GENETIC DIVERSITY STUDY AND MOLECULAR MARKER DEVELOPMENT IN CYAMOPSIS TETRAGONOLOBA

Abstract

Cluster bean (Cyamopsis tetragonoloba [L.] Taub.), commonly known as guar, is used as fodder, vegetable and green manure. In recent times it has become a major industrial crop due to the need of the guar galactomannan/gum present in the endosperm of its seeds. Guar gum is useful in various industries like paper, textile, petroleum, drilling, pharmaceutics, food, cosmaceutics, explosives, etc. Guar is a drought resistant, hardy, deep rooted annual legume. In India the crop is mainly grown in the dry habitats of Rajasthan, Haryana, Gujarat, Punjab and to a limited extent in Uttar Pradesh and Madhya Pradesh. Outside India, guar is grown in Pakistan, South Africa, Brazil, Australia and Oklohoma planes of North Texas in USA. Guar is a cultivated crop not found in wild conditions and hence its available landraces are the main source of genetic variability. Guar is strictly a self-pollinated diploid legume with chromosome number (2n) equal to 14 and genome size approximately 2.45 Giga Bases/C. Cross pollination is prevented due to the cleistogamous nature of flowers. Thus, the heterosis available is reduced, which makes commercial hybrid seed production difficult and non-economical. This limiting factor of yield gap can be overcome by production of improved varieties of guar through molecular marker based selection and breeding programs. The DNA based molecular markers reveal natural variation at the DNA sequence level; these markers are used in plant genotyping, diversity studies, genetic linkage studies, quantitative trait mapping and marker-assisted selection during plant breeding. Hence, an overview of the genetic diversity and the development of molecular markers are very important for breeding and crop improvement in guar. In the present study genetic diversity in 19 commercial varieties and 29 landraces of cluster bean belonging to Gujarat, Rajasthan, Haryana and Delhi regions of India were analyzed using 13 RAPD (Randomly Amplified Polymorphic DNA) and 7 1SSR (Inter Simple Sequence Repeat) markers. The amplification using RAPD primers produced a total of 118 bands, out of which 103 were polymorphic and 15 monomorphic. Out of the 13 primers used OPQ-09 produced the highest number of bands (12); the average percentage polymorphism for RAPD markers was 87.63. UPGMA tree was constructed using Jaccard's similarity. The accessions of cluster bean distinguished into two major clusters at 75% similarity and a third cluster at lower similarity. The observed number of alleles, effective number of alleles, Nei's genetic diversity, Shannon's information index for landraces and commercial varieties using 13 RAPD markers were found to be 1.872±0.335, 1.589±0.351, 0.333±0.170, and 0.490±0.230, respectively. The value of total genotypic diversity among population (Ht) was 0.333±0.029, whereas diversity within population (Hs) was found to be 0.283±0.026. Mean coefficient of gene differentiation (Gst) value was 0.148 which indicated that 86.2% of genetic diversity was present within the population. AMOVA was used to analyze molecular variance among and within the population. Percentage of molecular variance was found to be 27% among populations and 73% variance was attributed to variance within the population. Seven ISSR markers used in the study produced 64 bands out of which 50 were polymorphic. Among the ISSR primers used UBC-868 produced highest number of bands (13); the average percentage polymorphism for ISSR markers was 77.82. The dendrogram from ISSR data showed one major cluster at 75% similarity and five minor clusters at lower level of similarity. The major cluster possesses six sub-clusters. The dendrogram did not differentiate between landraces and commercial varieties. The observed number of alleles, effective number of alleles, Nei's genetic diversity, Shannon's information index for landraces and commercial varieties using 7 ISSR markers were found to be 1.7812±0.4167, 1.4627±0.3844, 0.267±0.1939, and 0.3988±0.2681, respectively. The value of total genotypic diversity among population (Ht) was 0.2639±0.0378 whereas diversity within population (Hs) was found to be 0.253±0.035. Mean coefficient of gene differentiation (Gst) value was 0.041 and the estimated gene flow in the population was 11.549. AMOVA was used to analyze variation among and within the populations. Molecular variances were 8% and 92% among and within the population, respectively. Pooled RAPD+ISSR data of cluster bean distinguished into two major clusters at 75% similarity and 3 minor clusters at lower similarity in the dendrogram constructed. The observed number of alleles, effective number of alleles, Nei's genetic diversity, Shannon's information index for landraces and commercial varieties using 13 RAPD and 7 ISSR markers were found to be 1.8407+0.367, 1.5446±0.3671, 0.3130±0.1817, and 0.4584±0.2478, respectively. The value of total genotypic diversity among population (Ht) was 0.3089±0.0333 whereas diversity within population (Hs) was found to be 0.272±0.029. Mean coefficient of gene differentiation (Gst) value was 0.116 and the estimated gene flow in the population was found to be 3.787. AMOVA was used to analyze variation among and within the populations. Molecular variance among populations was found to be 21% and that within the population was 79% indicating higher variation within the population. The Mantle test revealed a significant correlation between the molecular data and the geographic data. The correlation value (R) for RAPD, ISSR, RAPD+ISSR data with geographic data were 0.5252, 0.3144, 0.5303, respectively. This indicates that molecular variation corresponds to differences in geographic distribution of landraces. A recombinant inbred population is essential for various purposes in plant genetics. Hence a cross was made between guar cultivars M-83 X RGC-1066. The parental lines used were unbranched type with 0-3 branches. M-83 is a vegetable variety with low gum content, glabrous leaves and, white flowe. While, RGC- 1066 is a gum producing variety with high gum content, hairy leaves and purple flowers. The leaf pubescence is associated with uneasiness in handling the plants, yet no information on the morphology of trichomes in guar is available. Scanning Electron Microscopy (SEM) was done to observe the fine structure of leaf pubescence. The SEM pictures showed long, slender, porous hair like structures in the hairy plant. The Fl hybrid plants showed hairy leaf, purple flowers and 0-3 branches. The F2 population segregated in Mendelian ratio showing —3:1 ratio of hairy and glabrous plants. No observable difference in the density of the pubescence was observed in hairy plants. Similar ratio was observed for purple and white flowered phenotypes, showing single locus control for these two phenotypes. But, the branching behavior showed transgressive segregation leading to very high branching in some F2 plants. In the population majority of the plants showed less than 10 branches per plant. Expressed Sequence Tags (EST) are considered as a quick and inexpensive source for obtaining Simple Sequence Repeat (SSR) markers. Available C. tetragonoloba EST (16,476) were downloaded from dbEST of NCBI. The EST sequences were trimmed using EST trimmer. The sequences so obtained (16,108) were assembled into contigs using CAP3. The candidate SSR containing sequences in the EST sequences (16,108), assembled contigs (1755) and singlet sequences (4320) were mined using the PERL script MISA. The program showed a total of 1568 microsatellite repeats in the complete EST sequences; among them 91 were in compound formation and remaining were perfect microsatellites. The contigs showed 327 microsatellite repeats from 276 sequences of which 28 were in compound formation. Singlets had 580 microsatellite repeats in 506 sequences of which 41 were in compound formation. Mononucleotide repeats (435) were the most abundant among the SSR types in C. tetragonoloba ESTs, followed by di-nucleotide repeats (189). Only one hexa-nucleotide repeat and seven penta-nucleotide repeats were found. A/T repeats were the most abundant form of nucleotide repeats. The sum of the microsatellite repeats from the contigs and singlets (907) were used for designing primers. Primers flanking the SSR regions were designed using Primer3. To test the EST based SSRs, 226 primers were synthesized and tested on 3 accessions of C. tetragonoloba, viz., M-83, RGC-1002, RGC-1066 and one accession each of C. serrata and C. senegalensis. The amplification, transferability and polymorphism of these markers were analyzed. Out of the 226 primers used 190 amplified to produce the product in the expected range. The polymorphism between the C. tetragonoloba accessions was very low; only 5 of the markers were polymorphic, showing very high homogeneity in the cultivated genotypes. Euclidean similarity coefficient was used to generate a dendrogram using the SSR data. The dendrogram showed that C. senegalensis was very diverse from the other accessions. The accession of C. serrata was very close to the cultivated guar cultivars. DNA from the F2 population was extracted and used to study the inheritance of the markers developed. Bulk segregant analysis was also done to check the linkage of leaf pubescence and flower color traits with the polymorphic markers derived. The amplification in bulks showed both the marker alleles of parents. This shows that the traits are independent of the markers tested. This research work showed significant variation between the landraces and commercially grown cultivars. In the landraces the diversity correlates with the geographic distribution. The work also showed very high variation in the genomes of the wild Cyamopsis species in comparison to guar. In this work a total of 362 primers flanking the SSR regions in guar were designed; this would be a useful resource for genetic studies and breeding in guar.