Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2020 Dec 23;229(6):3330–3344. doi: 10.1111/nph.17097

The antagonistic MYB paralogs RH1 and RH2 govern anthocyanin leaf markings in Medicago truncatula

Chongnan Wang 1,*, Wenkai Ji 1,*, Yucheng Liu 2, Peng Zhou 3, Yingying Meng 1, Pengcheng Zhang 1, Jiangqi Wen 4, Kirankumar S Mysore 4, Jixian Zhai 5, Nevin D Young 3, Zhixi Tian 2, Lifang Niu 1, Hao Lin 1,✉
PMCID: PMC7986808  PMID: 33222243

Summary

  • Patterned leaf coloration in plants generates remarkable diversity in nature, but the underlying mechanisms remain largely unclear.

  • Here, using Medicago truncatula leaf marking as a model, we show that the classic M. truncatula leaf anthocyanin spot trait depends on two R2R3 MYB paralogous regulators, RED HEART1 (RH1) and RH2.

  • RH1 mainly functions as an anthocyanin biosynthesis activator that specifically determines leaf marking formation depending on its C‐terminal activation motif. RH1 physically interacts with the M. truncatula bHLH protein MtTT8 and the WDR family member MtWD40‐1, and this interaction facilitates RH1 function in leaf anthocyanin marking formation. RH2 has lost transcriptional activation activity, due to a divergent C‐terminal domain, but retains the ability to interact with the same partners, MtTT8 and MtWD40‐1, as RH1, thereby acting as a competitor in the regulatory complex and exerting opposite effects. Moreover, our results demonstrate that RH1 can activate its own expression and that RH2‐mediated competition can repress RH1 expression.

  • Our findings reveal the molecular mechanism of the antagonistic gene paralogs RH1 and RH2 in determining anthocyanin leaf markings in M. truncatula, providing a multidimensional paralogous–antagonistic regulatory paradigm for fine‐tuning patterned pigmentation.

Keywords: anthocyanin leaf marking, Medicago truncatula, MYB regulator, paralog antagonism, patterned pigmentation

Introduction

Patterned pigmentation in plants provides tremendous coloration diversity in nature. As important phytoprotective pigments, anthocyanin imparts vivid coloration ranging from red and purple to blue, often in complex patterns, such as spots, stripes and vein pigmentation (Grotewold, 2006). In flowers and fruits, these patterns fulfill an important ecological function in attracting pollinators and seed distributors, therefore influencing reproductive success (Petroni & Tonelli, 2011). Anthocyanin pigmentations in vegetative tissues are often produced in response to abiotic stress, while spatially restricted anthocyanin patterns in leaves (leaf markings) are established during leaf development, which is relatively uncommon, and may function as an aposematic signal that deters predators (Lev‐Yadun, 2001), as in some poisonous ornamental plants such as Caladium, and also occur in many forage legumes, such as some genera in Trifolium (Carnahan et al., 1955; Corkill, 1971) and most annual Medicago species (McComb, 1974). Since the 1950s, leaf marking traits have been widely used for the identification of cultivated varieties and the detection of varietal crosses/contamination, and their inheritance has been well studied in Trifolium and Medicago (Carnahan et al., 1955; Corkill, 1971; McComb, 1974), yet the underlying molecular basis remains largely unclear.

Previous studies have indicated that the MYB transcription factors (TFs) play a critical role in the regulation of anthocyanin biosynthesis at the transcriptional level. This regulation is coordinated by interacting with basic helix–loop–helix (bHLH) and WD‐repeat (WDR) proteins to form a conserved MYB‐bHLH‐WDR (MBW) complex (Baudry et al., 2004; Koes et al., 2005; Ramsay & Glover, 2005; Xu et al., 2015). The patterning and spatial localization of anthocyanins are determined primarily by the MYB TFs, with individual gene‐family members regulating separate patterns (Albert et al., 2011, 2014; Davies et al., 2012; Ding et al., 2020). Recently, an understanding of the role of MYB TFs in regulating patterned anthocyanin pigmentation in leaves has begun to emerge. In Trifolium repens, by taking a candidate gene approach, a family of R2R3‐MYB genes was found to be associated with the classic R‐ and V‐anthocyanin pigmentation loci, which correspond to several distinct anthocyanin leaf patterns (Carnahan et al., 1955; Corkill, 1971; Albert et al., 2015). In M. truncatula, the LAP1 MYB TF orchestrates anthocyanidin biosynthesis, and overexpression of LAP1 leads to the accumulation of anthocyanin pigments in the whole M. truncatula leaves (Peel et al., 2009). Two MYB homologs, MtMYBA and AN2, which regulate genes involved in the anthocyanin biosynthetic pathway, have been implicated in anthocyanin spot accumulation in M. truncatula leaves under normal conditions (Carletti et al., 2013). Despite these reports, the MYB TFs underlying specific leaf markings have not been conclusively identified and/or verified, and the molecular mechanisms of MYB TFs in the regulation of leaf markings are still not well understood.

Medicago truncatula is an annual legume species with a wide range of leaf markings (e.g. red spot markings). It has been reported that the M. truncatula leaf with a red/purple mark have anthocyanins in the upper epidermis, while the yellow color is probably due to reduced Chls and carotenoids (McComb, 1974). The availability of abundant leaf marking mutant resources in M. truncatula provides an ideal system for studying the underlying mechanisms of complex pigmentation patterns in vegetative tissues. In this study, we demonstrate that the distinct anthocyanin leaf spot marking of M. truncatula ecotype R108 is controlled by two uncharacterized MYB homologous genes, RED HEART1 (RH1) and RH2. RH1 acts as a master regulator of leaf marking and positively regulates the formation of anthocyanin spots in M. truncatula leaves, whereas the inactive RH2 exerts an opposite effect by competing with the RH1 protein while also repressing RH1 expression, suggesting a multidimensional paralogous–antagonistic regulatory paradigm for fine‐tuning the complex patterned pigmentation in M. truncatula leaves.

Materials and Methods

Plant materials and growth conditions

Medicago truncatula ecotypes R108 were used for all experiments described in this study. Mutant lines rh1d (NF7599), rh1‐1 (NF11022), rh1‐2 (NF19601), rh2‐1 (NF21135), rh2‐2 (NF21515), mttt8 (NF15995) and mtwd40‐1 (NF11228) were identified from the Tnt1 retrotransposon‐tagged mutant collection of M. truncatula R108 (Tadege et al., 2008). The M. truncatula Jemalong A17 (A17), which has been used for the whole‐genome sequencing project (Choi et al., 2004; Young et al., 2011), was used for generating the mapping population. Primers used for genotyping are listed in Supporting Information Table S1. Scarified M. truncatula seeds were germinated overnight in moist Petri dishes and placed at 4°C in the dark for 1 wk. Plants were grown at 24°C : 22°C, day : night temperatures with a 16 h : 8 h, day : night photoperiod and 60–70% relative humidity. Tobacco (Nicotiana benthamiana) seeds were sown into the soil and grown in the glasshouse for 1 month before being used for the subcellular localization assay.

Genetic mapping

F2 mapping populations were generated by self‐pollination of F1 plants derived from crosses between rh1d and M. truncatula ecotype A17. The RH1 locus was mapped primarily with insertion/deletion (InDel) markers, using 3777 F2 mutant plants. Developed molecular markers used are listed in Table S1.

BSA‐seq analysis

For bulk segregant analysis with whole‐genome sequencing (BSA‐seq) analysis, two bulked pools from 30 F2 wild‐type and mutant plants each, derived from crosses between rh1d and wild‐type R108, were sequenced on an Illumina Hi‐Seq 2500 platform. To ensure that reads were reliable and without artificial bias in the following analyses, raw data (raw reads) of fastq format was first processed through a series of quality control (QC) procedures. QC standards were as follows: removing reads with ≥ 10% unidentified nucleotides (N); removing reads with > 50% bases having phred quality < 5; removing reads with > 10 nt aligned to the adapter, allowing ≤ 10% mismatches; and removing putative PCR duplicates generated by PCR amplification in the library construction process (reads 1 and 2 of two paired‐end reads that were completely identical). The Burrows–Wheeler Aligner (Li & Durbin, 2009) was used to align the clean reads of each sample against the reference genome R108_v1.0 (http://www.medicagohapmap.org/downloads/r108). A 10 kb sliding window without overlap was used to present the sequencing depth of the whole genome. The average of all sequencing depths in each window was used as the depth for this window. The difference in sequencing depth of two progeny pools was calculated as the delta depth.

Synteny analysis

Synteny blocks between R108_v1.0 and A17_v5.0 (https://medicago.toulouse.inra.fr/MtrunA17r5.0‐ANR/) Chromosome 7 in this study were computed using mummer tools (Marçais et al., 2018). We used nucmer4 to calculate the synteny blocks with default parameters, and filtered the synteny blocks via a delta‐filter with parameter −1. The synteny blocks were visualized by symap v.4.2. (Soderlund et al., 2011).

Chromosomal walking and deletion border determination

Chromosomal walking was performed using PCR to detect the deletion in the rh1d mutant using the primers listed in Table S1. The deletion border was determined by sequencing the PCR products. All sequence‐specific primers were designed using primer3web v.4.1.0 (http://primer3.ut.ee/), and annealing (T m) temperatures ranged from 55 to 60°C. PCRs were conducted using 2 × Taq PCR Master Mix (#PM604‐1; Uptech) according to the manufacturer’s instructions, and the PCR amplicons were examined by electrophoresis using a 1% (w/v) agarose gel.

Sequence alignment and phylogenetic analysis

Full amino acid sequences of the RH1, RH2 and other R2R3‐MYB proteins were aligned using clustalw (http://www.genome.jp/tools/clustalw/). Bootstrap values of 1000 permutations for the maximum likelihood phylogenetic tree were generated by the mega 7.0 software using the Jones–Taylor–Thornton (JTT) model.

Expression profiling analysis

RNA‐seq libraries were constructed and sequenced on an Illumina NovaSeq platform. Samples from the folded leaves of rh1d and wild‐type at 3 wk old were sheared for RNA‐seq analysis. Three biological replicates of plant tissue were analyzed, and each one was mixed by three seedling leaves. An average of 6 Gb of data was generated for each sample. The RNA‐seq reads were mapped to the Medicago genome sequences MtrunA17r5.0 using hisat2 v.2.0.6 (Kim et al., 2015) with default parameters. The expression level of each predicted transcript in each RNA‐seq library was calculated as the FPKM with htseq v.0.11.2 (Anders et al., 2015). Differentially expressed genes (DEGs) between two groups were identified using the edger package v.3.16.5 (Robinson et al., 2010).

Gene expression analysis

RNA samples were isolated from the dissected leaves of 3‐wk‐old plants using TRIzol reagent (Invitrogen). For each biological replicate, at least 3–5 leaves were harvested. cDNA was synthesized by reverse transcription with TransScript‐Uni One‐Step gDNA Removal and cDNA Synthesis SuperMix (TRAN, #AU311). Quantitative reverse transcriptase PCR (qRT‐PCR) was performed as previously described (Meng et al., 2019), with three biological replicates. Gene expression was normalized using the expression of the housekeeping gene MtActin. Relative gene expression for each gene in the mutant plants or transgenic lines was compared with that obtained for the wild‐type, which was arbitrarily set to 1.0. All primers used are listed in Table S1.

Generation of CRISPR‐Cas9‐induced mutants

The CRISPR‐Cas9‐induced mutations were created as previously described (Meng et al., 2017). Guide RNAs (gRNAs) targeting RH1 were designed using crispr‐p (http://cbi.hzau.edu.cn/cgi‐bin/CRISPR), and an array containing gRNA expressed from the M. truncatula U6 promoter was synthesized, with gRNAs GATGATAAAGTTATAGGCCG or AAATTCTCCCGGTTGATGTA. This array was cloned into the pFGC5941‐Cas9 binary vector and introduced into rh1d by Agrobacterium tumefaciens‐mediated transformation. Mutants were identified by Sanger sequencing. The primers used are listed in Table S1.

Transgene construction and plant transformation

The coding sequence (CDS) of RH1 or RH2 was cloned into the vector pEarleyGate203 or pEarleyGate202 using the Gateway cloning system (Invitrogen), resulting in the constructs pEarleyGate203‐RH1 or pEarleyGate202‐RH2. The destination constructs were introduced into A. tumefaciens by electroporation. Agrobacterium tumefaciens strain AGL1 was used for M. truncatula transformation as described by Meng et al. (2017). The primers used are listed in Table S1.

Transactivation activity assay in yeast

The transactivation activity assay was carried out as previously described (Meng et al., 2019). Different domains of RH1, RH2 and RH21‐112 + RH1113‐241 were fused with the GAL4 DNA‐binding domain (BD) in the plasmid pGBKT7‐GW using a Gateway recombination system. These constructs were then transformed into the yeast strain Gold according to the instructions for the Frozen‐EZ Yeast Transformation II Kit (#T2001; Zymo Research). Yeast colonies were patched onto SD/‐Trp (‐W) and SD/‐Trp/‐His/‐Ade (‐W‐H‐Ade) plates and grown at 28°C for 3 d. The α‐galactosidase assay was performed according to the Yeast Protocols Handbook (#PT3024‐1; Clontech, Palo Alto, CA, USA). The primer sequences are listed in Table S1.

Trans‐activation assay in Arabidopsis protoplasts

The transient dual‐luciferase assay was performed according to a previously described procedure with minor modifications (Meng et al., 2019). The effector plasmids were constructed by cloning the RH1, RH2, MtTT8 and MtWD40‐1 CDSs into the vector pEarleyGate203 using the Gateway cloning system (Invitrogen), resulting in the constructs pEarleyGate203‐RH1, pEarleyGate203‐RH2, pEarleyGate203‐MtTT8 and pEarleyGate203‐MtWD40‐1, respectively. Plasmid pEarleyGate203‐GFP was used as the negative control. The c. 3 kb promoter fragments upstream of the transcription start sites of MtCHS, MtANS, RH1 and RH2 were cloned into the vector pGreenII‐0800‐Luc using the In‐Fusion cloning strategy (Clontech) to generate the corresponding reporter constructs. For the transcriptional activation assay, the reporter GAL4‐LUC plasmid was constructed as described previously (Lin et al., 2013). For effector plasmids, the CDSs of RH1, RH2 and truncated form RH21‐112 + RH1113‐241 were first cloned into pGBKT7. The coding regions of BD fusion were amplified using specific primers and cloned into the destination vector p2GW7 (Wang et al., 2010) using the Gateway system (Invitrogen) to yield effector plasmids. Arabidopsis protoplasts were cotransformed with different combinations of plasmids, incubated for 12–14 h in darkness, and then collected and lysed for the detection of luciferase activity. Detection was performed according to the manufacturer’s recommendations with the Dual‐Luciferase Reporter Assay System (Promega, E1910). The primers used are listed in Table S1.

DNA‐binding assay

The DNA‐binding assay was performed as previously described (Meng et al., 2019). The CDSs of RH1 and RH2 were amplified and inserted into the BamHI and XhoI sites of the pGEX‐4T‐1 vector using the In‐Fusion cloning strategy (Clontech). The recombinant constructs were transformed into Escherichia coli strain BL21 and induced with 0.2 mM isopropyl‐1‐thio‐d‐galactopyranoside (IPTG). Recombinant GST‐RH1, GST‐RH2 and GST (glutathione S‐transferase, control) were purified using Glutathione Sepharose 4B (#17‐0756‐01; GE Healthcare, Piscataway, NJ, USA) according to the manufacturer’s protocol and quantified with the GE Healthcare protein assay reagent. The MYB‐core or AC‐rich element binding fragment was incubated with the GST‐RH1, GST‐RH2 and GST proteins, respectively, in Glutathione Sepharose and then the DNA‐binding activity (protein‐bound DNA) was determined by qRT‐PCR after washing and elution. The primers used are listed in Table S1.

Electrophoretic mobility shift (EMSA) assay

The direct binding of RH1 and RH2 to the MtCHS and MtANS promoter was detected using an EMSA kit (GS009; Beyotime, Shanghai, China) following the manufacturer’s protocol with probes. The CDSs of RH1 and RH2 were cloned into the E. coli expression vector pCOLD‐TF with His tag (Takara, Shiga, Japan) using EcoRI and BamHI restriction enzymes. The recombinant constructs were transformed into E. coli strain BL21 and induced with 0.2 mM IPTG. Recombinant His‐TF‐RH1, His‐TF‐RH2 and His‐TF (control) were purified using Profinity IMAC Ni‐charged Resin (no. 1560131; Bio‐Rad) according to the manufacturer’s protocol and quantified by the Bio‐Rad protein assay reagent. The 3′biotin‐labeled oligonucleotide probes were synthesized and labeled by the Shanghai Sangon Company (Shanghai, China). Probe information is given in Table S1.

Yeast two‐hybrid (Y2H) assay

Y2H assays were performed as previously described (Meng et al., 2019). The CDSs of RH1, RH2, MtTT8 and MtWD40‐1 were amplified and then inserted into the bait plasmid pGBKT7‐GW or the prey plasmid pGADT7‐GW. All clones were validated by sequencing. The bait and prey plasmids were cotransformed into yeast strain Gold using the Frozen‐EZ Yeast Transformation II Kit (#T2001; Zymo Research, California City, CA, USA). Yeast colonies were patched onto SD/‐Leu/‐Trp (DDO, Double Dropout) and SD/‐Trp/‐Leu/‐His/‐Ade (QDO, Quadruple Dropout) plates and grown at 28°C for 3 d. The α‐galactosidase assay was performed according to the Yeast Protocols Handbook (#PT3024‐1; Clontech). The primer sequences are listed in Table S1.

Bimolecular fluorescence complementation (BiFC) assay

BiFC assays were performed as described previously (Lu et al., 2010; Meng et al., 2019). The CDSs of RH1, RH2, MtTT8 and MtWD40‐1 were cloned into the pEarleyGate 201‐YN or pEarleyGate 202‐YC vectors (Lu et al., 2010) using a Gateway recombination system. The prepared vectors were introduced into A. tumefaciens strain GV2260. Combinations of YN and YC plasmids together with plasmids mRFP‐AHL22 and P19 were agroinfiltrated into the leaves of 4‐wk‐old N. benthamiana plants. Signals were observed 3 d after infiltration by confocal microscopy (TCS SP2 microscope; Leica, Wetzlar, Germany). The primers used for plasmid construction are listed in Table S1.

Co‐immunoprecipitation (Co‐IP) assay

The Co‐IP assay was performed as previously described with minor modifications (Meng et al., 2019). The CDSs of RH1, RH2, MtTT8 and MtWD40‐1 were cloned into either pGWB17‐GW‐Myc or pGWB11‐GW‐Flag vectors, resulting in pGWB17‐MtTT8‐Myc, pGWB17‐MtWD40‐1‐Myc, pGWB11‐RH1‐Flag and pGWB11‐RH2‐Flag. To test for protein–protein interactions, A. tumefaciens strain GV2260 containing pairs of these constructs together with P19 was co‐infiltrated into 4‐wk‐old N. benthamiana leaves. Equal amounts of samples (0.3 g) were collected 3 d after infiltration, ground in liquid nitrogen and then homogenized in 1 ml of extraction buffer (50 mM Tris‐HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X‐100, 0.1% Tween 20, 10% glycerol, 1 mM phenylmethylsulfonyl fluoride (PMSF) and one tablet/50 ml of protease inhibitor cocktail). The lysates were incubated at 4°C for 15 min and centrifuged at 16 000 g for 10 min at 4°C. The supernatant was precleared with 30 µl of Dynabeads Protein A (#10001D; Novex) at 4°C for 1 h. After a brief spin, the supernatants were incubated with 30 µl of suspensions of Anti‐DYKDDDDK‐Tag mAb (Agarose Conjugated) (#M200018; Abmart) at 4°C for 4 h and then washed five times with the extraction buffer. The proteins were eluted from the beads with 30 µl of 6× Protein Loading Buffer (#L10215; TRAN), boiled for 5 min, and spun at 13 500 g for 10 min at room temperature. The supernatants were electrophoretically separated by 10% SDS‐PAGE and transferred to a nitrocellulose membrane (#A10190852; GE Healthcare). Immunoblots were performed using an anti‐Myc antibody (#M20002; Abmart) for probing MtTT8‐Myc or MtWD40‐1‐Myc and an anti‐Flag antibody (#M20008; Abmart) for probing RH1‐Flag or RH2‐Flag, sequentially. The primer sequences are listed in Table S1.

Anthocyanin analysis

Anthocyanin content was detected as previously described (Pang et al., 2009). In total, 50 mg of different regions of the wild‐type leaflets (from at least 50 plants) was added to 500 µl of 0.1% (v/v) HCl/methanol, which was allowed to stand overnight on a rotating wheel at 4°C in the dark. Following centrifugation at 2500 g for 10 min at 4°C, the supernatant was transferred to a fresh tube. An equal volume of water and chloroform was added to remove Chl, and the absorption of the aqueous phase was recorded at 530 nm. Total anthocyanin content was calculated using a standard curve of absorbance of cyanidin 3‐O‐glucoside chloride (#52976; Sigma‐Aldrich).

GenBank accession numbers

The sequence data that support the findings of this work have been deposited in GenBank with the following accession numbers: RH1 (XP_013442258), RH2 (XP_013442390), MtTT8 (AES60907), MtWD40‐1 (XP_003602392), MtCHS (XP_003601647), MtCHI (XP_003592761), MtF3H (XP_003629323), MtF3′H (XP_013457979), MtF3′5′H (XP_013459330), MtDFR1 (XP_013466134), MtDFR2 (XP_013466133), MtANS (XP_003611189), MtMYB114 (XP_003616344), MtLAP1 (XP_003628527), MtLAP2 (XP_003616362), MtLAP3 (XP_003621572), MtLAP4 (XP_003616359), WP1 (XP_013442371), TrRED_LEAF (AIT76557), TrRED V‐a (AIT76565), TrRED LEAF DIFFUSE‐a (AIT76556), TrRED LEAF DIFFUSE‐b (AIT76560), TrCA1 (AIT76562), TrBX1 (AIT76561), AtMYB113 (NP_176811), AtMYB114 (NP_176812), AtPAP1 (NP_176057), AtPAP2 (NP_176813), AtMYB11 (NP_191820), AtMYB12 (NP_182268), AtMYB111 (NP_199744), AtTT2 (NP_198405), AtGL1 (NP_189430), AtMYB23 (NP_198849), AtWER (NP_196979), PhAN2 (AAF66727), AtMYB4 (NP_195574) and PhMYB27 (AHX24372).

Results

Molecular cloning of RH1 and RH2 for anthocyanin leaf markings in M. truncatula

Plants of M. truncatula ecotype R108 have characteristic distinct leaf markings, which show yellow spots with strong red border on the adaxial surface in the basal part of the leaflets (Fig. 1a,e), which is assumed to be mainly pigmented by anthocyanins (McComb, 1974). In a screen of a collection of M. truncatula R108 Tnt1 insertional mutants (Tadege et al., 2008), we isolated a dominant leaf marking mutant with enlarged and enhanced leaf anthocyanin spots, which was named red heart1 (rh1d) (Fig. 1a). Genetic analysis of progeny from a backcross of rh1d with the wild‐type indicated that rh1d has a semidominant mutation in a single locus, given that the leaf marking phenotype of heterozygous F1 plants (rh1d/+) was intermediate between those of homozygous rh1d and wild‐type, and the F2 segregation of wild‐type : rh1d/+‐type : rh1d‐type fits a ratio of 1 : 2 : 1 (55 : 99 : 48) (Fig. S1). Flanking sequence analysis of the Tnt1 retrotransposon in the rh1d mutant by PCR‐based genotyping revealed that no flanking sequence segregated with the mutant phenotype, implying this mutation may not be caused by the Tnt1 insertion. Using 3777 F2 plant lines from a cross between rh1d and another M. truncatula ecotype A17, we localized the RH1 locus to an c. 590 kb region based on M. truncatula genome version MT4.0 (Fig. S2), but to a 6.9 Mb region using the latest version Mt5.0 of the A17 genome (Pecrix et al., 2018), both near the centromere of chromosome 7 (Figs 1b, S2).

Fig. 1.

Fig. 1

Open in a new tab

Identification of RH1 and RH2 in Medicago truncatula. (a) Phenotype of the rh1d mutant. WT, wild‐type; Ad, adaxial; Ab, abaxial. Inset: the enlarged leaf marking of WT. Bars, 1 cm (plants); 5 mm (leaves). (b) Genetic mapping narrowed the RH1 locus to the pericentromeric region on chromosome 7. (c) Sequencing depth information for the progeny pools on MWMB01000046.1. Gray box indicates the delta depth > 0, representing the 215 kb deletion in rh1d. (d) Molecular cloning of RH1. Above, schematic map showing the deletion region in rh1d; below, expression analysis of genes flanking the deletion region. Data are mean ± SD of three biological replicates; asterisks indicate significant differences from WT (**, P < 0.01, Student’s t‐test). (e) Transcript levels of RH1 and RH2 in the different regions of the WT leaflets determined by qRT‐PCR. Data are mean ± SD (n = 3 technical replicates). Three independent biological replicates showed similar results. (f) Leaf phenotypes of RH1 mutations in rh1d. Ho and He indicate homozygous and heterozygous mutant, respectively. Bars, 5 mm. (g) Above, leaf phenotypes of the rh1 mutants; below, enlargements of leaf marks of lateral leaflets. Bars, 5 mm. (h) Above, leaf phenotypes of rh2 mutants; below, enlargement of leaf marks. Bars, 5 mm. (i) Transcript levels of RH1 in the yellow regions of the WT and the corresponding region of rh2 mutant leaflets determined by qRT‐PCR. Data are mean ± SD of three biological replicates; asterisks indicate significant differences from WT (**, P < 0.01, Student’s t‐test).

To further identify RH1, we performed BSA‐seq using the F2 progeny from an rh1d × wild‐type R108 cross. We identified a significantly linked deletion on M. truncatula R108 scaffold MWMB01000046.1, which is highly syntenic to the mapped region in the A17 genome. Using PCR‐based chromosomal walking, we identified an c. 215 kb deletion containing multiple transposable elements (TEs) in rh1d (Figs 1d, S4; Table S2). Notably, parallel RNA‐seq analysis showed that a putative MYB gene, g47425 (Chr7g0229101 in A17 Mt5.0), neighboring the left border of the deletion region, was significantly upregulated in rh1d (Table S3). Further qRT‐PCR analysis revealed that, besides g47425, five predicted neighboring open reading frames (ORFs)/genes spanning the deletion were universally upregulated in rh1d (Fig. 1d; Table S2), indicating that this large genomic deletion affects expression of adjacent genes. We focused on the markedly upregulated MYB gene g47425, which is predominantly expressed in the leaf markings, particularly in the red border (Figs 1e, S5a).

To examine whether the upregulation of g47425 leads to the enhanced leaf marking phenotype, we interrupted g47425 in rh1d by CRISPR/Cas9 (CR) (Meng et al., 2017), using two 20 bp sequences in exons 3 and 2 of g47425, corresponding to the conserved [R/K]Px[P/A/R]xx[F/Y] motif and the R3 domain, which are important for the function of MYB proteins (Stracke et al., 2001; Dubos et al., 2008), as target sites for Cas9 cleavage (Fig. S6). Both mutations, when homozygous, abolished leaf markings in rh1d, whereas when heterozygous, produced severely diminished leaf anthocyanin markings (Fig. 1f). Further qRT‐PCR analysis showed that the transcript of g47425 was also decreased in the above CRISPR/Cas9 mutants (Fig. S7), suggesting the CRISPR/Cas9‐mediated edition not only leads to mutations in g47425 but also affects its transcription. Together, these results implied the elevated expression of g47425 is responsible for the enhanced leaf marking phenotype in rh1d. To further confirm the role of g47425 in regulating M. truncatula anthocyanin leaf markings, we identified two insertion lines from the Tnt1 insertion population by searching the flanking sequence tag (FST) database (Sun et al., 2019) (Fig. S8a,b). Compared to the wild‐type, homozygous/heterozygous g47425 mutant plants showed abolished/diminished leaf markings (Fig. 1g), the opposite phenotype to rh1d, confirming that g47425 acts as a key regulator in determining leaf marking formation in M. truncatula. Accordingly, we designated g47425 as RH1.

Notably, annotation of the 215 kb deletion region in rh1d identified another predicted MYB gene, g47401 (Chr7g0229131 in A17 Mt5.0), which is also highly expressed in leaf markings and the highest transcript was detected in yellow spots (Figs 1e, S5b). To examine whether deletion of this MYB gene causes the anthocyanin leaf spot phenotype in rh1d, we isolated g47401 loss‐of‐function mutants by screening the Tnt1 mutant population in M. truncatula R108 (Fig. S8c,d). g47401 knockout mutants showed no other obvious phenotypes, but an enhanced leaf anthocyanin spot, partially mimicking rh1d (Fig. 1h). Interestingly, loss of g47401 led to an increased RH1 expression (Fig. 1i), implying that the deletion of g47401 contributes to increased RH1 expression, and therefore to the RH1‐mediated rh1d phenotype. We concluded that g47401 regulates leaf markings negatively and accordingly named it as RH2.

RH1 encodes a subgroup 6 R2R3 MYB TF and functions as an anthocyanin biosynthesis activator

Phylogenetic analysis and sequence alignment showed that RH1 is related to the anthocyanin‐activating subgroup 6 R2R3 MYBs (Fig. 2a), which contain a typical bHLH‐interacting motif, an ANDV motif, and a [R/K]Px[P/A/R]xx[F/Y] motif downstream of the conserved R2 and R3 MYB DNA‐binding domains (Fig. S9) (Stracke et al., 2001; Dubos et al., 2008). A transactivation activity assay in yeast and Arabidopsis protoplasts confirmed that RH1 has strong transactivation activity (Fig. 2b,c). Individual and combined domain deletion analysis in the yeast assay indicated that the C‐terminal domain (CTD; amino acids 113–241) is essential for the transactivation activity of RH1 (Fig. 2b).

Fig. 2.

Fig. 2

Open in a new tab

RH1 acts as a transcriptional activator, and RH2 negatively regulates Medicago truncatula leaf marks indirectly. (a) Phylogenetic analysis of the RH1, RH2 (highlight in red diamonds) and other MYB proteins. Tree is midpoint‐rooted. Subgroup (S) names are indicated on the right, and the subgroups are shaded. Species: At, Arabidopsis thaliana; Mt, Medicago truncatula; Tr, Trifolium repens; Ph, Petunia hybrida. (b) Transactivation analysis of RH1 and RH2 using a yeast assay. Schematic representations of the RH1 and RH2 proteins. Numbers represent the amino acid number of RH1 or RH2. BD, GAL4 DNA‐binding domain; CTD, C‐terminal domain; ‐W, SD/‐Trp; ‐W‐H‐Ade, SD/‐Trp/‐His/‐Ade. Plate auxotroph and α‐galactosidase (α‐gal) assay showing transcriptional activation of each protein. Data are mean ± SD of three independent experiments. (c) Transactivation analysis of RH1, RH2 and RH21‐112 + RH1113‐241 in Arabidopsis protoplasts. The LUC/REN luminescence was normalized to the value of the control of GAL4 DNA‐binding domain (BD). Data are mean ± SD of three independent experiments; different letters denote means that are significantly different as determined by one‐way ANOVA and a post‐hoc Tukey test (P = 0.05). (d) Transcript levels of anthocyanin biosynthetic genes in leaves of wild‐type (WT), rh1d and rh2‐1. Data are mean ± SD of three biological replicates; asterisks indicate significant differences from WT (*, P < 0.05; **, P < 0.01, Student’s t‐test). (e) DNA‐binding assay corresponding to conserved MYB binding sites. GST was used as a negative control. Data are mean ± SD of three independent experiments; asterisks indicate significant differences from GST control (**, P < 0.01, Student’s t test). (f) Transcriptional activation by RH1 and RH2 on the promoters of the anthocyanin biosynthetic genes in Arabidopsis protoplasts. The LUC/REN luminescence was normalized to the value of the GFP control. Data are mean ± SD of three independent experiments; asterisks indicate significant differences from GFP control (**, P < 0.01, Student’s t‐test).

To determine the mechanism by which RH1 regulates leaf anthocyanin markings in M. truncatula, we analyzed the transcript levels of anthocyanin pathway genes (Fig. S10) in rh1d leaves. qRT‐PCR analyses showed that six of the eight anthocyanin biosynthetic genes, including early biosynthetic genes such as MtCHS, MtF3H and MtF3′H, and late biosynthetic genes such as MtDFR1, MtDFR2 and MtANS, were significantly upregulated in leaves of rh1d compared to wild‐type plants (Fig. 2d). Sequence analysis using the plantcare program (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/) predicted that multiple typical MYB‐binding sites (MYB‐core and/or AC‐rich elements) were present in the promoter regions of the six upregulated anthocyanin biosynthetic genes (Fig. S11). To confirm the prediction that RH1 serves as an anthocyanin biosynthesis activator, we first used an in vitro DNA‐binding assay and found that RH1 binds to the conserved MYB‐core and AC‐rich elements (Fig. 2e). To further confirm these binding sites, we performed the EMSA using purified recombinant proteins of the His‐TF‐RH1. The results showed that RH1 could bind to the biotin‐labeled probes containing MYB‐core or AC‐rich elements in the promoters of upregulated anthocyanin biosynthetic genes in rh1d, such as MtCHS and MtANS (Fig. S12). We further employed transient luciferase assays in Arabidopsis protoplasts and demonstrated that RH1 activates the expression of MtCHS and MtANS promoters (Fig. 2f). Together, these data indicate that RH1 functions as an anthocyanin biosynthesis activator and that RH1 activation of anthocyanin biosynthetic genes may be mediated by binding to their specific promoter regions.

RH2 is paralogous to RH1 but lost the transactivation activity

Sequence alignment showed that RH2 also contains R2 and R3 domains that are nearly identical to those in RH1. Phylogenetic analysis revealed that RH1 and RH2 group together, indicating that RH2 is paralogous to RH1, although with a short and divergent CTD (Figs 2a, S9). However, transactivation activity assay in both yeast and Arabidopsis protoplasts revealed that RH2 has neither transactivation nor obvious repressive activity (Fig. 2b,c), indicating a functional divergence between RH2 and RH1. Given that RH2 and RH1 have nearly identical R2 and R3 domains, we speculated that the loss of transactivation in RH2 is associated with its divergent C terminus. To test this hypothesis, we performed domain swapping by replacing the CTD of RH2 with that of RH1 and examined the transactivation activity of this chimeric protein. The results showed that the replacement of CTD restored the transactivation activity of RH2 (Fig. 2b,c), indicating that the reason RH2 has no transactivation/repressive activity is due to its inactive CTD, suggesting that the CTD divergence leads to a subfunctionalization between RH1 and RH2.

To determine how RH2 regulates leaf markings in M. truncatula, we analyzed the transcript levels of anthocyanin pathway genes in leaves of rh2. qRT‐PCR analyses showed that anthocyanin biosynthetic genes, MtCHS, MtF3H, MtF3′H, MtDFR1, MtDFR2 and MtANS, which are upregulated in rh1d, are also moderately increased in rh2 (Fig. 2d). However, a transient expression assay using the luciferase system in Arabidopsis leaf protoplasts showed that RH2 fails to activate the expression of anthocyanin biosynthetic genes such as MtCHS and MtANS, despite the fact that they bind to the MYB‐core and AC‐rich elements (Fig. 2e,f), indicating RH2 may not directly regulate anthocyanin biosynthesis, although RH2 negatively affects M. truncatula anthocyanin leaf markings.

RH1 and RH2 physically interact with MtTT8 and MtWD40‐1 to form the MBW complex

Previous studies have shown that the R2R3 MYB regulates transcription by interacting with bHLH and WDR proteins to form a conserved MBW activator complex (Baudry et al., 2004; Koes et al., 2005; Ramsay & Glover, 2005; Xu et al., 2015). Given that both RH1 and RH2 contain the highly conserved bHLH‐interacting motif, we investigated whether they function within the MBW complex as well. A Y2H assay showed that, despite RH1 showing self‐activation activities, both RH1 and RH2 interact with the bHLH TF MtTT8 and the WDR protein MtWD40‐1 (Fig. 3a,b), which modulate anthocyanin biosynthesis in M. truncatula (Pang et al., 2009; Li et al., 2016). These interactions among RH1, RH2, MtTT8 and MtWD40‐1 were verified via a BiFC assay in N. benthamiana leaves using split yellow fluorescent protein (YFP). Strong yellow fluorescence was clearly observed between RH1 or RH2 fused to the N‐terminal half of YFP (nYFP) and MtTT8 or MtWD40‐1 fused to the C‐terminal half of YFP (cYFP), whereas no YFP fluorescent signals were detected in the negative control (combination of RH1/RH2–nYFP and cYFP alone; Fig. 3c). These interactions were further confirmed in vivo by Co‐IP assays. We transiently co‐expressed 35S:RH1‐Flag with 35S:MtTT8‐Myc or 35S:MtWD40‐1‐Myc, or 35S:RH2‐Flag with 35S:MtTT8‐Myc or 35S:MtWD40‐1‐Myc in N. benthamiana leaves. Total proteins were isolated and incubated with anti‐C‐Flag agarose beads to immunoprecipitate RH1‐Flag or RH2‐Flag. Anti‐Flag and anti‐Myc antibodies were then used to detect immunoprecipitated proteins with corresponding tags. MtTT8 and MtWD40‐1 were detected in the immunoprecipitated RH1 or RH2 complex but not in the negative control without RH1‐Flag or RH2‐Flag input (Fig. 3d,e), indicating that both RH1 and RH2 physically associate with MtTT8 and MtWD40‐1 in planta.

Fig. 3.

Fig. 3

Open in a new tab

Medicago truncatula RH1 and RH2 interact with MtTT8 and MtWD40‐1 to form an MBW complex. (a, b) Interactions of RH1 (a) or RH2 (b) with MtTT8 and MtWD40‐1 in a Y2H assay. Plate auxotroph and α‐galactosidase (α‐gal) assay showing the interaction of each protein. Data are mean ± SD of three independent experiments. AD, activation domain; BD, GAL4 DNA‐binding domain; DDO, double dropout; QDO, quadruple dropout. (c) Interaction between RH1 or RH2 with MtTT8 and MtWD40‐1 in Nicotiana benthamiana leaf epidermal cells using a split YFP BiFC assay. Bars, 20 μm. (d, e) Interaction of RH1 (d) or RH2 (e) with MtTT8 and MtWD40‐1 in N. benthamiana leaf epidermal cells detected using a Co‐IP assay. Immunoblots of total protein extract (Input) and IP product were performed using anti‐Flag (α‐Flag) and anti‐Myc (α‐Myc), respectively.

To clarify the connection of MtTT8 and MtWD40‐1 to RH1 and RH2 function in regulating anthocyanin leaf markings, we performed a complementary genetic analysis by generating double mutants of rh1d, rh1 and rh2 with mttt8 or mtwd40‐1, respectively. In contrast to the wild‐type, both mttt8 and mtwd40‐1 plants showed less anthocyanin pigmentation at whole leaflets and petioles, but mttt8 leaves have pale green spots while mtwd40‐1 leaves have severely weakened leaf markings (faint spots) (Fig. 4a). Phenotypic analysis revealed that leaves of the rh1d mttt8 double mutant exhibit similar patterns to rh1d, but with pale green spots instead of red ones (Fig. 4b). Leaves of the rh1d mtwd40‐1 double mutant were similar to those of rh1d mttt8 but with very faint markings, indicating that both MtTT8 and MtWD40‐1 are required for RH1‐mediated control of leaf anthocyanin markings (Fig. 4b). By contrast, the rh1 mttt8 and rh1 mtwd40‐1 double mutants showed complete abolishment of leaf markings, similar to the RH1 knockout mutant, suggesting that RH1 fulfills the major function within the activation complex in determining leaf markings (Fig. 4c). Similarly, genetic analysis showed that the rh2 mttt8 and rh2 mtwd40‐1 double mutants showed similar patterns to mttt8 and mtwd40‐1, respectively (Fig. 4d), implying that MtTT8 and MtWD40‐1 are also required for RH2‐mediated control of anthocyanin leaf markings. Nevertheless, both MtTT8 and MtWD40‐1 are universally expressed in the three different leaf regions (Fig. S13), suggesting their function in the regulation of leaf anthocyanin marking is probably brought about by the specific expression of RH1 and RH2.

Fig. 4.

Fig. 4

Open in a new tab

Genetic analysis of RH1, RH2, MtTT8 and MtWD40‐1 in regulating leaf marking in Medicago truncatula. Leaf phenotypes of wild‐type (WT), mttt8, mtwd40‐1 (a), rh1d, rh1d mttt8, rh1d mtwd40‐1 (b), rh1‐1, rh1‐1 mttt8, rh1‐1 mtwd40‐1 (c), and rh2‐1, rh2‐1 mttt8 and rh2‐1 mtwd40‐1 (d). Insets are enlargements of leaf marks of lateral leaflets. Bars, 5mm.

RH2 negatively regulates anthocyanin leaf markings by multidimensionally antagonizing RH1

The observations that RH2 recognizes the same cis elements within promoters of anthocyanin biosynthetic genes as RH1, and involves in the MBW complex, but shows no transcriptional repressive activity, implied that RH2 might negatively regulate leaf markings either by competing for RH1’s function or its physical placement in the MBW complex. To test this hypothesis, we performed three complementary experiments. First, we tested the MtANS promoter activity in Arabidopsis leaf protoplasts cotransfected with RH1 and/or RH2 along with combinations of MBW components. RH1 alone activated the MtANS promoter, and co‐expression of RH1 with MtTT8 and MtWD40‐1 significantly increased the MtANS promoter activity (Fig. 5a). By contrast, RH2 did not activate the MtANS promoter, even in the presence of MtTT8 and MtWD40‐1 (Fig. 5a). However, co‐expression of RH1 and RH2, compared to RH1 alone, resulted in significantly lower MtANS promoter activity in the presence of MtTT8 and MtWD40‐1 (Fig. 5a), suggesting that RH2 competes with RH1 to suppress anthocyanin biosynthesis. Second, to further verify this observation in vivo, we crossed rh1d to wild‐type R108 and to the rh2 mutant. Phenotypic analysis showed that, in contrast to the diminished anthocyanin leaf spots in rh1d heterozygous plants (backcrossed with R108), rh1d × rh2 F1 progeny are indistinguishable from rh1d (Fig. 5b), confirming that RH2 represses RH1 function in vivo. Furthermore, we overexpressed RH2 under control of the cauliflower mosaic virus 35S promoter in rh1d and found that ectopic expression of RH2 in the rh1d background repressed the rh1d leaf marking phenotype, producing diminished anthocyanin leaf markings (Fig. 5c,d). These data collectively indicate that RH2 negatively regulates leaf anthocyanin markings by competing with RH1 in the MBW regulatory complex.

Fig. 5.

Fig. 5

Open in a new tab

RH2 competes with RH1 to negatively regulate Medicago truncatula anthocyanin leaf markings. (a) Transcriptional activation of the MtANS promoter in a transient luciferase assay using Arabidopsis protoplasts. Protoplasts were transfected with various combinations of M. truncatula effectors (RH1, RH2, MtTT8, MtWD40‐1 and GFP control) along with the MtANS promoter fused with luciferase reporter. LUC/REN luminescence was normalized to the value of the GFP control. Data are mean ± SD of three independent experiments; different letters denote means that are significantly different as determined by one‐way ANOVA and a post‐hoc Tukey test (P = 0.05). (b) Leaf phenotypes of wild‐type (WT), rh1d, rh2‐1 (above), and their F1 progeny (below). Bars, 5 mm. (c) Leaf phenotypes of rh1d and different 35S:RH2/rh1d transgenic lines. Bars, 5 mm. (d, e) Transcript levels of RH2 (d) and RH1 (e) in rh1d and different 35S:RH2/rh1d transgenic lines. Data are mean ± SD of three biological replicates; asterisks indicate significant differences from rh1d (**, P < 0.01, Student’s t‐test).

Notably, compared to rh1d, the transcript abundance of RH1 was severely decreased in RH2 overexpressors (Fig. 5e), suggesting that RH2 may also antagonize RH1 function by repressing its expression. However, given that RH2 has no obvious transcriptional repressive activity, one possibility is that RH1 can activate its own expression and RH2 may repress RH1 expression by competing RH1 function, in a similar manner as negatively regulating anthocyanin biosynthetic genes. Consistent with this scenario, sequence analysis showed that conserved MYB‐core and AC‐rich elements are present in the promoter regions of RH1 (Fig. S11), and transient expression assay in Arabidopsis leaf protoplasts showed that the RH1 effector significantly activated RH1 promoter activity compared to the green fluorescent protein (GFP) control, while no obvious activation was detected when the RH2 effector was co‐expressed with the luciferase reporter driven by the promoter of RH1 (Fig. 6a), indicating RH1 can activate its own expression. It is worth noting that although the MYB‐binding elements are also present in the promoter regions of RH2 (Fig. S11), co‐expression of the RH1 or RH2 effector protein and the luciferase reporter construct driven by the RH2 promoter resulted in little increased luminescence intensity compared to the GFP control, indicating that RH2 promoter cannot be activated by RH1 and itself (Fig. S14a). Moreover, transgenic M. truncatula plants overexpressing the RH1 CDS leads to ectopic anthocyanin accumulation and showed upregulation of native RH1 but not RH2 (Figs 6b,c, S14b), confirming RH1 functions as an anthocyanin biosynthesis activator and has a transcriptional self‐activating activity in vivo. Further transient expression assays in Arabidopsis protoplasts showed that co‐expression of RH1 and RH2, compared to RH1 alone, resulted in significantly lower RH1 promoter activity in the presence of MtTT8 and MtWD40‐1 (Fig. 6a), mimicking the downregulation of RH1 in 35S:RH2/rh1d plants. These data together demonstrated that RH2 negatively regulates M. truncatula leaf anthocyanin markings by multidimensionally antagonizing RH1 (Fig. 7a), uncovering a paralogous–antagonistic regulatory paradigm for fine‐tuning patterned pigmentation.

Fig. 6.

Fig. 6

Open in a new tab

Medicago truncatula RH1 has a transcriptional self‐activating activity. (a) Transcriptional activation of the RH1 promoter in a transient luciferase assay using Arabidopsis protoplasts. Protoplasts were transfected with various combinations of M. truncatula effectors (RH1, RH2, MtTT8, MtWD40‐1 and GFP control) along with the RH1 promoter fused to the luciferase reporter. LUC/REN luminescence was normalized to the value of the GFP control. Data are mean ± SD of three independent experiments; different letters denote means that are significantly different as determined by one‐way ANOVA and a post‐hoc Tukey test (P = 0.05). (b) Leaf phenotypes of wild‐type (WT) and RH1‐overexpressing plants. Scale bars = 5 mm. (c) Transcript abundances of native RH1 in transgenic M. truncatula expressing 35S:RH1 revealed by qRT‐PCR. Data are mean ± SD (n = 3 technical replicates). Three independent biological replicates showed similar results. Asterisks indicate significant differences from mock (**, P < 0.01, Student’s t‐test).

Fig. 7.

Fig. 7

Open in a new tab

Illustrations of leaf marking phenotypes and the RH1–RH2 regulatory paradigm in Medicago truncatula wild‐type and diverse mutants. (a) A proposed working model for leaf marking control in M. truncatula. The R2R3 MYB activator RH1 associates with MtTT8 and MtWD40‐1 and regulates leaf marking formation by directly activating the expression of anthocyanin biosynthetic genes (ABG) and itself, whereas its inactive paralog RH2 acts as a competitor in the regulatory complex to modulate RH1‐mediated anthocyanin pigmentation, governing the distinct anthocyanin leaf marking (yellow spot with strong red border) in wild‐type (WT) M. truncatula R108. (b) Semidominant rh1d mutant with enlarged and enhanced leaf spots, due to disruption of RH2 and other genes/elements within the 215 kb deletion leading to the significant upregulation of RH1. (c) Loss‐of‐function of the central regulator RH1 abolishes M. truncatula leaf anthocyanin markings. (d) Disruption of RH2 releases the antagonism to RH1, leading to enhanced leaf anthocyanin markings, partially mimicking rh1d.

Discussion

The genus Medicago contains diverse species that commonly display a set of anthocyanin leaf markings (McComb, 1974). However, neither the molecular basis that imparts diverse Medicago leaf markings nor their regulatory mechanisms have been clearly understood. Here, starting from the genetic analysis of a dominant leaf marking mutant rh1d, we identified two paralogous MYB regulators, RH1 and RH2, in controlling the distinct anthocyanin leaf markings of M. truncatula and elucidated their mechanism of action.

Several lines of evidence suggest that RH1, an R2R3 MYB TF, is the central regulator of anthocyanin leaf markings in M. truncatula. First, RH1 gene expression is predominantly confined to leaf markings, particularly in the red borders (Figs 1e, S5a). Second, RH1 acts as a transcription activator and positively regulates early and late anthocyanin biosynthetic genes (Figs 2, 6b). Third, loss‐of‐function rh1 mutants showed no aberrant phenotype apart from abolished leaf anthocyanin markings (Fig. 1g). To date, several R2R3 MYB TFs have been reported to be associated with anthocyanin leaf markings in M. truncatula and T. repens (Carletti et al., 2013; Albert et al., 2015), whereas, to our knowledge, RH1 is the conclusive functionally defined MYB regulator specifically controlling M. truncatula anthocyanin leaf markings. Given the sequence similarity, RH1’s closest known homolog is the CA1 protein of T. repens that remains uncharacterized (Albert et al., 2015). Future investigation may reveal the function of RH1 homologs in diverse M. truncatula varieties and show whether CA1 fulfills a role similar to RH1 in determining T. repens leaf marking traits.

It has been reported that the yellow color within the M. truncatula leaf markings is probably due to the reduction of Chls and carotenoids (McComb, 1974). Overexpression of RH1 leads to enlarged and enhanced leaf anthocyanin spots (Fig. 1a), suggesting that the anthocyanin marking is dominant to yellow coloration, while disruption of RH1 leads to the abolishment of the entire leaf markings including both red and yellow coloration (Fig. 1f), suggesting that RH1 may also play a role in the regulation of Chls and carotenoid production. Likewise, recent studies reported that an anthocyanin MYB‐like protein, WHITE PETAL1 (WP1), which is clustered with RH1 (Fig. 2a), could activate the carotenoid pathway in M. truncatula (Meng et al., 2019). Further investigation of the role of RH1 in the regulation of Chl and carotenoid biosynthesis will confirm this possibility.

In this study, RH2 was identified as an additional regulator with opposite effects on the regulation of leaf anthocyanin markings in M. truncatula. So far, several types of anthocyanin repressors have been characterized, including: R3 MYBs with or without repression motifs (e.g. CPC, MYBx and SIMYBATV), CgRuby2short with partial R2 and R3 domains without a repression motif, MYBL2 containing a partial R2 domain and complete R3 domain with a repression motif, and R2R3 MYBs with repression motifs (e.g. FaMYB1, PhMYB27 and MtMYB2) (Aharoni et al., 2001; Dubos et al., 2008; Zhu et al., 2009; Albert et al., 2014; Jun et al., 2015; D. Huang et al., 2018; Sun et al., 2020). In contrast to these examples, RH2 encodes a nonfunctional version of the anthocyanin repressor with complete R2 and R3 domains but exhibiting no transactivation/repressive activity. Intriguingly, RH2 can gain the transactivation activity when its short CTD is replaced with that of RH1, which contains multiple acidic amino acid residues (e.g. Asp/D and Glu/E) (Fig. S9), suggesting its short divergent CTD renders RH2 inactive. This idea received support from previous studies in maize C1 alleles (Paz‐Ares et al., 1987, 1990; Martin and Paz‐Ares, 1997). Our data suggest that RH2 is not a simple repressor but rather an antagonistic competitor of RH1 via binding to the same MYB elements of downstream promoters with RH1 and binding to the same bHLH and WD40 partners and thereby exerts a negative effect on anthocyanin biosynthesis within yellow spots, thus determining the distinct anthocyanin leaf marking in M. truncatula (Fig. 7a). In agreement with this, we observed that the transcripts of anthocyanin biosynthetic genes and anthocyanin production are severely decreased in the RH2‐expressing yellow spot, in contrast to the red border, where RH1 is predominantly expressed (Fig. S15). Furthermore, disruption of RH2 leads to no other phenotypes, except an enhanced leaf anthocyanin spot, partially mimicking rh1d (Fig. 1h), while overexpression of RH2 represses the enhanced leaf anthocyanin marking in rh1d (Fig. 5c), suggesting that RH2 may mainly serve as a molecular rheostat to modulate RH1‐mediated restricted anthocyanin pigmentation, thus providing new insight into understanding the subfunctionalization of the MYB TFs. Based on the above results, we propose a simple model depicting the RH1–RH2‐mediated regulatory paradigm for leaf anthocyanin accumulation in M. truncatula, and further schematic diagrams have been drawn to illustrate the diverse leaf anthocyanin marking phenotypes of individual mutants (Fig. 7).

Nevertheless, it is worth noting that the precise location of RH1 and RH2, and the cause of the restricted position of leaf spot marking remain unclear. One possible explanation is that the distinct anthocyanin biosynthesis/deposition in M. truncatula leaves might be suppressed in areas outside of the markings by one or more unidentified repressors epistatic to the RH1 and RH2 regulators. Further identification and characterization of these putative suppressors of RH1 and RH2, and demonstration of their precise expression locations and interactions with the RH1–RH2 regulatory module are needed to test these possibilities. Moreover, several MYB TFs such as LAP1, MtMYBA and AN2 have been reported to affect anthocyanin accumulation in M. truncatula leaves (Peel et al., 2009; Carletti et al., 2013). The exact relationship between the RH1–RH2 regulatory module and these MtMYB genes involved in the regulation of leaf anthocyanin marking in M. truncatula awaits elucidation in future studies.

that RH2‐mediated competition can repress RH1 expression (Fig. 6a), while disruption of RH2 leads to only a two‐fold increase of RH1 expression (Fig. 1i), suggesting that the significant upregulation of RH1 may involve other genes within the 215 kb deletion in rh1d or possibly caused by chromatin modifications due to this large genomic aberration. Given that TEs play key roles in chromosome architecture and gene regulation (Slotkin & Martienssen, 2007; Sigman & Slotkin, 2016; C. Huang et al., 2018), further investigation of the TEs and/or other genes within the 215 kb deletion and the chromatin status of RH1 should reveal the specific causative variants that are responsible for the increased RH1 expression in the context of rh1d.

Gene duplication is a major evolutionary force driving adaptation and speciation (Innan & Kondrashov, 2010; Lan & Pritchard, 2016). However, most young duplicates are degraded by loss‐of‐function mutations, and the factors that allow some duplicate pairs to survive in the long term remain controversial. Due to an ancient genome duplication event, the presence of duplicated gene pairs has been commonly found in most legume plants, including M. truncatula. The sequence similarity, phylogenetic relationship as well as their close physical position (RH2 located c. 180 kb downstream of RH1, in the centromere region of chromosome 7) suggest that RH1 and RH2 may derive from a gene duplication event, while the opposite expression patterns and biological functions suggest that RH1 and RH2 show a subfunctionalization. Our results suggest that RH2 may have lost transcriptional activity through simple impairment of a CTD, a rapid mechanism that may have been widely used in evolution. However, RH2 has acquired a new function as a passive repressor by antagonizing its active paralog RH1, thus providing an evolutionary outcome of gene duplication – functional antagonism – that can rapidly integrate into existing regulatory circuitry to provide biological benefits. Likewise, recent studies in mice reported that the antagonistic gene paralogs Upf3a and Upf3b are critical for the nonsense‐mediated RNA decay (Shum et al., 2016), suggesting this kind of paralogous antagonism mechanism is probably widespread in the evolution of plants and animals that allow the persistence of young duplications.

Together, our study demonstrates the molecular mechanism of the antagonistic gene paralogs RH1 and RH2 in determining leaf anthocyanin markings in M. truncatula. Our findings provide a paralogous–antagonism regulatory paradigm for fine‐tuning patterned pigmentation and offer mechanistic insights into the evolution of gene subfunctionalization through loss‐of‐function mutations.

Author contributions

CW, LN and HL designed the research. CW, WJ, YL, PZ, YM and PZ performed the experiments. CW, JZ, ZT, LN and HL analysed the data. JW, KSM and NDY contributed analytical tools. CW, LN and HL wrote the manuscript with contributions by all authors. CW and WJ contributed equally to this work.

Supporting information

Fig. S1 rh1d is a semidominant mutant.

Fig. S2 Map‐based cloning of the RH1 gene based on M. truncatula genome version Mt4.0.

Fig. S3 Synteny analysis between M. truncatula R108_v1.0 and the mapped region on A17 chromosome 7.

Fig. S4 Chromsome walking verification of the 215 kb deletion in rh1d.

Fig. S5 Expression analysis of RH1 and RH2.

Fig. S6 Targeted mutagenesis in RH1 using the CRISPR/Cas9 system.

Fig. S7 Transcript levels of RH1 in the CRISPR/Cas9 mutant leaflets determined by qRT‐PCR.

Fig. S8 Identification and characterization of rh1 and rh2 mutants.

Fig. S9 Sequence alignment of RH1, RH2 and their close orthologs in Trifolium and Arabidopsis.

Fig. S10 Simplified scheme of the enzyme genes involved in the anthocyanin biosynthetic pathway.

Fig. S11 The predicted MYB‐binding sites in the promoters of upregulated anthocyanin biosynthetic genes in rh1d, as well as RH1 and RH2.

Fig. S12 EMSA showed that RH1 and RH2 can bind to the promoters of MtCHS and MtANS.

Fig. S13 Transcript levels of MtTT8 and MtWD40‐1 in the different regions of the wild‐type leaflets determined by qRT‐PCR.

Fig. S14 RH2 cannot be activated by RH1 and itself.

Fig. S15 The transcripts of anthocyanin biosynthetic genes were decreased in the wild‐type yellow mark with reduced anthocyanin production.

Table S1 Primers used in this study.

Table S2 Annotation of the 215 kb deletion and flanking regions in rh1d.

Table S3 The 448 differentially expressed genes in rh1 d leaves compared with wild‐type leaves.

Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

Click here for additional data file. (994.2KB, pdf)

Acknowledgements

We thank Prof. Richard Dixon for providing p2GW7 and pRLC plasmids, Dr Yuhong Tang and Dr Guifen Li for assistance with histological analysis. This work was supported by grants from the National Natural Science Foundation of China (32070554 and 32071864), Agricultural Science and Technology Innovation Program of CAAS (CAAS‐ZDRW202009 and CAAS‐ZDXT2019004), and Fundamental Research Funds for Central Non‐profit Scientific Institution (Y2020YJ12 and No. 1610392020005). The development of M. truncatula Tnt1 insertion lines was supported by the US National Science Foundation (DBI 0703285 and IOS‐1127155) and Noble Research Institute LLC. The authors declare no competing financial interests.

References

  1. Aharoni A, De Vos CH, Wein M, Sun Z, Greco R, Kroon A, Mol JN, O’Connell AP. 2001. The strawberry FaMYB1 transcription factor suppresses anthocyanin and flavonol accumulation in transgenic tobacco. The Plant Journal 28: 319–332. [DOI] [PubMed] [Google Scholar]
  2. Albert NW, Davies KM, Lewis DH, Zhang H, Montefiori M, Brendolise C, Boase MR, Ngo H, Jameson PE, Schwinn KE. 2014. A conserved network of transcriptional activators and repressors regulates anthocyanin pigmentation in eudicots. Plant Cell 26: 962–980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Albert NW, Griffiths AG, Cousins GR, Verry IM, Williams WM. 2015. Anthocyanin leaf markings are regulated by a family of R2R3‐MYB genes in the genus Trifolium . New Phytologist 205: 882–893. [DOI] [PubMed] [Google Scholar]
  4. Albert NW, Lewis DH, Zhang H, Schwinn KE, Jameson PE, Davies KM. 2011. Members of an R2R3‐MYB transcription factor family in Petunia are developmentally and environmentally regulated to control complex floral and vegetative pigmentation patterning. The Plant Journal 65: 771–784. [DOI] [PubMed] [Google Scholar]
  5. Anders S, Pyl PT, Huber W. 2015. HTSeq–a Python framework to work with high‐throughput sequencing data. Bioinformatics 31: 166–169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Baudry A, Heim MA, Dubreucq B, Caboche M, Weisshaar B, Lepiniec L. 2004. TT2, TT8, and TTG1 synergistically specify the expression of BANYULS and proanthocyanidin biosynthesis in Arabidopsis thaliana . The Plant Journal 39: 366–380. [DOI] [PubMed] [Google Scholar]
  7. Carletti G, Lucini L, Busconi M, Marocco A, Bernardi J. 2013. Insight into the role of anthocyanin biosynthesis‐related genes in Medicago truncatula mutants impaired in pigmentation in leaves. Plant Physiology and Biochemistry 70: 123–132. [DOI] [PubMed] [Google Scholar]
  8. Carnahan HL, Hill HD, Hanson AA, Brown KG. 1955. Inheritance and frequencies of leaf markings in white clover. Journal of Heredity 46: 109–114. [Google Scholar]
  9. Choi HK, Kim D, Uhm T, Limpens E, Lim H, Mun JH, Kalo P, Penmetsa RV, Seres A, Kulikova O et al. 2004. A sequence‐based genetic map of Medicago truncatula and comparison of marker colinearity with M. sativa . Genetics 166: 1463–1502. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Corkill L. 1971. Leaf markings in white clover. Journal of Heredity 62: 307–310. [Google Scholar]
  11. Davies KM, Albert NW, Schwinn KE. 2012. From landing lights to mimicry: the molecular regulation of flower colouration and mechanisms for pigmentation patterning. Functional Plant Biology 39: 619–638. [DOI] [PubMed] [Google Scholar]
  12. Ding B, Patterson EL, Holalu SV, Li J, Johnson GA, Stanley LE, Greenlee AB, Peng F, Bradshaw HD Jr, Blinov ML et al. 2020. Two MYB proteins in a self‐organizing activator‐inhibitor system produce spotted pigmentation patterns. Current Biology 30: 802–814. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Dubos C, Le Gourrierec J, Baudry A, Huep G, Lanet E, Debeaujon I, Routaboul JM, Alboresi A, Weisshaar B, Lepiniec L. 2008. MYBL2 is a new regulator of flavonoid biosynthesis in Arabidopsis thaliana . The Plant Journal 55: 940–953. [DOI] [PubMed] [Google Scholar]
  14. Grotewold E. 2006. The genetics and biochemistry of floral pigments. Annual Review of Plant Biology 57: 761–780. [DOI] [PubMed] [Google Scholar]
  15. Huang C, Sun H, Xu D, Chen Q, Liang Y, Wang X, Xu G, Tian J, Wang C, Li D et al. 2018. ZmCCT9 enhances maize adaptation to higher latitudes. Proceedings of the National Academy of Sciences, USA 115: E334–E341. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Huang D, Wang X, Tang Z, Yuan Y, Xu Y, He J, Jiang X, Peng SA, Li L, Butelli E et al. 2018. Subfunctionalization of the Ruby2–Ruby1 gene cluster during the domestication of citrus. Nature Plants 4: 930–941. [DOI] [PubMed] [Google Scholar]
  17. Innan H, Kondrashov F. 2010. The evolution of gene duplications: classifying and distinguishing between models. Nature Reviews Genetics 11: 97–108. [DOI] [PubMed] [Google Scholar]
  18. Jun JH, Liu C, Xiao X, Dixon RA. 2015. The transcriptional repressor MYB2 regulates both spatial and temporal patterns of proanthocyandin and anthocyanin pigmentation in Medicago truncatula . Plant Cell 27: 2860–2879. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kim D, Langmead B, Salzberg SL. 2015. HISAT: a fast spliced aligner with low memory requirements. Nature Methods 12: 357–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Koes R, Verweij W, Quattrocchio F. 2005. Flavonoids: a colorful model for the regulation and evolution of biochemical pathways. Trends in Plant Science 10: 236–242. [DOI] [PubMed] [Google Scholar]
  21. Lan X, Pritchard JK. 2016. Coregulation of tandem duplicate genes slows evolution of subfunctionalization in mammals. Science 352: 1009–1013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lev‐Yadun S. 2001. Aposematic (warning) coloration associated with thorns in higher plants. Journal of Theoretical Biology 210: 385–388. [DOI] [PubMed] [Google Scholar]
  23. Li H, Durbin R. 2009. Fast and accurate short read alignment with Burrows–Wheeler transform. Bioinformatics 25: 1754–1760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Li P, Chen B, Zhang G, Chen L, Dong Q, Wen J, Mysore KS, Zhao J. 2016. Regulation of anthocyanin and proanthocyanidin biosynthesis by Medicago truncatula bHLH transcription factor MtTT8. New Phytologist 210: 905–921. [DOI] [PubMed] [Google Scholar]
  25. Lin H, Niu L, McHale NA, Ohme‐Takagi M, Mysore KS, Tadege M. 2013. Evolutionarily conserved repressive activity of WOX proteins mediates leaf blade outgrowth and floral organ development in plants. Proceedings of the National Academy of Sciences, USA 110: 366–371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lu Q, Tang X, Tian G, Wang F, Liu K, Nguyen V, Kohalmi SE, Keller WA, Tsang EW, Harada JJ et al. 2010. Arabidopsis homolog of the yeast TREX‐2 mRNA export complex: components and anchoring nucleoporin. The Plant Journal 61: 259–270. [DOI] [PubMed] [Google Scholar]
  27. Marçais G, Delcher AL, Phillippy AM, Coston R, Salzberg SL, Zimin A. 2018. MUMmer4: A fast and versatile genome alignment system. PLoS Computational Biology 14: e1005944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Martin C, Paz‐Ares J. 1997. MYB transcription factors in plants. Trends in Genetics 13: 67–73. [DOI] [PubMed] [Google Scholar]
  29. McComb JA. 1974. Leaf marks in Medicago, with special reference to their inheritance in Medicago truncatula . Australian Journal of Botany 22: 67–80. [Google Scholar]
  30. Meng Y, Hou Y, Wang H, Ji R, Liu B, Wen J, Niu L, Lin H. 2017. Targeted mutagenesis by CRISPR/Cas9 system in the model legume Medicago truncatula . Plant Cell Reports 36: 371–374. [DOI] [PubMed] [Google Scholar]
  31. Meng Y, Wang Z, Wang Y, Wang C, Zhu B, Liu H, Ji W, Wen J, Chu C, Tadege M et al. 2019. The MYB activator WHITE PETAL1 associates with MtTT8 and MtWD40‐1 to regulate carotenoid‐derived flower pigmentation in Medicago truncatula . Plant Cell 31: 2751–2767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Pang Y, Wenger JP, Saathoff K, Peel GJ, Wen J, Huhman D, Allen SN, Tang Y, Cheng X, Tadege M et al. 2009. A WD40 repeat protein from Medicago truncatula is necessary for tissue‐specific anthocyanin and proanthocyanidin biosynthesis but not for trichome development. Plant Physiology 151: 1114–1129. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Paz‐Ares J, Ghosal D, Wienand U, Peterson, Saedler H. 1987. The regulatory c1 locus of Zea mays encodes a protein with homology to myb proto‐oncogene products and with structural similarities to transcriptional activators. The EMBO Journal 6: 3553–3558. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Paz‐Ares J, Ghosal D, Saedler H. 1990. Molecular analysis of the C1‐I allele from Zea mays: a dominant mutant of the regulatory C1 locus. The EMBO Journal 9: 315–321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Pecrix Y, Staton SE, Sallet E, Lelandais‐Briere C, Moreau S, Carrere S, Blein T, Jardinaud MF, Latrasse D, Zouine M et al. 2018. Whole‐genome landscape of Medicago truncatula symbiotic genes. Nature Plants 4: 1017–1025. [DOI] [PubMed] [Google Scholar]
  36. Peel GJ, Pang Y, Modolo LV, Dixon RA. 2009. The LAP1 MYB transcription factor orchestrates anthocyanidin biosynthesis and glycosylation in Medicago . The Plant Journal 59: 136–149. [DOI] [PubMed] [Google Scholar]
  37. Petroni K, Tonelli C. 2011. Recent advances on the regulation of anthocyanin synthesis in reproductive organs. Plant Science 181: 219–229. [DOI] [PubMed] [Google Scholar]
  38. Ramsay NA, Glover BJ. 2005. MYB‐bHLH‐WD40 protein complex and the evolution of cellular diversity. Trends in Plant Science 10: 63–70. [DOI] [PubMed] [Google Scholar]
  39. Robinson MD, McCarthy DJ, Smyth GK. 2010. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 26: 139–140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Shum EY, Jones SH, Shao A, Dumdie J, Krause MD, Chan WK, Lou CH, Espinoza JL, Song HW, Phan MH et al. 2016. The antagonistic gene paralogs Upf3a and Upf3b govern nonsense‐mediated RNA decay. Cell 165: 382–395. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sigman MJ, Slotkin RK. 2016. The first rule of plant transposable element silencing: location, location, location. Plant Cell 28: 304–313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Slotkin RK, Martienssen R. 2007. Transposable elements and the epigenetic regulation of the genome. Nature Reviews Genetics 8: 272–285. [DOI] [PubMed] [Google Scholar]
  43. Soderlund C, Bomhoff M, Nelson WM. 2011. SyMAP v3.4: a turnkey synteny system with application to plant genomes. Nucleic Acids Research 39: e68. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Stracke R, Werber M, Weisshaar B. 2001. The R2R3‐MYB gene family in Arabidopsis thaliana . Current Opinion in Plant Biology 4: 447–456. [DOI] [PubMed] [Google Scholar]
  45. Sun C, Deng L, Du M, Zhao J, Chen Q, Huang T, Jiang H, Li CB, Li C. 2020. A transcriptional network promotes anthocyanin biosynthesis in tomato flesh. Molecular Plant 13: 42–58. [DOI] [PubMed] [Google Scholar]
  46. Sun L, Gill US, Nandety RS, Kwon S, Mehta P, Dickstein R, Udvardi MK, Mysore KS, Wen J. 2019. Genome‐wide analysis of flanking sequences reveals that Tnt1 insertion is positively correlated with gene methylation in Medicago truncatula . The Plant Journal 98: 1106–1119. [DOI] [PubMed] [Google Scholar]
  47. Tadege M, Wen J, He J, Tu H, Kwak Y, Eschstruth A, Cayrel A, Endre G, Zhao PX, Chabaud M et al. 2008. Large‐scale insertional mutagenesis using the Tnt1 retrotransposon in the model legume Medicago truncatula . The Plant Journal 54: 335–347. [DOI] [PubMed] [Google Scholar]
  48. Wang H, Avci U, Nakashima J, Hahn MG, Chen F, Dixon RA. 2010. Mutation of WRKY transcription factors initiates pith secondary wall formation and increases stem biomass in dicotyledonous plants. Proceedings of the National Academy of Sciences, USA 107: 22338–22343. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Xu W, Dubos C, Lepiniec L. 2015. Transcriptional control of flavonoid biosynthesis by MYB‐bHLH‐WDR complexes. Trends in Plant Science 20: 176–185. [DOI] [PubMed] [Google Scholar]
  50. Young, ND , Debellé, F , Oldroyd, GE , Geurts, R , Cannon, SB , Udvardi, MK , Benedito, VA , Mayer, KFX , Gouzy, J , Schoof, H et al. 2011. The Medicago genome provides insight into the evolution of rhizobial symbioses. Nature 480: 520–524. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zhu HF, Fitzsimmons K, Khandelwal A, Kranz RG. 2009. CPC, a single‐repeat R3 MYB, is a negative regulator of anthocyanin biosynthesis in Arabidopsis. Molecular Plant 2: 790–802 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Fig. S1 rh1d is a semidominant mutant.

Fig. S2 Map‐based cloning of the RH1 gene based on M. truncatula genome version Mt4.0.

Fig. S3 Synteny analysis between M. truncatula R108_v1.0 and the mapped region on A17 chromosome 7.

Fig. S4 Chromsome walking verification of the 215 kb deletion in rh1d.

Fig. S5 Expression analysis of RH1 and RH2.

Fig. S6 Targeted mutagenesis in RH1 using the CRISPR/Cas9 system.

Fig. S7 Transcript levels of RH1 in the CRISPR/Cas9 mutant leaflets determined by qRT‐PCR.

Fig. S8 Identification and characterization of rh1 and rh2 mutants.

Fig. S9 Sequence alignment of RH1, RH2 and their close orthologs in Trifolium and Arabidopsis.

Fig. S10 Simplified scheme of the enzyme genes involved in the anthocyanin biosynthetic pathway.

Fig. S11 The predicted MYB‐binding sites in the promoters of upregulated anthocyanin biosynthetic genes in rh1d, as well as RH1 and RH2.

Fig. S12 EMSA showed that RH1 and RH2 can bind to the promoters of MtCHS and MtANS.

Fig. S13 Transcript levels of MtTT8 and MtWD40‐1 in the different regions of the wild‐type leaflets determined by qRT‐PCR.

Fig. S14 RH2 cannot be activated by RH1 and itself.

Fig. S15 The transcripts of anthocyanin biosynthetic genes were decreased in the wild‐type yellow mark with reduced anthocyanin production.

Table S1 Primers used in this study.

Table S2 Annotation of the 215 kb deletion and flanking regions in rh1d.

Table S3 The 448 differentially expressed genes in rh1 d leaves compared with wild‐type leaves.

Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

Click here for additional data file. (994.2KB, pdf)

Articles from The New Phytologist are provided here courtesy of Wiley

RESOURCES