Mutation of MYB36 affects isoflavonoid metabolism, growth, and stress responses in Lotus japonicus

ORIGINAL RESEARCH Mutation of MYB36 affects isoflavonoid metabolism, growth, and stress responses in Lotus japonicus María Dolores Monje-Rueda 1 | Peter Pal'ove-Balang 2 | Kristina Trush 2 | Antonio J. Márquez 1 | Marco Betti 1 | Margarita García-Calder on 1 1 Departamento de Bioquímica Vegetal y Biología Molecular, Facultad de Química, Universidad de Sevilla, Sevilla, Spain 2 Institute of Biology and Ecology, Faculty of Science, P.J. ˇ Safárik University in Koˇ sice, Koˇ sice, Slovakia Correspondence Email: [email protected] Funding information MCIN/AEI/10.13039/501100011033/ and FEDER Una manera de hacer Europa, Grant/Award Number: PID2021-122353OB- I00; Ministerio de Ciencia, Innovaci on y Universidades, Agencia Estatal de Investigaci on and Fondo Europeo de Desarrollo Regional (FEDER), Grant/Award Number: RTI-2018-093571-B100; Ministry of education, science, research and sport of the Slovak Republic, Grant/Award Number: Project VEGA 1/0291/20 Edited by S.-.H. Li Abstract Isoflavonoids are mostly produced by legumes although little is known about why and how legumes are able to regulate the biosynthesis of these particular compounds. Understanding the role of potential regulatory genes of the isoflavonoid biosynthetic pathway constitutes an important topic of research. The LORE1 mutation of the gene encoding the transcription factor MYB36 allowed the identification of this gene as a regulator of isoflavonoid biosynthesis in Lotus japonicus plants. The levels of several isoflavonoid compounds were considerably lower in two lines of Ljmyb36 mutant plants compared to the WT. In addition, we found that Ljmyb36 mutant plants were significantly smaller and showed a substantial decrease in the chlorophyll levels under normal growth conditions. The analysis of plants subjected to different types of abiotic stress conditions further revealed that mutant plants presented a higher sensitivity than WT plants, indicating that the MYB36 transcription factor is also involved in the stress response in L. japonicus plants. 1|INTRODUCTION It is well established that flavonoids and isoflavonoids are important secondary metabolites which have rolesinplantgrowthanddevelopment, and in the defense against biotic and abiotic stress acting as phytoalexins (García-Calder on et al., 2020; Liu et al., 2015). Therefore, the study of the biosynthesis of flavonoids and isoflavonoids is a very important topic of research which, in addition, may have important consequences for human nutrition. For example, soybean isoflavonoids (isoflavones) have several beneficial effects on human health including the prevention of different types of cancers and cardiovascular diseases (Chu et al., 2017). (Iso)flavonoid compounds are of special interest because of their role in the nodulation process in legumes. They are crucial for the determination of the host range, since only a specific combination of (iso)flavonoids exudated by the legume root is able to attract the symbiont and activate NOD genes transcription in a compatible rhizobial strain (García-Calder on et al., 2020; Liu & Murray, 2016). It has been described that isoflavonoids are also involved in the systemic autoregulation of nodule number (Dong & Song, 2020). Different (iso)flavonoids can be used by the host in order to exert an inhibitory effect on incompatible rhizobial strains (Dong & Song, 2020). Although the main role for flavonoids or isoflavonoids in legume nodulation is to induce the Nod factor production by rhizobia, they also regulate other rhizobial responses that are important for symbiosis, including alterations in growth and motility, P and Fe acquisition and others (Liu & Murray, 2016 and references therein). Interestingly, isoflavonoids are mostly produced by legume species. However, little is known about why this group of plants produces these particular compounds and how they regulate them. The response is also quite species-specific, for example, while several legumes accumulate flavonols in response to abiotic stress, different Received: 12 August 2023 Revised: 11 October 2023 Accepted: 29 October 2023 DOI: 10.1111/ppl.14084 Physiologia Plantarum This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2023 The Authors. Physiologia Plantarum published by John Wiley & Sons Ltd on behalf of Scandinavian Plant Physiology Society. Physiologia Plantarum. 2023;175:e14084. wileyonlinelibrary.com/journal/ppl 1of12 https://doi.org/10.1111/ppl.14084 works from our group indicate that the model legume Lotus japonicus uses a peculiar strategy accumulating the isoflavonoid vestitol in response to different kinds of abiotic stresses (García-Calder on et al., 2015; Kaducová et al., 2019). A recent co-expression network was built in our laboratory to establish the possible interconnections between expression patterns of genes of the isoflavonoid biosynthetic pathway and other MYB and MYB-related genes, in order to search for putative regulatory genes of isoflavonoid biosynthesis (García- Calder on et al., 2020). By using this new gene co-expression network, six different MYB transcription factors (TFs) were identified. In the present paper mutants affected in one of these genes have been characterized phenotypically under different stress conditions. 2|MATERIALS AND METHODS 2.1 |Plant material, growth conditions and experimental designs The model legume L. japonicus (Regel) K. Larsen ecotype Gifu (B- 129-S9) was used in all experiments. Seeds were originally obtained from Jens Stougaard (Aarhus University, Denmark) and selfpropagated at the University of Seville. Homozygous mutant plants affected in the MYB36 gene were identified and selected from the progeny of different lines containing the LORE1 transposon inserts. WT and mutant seeds were scarified and surface sterilized and germinated in 1% (w/v) agar in Petri dishes. Phenotype characterization was determined with five seedlings planted in each pot using vermiculite as a solid support and grown in a growth chamber under 16/8 h day/ night, 20/18C with a photosynthetic photon flux density of 200 μmol m 2 s 1 and a constant humidity of 70%. Non-nodulated plants were irrigated with ‘Hornum’nutrient solution containing 5 mM NH 4 NO 3 and 3 mM KNO 3 . Nodulated plants were inoculated with the compatible symbiont Mesorhizobium loti and watered with nitrogen-free ‘Hornum’medium supplemented with 3 mM KCl (Handberg & Stougaard, 1992). Mesorhizobium loti TONO JA76 (Kawaguchi et al., 2002)wasgrowninliquidYM medium (Vincent, 1970)at28 C to an optical density of 1.0 at 600 nm, and then collected by centrifugation for 30 min at 2408 g and resuspended in 0.75% (w/v) NaCl. Once sown in the pots, the plants were inoculated by the addition of 200 μLofthisbacterial suspension. 15 plants per genotype were analyzed to determinate the growth patterns under non-symbiotic and symbiotic conditions. For the stress treatments, WT and mutant plants were irrigated with a ‘Hornum’nutrient solution and grown under the above conditions, with some exceptions according to the different experimental designs described below. UV-B radiation was applied together with the continuous visible-light illumination on plants that had been previously grown as described above for 30 days, in pots filled with vermiculite. Three biological replicates grown in the same pot, consisting of five plants each, were used for each genotype. Leaf tissue samples collected at the beginning of the experiment just before the UV treatment (t =0 h) were used as controls. Samples were exposed to UV-B treatment for 16 h using a UV-B G8T5E fluorescence lamp (Sayko Denki, 310 nm). The distance between the leaves and the lamp was approximately 23 ± 3 cm. The spectral output of the UV tubes at the top of the plant canopy was determined with a spectral radiometer (Sola-Scope 2000 from Solatell, resolution 0.5 nm) and weighted using the biological spectral weighting function (Flint & Caldwell, 2003); the bioactive UV-B dose was 0.68 mW cm 2 . Tissues for metabolite analysis were dried at 70C for 24 h. All leaves from five control or treated plants were collected and pooled for the different measurements carried out (Kaducová et al., 2019). All leaf samples used in this work were harvested 4 h after the beginning of the light period. For salt stress treatments, a single seedling was sown per pot four days after imbibition, and salt stress treatment was applied four days after sowing. In total, 180 plants per genotype were analyzed. The salt content in the nutrient solution was gradually increased by 50 mM for salinity acclimation in steps of four days till reaching lethal salt conditions (300 mM NaCl). Mortality rate was determined when the whole plant or all leaves were wilted or dead as described by Sánchez et al. (2011) with minor modifications. The experiment to determinate the irradiance sensitivity was performed by growing control plants at the standard photosynthetic flux density described above. Treated plants were provided with a higher light intensity corresponding to 300 μmol m 2 s 1 . All plants were grown in these conditions for 56 days. 30 plants were examined for each genotype. The sensitivity to high temperature was analyzed by increasing the temperature of the growth chamber to 25Cand30 C during the day photoperiod while maintaining 18C during the night. Control plants were grown under 20/18C day/night as mentioned above. All plants were grown in these conditions for 56 days. 15 plants were examined for each genotype. 2.2 |Identification of Ljmyb36 LORE1 mutants DNA extraction and cDNA synthesis methods were carried out as described by Pérez-Delgado et al. (2016). Ljmyb36 genotypes were detected through genotyping by PCR. Forward primers were used with LORE1-specific primer P2 to detect loci with LORE1 insertions. Reverse primers together with forward primers yielded a PCR product only when the locus they flank does not contain an insertion (Urba nski et al., 2012). Knockout lines were detected through forward and reverse specific primers. Real-time qPCR of cDNA samples from WT and mutant plants was performed as described by Kaducová et al. (2019). The housekeeping gene ubiquitin carrier protein 10 from L. japonicus (LjUBC10)wasused as cDNA loading control gene. All the primers used are listed in Table S2. 2.3 |Determination of chlorophyll content 0.05 g of fresh leaves were homogenized with 1 mL of 100% pure acetone. The extracts were centrifuged at room temperature for 2of12 MONJE-RUEDA ET AL. Physiologia Plantarum 13993054, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ppl.14084 by Universidad De Sevilla, Wiley Online Library on [23/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 10 min at 19 510 g. The absorbance was measured at 661.6 and 644.8 nm, and chlorophyll aand bcontents were calculated according to Lichtenthaler and Buschmann (2001). 2.4 |Metabolite Profiling Analysis For HPLC analysis, the leaves were dried for 24 h at 60C and stored in the dark in desiccator at room temperature until used. Previous studies showed that the incubation of leaf samples at this temperature did not significantly change the main flavonoid profiles (García- Calder on et al., 2015). 50 mg of dry tissue were extracted with 1.5 mL of 80% (v/v) methanol for analysis of flavonoid glycosides, phenylpropanoids and isoflavonoids. The extracts were centrifuged at 760 gfor 5 min and filtered using a 0.2 μm nylon membrane. Samples were analyzed by gradient reversed phase HPLC using an Agilent 1260 Infinity Quaternaly LC system with 1260 Infinity DAD detector and Kromasil C 18 250x4.6mmx5μm column (Nouryon). The solvent was delivered to the column at a 0.7 cm 3 min flow rate. A gradient system with two mobile phases was used: 5% (v/v) acetonitrile with 3% (v/v) trifluoroacetic acid (A) and 80% (v/v) acetonitrile (B). The gradient program utilized for determination of flavonoids, phenypropanoids, and isoflavonoids was as follows: 0 min A/B (90:10); 5 min A/B (86:14); 30 min A/B (76:24); 35 min A/B (60:40); 50 min A/B (0:100); 55 min A/B (0:100); 60 min A/B (90:10). The gradient program utilized for determination of isoflavonoids was as follows: 0 min A/B (90:10); 5 min A/B (60:40); 24 min A/B (0:100); 30 min (0:100); 35 min (90:10). Compounds were detected at 220 and 350 nm. Peak identities were identified based on their retention time and UV–VIS spectra measurements carried out during the analysis by comparison of commercially available standards of p-coumaric acid, p-ferulic acid, isoferulic acid, cinnamic acid, isoliquiritigenin, biochanin A, kaempferol-3-O-glucoside (Sigma Aldrich), quercetin-3-O-rhamnoside (Roth), vestitol (Biorbit), medicarpin (Plantech, UK) or using the standards previously isolated from L. japonicus leaves and identified by NMR methods (García-Calder on et al., 2015). The identities of some peaks were also verified according to Kaducová et al. (2019). For quantification, kaempferol-3-O-glucoside was used as a standard for MS identified kaempferol glycosides, quercetin-3-O- rhamnoside was used for MS identified quercetin glycosides, p-coumaric acid was used as a standard for coumaric acid isomers and vestitol was used as the reference for its derivative and sativan. All flavonols and isoliquiritigenin were evaluated at λ=350 nm, phenylpropanoids at λ=280 nm, vestitol, vestitol derivatives, sativan and medicarpin at λ=220 nm and expressed as μmol g 1 dry weight. FIGURE 1 Phylogenetic analysis of the LjMYB36 gene. (A) Phylogenetic tree of complete amino acid sequence from the sequence Lj4g3v2079370.1 from Lotus japonicus and other R2R3-MYB proteins from different plant species (Arabidopsis thaliana,Glycine max, Medicago truncatula and Trifolium pratense). MONJE-RUEDA ET AL.3of12 Physiologia Plantarum 13993054, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ppl.14084 by Universidad De Sevilla, Wiley Online Library on [23/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2.5 |Phylogenetic analyses A phylogenetic tree was constructed using the complete amino acid sequence from sequence Lj4g3v2079370.1 from L. japonicus together with those of R2R3-MYB proteins from different plant species. In this case, three other commonly used legume species (Glycine max,Medicago truncatula and Trifolium pratense) were analyzed, together with Arabidopsis thaliana as reference. The BLAST algorithm was used to select the 10 sequences most closely related to each of the plant species analyzed. These sequences were identified by searching the genome in specific databases. SoyBase (https://www.soybase.org/)forG. max,TAIR (https://www.arabidopsis.org/)forA. thaliana,LotusBase(https://lotus. au.dk/)forL. japonicus,andPlantTFDB(http://planttfdb.gao-lab.org/)for the rest of the species. The amino acid sequences were aligned using the MUSCLE multiple alignment method in MEGA11 ver. 11.0.13 (Tamura et al., 2021). Phylogenetic relationships between sequences and tree reconstruction were inferred by using the Maximum Likelihood method and the w/freq. model (Jones et al., 1992). The percentage of trees in which the associated taxa are clustered is shown beside the branches. The initial tree for the heuristic search was obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the JTT model, and then selecting the topology with superior log likelihood value. A discrete Gamma distribution was used to model evolutionary rate differences among sites (five categories [+G, parameter =1.1158]). The rate variation model allowed some sites to be evolutionarily invariable ([+I], 8.61% sites). This analysis involved 45 amino acid sequences. All positions with less than 95% site coverage were eliminated, i.e., fewer than 5% alignment gaps, missing data, and ambiguous bases were allowed at any position (partial deletion option). 2.6 |Statistical data analysis Significant differences between control and mutants under the different growth conditions analyzed were determined for each genotype by ANOVA (p < 0.05) with a post-hoc Turkey HSD test. 3|RESULTS 3.1 |Isolation of LORE1 mutants affected in the L. japonicus MYB36 gene Previous results established a high degree of interconnections between different transcription factor genes from L. japonicus and TABLE 1 Percent amino acid (aa) and DNA sequence identity of Lj4g3v2079370.1 with the most closely related MYBs from A. thaliana ID Gene name % Identity (aa) % Identity (DNA) At5g57620.1 AtMYB36 58.57 63.21 At4g37780.1 AtMYB87 52.65 53.53 At3g49690.1 AtMYB84 50.80 55.40 At5g65790.1 AtMYB68 47.02 54.50 At2g36890.1 AtMYB38 45.56 52.37 At2g36890.2 AtMYB38 45.56 52.37 At5g23000.1 AtMYB37 44.94 49.47 At5g56110.1 AtMYB103 40.00 45.67 At4g05100.1 AtMYB74 39.83 43.63 At4g28110.1 AtMYB41 39.07 43.51 FIGURE 2 Molecular analysis of Ljmyb36 mutants. (A) Mapping of LORE1 insertions in Ljmyb36-1 and Ljmyb36-2 mutants, (B) genotyping of DNA products by PCR to detect genotypes of Ljmyb36, (C) PCR of cDNA to confirm Ljmyb36 gene knockout, (D) PCR carried out to confirm the correct extraction of cDNA using LjUBC10, a housekeeping gene, and (E) real-time qPCR of cDNA samples from WT and mutant plants. M ladder, 1: WT, 2: Ljmyb36-1,3:Ljmyb36-2. 4of12 MONJE-RUEDA ET AL. Physiologia Plantarum 13993054, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ppl.14084 by Universidad De Sevilla, Wiley Online Library on [23/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License several genes of the isoflavonoid biosynthetic pathway from this plant, based on a co-expression analysis (García-Calder on et al., 2020). One of the sequences identified in this co-expression network (Lj4g3v2079370.1) was initially assigned to the LjMYB84 gene based on database descriptions. In the present paper the Figure 1shows a detailed phylogenetic analysis carried out comparing the complete amino acid sequence from this TF to other homologous amino acid sequences from Arabidopsis thaliana and different legume species. This analysis showed that Lj4g3v2079370.1 was grouped in a clade which included the corresponding gene from the other legume species analyzed, most closely related, together with one particular gene from Arabidopsis (Figure 1A). The results obtained indicate that among all Arabidopsis sequences analyzed, the sequence Lj4g3v2079370 from L. japonicus shows the highest degree of sequence identity of around 58–63% to the At5G57620.1 sequence, corresponding to the AtMYB36 gene (Table 1). Homozygous mutant lines affected in the different TF genes from L. japonicus determined in the co-expression network were identified by genotyping the self-fertilization progeny of the LORE1 mutant seed resource available for these particular genes in our laboratory (Małolepszy et al., 2016). A preliminary analysis of these mutant plants subjected to stress conditions showed that the mutant plants affected in the MYB36 gene presented a higher sensitivity to stress than the other mutant plants tested. For this reason, MYB36 mutant plants were selected to study the role of this gene in L. japonicus plants under different growth conditions. Figure 2A indicates the position of the two different mutations corresponding to the two mutant lines studied in the present work. For the rest of the paper, mutant line 30120982 will be called Ljmyb36-1 while mutant line 30106994 will be Ljmyb36-2,respectively, according to standard nomenclature given for this plant (Stougaard et al., 1999). PCR analysis further confirmed that both mutant lines show the presence of LORE1 insertion where expected (Figure 2B) and lack of the corresponding MYB36 transcripts (Figure 2C) while having normal levels of the LjUBC10 gene, used as a control by cDNA extraction and loading in the gels (Figure 2D). No expression of LjMYB36 gene in mutant plants was confirmed by real-time qPCR of cDNAs (Figure 2E). Therefore, these mutant lines were considered complete knockout mutants for further studies. 3.2 |Ljmyb36 mutants are affected in isoflavonoid biosynthesis Previous studies have reported that UV-B irradiation increases the levels of isoflavonoids in L. japonicus (García-Calder on et al., 2015; Kaducová et al., 2019). Therefore, Ljmyb36-1 and Ljmyb36-2 mutants were subjected to UV-B treatments and the determination of the levels of different flavonoids and isoflavonoids was carried out. At 16 h of UV-B treatment, important differences in the levels of some compounds were detected, particularly in the levels of isoflavonoids isoliquiritigenin and vestitol, which were substantially reduced by more than 2-fold in both mutant lines, compared to the WT (Figure 3). A full list of the levels of metabolites can be found in Supplemental Table 1, where all significant differences of mutants with regard to the WT are shown, as well as the differences obtained before and after UV-B treatment which increased specifically the levels of these isoflavonoids. These results indicate that the mutation of the transcription factor MYB36 from L. japonicus resultsinanalterationofisoflavonoidlevels,thussuggesting that this TF may be involved in the regulation of the biosynthesis of these compounds. However, the measurements carried out before UV-B treatment indicates that the levels of a coumaric acid derivative and several kaempferol glycosides were also significantly diminished in mutant plants compared to WT. In addition, increased levels of sativan and a quercetin glycoside were obtained in Ljmyb36 mutant plants in the absence of UV-B irradiation. FIGURE 3 Scheme of major changes observed in flavonoids / isoflavonoids biosynthetic pathways in Ljmyb36 mutants. Samples were analyzed after 16 h of UV-B irradiation treatment. In bold are the compounds with significant increase or decrease. These changes are highlighted with arrows. Abbreviations: hex, hexose; pen, pentose; 6DH, 6-deoxyhexose. MONJE-RUEDA ET AL.5of12 Physiologia Plantarum 13993054, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ppl.14084 by Universidad De Sevilla, Wiley Online Library on [23/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License FIGURE 4 Growth patterns under non-symbiotic conditions of WT and Ljmyb36 mutant plants. (A) Phenotype of WT and Ljmyb36 mutants after eight weeks of growth, (B) shoots and (C) roots fresh weight at different times of growth. Growth patterns under symbiotic conditions of WT and Ljmyb36 mutant plants. (D) Phenotype of WT and Ljmyb36 mutants after eight weeks of growth, (E) roots and (F) shoots fresh weight at different times of growth, (G) fresh weight of nodules per nodule, (H) fresh weight of total plant nodules per fresh weight of roots. Values are the means ± SE of 15 plants per genotype. Letters indicate significant differences (P < 0.05) between treatments according to ANOVA with a posthoc Tukey HSD test. 6of12 MONJE-RUEDA ET AL. Physiologia Plantarum 13993054, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ppl.14084 by Universidad De Sevilla, Wiley Online Library on [23/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 3.3 |Effects in growth of Ljmyb36 mutants The differences in growth observed for WT and Ljmyb36 mutant plants grown either under non-symbiotic or symbiotic conditions are shown in Figure 4. A much lower size of the plants was detected under nitrate-grown conditions (Figure 4A)affecting both the growth of roots (Figure 4B) and shoots (Figure 4C)which were significantly smaller in both Ljmyb36-1 and Ljmyb36-2 mutants compared to WT plants after five weeks of growth. A similar result was detected for plants grown in the absence of nitrate under dinitrogen fixing conditions as a result of symbiosis with Mesorhizobium loti (Figure 4D). In this case differences in growth were mostly observed at seven and eight weeks of growth (Figure 4E and F). These results indicate that the mutation of the LjMYB36 gene causes an important effect in the growth of the plants, independently of these plants being nodulated or nonnodulated. Interestingly, the fresh weight of the nodules produced by the plants was also substantially reduced, particularly at seven andeightweeksofgrowth(Figure4G). However, the fresh weight of nodules referred to root biomass was significantly increased (Figure 4H) due to the lower growth of the roots. Analysis of Ljmyb36 mutants was also carried out at a later stage of growth (three months). Figure 5illustrates that at this time of growth the mutant plants showed not only a reduction in growth but also an increase in the proportion of yellow leaves (Figure 5A). In fact, a substantial decrease in the levels of chlorophyll aand bwas detected in these plants (Figure 5B). Interestingly, Figure 5B also shows that even at one month of growth, a lower content of chlorophyll could be detected in Ljmyb36 mutants in comparison with the WT, although this change in chlorophyll is not properly noticeable by eye. In addition, Figure 5C shows that the number of leaves was reduced by 50% in older plants (three months grown), thus indicating that Ljmyb36 mutations result not only in retardation of growth, as mentioned above, but also in an early senescence phenotype. 3.4 |Ljmyb36 mutants show higher sensitivity to salt, light and thermal stress The response of Ljmyb36 mutants to a gradual increase in NaCl, carried out as previously reported for this plant species (Sánchez et al., 2011) is shown in Figure 6A. The results obtained indicate that a much higher proportion of dead plants were found at 200–300 mM NaCl for both lines of Ljmyb36 mutants, compared to the WT. In fact, more than 50% of mutant plants were dead at the beginning of the 300 mM NaCl treatment, while no significant death was observed for the WT plants. Response of Ljmyb36 mutants to an increase in light irradiance was also analyzed. A much higher proportion of dead plants were obtained when mutant plants were provided with 300 μmol m 2 s 1 light intensity for 56 days (Figure 6B). In fact, 60–80% of the mutant plants died as a consequence of this treatment, while only 10% of WT plants were affected. In contrast, only a minor proportion of dead plants could be detected under control standard illumination conditions. Growth of WT and Ljmyb36 mutants was comparatively examined at 20C, 25C and 30C (Table 2). The results obtained indicated an important reduction of shoot and root growth in the mutant plants compared to the WT when grown at 20C (standard growth conditions). However, when temperature was increased to 25C and 30C, a much lower growth rate was obtained for the mutants compared to the WT, depending on the temperature. FIGURE 5 Growth differences between WT and Ljmyb36 mutant mature plants. (A) Phenotype of WT and Ljmyb36 mutants after three months of growth, (B) chlorophyll aand bcontent in Lotus japonicus leaves at one month (left) or three months (right) of growth, (C) percent of leaves present in mutant plants at one or three months of growth with respect to WT plants. MONJE-RUEDA ET AL.7of12 Physiologia Plantarum 13993054, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ppl.14084 by Universidad De Sevilla, Wiley Online Library on [23/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 4|DISCUSSION The results shown in this paper indicate that in the model legume L. japonicus, two different knockout mutations of the MYB36 gene lead to significant alterations in the isoflavonoid biosynthetic pathway. These results are in accordance with the previous identification of this gene within a co-expression network of isoflavonoid genes together with other potential regulatory genes from this plant. In this co-expression network, it was found that the LjMYB36 gene is highly connected to the LjHID4 (2-hydroxyisoflavanone dehydratase) and LjI2’H2 (isoflavone-20-hydroxylase) genes, both involved in the isoflavonoid pathway (García-Calder on et al., 2020). The fact that isoflavonoid levels are altered in the mutant plants suggests that the MYB36 TF is involved in the regulation of isoflavonoid biosynthesis. In addition, the results obtained clearly support a major role of isoflavonoids in growth and abiotic stress protection in this plant, since mutants show an important retardation in growth as well as early senescence FIGURE 6 Response of the mutant lines to abiotic stress. (A) Lethality of WT and Ljmyb36 mutant plants under NaCl stress treatment. The stepwise increase in total NaCl added to each pot is shown at the top. Values are the means ± SE of 180 plants per genotype. (B) Increase in the proportion of dead plants after 56 days of growth under high irradiance versus control conditions. Values are the means ± SE of 30 plants per genotype and treatment. Letters indicate significant differences (P < 0.05) between treatments according to ANOVA with a post-hoc Tukey HSD test. TABLE 2 Changes in fresh weight of shoots and roots between WT and Ljmyb36 mutants in response to temperature increase. Values are the means of 15 plants per genotype and treatment FW Shoots FW Roots 20C25 C30 C20 C25 C30 C WT 100 100 100 100 100 100 Ljmyb36-1 55 43 22 40 24 11 Ljmyb36-2 49 37 35 42 18 13 8of12 MONJE-RUEDA ET AL. Physiologia Plantarum 13993054, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ppl.14084 by Universidad De Sevilla, Wiley Online Library on [23/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License symptoms, and a higher sensitivity to saline, high irradiance and temperature stresses. Previous works have indicated that stress conditions increase the levels of isoflavonoids in L. japonicus WT plants (García-Calder on et al., 2015; Kaducová et al., 2019). The observed alterations in the levels of different flavonoid compounds and precursors in Ljmyb36 mutants (Supplemental Table 1) indicate that LjMYB36 is not only involved in isoflavonoid biosynthesis control, but also in the levels of several flavonols and other steps of the phenylpropanoid metabolism, thus revealing a crucial importance of this TF. In plants, flavonoids carry out many different functions like regulating cell growth, attracting pollinator insects and protecting against biotic and abiotic stresses (Agati et al., 2012;Liu et al., 2021; Winkel-Shirley, 2001). Previous works have described that flavonoids may effectively control key steps of cell growth and differentiation, thus acting by regulating the development of the whole plant and individual organs (Agati et al., 2012). Quercetin modulates root growth by limiting cell proliferation and enhancing the cell elongation phase. It is assumed that the meristematic region of the roots faces high limitations of cell proliferation under quercetin treatment (Silva-Navas et al., 2016; Singh et al., 2021). The hormone auxin plays an important role in defining root meristem size through formation of the PLETHORA (PLT) gradient that regulates cell proliferation. Treatment of plant roots with quercetin reduced this gradient along with the meristem size, so quercetin content regulates meristem size through regulation of the auxin-PLT gradient (Silva-Navas et al., 2016). Therefore, the lower growth rate and those growth defects observed in Ljmyb36 mutant plants could be related with the deficiency in the MYB36 TF. However, the results obtained indicate no clear connection between the deficiency of the MYB36 gene and the nodule numbers in L. japonicus, since mutants are still able to nodulate in a similar pattern than the WT. Nevertheless, further studies of the infection process such as the quantification of the number of root hair infection threads and others are necessary to elucidate the involvement of this TF in the symbiosis process in L. japonicus plants. Recent work has been able to show that different phenolic acids, particularly ferulic acid, but not (iso)flavonoids, are a novel type of NOD gene inducers in L. japonicus-Mesorhizobium symbiosis (Shimamura et al., 2022). Since the levels of ferulic and other phenolic acids were found not to be altered in Ljmyb36 mutants, all these results could explain why the nodule number of the mutant plants may not be significantly affected in these mutants. Different TFs have been identified to regulate the phenylpropanoid pathway in higher plants, such as MYB, bZIP, WRKY, MADS box and WD40. Plant MYB proteins can be classified into four subfamilies according to the DNA-binding domains: the R1-MYB related group (one single domain), the R2R3-type group (two domains), the R1R2R3-MYB group (three domains) and a minor subfamily of MYB genes carrying four domains (Campos et al., 2016). A high number of genes encoding R2R3-MYB TFs have been identified in plants (Paz-Ares et al., 1987; Pucker et al., 2020; Stracke et al., 2001; Zheng et al., 2017). These TFs are the most abundant type and constitute one of the largest transcription factor families being involved in the regulation many biological processes like cell shape and identity, developmental control, secondary metabolite biosynthesis, hormone signal transduction and response to abiotic stresses and pathogen defense (Zheng et al., 2017). In legumes, less than 1% of these have been characterized genetically, although TFs likely played seminal roles in legume evolution and clearly now play crucial roles in plant development and differentiation (Udvardi et al., 2007). For this reason, it is of paramount importance to elucidate the role of every single TF in legume plants like in this study. In L. japonicus the involvement of different subgroups of R2R3MYB TF that act coordinately to induce the flux to isoflavonoids and/or reduce the flux of metabolites through competing branches of phenylpropanoid metabolism has been shown (Shelton et al., 2012). However, MYB36 TF has not been analyzed so far in this plant. Different roles of the MYB36 gene have been described in several works. Liberman et al. (2015) showed that the MYB36 TF from Arabidopsis plants activated genes involved in Casparian strip formation and represses genes involved in proliferation, suggesting that MYB36 is a critical regulator of developmental timing in the root endodermis by promoting the transition from proliferation to differentiation. Other authors demonstrated that MYB36 directly and positively regulates the expression of the Casparian strip genes CASP1 (Casparian strip domain proteins), PER64 (Peroxidase 64) and ESB1 (Enhanced Suberin 1) in Arabidopsis (Hosmani et al., 2013; Kamiya et al., 2015). ESB1 was suggested to mediate bimolecular coupling during the lignin biosynthesis (Hosmani et al., 2013). This sparks an intricate developmental program, in which the transcription factor MYB36 plays a role in the synthesis and deposition of lignin between endodermal cells (Li et al., 2018; Kamiya et al., 2015). Likewise, another study identified PpMYB36 as a regulator involved in lignin accumulation and russet coloration in Pyrus pyrifolia (Ma et al., 2021). Furthermore, MYB36 plays a critical role in restricting cell division at larger stages of lateral root primordia development by regulating a subset of PER genes, and perhaps other genes that modulate ROS balance and cell proliferation potential (Fernández-Marcos et al., 2016). In addition, through a new network with connected genes related to anthocyanin synthesis, a high co-expression between VvMYB36 with bHLH041 and WRKY75 was found, regulating four PAL genes and one 4CL gene (Zhong et al., 2022). However, effects in (iso)flavonoid metabolism, growth and stress responses of MYB36 mutants have not been described. The results shown in the present paper indicate that LjMYB36 is involved in controlling phenylpropanoid metabolism and, particularly, (iso)flavonoid biosynthetic pathways. We establish a major role of MYB36 in the biosynthesis of phenolic compounds and growth of L. japonicus plants. Consequently, MYB36 is a new interesting TF identified that can be added to the complex picture of TFs involved in the cooperative and homeostatic regulation of the biosynthesis of flavonoid and other phenolic compounds such as lignin, which constitute a very interesting topic of research (Shi et al., 2022; Zhang et al., 2021). In addition, it has been also reported that the MYB36 TF could be a key TF that may drive floral integrators for flowering time regulation in orchids (Ahmad et al., 2022). All these studies emphasize the importance of the MYB36 TF in plants. MONJE-RUEDA ET AL.9of12 Physiologia Plantarum 13993054, 2023, 6, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ppl.14084 by Universidad De Sevilla, Wiley Online Library on [23/02/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License