Assessment of genetic diversity in Indian rice germplasm (Oryza sativa ...

c Indian Academy of Sciences

RESEARCH ARTICLE

Assessment of genetic diversity in Indian rice germplasm (Oryza sativa L.): use of random versus trait-linked microsatellite markers SHEEL YADAV1,5 , ASHUTOSH SINGH1 , M. R. SINGH2 , NITIKA GOEL1 , K. K. VINOD3 , T. MOHAPATRA4 and A. K. SINGH1 ∗ 1

Division of Genetics, Indian Agricultural Research Institute, New Delhi 110 012, India 2 Post Graduate College, Ghazipur 233 001, India 3 Indian Agricultural Research Institute, Rice Breeding and Genetics Research Centre, Aduthurai 612 101, India 4 Central Rice Research Institute, Cuttack 753 006, India 5 Division of Genomic Resources, National Bureau of Plant Genetic Resources, New Delhi 110 012, India

Abstract Assessment of genetic diversity in a crop germplasm is a vital part of plant breeding. DNA markers such as microsatellite or simple sequence repeat markers have been widely used to estimate the genetic diversity in rice. The present study was carried out to decipher the pattern of genetic diversity in terms of both phenotypic and genotypic variability, and to assess the efficiency of random vis-à-vis QTL linked/gene based simple sequence repeat markers in diversity estimation. A set of 88 rice accessions that included landraces, farmer’s varieties and popular Basmati lines were evaluated for agronomic traits and molecular diversity. The random set of SSR markers included 50 diversity panel markers developed under IRRI’s Generation Challenge Programme (GCP) and the trait-linked/gene based markers comprised of 50 SSR markers reportedly linked to yield and related components. For agronomic traits, significant variability was observed, ranging between the maximum for grains/panicle and the minimum for panicle length. The molecular diversity based grouping indicated that varieties from a common centre were genetically similar, with few exceptions. The trait-linked markers gave an average genetic dissimilarity of 0.45 as against that of 0.37 by random markers, along with an average polymorphic information constant value of 0.48 and 0.41 respectively. The correlation between the kinship matrix generated by trait-linked markers and the phenotype based distance matrix (0.29) was higher than that of random markers (0.19). This establishes the robustness of trait-linked markers over random markers in estimating genetic diversity of rice germplasm. [Yadav S., Singh A., Singh M. R., Goel N., Vinod K. K., Mohapatra T. and Singh A. K. 2013 Assessment of genetic diversity in Indian rice germplasm (Oryza sativa L.): use of random versus trait-linked microsatellite markers. J. Genet. 92, 545–557]

Introduction Rice (Oryza sativa L.) is one of the world’s most widely cultivated crop species. As a major cereal crop, it is one of the most diversified crop species due to its adaptation to a wide range of geographical, ecological and climatic regions. India is remarkably rich in rice diversity, including cultivars, landraces, wild and weedy relatives (DRR, Hyderabad). The rice germplasm is a rich reservoir of valuable genes that plant breeders can harness for crop improvement. There are several ways for estimation of diversity in germplasm, such as evaluation of phenotypic variation, ∗ For correspondence. E-mail: [email protected]. Sheel Yadav and Ashutosh Singh contributed equally to this work.

biochemical and DNA polymorphisms. However, both phenotypic and biochemical characterizations are unreliable because they are environmentally challenged, labour demanding, numerically and phenologically limited. On the contrary, DNA-based molecular markers are ubiquitous, repeatable, stable and highly reliable (Ford-Lloyd et al. 1997; Virk et al. 2000; Song et al. 2003). Among the several classes of available DNA markers, microsatellite or simple sequence repeat (SSR) markers are considered the most suitable due to their multiallelic nature, high reproducibility, codominant inheritance, abundance and extensive genome coverage. A large number of SSR markers have been developed and mapped in rice (Temnykh et al. 2000; McCouch et al. 2002), which vary in the degree of polymorphism depending on

Keywords. germplasm; genetic diversity; microsatellite markers; principal component analysis. Journal of Genetics, Vol. 92, No. 3, December 2013

545

Sheel Yadav et al. their position in the coding or noncoding segments, nature of their repeat motifs and the genomewide abundance. Therefore, an ideal set of SSR markers providing genomewide coverage will facilitate an unbiased assay of genetic diversity which in turn will provide a robust, unambiguous molecular description of rice cultivars. The assessment of diversity of Indian rice germplasm has been carried out by many researchers in the past (Saini et al. 2004; Ram et al. 2007; Sundaram et al. 2008; Kumar et al. 2010; Sivaranjani et al. 2010; Vanniarajan et al. 2012). However, most of these studies have used a set of random, nondescript SSR markers. It has been hypothesized that the use of random markers for assessing genetic diversity might not reflect the functionally useful variations prevalent at the coding regions of the genome (Zhang et al. 2010), a crucial requisite for the breeding programmes, where the diversity occurring at identified ‘heterotic loci’ should be taken into account for selection of suitably diverse parental lines. Therefore, it is pertinent to study and compare the pattern of genetic diversity, as established, by using random visà-vis trait-linked SSR markers, which would confirm their suitability to assess genetic diversity. In this investigation, we selected 88 genotypes that included landraces, mega-varieties, elite and popular cultivars, and breeding lines developed and sourced from different rice breeding centres across India having common knowledge on their variability, adaptability and popularity with the objective of identifying genotypic heterotic pools to exploit heterosis and recombination advantage in future breeding programmes. Special selection emphasis was given to yield and yield components. While determining genetic diversity of 88 rice germplasm accessions, we compared the efficacy of a random set of SSR markers as against another set of SSR markers reportedly linked to yield and yield components.

Materials and methods Plant materials

Study materials consisted of 88 rice accessions representing landraces, released varieties, cultivars and breeding lines from different breeding centres across India (table 1). The accessions were field-grown during kharif season of 2010 at Indian Agricultural Research Institute, New Delhi, India, in randomized blocks with three replications. The seedlings were raised through panicle to row method to maintain genetic purity. Three rows of four metre each formed the plot for each genotype. The experiment was subjected to standard agronomic management (includes aspects regarding plant geometry, fertilizer application and insect management. These have been standardized for each crop and were implemented here for rice). Evaluation of morphological traits

Data were recorded on five competitive plants per genotype from the middle row for yield and yield components such 546

as days to 50% flowering (DFL), days to maturity (DMT), plant height (PLH), number of tillers per plant (TLN), panicle length (PNL), single plant yield (SPY), 1000-grain weight (TGW), total grains per panicle (TGN), number of filled grains per panicle (FGN), number of unfilled grains per panicle (UGN) and spikelet fertility (SFP). Selection of SSR markers

A priori information was used to select random as well as trait-linked SSR markers. Selected random markers consisted of a set of 50 markers having genomewide distribution that could effectively illustrate the diversity among the germplasm, shortlisted by the International Rice Research Institute under the Generation Challenge Program (http://gramene.org/markers/microsat/50_ssr.html). Another set of 50 SSR markers linked to yield and related traits were selected based on the mapping studies conducted and the results that have been documented hitherto in rice (Zhang et al. 2010). However, there was a perceptible limitation with the latter as they did not provide genomewide coverage. SSR marker assay

Total genomic DNA isolation was carried out using a microextraction method (Prabhu et al. 1998) and assayed with a total of 100 SSR markers. The PCR mixture contained 25–30 ng template DNA, 5 pmol of each primer, 0.05 mM dNTPs, 10× PCR buffer and 0.5 U of Taq DNA polymerase in a reaction volume of 10 μL. Template DNA was initially denatured at 94◦ C for 5 min followed by 35 cycles (1 min denaturation at 94◦ C, 1 min annealing at 55◦ C and 2 min of primer extension at 72◦ C) of PCR amplification followed by final extension of 72◦ C for 7 min. The amplification products were separated on 3.5% metaphor agarose gels and run for 3 h in 1× TAE buffer. DNA fragments were visualized under UV transillumination using Bio-Rad Molecular Imager (GelDocTM XR+ Imager, New Delhi, India). Data analysis

The phenotypic data were subjected to statistical analysis for descriptive purposes, as well as for character association. Since the data contained multicollinear variables, a principal component analysis (PCA) based clustering was performed. The genotypic factor scores corresponding to significant principal components (PCs) were used to perform cluster analysis using pair group distance with Euclidean similarity measure. The dendrogram was constructed using the software package PAST (paleontological statistics software package for education and data analysis, Hammer et al. (2001). Amplicons were scored for their product size for each primer genotype combination. The informativeness of the markers in terms of polymorphism was determined by

Journal of Genetics, Vol. 92, No. 3, December 2013

Genetic diversity in Indian rice germplasm Table 1. Details of 88 genotypes used in the study. Variety (other name)

Parentage

Breeding/source location

1 2 3 4

Abhaya ARS36 Bamleshwari (IET14444) Basmati 370

IGKV, Raipur UAS, Bangalore IGKV, Raipur RRI, Kala Shah Kaku

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52

Basmati 386 Basmati 564 (RR564) BhriguDhan (HPR179) Chandrahasini (IET16800; R979-528-2-1) China 988 Chinikamini Danteshwari (IET15450) Dhanaprasad Govind Hassan Serai Himalaya 1 Himalaya 741 Himalaya 799 (HPU799) Himdhan (RxT42) HPR1068 HPR2143 HUBR2-1 (MalviyaBasmati 1) HUR105 (MalviyaSugandh 105) HUR3022 (MalviyaDhan 2) HUR36 (MalviyaDhan 36) HUR4-3 (MalviyaSugandh 4-3) Indira Sona Indira SugandhitDhan 1 IR24 IR64 Jaiphulla Jaldubi (IET 17153; AR1023) JR201 (Rashmi) JR75 K332 K429 (Kohsaar) Karma Mahsuri Kasturi Kranti Mahamaya MalaviyaDhan Manhar MAS109 MAS25 MAS868 MAS946 MAS946-1 (Sharada) MR219 MR220 NaggarDhan OYR128 OYR69 PalamDhan 957

CR157-392/OR57-21 RP2151-40-1/IR9828-23-1 Pureline selection from local Dehraduni basmati landraces Selection from Pakistan Basmati Sel from Basmati 370 Chucheng/ Daval // Matali Abhaya / Phalguna

CSKHPKV, Palampur OUAT, Bhubaneswar IGKV, Raipur OUAT, Bhubaneswar GBPUAT, Pantnagar CSKHPKV, Palampur CSKHPKV, Palampur CSKHPKV, Palampur CSKHPKV, Palampur CSKHPKV, Palampur CSKHPKV, Palampur CSKHPKV, Palampur BHU, Varanasi. BHU, Varanasi. BHU, Varanasi. BHU, Varanasi. BHU, Varanasi. IGKV, Raipur IGKV, Raipur IRRI, Los Baños IRRI, Los Baños OUAT, Bhubaneswar IGKV, Raipur JNKV, Jabalpur JNKV, Jabalpur SKUAST, Srinagar SKUAST, Srinagar IGKV, Raipur CSKHPKV, Palampur JNKV, Jabalpur IGKV, Raipur BHU, Varanasi GBPUAT, Pantnagar UAS, Bangalore UAS, Bangalore UAS, Bangalore UAS, Bangalore UAS, Bangalore MARDI, Selangor MARDI, Selangor CSKHPKV, Palampur UAS, Bangalore UAS, Bangalore CSKHPKV, Palampur

53 54 55 56 57 58

Pant Dhan 10 Pant Dhan 11 Pant Dhan 12 (IET10955) Pant Dhan 15 Pant Dhan 16 Pant Dhan 18

Introduction from China Landrace Samridhi/ IR8608-298 Landrace IR20/ IR24 Introduction from Iranian Basmati IR8/ Tadukan CR125-42-5/ IR2061-213 IR28/ Shensi Var// IR28 R575/TN1 IR42015-83-3-2-2/ IR9758-K2 PhulPatas 72/ HPU741 HBR92/ Pusa Basmati/ Kasturi Mutant of MPR 7-2 IR 36/ HR 137 Mutant of Mahsuri Mutant of Lanjhi IR58025A/ R710-437-1-1 Madhuri/ Surekha IR8/ IR127-2-2 IR5657-33-2-1/ IR2061-465-1-5-5 Landrace Sel. Surguja local IR36/ JR75 IR20/ L14// BSJ205 Shinei/ Marin11 Shinei/ Ginmsari Mahsuar/ R 296-260 Basmati 370/ CR88-17-1-5 Cross 116/ IR8 Asha/ Kranti Mutant of Mahsuri IR24/ Cauvery IR64/ Azucena IR64/ Azucena IR64/ Azucena IR64/ Azucena IR64/ Azucena MR137/ MR151 MR137/ MR151 ChingSai 25 – – IR32429-122-3-1-2/ IR31851-64-2-3-2 IR32/ Mahsuri// IR28 VL206/ Dagi Govind/ UPR201-1-1 Basmati 370/ Sudari/ Behral/ Muskan 41 BG380/ BG367-4 IR25393-57/ RD23// IR27316-96/// SPRLR77205-3-2/ SPRLR79234-51-2


RRI, Kala Shah Kaku SKUAST, Jammu CSKHPKV, Palampur IGKV, Raipur

GBPUAT, Pantnagar GBPUAT, Pantnagar GBPUAT, Pantnagar GBPUAT, Pantnagar GBPUAT, Pantnagar GBPUAT, Pantnagar

547

Sheel Yadav et al. Table 1 (contd). (contd) No.

Variety (other name)

Parentage

Breeding/source location

59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88

Pant Dhan 19 Pant Dhan 6 Pant SankarDhan 1 Pant SankarDhan 3 Pant SugandhDhan 17 PAU201 Poornima PR111 (IET13576) PR114 PR115 PR116 PR118 PR120 Prasad Punjab Basmati 2 Ranbir Basmati (IET11348) Saanwal Basmati (IET15815) SambhaMahsuri (BPT5204) Samleshwari (IET17455; R1027-2282-2-1) Sarjoo 52 Shyamala SKAU23 (Chenab) SKAU27 (Jhelum) SKAU5 (K39) SonaMahsuri (BPT 3291) SR1 (Shalimar Rice 1) SukaraDhan 1 (HPR1156) Super Basmati Swarna (MTU7029) T23

BG132/ UPRI95-141 IR8608-298-3-1/ IR10179-23 UPR195-178A/ UPR192-133R UPRI95-17A/ UARI93-287 R Pusa Basmati 1/UPRM500 PR103 / PAU1126 Poorva/ IR8608-298 IR54/ PR106 TN1/ Patong 32// PR106*4/// IR8 RP2151-173-1-8/ PR103*3 PR108/// TN1/ Patong 32// PR106*6//// PR108 Pusa 44/ PR110// Pusa 44*3 PAU1196-14-2-5-1-3/ SR817-255 IR747-B-26-3/ IR57948 Basmati 386/ Super Basmati Sel from Basmati 370-90-95 Sel from Basmati 370 GEB24/ TN1// Mashuri R310-37/ R308-6 TN1/ Kashi R60-2713/ R238-6 K21-9-10-1/ IR2053-521-1-2 Jinoku/ IET4444 China 1039/IR580-19-2-3-3 Sona/ Mashuri CH1007/ IET1444 IR32429-122-3-1-2/ IR31868-64-2-3-3-3 IR662/Bamati 320 Vasista / Mahsuri Sel from Kala Sukhdas

GBPUAT, Pantnagar GBPUAT, Pantnagar GBPUAT, Pantnagar GBPUAT, Pantnagar GBPUAT, Pantnagar PAU, Ludhiana JNKV, Jabalpur PAU, Ludhiana PAU, Ludhiana PAU, Ludhiana PAU, Ludhiana PAU, Ludhiana PAU, Ludhiana GBPUAT, Pantnagar PAU, Ludhiana SKUAST, Jammu SKUAST, Jammu ANGRAU, Bapatla IGKV, Raipur NDUAT , Faizabad IGKV, Raipur SKUAST, Srinagar SKUAST, Srinagar SKUAST, Srinagar ANGRAU, Bapatla SKUAST, Srinagar CSKHPKV, Palampur RRI, Kala Shah Kaku APRRI, Maruteru CSKHPKV, Palampur

JNKV, Jawaharlal Nehru Krishi Vishvavidyalaya; OUAT, Orissa University of Agriculture and Technology; BHU, Banaras Hindu University; UAS, University of Agricultural Sciences; PAU, Punjab Agricultural University; SKUAST, Sher-e-Kashmir University of Agricultural Sciences and Technology; RRI, Rice Research Institute; IGKV, Indira Gandhi Krishi Vishwavidyalaya; CSKHPKV, Chaudhary Sarwan Kumar Himachal Pradesh Krishi Vishvavidyalaya; GBPUAT, Govind Ballabh Pant University of Agriculture and Technology; IRRI, International Rice Research Institute; ANGRAU, Acharya NG Ranga Agricultural University; APRRI, Andhra Pradesh Rice Research Institute; MARDI, Malaysian Agricultural Research and Development Institute; NDUAT, Narendra Dev University of Agriculture and Technology; IARI, Indian Agricultural Research Institute

calculating the polymorphism information content (PIC, Botstein et al. 1980) for each SSR locus, according to the formula: k k−1 k 2n 2 − pˆ i 2ˆp2i pˆ 2j , PIC = 1 − 2n − 1 i=1 i=1 j=i+1 where pˆ i was the estimated allele frequencies of k alleles (i = 1 to k) and n, the number of individuals sampled. The allelic data was analysed using the computer package Darwin (ver. 5.0.158) (Perrier and Collet 2006). The dissimilarity coefficients were calculated between genotypes using simple matching coefficient (dij ). 1 ml . L l=1 π L

dij = 1 −

548

Where L is the total number of loci (l = 1 to L), mi is the number of matching alleles for locus l, and π is the ploidy level. The dissimilarity matrix was used for clustering of genotypes, based on unweighted neighbour-joining method. Confidence limits of different clades were tested by bootstrapping 10,000 times to assess the repetitiveness of genotype clustering (Felsenstein 1985). To study relative efficiency of diversity assessment of two sets of molecular markers, relative kinship coefficients between the genotypes were estimated as pair-wise dissimilarity for random, gene-linked and pooled marker data sets. The Euclidean distance matrix from the genotypewise factor loadings for the significant PCs was also used in comparison. A Mantel test was done to test similarities of distance and kinship matrices, with 30,000 permutations for two-tailed


Genetic diversity in Indian rice germplasm test. Mantel test was performed using the software package PASSaGE (Rosenberg and Anderson 2011).

Results Morphological characterization

Evaluation of morphological traits in the rice germplasm lines revealed that wide range of variation existed among the genotypes, for traits such as UGN, SPY, FGN, TLN and TGN showing high coefficients of variation (table 2) and SPF, PNL and DMT showing the lowest. The earliest flowering line was SKAU 5 (79 days) while flowering in Chinikamini took as long as 159 days. The average TLN varied from 8.8 in Super Basmati to 42.6 for BhriguDhan. The SPY was highest for MAS 25 (43.0 g), while it was lowest in Jaldubi (4.5 g). Pant Dhan 19 had the maximum

TGW (31.1 g) whereas it was least for Chinikamini (8.5 g). The SPF of the accessions ranged from 38.2% in HPR 2143 to 95.7% in Mahamaya (table 1 in electronic supplementary material at http://www.ias.ac.in/jgenet/). Pearson’s correlation coefficients between the phenotypic traits are given in table 3. There were some highly correlated characters, such as DMT and DFL (0.99) and FGN and TGN (0.91), which may bias the estimates of diversity pattern due to multicollinearity. Therefore, an independent estimate of variability was performed by a PCA. The active characters with their eigenvalues obtained from the PCA of the correlation matrix are given in table 4. There were four major PCs that accounted for the significant part of total variation (78.3%). The first and the second PCs explained 57.2% of the total variation. The variable contributions to the first two PCs are given in figure 1. It is seen that all traits except for SPF, SPY, FGN and UGN significantly

Table 2. Variability observed for agronomic traits. Trait

Mean (±SE)

Range

CV (%)

Days to 50% flowering (DFL) Days to maturity (DMT) Plant height (cm) (PLH) Number of tillers (TLN) Panicle length (cm) (PNL) Yield/plant (g) (SPY) Test weight (g) (TGW) Total grains per panicle (TGN) Filled grains per panicle (FGN) Unfilled grains per panicle (UGN) Spikelet fertility (%) (SPF)

104.73 (2.11) 134.61 (2.11) 116.60 (2.37) 14.06 (0.49) 26.93 (0.36) 25.29 (0.93) 21.60 (0.43) 172.6 (5.61) 139.88 (5.00) 32.43 (2.33) 81.06 (1.04)

79.0–159.0 109.0–189.0 84.1–177.5 8.8–42.6 18.8–36.3 4.5–43.0 8.5–31.1 72.4–326.8 60.8–292.6 6.4–126.2 38.2–96.7

18.89 14.7 19.07 32.4 12.37 34.32 18.67 30.47 33.55 67.27 12.07

FGN

Table 3. Phenotypic correlation coefficients among the traits.

DMT PLH TLN PNL SPY TGW TGN FGN UGN SPF

DFL

DMT

PLH

TLN

PNL

SPY

TGW

TGN

0.999 0.426 −0.225 0.209 0.106 −0.412 0.608 0.535 0.317 0.005

0.421 −0.219 0.208 0.106 −0.416 0.603 0.531 0.312 0.005

−0.143 0.504 −0.048 −0.180 0.212 0.155 0.169 −0.045

−0.052 0.047 −0.075 −0.174 −0.142 −0.108 0.047

0.088 −0.095 0.230 0.123 0.304 −0.182

0.279 0.243 0.429 −0.323 0.477

−0.446 −0.318 −0.391 0.213

0.909 0.470 −0.018

0.066 0.384

UGN

−0.828

Critical values of Pearson’s correlation coefficient at df=86, 0.2096 (P = 0.05) and 0.2732 (P = 0.01). DFL, days to 50% flowering; DMT, days to maturity; PLH, plant height in cm; TLN, number of tillers per plant; PNL, panicle length in cm; SPY, single plant yield in g; TGW, 1000-grain weight in g; TGN, total grains per panicle; FGN, number of filled grains per panicle; UGN, number of unfilled grains per panicle; SFP, spikelet fertility per cent. Significant values are given in bold. Journal of Genetics, Vol. 92, No. 3, December 2013

549

Sheel Yadav et al. Table 4. Eigenvalues of correlation matrix and related statistics for the agronomic traits. Eigenvalue

Cumulative eigenvalue

Per cent total variance

Cumulative variance %

3.90 2.38 1.28 1.04 0.97 0.65 0.39 0.32 0.06 0.00 0.00

3.90 6.29 7.57 8.61 9.58 10.24 10.63 10.99 10.99 10.99 11.00

35.48 21.68 11.64 9.49 8.85 5.94 3.52 2.87 0.53 0.02 0.01

35.48 57.16 68.79 78.28 87.13 93.06 96.58 99.45 99.98 99.99 100.00

Mashuri and NaggarDhan that had either significant contributions to either PC1 or PC2 or both. Further, the cluster analysis based on the unweighted pair group averaging of Euclidean distances of the factor ladings of all four significant PCs revealed two major clusters at 100% level and four clusters at 80% level (figure 3). The two major clusters correspond to those identified by the two-dimensional scaling between first two PCs. At 80% level, HPR2143 formed a separate cluster, followed by Dhanaprasad and Jaldubi in second cluster. BhirguDhan 1 formed another single member cluster, followed by a large cluster of all the remaining entries. Molecular characterization Random SSR markers: The molecular diversity pattern of ran-

Figure 1. Contribution of the phenotypic traits towards first two PCs.

contributed to the PC1. SPF and SPY had significant contribution to PC2, while UGN and FGN contributed significantly to the both the components. Contributions of genotypes to major PCs are given in table 5. Highest contribution of variability to the first PC was observed for Dhanaprasad (11.1%), while HPR2143 (15.8%) and Jaldubi (10.9%) showed significant contribution towards PC2. Two dimensional scaling of the genotypes by the first two PCs (figure 2) showed two distinct groups of genotypes, first with Dhanaprasad, Jaldubi and HPR2143, and the second with rest of all other genotypes. Among the second and the largest group, there were many significant variants, such as Jaiphulla, Indira Sona, MalaviaDhan, Chinikamini, Abhaya, Mahamaya, Swarna, Samba 550

dom SSR markers showed allelic variation ranging from 1 to 7 (figure 4). Forty-six out of 50 markers were polymorphic. The PIC values of the polymorphic markers ranged from 0.02 (RM 489) to 0.74 (RM 152). The average PIC for these markers was 0.41. The marker RM413 produced a maximum of seven alleles. The average number of polymorphic alleles was 3.63 (table 2 in electronic supplementary material). Clustering of the germplasm based on 46 polymorphic markers produced two major clusters at the dissimilarity coefficient of 0.57, and further into four clusters at dissimilarity coefficient of 0.60 (figure 5a). The first cluster consisted of the landraces, japonica varieties and basmati lines. The second and third cluster consisted primarily of indica lines. Pant Dhan 15 and HPR2143 formed the fourth cluster. The average genetic diversity depicted by these markers was 0.37. It was observed that the germplasm/varieties from Bangalore were the least diverse whereas the varieties from Srinagar exhibited maximum genetic diversity (table 6). SSR Markers for functional genes

Forty-nine SSR markers out of 50 were found to be polymorphic. The average PIC of these markers was 0.48, with the functional marker for Monoculm 1 (Moc1) on chromosome 6


Genetic diversity in Indian rice germplasm Table 5. Contributions of genotypes in percentage to major PCs based on correlations. Genotypes

PC1

PC2

PC3

PC4

Abhaya ARS36 Bamleshwari Basmati370 Basmati386 Basmati564 BhirguDhan1 Chandrahasini China988 Chinikamini Danteshwari Dhanaprasad Govind Hassanserai Himalaya1 Himalaya741 Himalaya799 Himdhan HPR1068 HPR2143 HUBR2-1 HUR105 HUR302 HUR36 HUR4-3 IndiraSona IndiraSugandhitDhan1 IR24 IR64 Jaiphulla Jaldubi JR201 JR75 K332 K429 KarmaMahsuri Kasturi Kranti Mahamaya MalviyaDhan Manhar MAS109 MAS25 MAS868 MAS946 MAS946-1 MR219 MR220 NaggarDhan OYR128 OYR69 PalamDhan957 PantDhan10 PantDhan11 PantDhan12 PantDhan15 PantDhan16 PantDhan18 PantDhan19 PantDhan6 PantSankarDhan1 PantSankarDhan3

0.32 0.01 0.54 0.92 0.29 0.10 3.40 0.13 0.47 5.24 0.67 11.10 1.71 0.33 0.59 1.57 3.62 0.76 2.46 0.01 0.00 0.09 0.03 3.10 0.01 1.48 0.26 0.26 0.22 6.46 6.80 2.26 0.96 2.40 2.76 0.46 0.32 0.13 0.82 4.52 0.26 0.97 0.09 0.44 0.12 0.00 0.22 0.21 3.67 0.01 0.11 1.26 0.01 1.22 0.22 0.50 0.23 0.14 0.13 0.95 0.00 0.01

2.58 1.07 2.67 0.03 0.20 0.01 0.01 0.40 0.18 0.04 0.23 8.67 0.91 1.20 0.19 0.10 0.00 0.04 0.00 15.77 0.21 0.09 0.97 2.58 0.03 1.55 0.30 0.00 0.34 1.11 10.92 0.01 3.61 0.02 0.09 0.34 0.01 0.15 2.80 0.17 1.89 0.17 2.13 0.83 0.11 0.14 2.10 2.24 0.23 0.83 0.29 0.02 2.37 0.00 0.23 0.00 0.01 2.11 0.68 0.10 0.18 0.27

1.24 0.64 0.91 6.39 5.59 7.79 1.57 0.00 2.29 1.41 0.78 0.28 0.19 1.72 2.10 0.67 0.32 0.97 0.18 0.07 0.14 0.77 0.10 0.72 0.27 0.02 0.25 0.24 0.00 3.13 0.08 0.15 0.37 0.09 1.80 1.65 5.18 0.16 1.00 0.55 0.06 1.93 0.51 2.74 0.63 0.19 0.00 0.08 0.12 0.82 1.06 0.36 1.39 0.01 0.72 1.49 0.40 1.85 0.18 1.16 0.00 0.45

0.01 0.00 1.73 0.27 0.20 0.16 18.98 0.28 0.86 1.06 0.25 0.46 0.06 0.33 0.35 0.71 1.10 0.18 0.67 2.24 0.37 1.13 0.07 0.00 0.28 3.01 1.89 0.90 0.36 0.36 1.47 0.49 0.61 1.65 1.65 0.78 0.01 0.50 1.00 0.45 0.23 0.02 8.70 0.03 0.02 0.09 0.04 0.02 1.84 1.89 3.18 0.02 0.88 0.48 0.80 0.65 0.20 0.07 1.88 0.24 0.07 0.00


551

Sheel Yadav et al. Table 5 (contd). (contd) Genotypes

PC1

PantSugandhDhan17 PAU201 Poornima PR111 PR114 PR115 PR116 PR118 PR120 Prasad Punjab Basmati2 Ranbir Basmati Saanwal Basmati Sambha Mahsuri Samleshwari Sarjoo52 Shyamala SKAU23 SKAU27 SKAU5 SonaMahsuri SR1 SukaraDhan1 SuperBasmati Swarna T23

1.78 0.01 0.92 0.03 0.00 0.01 0.35 0.07 0.14 0.18 0.11 0.85 0.34 2.77 0.02 0.12 0.21 1.72 1.25 0.81 0.94 0.75 1.49 0.14 4.67 2.02

PC2 0.16 1.04 0.80 0.68 0.10 0.24 0.01 2.33 0.87 1.53 0.14 0.67 0.17 3.30 0.01 1.78 0.32 0.09 0.54 2.56 1.97 0.45 0.14 0.00 3.53 0.07

PC3 2.04 0.05 0.24 0.62 0.49 0.00 1.04 0.37 0.09 0.10 1.21 0.65 2.18 3.41 0.95 1.88 0.04 0.42 0.05 0.12 6.43 3.28 0.08 1.15 4.31 0.87

Figure 2. Two-dimensional scaling of 88 rice genotypes by PCA.

552


PC4 2.88 0.36 0.31 4.03 0.03 0.68 0.84 1.26 0.08 2.25 0.02 3.28 1.12 0.07 1.71 0.70 4.10 0.00 0.44 0.89 0.38 0.48 0.06 2.22 0.09 3.53

Genetic diversity in Indian rice germplasm

Figure 3. Unweighted pair group average dendrogram of 88 rice genotypes based on Euclidean distances.

Figure 4. A representative gel picture showing allelic diversity for different SSR markers; M, 100-bp ladder; 1–24, rice germplasm lines; A, B, C, D and E, different alleles amplified. Journal of Genetics, Vol. 92, No. 3, December 2013

553

Sheel Yadav et al.

Figure 5. Dendrogram of 88 genotypes colour coded against their location of origin (a) as revealed by random markers (b) as revealed by gene linked/based markers.

Table 6. Genetic dissimilarity values calculated from random SSR markers.

Centre Overall UAS, Bangalore PAU, Ludhiana SKUAS&T, Srinagar JNKVV, Jabalpur OUAT, Bhubaneswar IGKV, Raipur CSHPKV, Palampur GBPUAT, Pantnagar BHU, Varanasi

Average GD 0.37 0.23 0.47 0.50 0.30 0.32 0.28 0.45 0.39 0.35

Maximum genetic distance Between genotypes Jaiphulla MAS- 868 Punjab Basmati-2 SR1 JR201 Dhanaprasad Indira Sugandh Dhan 1 Palam Dhan 957 Pant Dhan 10 HUBR-2-1

Minimum genetic distance GD

Bhirgu Dhan OYR 128 PAU201 K332 Poornima Jaiphulla Samleshwari Bhirgu Dhan P. Sankar Dhan 1 HUR - 302

0.77 0.28 0.70 0.60 0.35 0.51 0.24 0.58 0.40 0.32

Between genotypes MR 219 MAS-946 Punjab Basmati-2 SKAU 5 JR75 Dhanaprasad Abhaya Himalaya 1 Govind HUR-36

MR 220 MAS-946-1 Basmati 386 SKAU 27 Kranti Chinikamini Indira Sona Himalaya 741 Manhar HUR-302

GD 0.01 0.01 0.04 0.07 0.14 0.19 0.18 0.10 0.19 0.10

GD, genetic distance between two genotypes as determined by the genetic dissimilarity matrix.

for number of tillers, exhibiting the highest PIC of 0.73. The average number of polymorphic alleles amplified by these markers was 3.67 with RM252 producing a maximum of six alleles (table 3 in electronic supplementary material). 554

The lines were grouped into two major clusters at the dissimilarity coefficient of 0.55 and further into four clusters at a dissimilarity coefficient of 0.58. The first cluster consisted of the landraces, basmati lines and japonica varieties.


Genetic diversity in Indian rice germplasm Table 7. Genetic dissimilarity values calculated from SSR markers for functional genes. Maximum genetic distance

Average GD

Centre Overall UAS, Bangalore PAU, Ludhiana SKUAS&T, Srinagar JNKVV, Jabalpur OUAT, Bhubaneswar IGKV, Raipur CSHPKV, Palampur GBPUAT, Pantnagar BHU, Varanasi

0.45 0.20 0.50 0.58 0.40 0.38 0.40 0.54 0.45 0.40

Between genotypes PR118 MAS- 946-1 PR118 SKAU23 JR201 Dhanaprasad I. Sugandh Dhan 1 Palam Dhan 957 Pant Dhan 15 HUR-4-3

K429 OYR69 Super Basmati K429 Poornima Jaiphulla Jaldubi Naggar Dhan Pant Dhan 19 HUR 36

The second cluster consisted of varieties from Kashmir and Palampur. The third cluster comprised of varieties from UAS and PAU. The fourth cluster consisted of Pant Dhan varieties (figure 5b). These markers revealed a higher magnitude of genetic diversity among the germplasm lines. The average genetic diversity was 0.45, with the germplasm from Bangalore being the least diverse and the varieties Srinagar and Palampur being the most diverse (table 7).

Correlation between genotypic and phenotypic data

The correlation value was calculated from (88 × 88) distance matrices of phenotypic and genotypic datasets using the Mantel test. A moderate but significant correlation (0.29) was observed between kinship matrices of trait-linked markers and phenotype distance matrix. This was higher than the correlation between the kinship matrix of random markers and phenotypic distance matrix (table 8). This indicates that trait-linked markers are more accurate and useful in studying the diversity existing in rice germplasm.

Table 8. Mantel correlations rM (upper diagonal) and two-tailed P values for 30000 permutations (Lower diagonal) between phenotype distance and kinship coefficient matrices for 88 rice genotypes using different sets of marker data. Estimated Mantel correlations (rM)

Kfunc Krand Kpooled Dphen

Kfunc

Krand

Kpooled

Dphen

– 0.000 0.000 0.000

0.749 – 0.000 0.006

0.948 0.921 – 0.003

0.290 0.189 0.261 –

Kfunc , kinship matrix of trait-linked markers; Krand , kinship matrix of random (diversity panel) markers; Kpooled , kinship matrix of pooled markers, Dphen , phenotype distance matrix.

Minimum genetic distance GD 0.81 0.31 0.76 0.75 0.50 0.59 0.37 0.71 0.54 0.29

Between genotypes MR 219 MAS-946 Basmati 386 K332 JR75 Dhanaprasad Danteshwari Himalaya 1 Pant Sankar Dhan 3 HUR-4-3

MR 220 MAS-946-1 P. Basmati 2 K429 Kranti Chinikamini Chandrahasini Himalaya 741 Pant Sankar Dhan 1 HUR-302

GD 0.02 0.04 0.02 0.09 0.27 0.19 0.26 0.19 0.12 0.25

Discussion The Indian subcontinent has a very rich diversity in rice germplasm which includes landraces, wild Oryza species, natural hybrids between the cultivars and wild relatives, and the germplasm resources generated in the breeding programmes (Rai 1999). The primary objective of the investigation was to study the efficacy of markers in differentiating the rice germplasm from diverse source, by comparing clustering based on random SSR markers vs the trait linked/gene based markers. A total of 88 rice germplasm accessions, representing the gamut of rice diversity from land races to elite cultivars were sourced from different parts of India. These were then characterized for morphological and molecular variability. To study genetic diversity, it is important to know which type of markers can precisely predict the actual or functional variability existing in the germplasm of a crop. In this study, we compared the data generated using random SSR markers vs the SSR markers which have been reported to be gene based or linked to genes related to yield and yield component traits in rice (Zhang et al. 2010). It has been considered by earlier researchers that the diversity established by using random markers, which are essentially derived from the functionally inactive or noncoding regions of the genome might not illustrate the ‘functional’ diversity. Therefore, it is important to use markers which could unravel the diversity in functionally related coding regions of the genome to identify the variability. The results based on the analysis of diversity of germplasm with two different sets of SSRs markers showed that germplasm originating from specific centre clustered together irrespective of the marker sets reflecting the narrow genetic base in the rice breeding programmes at these centres. Therefore, there is an urgent need for broadening the genetic base of breeding programme to improve the per se performance of varieties and the degree of heterosis. The varieties procured from Bangalore exhibited least genetic diversity since most of these genotypes are RILs


555

Sheel Yadav et al. derived from the IR 64/Azucena cross. The high level of diversity exhibited by the varieties from Srinagar can be attributed to the inclusion of both indica and japonica varieties in the collection and also because these varieties are well adapted to different ecologies in Jammu and Kashmir, namely intermediate zone (1300–2000 m msl) of Jammu and Kashmir (varieties including SR1, SKAU 5, SKAU 23 and SKAU 27) and varieties suited for high altitude areas (> 2000 m msl) of Jammu and Kashmir (K 332 and K 429). Both the set of marker data were able to discriminate the landraces from the other varieties. The presence of high genetic similarity among landraces of a common geographic region has been described in earlier studies (Ram et al. 2007). However, this rich genetic variability which has accumulated over a period of time is yet to be exploited by plant breeders. Analysis with both sets of markers classified the germplasm into different groups of landraces, indica varieties, japonica varieties and Basmati groups. These subpopulations seem to have been created during the course of rice domestication (Kovach et al. 2007). The indica and Basmati group were found to be the most polymorphic. The Basmati varieties were placed closer to japonica group than indica group. Basmati varieties have conventionally been classified as members of the indica varietal group essentially because of their long and slender grain shape. Previous studies have also, however, reported that Basmati lines are more closely related to japonica than indica group, with the FST value for tropical japonica-aromatic reported to be 0.23 and the FST for indica-aromatic to be 0.39 (Garris et al. 2005). The conspicuous differentiation between Basmati and nonBasmati varieties is indicative of divergence and subsequent independent evolution of the former through artificial selection (Nagaraju et al. 2002). This kind of classification has been well documented as Glaszmann (1987) classification of Asian rice germplasm. He proposed a six group classification of Asian rice germplasm, with two major groups, group I and group VI corresponding to the indica and japonica group of varieties respectively. Group V represented the Basmati group from Iran, Pakistan, India and Nepal. MR 219 and MR 220 were the least dissimilar lines. These two varieties are sister lines released in Malaysia. Jaiphulla, a landrace procured from Orissa was distant from the other landraces from the same and different centres. There is little phenotypic or ancestry information about the landraces that can justify this observation. However, previous studies (Singh et al. 2010) have pointed out that Jaiphulla is medium in height, early maturing and high yielding variety as opposed to the other landraces which are mostly tall, medium to long duration type with long panicle, having moderate to high yield. The average genetic diversity at genomic loci as assessed by trait-linked markers was found to be more than that revealed by random markers. This indicates that there exists greater variability at the functional regions of the genome which is rightly established by the functional markers in the selected germplasm. There is an argument that genic 556

SSRs display a low level of polymorphism vis-à-vis genomic SSRs, since SSRs located in the coding regions are under strong selection pressure and therefore accumulate few mutations (Li et al. 2004; Varshney et al. 2005). This was demonstrated in many crops such as rice (Cho et al. 2000), sugarcane (Cordeiro et al. 2001), wheat (Gupta et al. 2003) and chickpea (Choudhary et al. 2009). However, in the present study we have observed slightly higher average PIC values for the trait-linked SSRs (0.48) than the random SSRs (0.41). Apparently, we have used rice microsatellite (RM) markers linked to QTLs affecting yield and yield components (trait-linked markers), but not truly functional gene based markers. Although, they are linked to genomic regions that affect agronomic traits, but not located within the coding region. Therefore, it is unreasonable to expect such markers to fall on conserved regions all the time and to have lesser polymorphism than the random markers. Further, when markers linked to known QTLs are tested in a germplasm with large variation for the respective traits (yield and its components in this study), it is likely that we are bringing in more variable alleles together with respect to the target trait. Therefore, allele richness of any QTL linked marker would have slight advantage over that of a random marker resulting into relatively higher average PIC value for such specific QTL linked markers than random markers. There was a moderately significant correlation between the phenotypic diversity and kinship as revealed by trait-linked markers. The use of such markers will therefore enable plant breeders to rightly estimate the diversity at heterotic loci. However, the low correlation value confirms role of environment in influencing the phenotype and how classification based on only phenotypic diversity can be ambiguous. Therefore, it can be concluded that both the phenotypic diversity and genetic diversity at functional regions of the genome, should be targeted and utilized in heterotic rice breeding programmes, for predicting hybrid performance. Acknowledgements This study was carried out under the project ‘Marker Assisted Creation of Heterotic Pools and Diversification of Male Sterility and Fertility Restoration System for Hybrid Seed Production in Rice’ funded by the Council of Scientific and Industrial Research (CSIR), Government of India, under The New Millennium Indian Technology Leadership Initiative (NMITLI).

References Botstein D., White R. L., Skolnick M. and Davis R. W. 1980 Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am. J. Hum. Genet. 32, 314–331. Cho Y. G., Ishii T., Temnykh S., Chen X., Lipovich L., McCouch S. R. et al. 2000 Diversity of microsatellites derived from genomic libraries and GenBank sequences in rice (Oryza sativa L.). Theor. Appl. Genet. 100, 713–722. Choudhary S., Sethy N. K., Shokeen B. and Bhatia S. 2009 Development of chickpea EST-SSR markers and analysis of allelic


Genetic diversity in Indian rice germplasm variation across related species. Theor. Appl. Genet. 118, 591– 608. Cordeiro G. M., Casu R., McIntyre C. L., Manners J. M. and Henry R. J. 2001 Microsatellite markers from sugarcane (Saccharum spp.) ESTs cross transferable to erianthus and sorghum. Plant Sci. 160, 1115–1123. Felsenstein J. 1985 Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39, 783–791. Ford-Lloyd B. V., Jackson M. T. and Newbury H. J. 1997 Molecular markers and the management of genetic resources in seed genebanks: a case study of rice. In Biotechnology and plant genetic resources: conservation and use (ed. J. A. Callow, B. V. Ford-Lloyd and H. J. Newbury). CAB International, Wallingford, UK. Garris A. J., Tai T. H., Coburn J., Kresovich S. and McCouch S. 2005 Genetic structure and diversity in Oryza sativa L. Genetics 169, 1631–1638. Glaszmann J. C. 1987 Isozymes and classification of Asian rice varieties. Theor. Appl. Genet. 74, 21–30. Gupta P. K., Rustgi S., Sharma S., Singh R., Kumar N. and Balyan H. S. 2003 Transferable EST-SSR markers for the study of polymorphism and genetic diversity in bread wheat. Mol. Genet. Genomics 270, 315–323. Hammer Ø., Harper D. A. T. and Ryan P. D. 2001 PAST: Paleontological statistics software package for education and data analysis. Palaeontol. Electron. 4, 9. Kovach M. J., Sweeney M. T. and McCouch S. R. 2007 New insights into the history of rice domestication. Trends Genet. 23, 578–587. Kumar S., Bisht I. S. and Bhatt K. V. 2010 Assessment of genetic diversity among rice (Oryza sativa L.) landrace populations under traditional production using microsatellite (SSR) markers. Electron. J. Plant Breed. 1, 474–483. Li Y.C., Korol A. B., Fahima T. and Nevo E. 2004 Microsatellites within genes: structure, function, and evolution. Mol. Biol. Evol. 21, 991–1007. McCouch S. R., Chen X. and Cho Y. G. 2002 Sequence divergence of rice microsatellites in Oryza and other plant species. Mol. Genet. Genomics 268, 331–343. Nagaraju J., Kathirvel M., Kumar R. R., Siddiq E. A. and Hasnain S. E. 2002 Genetic analysis of traditional and evolved basmati and non-basmati rice varieties by using fluorescence based ISSR PCR and SSR markers. Proc. Natl. Acad. Sci. USA 99, 5836–5841. Perrier X. and Collet J. P. 2006 DARwin software (http://darwin. cirad.fr/). Prabhu K. V., Somers D. J., Rakow G. and Gugel R. K. 1998 Molecular markers linked to white rust resistance in mustard Brassica juncea. Theor. Appl. Genet. 97, 865–870. Rai M. 1999 Rice germplasm evaluation and enhancement in India: issues, status, options, and future plan of action. Proceedings

of the International Symposium on Rice Germplasm Evaluation and Enhancement (ed. J. N. Rutger, J. F. Robinson and R. H. Dilday), pp. 83–91, Arkansas Agricultural Experiment Station, Fayetteville, Arkansas. Ram S. G., Thiruvengadam V. and Vinod K. K. 2007 Genetic diversity among cultivars, landraces and wild relatives of rice as revealed by microsatellite markers. J. Appl. Genet. 48, 337– 345. Rosenberg M. S. and Anderson C. D. 2011 PASSaGE: pattern analysis, spatial statistics and geographic exegesis, version 2. Methods Ecol. Evol. 2, 229–232. Saini N., Jain N., Jain S. and Jain R. K. 2004 Assessment of genetic diversity within and among basmati and non-basmati rice varieties using AFLP, ISSR and SSR markers. Euphytica 140, 133–146. Singh B., Mishra M. K. and Naik R. K. 2010 Genetic diversity among some traditional aromatic rice (Oryza sativa L) varieties of orissa. Indian J. Agric. Res. 44, 141–145. Sivaranjani A. K. P., Pandey M. K., Sudharshan I., Kumar G. R., Madhav M. S., Sundaram R. M. et al. 2010 Assessment of genetic diversity among basmati and non-basmati aromatic rices of India using SSR markers. Curr. Sci. 99, 221–226. Song Z. P., Xu X., Wang B., Chen J. K. and Lu B. R. 2003 Genetic diversity in the northernmost Oryza rufipogon populations estimated by SSR markers. Theor. Appl. Genet. 107, 1492–1499. Sundaram R. M., Naveenkumar B., Biradar S. K., Balachandran S. M., Mishra B., Ahmed M. I. et al. 2008 Identification of informative SSR markers capable of distinguishing hybrid rice parental lines and their utilization in seed purity assessment. Euphytica 163, 215–224. Temnykh S., Park D. W., Ayres N., Cartinhour S., Hauck N., Lipovich L. et al. 2000 Mapping and genome organization of microsatellite sequences in rice (Oryza sativa L.). Theor. Appl. Genet. 100, 697–712. Vanniarajan C., Vinod K. K. and Pereira A. 2012 Molecular evaluation of genetic diversity and association studies in rice (Oryza sativa L.). J. Genet. 91, 9–19. Varshney R. K., Graner A. and Sorrells M. E. 2005 Genic microsatellite markers in plants: features and applications. Trends Biotechnol. 23, 48–55. Virk P. S., Newbury J. H., Bryan G. J., Jackson M. T. and Ford-Lloyd B. V. 2000 Are mapped or anonymous markers more useful for assessing genetic diversity? Theor. Appl. Genet. 100, 607–613. Zhang T., Ni X-L., Jiang K-F., Deng H-F., He Q., Yang QH. et al. 2010 Relationship between heterosis and parental genetic distance based on molecular markers for functional genes related to yield traits in rice. Rice Sci. 17, 288– 295.

Received 8 May 2013, in revised form 8 August 2013; accepted 19 August 2013 Published on the Web: 11 December 2013


557