Two <scp>LysM</scp> receptor‐like kinases regulate arbuscular mycorrhiza through distinct signaling pathways in <i>Lotus japonicus</i>

Abstract

Arbuscular mycorrhiza (AM) is a mutualistic plant–fungal interaction that greatly benefits the growth of both organisms (Parniske, 2008; Smith & Read, 2010). AM fungi enter the host root, and their hyphae elongate through the cortical cell layer, forming a tree-like symbiotic structure called ‘arbuscule’, which facilitates nutrient supply to the plant (Harrison, 2012). AM fungi provide nutrients, including phosphate, minerals, and water to the host plant. In return, the fungi receive photosynthetic products from the host plant via the arbuscules (Bago et al., 2003; Zhu & Miller, 2003). AM fungi are known to secrete multiple symbiotic signaling molecules recognized by the host plant, such as chito-oligosaccharides (COs) and lipo-chito-oligosaccharides (Maillet et al., 2011; Genre et al., 2013). Signal perception by the host plant activates symbiosis-related signaling and responses, such as symbiosis-related gene expression and the oscillation of intracellular calcium (Ca) concentration, called Ca spiking (Genre et al., 2013). Lysin motif (LysM)-type receptor proteins are known to perceive COs and regulate AM in rice (Oryza sativa), tomato (Solanum lycopersicum), legumes, and many other plants (Miyata et al., 2014; Buendia et al., 2016; Liao et al., 2018; Feng et al., 2019; Girardin et al., 2019). In rice, chitin elicitor-receptor kinase 1 (OsCERK1) and chitin elicitor-binding protein (CEBiP) form heterodimers that mediate the long-chain CO perception and signaling (Kaku et al., 2006; Shimizu et al., 2010). The perception of COs by the LysM receptor plays an important role in triggering plant immunity (Desaki et al., 2018). In the leguminous plant Lotus japonicus, LYS6 (CERK6) is a LysM receptor involved in CO perception and pathogenic responses; however, its relationship with AM has not yet been demonstrated (Bozsoki et al., 2017). Furthermore, AM fungi infection has been shown to induce LYS11 expression in L. japonicus; however, the phenotype of the lys11 mutant was comparable to that of the wild-type (WT) (Rasmussen et al., 2016). In this study, we searched for receptors involved in AM among CERK1 homologs in L. japonicus, revealing the involvement of two LysM receptors in regulating AM. Seeds of WT L. japonicus MG-20 and the LysM receptor mutant lines were scarified with sandpaper and sterilized with sodium hypochlorite (effective chloride 1%). Once sterilized, the seeds were soaked overnight in sterilized water, germinated on 0.8% agar plates, and grown in a growth chamber (16 h : 8 h, 24°C, light : dark). Spores of Rhizophagus irregularis (DAOM197198; PremierTech, Rivière-du-Loup, QC, Canada) were inoculated on the host plant (> 100 spores per plant) planted in pots containing 300 ml of soil supplied with ½-strength Hoagland solution containing 0.1 mM KNO3 (100 ml per pot). A chive (Allium schoenoprasum) nurse pot system was used to count hyphal attachments and entry points in the host root (Demchenko et al., 2004). The LysM receptors homologous to OsCERK1 in L. japonicus (phylogenetic group LYS-1 in Lohmann et al., 2010) and other homologs known to be involved in AM and root nodule symbiosis were selected for phylogenetic analysis. Amino acid sequences of the LysM receptors were obtained from the National Center for Biotechnology Information and genome database of L. japonicus (http://viewer.shigen.info/lotus/index-j.php). Full-length amino acid sequences were aligned using ClustalW, and a phylogenetic tree was constructed using the neighbor-joining method implemented in the Mega11 software (v.11.0.13, https://www.megasoftware.net/) with a bootstrap value of 1000 replicates (Tamura et al., 2021). The CRISPR/Cas9 system was used to mutate the LYS6 and LYS7 genes of L. japonicus MG-20 plants. Two target sites were selected for each target gene using the CRISPR-P program (http://cbi.hzau.edu.cn/crispr/) (Lei et al., 2014). Each target site sequence harbored > 3 bp mismatches with potential off-target sites in the L. japonicus genome. Target oligonucleotides (primer sets 1–8 in Supporting Information Table S1) were annealed and cloned into the BbsI site of the single-guide RNA (sgRNA) vector pUC19_AtU6oligo (sgRNA targets 1–4; Fig. S1) (Ito et al., 2015). The sgRNA cassette was amplified, and the amplicon was ligated to the PCR product of the other sgRNA (primer sets 9–12 in Table S1). It yielded a vector with the four sgRNA cassettes arranged in tandem. The sgRNA cassettes were then subcloned into the I-SceI site of pZH_gYSA_FFCas9, which already contained the Cas9 and HPT expression cassettes (Ito et al., 2015). CRISPR/Cas9 constructs were introduced into L. japonicus MG-20 using Agrobacterium tumefaciens AGL1, as previously described (Stiller et al., 1997). Mutations around the target sites in the transgenic plants were first evaluated by PCR and then confirmed by sequencing. Homozygous mutants were selected from T1 transgenic plants, and lys6lys7 double mutants (#24, #28, #29) and a lys7 single mutant (#15) were obtained. Progeny (T2) was used in this study. Subsequently, the double mutants were crossed with the WT, and lys6 (lys6#24, #28, #29) and lys7 (lys7#24, #28) single mutants were obtained from the F2 plants. The F3 generation of the single mutants was also used for subsequent analyses. WT seeds from the same parent plants used for the transformation or crossing were used as WT controls in all experiments. Ca spiking was observed using nuclear-localized yellow-cameleon 2.60 (NLS-YC) (Nagai et al., 2004). NLS-YC, under the control of the Ubiquitin promoter, was introduced into L. japonicus via hairy root transformation using A. rhizogenes AR1193 (Takeda et al., 2009). Transgenic roots were treated with a CO mixture (1–8 mers, 1 mg ml−1) (Seikagaku Corporation, Tokyo, Japan), chitin tetramer (CO4, 10−6 M) (IsoSep AB, Tullinge, Sweden), or chitin octamer (CO8, 10−6 M) (IsoSep AB) in BNM medium (Ehrhardt et al., 1996). Ca imaging was performed using a Nikon microscope TE2000-U equipped with 20× dry objectives and imaging systems (MAC5000; Ludl electronic products, Hawthorne, NY, USA, and CoolSNAP Dyno; Teledyne Photometrics, Tuscon, AZ, USA). Fluorescent images of the NLS-YC root were acquired every 5 s with CFP (CFP ex. 440 nm/CFP em. 480 nm) and fluorescence resonance energy transfer (FRET; CFP ex. 440 nm/YFP em. 535 nm) filter sets. Fluorescence intensities of FRET and CFP around the nucleus were measured, and the ratio (FRET int. : CFP int.) was calculated using NIS-Elements AR (Nikon, Tokyo, Japan). Cells with more than two Ca increases were counted as spiking+ cells, and the ratio (spike+ cells : total cells) was calculated. More than three roots (5–15 cells per root) were analyzed in WT and each mutant line. Total RNA was extracted from whole root samples obtained from 5 to 7 plants using the PureLink™ Plant RNA Reagent (Thermo Fisher Scientific, Waltham, MA, USA). Total RNA concentration was quantified using a NanoDrop ND-1000 spectrophotometer (Thermo Fisher Scientific). Reverse transcription and real-time PCR were performed using the ReverTra Ace qPCR RT kit (Toyobo, Osaka, Japan) and the Thunderbird qPCR Mix (Toyobo) on the AriaMx Real-Time PCR System (Agilent, Santa Clara, CA, USA), according to the manufacturers' instructions. cDNA was synthesized from 250 ng of total RNA, and each reaction was performed in duplicate or triplicate. CO-induced genes were identified through transcriptome analysis of the root infected with AM fungi (Takeda et al., 2015), and of the roots treated with chitin pentamer (data not shown). The primer sets used for this analysis are listed in Table S1 (primer sets 13–26). PCR conditions were set as described previously by Takeda et al. (2013). Relative gene expressions were compared against Ubiquitin and Elongation factor1 transcript levels using the 2−ΔΔCt method. The relative expression levels did not significantly change between both internal controls. The relative values corrected by Ubiquitin are shown in the graphs of qRT polymerase chain reaction analysis. The fungal structures in the host roots were stained using the ink-staining method (Demchenko et al., 2004) or with a wheat germ agglutinin (WGA)-Alexa Fluor 594 conjugate (Invitrogen) (Harrison et al., 2002). Briefly, the root samples were heated in 5% KOH at 95°C for 1 h (for ink staining) or 15 min (for WGA-Alexa Fluor staining), and washed three times with water or phosphate-buffered saline (PBS). For ink staining, the roots were incubated at 95°C in a solution containing 3% black ink and 5% acetic acid for 20 min. For fluorescent staining, the roots were immersed in PBS containing WGA-Alexa Fluor 594 (1 μg ml−1 final concentration) and kept at room temperature for 1 h. Bright-field and fluorescence microscopy were performed with a stereomicroscope (SZX16; Olympus, Tokyo, Japan) and a confocal microscope (A1; Nikon). The root infected with AM fungi was stained with either ink or WGA-Alexa Fluor 594, and the frequency of colonization (the intercellular fungal structures per unit of root length) was calculated using the magnified intersection method (McGonigle et al., 1990). Finally, the number of hyphal entries in the infected roots was counted and divided by the root length. NFR1 and LYS1 (NFRe), which are involved in root nodule symbiosis (Radutoiu et al., 2003; Murakami et al., 2018), LYS2, LYS6 (CERK6), and LYS7 are close homologs of OsCERK1 in L. japonicus (Fig. 1a). Of these, LYS6 and LYS7 showed a high amino acid identity with OsCERK1 (59.2% and 57.4%, respectively). Furthermore, qRT polymerase chain reaction of these LysM receptors in the roots infected with AM fungi (Rhizophagus irregularis) revealed LYS7 upregulation in the host root, indicating potential involvement of LYS7 in AM (Fig. 1b). Previous studies involving transcriptome analyses of the roots infected with AM fungi also showed either LYS7 induction and no upregulation of the other LysM receptors (Handa et al., 2015; Takeda et al., 2015). Next, we created lys6 and lys7 single mutants and lys6lys7 double mutants using the CRISPR/Cas9 system in L. japonicus MG-20 Miyakojima to assess the function of LysM receptors in AM (Figs S1–S3). The AM fungi-infected roots of the single mutant lines exhibited significantly reduced hyphal colonization and arbuscule formation than WT roots (Figs 1c, S4). This result was consistent with the finding of a previous study reporting the involvement of OsCERK1 and MtLYK9, a close homolog of LYS6, in AM responses in O. sativa and Medicago truncatula, respectively (Miyata et al., 2014; Gibelin-Viala et al., 2019). Another study reported that OsRLK10 (OsCERK2) knockout in rice or RNAi knockdown of SlLYK12 in tomato notably reduced AM fungi colonization in the infected roots (Liao et al., 2018; Miyata et al., 2022). Both proteins had a high amino acid sequence identity with LYS7 in L. japonicus (52.0% and 65.5%, respectively). These findings suggested that LYS6 and LYS7 play a key role in AM in L. japonicus. However, despite reduced fungi colonization in the roots of single mutants compared with WT roots, the single mutants exhibited normal inner hyphae morphology and arbuscule formation (Figs 1d, S5). By contrast, the roots of double mutant plants did not exhibit fungal infection in the roots at 4 wk (Figs 1c, S6a,b) and 6 wk (Fig. S6c,d) after infection with AM fungi. In the double mutant plants, AM fungal hyphae were attached to the root surface, and the morphology of the hyphae was comparable to that in the WT roots (Fig. S7). However, the double mutant roots did not exhibit any hyphal colonization, similar to common symbiosis mutants such as pollux and ccamk (Figs 1c, S4) (Parniske, 2008). The similar phenotype was also observed in the double mutants of the M. truncatula orthologs MtLYK8 and MtCERK1 (Zhang et al., 2024). OsCEKR1 and OsRLK10 (OsCERK2) are considered functional homologs of LjLYS6/LjLYS7; however, the Oscerk1/Oscerk2 double mutant does not exhibit an additive phenotype (Miyata et al., 2022). The LysM receptor genes are considered rapidly evolving genes, and differences in the number of LysM receptor genes have been reported among M. truncatula ecotypes (Luu et al., 2022). These findings suggest that the mechanisms underlying LysM receptor-mediated AM signaling pathway might differ between the monocotyledonous rice and the dicotyledonous L. japonicus. In addition, the additive phenotype of the double mutants suggests that LYS6 and LYS7 regulate AM fungi infection via genetically distinct pathways. However, the impaired fungal infection in the double mutants might be attributed to the redundancy in the functions of these two receptors. Therefore, we investigated the differences in the symbiotic phenotypes of lys6 and lys7 mutants. A previous study reported that the hyphal colonization in the lys6 mutant root was not significantly different from the WT in the longer term (6 wk) infection with AM fungi (Bozsoki et al., 2017). Therefore, we speculated that AM fungal infection in the mutants was aborted in the early stages. To examine the effects of loss-of-function mutations in LYS6 and LYS7 on the initial stage of AM fungi infection, we assessed the number of hyphopodia and the number of hyphal entries into the host roots (Fig. 1e). The number of hyphopodia in the single mutant roots was comparable to that of the WT. Moreover, no morphological abnormalities of hyphopodia, inner hyphae, or arbuscules were observed in the roots of any single mutants (Fig. S5a–c). However, the lys6 mutants exhibited significantly fewer hyphal entries, with a lower entry/hyphopodia ratio than in WT (Fig. 1f). This result indicated that LYS6 knockout restricted hyphae penetration into the host root, suggesting that LYS6 mediates hyphal entry into the host root via the hyphopodia. By contrast, the entry/hyphopodia ratio for lys7 mutants was comparable to that for WT (Fig. 1e,f). The phenotypic difference between lys6 and lys7 mutants during the early stage of infection demonstrates the distinct function of these receptors in AM-related signaling pathways. The expression of AM-induced genes, including RAM1, RAM2, STR1, SbtM1, PT4, VAPYRIN1(VPY1), and EXO70I, was analyzed in mutant roots inoculated with AM fungi (Figs 1g, S8). None of these genes were induced in the lys6lys7 double mutant, reflecting the absence of an infection phenotype. On the contrary, RAM1, RAM2, STR1, SbtM1, and PT4 were induced in the single mutants after AM fungi infection. In the lys7 mutants, all the AM marker genes were induced by AM fungi infection (Fig. 1g); however, it exhibited significantly lower RAM1 and EXO70I expression levels than WT (Fig. S8). The decrease in expression could be reflected by the reduced hyphal colonization and arbuscule formation in the lys7 root. Notably, the lys6 mutants did not exhibit VPY1 and EXO70I induction, suggesting that VPY1 and EXO70I might be located downstream of the LYS6-mediated perception and signaling pathways. The normal arbuscule formation with PT4 induction in the lys6 mutants observed in this study was not consistent with the abnormal arbuscule development previously reported in the M. truncatula exo70I mutant (Fig. 1d,g) (Zhang et al., 2015). The basal expression level of EXO70I in the lys6 mutants may be sufficient to induce arbuscule formation. Meanwhile, VAPYRIN gene knockdown has been shown to reduce hyphal entry in M. truncatula (Pumplin et al., 2010). Although the lys6 mutants did not show the aberrant hyphopodia structures such as the swollen hyphopodia previously reported in the MtVAPYRIN knockdown plants (Fig. S5b), the reduced AM fungi infection in the lys6 mutants might be attributed to the lack of induction of VPY1 and other AM-induced genes regulated by the LYS6-mediated signaling pathway. Chito-oligosaccharides are AM signaling molecules, and LYS6 mediates CO perception and subsequent signaling pathways (Bozsoki et al., 2017). A previous report showed a lack of periodic oscillation in Ca concentration ‘Ca spiking’ in the rice Oscerk1 mutant (Carotenuto et al., 2017). Therefore, in this study, we analyzed the Ca spiking after CO perception in the mutants. Treatment with high concentrations of CO mixture (1–8 mers) induced Ca spiking in the lys7 mutants, with Ca spiking patterns and spike+ cell/total cell ratios comparable to those in WT (Fig. 2a,b). By contrast, the CO-treated lys6 roots did not exhibit a typical Ca spiking (Fig. 2a,b; lower pattern of lys6). Similarly, the lys6lys7 double mutant did not show a typical Ca spiking pattern by CO treatment (Fig. S9). The residual Ca concentration changes observed in the lys6 and lys6lys7mutants, which were counted in spiking + cells (Figs 2a,b, S9 top and middle pattern of lys6), differed from normal Ca spiking in the shape and frequency of Ca concentration increases, suggesting that they represent other physiological Ca responses in the root cells. A previous study using aequorin-based Ca tracing revealed an increase in nuclear Ca concentration in the lys6 mutants of L. japonicus by the CO treatment, which was suggested to be a CO-induced Ca spiking (Binci et al., 2024). However, this study also reported a marked reduction in the number of responding cells. Thus, this increase in nuclear Ca concentration would correspond to the residual Ca response in the lys6 mutant cells observed in our study (Fig. 2b). These results indicate that the lys6 mutant cells lost the ability to induce Ca spiking in response to CO treatment. Previously, rice has been shown to exhibit different Ca spiking responses between relatively short-chain chitin chains (tetramers or and long-chain chitin or (Genre et al., 2013). In this study, WT and lys7 mutants exhibited Ca spiking after treatment with chitin tetramer and octamer with no difference observed between Ca spiking patterns after and (Fig. 2b). This result indicated the of chitin did not the Ca spiking response in L. japonicus and that LYS7 is not involved in CO-induced Ca By contrast, the lys6 mutants did not exhibit any Ca spiking after either or treatment, indicating that LYS6 is involved in the or signaling of both and long-chain chitin and activates the signaling pathways to Ca the expression of genes that were also induced after AM fungi infection (Fig. Treatment with the CO mixture (1–8 mers) induced these genes in WT and lys7 mutants not in lys6 mutants (Fig. indicating the key role of LYS6 in and AM-induced gene that LYS6 (CERK6), involved in the pathogenic response to chitin (Bozsoki et al., is also involved in the expression of AM-induced genes and the induction of Ca LYS6-mediated symbiotic responses play an important role in the entry of AM fungi via the root The hyphal entry was comparable between lys7 mutants and WT, indicating that LYS7 might be involved in the infection after the hyphal entry (Fig. However, the residual entry of AM fungi in the lys6 mutants and the absence of the entry in the lys6lys7 mutants suggest of hyphal entry by symbiotic LysM receptors other than LYS6 and LYS7 are also involved in AM and chitin responses in L. japonicus (Rasmussen et al., 2016; et al., to the chitin receptor by OsCERK1 and LYS6 and LYS7 might form receptor the AM and chitin In addition, it be that the receptor and the AM signaling pathway. analysis is to the functions of LYS6 and LYS7 and their with other AM-related signaling In we demonstrated that the receptors LYS6 and LYS7 and their downstream signaling pathways are for the initial response to AM in L. japonicus. CRISPR/Cas9 vector was by Japan). and Japan) plant growth and of is by the for the of None this the mutant plants and and analyzed the mutants. The that the findings of this study are in the Supporting Information of this Fig. S1 CRISPR/Cas9 motif receptor Fig. of mutations in each mutant obtained by genome Fig. sequences around the site of each mutant obtained by genome Fig. Arbuscular mycorrhiza fungi colonization in the roots of lys6lys7 double mutants. Fig. Arbuscular mycorrhiza fungi structures in the roots of lys6 and lys7 mutants. Fig. Arbuscular mycorrhiza fungi colonization in the roots of lys6lys7 double mutants of Lotus japonicus. Fig. Arbuscular mycorrhiza fungi structures on the root of lys6lys7 double mutant of Lotus japonicus. Fig. analysis of genes in and lys6lys7 mutants. Fig. S9 spiking responses in lys6lys7 double mutants of Lotus japonicus. Fig. of and mycorrhiza genes in the roots of Lotus japonicus wild-type at 4 wk after AM fungi infection. Table S1 sets used in this study. is not for the or of any Supporting Information supplied by the than be to the The is not for the or of any supplied by the than be to the for the

Topics & Concepts