EST-SSR based genetic polymorphism among Lablab (Lablab purpureus L. Sweet) accessions contrasting for drought stress at seedling stage Authors: Julius S. Missanga, Patrick A. Ndakidemi, Pavithravani B. Venkataramana Published: 2026 ABSTRACT Lablab is a multipurpose and the most drought-tolerant (DT) crop compared with its relatives. Despite its potential, Lablab is still an underutilized crop with a lack of improved varieties in many countries. The DT (D349, D147, HA4, D363, D352, D359, D348, D311, D55 and D250) and drought-susceptible (DS) (D271, D66, D106, D6, D26, D255, D28, D186, D95, and D258) accessions were earlier identified according to their morphological and biochemical responses to moisture stress at the seedling stage. These accessions were used to establish genetic polymorphism among the accessions contrasting for drought stress based on the Expressed Sequence Tag-Simple Sequence Repeats (EST-SSR) markers. The CTAB protocol was employed for the genomic DNA extraction. After DNA quality and quantity verification, the PCR was conducted using 16 EST-SSR primer pairs specific to the Lablab. The products were separated through the horizontal polyacrylamide gel electrophoresis (hPAGE). Discriminating ability of the markers and primers’ efficiency were evaluated based on various genetic parameters. Principal Coordinate Analysis (PCoA) was performed to estimate the distance matrix among the population and among the accessions. While cluster analysis was processed to trace the genetic relationship among the accessions, dendrogram was constructed to decipher their genetic relationship. Analysis of Molecular Variance (AMOVA) was finally computed to quantify the diversity level and genetic relationship among the population, and among the accessions. A low polymorphism (GD = 0.19) was observed between the DT and DS accessions, likely due to limited discriminatory power of the EST-SSR markers. However, the PCoA, cluster analysis and AMOVA identified DT (D147, HA4, and D349) and DS (D106, D95, and D271) accessions as strongly contrasting populations under drought stress, with D147, HA4, D349, D363, D359, D352, and D348 further recommended as DT accessions. Given the low polymorphism observed, further validation using more informative molecular markers and advanced genomic approaches is recommended to improve the identification of drought-tolerance genes and related QTLs to support Lablab breeding programs.Supplementary InformationThe online version contains supplementary material available at https://doi.org/10.1038/s41598-026-58070-4. Introduction Drought is the main climatic factor that significantly limits crop production, especially in tropical and subtropical regions1. The basis for drought tolerance in crops involves different morphological, biochemical, and molecular mechanisms2. Lablab (Lablab purpureus L. Sweet) is a Fabaceae legume with chromosome counts of 2n = 20, 22, or 243. Compared with its close relatives, Lablab exhibits greater drought tolerance and provides numerous benefits4,5. The crop is a multipurpose legume utilized in diverse production systems, including its use as a pulse, vegetable, fodder, green manure, and intercrop6. It plays a key role in conservation agriculture, soil fertility improvement, sustainable farming practices, and biodiversity conservation7,8. Lablab serves as a valuable source of nutritious food for both humans and animals. The nutritional profile of Lablab includes substantial amounts of protein (30 g), dietary fiber (25.6 g), carbohydrates (60.74 g), and minerals like Zn (9.3 mg), P (0.372 mg), and Fe (5.1 mg)9. It contains a low fat content of 1.69 g10. Lablab contributes significantly to both health benefits11 and the socio-economic well-being of farmers, especially in rural communities12. Despite its multitude of advantages, Lablab is still largely underutilized and primarily grown by smallholder farmers, especially in African countries13. Many countries still do not have access to the improved drought-tolerant (DT) varieties. Given the projected rise in drought conditions from climate change, analyzing molecular marker-based polymorphisms in the Lablab accessions with contrasting drought responses is critical for breeding high-performing varieties. According to Popoola et al.14 and Missanga et al.15, Lablab employs three mechanisms to defend itself against drought stress. The morphological mechanisms underlying drought tolerance in Lablab include changes in leaf size and inclinations, glabrous and trailing stems, and enhanced shoot growth to reduce radiation load,variation in chlorophyll content, and stomatal regulation to limit water loss; as well as extensive root systems with deep penetration and abundant lateral and adventitious roots to enhance water absorption, microbial interactions and nutrients circulations16. Lablab defends itself against reactive chemical species such as reactive oxygen species (ROS), sulfur species (RSS) and nitrogen species (RNS) using both enzymatic and non-enzymatic biochemical pathways17. Peroxidase (POX), catalase (CAT), and polyphenol oxidase are some of the key antioxidant enzymes involved in the detoxification of these chemical species. The accumulation of non-enzymatic-secondary metabolites such as sugars, phenols, proline, organic solutes, and flavanols contributes also to the protection of cellular structures against reactive species during drought stress18. However, integration of the reactive species signaling is a central mechanism by which some legumes perceive and respond to environmental stresses such as drought. Rather than being purely harmful byproducts, reactive species act as key signaling molecules when tightly regulated. They act as a double-edged sword, harmful at high levels but essential for signaling and stress tolerance at controlled levels19,20. The molecular response of Lablab to drought stress involves expression regulations of different classes of the DT genes and their transcription factors, including osmo-protectant biosynthesis genes, abscisic acid (ABA) responsive genes, antioxidant defense genes, reactive species detoxification genes, heat shock proteins (HSPs), late embryogenesis abundant (LEA) proteins e.g. dehydrins, and water transport proteins e.g. aquaporins (Yao et al., 2013; Wang et al., 2018). The phenomena of gene expressions to control drought stress in some crops were formerly studied through molecular markers. However, the development of marker technology has been slower in Lablab, and some other legumes than in cereal crops to the point of considering them as “orphan crops”21, Dhaliwal et al. (2020). In their steps of advancement, the first group of markers; amplified fragment length polymorphism (AFLP), restriction fragment length polymorphism (RFLP), and random amplification of polymorphic DNA (RAPD) were mainly employed to understand the diversity and genetic characterization of studied crops. The second group namely sequence-based markers including single nucleotide polymorphism (SNPs), microsatellite or simple sequence repeats (SSRs), and expressed sequence tags (EST) were mostly utilized in mapping studies22,23. Despite numerous applications of molecular markers in different legume species, their deployment in enhancing drought tolerance in Lablab has been minimal. Since drought begins to affect crops during the seedling stage, early-stage tolerance to water stress improves their chances of progressing to flowering and harvest24,25. In this regards, DT and DS accessions were earlier identified from a large-scale screening of Lablab accessions for seedling-stage drought tolerance using physiological indicators of drought stress. However, the lack of detailed knowledge about molecular responses to drought at this stage constrained the confirmation of DT and DS accessions. Therefore, this Lablab population was employed to determine the genetic polymorphism among the accessions contrasting for drought stress using EST-SSR markers, to support drought breeding programs. Methods Study site This study was carried out at the Nelson Mandela African Institution of Science and Technology (NM-AIST) laboratory, Arusha, Tanzania found at Latitude 3° 24′ 7″ S and Longitude 36° 47′ 51″ E at an elevation of 1208 m above sea level (masl). The region experiences tropical to subtropical climatic conditions influenced by highlands, with mean annual rainfall of approximately 870 mm and temperatures ranging from 22–28 °C (maximum) to 12–15 °C (minimum)26,27. Plant materials and the sources of seeds The plant materials used in the present study comprised of 20 Lablab accessions (Table 1); 10 as the DT accessions including D349, D147, HA4, D363, D352, D359, D348, D311, D55 and D250 as well as the drought susceptible (DS) accessions including D271, D66, D106, D6, D26, D255, D28, D186, D95, and D258. These accessions were selected following the screening of approximately 300 Lablab accessions at the seedling stage, where drought tolerance among them was evaluated using qualitative and quantitative physiological responses to drought stress. During this experiment, 10 best accessions performing well under water stressing and recovery phases were selected as DT accessions while highly susceptible ones performing poorly or drying out entirely during both phases were determined as DS accessions. The DT Lablab accessions D363 (NM-D 19) and D348 (NM-D 20) are Lablab commercial varieties in Tanzania that were initially introduced from Uganda and Kenya, respectively. The sources of seeds (Table 1) included the following institutions: -(i)The Indian Council of Agricultural Research—National Bureau of Plant Genetic Resources (ICAR-NBPGR) through the Stress Tolerant Orphan Legumes (STOL)—Kirkhouse Trust (KT) supported project: for the accessions HA4, D250, D55, D147, D271, D66, D6, D106, D258, D26, D186, D255, D28, and D9528(ii)The ICAR-NBPGR—University of Eldoret (UoE) in Kenya through the Stress Tolerant Orphan Legumes (STOL)—Kirkhouse Trust (KT) supported project: for the accessions D348, D352, and D349 (www.uoeld.ac.ke/research/wheat-and-dolichos-varieties-research-university-eldoret)(iii)Uganda through Educational Concerns for Hunger Organization (ECHO) East Africa Community: for the accession D3638(iv)The International Livestock Research Institute (ILRI), gene bank in Kenya: for the accession D359 formerly known as ILRI. 14,491.(v)Local landraces collected from the central Tanzania: for the accession D31129.Table 1Lablab accessions used for the molecular marker assessment.S/NAccessionAccession nameSeed colorFlower colorCountry of OriginDrought tolerant (DT) accessions 1D 348Eldoret KT Black-2 (NM-D 20)BlackWhiteKenya 2D 363Karamoja Red (NM-D 19)RedWhiteUganda 3HA4Hebbal Avare 4CreamYellowishIndia 4D 352Eldoret KT CreamCreamPinkKenya 5D 349Eldoret KT MaridadiBlackPurpleKenya 6D 359ILRI. 14,491BrownWhiteKenya 7D 311Kondoa, WhiteWhiteYellowTanzania (Kondoa) 8D 250UnidentifiedBrownish-creamyYellowishEthiopia 9D 55UnidentifiedWhiteCambodia 10D 147UnidentifiedBrownYellowishEthiopiaDrought susceptible (DS) accessions 1D 271UnidentifiedReddishWhiteIndia 2D 66UnidentifiedBrownishWhiteUzbekistan 3D 6UnidentifiedRedUnidentifiedThailand 4D 106UnidentifiedCreamUnidentifiedTanzania 5D 258UnidentifiedReddishUnidentifiedIndia 6D 26UnidentifiedBlackUnidentifiedLao People’s Democratic Rep 7D 186UnidentifiedBrownUnidentifiedJapan 8D 255UnidentifiedWhiteUnidentifiedIndia 9D 28UnidentifiedBlackUnidentifiedBangladesh 10D 95UnidentifiedCreamUnidentifiedTanzania (i)The Indian Council of Agricultural Research—National Bureau of Plant Genetic Resources (ICAR-NBPGR) through the Stress Tolerant Orphan Legumes (STOL)—Kirkhouse Trust (KT) supported project: for the accessions HA4, D250, D55, D147, D271, D66, D6, D106, D258, D26, D186, D255, D28, and D9528 The Indian Council of Agricultural Research—National Bureau of Plant Genetic Resources (ICAR-NBPGR) through the Stress Tolerant Orphan Legumes (STOL)—Kirkhouse Trust (KT) supported project: for the accessions HA4, D250, D55, D147, D271, D66, D6, D106, D258, D26, D186, D255, D28, and D9528 (ii)The ICAR-NBPGR—University of Eldoret (UoE) in Kenya through the Stress Tolerant Orphan Legumes (STOL)—Kirkhouse Trust (KT) supported project: for the accessions D348, D352, and D349 (www.uoeld.ac.ke/research/wheat-and-dolichos-varieties-research-university-eldoret) The ICAR-NBPGR—University of Eldoret (UoE) in Kenya through the Stress Tolerant Orphan Legumes (STOL)—Kirkhouse Trust (KT) supported project: for the accessions D348, D352, and D349 (www.uoeld.ac.ke/research/wheat-and-dolichos-varieties-research-university-eldoret) (iii)Uganda through Educational Concerns for Hunger Organization (ECHO) East Africa Community: for the accession D3638 Uganda through Educational Concerns for Hunger Organization (ECHO) East Africa Community: for the accession D3638 (iv)The International Livestock Research Institute (ILRI), gene bank in Kenya: for the accession D359 formerly known as ILRI. 14,491. The International Livestock Research Institute (ILRI), gene bank in Kenya: for the accession D359 formerly known as ILRI. 14,491. (v)Local landraces collected from the central Tanzania: for the accession D31129. Local landraces collected from the central Tanzania: for the accession D31129. Table 1Lablab accessions used for the molecular marker assessment.S/NAccessionAccession nameSeed colorFlower colorCountry of OriginDrought tolerant (DT) accessions 1D 348Eldoret KT Black-2 (NM-D 20)BlackWhiteKenya 2D 363Karamoja Red (NM-D 19)RedWhiteUganda 3HA4Hebbal Avare 4CreamYellowishIndia 4D 352Eldoret KT CreamCreamPinkKenya 5D 349Eldoret KT MaridadiBlackPurpleKenya 6D 359ILRI. 14,491BrownWhiteKenya 7D 311Kondoa, WhiteWhiteYellowTanzania (Kondoa) 8D 250UnidentifiedBrownish-creamyYellowishEthiopia 9D 55UnidentifiedWhiteCambodia 10D 147UnidentifiedBrownYellowishEthiopiaDrought susceptible (DS) accessions 1D 271UnidentifiedReddishWhiteIndia 2D 66UnidentifiedBrownishWhiteUzbekistan 3D 6UnidentifiedRedUnidentifiedThailand 4D 106UnidentifiedCreamUnidentifiedTanzania 5D 258UnidentifiedReddishUnidentifiedIndia 6D 26UnidentifiedBlackUnidentifiedLao People’s Democratic Rep 7D 186UnidentifiedBrownUnidentifiedJapan 8D 255UnidentifiedWhiteUnidentifiedIndia 9D 28UnidentifiedBlackUnidentifiedBangladesh 10D 95UnidentifiedCreamUnidentifiedTanzania Lablab accessions used for the molecular marker assessment. Extraction of genomic DNA (gDNA) The extraction of gDNA involved grinding pooled leaf samples from three-week-old plants of each accession into a fine powder using a mortar and pestle. The DNA extraction followed the slightly modified cetyltrimethyl ammonium bromide (CTAB) DNA extraction method30,31. Quality and quantity of the DNA The DNA quality and quantity were verified through gel electrophoresis and Nano-Drop spectrophotometry, respectively. The gel electrophoresis for DNA quality determination involved 0.8% (w/v) agarose gel (aMReSCO, Solon, Ohio, USA) in 1X TAE (Tris–acetate EDTA) buffer, run for 40 min at 80 V (v). A gel was stained with 5 µL of Ethidium Bromide (EtBr) (10 mg mL-1) before pouring it in the electrophoresis tank to enhance a good view of DNA under ultraviolet (UV) trans-illuminator. The DNA bands in gels stained with ethidium are fluorescent on exposure to the UV light. Bromophenol blue loading dye was added before electrophoresis. The quality of DNA was confirmed by presence of one big DNA band in each well without any smear (degraded DNA). The DNA quantification involved concentration and purity of the DNA. Both were determined through the Nano-Drop spectrophotometer ND 2000 (Thermo Scientific, Cambridge, MA, USA). Finally, the DNA samples were diluted to a concentration of 250 ng µL-1 with deionized distilled water (ddH2O) for the EST-SSR PCR amplification. DNA amplification based on the selected EST-SSR primers The polymerase chain reaction (PCR) of the isolated gDNA was conducted using 16 EST-SSR primer pairs specific to the Lablab (Table 2) newly designed by the research team for this study. The PCR was carried out in a 15 µL containing 6.5 µL of One Taq Quick-Load 2X Master Mix with Standard Buffer (Inqaba Biotec, Pretoria, South Africa), 0.4 µM µL-1 of each the forward and reverse primers, 25 ng µL-1 of the genomic DNA and 6.0 µL of PCR water. PCR amplification was performed using a thermal cycler (Veriti™, Applied Biosystems, Foster City, CA, USA) following the modified protocol of Kamau et al.32. This process involved 2 min of initial denaturation at 94 °C followed by 35 cycles at 94 °C for 30 s (actual denaturation), 50–59 °C for 45 s (annealing depending on primer used: Table 2), and 72 °C for 1 min (extension). The final extension was performed at 72 °C for 5 min.Table 2Description of the EST-SSR primers used in the present study.S/NPrimer NameForward primerReverse primerProduct Size (bp)Annealing Temp (oC)1LabRRT36CCACACGTCATCGCTATCAGTTGGGTTGATTCTGAGTGGA15059, 62, 652LabRRT114TGGTAGGTGAGGAGGCCATAGTTATCCACGTCTCAACTAGTGGTCATT < 10059, 62, 653LabRRT63GACATGTACAGTCTTTGAGAAAGCTGGCCTTGAACTCAAGATACAACC320564LabRRT61TGCGAGGCTCTTTTGACTCTACTTAGTTATTCCACTGCATTGCATCCT300565LabS1GAAGAGGGAAGATTGTTGGAGTAAGTATCTTTCTTCGCTATGGTCCTG120566LabS23GGAATGGAATCAGATTGAACTGTGCTCTTCAAACCCAGTCTCAAAGAAACC100567KTD 45AGCTTTCCGTAGAGTCTCCTATAACCTGTTACCTTGTCTCCTTTATC80–100568KTD 96GCCTTAATAGAGTTCTTGAGAGAGGCTGGTGATCTTGTAAACAAAC90569KTD 130CAGAGTATAAAGGAGAGGAGTCATCTCACAATTGTTTAGGTGGAAG1755610KTD 133GATGAAGGTGAAGAGAGTATGAGTTAGCAGTGAAGAAGTGAGTGAGTA1485611KTD 195TGGTTGAATGAGAGAGTAAAGGGTTTCTTCAAGGTACATGTCTCAC1505612KTD 199TTCTTCTCTTCAACTTCACTCCACGAAGACAAGGAAGAGAAATC2505613KTD 278GCTCGTAAGTCACATCATATATCCGATGCGTTTCAAGTACACAAGT2405614LabT25GGGTTGAAGCTCACACAAATTCTTCCAATGATGGTTGTATGAGTAGCAC1265615LabT32CTAGCTTCCACGACATTTCCTTGTCTCGGAGTGAGAGTTCATCTCGTT1695616LabT28CTTTCTCCATGCAGACCAAACTTCCCTGTAAATAACTGTCCTGGGAAGC20456 Table 2Description of the EST-SSR primers used in the present study.S/NPrimer NameForward primerReverse primerProduct Size (bp)Annealing Temp (oC)1LabRRT36CCACACGTCATCGCTATCAGTTGGGTTGATTCTGAGTGGA15059, 62, 652LabRRT114TGGTAGGTGAGGAGGCCATAGTTATCCACGTCTCAACTAGTGGTCATT < 10059, 62, 653LabRRT63GACATGTACAGTCTTTGAGAAAGCTGGCCTTGAACTCAAGATACAACC320564LabRRT61TGCGAGGCTCTTTTGACTCTACTTAGTTATTCCACTGCATTGCATCCT300565LabS1GAAGAGGGAAGATTGTTGGAGTAAGTATCTTTCTTCGCTATGGTCCTG120566LabS23GGAATGGAATCAGATTGAACTGTGCTCTTCAAACCCAGTCTCAAAGAAACC100567KTD 45AGCTTTCCGTAGAGTCTCCTATAACCTGTTACCTTGTCTCCTTTATC80–100568KTD 96GCCTTAATAGAGTTCTTGAGAGAGGCTGGTGATCTTGTAAACAAAC90569KTD 130CAGAGTATAAAGGAGAGGAGTCATCTCACAATTGTTTAGGTGGAAG1755610KTD 133GATGAAGGTGAAGAGAGTATGAGTTAGCAGTGAAGAAGTGAGTGAGTA1485611KTD 195TGGTTGAATGAGAGAGTAAAGGGTTTCTTCAAGGTACATGTCTCAC1505612KTD 199TTCTTCTCTTCAACTTCACTCCACGAAGACAAGGAAGAGAAATC2505613KTD 278GCTCGTAAGTCACATCATATATCCGATGCGTTTCAAGTACACAAGT2405614LabT25GGGTTGAAGCTCACACAAATTCTTCCAATGATGGTTGTATGAGTAGCAC1265615LabT32CTAGCTTCCACGACATTTCCTTGTCTCGGAGTGAGAGTTCATCTCGTT1695616LabT28CTTTCTCCATGCAGACCAAACTTCCCTGTAAATAACTGTCCTGGGAAGC20456 Description of the EST-SSR primers used in the present study. Horizontal polyacrylamide gel electrophoresis (hPAGE) The PCR products were separated through horizontal polyacrylamide gel electrophoresis (hPAGE) as described by Kamau et al.32 and Male et al.33. The PAGE gel mixture was prepared by mixing 6% Acrylamide Bis-Acrylamide stock solution (19:1, 40%), 1X of 50X TAE buffer (2 Tris, 1 glacial acetic acid and 0.05 EDTA) and dH2O. To initiate polymerization, 20 µL of Tetramethylethylenediamine (TEMED) and 200 µL of freshly prepared 10% ammonium per sulphate (APS) were added to the gel solution immediately prior to casting. The gel mould was covered with glass plate pre-treated with silane solution to enhance gel binding. The glass plate with PAGE gel was transferred to the hPAGE tank (Kirkhouse Trust-Cleaver Scientific LTD. UK) containing 1X TAE buffer. Following loading of the samples and DNA ladder, the gel was subjected to electrophoresis at 120 V for 3 h. After separation of the PCR products, the gel was stained with 0.5 µg mL-1 EtBr for 30 min. The bands were finally visualized and photographed on UV trans-illuminator. Scoring of EST-SSR data The hPAGE ends up separating the EST- SSR amplicons (amplified DNA fragment) at different loci (levels). Since these amplicons from the EST-SSR are inherited as codominant markers, allowing detection of allelic variation, their fragment sizes were determined based on the standard 100 bp molecular ladder (AXYGEN Biosciences) that ranges between 100 and 1000 bp. The binary system was used to manually score the bands by assigning ‘1’ or ‘0’ for the presence or absence of the band at each locus, respectively. Analysis of markers’ data Genetic diversity parameters among the accessions The discriminating ability of the EST-SSR markers and the primers’ efficiency to disclose the genetic diversity existing among the accessions were evaluated based on various genetic diversity parameters. The genetic parameters across the accessions included polymorphic information content (PIC) computed through PowerMarker software version 3.25 (North Carolina State University, Raleigh, NC, USA)34,35. The major allele frequency (MAF) and average gene diversity (AGD) as well as the genetic parameters across the populations including the number of effective alleles (Ne), Shannon index (I), observed heterozygosity (Ho), expected heterozygosity (He), unbiased expected heterozygosity (uHe), and fixation index (F) were computed through the GenAlEx software version 6.50336,37. Principal coordinate analysis (PCoA) Principal Coordinate Analysis (PCoA) of the binary data was performed to estimate the genetic distance matrix among the population and among Lablab accessions using GenAlEx software version 6.50332,37. Cluster analysis Cluster analysis was processed to trace the genetic relationship among the Lablab accessions. The dendrogram was constructed using the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) in PowerMarker software version 3.25 (North Carolina State University, Raleigh, NC, USA)34 to decipher the genetic relationship. The clusters in dendrogram were visualized using Molecular Evolutionary Genetics Analysis (MEGA-X) software version 10.1.838. Analysis of molecular variance (AMOVA) AMOVA was computed to quantify the diversity level and genetic relationship among the population, and among 20 Lablab accessions using GenAlEx software version 6.50337. Data records EST-SSR marker based polymorphism among the lablab accessions contrasting for responses to the drought stress The 16 EST-SSR primer pairs (Table 2) were used to estimate polymorphism among 20 Lablab accessions. The polymorphism parameters among the accessions were reported as shown. DNA fragments All the 16 EST-SSR markers produced clear and scorable bands (Supplementary: Fig. 1) with a small variation among few accessions. The amplicon sizes and their binary scores are found in the (Supplementary: Table 1).Figure 1Principal Coordinate Analysis (PCoA) to estimate the genetic distance matrix among the 20 Lablab accessions and between the populations: Two populations i.e. DT and DS were determined. Principal Coordinate Analysis (PCoA) to estimate the genetic distance matrix among the 20 Lablab accessions and between the populations: Two populations i.e. DT and DS were determined. Genetic parameters among the lablab accessions The main genetic parameters among the Lablab accessions have been displayed in Table 3. A total of 33 alleles were discovered by 16 EST-SSR markers with a mean of 2.06 alleles per loci. The allele number per locus ranged from 3 for locus Lab61 to 2 for the rest of loci. The major allele frequency (MAF) ranged from 0.95 at almost half of the loci (Lab63; LabS1; LabS23; KTD195; KTD278; LabT32; and LabT28) to 0.75 at locus LB36. The MAF mean per locus was 0.89.Table 3Main genetic parameters among the Lablab accessions based on EST – SSR markers.No of MarkersLocus/ MarkerANMAFPICGD1Lab3620.750.300.382Lab11420.800.270.323Lab6320.950.090.104Lab6130.850.250.275LabS120.950.090.106LabS2320.950.090.107KTD4520.900.160.188KTD9620.850.220.269KTD13020.900.160.1810KTD13320.850.220.2611KTD19520.950.090.1012KTD19920.900.160.1813KTD27820.950.090.1014LabT2520.800.270.3215LabT3220.950.090.1016LabT2820.950.090.10Mean2.060.890.170.19AN Allele number, MAF Major allele frequency, PIC Polymorphic information content, GD Gene diversity. Table 3Main genetic parameters among the Lablab accessions based on EST – SSR markers.No of MarkersLocus/ MarkerANMAFPICGD1Lab3620.750.300.382Lab11420.800.270.323Lab6320.950.090.104Lab6130.850.250.275LabS120.950.090.106LabS2320.950.090.107KTD4520.900.160.188KTD9620.850.220.269KTD13020.900.160.1810KTD13320.850.220.2611KTD19520.950.090.1012KTD19920.900.160.1813KTD27820.950.090.1014LabT2520.800.270.3215LabT3220.950.090.1016LabT2820.950.090.10Mean2.060.890.170.19AN Allele number, MAF Major allele frequency, PIC Polymorphic information content, GD Gene diversity. Main genetic parameters among the Lablab accessions based on EST – SSR markers. AN Allele number, MAF Major allele frequency, PIC Polymorphic information content, GD Gene diversity. The polymorphic information content (PIC) varied among the loci with a mean of 0.17. Its variation ranged from 0.30 at the locus Lab36 to 0.09 at loci Lab63; LabS1; LabS23; KTD195; KTD278; LabT32; and LabT28. All the loci had less than 0.35 PIC. The gene diversity (GD) ranged from 0.38 at Lab36 to 0.10 among seven loci (Lab63; LabS1; LabS23; KTD195; KTD278; LabT32; and LabT28) with a mean of 0.19 per locus. Only loci Lab36, Lab114, and LabT25 had gene diversity in a range of 3.0 to 4.0, showing relatively higher gene diversity values. Genetic parameters between the populations The main genetic parameters between the populations have been displayed in the Table 4. The mean number of different loci (Nl), number of effective alleles (Ne), Shannon index (I) and unbiased expected heterozygosity (uHe) were observed more (2; 1.36; 0.34; 0.23) in the DT population than (1; 1.14; 0.18; 0.11) in the DS populations, respectively. All markers had lower observed heterozygosity (Ho) than expected heterozygosity (He), suggesting possible inbreeding or limited genetic variation among the accessions. The overall means for Ho and %Pm: Percentage polymorphism; SE: Standard error; N/A: Not applicable. The heterozygosity values were 0.03 and 0.00, respectively while fixation index (F) was 1.00 to both populations. The percentages of polymorphism (%Pm) among the DT and DS accessions were 60% and 40%, respectively.Table 4Main genetic parameters between the populations based on EST-SSR markers.PopulationLocus/ markerNlNeIuHeHoHeF%PmDrought Tolerant AccessionsKTD13321.220.330.190.000.181.0060.0KTD4511.000.000.000.000.00N/AKTD27811.000.000.000.000.00N/AKTD9621.470.500.340.000.321.00KTD19921.470.500.340.000.321.00KTD19521.220.330.190.000.181.00KTD13011.000.000.000.000.00N/ALabS121.220.330.190.000.181.00LabT3211.000.000.000.000.00N/ALabS2321.220.330.190.000.181.00LabT2811.000.000.000.000.00N/ALabT2521.920.670.510.000.481.00Lab6321.220.330.190.000.181.00Lab3622.000.690.530.000.501.00Lab6131.850.800.480.000.461.00Lab11421.920.670.510.000.481.00Mean21.360.340.230.000.221.00SE00.090.070.050.000.050.00Drought Susceptible AccessionsKTD13321.470.500.340.000.321.0040.0KTD4521.470.500.340.000.321.00KTD27821.220.330.190.000.1801.00KTD9621.220.330.190.000.181.00KTD19911.000.000.000.000.00N/AKTD19511.000.000.000.000.00N/AKTD13021.470.500.340.000.321.000LabS111.000.000.000.000.00N/ALabT3221.220.330.190.000.181.000LabS2311.000.000.000.000.00N/ALabT2821.220.330.190.000.181.00LabT2511.000.000.000.000.00N/ALab6311.000.000.000.000.00N/ALab3611.000.000.000.000.00N/ALab6111.000.000.000.000.00N/ALab11411.000.000.000.000.00N/AMean11.140.180.110.0000.111.00SE00.050.050.040.0000.030.00TotalMean21.250.260.170.0000.161.00100.0SE00.060.050.030.0000.030.00Nl Mean number of different loci, Ne Number of effective alleles, I Shannon index, uHe Unbiased expected heterozygosity, Ho Observed heterozygosity, He Expected heterozygosity, F Fixation Index. Table 4Main genetic parameters between the populations based on EST-SSR markers.PopulationLocus/ markerNlNeIuHeHoHeF%PmDrought Tolerant AccessionsKTD13321.220.330.190.000.181.0060.0KTD4511.000.000.000.000.00N/AKTD27811.000.000.000.000.00N/AKTD9621.470.500.340.000.321.00KTD19921.470.500.340.000.321.00KTD19521.220.330.190.000.181.00KTD13011.000.000.000.000.00N/ALabS121.220.330.190.000.181.00LabT3211.000.000.000.000.00N/ALabS2321.220.330.190.000.181.00LabT2811.000.000.000.000.00N/ALabT2521.920.670.510.000.481.00Lab6321.220.330.190.000.181.00Lab3622.000.690.530.000.501.00Lab6131.850.800.480.000.461.00Lab11421.920.670.510.000.481.00Mean21.360.340.230.000.221.00SE00.090.070.050.000.050.00Drought Susceptible AccessionsKTD13321.470.500.340.000.321.0040.0KTD4521.470.500.340.000.321.00KTD27821.220.330.190.000.1801.00KTD9621.220.330.190.000.181.00KTD19911.000.000.000.000.00N/AKTD19511.000.000.000.000.00N/AKTD13021.470.500.340.000.321.000LabS111.000.000.000.000.00N/ALabT3221.220.330.190.000.181.000LabS2311.000.000.000.000.00N/ALabT2821.220.330.190.000.181.00LabT2511.000.000.000.000.00N/ALab6311.000.000.000.000.00N/ALab3611.000.000.000.000.00N/ALab6111.000.000.000.000.00N/ALab11411.000.000.000.000.00N/AMean11.140.180.110.0000.111.00SE00.050.050.040.0000.030.00TotalMean21.250.260.170.0000.161.00100.0SE00.060.050.030.0000.030.00Nl Mean number of different loci, Ne Number of effective alleles, I Shannon index, uHe Unbiased expected heterozygosity, Ho Observed heterozygosity, He Expected heterozygosity, F Fixation Index. Main genetic parameters between the populations based on EST-SSR markers. Nl Mean number of different loci, Ne Number of effective alleles, I Shannon index, uHe Unbiased expected heterozygosity, Ho Observed heterozygosity, He Expected heterozygosity, F Fixation Index. Principal coordinate analysis (PCoA) The population structure analysis by PCoA (Fig. 2) confirmed the presence of two populations of accessions i.e. the drought tolerant (DT) and drought susceptibility (DS) populations with a clear distance between them. The DT accessions were mainly scattered on the center of coordinate one and upper part of coordinate two (i.e. in the quadrant 1, 2 and 4) while the DS accessions were aligned together on the lower part of the coordinate two (i.e. in the quadrant 3).Figure 2Dendrogram to decipher the genetic relationship among 20 Lablab accessions based on 16 EST-SSR markers; two main clusters: A and B were identified. Dendrogram to decipher the genetic relationship among 20 Lablab accessions based on 16 EST-SSR markers; two main clusters: A and B were identified. Cluster analysis The UPGMA-based method grouped the accessions into two major clusters (Cluster A and B) (Fig. 2). Cluster A has the accessions that were morphologically reported as the DT accessions while cluster B had DS accessions (Table 1). D147 was the remarkable accession in the DT population. The accessions HA4 and D349 as well as D359 and D352 had the similar genetic distance. The D311 was grouped in a population of DT accessions just closer to the cluster B. Analysis of molecular variance (AMOVA) The percentage (%) of variation between the populations and among the accessions by AMOVA revealed the presence of more (75%) genetic variation among the accessions rather than between (25%) the populations (Fst = 0.251) (Table 5). The p-value (0.01) also verified the significant difference among the accessions rather than between the populations.Table 5Molecular variation among the population and among the Lablab accessions based on the EST-SSR markers.Source of variationsDFSSMSSEstimated variance% Variationp-valueBetween the population110.10010.1000.439250.001Among the accessions3849.8001.3111.31175Total3959.9001.750100Fst = 0.251Fst F-Statistics (Genetic differentiation index), DF Degree of Freedom, SS Sum of Square, MSS Mean Sum of Square. Table 5Molecular variation among the population and among the Lablab accessions based on the EST-SSR markers.Source of variationsDFSSMSSEstimated variance% Variationp-valueBetween the population110.10010.1000.439250.001Among the accessions3849.8001.3111.31175Total3959.9001.750100Fst = 0.251Fst F-Statistics (Genetic differentiation index), DF Degree of Freedom, SS Sum of Square, MSS Mean Sum of Square. Molecular variation among the population and among the Lablab accessions based on the EST-SSR markers. Fst F-Statistics (Genetic differentiation index), DF Degree of Freedom, SS Sum of Square, MSS Mean Sum of Square. Technical validation EST-SSR marker based polymorphism among the lablab accessions contrasting for responses to the drought stress Molecular markers have a significant role in evaluating and developing improved varieties39. The SSR markers arise from repeated mutational events caused by the failure of DNA repair mechanisms to correct replication errors40. According to de Jesus Pires et al.41, these markers are valuable in breeding programs, especially for complementing morphological and biochemical evaluations in the development of elite breeding materials. Since they are PCR based, co-dominant and multiple allelic42,43, the SSR markers have become accurate, easier to score, and efficient in providing genetic information in crops when compared with some other markers44. In the present study, EST-SSR markers (Table 2) were used to establish genetic information correlating drought tolerance and susceptibility among Lablab accessions identified through the moisture stress screening experiment (Table 1)15. Both groups of accessions formed a Lablab population that exhibited contrasting responses to drought stress. The morphological45 and biochemical46 bases of their drought tolerance and susceptibility was established under semi-arid climatic conditions, producing similar findings. The overall genetic diversity recorded in this study was low (0.19), making it difficult to establish marker specificity for drought tolerance. Only three markers, Lab36, Lab114, and LabT25 were able to express at least a low level of specificity for drought tolerance (Supplementary: Table 1, and Table 3). In essence, these markers were at least able to correlate with some prior morphological45 and biochemical46 traits. Similarly, the polymorphic information content (PIC) especially on marker Lab36 conveyed little genetic information on drought tolerance among the accessions. Low specific and informative markers among the accessions in the present study could be due to their little discriminatory ability on drought tolerance. The markers used in this study, or closely related ones, have also been applied in Lablab diversity assessments across different countries, including India47, Thailand10, and Kenya32, generating similar findings to this study. However, a deeper exploration of molecular mechanisms is crucial for improving our understanding of drought tolerance in crops such as Lablab20. To gain a more detailed understanding of how individual markers contribute to cellular adaptation under stress conditions, the mechanisms of each marker, such as Lab 36, could be further investigated through high-resolution stress response profiling at the cellular level48,49. In addition, drought-induced gene expression profiling can facilitate functional validation in stress responses by linking markers to key drought-tolerance traits, such as belowground structural characteristics and biochemical detoxification pathways50,51. Even though genetic diversity was low among the accessions in this study, multivariate analyses including PCoA, cluster analysis, and AMOVA were able to resolve them into two distinct populations. This suggests the presence of underlying genetic structure despite overall limited variation among the samples. Such differentiation may reflect historical separation, selective pressures, or adaptation to distinct environmental conditions. Assessment of population structure by PCoA, cluster analysis, and AMOVA The matrix assessment of the accessions through PCoA in Fig. 1, revealed the presence of two Lablab populations in a separated distance. Accessions within the DT population showed a higher degree of distribution in the scatter plot compared with the DS population, attributable to the presence of both improved and non-improved genotypes. The DT population comprised improved varieties from different countries, as well as local cultivars and farmers’ landraces originating from both within and outside the country. Morphological responses under seedling-stage moisture stress reflected the differentiation of these two Lablab populations. The results of the cluster analysis indicated also the presence of two clusters with different sets of accessions. Population clusters aligned well with the recorded morphological traits. Cluster A (Fig. 2) included several promising DT accessions, as also revealed in both seedling screening experiment15 and multi-locational field evaluation52. In this context, accession D147 was among the top-performing accessions in both scenarios. The accession D147 originated from Ethiopia, which is considered a potential site for the domestication of Lablab53. Lablab cultivars from Ethiopia have been reported to exhibit high genetic diversity10,32. Cluster B (Fig. 2) comprised DS accessions that were positively correlated with morphological traits during the seedling stage45. These accessions were merely germplasm and landraces without any genetic improvement. Genetic relationships within Cluster ‘A’ revealed that certain accessions were closely related to one another. The observed close genetic relationships, HA4 with D349 and D359 with D352 suggest that they belong to improved cultivars. The HA4 is a photoperiodic insensitive and commercial Lablab variety from India[28,54] while D349 (Eldo-KT Maridadi), D352 (Eldo-KT cream), and D348 (Eldo-KT Black 2) are the commercial varieties from Kenya28,55. As the commercial varieties in Tanzania, D363 and D348, have been named NM-D 19 and NM-D 20, respectively52,56 while the accessions D55 and D250 are exotic germplasm introduced for research from Southeast Asia where Lablab is a major pulse crop (Venkatesha et al., 2013). Both D55 and D250 performed well in the present study, similar to D147, HA4, D349, D352, D363, and D359. In contrast, their seed yield in field conditions was inferior compared to the other accessions52. Similar genetic relationship among several accessions in Cluster B [(D106, D95, and D271); (D255, and D28); (D6 and D66)] could have been due to lack of genetic improvement among them. Despite its promising seed yield during field evaluation, the presence of the accession D311 closer to the DS accessions contradicts their morphological performance during the drought tolerance screening experiment. This controversy may be due to the epigenetic influence to their drought stress tolerance. The DNA methylation, histone modifications, and noncoding RNAs are some of the epigenetic mechanisms discussed to relate with stress responses in different legumes57 and even in some other plants58. Given that D311 is a farmers’ landrace, it is likely to possess fewer improved traits. However, specific epigenic drivers and other related factors can be determined through advanced genomic and high-resolution stress response profiling at the cellular level to identify genes and QTLs linked to drought tolerance48. D311 showed contrasting molecular and phenotypic patterns. This suggests possible complex genetic or environmental influences on drought response in this accession. A diverse resilience to drought stress among the Lablab accessions by the cluster analysis and AMOVA (Table 5) indicates a greater genetic differentiation among the accessions rather than between the populations. Genetic differentiation among crop accessions is the extent of genetic variation that distinguishes different accessions within a crop species, reflecting their evolutionary history and breeding background59. The results of the present AMOVA were consistent with some previous molecular studies in Lablab31,35,60, all of which some genetically distinct groups of accessions were reported. The variation between the two groups in Lablab favored the individuals within the population rather than among the population. The limited variability detected between populations could be explained by extensive gene flow among the Lablab accessions, the self-pollinating behavior of the crop, influences of cultural and management practices, soil parameters, varying farming systems, or a short separation period inadequate for measurable genetic differentiation. Similar causes for the observed patterns were described in earlier Lablab studies, including those by Kamau et al.32, Kamotho et al.31,61 in Kenya. Molecular diversity analysis using inter-simple sequence repeats (ISSR) was carried out by Abdallah et al.62 in Ethiopia to reveal polymorphic alleles among the Lablab genotypes. The cluster analysis performed with the UPGMA method grouped the genotypes into three distinct clusters, demonstrating that ISSR markers are effective for differentiating the studied Lablab genotypes. The SSR characterization of 96 Lablab accessions from across Kenya revealed a high level of genetic relatedness, as shown by UPGMA cluster analysis61. In that particular study, 10 SSR primers were employed, detecting an average of 4.3 alleles per primer. The mean He was 0.38, while the average PIC was 0.63. AMOVA indicated that 15% of the genetic variation existed among populations, with the remaining 85% occurring within populations. SSR markers were also employed Keerthi et al.47 in India to evaluate diversity at marker loci among 16 phenotypically diverse Lablab genotypes. In this study, 52 out of 55 SSR markers were polymorphic, representing 94.55% polymorphism. Amplification using 55 SSR primers produced 133 scorable alleles, averaging 2.5 alleles per locus. The SSR markers showed varying capacities to discriminate among the 16 genotypes, as reflected by effective multiplex ratio values ranging from 1.89 to 4.73 and marker indices ranging from 0.69 to 3.40. A similar kind of study by Kamau et al.32 in Kenya, this time involving both local and exotic accessions, detected a total of 39 alleles using 8 SSR markers. A PIC of 0.42 and an average of 4.88 alleles per PIC were detected. Gene diversity among the accessions ranged from 0.26 to 0.52, averaging 0.38, indicating moderate genetic variation. The AMOVA in this study attributed 94% of allele diversity to individuals within populations and only 6% among populations. Their Bayesian model-based Structure analysis and PCoA grouped the accessions into three clusters, with Kenyan germplasm distributed across all groups. Wild accessions clustered primarily with those from Southern and Eastern Africa, supporting previous evidence that Lablab originated in Africa6. Vaijayanthi et al.21 performed genome-wide association analysis in India to identify marker-trait relationships within a core collection of 64 Lablab accession using SSR markers. Substantial diversity was detected among the core set accessions at loci controlling quantitative traits (QTs), with 95 out of 234 SSR markers showing some polymorphism among them. Genetic diversity and population structure in 299 Lablab genotypes from Thailand were evaluated using 13 SSR markers22. The Thai study on Lablab revealed an overall gene diversity of 0.360, with comparable levels of gene diversity (He) and allelic richness across different geographic regions. While its STRUCTURE and neighbor-joining analyses identified two main clusters among these 299 individuals, PCoA revealed admixture patterns within the Lablab germplasm. To overcome the limitations of bulk analyses in heterogeneous accessions such as those examined in this study, single-cell RNA sequencing and spatial transcriptomics could be employed to characterize cell-type-specific transcriptional responses to drought stress. Nevertheless, additional field studies are compulsory to verify the molecular and genomic results. Field studies serve as a crucial step in assessing the effectiveness of markers’ technologies63. Some accessions from the present study were field-evaluated under dry climatic conditions of Northern Tanzania, where they exhibited promising performance trends52. Conclusion and recommendations Low level of polymorphism was observed between the DT and DS accessions in the present study, likely due to the limited discriminatory ability of the markers used. The low polymorphism observed highlights the need for other SSR markers with high discrimination ability or more informative markers such as SNPs. However, the PCoA, cluster analysis and AMOVA confirmed the presence of DT and DS populations with a clear separation between them. Therefore, the DT accessions (D147; HA4; D349) and DS accessions (D106, D95; D271) were selected as highly diverse breeding material contrasting for drought stress. Based on these findings, the following recommendations could be considered in unveiling all potentials of Lablab towards wide benefits of this crop among the farmers, consumers, scientific community and entire society at large. The contrasting accessions identified in this study, particularly D147, HA4, D349, D106, D95, and D271, may serve as valuable materials for future genomic analyses related to drought tolerance. Moreover, these approaches can be used to associate markers with critical processes for drought tolerance in plants, including reactive species detoxification and root architecture. Even the recommended drought-tolerant accessions (D147, HA4, D349, D363, D359, D352, and D348) could be further evaluated under field conditions in arid or semi-arid regions to support the selection of commercially promising varieties that meet farmers’ preferences. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1. Supplementary Material 1.