Small RNA sequencing reveals drought-responsive microRNAs and regulatory networks in roots of mungbean genotypes differing in drought tolerance

Abstract

Background

Drought stress is a major environmental constraint limiting mungbean productivity, particularly in arid and semi-arid regions. However, the molecular mechanisms underlying drought tolerance in mungbean remain poorly understood. MicroRNAs (miRNAs), which function as important post-transcriptional regulators of gene expression, play critical roles in modulating plant responses to abiotic stresses. This study aimed to identify drought-responsive miRNAs and elucidate their regulatory roles in contrasting mungbean genotypes under drought stress.

Results

Seven widely cultivated mungbean genotypes were evaluated under controlled drought stress conditions using physiological, molecular, and yield-related parameters. Traits including root and shoot length, biomass accumulation, nodule number, expression of drought-responsive genes under prolonged stress (30 days), and yield attributes were assessed. Based on these analyses, genotype cv. ML267 was identified as drought tolerant, whereas cv. PDM139 was identified as drought susceptible. To investigate the molecular basis of drought tolerance, small RNA sequencing was performed using root tissues of the contrasting genotypes. A total of 61 novel drought-responsive miRNAs and 274 conserved miRNA homologs were identified, showing genotype-specific differential expression under drought stress conditions. Quantitative RT-PCR validation of selected miRNAs and their predicted target genes demonstrated inverse expression patterns, confirming miRNA-mediated post-transcriptional regulation. Functional annotation and network analyses revealed that these miRNA-target regulatory modules are associated with hormone signaling, antioxidant defense, root development, and stress adaptation pathways. The study also revealed complex miRNA-mediated regulatory networks that regulate mungbean gene expression in response to drought.

Conclusions

This study offers a thorough understanding of small RNA-mediated regulation associated with drought tolerance in mungbean. The identified drought-responsive miRNAs along with their target genes indicate potential molecular candidates for enhancing drought resistance and could aid in the creation of climate-resilient mungbean varieties through molecular breeding and biotechnological methods.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12870-026-09401-0.

Background

Drought stands as one of the major abiotic challenges hindering legume crop production on a global scale, and its increasing occurrence and severity due to climate change represent a considerable risk to food security, especially in rainfed agricultural systems [1]. Mungbean (Vigna radiata L.), is an important short-duration pulse crop predominantly grown in Asia, is valued for its nutritional benefits and its role in improving soil fertility through biological nitrogen fixation. Nevertheless, its yield is highly vulnerable to insufficient soil moisture, particularly during the early growth and reproductive phases [1, 2]. Given that drought resistance is a complex characteristic influenced by various physiological, biochemical, and molecular mechanisms, combining phenotypic assessments with molecular analysis is vital for identifying resilient genotypes and understanding their adaptive strategies [1, 3].

The effects of drought are especially harsh during the reproductive phase, resulting in significant yield reductions in mungbean. Even though mungbean is significant for agriculture, its genetic enhancement has advanced at a slower rate compared to major cereal crops like rice, maize, and wheat [4]. Furthermore, the limited genetic diversity of cultivated mungbean, due to the recurrent use of a small selection of parental lines, restricts breeding initiatives focused on enhancing drought resilience and yield consistency.

Recent developments in genomics, particularly the release of the mungbean draft genome, have offered a significant resource for gene identification and functional research [5]. Although these advancements aid molecular studies, a thorough comprehension of gene function necessitates high-resolution transcriptomic analysis. Past research on mungbean has primarily focused on its morphological, physiological, and biochemical reactions to drought stress, providing limited understanding of the adaptive mechanisms involved. Nonetheless, comprehensive investigations at the transcriptome level, especially those focused on regulatory networks, are still scarce.

Besides transcriptional regulation, epigenetic mechanisms are essential in influencing how plants respond to environmental stress. Among these, small non-coding RNAs, especially microRNAs (miRNAs), represent a key layer of post-transcriptional regulation that fine-tunes gene expression during stress adaptation. miRNAs regulate the expression of transcription factors and signaling molecules related to root development, hormone signaling, osmotic equilibrium, and antioxidant defense [6–8]. Integration of small RNA sequencing with transcriptomic approaches has facilitated the discovery of miRNA–mRNA regulatory networks that govern complex stress responses [6, 9].

Several conserved miRNAs have been implicated in drought responses across various plant species. For example, miR169 controls NF-YA transcription factors in rice [10], while miR398 influences ROS detoxification by targeting Cu/Zn-superoxide dismutase in Arabidopsis and Medicago truncatula [11]. In maize, miR164, miR167, and miR393 play roles in auxin-driven root development during water scarcity [12], while in wheat, miR156, miR159, and miR396 modulate transcriptional and growth responses to drought [13]. Comparable regulatory functions have been observed in legumes such as soybean, chickpea, common bean, and cowpea [14–17], emphasizing both conserved and species-specific functions of miRNAs in stress adaptation.

While drought-responsive miRNAs have been discovered in mungbean, the majority of research has used short-term polyethylene glycol (PEG)-induced osmotic stress, which might not effectively replicate drought conditions found in the field [18]. Furthermore, there is a lack of integrative analyses that connect physiological performance with miRNA expression across different genotypes experiencing extended soil-induced drought stress. Specifically, root-specific miRNA-mediated regulatory networks, which are critical for sensing and responding to water deficit, are still poorly understood.

To address this gap, the present study evaluated seven widely cultivated mungbean genotypes under 30 days of soil-imposed drought stress to identify tolerant and susceptible lines. Based on comprehensive physiological, biochemical, and molecular analyses, contrasting genotypes were selected for small RNA sequencing to characterize drought-responsive miRNAs and their target genes. Given that miRNAs constitute an important layer of epigenetic regulation, this study aims to elucidate small RNA–mediated regulatory mechanisms underlying drought adaptation. Validation of selected miRNAs and their targets was performed using qRT-PCR to confirm their regulatory relationships. By integrating phenotypic screening with transcriptomic and small RNA profiling, this study provides new insights into miRNA-mediated gene regulatory networks associated with drought tolerance in mungbean. To our knowledge, this is the first comprehensive study integrating physiological screening with root-specific small RNA and transcriptome profiling to elucidate miRNA-mediated gene regulatory networks underlying drought tolerance in contrasting mungbean genotypes.

Materials and methods

Plant material and drought stress treatment

Seven mungbean genotypes cv. PUSA-105, K-851, MUM-2, ML-267, PDM-139, SML-667, and PUSA-Ratna widely cultivated across India during kharif, summer, and spring seasons, were evaluated for drought tolerance at the reproductive stage under 30-day drought stress conditions. Seeds of these genotypes were obtained from Indian Agricultural Research Institute (IARI), New Delhi. The genotypes are cultivated breeding lines/varieties developed by recognized agricultural research institutes in India and were used solely for research purposes.

Seeds of all genotypes were surface-sterilized, germinated on moist absorbent cotton layered with filter paper in Petri dishes, and maintained in a temperature-controlled chamber at 22 °C under dark conditions for 2 days, followed by 2 days under a 14 h/10 h light/dark photoperiod with 70 ± 5% relative humidity. The uniform seedlings were then transplanted into soil under net house conditions. Plants were watered at 4–5-day intervals to promote normal vegetative growth, and drought stress was imposed after 30 days by withholding irrigation. Plants were monitored daily for visible drought stress symptoms.

Phenotypic and agronomic parameters—including root and shoot length, fresh and dry biomass, plant height, number of branches, pods per plant, seeds per pod, and seed yield per plant were recorded under both control and drought conditions. In addition, molecular analyses were performed to assess the relative expression of drought-responsive and antioxidant enzyme genes such as SOD, APX, CAT, and POX, which are associated with reactive oxygen species (ROS) detoxification. The root samples were collected from all seven genotypes were used for dry weight and Relative water content (RWC) as per the procedure described by Kumar et al. [19]. Comparative evaluation of these traits enabled the identification of contrasting genotypes differing in drought tolerance.

RNA isolation, small RNA library construction and small RNA transcriptome sequencing

Total RNA was extracted from 150 mg of root tissue collected from both control and drought-stressed plants using the Plant RNA Mini Kit (Genetix, USA) following the manufacturer’s protocol. The purity and concentration of RNA were evaluated using a NanoDrop 8000 spectrophotometer (Thermo Fisher Scientific, USA) and confirmed by agarose gel electrophoresis. RNA samples exhibiting an A260/A280 ratio of 1.8–1.9, an A260/A230 ratio of 2.0–2.4, and a RNA integrity number (RIN) ≥ 8 were considered suitable for sequencing. The small RNA libraries were generated from equal quantities of high-quality RNA from three independent biological replicates per treatment using the TruSeq RNA Library Preparation Kit v2 (Illumina, San Diego, CA, USA) as per the manufacturer instructions.

RNA sequencing and data analysis

Small RNA sequencing was employed to investigate miRNA-mediated post-transcriptional regulatory mechanisms under drought stress. The small RNA libraries were sequenced on the Illumina NovaSeq 6000 platform to generate approximately 101 bp paired-end reads for each sample. The raw sequencing data were obtained in FASTQ format and subjected to quality control using the NGS QC Toolkit [20]; http://www.nipgr.res.in/ngsqctoolkit.html. Low-quality reads and adaptor sequences were filtered out, retaining only high-quality reads with a Phred quality score (Q) > 20. Genome indexing was performed using Bowtie v2.2.6 to optimize data compression and minimize memory usage. The filtered high-quality reads were aligned to the mungbean miRNAs from Plant MicroRNA Encyclopedia (PmiREN) Database (www.pmiren.com).

Data processing and identification of miRNAs

Following quality control (QC) filtering, small RNA reads containing unique adaptor sequences were processed using the sRNA toolkit 2.4 (Plant version) [21]. This step included trimming of 3′ adaptors, size selection, and removal of sequences unrelated to small RNAs [22]. Only 17 to 25 nucleotide long reads were used for miRNA prediction. The processed reads were then mapped to the mungbean miRNA taken from PmiREN using bowtie [23]. Read counts were normalized as transcript per million (RPM) prior to fold-change estimation, a standard normalization approach for RNA-Seq studies [24]. Normalized expression values for miRNAs were then used to calculate fold change and visualize expression patterns. Differentially expressed miRNAs were defined using a log2 fold-change threshold of ± 2. Both up- and down-regulated miRNAs were visualized in heatmaps using meV tool https://sourceforge.net/projects/mev-tm4/ color gradient (from red to green) shows differential expression of miRNAs, where shades of red indicate higher expression and shades of green indicate lower expression levels.

Target prediction and functional annotation

Potential target genes of the identified miRNAs were predicted using the psRNATarget server [25]. The analysis applied standard parameters, including an expectation score < 3.0, complementarity length < 20 nt, flanking regions of 17 bp upstream and 13 bp downstream for accessibility, mismatch allowance at positions 9–11 for translational inhibition, and UPE < 25. Mature miRNA sequences were used as queries against the mungbean genome.

Functional and pathway analysis of target genes

GO annotations for the predicted target genes of differentially expressed miRNAs were retrieved from the ShinyGO 0.82 (https://bioinformatics.sdstate.edu/go82/). GO terms were categorized into biological process, molecular function, and cellular component KEGG pathway enrichment was performed using gprofiler (https://biit.cs.ut.ee/gprofiler/gost).Pathways and Gene Ontology (GO) categories with p ≤ 0.05 were considered significantly enriched.

qRT-PCR analysis of drought stress responsive genes, potential drought-related miRNAs and their targets

To evaluate the effect of drought stress across seven mungbean genotypes cv. ML-267, SML-668, PUSA-105, PDM-139, K-851, PUSA-Ratna, and MUM2, a quantitative reverse transcription PCR (qRT-PCR) was conducted under both normal and drought-stressed conditions. Root tissues were collected from control and drought-stressed plants after 30 days of drought imposition at the maturity stage. Total RNA was extracted as described earlier and used to synthesize first-strand cDNA using the cDNA Synthesis Kit (Helix Biosciences, India) following the manufacturer’s protocol [19].

Real-time quantitative PCR (qRT-PCR) was performed to assess the expression of drought stress responsive genes among seven mungbean genotypes and further also to validate the expression profiles of mature miRNAs identified through high-throughput sequencing and bioinformatic prediction in drought-stressed and control mungbean samples following the protocol described previously [19]. qRT-PCR reactions were carried out using SYBR Premix Ex Taq II (Helix Biosciences, India) on an ABI 3000 Real-Time PCR System (Applied Biosystems, USA). Each reaction was run for 40 amplification cycles under the following conditions: initial denaturation at 95 °C for 5 s, annealing at 60 °C for 30 s, and extension at 72 °C for 30 s. For target gene validation, PCR conditions included a primary denaturation at 94 °C for 2 min, followed by 40 cycles of denaturation at 94 °C for 30 s, annealing at 60 °C for 30 s, and extension at 72 °C for 30 s. All reactions were performed with three biological and two technical replicates, using β-tubulin as an internal reference gene. Relative expression levels of miRNAs and their target genes were calculated using the ΔΔCt method, and fold-change values were determined by comparing drought-stressed and control samples. Statistical analysis of gene expression data was conducted using one-way ANOVA in SPSS v13.0 (SPSS Inc., Chicago, USA), and fold-change expression patterns were visualized as bar graphs.

Data availability

The Illumina sequencing reads of all the four samples (drought tolerant cv. ML-267 and drought sensitive cv. PDM-139 both under control and stress conditions) have been submitted to NCBI BIO PROJECT (PRJNA1345533) accession number SRP632673. The BIOSAMPLE accession number for cv. ML-267 in triplicates under untreated control condition is SRR35796975, SRR35796974, SRR35796971, for sample cv. ML-267 (drought stress) in triplicates is SRR35796970, SRR35796969, SRR35796968, for sample cv. PDM-139 under untreated control condition in triplicates is SRR35796967, SRR35796966, SRR35796965, and for sample cv. PDM-139 (drought stress) in triplicates is SRR35796964, SRR35796973, SRR35796972 respectively.

Results

Analysis of plant physiological parameters under drought stress

The measurement of root length, shoot length, no. of nodules, fresh weight and dry weight were recorded under 30-d drought stress and normal conditions among all seven genotypes at maturity stage (Fig. 1a-g). We observed 0.2–0.6-fold increase in shoot length as ML267 > PUSA-105 > PUSA-RATNA > K-851 > SML-668 > MUM-2 > PDM-139, the highest 0.6-fold were recorded in ML267 in comparison to PDM-139. Similarly, 0.3–1.4-fold increase in root length as ML267 > PUSA-105 > PUSA-RATNA > K-851 > SML-668 > MUM-2 > PDM-139, the highest 1.4-fold were observed in ML-267 when compared to PDM-139 under 30-d drought stress at maturity stage (Fig. 2a-b). We observed maximum nodule formation in cv. ML267 while least in cv. PDM139 under drought stress ML267 > PUSA-105 > PUSA-RATNA > K-851 > SML-668 > MUM2>PDM139 (Fig. 2c).

Fig. 1.

a-g Phenotypic response and root morphology of mungbean genotypes under drought stress. Representative images showing whole-plant phenotypes and root architecture of mungbean genotypes (cv. PUSA-105, cv. PUSA-RATNA, cv. SML-668, cv. ML-267, cv. K-851, cv. PDM-139, and cv. MUM-2) grown under control and drought stress conditions for 30 days. Scale bar = 10 cm

Fig. 2.

Effects of drought stress on physiological parameters of mungbean genotypes. Quantitative assessment of plant physiological traits in seven mungbean genotypes (cv. PUSA-RATNA, PUSA-105, ML-267, SML-668, K-851, MUM-2, and PDM-139) after 30 days of drought stress. Parameters measured include (a) root length, (b) shoot length, (c) nodule number, and (d) relative water content (RWC). Values represent means ± standard error (SE) from three independent biological replicates, with at least five plants analyzed per replicate. Asterisks (*) indicate significant differences compared to the control (P < 0.05)

Compared with the control, the root water content of all seven mungbean genotypes (ML-267 > PUSA-105, PUSA RATNA, K-851, SML-668, MUM-2 and PDM-139) decreased after 30 days of drought stress. Among these, the highest root water content was recorded in cv. ML-267, while cv. PDM-139 exhibited the lowest value at the maturity stage. These observations indicate that, under 30-day drought stress, cv. ML267 maintained significantly higher root water content compared with the other genotypes (Fig. 2d). Our results indicated that among seven genotypes of mungbean screened for drought tolerance, cv. ML-267 retained significantly higher water content than PUSA-105, PUSA RATNA, K-851, SML-668, MUM-2 and PDM-139, in the root tissues under 30-d drought stress. We also measured various agronomic parameters of all seven genotypes under 30-d drought stress and normal conditions. We observed increase in plant height, number of lateral branches, no of flowers, no of pods and number of seeds in ML-267 in compare to PUSA-105, PUSA RATNA, K-851, SML-668, MUM-2 and PDM-139, which suggest better drought tolerance capacity of ML-267 in comparison to PDM-139 (Supplementary Table-1).

Expression analysis of stress responsive genes regulated in drought stress

We checked the transcript level of stress responsive genes such as Chitin synthase (CHS), Map kinase 3 (MAPK-3), Heat shoch protein 70 (HSP70), Transcription Factor WRKY53, and proline (P5CS) in all seven mungbean genotypes under 30-d drought stress. We observed enhanced gene expression ranging from 0.2- 4.2-fold higher in cv. ML-267 and least in cv. PDM-139 in qRT-PCR analysis (Fig. 3a-e). We also estimated the expression of ROS dependent enzymes such as Catalase (CAT), Glutathione Peroxidase (GPX), Ascorbate peroxidase (APX), Superoxide dismutase (SOD) in all seven mungbean genotypes upon exposure of 30 days drought stress. We observed similar trend in gene expression of CAT, SOD, APX, and GPX in cv. ML-267 followed by PUSA-105, PUSA RATNA, K-851, SML-668, MUM-2 and least in PDM-139 genotypes (Fig. 3f-i).

Fig. 3.

a-i Expression analysis of drought-responsive genes in mungbean under drought stress. Quantitative Real-Time PCR (qRT-PCR) analysis of selected drought-responsive genes in mungbean plants subjected to 30 days of drought stress. The genes analyzed include chitin synthase (CHS), mitogen-activated protein kinase 3 (MAPK3), heat shock protein 70 (HSP70), transcription factor WRKY53, pyrroline-5-carboxylate synthetase (P5CS), and antioxidant enzymes catalase (CAT), superoxide dismutase (SOD), ascorbate peroxidase (APX), and glutathione peroxidase (GPX). The Vr-β-tubulin gene was used as an internal control. Expression levels are presented as mean relative fold changes from three independent biological replicates, and error bars indicate standard deviation (SD) (n = 3)

Small RNA sequencing and identification of microRNAs

Small RNA sequencing was used to construct four libraries (NCBI Bio Project: PRJNA1345533; SRA accessions in triplicates SRR35796975, SRR35796974, SRR35796971, SRR35796970, SRR35796969, SRR35796968, SRR35796967, SRR35796966, SRR35796965, SRR35796964, SRR35796973, SRR35796972), representing control and drought-stress conditions for both the tolerant and sensitive mungbean genotypes (Table 1). After trimming adapters and removing low-quality sequences, 240.12 million high-quality reads were retained for further analysis. Reads shorter than 18 nt or longer than 24 nt were subsequently filtered out to obtain a refined set of unique reads for miRNA prediction. Among the retained reads, 24-nt small RNAs were the most abundant (22.05%), followed by 23-nt (11.21%), 21-nt (9.13%), and 18-nt (6.24%) classes (Supplementary Fig. 1). From these processed reads, 266 miRNAs belonging to 51 conserved miRNA families were identified. Most of these miRNAs were present in all four libraries, with miR156 (10 members) being the largest family, followed by miR165 (9 members), miR159 (7 members), and miR167, miR394, and miR396 (each with 4 members). Several other families, including miR162, miR390, miR399, miR408, and miR3307, contained two members, while many families such as miR166, miR168, miR482, miR403, miR398, and miR171 were represented by a single member each. Ab initio prediction approaches additionally identified 61 novel putative miRNAs, bringing the total to 274 miRNAs across all libraries. Notably, most conserved and novel miRNAs showed a strong preference for uridine at the 5′-terminal position. Genomic mapping against the mungbean reference genome revealed that several of these miRNAs occurred in clustered arrangements within specific genomic regions.

Table 1.

Small RNAs from mungbean roots

Mungbean variety and Replicates	Total reads	Unique reads	Filtered Read %	GC%	No. of identified microRNAs
ML-267 C1	19,281,915	18,705,378	97.01	55	55
ML-267 C2	20,597,261	19,966,909	96.94	54	62
ML-267 C3	18,385,817	17,882,222	97.26	54	52
ML-267-DS1	18,741,937	18,185,654	97.03	55	40
ML-267-DS2	24,213,685	23,532,252	97.19	55	53
ML-267-DS3	18,519,981	17,986,117	97.12	55	57
PDM-139-C1	21,928,278	21,322,288	97.24	54	69
PDM-139-C2	20,875,762	20,302,362	97.25	55	62
PDM-139-C3	19,438,677	18,891,716	97.19	55	64
PDM-139-DS1	31,385,044	30,516,412	97.23	53	78
PDM-139-DS2	30,542,683	29,706,627	97.26	54	75
PDM-139-DS3	25,949,125	25,277,161	97.41	53	72

High-throughput miRNA expression profiling

To assess the regulatory involvement of miRNAs in drought adaptation, expression patterns from the four small RNA libraries were compared, and a heatmap was constructed to visualize the differential expression trends (Fig. 4, Supplementary Fig. 2). A total of 86 miRNAs showed significant expression changes across the treatment groups, including 53 in ML267-DS vs. ML267-C, 42 in PDM139-DS vs. PDM139-C, 38 in ML267-DS vs. PDM139-DS, and 24 in ML267-C vs. PDM139-C.

Fig. 4.

Differential expression profiling of miRNAs in drought-tolerant cv. ML267 and drought-susceptible cv. PDM139. Comparative analysis showing the expression patterns of identified miRNAs between the drought-tolerant genotype ML267 and the drought-susceptible genotype PDM139 under drought stress conditions. Differentially expressed miRNAs exhibiting up- or down-regulation between the two genotypes are represented

In the drought-tolerant genotype ML267, drought stress induced the upregulation of three conserved miRNAs, namely miR167 (1.2-fold), miR156 (0.8-fold), and miR160 (0.07-fold). In contrast, several conserved miRNAs were downregulated under drought conditions compared with the control, including miR5770 (− 4.7-fold), miR166 (− 2.9-fold), miR159 (− 2.4-fold), miR165 (− 2.1-fold), miR162 (− 2.1-fold), miR482 (− 2.1-fold), miR394 (− 1.4-fold), miR396 (− 1.2-fold), and miR168 (− 0.8-fold). Similarly, all identified novel miRNAs exhibited reduced expression in response to drought stress. These included nov-miR6632 (− 2.46-fold), nov-miR2591 (− 2.46-fold), nov-miR6635 (− 1.54-fold), nov-miR6631 (− 1.12-fold), nov-miR6637 (− 0.92-fold), nov-miR6630 (− 0.52-fold), nov-miR6636 (− 0.30-fold), and nov-miR3307 (− 0.18-fold). Overall, the drought-tolerant genotype ML267 displayed a distinct miRNA expression profile characterized by the upregulation of a limited number of conserved miRNAs and the downregulation of several conserved and novel miRNAs under drought stress. In contrast, the drought-sensitive genotype PDM139 displayed a different expression landscape. Significant upregulation was observed for several conserved miRNAs, including miR396 (5.7-fold), miR156 (4.6-fold), miR159 (3.4-fold), miR162 (2.8-fold), miR394 (1.2-fold) and miR390 (0.5-fold). Conversely, miR482 (− 2.8-fold), miR408 (− 2.6-fold), miR398 (− 2.0-fold), miR164 (− 1.6-fold), miR160 (− 1.8-fold), miR167 (− 0.8-fold), and miR399 (− 0.5-fold) exhibited reduced expression following drought treatment. Among the nine novel miRNAs detected, only nov-miR6632 (− 0.81-fold) and nov-miR6635 (− 1.51-fold) were downregulated, while seven novel miRNAs—nov-miR3307a (4.17-fold), nov-miR6636 (1.22-fold), nov-miR3307b (4.05-fold), nov-miR2591 (0.01-fold), nov-miR6631 (0.63-fold), nov-miR6630 (1.92-fold), and nov-miR6637 (2.12-fold)—showed increased expression in PDM139 under drought stress.

Prediction of miRNA target genes and their categorization

A computational analysis was conducted to identify possible target genes for both conserved and novel miRNAs. Out of the 335 miRNAs discovered via sequencing (274 known and 61 novel), there were 704 potential target mRNAs forecasted using the psRNATarget tool. Single miRNAs were linked to anywhere from one to twelve anticipated targets. The majority of miRNA–mRNA interactions indicated regulatory impacts via transcript cleavage, which is the main mechanism of action in plants (Supplementary Table 2; unannotated targets not included). Given that plant miRNA binding sites usually exist within open reading frames (ORFs), the examination concentrated on coding sequence areas and identified targets for 34 conserved miRNA families, accounting for 704 predicted mRNAs found in mungbean roots. Interestingly, several conserved families, such as miR168, miR170, miR172, miR396, miR482, miR1507, miR5770, and miR6630, exhibited no anticipated targets in this dataset.

Small RNA sequencing also indicated that 37 of the novel miRNAs each possessed over 100 potential targets, excluding those lacking functional annotations. Many conserved miRNAs were predicted to regulate transcription factors homologous to well-characterized regulatory proteins in Arabidopsis and other crops (Supplementary Table 3). For instance, dicot-specific families such as miR160, miR164, miR390, miR394 and miR2118 were predicted to target key regulators including ARF10/16/17/18, NAC-domain transcription factors, LRR receptor-like serine/threonine kinase RCH1, F-box domain-containing proteins, and TIR-NBS-LRR resistance gene analogs.

Similarly, several novel miRNAs (e.g., nov-miR6632, nov-miR3307, nov-miR6636, nov-miR6635, nov-miR2591, nov-miR6631, nov-miR6630 and nov-miR6637) were associated with targets involved in diverse molecular and signaling functions. These included genes encoding GDSL esterase/lipase, calcium-binding protein CML23/ F-box protein, protein phosphatase 2 C, serine hydroxymethyl transferase, L-type lectin receptor kinases, MAN1/receptor-like kinase At5g47070, glycosyltransferases, MYB-related transcription factors, and transporters such as nicotianamine transporter YSL7 or thermospermine synthase ACAULIS5-like proteins. Collectively, these predictions support a wide regulatory role for both conserved and novel miRNAs in coordinating metabolic and defense-related responses under drought stress in mungbean.

This pie chart illustrates the functional distribution of protein targets in plants, revealing how cellular proteins are allocated across different biological processes: Signal Transduction & Receptor Kinases (22%) - The largest category, reflecting plants’ need to respond to environmental stimuli like light, hormones, pathogens, and stress. This encompasses receptor-like kinases and different signaling pathways crucial for plant growth and adjustment. Uncharacterized Proteins (15%) - A considerable segment stays functionally unidentified, underscoring continuing research prospects in plant molecular biology. Enzymes and Metabolism (13%) - Essential metabolic enzymes facilitating primary and secondary metabolic processes, such as photosynthesis, respiration, and the biosynthesis of specialized compounds. Transcription Factors & Gene Regulation (12%) - Essential for managing gene expression in development, responses to stress, and adaptation to the environment. Transporters and Channels (11%) - Crucial for nutrient absorption, ion balance, and the distribution of metabolites across cell compartments and throughout the plant. Protein Modification & Ubiquitination (10%) - mechanisms of post-translational regulation that manage protein activity, stability, and breakdown. The other categories consist of RNA processing (9%), stress response and defense (6%), and structural proteins (2%), all of which fulfill distinct yet crucial roles in the functioning and survival of plant cells. This distribution showcases the distinct challenges faced by plants as sessile organisms that need complex sensing, signaling, and adaptation mechanisms.

Functional significance of putative target genes

The anticipated target genes of both conserved and novel miRNAs were identified as being involved in multiple essential biological processes related to drought tolerance in mungbean. GO enrichment analysis indicated that most of these targets participated in cellular components, followed by molecular functions and biological processes (Fig. 5). Numerous targets encoded proteins involved in nucleotide binding, ribonuclease activity, carbohydrate derivative binding, and other crucial biochemical functions, highlighting their importance in controlling gene expression at various levels. Within the biological process category, various targets were linked to defense mechanisms, regulation of cellular and physiological processes, and auxin-responsive signaling pathways, highlighting their role in adaptive responses during stress (Fig. 5). KEGG pathway analysis revealed notable enrichment of 64 genes in pathways associated with sugar metabolism, ribosome biogenesis, secondary metabolite biosynthesis, amino acid metabolism, pantothenate and CoA production, and selenocompound metabolism (Supplementary Fig. 3). Collectively, these findings highlight that the miRNA-regulated target genes are functionally diverse and play central roles in metabolic adjustment and stress signaling mechanisms that support drought resilience in mungbean.

Fig. 5.

Target gene distribution and Gene Ontology (GO) classification of differentially expressed miRNAs in mungbean under drought stress. a Distribution of predicted target genes of differentially expressed miRNAs identified under 30 days of drought stress. b Gene Ontology (GO) enrichment analysis of drought-responsive miRNA target genes, classified into three main categories: biological process (BP), cellular component (CC), and molecular function (MF). The right x-axis indicates the percentage of genes associated with each GO category within the respective main category

qRT-PCR expression profiling of miRNAs

Expression profiling of seven drought-responsive miRNAs showed clear differential regulation in both ML267 and PDM139 under stress conditions compared to their respective controls (Fig. 6). In ML267, three miRNAs—miR167 (1.32-fold), miR156 (0.93-fold), and miR160 (0.08-fold)—were upregulated, while miR159 (− 2.54-fold), miR394 (− 1.66-fold), miR396 (− 1.23-fold), and nov-miR3307 (− 0.30-fold) showed reduced expression under drought stress. In contrast, PDM139 displayed upregulation of miR396 (5.6-fold), miR394 (1.4-fold), miR156 (4.8-fold), miR159 (3.5-fold), and nov-miR3307 (4.2-fold), whereas miR160 (− 1.9-fold) and miR167 (− 0.86-fold) were downregulated under drought treatment (Fig. 6a).

Fig. 6.

qRT-PCR validation of drought-responsive miRNAs and their corresponding target genes in mungbean roots. a Relative expression profiles of seven selected conserved and novel miRNAs, and (b) expression patterns of their corresponding target genes in root tissues of the drought-tolerant genotype ML267 and the drought-susceptible genotype PDM139 under control and drought stress conditions. The target genes analyzed included auxin response factor (ARF), heat shock protein 70 (HSP70), GAMYB-like MYB transcription factor (MYB), squamosa promoter-binding-like protein 6 (SPL6), F-box protein 6 (F-BOX), probable calcium-binding protein CML23, and aspartic proteinase nepenthesin-1 (APN1). Panel (a) illustrates drought-induced changes in miRNA abundance, whereas panel (b) shows the expression responses of their putative target genes, enabling assessment of potential miRNA-mediated regulatory relationships. Upregulated and downregulated expression patterns are presented relative to their respective control samples. The Vr-β-tubulin gene was used as the internal reference, and relative expression levels were calculated using the comparative Ct (2⁻ΔΔCt) method. Expression levels in control plants (ML267-C and PDM139-C) were normalized to 1. Data represent the mean ± standard error (SE) of three biological replicates, each with three technical replicates. Asterisks (*) indicate statistically significant differences at P ≤ 0.05. ML267-C, drought-tolerant control; ML267-DS, drought-tolerant under drought stress; PDM139-C, drought-susceptible control; PDM139-DS, drought-susceptible under drought stress

Expression profiling of target genes under drought stress

Expression profiling of selected conserved and novel putative miRNAs and their corresponding target genes was performed using qRT-PCR in the root tissues of the drought-tolerant genotype ML-267 and the drought-sensitive genotype PDM-139. Overall, an inverse regulatory relationship was observed between the miRNAs and their target transcripts under drought stress in both ML267-DS and PDM139-DS samples (Fig. 6). In ML267-DS, miR167 showed more than a 1.2-fold increase in expression, while its target gene encoding Aspartic Proteinase Nepenthesin-1 (APN-1) was markedly suppressed (0.11-fold). In contrast, PDM139-DS exhibited a 0.8-fold reduction in miR167 expression, with a corresponding increase in APN-1 transcript levels, displaying an opposite trend to ML-267. Similar patterns were also detected for other miRNA–target pairs. miR160 and its target Auxin Response Factor (ARF), along with miR156 and its target Squamosa Promoter-Binding-Like Protein 13 A (SPL13A), demonstrated consistent inverse expression behavior. Furthermore, in ML267-DS, miR159 was downregulated (–2.4-fold), whereas its target MYB transcription factor was upregulated by 2.4-fold. A similar inverse pattern was observed for miR394 (–1.4- fold), miR396 (–1.2-fold), and nov-miR3307 (–0.1-fold), whose target genes—F-Box only protein-6 (2.3-fold), HSP70 (3.4-fold), and calcium-binding protein-23 (3.2-fold), respectively—displayed strong upregulation under drought conditions. In PDM139-DS, these miRNA–target pairs exhibited reverse trends compared to the tolerant genotype ML-267 (Fig. 6b).

Construction of miRNA–target interaction networks using Cytoscape

Cytoscape was used to create and visualize miRNA–target interaction networks in order to comprehend the regulatory function of differentially expressed miRNAs under the experimental settings. A number of highly connected miRNAs interacting with numerous functional target genes linked to transcriptional regulation, stress signaling, transport activities, protein turnover, and metabolic pathways were found by the network analysis. The miR159, miR156, miR396, miR160, miR394, miR3307a, and miRN6630 were the notable regulatory modules found in the network.

The interaction maps demonstrated that a single miRNA family could control several target transcripts at once, indicating coordinated post-transcriptional regulation during physiological modulation and stress adaptation (Fig. 7).

Fig. 7.

Cytoscape-based computationally predicted miRNA–target interaction network illustrating major drought stress-responsive regulatory modules in mungbean. The network was constructed using Cytoscape based on bioinformatically predicted miRNA–target gene interactions. Key regulatory miRNA families include Vra-miR159, Vra-miR396, Vra-miR160, Vra-miR156, Vra-miR394, Vra-miR3307a, and Vra-miR6630, along with their putative target genes. Prominent predicted targets include MYB transcription factors, GRFs (Growth-Regulating Factors), ARFs (Auxin Response Factors), SPL proteins (SQUAMOSA promoter-binding-like proteins), F-box proteins, CML23 (calcium-binding protein-like 23), and 5-methyltetrahydropteroyltriglutamate–homocysteine methyltransferase. These predicted targets are associated with transcriptional regulation, hormone signaling, calcium signaling, protein degradation, and metabolic adaptation pathways. The network represents in silico predicted regulatory relationships and does not include experimentally validated miRNA–target interactions

miRNA–target interaction networks underlying drought stress responses in mungbean

We explored the relationships between mungbean microRNAs (miRNAs) and their target genes to clarify their functional roles in response to drought stress. Our investigation showed that various drought-responsive miRNAs are involved in essential regulatory pathways that enhance drought tolerance, such as hormone signaling, transcriptional control, reactive oxygen species (ROS) detoxification, and iron homeostasis. In particular, miR393, miR160, and miR156 were identified as key players in controlling plant growth and responses mediated by hormones. miR396, miR6630, miR172, and miR159 were associated with sustaining iron homeostasis and transport, thus aiding cellular metabolic equilibrium during stress scenarios. Moreover, miR482, miR1507, and miR396 were linked to stress responses and defense processes, whereas miR159, miR164, and miR165 played roles in ROS removal and antioxidant enzyme activity. Additionally, miR162, miR515, and miR6632 played a role in miRNA production and RNA splicing activities, suggesting a complex regulatory network influencing gene expression during drought stress (Fig. 8). Together, these results emphasize the collaborative role of miRNAs in managing physiological and molecular changes that improve drought resistance in mungbean.

Fig. 8.

microRNA-mediated regulatory network involved in drought and salt stress responses in mungbean; Schematic representation of key mungbean microRNAs (miRNAs) implicated in the regulation of drought and salinity stress responses. Conserved and novel miRNAs identified in this study are shown along with their predicted or validated target genes involved in stress perception, signal transduction, transcriptional regulation, osmotic adjustment, antioxidant defense, and hormonal signaling pathways. Differential expression patterns of miRNAs under drought and salt stress conditions are indicated by distinct color gradients, representing up-regulation and down-regulation relative to control plants. The network highlights miRNA–target interactions that contribute to stress adaptation by modulating physiological and molecular processes essential for mungbean stress tolerance

Discussion

Drought represents a significant among abiotic stress limitation that hinders mungbean yield, especially in rainfed farming systems [26]. Understanding the regulatory mechanisms that control drought adaptation is crucial for developing stress-resilient crop varieties. In this scenario, microRNAs (miRNAs) have surfaced as essential elements of epigenetic control, adjusting gene expression after transcription and allowing plants to precisely calibrate their reactions to environmental pressures. Mungbean genotypes show varied reactions to drought based on their genetic makeup, growth phase, and intensity of stress. At the molecular level, drought triggers significant transcriptional changes, activating genes related to stress that improve resilience to water scarcity [27]. Nevertheless, drought tolerance continues to be a complicated characteristic influenced by various interacting pathways and regulatory networks [28, 29].

This study systematically assessed seven mungbean genotypes under drought stress by utilizing a combination of physiological, biochemical, molecular, and agronomic parameters to identify tolerant and sensitive lines. Important physiological characteristics, such as root length, shoot length, relative water content (RWC), and nodule count, acted as markers for water uptake efficiency and the preservation of tissue hydration (Figs. 1 and 2). The ML267 genotype demonstrated longer roots, increased RWC, and better nodulation relative to other genotypes, while PDM139 displayed the lowest values. These findings indicate that ML267 exhibits enhanced adaptive ability in drought conditions, aligning with previous research emphasizing the significance of root structure and cellular water preservation in drought resilience [30, 31].

Molecular profiling additionally uncovered transcriptional responses linked to stress adaptation that vary by genotype. Genes encoding antioxidant enzymes like SOD, CAT, GPX, and APX were more significantly upregulated in ML267 compared to PDM139, reflecting increased scavenging of reactive oxygen species (ROS) and greater protection from oxidative damage [32]. Moreover, stress-responsive genes such as CHS, HSP70, MAPK3, and WRKY53 showed increased expression in ML267, indicating the activation of several protective signaling pathways during drought stress (Fig. 3a-d). These results align with earlier studies showing that coordinated transcriptional responses play a key role in providing drought tolerance in legumes [19].

Drought stress negatively impacted agronomic traits such as plant height, biomass, pod count, seed count, and yield per plant, but ML267 showed relatively smaller decreases (Supplementary Table 1). This reinforces the well-known connection between physiological resilience and yield consistency in water-scarce environments [33]. Integrating physiological performance and molecular responses creates a strong foundation for selecting drought-tolerant genotypes.

small RNA transcriptome profile of root tissues from different genotypes revealed greater understanding of miRNA-mediated regulatory processes. Roots are crucial for detecting soil moisture shortages and triggering adaptive actions [7]. High-throughput sequencing discovered both conserved and novel drought-responsive miRNAs, many of which showed genotype specific expression patterns (Figs. 5 and 6). Conserved miRNA families such as miR156, miR159, miR160, miR167, miR394, and miR396 exhibited distinct regulatory patterns between ML267 and PDM139. The upregulation of miR167, miR156, and miR160 in ML267 indicate improved control of auxin-related developmental pathways and preservation of root structure during drought stress [18]. Conversely, the differential modulation of miR396, miR159, and miR394 in PDM139 might indicate stress-related growth suppression and modified signaling pathways linked to drought sensitivity. These results emphasize the importance of conserved miRNAs as essential regulators of gene expression in response to stress. Comparable functions of miR396 in controlling GRF transcription factors and growth mechanisms [34–37], as well as of miR159 and miR167 in hormone signaling and MYB/ARF mediated response [38–40] have been reported across multiple plant species, indicating evolutionary conservation of these regulatory modules.

The identification of novel miRNAs enhances our comprehension of species-specific regulatory processes. Multiple novel miRNAs, such as nov-miR6632, nov-miR3307, nov-miR6636, nov-miR6635, nov-miR2591, nov-miR6631, nov-miR6630, and nov-miR6637, exhibited differing expression patterns in tolerant versus sensitive genotypes. They anticipate targeting genes related to signaling, transcriptional control, and metabolic functions, including GDSL esterase/lipase, calcium-binding protein CML23, F-box proteins, protein phosphatase 2 C, and MYB transcription factors (Supplementary Tables 2 and 3). These results indicate that novel miRNAs might play a role in refining stress responses by regulating various biological pathways. Significantly, nov-miR6635, anticipated to target serine hydroxymethyl transferase, exhibited reduced expression under drought stress in both genotypes, aligning with earlier reports connecting metabolic changes to stress adaptation [41, 42]. Nevertheless, functional validation is necessary to verify these regulatory interactions. The identified target genes are computationally predicted targets and their regulatory relationships require additional experimental confirmation, such as transcriptome sequencing, degradome analysis, or expression-based validation approaches.

Functional enrichment and pathway analyses indicated that miRNA targets participate in hormone signaling, transcriptional regulation, metabolic adaptation, and ROS detoxification (Fig. 5). KEGG pathway analysis further emphasized the role of MAPK signaling and carbohydrate metabolism, both essential for drought adaptation [43–45]. The contrasting enrichment patterns observed between ML267 and PDM139 suggest the presence of distinct but coordinated regulatory networks underlying drought tolerance.

qRT-PCR validation verified the expression patterns of selected miRNAs and their targets, supporting the reliability of sequencing data (Fig. 6). The inverse relationship between miRNAs and their targets highlights their function in post-transcriptional gene control. These miRNAs might act as possible biomarkers for drought tolerance and targets for genetic improvement.

Cytoscape-driven interaction analysis uncovered complex miRNA–target networks, in which specific miRNAs control several genes associated with transcriptional regulation, hormone signaling, protein degradation, and metabolic activities (Fig. 7). These interconnected networks imply that miRNAs serve as central regulatory hubs orchestrating multiple stress-responsive pathways (Fig. 8). This coordinated regulation illustrates the wider epigenetic landscape, where small RNAs interact with transcriptional and chromatin-level mechanisms to precisely fine-tune gene expression during stress situations.

In summary, our results reveal that drought resistance in mungbean is regulated by a complex miRNA-mediated regulatory network connecting hormone signaling, redox homeostasis, and stress defense pathways. The contrasting responses observed between ML267 and PDM139 emphasize the significance of coordinated post-transcriptional regulation in influencing drought adaptation. Combining physiological screening with small RNA profiling offers a thorough framework for comprehending regulatory mechanisms responsive to drought adaptation. These findings offer valuable targets for forthcoming functional validation and molecular breeding approaches, focused on developing drought-resilient mungbean cultivars.

Conclusion

This study offers a comprehensive insight into drought tolerance in mungbean through the integration of physiological, molecular, and small RNA transcriptomic studies. Among the assessed genotypes, ML267 showed greater drought tolerance, marked by superior physiological performance, boosted antioxidant capacity, and more reliable yield stability in comparison to the sensitive genotype PDM139. Small RNA profiling uncovered a variety of conserved and novel drought-responsive microRNAs (miRNAs) exhibiting genotype-specific expression patterns and influencing essential pathways associated with hormone signaling, stress response, ROS detoxification, and metabolic adaptation. Functional enrichment and network analyses highlighted the involvement of coordinated regulatory pathways, such as MAPK signaling and carbohydrate metabolism, that contribute to drought adaptation. Importantly, the inverse expression patterns of miRNAs and their targets underscore the role of miRNA-mediated post-transcriptional regulation as a critical component of the epigenetic landscape governing stress responses. The identified candidate miRNAs and their targets provide valuable molecular resources for future functional validation and offer promising targets for marker-assisted breeding and genome-editing approaches aimed at developing drought-resilient mungbean cultivars.

Authors’ contributions

SK: Conceived project, designed and performed experiments, written original manuscript, PK: Molecular and phenotypic analysis of mungbean genotypes, KG: Transcriptomic data analysis, BOO: Molecular analysis, NSM: Manuscript Review and editing. All authors contributed to the article and approved the submitted version.

Funding

The authors are thankful for the financial support from the Department of Biotechnology, DBT- M. K. BHAN Young Researcher Fellowship program (161016/1/2024-HRD-DBT) Govt. of India, for this study.

Data availability

The datasets generated and/or analyzed during the current study are available in the NCBI BioProject database under the accession number PRJNA1345533 and in the Sequence Read Archive (SRA) under the accession number SRP632673. The original contributions presented in this study are included in the manuscript and supplementary materials. Further inquiries regarding the data can be directed to the corresponding author(s).

Declarations

Ethics approval and consent to participate

This research did not contain any studies on human and/or animal participants performed by any of the authors.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.