays a role in natural accessions that show microbiota-dependent autoimmunity.

Overall, results from this study begin to illustrate conceptual parallels in microbiota interactions with plants and animals and may have broad implications in understanding host–microbiota interactions in general. In mammalian–microbiota interactions, for example, dysbiotic shifting/reducing diversity in microbiome composition is often associated with inflammation and dysregulated immune responses that fail to distinguish self from non-self, which are characteristics of autoimmune disorders^63,64,65. The similarities in autoimmune symptoms between the Arabidopsis tip1 mutant and mammalian inflammatory autoimmunity are notable that they include a dysbiotic microbial community, tissue lesions and dysregulated immune responses. The most renowned substrate of zDHHC17, the human homologue of Arabidopsis TIP1, is huntingtin^50,66. In light of the connection between the tip1 mutation and autoimmunity in Arabidopsis, it would be interesting for future research to examine whether mutations in zDHHC17 or other TIP1 orthologues are associated with dysbiosis and/or whether dysbiosis is involved in Huntington’s disease development. Indeed, recent reports have suggested an association between Huntington’s disease and dysbiosis⁶⁷. The advance in biochemistry studies of zDHHC17 (refs. ^68,69,70) and the more easily amenable mutant studies at the whole organismal level in Arabidopsis should facilitate further understanding of a possibly broad role of DHHC S-acyltransferases in microbiome homeostasis and immunity across the kingdoms of life.

Methods

Plant materials and growth conditions

All seeds were surface-sterilized using 15% diluted bleach (containing final concentration of 1.2% active ingredient sodium hypochlorite (NaOCl)) before being sown onto potting soil. All plants were grown under a 12 h day/12 h night regimen with 100 µE light intensity and ~50% RH, unless otherwise indicated. See Supplementary Table 3 for a complete list of Arabidopsis mutants and accessions used in this study.

For plant images and microbiota quantification on plants grown on different soil types, soil was sourced from Miscanthus plot at MSU (42° 43′ 01.2″ N, 84° 27′ 45.8″ W)²⁶, Michigan Benton Harbor (42° 05′ 06.2″ N, 86° 21′ 12.1″ W) and North Carolina (36° 13′ 07.2″ N, 79° 10′ 45.5″ W). ‘Arabidopsis mix’ potting soil was prepared in-house by mixing equal volumes of SureMix (Michigan Grower Products), perlite and vermiculite. After preparing pots using different soil types, all pots were treated with NEMAforce SF for insect pest control before sowing surface-sterilized seeds.

For experiments using GnotoPots, a peat-based gnotobiotic system²⁶, nutrients were supplemented with buffered half-strength Linsmaier and Skoog liquid media (0.5× LS, pH 5.7; Caisson Labs LSP03). Soil for natural microbiota inoculation was collected from MSU. Holoxenic plants were inoculated with soil slurry (10 g soil per litre of 0.5× LS), whereas axenic plants were inoculated with 0.5× LS liquid medium. Heat-killed natural microbiota input was prepared by inoculating with an autoclaved soil slurry.

Genetic screen

Roughly 30,000 Arabidopsis bak1-5 bkk1-1 cerk1-2 (bbc) seeds were mutagenized using 0.2% ethyl methanesulfonate (EMS). Mutagenized M1 seeds were sown on soil and allowed to grow to set seeds. Seeds from two to three M1 plants were pooled, and approximately 1,700 pools were collected. This EMS population was estimated to cover the Arabidopsis genome more than 10 times. The primary screen was conducted by seedling flood-inoculation assay⁷¹. In brief, roughly 50 M2 seeds from each pool were sown onto 0.5× LS agar plates followed by flooding 3-week-old seedlings with 1 × 10⁸ c.f.u. ml⁻¹ Pst D28E (suspended in 2.5 mM MgCl₂ and 0.015% Silwet L-77) for 4 min. After 4 min, inoculum was removed, and plates were returned to Percival chambers for disease symptom development. Mutants showing signs of chlorosis and/or necrosis were transplanted to potting soil and transferred to a growth chamber to collect M3 seeds. Secondary screen was conducted by monitoring symptoms after either syringe infiltration of non-pathogenic Pst ΔhrcC mutant strain to 4-week-old soil-grown M3 plants or after growth in holoxenic condition in the FlowPot gnotobiotic plant growth system²⁶.

Mutation identification using mapping-by-sequencing approach

To identify the causative mutation in the grm1 mutant, a mapping-by-sequencing population was generated by backcrossing the grm1 mutant with the bbc mutant. All four F1 plants showed bbc-like morphology, suggesting that the mutant trait is recessive. F1 plants were selfed to produce F2 populations. Of 674 F2 plants screened, 633 had bbc-like morphology, and 41 were grm1-like. This ratio deviates from the 3:1 single nuclear gene inheritance pattern. However, mapping-by-sequencing data does not support the idea that the mutant phenotype in grm1 plants was caused by two or more unlinked loci as the allele frequency only peaks at around 7 Mb on chromosome 5 (Extended Data Fig. 2a); the G-to-A mutation in the TIP1 (At5g20350) gene is tightly associated with the grm1 phenotype (Extended Data Fig. 2 and Supplementary Table 1). A literature search found that a similar genetic inheritance pattern deviation was observed in the characterization of the first mutant allele of TIP1 (tip1-1; ref. ⁴⁸). The authors suggested that ‘deficiency in the number of homozygous tip1 mutant seeds’ as a potential cause of such inheritance ratio deviation.

Genetic complementation of the grm1 mutant

High-fidelity PCR was performed using Col-0 genomic DNA as template and with a sense primer covering roughly 2 kb upstream of the At5g20350 start codon and an anti-sense primer roughly 1 kb downstream of the stop codon. Cloned PCR product was inserted into pDONR207 entry vector and verified using Sanger sequencing. Recombination reaction was conducted using verified entry clone and pMDC123 destination vector to create the TIP1 genomic clone driven by the TIP1 native promoter. The construct was transformed into the grm1 mutant plants via floral dipping method⁷² using Agrobacterium tumefaciens GV3101 as the vehicle strain. See Supplementary Table 2 for the primer sequences used in this study.

Quantification of endophytic leaf bacterial microbiota

Four-week-old potting soil-grown plants were sprayed with distilled water and fully covered with a clear dome to maintain high humidity (~95% RH) for 5 days (or 7 days for natural Arabidopsis accessions). After high humidity treatment, one to three leaves from each plant were surface sterilized with diluted bleach (containing final concentration of 0.3% active ingredient NaOCl) for 1 min followed by two rinses with sterile water. Surface-sterilized leaves were blotted dry using paper towels and weighted. Sterile water was added to leaf samples, which were homogenized using TissueLyser II (QIAGEN) for 2 × 45 s at 30 Hz. Homogenized samples were serial diluted and spotted onto R-2A plates (Sigma-Aldrich catalogue number 17209) supplemented with cycloheximide (15 mg l⁻¹) and 0.5% methanol to enumerate culturable colonies.

The tip1 mutant plants showed most prominent dysbiosis phenotypes under high humidity; we therefore performed most of our dysbiosis-related experiments under high humidity.

Preparation of bacterial bulk culture

Lysates prepared from surface-sterilized leaf tissues (10 µl autoclaved Milli-Q water per mg of leaf fresh weight) from Col-0 and tip1 plants treated with high humidity (~95% RH) were mixed with glycerol to final concentration of 20%, aliquoted and stored in −80 °C.

Seven days before the experiment, one tube of Col-0 and one tube of tip1 leaf tissue lysate glycerol stock were taken out from the −80 °C, serial diluted and plated onto R-2A plates (supplemented with 15 mg l⁻¹ cycloheximide and 0.5% methanol). Plates were incubated at 10 °C. On the day of the experiment, plates containing mostly isolated colonies were used to prepare bacterial bulk culture (BC) as described³⁵ with modifications. To prepare bacterial BC, bacteria were scraped from the agar with a sterile loop and transferred to 15 ml tubes containing 2.5 mM MgCl₂; cells were pelleted down at 5,000 × g for 5 min and resuspended in 1 ml 2.5 mM MgCl₂. BC^Col-0 was diluted to optical density at 600 nm of 0.05 and BC^tip1 to 0.04, which gave ~5 × 10⁷ c.f.u. ml⁻¹.

Four-week-old soil-grown Col-0 plants were hand-infiltrated using a needleless syringe with either BC^Col-0 or BC^tip1. After infiltration, day 0 samples were collected, and the remaining infiltrated plants were returned to growth chambers and kept at high humidity (~95% RH) for symptom progression. Samples were collected 6 days post inoculation (d.p.i.).

16S rRNA gene amplicon sequencing

For this experiment, all consumables and kits came from the same lot to avoid any background contamination or variations. Seeds of indicated Arabidopsis genotypes were surface sterilized using 15% bleach (1.2% active ingredient NaOCl) and washed twice using autoclaved Milli-Q water before sowing onto potting soil. Plants were grown in growth chambers. Four-week-old plants were sprayed with distilled water and kept at ≥95% RH for 5 days. One to two leaves per plant were collected and surface-sterilized using diluted bleach (0.3% active ingredient NaOCl) for 1 min, followed by two washes using autoclaved Milli-Q water. Excess water on leaf surfaces was blotted dry, put into Safe-Lock Eppendorf tubes containing 3 mm zirconium beads (Glen Mills), snap froze in liquid nitrogen and stored at −80 °C.

Total DNA (host and microbes) was extracted using DNeasy PowerSoil Pro Kit (QIAGEN catalogue number 47014) following the manufacturer’s protocol. Extracted DNA was used as template for PCR amplification of the v5/v6 region of 16S rRNA gene using 799 F and 1,193 R primers (see Supplementary Table 2 for the sequence of primers) and high-fidelity AccuPrime Taq DNA Polymerase (Invitrogen catalogue number 12346086). Amplified products were run in 1% agarose gels to separate bacterial and chloroplast 16S rRNA gene amplicons (~400 bp) from the mitochondrial 18S amplicon (~750 bp). DNA in the ~400 bp band was recovered using the Zymoclean Gel DNA Recovery Kit (Zymo Research D4008). Concentration of recovered DNA was measured using the Quant-iT PicoGreen dsDNA Assay Kits (Invitrogen P7589) and normalized to 3–8 ng µl⁻¹ for sample submission. Library preparation and sequencing using MiSeq (Figs. 1 and 4) or NovaSeq (Extended Data Fig. 8) platforms (2 × 250 bp paired-end format) was conducted by the Genomic Core Facility at MSU.

16S rRNA gene amplicon sequencing data processing

Raw Illumina data for 16S rRNA gene amplicon sequencing were processed as described previously²⁰ using QIIME2 version 2022.2⁷³. In brief, primer sequences were removed using Cutadapt⁷⁴ followed by filtering, denoising and creating an ASV table by DADA2 (ref. ⁷⁵). For data from NovaSeq platform, ASVs contributing to less than 0.005% of the total or only present in a single sample were removed from analysis. Taxonomic assignment of each ASV was performed using a Naive Bayes classifier⁷⁶, pre-trained on the SILVA 16S rRNA gene reference database^77,78 (release 138; https://www.arb-silva.de/) formatted for QIIME using RESCRIPt⁷⁹. Unassigned sequences and sequences annotated as mitochondria and chloroplast were removed. Diversity analyses were performed within QIIME2. Samples were rarified to 1,952 and 74,693 reads for data from MiSeq and NovaSeq platforms, respectively, for calculating diversity metrics. The entire sequence analysis workflow is available on GitHub (https://github.com/BradCP/Roles-of-microbiota-in-autoimmunity-in-Arabidopsis).

Bacterial infection assays

Four-week-old soil-grown plants were hand-infiltrated using a needleless syringe with either Pst DC3000 at 1 to 2 × 10⁵ c.f.u. ml⁻¹ or Pst ΔhrcC at 2 to 3 × 10⁶ c.f.u. ml⁻¹. After infiltration, excess bacterial suspensions were blotted dry, and plants were returned to growth chambers and kept at high humidity (~95% RH) for disease progression. Samples were collected 3 d.p.i. for Pst DC3000 or 5 d.p.i. for Pst ΔhrcC. To determine bacterial population in leaves, leaf discs were collected and ground in autoclaved Milli-Q water using a TissueLyser II (QIAGEN; 45 s at 30 Hz). Serial dilutions of the ground tissue lysates were spotted onto low-salt Luria–Bertani plates (10 g l⁻¹ tryptone, 5 g l⁻¹ yeast extract and 5 g l⁻¹ NaCl; pH 7.0) with appropriate antibiotics. Colony-forming units per cm² were determined for each sample.

Reverse transcription quantitative PCR analysis gene expression

For gene expression analysis, plant tissues at the indicated conditions were collected, snap frozen in liquid N₂ and stored at −80°C until further processing.

Total RNA was extracted from plant tissues using TRIzol Reagent (Thermo Fisher catalogue number 15596026) according to the manufacturer’s instructions. Complementary DNA synthesis was accomplished in 10 µl volume with SuperScript IV VILO Master Mix (Thermo Fisher catalogue number 11756050) according to the manufacturer’s instructions with 1 µg total RNA as input. Upon reverse transcription, the product was diluted fivefold using TE buffer (10 mM Tris–HCl pH 8.0, 1 mM EDTA). Quantitative PCR was performed in a 10 µl reaction volume containing 5 µl SYBR Green PCR master mix (Thermo Fisher catalogue number 4309155), 500 nM of each primer, and 1 µl of template cDNA using a QuantStudio 3 real-time PCR system (Applied Biosystems). PP2AA3 was used for normalization. The primer sets used to quantify gene expression in this study are listed in Supplementary Table 2.

Transcriptome analysis

Total RNA was extracted from plant tissues using TRI Reagent (Sigma T9424) and cleaned up using Zymo RNA Clean & Concentrator-5 (R1013). RNA Integrity was assessed using Bioanalyzer Plant RNA Pico Assay (Agilent RNA 6000 Pico Kit 5067-1513) with RIN values 7.6 ± 0.3.

Library preparation and transcriptome sequencing was complete by the MSU Genomics Core Facility. In brief, libraries were prepared from plant total RNA using the Illumina Stranded mRNA Prep, Ligation kit with IDT for Illumina RNA UD Indexes, following manufacturer’s recommendations except that half volume reactions were performed. Completed libraries were quality controlled and quantified using a combination of Qubit dsDNA HS and Agilent 4200 TapeStation HS DNA1000 assays. The libraries were pooled in equimolar amounts and the pool quantified using the Invitrogen Collibri Quantification qPCR kit. The pool was loaded onto one lane of an Illumina S4 flow cell, and sequencing was performed in a 2 × 150 bp paired-end format using a NovaSeq v1.5, 300 cycle reagent kit.

Base calling was done by Illumina Real Time Analysis v3.4.4, and output of Real Time Analysis was demultiplexed and converted to FastQ format with Illumina Bcl2fastq v2.20.0.

Abundance of transcripts was quantified via Salmon v1.2.1 (ref. ⁸⁰), and differential expression analysis and plotting was accomplished via R packages, including DESeq2 v1.42.0, apeglm v1.24.0, ggplot2 v3.5.0 and pheatmap v1.0.12.

Phylogenetic analysis

To find all possible TIP1 orthologues, DHHC domain and ankyrin-repeat domain of TIP1 were blasted against 131 proteomes (23 species outside of the plant kingdom and 108 species inside the plant kingdom; Supplementary Table 4), via Phytozome⁸¹, Ensemble⁸² and PLAZA⁸³ databases. Proteins were confirmed via InterPro⁸⁴ for the presence of both DHHC and ankyrin-repeat domains. In total, 329 proteins from 123 species, including TIP1 (AT5G20350) and TIP1 paralog (AT2G14255), were then used to construct the phylogenetic tree via maximum likelihood algorithm using MEGA7 (ref. ⁸⁵).

ROS burst assay

Leaf discs (4 mm in diameter) were taken from the centre of leaves from 4-week-old soil-grown plants and floated with abaxial side down in wells of a white 96-well plate containing 200 µl sterile water in each well. Plates were covered with foil, and leaf discs were kept in sterile water overnight to attenuate wounding response. After 24 h in the dark, water was removed from wells and replaced with 100 µl of an immune-eliciting solution containing 68 µg ml⁻¹ luminol (Sigma, A8511), 20 µg ml⁻¹ horseradish peroxidase (Sigma, P6782) and 100 nM of the indicated Pathogen/Damage-Associated Molecular Patterns (PAMP/DAMP). Luminescence measurements were collected (total photon counting) over 60 min immediately after the addition of immune-eliciting solution using a SpectraMax L microplate reader with SoftMax Pro v.7.0.3 (Molecular Devices). Total ROS was calculated for each sample in Prism v.10.0.0 (GraphPad) using the ‘area under curve’ analysis.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Raw Illumina data for 16S rRNA gene amplicon sequences for the grm1 mutant and related controls are available in the Sequence Read Archive database (SRA) under BioProject PRJNA934331, with accession numbers SAMN33271678 to SAMN33271728. Raw Illumina data for 16S rRNA gene amplicon sequences for Col-0, tip1 and snc1 are available in the SRA database under BioProject PRJNA934350, with accession numbers SAMN33272493 to SAMN33272548. Raw Illumina data for 16S rRNA gene amplicon sequences for the tip1-like lesion-mimic autoimmune mutants are available in the SRA database under BioProject PRJNA1101553, with accession numbers SAMN40996484 to SAMN40996564. RNA-sequencing reads data have been deposited in the SRA database under BioProject PRJNA1103072, with accession numbers SAMN41039378 to SAMN41039404. Source data are provided with this paper.

Code availability

There is no custom code generated for this study.

References

Staskawicz, B. J., Ausubel, F. M., Baker, B. J., Ellis, J. G. & Jones, J. D. Molecular genetics of plant disease resistance. Science 268, 661–667 (1995).
Article CAS PubMed Google Scholar
Jones, J. D. & Dangl, J. L. The plant immune system. Nature 444, 323–329 (2006).
Article CAS PubMed Google Scholar
Ngou, B. P. M., Ding, P. & Jones, J. D. G. Thirty years of resistance: zig-zag through the plant immune system. Plant Cell 34, 1447–1478 (2022).
Article PubMed PubMed Central Google Scholar
Lewis, J. D., Guttman, D. S. & Desveaux, D. The targeting of plant cellular systems by injected type III effector proteins. Semin. Cell Dev. Biol. 20, 1055–1063 (2009).
Article CAS PubMed Google Scholar
Dou, D. & Zhou, J. M. Phytopathogen effectors subverting host immunity: different foes, similar battleground. Cell Host Microbe 12, 484–495 (2012).
Article CAS PubMed Google Scholar
Toruno, T. Y., Stergiopoulos, I. & Coaker, G. Plant–pathogen effectors: cellular probes interfering with plant defenses in spatial and temporal manners. Annu. Rev. Phytopathol. 54, 419–441 (2016).
Article CAS PubMed PubMed Central Google Scholar
Xin, X. F., Kvitko, B. & He, S. Y. Pseudomonas syringae: what it takes to be a pathogen. Nat. Rev. Microbiol. 16, 316–328 (2018).
Article CAS PubMed PubMed Central Google Scholar
Zhou, J. M. & Zhang, Y. Plant immunity: danger perception and signaling. Cell 181, 978–989 (2020).
Article CAS PubMed Google Scholar
Yuan, M., Ngou, B. P. M., Ding, P. & Xin, X. F. PTI-ETI crosstalk: an integrative view of plant immunity. Curr. Opin. Plant Biol. 62, 102030 (2021).
Article CAS PubMed Google Scholar
Huot, B., Yao, J., Montgomery, B. L. & He, S. Y. Growth–defense tradeoffs in plants: a balancing act to optimize fitness. Mol. Plant 7, 1267–1287 (2014).
Article CAS PubMed PubMed Central Google Scholar
He, Z., Webster, S. & He, S. Y. Growth–defense trade-offs in plants. Curr. Biol. 32, R634–R639 (2022).
Article CAS PubMed Google Scholar
Muller, D. B., Vogel, C., Bai, Y. & Vorholt, J. A. The plant microbiota: systems-level insights and perspectives. Annu. Rev. Genet. 50, 211–234 (2016).
Article CAS PubMed Google Scholar
Wang, N. R. & Haney, C. H. Harnessing the genetic potential of the plant microbiome. Biochemist 42, 20–25 (2020).
Article CAS Google Scholar
Bulgarelli, D., Schlaeppi, K., Spaepen, S., van Themaat, VerLoren & Schulze-Lefert, E. Structure and functions of the bacterial microbiota of plants. Annu. Rev. Plant Biol. 64, 807–838 (2013).
Article CAS PubMed Google Scholar
Paasch, B. C. & He, S. Y. Toward understanding microbiota homeostasis in the plant kingdom. PLoS Pathog. 17, e1009472 (2021).
Article CAS PubMed PubMed Central Google Scholar
Hacquard, S., Spaepen, S., Garrido-Oter, R. & Schulze-Lefert, P. Interplay between innate immunity and the plant microbiota. Annu. Rev. Phytopathol. 55, 565–589 (2017).
Article CAS PubMed Google Scholar
Nobori, T. et al. Dissecting the cotranscriptome landscape of plants and their microbiota. EMBO Rep. https://doi.org/10.15252/embr.202255380 (2022).
Article PubMed PubMed Central Google Scholar
Velasquez, A. C., Huguet-Tapia, J. C. & He, S. Y. Shared in planta population and transcriptomic features of nonpathogenic members of endophytic phyllosphere microbiota. Proc. Natl Acad. Sci. USA 119, e2114460119 (2022).
Article CAS PubMed PubMed Central Google Scholar
Xin, X. F. et al. Bacteria establish an aqueous living space in plants crucial for virulence. Nature 539, 524–529 (2016).
Article CAS PubMed PubMed Central Google Scholar
Chen, T. et al. A plant genetic network for preventing dysbiosis in the phyllosphere. Nature 580, 653–657 (2020).
Article CAS PubMed PubMed Central Google Scholar
Pfeilmeier, S. et al. The plant NADPH oxidase RBOHD is required for microbiota homeostasis in leaves. Nat. Microbiol. 6, 852–864 (2021).
Article CAS PubMed PubMed Central Google Scholar
Gimenez-Ibanez, S. et al. AvrPtoB targets the LysM receptor kinase CERK1 to promote bacterial virulence on plants. Curr. Biol. 19, 423–429 (2009).
Article CAS PubMed Google Scholar
Schwessinger, B. et al. Phosphorylation-dependent differential regulation of plant growth, cell death, and innate immunity by the regulatory receptor-like kinase BAK1. PLoS Genet. 7, e1002046 (2011).
Article CAS PubMed PubMed Central Google Scholar
Cunnac, S. et al. Genetic disassembly and combinatorial reassembly identify a minimal functional repertoire of type III effectors in Pseudomonas syringae. Proc. Natl Acad. Sci. USA 108, 2975–2980 (2011).
Article CAS PubMed PubMed Central Google Scholar
Wei, W. et al. The gene coding for the Hrp pilus structural protein is required for type III secretion of Hrp and Avr proteins in Pseudomonas syringae pv. tomato. Proc. Natl Acad. Sci. USA 97, 2247–2252 (2000).
Article CAS PubMed PubMed Central Google Scholar
Kremer, J. M. et al. Peat-based gnotobiotic plant growth systems for Arabidopsis microbiome research. Nat. Protoc. 16, 2450–2470 (2021).
Article CAS PubMed PubMed Central Google Scholar
Alonso, J. M. et al. Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science 301, 653–657 (2003).
Article PubMed Google Scholar
Chinchilla, D., Bauer, Z., Regenass, M., Boller, T. & Felix, G. The Arabidopsis receptor kinase FLS2 binds flg22 and determines the specificity of flagellin perception. Plant Cell 18, 465–476 (2006).
Article CAS PubMed PubMed Central Google Scholar
Gomez-Gomez, L., Felix, G. & Boller, T. A single locus determines sensitivity to bacterial flagellin in Arabidopsis thaliana. Plant J. 18, 277–284 (1999).
Article CAS PubMed Google Scholar
Boudsocq, M. et al. Differential innate immune signalling via Ca(2+) sensor protein kinases. Nature 464, 418–422 (2010).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Y., Goritschnig, S., Dong, X. & Li, X. A gain-of-function mutation in a plant disease resistance gene leads to constitutive activation of downstream signal transduction pathways in suppressor of npr1-1, constitutive 1. Plant Cell 15, 2636–2646 (2003).
Article CAS PubMed PubMed Central Google Scholar
Cheng, Y. T. et al. Stability of plant immune-receptor resistance proteins is controlled by SKP1-CULLIN1-F-box (SCF)-mediated protein degradation. Proc. Natl Acad. Sci. USA 108, 14694–14699 (2011).
Article CAS PubMed PubMed Central Google Scholar
Kunze, G. et al. The N terminus of bacterial elongation factor Tu elicits innate immunity in Arabidopsis plants. Plant Cell 16, 3496–3507 (2004).
Article CAS PubMed PubMed Central Google Scholar
Krol, E. et al. Perception of the Arabidopsis danger signal peptide 1 involves the pattern recognition receptor AtPEPR1 and its close homologue AtPEPR2. J. Biol. Chem. 285, 13471–13479 (2010).
Article CAS PubMed PubMed Central Google Scholar
Lundberg, D. S. et al. Contrasting patterns of microbial dominance in the Arabidopsis thaliana phyllosphere. Proc. Natl Acad. Sci. USA 119, e2211881119 (2022).
Article CAS PubMed PubMed Central Google Scholar
van Wersch, R., Li, X. & Zhang, Y. Mighty dwarfs: Arabidopsis autoimmune mutants and their usages in genetic dissection of plant immunity. Front. Plant Sci. 7, 1717 (2016).
PubMed PubMed Central Google Scholar
Bruggeman, Q., Raynaud, C., Benhamed, M. & Delarue, M. To die or not to die? Lessons from lesion mimic mutants. Front. Plant Sci. 6, 24 (2015).
Article PubMed PubMed Central Google Scholar
Boursiac, Y. et al. Disruption of the vacuolar calcium-ATPases in Arabidopsis results in the activation of a salicylic acid-dependent programmed cell death pathway. Plant Physiol. 154, 1158–1171 (2010).
Article CAS PubMed PubMed Central Google Scholar
Liang, H. et al. Ceramides modulate programmed cell death in plants. Genes Dev. 17, 2636–2641 (2003).
Article CAS PubMed PubMed Central Google Scholar
Zeng, H. Y. et al. The immune components ENHANCED DISEASE SUSCEPTIBILITY 1 and PHYTOALEXIN DEFICIENT 4 are required for cell death caused by overaccumulation of ceramides in Arabidopsis. Plant J. 107, 1447–1465 (2021).
Article CAS PubMed Google Scholar
Dietrich, R. A., Richberg, M. H., Schmidt, R., Dean, C. & Dangl, J. L. A novel zinc finger protein is encoded by the Arabidopsis LSD1 gene and functions as a negative regulator of plant cell death. Cell 88, 685–694 (1997).
Article CAS PubMed Google Scholar
Yang, H. et al. A mutant CHS3 protein with TIR-NB-LRR-LIM domains modulates growth, cell death and freezing tolerance in a temperature-dependent manner in Arabidopsis. Plant J. 63, 283–296 (2010).
Article CAS PubMed Google Scholar
Clough, S. J. et al. The Arabidopsis dnd1 “defense, no death” gene encodes a mutated cyclic nucleotide-gated ion channel. Proc. Natl Acad. Sci. USA 97, 9323–9328 (2000).
Article CAS PubMed PubMed Central Google Scholar
Jurkowski, G. I. et al. Arabidopsis DND2, a second cyclic nucleotide-gated ion channel gene for which mutation causes the “defense, no death” phenotype. Mol. Plant Microbe Interact. 17, 511–520 (2004).
Article CAS PubMed Google Scholar
Velasquez, A. C., Oney, M., Huot, B., Xu, S. & He, S. Y. Diverse mechanisms of resistance to Pseudomonas syringae in a thousand natural accessions of Arabidopsis thaliana. New Phytol. 214, 1673–1687 (2017).
Article CAS PubMed PubMed Central Google Scholar
Ton, J., Pieterse, C. M. & Van Loon, L. C. Identification of a locus in Arabidopsis controlling both the expression of rhizobacteria-mediated induced systemic resistance (ISR) and basal resistance against Pseudomonas syringae pv. tomato. Mol. Plant Microbe Interact. 12, 911–918 (1999).
Article CAS PubMed Google Scholar
Todesco, M. et al. Natural allelic variation underlying a major fitness trade-off in Arabidopsis thaliana. Nature 465, 632–636 (2010).
Article CAS PubMed PubMed Central Google Scholar
Schiefelbein, J., Galway, M., Masucci, J. & Ford, S. Pollen tube and root-hair tip growth is disrupted in a mutant of Arabidopsis thaliana. Plant Physiol. 103, 979–985 (1993).
Article CAS PubMed PubMed Central Google Scholar
Hemsley, P. A., Kemp, A. C. & Grierson, C. S. The TIP GROWTH DEFECTIVE1 S-acyl transferase regulates plant cell growth in Arabidopsis. Plant Cell 17, 2554–2563 (2005).
Article CAS PubMed PubMed Central Google Scholar
Singaraja, R. R. et al. HIP14, a novel ankyrin domain-containing protein, links huntingtin to intracellular trafficking and endocytosis. Hum. Mol. Genet 11, 2815–2828 (2002).
Article CAS PubMed Google Scholar
Babu, P., Deschenes, R. J. & Robinson, L. C. Akr1p-dependent palmitoylation of Yck2p yeast casein kinase 1 is necessary and sufficient for plasma membrane targeting. J. Biol. Chem. 279, 27138–27147 (2004).
Article CAS PubMed Google Scholar
Kumar, M., Carr, P. & Turner, S. R. An atlas of Arabidopsis protein S-acylation reveals its widespread role in plant cell organization and function. Nat. Plants 8, 670–681 (2022).
Article CAS PubMed Google Scholar
Hemsley, P. A., Weimar, T., Lilley, K. S., Dupree, P. & Grierson, C. S. A proteomic approach identifies many novel palmitoylated proteins in Arabidopsis. New Phytol. 197, 805–814 (2013).
Article CAS PubMed Google Scholar
Hurst, C. H. et al. Juxta-membrane S-acylation of plant receptor-like kinases is likely fortuitous and does not necessarily impact upon function. Sci. Rep. 9, 12818 (2019).
Article PubMed PubMed Central Google Scholar
Hurst, C. H. et al. S-acylation stabilizes ligand-induced receptor kinase complex formation during plant pattern-triggered immune signaling. Curr. Biol. 33, 1588–1596 e1586 (2023).
Article CAS PubMed Google Scholar
Chen, D. et al. S-acylation of P2K1 mediates extracellular ATP-induced immune signaling in Arabidopsis. Nat. Commun. 12, 2750 (2021).
Article CAS PubMed PubMed Central Google Scholar
Gomez-Gomez, L. & Boller, T. FLS2: an LRR receptor-like kinase involved in the perception of the bacterial elicitor flagellin in Arabidopsis. Mol. Cell 5, 1003–1011 (2000).
Article CAS PubMed Google Scholar
Zipfel, C. et al. Perception of the bacterial PAMP EF-Tu by the receptor EFR restricts Agrobacterium-mediated transformation. Cell 125, 749–760 (2006).
Article CAS PubMed Google Scholar
Noutoshi, Y. et al. A single amino acid insertion in the WRKY domain of the Arabidopsis TIR-NBS-LRR-WRKY-type disease resistance protein SLH1 (sensitive to low humidity 1) causes activation of defense responses and hypersensitive cell death. Plant J. 43, 873–888 (2005).
Article CAS PubMed Google Scholar
Li, L. & Weigel, D. One hundred years of hybrid necrosis: hybrid autoimmunity as a window into the mechanisms and evolution of plant–pathogen interactions. Annu. Rev. Phytopathol. 59, 213–237 (2021).
Article CAS PubMed Google Scholar
Bomblies, K. et al. Autoimmune response as a mechanism for a Dobzhansky–Muller-type incompatibility syndrome in plants. PLoS Biol. 5, e236 (2007).
Article PubMed PubMed Central Google Scholar
Song, S. et al. PSKR1 balances the plant growth–defence trade-off in the rhizosphere microbiome. Nat. Plants 9, 2071–2084 (2023).
Article CAS PubMed Google Scholar
Choi, S. C. et al. Gut microbiota dysbiosis and altered tryptophan catabolism contribute to autoimmunity in lupus-susceptible mice. Sci. Transl. Med. 12, https://doi.org/10.1126/scitranslmed.aax2220 (2020).
Cox, L. M. et al. Gut microbiome in progressive multiple sclerosis. Ann. Neurol. 89, 1195–1211 (2021).
Article CAS PubMed PubMed Central Google Scholar
Chriswell, M. E. et al. Clonal IgA and IgG autoantibodies from individuals at risk for rheumatoid arthritis identify an arthritogenic strain of Subdoligranulum. Sci. Transl. Med. 14, eabn5166 (2022).
Article CAS PubMed PubMed Central Google Scholar
Yanai, A. et al. Palmitoylation of huntingtin by HIP14 is essential for its trafficking and function. Nat. Neurosci. 9, 824–831 (2006).
Article CAS PubMed PubMed Central Google Scholar
Khoshnan, A. Gut microbiota as a modifier of Huntington’s disease pathogenesis. J. Huntingtons Dis. 13, 133–147 (2024).
Article CAS PubMed PubMed Central Google Scholar
Lemonidis, K., Sanchez-Perez, M. C. & Chamberlain, L. H. Identification of a novel sequence motif recognized by the ankyrin repeat domain of zDHHC17/13 S-acyltransferases. J. Biol. Chem. 290, 21939–21950 (2015).
Article CAS PubMed PubMed Central Google Scholar
Lemonidis, K., MacLeod, R., Baillie, G. S. & Chamberlain, L. H. Peptide array-based screening reveals a large number of proteins interacting with the ankyrin-repeat domain of the zDHHC17 S-acyltransferase. J. Biol. Chem. 292, 17190–17202 (2017).
Article CAS PubMed PubMed Central Google Scholar
Verardi, R., Kim, J. S., Ghirlando, R. & Banerjee, A. Structural basis for substrate recognition by the ankyrin repeat domain of human DHHC17 palmitoyltransferase. Structure 25, 1337–1347 e1336 (2017).
Article CAS PubMed PubMed Central Google Scholar
Ishiga, Y., Ishiga, T., Uppalapati, S. R. & Mysore, K. S. Arabidopsis seedling flood-inoculation technique: a rapid and reliable assay for studying plant–bacterial interactions. Plant Methods 7, 32 (2011).
Article CAS PubMed PubMed Central Google Scholar
Clough, S. J. & Bent, A. F. Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735–743 (1998).
Article CAS PubMed Google Scholar
Bolyen, E. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37, 852–857 (2019).
Article CAS PubMed PubMed Central Google Scholar
Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet. J. 17, 10–12 (2011).
Article Google Scholar
Callahan, B. J. et al. DADA2: high-resolution sample inference from Illumina amplicon data. Nat. Methods 13, 581–583 (2016).
Article CAS PubMed PubMed Central Google Scholar
Mohammad, N. S., Nazli, R., Zafar, H. & Fatima, S. Effects of lipid based multiple micronutrients supplement on the birth outcome of underweight pre-eclamptic women: a randomized clinical trial. Pak. J. Med Sci. 38, 219–226 (2022).
PubMed PubMed Central Google Scholar
Yilmaz, P. et al. The SILVA and “All-species Living Tree Project (LTP)” taxonomic frameworks. Nucleic Acids Res. 42, D643–D648 (2014).
Article CAS PubMed Google Scholar
Quast, C. et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 41, D590–D596 (2013).
Article CAS PubMed Google Scholar
Robeson, M. S. et al. RESCRIPt: reproducible sequence taxonomy reference database management. PLoS Comput. Biol. 17, e1009581 (2021).
Article PubMed PubMed Central Google Scholar
Patro, R., Duggal, G., Love, M. I., Irizarry, R. A. & Kingsford, C. Salmon provides fast and bias-aware quantification of transcript expression. Nat. Methods 14, 417–419 (2017).
Article CAS PubMed PubMed Central Google Scholar
Goodstein, D. M. et al. Phytozome: a comparative platform for green plant genomics. Nucleic Acids Res. 40, D1178–D1186 (2012).
Article CAS PubMed Google Scholar
Martin, F. J. et al. Ensembl 2023. Nucleic Acids Res. 51, D933–D941 (2023).
Article CAS PubMed Google Scholar
Van Bel, M. et al. PLAZA 5.0: extending the scope and power of comparative and functional genomics in plants. Nucleic Acids Res. 50, D1468–D1474 (2022).
Article PubMed Google Scholar
Paysan-Lafosse, T. et al. InterPro in 2022. Nucleic Acids Res. 51, D418–D427 (2023).
Article CAS PubMed Google Scholar
Kumar, S., Stecher, G. & Tamura, K. MEGA7: molecular evolutionary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870–1874 (2016).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank X. Li from the University of British Columbia for kindly providing us snc1 seeds. We thank MSU Growth Chamber Facility, MSU Genomics Core Facility and Duke Phytotron for technical assistance. This work was supported by the Natural Sciences and Engineering Research Council of Canada (Y.T.C.) and United States National Institutes of Health (1R01AI155441; to S.Y.H.). S.Y.H. is an Investigator at Howard Hughes Medical Institute.

Author information

Authors and Affiliations

Department of Biology, Duke University, Durham, NC, USA
Yu Ti Cheng, Li Zhang, Bradley C. Paasch, Reza Sohrabi & Sheng Yang He
Howard Hughes Medical Institute, Duke University, Durham, NC, USA
Yu Ti Cheng, Li Zhang, Bradley C. Paasch, Reza Sohrabi & Sheng Yang He
Department of Energy Plant Research Laboratory, Michigan State University, East Lansing, MI, USA
Yu Ti Cheng, Caitlin A. Thireault, Li Zhang, Reza Sohrabi & Sheng Yang He
School of Biology and Basic Medical Science, Soochow University, Suzhou, China
Li Zhang

Authors

Yu Ti Cheng
View author publications
You can also search for this author in PubMed Google Scholar
Caitlin A. Thireault
View author publications
You can also search for this author in PubMed Google Scholar
Li Zhang
View author publications
You can also search for this author in PubMed Google Scholar
Bradley C. Paasch
View author publications
You can also search for this author in PubMed Google Scholar
Reza Sohrabi
View author publications
You can also search for this author in PubMed Google Scholar
Sheng Yang He
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Y.T.C. and S.Y.H. conceptualized and designed the project. Y.T.C. led and conducted most of the experimental work. C.A.T. generated EMS-mutagenized population in the Arabidopsis bbc mutant background. C.A.T. and L.Z. carried out genetic screens and characterization of grm mutants. L.Z. performed phylogenetic analysis of TIP1 homologs. L.Z. performed data analysis of transcriptional profiling using RNA sequencing. B.C.P. contributed materials for gene expression analysis. B.C.P. and Y.T.C. performed data analysis of the 16S rRNA gene amplicon sequencing. R.S. contributed materials and analysis tools. Y.T.C and S.Y.H. wrote the manuscript with input from all authors.

Corresponding authors

Correspondence to Yu Ti Cheng or Sheng Yang He.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Plants thanks Wangsheng Zhu and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Mutants isolated from the bbc genetic screen.

Plant images of 4-week-old soil grown grm mutants; scale bar indicates 2 cm. All grm mutant phenotypes were inherited as recessive traits with exception of grm2. grm2 phenotypes were inherited as semi-dominant trait; based on the 1:2:1 segregation ratio, the grm2 mutant image panel contains bbc-like (left), heterozygous (middle), and homozygous (right) plants.

Extended Data Fig. 2 Identifying and confirming the causative mutation in the grm1 mutant.

a, grm1 genomic mapping. Red line represents allele frequency. Blue and purple lines denote 95% and 99% confidence intervals, respectively. b, RT-PCR products using primers flanking the grm1 mutation locus. Genomic (top panel) and complementary (bottom panel) DNA from indicated genotypes were used as templates. The retained intron is expected to lead to a premature STOP codon. c, Images of four-week-old, potting soil-grown Col-0, bbc, grm1 and two independent grm1 complementation lines under ambient humidity ( ~ 50% RH basal control condition; upper panel) or high humidity ( ~ 95% RH; bottom panel) for five days. Scale bar equals 2 cm. d, Transgene complementation of the grm1 mutant. Population sizes of leaf endophytic microbiota after five days of plant growth under ambient humidity ( ~ 50% RH, basal control condition) or high humidity ( ~ 95% RH). Results represent the mean values ± SEM of four plants. Statistical analysis was performed using two-way ANOVA with Tukey’s HSD test. Exact p-values for all comparisons are shown in the Source Data. Experiment was independently performed twice with similar results.

Source data

Extended Data Fig. 3 Appearance of tip1 mutant alleles.

a, A schematic diagram illustrating various mutant alleles of TIP1 protein. tip1^W171STOP is the allele isolated from this study which contains a G to A mutation at the splicing junction that is expected to cause a pre-mature STOP codon at amino acid residue Trp¹⁷¹ in the ankyrin-repeat domain. SALK_020996 and SALK_052842 have T-DNA insertions that would affect the catalytic DHHC cysteine-rich domain (DHHC-CRD). b, Images of four-week-old, potting soil-grown Col-0, bbc, grm1 and various tip1 single mutant plants under ambient humidity ( ~ 50% RH, basal control condition; upper panel) or high humidity ( ~ 95% RH) for five days. Scale bar equals 2 cm.

Extended Data Fig. 4 TIP1 and SNC1 gene expression is responsive to PTI elicitor flg22.

Expression of TIP1 (a), SNC1 (b), and FRK1 (c) in wild-type Col-0 plants after flg22 treatment. PP2AA3 expression was used for normalization. Plants were grown under axenic (Ax) or holoxenic (Holo) conditions in GnotoPots. Open circles (∘) indicate the basal, uninduced state whereas closed circles (•) indicate the expression 90 min after 250 nM flg22 induction. Results represent the mean values ± SEM of four biological replicates. Statistical analysis was done by the two-way ANOVA with Fisher’s LSD test. Experiment was independently performed twice times with similar results.

Source data

Extended Data Fig. 5 ROS burst dynamics induced by PTI elicitors.

ROS burst dynamics induced by 100 nM of flg22 (a), elf18 (b) and AtPep1 (c) in 4-week-old soil grown plants. Results represent the mean values ± SEM (n = 8 plants). Bar plots show cumulative ROS production over 60 min (right). Results represent the mean values ± SEM of eight biological replicates. Statistical analysis was done by one-way ANOVA with Tukey’s HSD test. Experiment was independently performed twice times with similar results.

Source data

Extended Data Fig. 6 Effects of different soil types on tip1 and snc1 plants.

a, soil appearance of Arabidopsis mix, North Carolina (NC) soil, Michigan State University (MSU) soil, and Michigan Benton Harbor (MI-BH) soil. b, Images of four-week-old plants growing on pots prepared using indicated soil types (left) and plant images after five days high humidity condition ( ~ 95% RH; right). c, Population sizes of endophytic leaf microbiota after five days high humidity condition ( ~ 95% RH). Results represent the mean values ± SEM (n = 4 biological replicates; each biological replicate contains leaves from one plant). Different letters represent a significant difference (p < 0.05, two-way ANOVA with Tukey’s HSD test). Exact p-values for all comparisons are shown in the Source Data. Experiment was independently performed twice with similar results.

Source data

Extended Data Fig. 7 Effects of bacterial bulk culture (BC) collected from dysbiotic tip1 plants on wild type plants.

a, Col-0 leaf images on Day 0 and Day 6 after infiltrated with BC^Col-0 or BC^tip1. b, Population sizes of endophytic leaf microbiota immediate (Day 0; left) and six days (Day 6; right) after BC infiltrations. Results represent the mean values ± SEM (n = 4 biological replicates; each biological replicate contains 1-2 leaves from one plant). Statistical analysis was done by two-way ANOVA with Fisher’s LSD test. Experiment was independently performed twice with similar results.

Source data

Extended Data Fig. 8 Endophytic bacterial microbiota profiling in tip1 and tip1-like autoimmune mutants.

a, PCoA of weighted UniFrac distances obtained from 16 S rRNA gene sequence profiles of endophytic bacterial microbiota in Col-0, tip1 and tip1-like autoimmune mutants under high humidity for 5 days. Pairwise PERMANOVA results are provided in the Source Data. b, The number of ASVs indicating endophytic bacterial microbiota alpha diversity in the indicated genotypes. n = 17 (Col-0) and n = 16 (for tip1 and tip1-like mutant plants). The center lines of the box plot represent means, the box edges are the 75^th and 25^th percentiles, whiskers extend to 10-90 percentiles, and dots are outliers. Statistical analysis was done by one-way ANOVA with Dunnett’s test. c, Relative abundance of bacterial populations at the phylum level. Members of Proteobacteria phylum are further separated into class.

Source data

Extended Data Fig. 9 Transcriptomic analysis of Col-0, tip1 and snc1 plants grown in GnotoPots.

a, PCA analysis of genes expressed in Col-0, tip1 and snc1 plants grown in GnotoPots under axenic (Ax), heat-killed “MSU” microbial community (HK) or live “MSU” microbial community (holoxenic; Holo) conditions. Zoomed-in image contains PCA analysis of Col-0 under all three conditions and tip1 grown in GnotoPots under Ax and HK conditions. b, Heat map of DEGs significantly changed between Col-0 and tip1 and/or between Col-0 and snc1 plants grown in GnotoPots under “MSU” microbial community holoxenic condition. Clusters 1 and 2 are enriched for defence-related Gene Ontology (GO) terms, whereas cluster 3 is enriched for growth and photosynthesis-related GO terms. See Supplementary Data 4 for the complete list of genes and their normalized counts. c, Selected defence-related gene expression in Col-0, tip1 and snc1 plants grown in GnotoPots under “MSU” microbial community holoxenic condition. See Supplementary Table 5 for the gene IDs.

Supplementary information

Supplementary Information

Supplementary Figs. 1–6.

Reporting Summary

Supplementary Data 1

Four tables; ASV counts from 16S amplicon profiling of leaf endophytic microbiota (Tables 1–3) and differentially expressed genes of transcriptomic data (Table 4).

Supplementary Data 2

Nine tables; statistical source data for Supplementary Fig. 6.

Supplementary Table 1

Five tables; list of primers, mutant lines and gene/protein IDs as supporting information.

Source data

Source Data Fig. 1

Statistical source data.

Source Data Fig. 2

Statistical source data.

Source Data Fig. 3

Statistical source data.

Source Data Fig. 4

Statistical source data.

Source Data Fig. 5

Statistical source data.

Source Data Fig. 6

Statistical source data.

Source Data Fig. 7

Statistical source data.

Source Data Extended Data Fig. 1

Unprocessed gel image.

Source Data Extended Data Fig. 1

Statistical source data.

Source Data Extended Data Fig. 2

Statistical source data.

Source Data Extended Data Fig. 3

Statistical source data.

Source Data Extended Data Fig. 4

Statistical source data.

Source Data Extended Data Fig. 5

Statistical source data.

Source Data Extended Data Fig. 6

Statistical source data.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Cheng, Y.T., Thireault, C.A., Zhang, L. et al. Roles of microbiota in autoimmunity in Arabidopsis leaves. Nat. Plants 10, 1363–1376 (2024). https://doi.org/10.1038/s41477-024-01779-9

Download citation

Received: 28 February 2023
Accepted: 01 August 2024
Published: 06 September 2024
Issue Date: September 2024
DOI: https://doi.org/10.1038/s41477-024-01779-9

Subjects

Abstract

Similar content being viewed by others

Main

Results

Identifying Arabidopsis mutants with altered leaf microbiota

Characterization of the grm1 mutant

Identification of the causal mutation in the grm1 mutant

The tip1 mutant has features of autoimmune mutants

tip1- and snc1-mediated autoimmunity are distinct

Role of microbiota in Arabidopsis lesion-mimic autoimmunity

Role of microbiota in autoimmunity in wild accessions

Discussion

Methods

Plant materials and growth conditions

Genetic screen

Mutation identification using mapping-by-sequencing approach

Genetic complementation of the grm1 mutant

Quantification of endophytic leaf bacterial microbiota

Preparation of bacterial bulk culture

16S rRNA gene amplicon sequencing

16S rRNA gene amplicon sequencing data processing

Bacterial infection assays

Reverse transcription quantitative PCR analysis gene expression

Transcriptome analysis

Phylogenetic analysis

ROS burst assay

Reporting summary

Data availability

Code availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding authors

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Extended data

Supplementary information

Source data

Rights and permissions

About this article

Cite this article

Share this article

Search

Quick links