Analysis of Genetic Diversity in Twelve Cultivars of Pea Based on Morphological and Simple Sequence Repeat Markers

Pea(Pisum sativum L.)is the second most important legume crop worldwide after chickpea (Cicer arietinum L.) and valuable resources for their genetic improvement. This study aimed to analyze genetic diversity of pea cultivars through morphological and molecular markers. The present investigation was carried out with 12 pea cultivars using 28 simple sequence repeat markers. A total of 60 polymorphic bands with an average of 2.31 bands per primer were obtained. The polymorphic information content, diversity index and resolving power were ranged from 0.50 to 0.33, 0.61 to 0.86 and 0.44 to 1.0 with an average of 0.46, 0.73 and 0.76, respectively. The 12 pea cultivars were grouped into 3 clusters obtained from cluster analysis with a Jaccardd's similarity coefficient range of 0.47-0.78, indicating the sufficient genetic divergence among these cultivars of pea. The principal component analysis showed that first three principal components explained 86.97% of the total variation, suggesting the contribution of quantitative traits in genetic variability. The contribution of 32.59% for number of seeds per plant, stem circumference, number of pods per plant and number of seeds per pod in the PC1 leads to the conclusion that these traits contribute more to the total variation observed in the 12 pea cultivars and would make a good parental stock material. Overall, this SSR analysis complements morphological characters of initial selection of these pea germplasms for future breeding program.


INTRODUCTION
Pea (Pisum sativum L.) is an economically valuable coolseason pulse crop grown worldwide for its seed and soil fertility benefits (McPhee 2003). Being a short duration legume crop belonging to Leguminosae or Fabaceae family, it fixes its own nitrogen and therefore, can become an excellent candidate for bio-energy especially in temperate regions. In India, it is an important legume crop after chickpea and pigeon pea due to its high nutritive value, particularly proteins; 7.2 g 100 -1 g (Singh 2007). Because of its important source of vegetable protein (21-32%) in major part of the world, recent studies have underlined the latent health profits of pulses in reduction of type II diabetes as well as cardiovascular diseases (Boye et al. 2010).In Asian countries, this crop is consumed as green vegetable (whole pods or immature seeds), whereas dry seeds are consumed in Europe, Australia, America and Mediterranean regions (Ghafoor and Arshad 2008). An increasing demand for proteinrich food and feed around the world has highlighted the commercial importance of pulses as protein source (Santalla et al. 2001).
The comparatively narrow gene pool (Hebblethwaite et al. 1985) as well as the hefty use of a petty numbers of cultivars as parents in competing breeding programs have directed to a little genetic mixture among pea cultivars (Simioniuc et al. 2002;Baranger et al. 2004), resulting in vulnerability to pests and diseases (Duvick 1984;Cox et al. 1986). This study reported that at proliferation of grain legume production by thorough utilization of high yielding cultivars enriched with tolerance to biotic and abiotic stresses.
The major pea diversity analysis is based mainly on pedigree data, morphological characters and molecular markers (Simioniuc et al. 2002). Since morphological characters are commonly influenced by environmental factors and in some species, adequate level of morphological polymorphism is inaccessible, they are of restricted status in an evaluation of genetic diversity (Patto et al. 2004).
There are several methods present to harness genetic diversity among the genotypes in crop improvement which includes allele mining. The sequence based method entails detection of variation in DNA sequences of various lines by PCR amplification of alleles. Another technique to identify DNA sequence polymorphism is targeted induced local lesions in genomes (TILLING). However, these approaches are costly and time intense. Alternatively, molecular markers can assist to study the genetic diversity in crops. Molecular markers are suitable to supplement the morphological and phonological characterization because they are abundant, free from tissue or environmental effects and permit genotype identification in the early stages of development.
Microsatellites or simple sequence repeats (SSR) are generally used for assessing genetic diversity in peas due to their precision, consistency, co-dominance and reproducibility. Moreover, easy detection of high polymorphism markers with PCR procedure, performs as the best existing choice of markers for pea diversity assessment and characterization (Loridon et al. 2005). For these reasons, microsatellites have been extensively used in gene tagging, genome fingerprinting, genome mapping and marker-assisted selection for numerous crops including pea (Burstin et al. 2001;Loridon et al. 2005;Nasiri et al. 2009;Cupic et al. 2009;Sarıkamış et al. 2010;Gong et al. 2010).
The objective of this study was to characterize pea cultivars on molecular level to support morphology in order to give information potentially utilized for selection of better parents for effective breeding programmes.

Plant Material
A total of 12 pea cultivars were used in this study. The genetically pure nucleus seed were acquired from the pea breeder Chandrasekhar, Azad Agricultural University, Kanpur (U.P), India. The 12 pea cultivars were grown under greenhouse and average day temperature was adjusted around 18-20 °C for 24 h.

Genomic DNA Isolation, Quality and Quantity Determination
Genomic DNA was extracted from each plant, selecting fresh, young disease free leaves at 8-10 leaf stage. Fresh young leaves were grinded into powder with liquid N 2 using a mortar and pestle. After that, plant genomic DNA was extracted following the method of Doyle and Doyle (1987) with slightly modified by Bhattacharyya and Mandal (1999).
The quality of extracted DNA was analyzed by agarose gel electrophoresis (0.8%), followed by ethidium bromide staining. The purity of the DNA was estimated by spectrophotometer using A260/A280 ratio, and the yield was estimated by measuring absorbance at 260 nm.

Phenotypic Performances
The cultivars were raised with recommended agronomical practices and observations were recorded on a randomly selected five competitive plants from each replication for morphological characters viz. plant height at maturity (cm), germination percentage, number of nodes per plant, number of leaves per plant, length of branch from main axis (cm), internode distance (cm), number of branches per plant, stem circumference (cm), number of seeds per plant, number of pods per plant, weight of 100 seeds (g) and number of seeds per pod.
The collected data were mean of five randomly selected plants from each replication.

Data Analysis
For SSR marker data the presence or absence of the bands was scored as 1 or 0, respectively, obtaining the molecular identification profile for each individual.

Cluster analysis was implemented by Unweighted Pair
Group Method with Arithmetic Mean (UPGMA) and the corresponding dendrogram was constructed. To estimate the goodness of fit between similarity matrix and the dendrogram, the coefficient of cophenetic correlation was calculated using the NTSYSpc (Rohlf 2009) software. The capacity of each primer to distinguish among the cultivars studied was evaluated by the resolving power (RP) (Prevost and Wilkinson 1999) and the polymorphic information content (PIC) (Weising et al. 2005). PIC of dominant bi-allelic data was estimated by the formula: PIC = 1-pi2-qi 2, where "p" is frequency of visual alleles and "q" is the frequency of null alleles. PIC for the SSR marker was estimated by using the formula PICi = 2fi (1-fi). Where, fi is the frequency of the marker fragments that were present and (1-fi) is the frequency of the marker fragments that were absent. RP is defined per primer as: RP = Σ Ib, where "Ib" is the band informativeness, that takes the values of 1-(2x [0.5-p]), being "p" the proportion of each genotype containing the band. The phenotypic data of the 12 pea cultivars were analyzed using XLSTAT software. The PC was used to determine the extent of genetic variation. Eigen-values were obtained from PC, which were used to determine the relative discriminative power of the axes and their associated characters. Ward method was used to group the twelve accessions based on their genetic relationship.

SSR Polymorphism
The DNA amplification of 12 pea cultivars using 28 SSR primers (Loridon et al. 2005;Kumari et al. 2013) revealed that only 26 primers were polymorphic, but not for AB91 and AD60. They showed different abilities to identify unique multiband among 12 cultivars. The example of DNA banding pattern on gel electrophoresis using primers AD148 and AA1 was illustrated in Figure  1. The primers produced 1-4 bands, of which, AD148 yielded the highest.
The polymorphism information content (PIC) differed between SSR primers, highest for AD174 and AA416 (0.50), and lowest for AA355 (0.33) with mean value of 0.46 (Table 2), suggesting their good indicators of the pea genetic diversity. The twenty six SSR markers revealed the usefulness of a marker in distinguishing accessions with DI values ranging from 0.61 (D21) to 0.86 (AA205) with an average of 0.73 (Table 3). The high DI values (more than 0.50) indicated that the SSR markers were informative. The estimates of RP were found to be the highest for AD174 and AA416 (

Cluster Analysis Using SSR Markers
Based on the constructed dendrogram, with NTSYSpc (Rohlf 2009), the 12 pea cultivars were grouped into three major clusters ranging from 0.47 to 0.78 coefficient scale ( Figure 2A). Clusters I contains the highest number cultivars including KPMR-763, KPMR-906, KPMR-918, KPMR-921, KPMR-913 and KPMR-820. The cluster II and III grouped four and two cultivars, respectively. Dendrogram summarized the existing genetic similarity/ dissimilarity among pea cultivars within the cluster based on molecular marker (SSR) characteristics. The PCA analysis of the SSR marker data from twelve pea cultivars also supported the cluster analysis. A three dimensional plot (3D plot) diagram was prepared using the first three principal components ( Figure 2B). Similar to the dendrogram, six cultivars preferentially were grouped together (cluster I). Whereas, the cultivars KPMR-922, KPMR-400 and KPMR-902 were in cluster II, and the remaining cultivars were in cluster III. These distinctive clusters demonstrated their varied genetics in support to phenotypes.

Phenotypic Performances
All of the quantitatively measured traits showed a high range of variation among 12 pea cultivars. The analysis of variance revealed significant variation among the cultivars for all of the studied morphological parameters at 1% critical level. The mean performance of the cultivars based on the phenotypic parameters indicated their cumulative performance with regards to multiple parameters (Table 4) The Pearson correlation analysis (Table 5) indicated significant correlation of plant height with germination percentage and stem circumference. The number of seeds per plant was positively correlated with number

A B
of branches and stem circumference. The number of pods per plant was positively correlated with number of nodes, stem circumference and number of seeds per plant. A significant positive correlation was also found for number of seeds per pod with number of branches and number of seeds per plant.
Eigen vectors indicate the degree of association among the original data and each principal component. The first three PC axes accounted for 86.97% (Table   6) of the multivariate variation among the entries indicating a higher degree of correlation among characters for these entries.    influenced number of seeds per pod. High magnitude of negative relationships were also observed for number of branches (PC2) and internode distance (PC3). It is interesting to note that few characters as plant height, length of branches from main axis and number of seeds per pod influenced multiple principle components most of which did not influence principle component 1.
These characters have a maximum variability among cultivars.
Hierarchical clustering using Euclidean distance as a distance measure was done for grouping the germplasm suitably. On the basis of dendrogram constructed by average linkage between groups (Figure 3) These clusters clearly indicate the relation among the cultivars and the distance based on their overall parameter variables.

CONCLUSION
The diversity among cultivars may be assessed based on morphological and molecular markers. However, systematic studies regarding the genetic diversity of pea through molecular markers in India are meagre. Hence, in-depth studies based on morphological and molecular markers SSR will help in understanding the genetic diversity of germplasm as well as identification, conservation and utilization of authentic and superior crop materials. The results indicate the presence of moderate genetic variability among the elite green pea cultivars. Among all the cultivars, the KPMR-870 and KPMR-920 genotypes from cluster III have wider genetic diversity and suggested to utilize in crop improvement program.