Morpho-molecular Diversity Analysis in Rice (Oryza sativa L.) Genotypes using Microsatellite Markers

Rice (Oryza sativa L.) is a source of instant energy and has a high content of carbohydrates and is known to be the staple food for Indian consumption at a large scale. A large fraction of the world’s population consumes rice, making it a major cereal food crop. According to United Nation, it has been expected that the world population might reach up to 8.5 billion by 2030 and thus, will create high pressure on rice breeding programs worldwide (Van Bavel, 2013). Therefore, to meet the increasing demand of the population, rice production must be increased by 50%.
 
Improvement in the genetic potential of the crop cultivars is one of the best approaches among very few options available to increase the yield potential of rice. Rice is rich in natural genetic diversity and there is humungous scope to exploit diversity for improvement of genetic base. Genetic diversity contributes to the production of superior recombinants and so it becomes an important element in crop improvement programs. Successful crop improvement program, mainly hybrid breeding, depends on genetic diversity present among the parents (Rana et al., 2004); a lack of genetic variability would have the potential to significantly limit breeding progress and/or yield and quality crop improvements (Cornelious et al., 2002). A quantitative assessment carried out on genetic diversity guides the breeder for the advancement in breeding programs. It aids in the selection of agronomically suitable diverse parents for hybridization and is also important to obtain desired recombinants in the segregating generations (Goulet et al., 2017). The development of novel biotechnological techniques contributes towards the evaluation of genetic variation at both phenotypic and genotypic levels. The results derived from analyses of genetic diversity at the DNA level could be effective for designing better breeding programs with a promising aim to broaden the genetic basis of commercially grown varieties (Gaballah et al., 2021).
 
Molecular markers are known to be a powerful tool for assessing genetic variation in rice genotypes. In contrast to morphological traits, molecular markers can provide insights about the variations among genotypes at the molecular level, providing more direct, reliable and efficient results for germplasm characterization, conservation and management. Additionally, the results are unaffected by environmental influence (Singh et al., 2015). Among the various kinds of markers available, microsatellite markers or Simple Sequence Repeat (SSR) markers can detect a high level of allelic diversity and they have been extensively used to identify genetic variation among rice subspecies (Chukwu et al., 2020).

Morphological, genetic and physiological characters have been used efficiently to improve hybrid breeding in rice along with genetic estimates. D² analysis is done to evaluate the conventional heritable diversity (Manomani and Khan 2003). This technique was developed by Mahalanobis (1936), is a multivariate analysis technique that is useful to the breeders for the estimation of genetic diversity in base population or germplasm collections concerning the characters considered together (Tejaswini et al., 2018).

However, with the increasing world population, environmental condition is deteriorating every day, so in such conditions, food security has become a major challenge around the world, especially in Asia and Africa. Exploring the genetic aspects of germplasm using genotypic and phenotypic analysis is much dependable for the development of rice varieties that could pave the way towards the attainment of food security (Mohammad et al., 2020). Thus, in the current study, a morpho-molecular approach was used to explore the diversity present among different rice varieties. 

The current investigation was carried out during Kharif -2019 at Agricultural Research Farm, Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, U.P., India. The molecular analysis was performed at the Central Laboratory (Niche Area Lab) of Banaras Hindu University. The experimental material for this investigation comprised of forty-seven rice genotypes and are listed in Table 1. Field observations were recorded on ten quantitative characters viz., days to 50% flowering, days to maturity, plant height(cm), total number of effective tillers per plant, panicle length (cm), plot yield(kg), biomass (kg/plot), 1000 grain weight (gm), harvest index and grain yield in kilogram per hectare (GYKGH) (kg). The calculated mean value for the recorded data was subjected to analysis of variance.


 
DNA isolation and Polymerase chain reaction
 
Genomic DNA was extracted from young leaves collected from 15-20 days old rice seedlings and stored immediately at -20^oC till further processing. DNA was extracted following the CTAB extraction method suggested by Doyle and Doyle, 1987 including few modifications made in the composition of DNA extraction buffer.

The polymerase chain reaction (PCR) was performed in a total volume of 15 µl per reaction containing 1.0 µl of template DNA (40 ng/µl), 1.0 µl of forward and reverse primers (10 µM stock concentration), 0.20 µl dNTPs (1 mM), 0.20 µl Taq polymerase (5 U/µl), 0.20µl of MgCl2 (50 mM), 1.50 µl 10 x PCR buffer and remaining volume was made up by HPLC Water. The PCR amplification was carried out on a thermal cycler at an initial temperature of 94^oC for 3 min, followed by 35 cycles of 94^oC for 30 s, 55^oC for 30 s (primer annealing occurred with most of the primers while some were adjusted), 72^oC for 1 min and at least 72^oC for 5 min and then stored at 4^oC. A total of twenty-four simple sequence repeat (SSR) markers were used for studying molecular diversity as denoted in Table 2.


 
Visualization and scoring of the PCR amplified fragments.
 
The PCR products generated through SSR primers were separated by electrophoresis in 2.5% concentration agarose gel prepared in 1x TAE buffer (electrophoresis buffer). The amplified product mixed with 2 µl of 6x gel loading dye (Bromophenol blue 0.25%; Xylene. cyanol 0.25% and glycerol 30%) were loaded in each well using a micropipette. After the run, the amplified products were visualized and photographed under UV light source in a gel documentation system (Gel Doc TM XR+, BIORAD, USA). The genotypes were manually scored using the binary coding system, ‘1’ for the presence of band and ‘0’ for absence of a particular band.
 
Data analysis
 
The quantitative trait mean values were calculated based on data of five randomly tagged plants in each genotype and were used for statistical analysis. Analysis of variance was performed to partition the total variation among the entries following randomized complete block design using WINDOSTAT version 9.0. Adjusted trait means of each of the genotype accession was estimated (Federer, 1956) and the same was used for all subsequent statistical analysis. In the present study, Mahalanobis’ generalized distance as described by Rao (1952) was used to estimate the genetic divergence.

Standardization was done by dividing the deviation of the mean for a line from the mean for forty-seven genotypes with the standard deviation for the given trait; the STAND module of NTSYS software was used to furnish the same. The binary data matrices generated by polymorphic SSR markers were subjected to further analysis using NTSYSpc version 2.11W. The Jaccard’s coefficient was calculated using SIMQUAL program. Unweighted pair group average (UPGMA) based clustering was done using SAHN module of NTSYSpc for dendrogram construction.
 
Polymorphic information content (PIC)
 
PIC value is used to estimate the discriminatory power of the SSR marker. The formula for PIC calculation for SSR markers:

                            

Where,
pi stands for the frequency of ith allele and summation extends over n patterns.

Analysis of variance (ANOVA) based on the mean values of ten quantitative traits in forty-seven rice genotypes recorded significant differences among the genotypes with respect to all the characters considered for the study (Table 3). This suggests that there is an inherent genetic difference among the genotypes for the traits studied.


 
Diversity analysis using morphological data.
 
D2 analysis revealed the presence of significant diversity in the set of forty-seven genotypes assessed. The germplasm was distributed into six different clusters (Table 4). The genotypes within each cluster were comparatively closer to each other than the genotypes present in different clusters. Maximum number of genotypes (20) were included in cluster II followed by 18 genotypes in cluster I, cluster III and V are with 4 and 3 genotypes, respectively. Clusters IV and VI are solitary clusters. Similar results were observed by Islam et al., (2020), as they used 113 rice genotypes for D2-analysis and observed that these genotypes were grouped into 4 clusters. Diversity analysis by Tejawini et al., (2018) and Salem and Sallam (2016) have also shown similar findings. 



Cluster I showed the maximum intra-cluster distance (64.47) with 18 genotypes followed by cluster III (64.05) with 4 genotypes. The intra-cluster distance for clusters IV and VI was observed to be zero as these two clusters were monogenotypic. The maximum inter-cluster distance was observed between clusters II and VI (1134.14), followed by clusters II and III (802.93) and clusters V and VI (788.86). This suggests that the hybrids developed from the selected genotypes of these clusters would have a high chance to obtain heterotic combinations and produce a highly variable population in the segregating generations. One should not forget the fact that, along with the genetic distance, per se yield and yield contributing characters should also be considered while selecting the genotypes for breeding programs. In this respect, genotypes like Desi Dhan, IR 82475-110-2-2-1-2, DRR Dhan 48, etc. have the highest inter-cluster distance and high per se yield so they can be considered as high yielding parents in hybridization programs.

The lowest inter-cluster distance was observed between clusters IV and VI (112.81), followed by clusters I and IV (151.44) and clusters III and IV (159.57). The genotypes in these clusters are genetically very close and hence hybridization among these lines may not be very fruitful.  These results are in accordance with the earlier reports of Thippeswamy et al., (2016), Devi et al., (2019) and Dey et al., (2020). The intra and inter-cluster distances (D2 values) were given in Table 5 and the cluster diagram was shown in Fig 1.  The inter-cluster distances between all the clusters were higher than the intracluster distances suggesting wider genetic diversity among the genotypes of different cluster groups. This is in parallel with the findings of Mohan et al., (2015) and Singh et al., (2020). Fig 2 represents the dendrogram of forty-seven genotypes of rice by Tocher method.


 




Molecular diversity using microsatellite (SSR) markers
 
Molecular diversity study has been assessed by using chemically designed molecular markers. These sequences are the complementary sequences of DNA that lie close to a particular gene or QTL and through primer annealing, they amplify the target gene. In the present study diversity analysis among forty-seven rice genotypes was done using 24 rice SSR markers. Out of the 24 SSRs used, all the markers showed polymorphism.
 
Scoring of SSR bands and PIC value
 
The polymorphic information content (PIC) was calculated for each locus to assess the information of each marker and its discriminatory ability and it is an evidence of allele diversity and frequency among varieties (Table 2). The highest PIC value was observed for the locus RM 507 (0.86) followed by RM 293 (0.84), RM 5791 (0.74), RM 545 (0.73) and RM 312 (0.71) and lowest by RM 552 (0.12) followed by RM 410 (0.15), RM 162 (0.22), RM 284 (0.28) and RM 19 (0.29). So, PIC value ranged from 0.86 to 0.12 with a mean value of  0.465. All the twenty-four primers showed polymorphism and the number of alleles were 2 in every case. A total of 50 alleles were amplified from 47 genotypes and this shows significant variability among the genotypes. PIC value revealed that RM 507 was considered as best marker. Rafii et al., (2018) reported the genetic relationship and diversity analysis among 8 aromatic rice cultivars using 32 SSR primers, detecting a total of 131 alleles. The average number of alleles per locus was 4.09. Similar results were reported by Pathaichindachote et al. (2019) evaluated a set of 167 rice genotypes using 13 polymorphic SSRs to assess the genetic diversity and genetic relationship. A total of 110 alleles were amplified and the Polymorphic Information Content (PIC) values for SSR primers ranged from 0.27(RM542) to 0.87(RM21) with a mean of 0.59. So, PIC value is related to the polymorphism of markers and a higher PIC value indicates high polymorphism.
 
Dendrogram analysis
 
Dissimilarity coefficient was used to determine the level of relatedness among the genotypes. The dissimilarity coefficient varies from zero to one, a value closer to one reflects a higher dissimilarity, whereas, closer to zero reflects a higher similarity. The average dissimilarity ranged from 0.5516 to 0.7862. A dendrogram (Fig 3) based on Jackard’s dissimilarity coefficient was constructed using UPGMA (Unweighted Pair Group Method with Arithmetic Averages) method and forty-seven rice genotypes were grouped into three main clusters i.e., cluster I, cluster II and cluster III. Cluster I was further sub-divided into two minor sub-groups IA and IB and IA was differentiated into IA -1 and 1A-2. Cluster IIB  was further sub-divided into two subgroups i.e., IIB-1 and IIB-2. Cluster III was divided into three subgroups IIIA, IIIB and IIIC. The genotypes included in these clusters are presented in Table 6.





The total average of dissimilarity coefficient of all the 47  genotypes is 0.701. The dissimilarity coefficient varied with a highest value of  0.78 among the cultivar IR 85850 and Lal Sundiya followed by Chauli and Swarna (0.76) and DRR Dhan 45 and URG-1 (0.73). These values reflect high dissimilarity between them showing that they are highly dissimilar from each other. The lowest value (0.48) was found between Jlgaam and URG-24 followed by 0.50 between Lal Sundiya and Swarna. This suggests there is a considerable diversity present among the genotypes in the study. Similar clustering of genotypes into various clusters and sub-clusters were reported by Das et al., (2013), Shahriar et al., (2014), Vengadessan et al., (2016), Dey et al. (2020) and Pathak et al., (2020).

Finally, according to the dendrogram and Jackard dissimilarity coefficient values, the most diverse cultivars among the 47 rice genotypes studied are IR 85850 and Lal Sundiya followed by Chauli and Swarna, DRR Dhan 45 and URG-1, there is a high chance that upon utilization of these genotypes in hybridization programs for crop improvement, will give useful segregants as they showed greater genetic distances. Additionally, these genotypes have lower relatedness making them genetically diverse which is of prime importance for any breeding program.

The present study shows the existence of a considerable level of morphological and molecular diversity among the 47 rice genotypes assessed. Analysis of variance discovered highly significant differences among all the characters and reveals the presence of a high degree of variability in the studied material for effective selection. Based on morphological and molecular diversity analysis the most diverse genotypes observed were Desi Dhan and IR 91143-AC 293-1, Desi Dhan and BD 105, IR 85850 and Lal Sundiya, Chauli and Swarna. Breeders may attempt hybridization among the above genotypes that showed maximum diversity, for creating more variability in rice and further to obtain desirable segregants.

The authors highly acknowledge Harvest Plus Project funded by IFPRI, USA and CIAT, Columbia for providing the germplasm and financial support for this work to be done.