2024 Volume 74 Issue 2 Pages 83-92
Flowering time is an important agronomic trait that is highly correlated with plant height, maturity time and yield in mungbean. Up to present, however, molecular basis of flowering time in mungbean is poorly understood. Previous studies demonstrated that flowering time in mungbean is largely controlled by a major QTL on linkage group 2 (LG2). In this study, the QTL on the LG2 in mungbean was investigated using F2 and F2:3 populations derived from a cross between mungbean cultivar Kamphaeng Saen 2 (KPS2) and wild mungbean accession ACC41. The QTL was narrowed down to a genome region of 164.87 Kb containing a phytochrome gene, designated VrPHYE, encoding phytochrome E (phyE), a known photoreceptor modulating flowering time. Compared to VrPHYE of the wild ACC41, VrPHYE of KPS2 contained several single nucleotide polymorphisms (SNPs) causing amino acid changes. Those SNPs were also found in other mungbean cultivars. Some amino acid changes were predicted to occur in the regulatory region of phytochromes. Gene expression analysis revealed that VrPHYE in KPS2 was expressed significantly higher than that in ACC41. These results showed that VrPHYE is the candidate gene controlling flowering time in the mungbean.
Mungbean (Vigna radiata (L.) R. Wilczek var. radiata) is an ancient and important legume crop of tropical and subtropical regions. The crop is domesticated from its wild form, Vigna radiata var. sublobata (Roxb.) Vercourt, in India about 4,000–4,500 years ago (Fuller and Harvey 2006). After domestication, mungbean has been spread into East, West and Southeast Asia via silk road (Tomooka et al. 2000). At present, mungbean is widely grown and consumed in Asia and is now gaining popularity for cultivation and consumption in several regions of Africa, America and Australia. India is the largest producer of mungbean, followed by Myanmar and China with the cultivation area of about 4.5 (Anonymous 2021), 1.2 (MAOLI 2019) and 0.8 (Nair and Schreinemachers 2020) million hectares, respectively. Popularity of the mungbean cultivation is due to the facts that the crop has a short life cycle of only about 55–75 days and ability to fix atmospheric nitrogen to the soil, use less water than other crops, and is relatively drought tolerance (Fernandez and Shanmugasundaram 1988). The crop fits well with several cropping systems, especially cereal-based system. For example, in Thailand, the cultivation of mungbeans starts after the cultivation of rice and maize finish. Mungbean is mainly grown for dry seeds. Dry seeds of mungbean contain high protein (~20–25%) and carbohydrate (~65–75%) (Somta and Srinives 2007). The seeds are used to prepared several foods and processed into sprouts, noodles, starch, splited beans, egg substitutes, and plant-based meats. Phytochemicals presented in the mungbean seeds are believed to have detoxification activity that can refresh the mind, alleviate heat stroke, and reduce swelling in summer (Tang et al. 2014).
Flowering time is a key phenological trait of plant adaptation and reproduction. In general, flowering time is largely influenced by photoperiod and temperature (Song et al. 2013). In crop plants, flowering time is an important agronomic trait that is highly correlated with plant height, maturity time and yield. Therefore, flowering time is correlated with crop management and harvest. Mungbean is a tropical and quantitative short-day plant (Summerfield and Lawn 1987). Therefore, knowledge and understanding on the molecular genetic mechanism underlying the flowering time in mungbean is crucial for expanding mungbean production to new regions. There are some reports on quantitative trait loci (QTLs) and candidate genes controlling flowering in the mungbean. However, most of the QTLs or candidate genes identified for flowering time are different and have not been confirmed/validated (reviewed in Somta et al. 2022). In the previous studies by Isemura et al. (2012), Kajonphol et al. (2012), and Somta et al. (2015), major QTLs controlling flowering time were always detected on linkage group 2 (LG2) and they are possibly the same locus (reviewed in Somta et al. 2022). Kajonphol et al. (2012) identified four QTLs on linkage group (LG) 2, 4, and 11 for days to first flowering (DFF) in an F2 population grown during long-day (LD) and short-day (SD) conditions in Thailand. The QTL Fld2 on LG2 and Fld4.2 on LG4 were the major QTLs explaining for 15.9 and 28.6% of the flowering time variation, respectively. Isemura et al. (2012) identified four QTLs, one each on LGs 2, 4, 6, and 11 in an F2 population (wild × cultivated) grown under LD and SD conditions in Japan where the natural day length gradually decreased from 14 h 12 min in July to 9 h 43 min in December. The Fld5.2.1- on LG2 and Fld5.4.1- on LG4 had the largest effect with the phenotypic variation explained by the QTL (coefficient of determination (R2)) of 32.9% and 24.0%, respectively. Somta et al. (2015) detected seven QTLs for DFF in F2:3 population grown in SD and LD conditions and F2:4 population grown in SD condition. Five QTLs, two on LG2 and one each on LGs 4, 5, and 6, were associated with LD condition, while two QTLs on LG2 were associated with SD condition. The QTLs qDFL2.1 and qDFL2.2 were detected in both conditions and expressed similar effect with R2 of about 29.0%. The QTLs on LGs 2, 4, 6, and 11 were always detected, especially those on LGs 2 and 4. So, the results in these reports showed that QTLs on LG2 were always detected in both the SD and LD conditions, while the QTLs on LGs 4, 6, and 11 were always identified in the LD condition.
The QTLs on LG2 appeared to be the major genetic factors playing important role in regulating flowering time in mungbean. Nonetheless, marker density of the linkage maps used in the studies of Isemura et al. (2012), Kajonphol et al. (2012), and Somta et al. (2015) are sparse and, up to present, molecular basis of the flowering QTLs on the LG2 has not yet been investigated. In this study, we report fine mapping of the QTL on LG2 and identification of candidate gene controlling flowering time in the mungbean.
Two mapping populations, F2 and F2:3 population, were used in this study. These populations were derived from a cross between Kamphaeng Saen 2 (hereafter called KPS2) and ACC41. KPS2 is a commercial mungbean cultivar of Thailand. It is developed by Kasetsart University, Thailand. ACC41 is a wild mungbean originated from Australia. The seeds of ACC41 were obtained from Australian Tropical Crops and Forages Genetic Resources Centre, Australia. The F2 population comprised 575 individuals. The F2 population and 10 plants of each parent were grown under a field condition at Kasetsart University, Kamphaeng Saen Campus, Nakhon Pathom, Thailand from February to May 2018. Twenty-five days after planting, days to first flowering of each plant was observed and recorded. DNA extraction was performed following the procedures reported by Lodhi et al. (1994). Details of the F2:3 generation is described in the section “Confirmation of the QTL for flowering time”. It is noteworthy that the F2 population used in this study was the same population used to identify the gene for seed dormancy previously reported by Laosatit et al. (2022).
Bioinformatics analysis of the mungbean QTLs on LG2 controlling flowering time and development of new markers for fine mappingSince Somta et al. (2015) reported that the qDFL2.1 and qDFL2.2 on LG2 controlling DFF were located between marker intervals GBssr-MB87–DMB-SSR160 and DMB-SSR160–CEDAAG002, respectively, while Kajonphol et al. (2012) reported that the QTL Fld2 on LG2 governing DFF was localized to the marker interval DMB-SSR160–VR0364 and Isemura et al. (2012) reported that the Fld5.2.1- on LG2 regulating DFF was delimited by the marker interval GMES0477–CEDG026. So, physical locations of the markers GBssr-MB87, DMB-SSR160, CEDAAG002 and CEDG026 on the reference genome sequence of the mungbean cultivar Sulv1 (Yan et al. 2020) were determined by BLASTN analysis. Once determined, all the genes in genome region harboring the QTLs for DFF were checked and genes having function(s) involving in the flowering time (Lin et al. 2020) were selected. Sequence data of genome region of 800-Kb covering each selected gene was downloaded and identified for simple sequence repeats (SSR) using SSRIT (Temnykh et al. 2001). Subsequently, primers were designed to amplify SSRs using Primer3 (Untergasser et al. 2012). In addition, each of the 800-Kb region was aligned against the genome sequence of Vigna sp. accession NI1135 (https://doi.org/10.1101/2022.03.28.486085), an accession genetically closest to wild mungbean (Takahashi et al. 2018), to identify insertions/deletions (InDels). Subsequently, primers for InDels were designed using Primer3 (Untergasser et al. 2012).
DNA marker, linkage and QTL analysesThe primers (Supplemental Table 1) were screened for polymorphism between KPS2 and ACC41. DNA marker analysis was carried out as per Laosatit et al. (2022). Briefly, polymerase chain reaction (PCR) was carried out in a total volume of 10 μl containing 5 ng of DNA template, 1 × Taq buffer, 2 mM MgCl2, 0.2 mM dNTPs, 1 U Taq DNA polymerase (Thermo Fisher Scientific Inc., Waltham, MA, USA), and 2.5 μM each of forward and reverse primers. PCR products were electrophoresed on 5% polyacrylamide gel electrophoresis and visualized by silver staining. Markers showing clear polymorphic DNA bands were used to analyze the F2 population.
Linkage map was constructed using software QTL IciMapping 4.2 (Meng et al. 2015). Markers were grouped with logarithm of odds (LOD) values of 5.0, and then ordered using REcombination Counting and ORDering (RECORD) method (Van Os et al. 2005). Distances between markers were calculated using Kosambi’s mapping function (Kosambi 1943).
QTL controlling DFF was localize onto the linkage map using the inclusive composite interval mapping method (ICIM) (Li et al. 2007) by the QTL IciMapping 4.2. A significant LOD threshold for QTL identification was determined by a 10,000 permutation test at P = 0.001. DNA marker and DFF data used in the QTL analysis of the F2 population were shown in Supplemental Table 2.
Confirmation of the QTL for flowering timeOnce the QTL for flowering time was identified, the locus was confirmed using an F2:3 population. This population comprised 217 F3 plants that were progenies of 15 F2 (KPS2 × ACC41) plants showing heterozygous genotype at the markers flanking the QTL for flowering time. The number of F3 plants derived from each F2 plant was between 10 and 25. The F2:3 population together with 10 plants of each parent were grown in a field condition from February to May 2020 in the same location with the F2 population. Data collection for flowering time (DFF), DNA extraction, DNA marker analysis, linkage analysis and QTL analysis of the F2:3 population were the same as described for the F2 population. DNA marker and DFF data used in the QTL analysis of this population were shown in Supplemental Table 3.
Sequence and RT-qPCR analyses of candidate geneBased on the result from fine mapping, the gene EVM0010407 (VrPHYE) was selected as the single candidate gene controlling the flowering time. A genomic region harboring the gene VrPHYE in KPS2 and ACC41 was amplified using primers listed in the Supplemental Table 1. PCR was conducted as described above. The PCR products were electrophoresis in 1.0% agarose gel to confirm whether a single band is amplified. Then, the products were eluted from the gel, cleaned and sequenced. Sanger sequencing was performed using ABI 3730xl DNA Analyzer (Applied Biosystems, CA, USA) by AGTC Genomics (Kuala Lumpur, Malaysia). The VrPHYE sequences from KPS2 and ACC41 were aligned with the reference sequences of the mungbean cultivar Sulv1 (Yan et al. 2020), Jilv7 (Liu et al. 2022) and VC1973A (Ha et al. 2021) using Clustal Omega (Sievers et al. 2011). Predicted protein sequence encoded by the VrPHYE in all the mungbean cultivars were also aligned. The protein sequence was also subjected to InterProScan (Paysan-Lafosse et al. 2023) to identify protein domain(s).
Expression of the VrPHYE in KPS2 and ACC41 was detected and quantified by quantitative reverse transcription PCR (RT-qPCR). KPS2 and ACC41 were grown in a crossing block. At 35 and 45 days after planting (DAP), total RNA was extracted from leaves of both accessions. RNA extraction and complementary DNA (cDNA) synthesis were conducted as described by Laosatit et al. (2022). Primers for RT-qPCR of the VrPHYE and reference gene VrACTIN (LOC106770112) were listed in the Supplemental Table 1. The RT-qPCR was performed with three biological and technical replicates were conducted using ViiA 7 Real-Time PCR System (Applied Biosystems). PCR reaction mixtures and thermal cycle conditions were the same as described by Laosatit et al. (2022). Expression levels of the gene EVM0010407 were calculated based on the 2–ΔΔCT method (Livak and Schmittgen 2001). Statistical difference in the gene expression between KPS2 and ACC41 was determined by t-test at 5% probability using R-program. The expression of the VrPHYE gene was detected at only 35 and 45 DAP because at later stages KPS2 was at maturing and harvesting stages and the leaves were old, while ACC41 was still at the vegetative stage or may be at the beginning of the flowering initiation, and thus comparison at the later stages was not suitable.
Phylogenetic analysisThe predicted proteins encoded by the VrPHYE in the mungbean together with their homologous proteins from other plants including azuki bean (Vigna angularis (Ohwi) Ohwi and Ohashi), black gram (Vigna mungo (L.) Hepper), cowpea (Vigna unguiculata (L.) Walp.), common bean (Phaseolus vulagaris L.), lima bean (Phaseolus lunatus L.), lablab (Lablab purpureus (L.) Sweet), soybean (Glycine max (L.) Merr.), wild soybean (Glycine soja Siebold and Zucc.), groundnut (Arachis hypogaea L.), wild groundnut (Arachis duraensis L.), pigeon pea (Cajanus cajan (L.) Millsp.), chickpea (Cicer arietinum L.), barrel medic (Medicago truncatula Gaertn.), Lotus japonicus (Regel) K. Larsen and subterranean clover (Trifolium subterraneum L.) were subjected to phylogenetic analysis using MEGA 11 (Tamura et al. 2021). The sequences were aligned using MUSCLE and the phylogenetic tree was constructed using maximum likelihood method with 1,000 bootstraps. Details of the protein sequences used in the phylogenic analysis are shown in Supplemental Table 4.
Previous studies showed that major QTLs (qDFL2.1, qDFL2.2, Fld, and Fld5.2.1-) controlling mungbean flowering time are on LG2, but location(s) of those QTLs on the mungbean reference genome is not known. In this study, bioinformatics analysis showed that markers GBssr-MB87, DMB-SSR160, CEDAAG002 and CEDG026 associating with the major QTL(s) for flowering time located between the positions 36.172 and 42.480 Mb on the chromosome 4 of the Sulv1 genome. So, the qDFL2.1, qDFL2.2, Fld, and Fld5.2.1- controlling the flowering time were believed to reside in this region. We checked all the genes in the region focusing on the one(s) with annotated function relating to flowering time. Four flowering-related genes were identified including EVM0010407, EVM0000157, EVM0026933 and EVM0031852. EVM0010407 and EVM0000157 were annotated to encode phytochrome E (phyE) and agamous-like MADS-box protein AGL82 (AGL82), respectively, while both EVM0026933 and EVM0031852 were annotated to encode clock-associated PAS protein ZEITLUPE (ZTL). These genes located at the positions 39.700, 39.128, 42.035, and 42.041 Mb, respectively.
QTL region controlling flowering and identification of candidate geneThe QTL(s) controlling flowering time appeared to locate on region between 36.172 and 42.480 Mb on the chromosome 4 of the Sulv1 mungbean reference sequence. In order to finely mapped the QTLs on LG2, SSR and InDel markers were developed from the 38.00–43.00 Mb region. In total, 83 SSR and 64 InDel markers were developed (Supplemental Table 1). We narrowed down the QTL region using an F2 population derived from parents showing contrasting flowering time (days to first flowering (DFF)). DFF in the F2 population varied between 32 and 80 days with a mean of 46.2 days. DFF in the mapping parents KPS2 and ACC41 was of 38.2 and 64.0 days, respectively. DFF in the F2 population showed continuous distribution (Fig. 1A).
Frequency distribution of days to first flowering in F2 (A) and F2:3 (B) populations derived from a cross between cultivated mungbean Kamphaeng Saen 2 (KPS2) and wild mungbean ACC41.
Screening of 147 markers in the parents revealed that 57 showed polymorphisms (Supplemental Table 1). Fifteen markers were selected and used to genotype the F2 population. Linkage analysis of the LG2 for the F2 population showed that the LG2 spanned 13.13 cM in length with average distance between adjacent markers of 0.94 cM. QTL analysis by ICIM method identified a single QTL controlling the days to first flowering (Fig. 2A, Table 1). The QTL was mapped between the markers VrE1-SSR2 and Vr05-ID19. It accounted for 17.79% of the total variation of the days to first flowering in the F2 population and expressed the additive effect of –3.96 and dominant effect of 0.64. The QTL was designated qFld2.1.
LOD graph of qFld2.1 on linkage group 2 controlling days to first flowering in F2 (A) and F2:3 (B) populations derived from a cross between cultivated mungbean Kamphaeng Saen 2 (KPS2) and wild mungbean ACC41.
Location of effects of QTLs for days to first flowering detected in the F2 and F2:3 populations derived from a cross between mungbean cultivar Kamphaeng Saen 2 and wild mungbean accession ACC41
| Population | QTL name | Position (cM) | Marker interval | LOD score | PVE (%) | Additive effect | Dominant effect |
|---|---|---|---|---|---|---|---|
| F2 | qFld2.1 | 5.5 | VrE1-SSR2–Vr05-ID019 | 24.34 | 17.79 | –3.96 | 0.64 |
| F2:3 | qFld2.1 | 6.3 | VrE1-SSR11–VrE1-SSR2 | 6.29 | 12.44 | –4.09 | 1.49 |
The qFld2.1 was confirmed using an F2:3 population derived from F2 plants. In this population, DFF ranged from between 30 to 75 days with a mean of 53.2 days. DFF in KPS2 was 38.7 days, while that in ACC41 was 64.4 days. DFF in the F2:3 population showed continuous distribution (Fig. 1B). ICIM detected a single QTL, qFld2.1, for the days to first flowering (Fig. 2B, Table 1). The qFld2.1 was located between the markers VrE1-SSR11 and VrE1-SSR2. It explained 12.44% of the total variation of the days to first flowering in the F2:3 population with the additive and dominant effects of –4.09 and 1.49, respectively.
Based on the results in the F2 and F2:3 populations, the location of the qFld2.1 was in region from the marker VrE1-SSR10 to VrE1-SSR21 with VrE1-SSR2 always the flanking marker to the qFld2.1. Candidate gene(s) was identified using the reference genome sequence of mungbean cultivar Sulv1. BLASTN analysis showed that the markers VrE1-SSR10 and VrE1-SSR21 were 164.87 Kb apart, being at the positions 38,936,736 and 39,101,608 on the chromosome 4 (Fig. 3). There were 14 predicted genes in the 164.87-Kb region, including the gene EVM0010407 (Fig. 3, Supplemental Table 5). So, the gene EVM0010407, designated VrPHYE, was chosen as the candidate gene at the qFld5.1 for the days to first flowering.
A comparative map showing location of qFld2.1 detected on linkage group 2 in the F2 (left) and F2:3 (middle) populations and position of the qFld2.1 on the reference genome sequence of the mungbean cultivar Sulv1 (left).
VrPHYE in KPS2 and ACC41 was sequenced. The VrPHYE sequences of KPS2 and ACC41 were aligned with the Sulv1 reference sequence and VC1973A reference sequence. The alignment revealed that several polymorphisms in 5ʹ untranslated region (UTR), exons, introns, and 3ʹ UTR (Fig. 4). Most of the polymorphisms discriminated between the cultivated (KPS2, Sulv1, Jilv7 and VC1973A) and wild (ACC41) mungbeans. Nonetheless, coding sequence (CDS) comparison between the mapping parents (KPS2 and ACC41) revealed 31 single nucleotide polymorphisms (SNPs). In addition, 5ʹ UTR sequence comparison between ACC41 and KPS2 showed 15 SNPs and 2 InDels (Supplemental Fig. 1). One of the InDel was a 38 bp (insertion in KPS2). This insertion also existed in Sulv1, Jilv7, and VC1973A.
PHYE sequence polymorphisms among mungbean accessions ACC41, Kamphaeng Saen 2 (KPS2), Sulv1, VC1973A and Jilv1.
VrPhyE protein sequence alignment among ACC41, KPS2, Jilv7, Sulv1 and VC1973A revealed the ACC41 and KPS2 had the same length of 1,121 amino acids and differed from Sulv1, Jilv7 and VC1973A which had the same length (1,125 amino acids) and sequence (Fig. 5). There were several amino acid changes between the wild mungbean ACC41 and all the cultivated mungbeans. Comparison between ACC41 and KPS2 revealed 17 amino acid changes. Nearly all of these amino acid changes were also found in other cultivated mungbeans. Based on InterProScan, the VrPHYE was composed of 6 protein domains including 3 Period/Arnt/Single-Minded (PAS), a cGMP phosphodiesterase/adenylyl cyclase/FhlA (GAF), a phytochrome-specific (PHY) domain and a histidine kinase-related domain (HKRD) domain (Fig. 5). All of the 17 amino acid changes between ACC41 and KPS2 existed in the C-terminal region: 7 and 9 of the changes occurred in the PAS and HKRD domains, respectively (Fig. 5).
VrPhyE sequence polymorphisms among mungbean accessions ACC41, Kamphaeng Saen 2 (KPS2), Sulv1, VC1973A and Jilv1. Protein domains of the PhyE are also shown. PAS = Period/Arnt/Single-Minded domain, GAF = cGMP phosphodiesterase/adenylyl cyclase/FhlA domain, PHY = phytochrome-specificdomain and HKRD = histidine kinase-related domain.
RT-qPCR analysis showed that expression level of the VrPHYE in KPS2 was significantly higher than that in ACC41 at both 35 and 45 DAP (Fig. 6), being 14.2 and 2.1 times higher than ACC41, respectively.
Expression level of VrPHYE in leaves of Kamphaeng Saen 2 (KPS2) and ACC41 at 35 (A) and 45 (B) days after planting. Error bars are standard error.
A phylogenic tree constructed from the phyE proteins from different legumes showed that the proteins are largely clustered into two groups, I and II (Fig. 7). Group I comprised phyE proteins from chickpea, subterranean cover and barrel medic, while group II comprised protein from mungbean and the other legumes. phyE from mungbean was closely related with those from black gram, azuki bean and cowpea. All the legume species in the group I were long-day plants, while all the species in the group II except Lotus japonicus were short-day plants.
A phylogenic tree depicting relationship of phytochrome E proteins from different legume species. The tree is constructed by maximum likelihood method.
Flowering time is an important agronomic and adaptive trait in crop production, and is therefore a major selection criterion in plant breeding. Usually, wild progenitor(s) and landrace cultivars of tropical crops are short-day plants showing high degree of sensitivity to photoperiod and causing them their flowering is progressively delayed as the photoperiod is increased. Mungbean is a quantitative short-day plant that time to flowering varies appreciably depending on the genotype, photoperiods and temperatures prevailing during the period after sowing (Summerfield and Lawn 1987, Vas Aggarwal and Poehlman 1977). However, mungbean is probably the most short-duration field crop that many improved cultivars mature and can be reaped within 55–65 days after planting (Fernandez and Shanmugasundaram 1988), making the crop useful for diversification in cropping systems. Therefore, understanding and maintaining reduced sensitivity to photoperiod and temperature is important in mungbean breeding. Although QTLs for flowering time have been identified in the mungbean, very little is known about molecular genetic architecture of flowering time in this crop. In this study, we narrowed down the major QTL on LG2 controlling the flowering time in the mungbean from a region of 6.31 Mb to a region of 164.87 Kb (Fig. 3) and showed that VrPHYE encoding phytochrome E (phyE) is the candidate gene for the flowering time in the mungbean (Figs. 3–6). In land plants, phytochromes are photoreceptors of red and far-red light that are responsible for triggering responses to specific light signals. The phytochromes play roles in germination, de-etiolation, shade avoidance, and flowering of plants (Legris et al. 2019). Five phytochromes have been identified in plants, phyA to phyE (Clack et al. 1994, Rockwell et al. 2006). Based on phylogenetic analysis in Arabidopsis thaliana L., the phytochromes are grouped into three classes, viz. phyA, phyB and phyC in which the phyB class is composed of phyB, phyD and phyE (Clack et al. 1994). Based on BLASTP analysis using A. thaliana phytochrome protein sequences as queries, genome of mungbean cultivar Sulv1 (Yan et al. 2020) contained four phytochrome genes including two phyA (EVM0031724 and EVM0012118), one phyB (EVM0002707) and one phyE genes (P. Somta, unpublished data). In A. thaliana, phye single mutants showed the same phenotypes with wild-type plants but showed earlier flowering in a phya phyb background, indicating that the function of phyE overlaps with that of phyA and phyB to delay flowering (Devlin et al. 1998). Later, again in A. thaliana, Clack et al. (2009) demonstrated that (1) phyE do not homodimerize, (2) heterodimeric phytochromes containing phyE or phyC play roles in light regulation of flowering time, (3) in the obligate heterodimerization of phyE and phyC, the latter become unstable by removal of its phyB binding partner, and (4) under SD condition, phyE and phyB mutants flower earlier than the wild type, and combination of these two mutants exacerbates early flowering. They further illustrated that in the absence of its phyB and phyD dimerization partners, phyE has biological activity. Recently, however, Sanchez-Lamas et al. (2016) showed that none of the phytochromes alone conferred a photoperiodic response and phyE and phyB are the strongest repressor of flowering time in A. thaliana and the repression is highly dependent on ambient temperature under long-day conditions in A. thaliana, albeit both phyB and phyC are needed to confer flowering response to photoperiod. Shapulatov et al. (2021) demonstrated that upregulation of both PHYA and PHYB under far-red light is dependent on PHYE. A recent study in soybean demonstrated that PHYE1 (Glyma.09g088500) may delay flowering (Dietz et al. 2022). These supported our findings that the gene VrPHYE is the candidate for the flowering time in the mungbean. Several SNPs resulted in phyE amino acid changes between the wild mungbean ACC41 and the cultivated mungbeans (Fig. 5). All these cultivated mungbeans are early flowering. The amino acids changes occurred in the PAS and HKRD domains in the C-terminal (Fig. 5). These domains constitute the C-terminal output module (OPM) of the phytochromes (Cheng et al. 2021, Rockwell et al. 2006) which mediates dimerization and signal transmission to the downstream effectors (Cheng et al. 2021). So, the mutation(s) in the OPM of the phyE is likely the cause of early flowering in the mungbean. It is also worth noting that the VrPHYE showed different expression levels between the wild mungbean ACC41 and the cultivated mungbean KPS2 (Fig. 6). The different expression may stem from the sequence variations in the promotor and/or 5ʹ UTR regions between ACC41 and KPS2, especially the 38-bp InDel in the 5ʹ UTR region (Supplemental Fig. 1). Nonetheless, although fine mapping (Figs. 2, 3), gene sequencing (Figs. 4, 5), and gene expression analysis (Fig. 6) strongly suggested that VrPHYE is the candidate gene controlling flowering time in mungbean, additional study is necessary due to the fact that VrPHYE was not narrowed to a single gene and functional study of the VrPHYE was not conducted to confirm the gene has function in photoperiod regulation of flowering in the mungbean.
QTL mapping in azuki bean (Vigna angularis (Ohwi) Ohwi and Ohashi) using a population from a cross between cultivated and wild azuki beans demonstrated that phyE gene LOC108331824 is one of the two candidate genes controlling flowering time at the QTL qVaFld4.1, the strongest-effect QTL for the flowering time (Li et al. 2017). Similarly, QTL mapping in cowpea (Vigna unguiculata (L.) Walp.) using a population of a cross between cultivated and wild cowpeas showed that the gene Vigun09g050600 encoding phyE is a candidate gene controlling flowering time at the locus CFt9, the strongest-effect QTL for the flowering time (Lo et al. 2018). In our study, VrPHYE is the candidate gene flowering time in populations deriving from a cross between the cultivated and wild mungbeans. These suggest that phyE gene is the most important gene playing role in reducing sensitivity to photoperiod during the process of domestication of legume crops of the genus Vigna. It is worth noting that phyE proteins in mungbean, azuki bean, and cowpea were closely related (Fig. 7). Additional study should be conducted to determine how the phyE modulate the flowering time in the mungbean and other Vigna species.
In mungbean and azuki bean, flowering time and pod maturity time are highly correlated and the largest-effect QTLs controlling flowering time and maturity time are closely linked or mapped to the same position on the LG2 (Isemura et al. 2012, Kaga et al. 2008, Kajonphol et al. 2012, Li et al. 2017). The QTLs for these two traits on the LG2 are also closely linked or locate to the same position with QTL controlling seed weight (Isemura et al. 2012, Kaga et al. 2008, Li et al. 2017, Somta et al. 2015) and internode length (Isemura et al. 2012). It is possible the QTL on LG2 has a pleiotropic effect controlling flowering time, maturity time, seed weight and internode length in mungbean and azuki bean. Devlin et al. (1998) showed that phyE plays a predominant role in controlling the internode elongation response to end-of-day far-red light of phyA phyB mutants. Since the phyE gene is the candidate gene at the QTL on the LG2 responsible for flowering time in both mungbean (this study) and azuki bean (Li et al. 2017), the gene may also be responsible for maturity time and seed weight. If this is really the case, it would suggest that mutation(s) in phyE gene contributes greatly to phenology changes and seed weight during domestication of these legumes. Additional study is necessary to determine the effect of the phyE gene to the maturity and seed weight/seed yield.
PS, KL and XC conceived idea, designed the study and secured research fund. PS and KL supervised the study. KA, KL and TY conducted field experiment. KA, KL, YL, JC and XY performed molecular experiments. KA carried out all data analyses. PS and KA wrote and revised the manuscript. All authors read and approved the manuscript.
This research was supported by the Office of National Higher Education Science Research and Innovation Policy Council via the Program Management Unit for Human Resources and Institutional Development, Research and Innovation (Grant No. B16F640185), and the China Agriculture Research System of MOF and MARA-Food and Legumes (CARS-08).
