24(2)-15.pdf1.20MB

• MATERIALS AND METHODS
• Materials
• DNA extraction and SSR analysis
• Data analysis
• Population genetic structure analysis
• RESULT AND DISCUSSION
• Genetic diversity in Meedon rice
• Cluster analysis
• Population genetic structure
• CONCLUSION
• ACKNOWLEDGEMENT

Rice (Oryza sativa L.) which feeds more than half of the world population has possessed wide genetic variation. Myanmar is one of the centers of genetic diversity of rice and varietal group which spreads along the Himalayas from Iran to Myanmar consists of very diverse varieties (Glaszmann & J. C. 1986). Many local varieties have been grown in Myanmar for centuries. Among the local varieties, Meedon rice varieties are important for local adaptability, grain quality, premium price and market availability. Because of temporal variability and spatial heterogeneity, Meedon group rice varieties in rainfed lowland areas may be diverse. However, information on molecular diversity among Meedon rice germplasm is limited.

DNA marker technology has provided an efficient tool to facilitate plant genetic resource conservation and management(Ni et al. 2002). In rice, molecular markers have been used to identify accessions (Virk et al. 1995 and Olufowote et al. 1997), to determine the genetic structure and pattern of diversity for cultivars of interest (Zhang Q. F. et al. 1992, Yang et al. 1994, Mackill D. J. 1995 and Akagi et al. 1997), and to optimize the assembly of core collections (Schoen et al. 1995). Compared to morphological analysis, molecular markers can reveal differences among accessions at the DNA level and thus provide a more direct, reliable, and efficient tool for germplasm conservation and management. Among several markers available for evaluating the extent of genetic variation in rice, SSR or microsatellites are codominant markers, their map positions on the rice genome are well known, and PCR-based markers that are both technically efficient and cost-effective to use (Chen et al. 1997 and Temnykh et al. 2001). Moreover, microsatellite markers detect a significantly higher degree of polymorphism than RFLP in rice (Wu et al. 1993 and Yang et al. 1994), and are especially suitable for evaluating genetic diversity among closely related rice cultivars (Akagi et al. 1997). The objectives of this study were to assess genetic diversity and to analyze population structure of Meedon rice germplasm conserved in Myanmar Seed Bank using SSR markers.

MATERIALS AND METHODS

Materials

In this study, 154 accessions of Meedon rice germplasm conserved in Myanmar Seed Bank, Department of Agricultural Research (DAR), Yezin, Nay Pyi Taw, Myanmar were used. Meedon is one of the five rice varietal groups in Myanmar, and most of the varieties in Meedon group are local quality rice.

DNA extraction and SSR analysis

Molecular analysis using SSR markers was carried out at the laboratory of the Genebank (Global Seed Vault), Suwon, Republic of Korea in 2008. DNA was extracted from the dehulled rice seeds. Ten seeds per accession were grinded, and DNA was extracted. The results of Ni et al. (2002) suggested that a relatively small number of SSR markers could be employed to evaluate the genetic diversity, to identify different accessions and to reveal the genetic relationship among accessions. In this study, nine SSR markers distributed on 8 different chromosomes were used (Table 1). The original sources and motifs for these markers can be found in the RiceGenes database (http://ars-genome.cornell.edu/rice). The size of polymorphic polymerase chain reaction (PCR) products was measured accurately by following the M13 tail PCR method of Schuelke & M. (2000). Amplification reactions were carried out in a total volume of 25 μl containing 300 ng template DNA, 10X PCR buffer, 0.5 mM of each dNTP, 1U Taq DNA polymerase, 10 pmole of each reserve and fluorescent-labeled M13 primer and 10 pmole of the forward primer with M13 tail at its 5’ end. Conditions of the PCR amplification were as follows: (i) an initial denaturation step of 15 min at 95˚C, (ii) 32 cycles each at 95˚C (20 sec), 52˚C (40 sec), 72˚C (1 min) followed by 15 cycles of 95˚C (30 sec), 52˚C (45 sec), 72˚C (1 min), and (iii) a final extension step for 15 min at 72˚C. Microsatellite alleles were resolved on 3130 x1Genetic Analyzer (Applied Biosystems) using GENESCAN 3.7 software and sized precisely against 6-carboxy-X-rhodamine (ROX) molecular size standards using GENOTYPER 3.7 software (Applied Biosystems).

Table 1. Diversity statistics at nine SSR loci among 154 accessions of Meedon rice germplasm.

Data analysis

Basic statistics were calculated using the genetic analysis package, Power Marker version 1.31 (Liu et al. 2005) for diversity measurements among 154 rice accessions at each SSR locus including the number of alleles, allele frequency, rare alleles (accession-specific alleles), observed heterozygosity (HO), expected heterozygosity (HE), and polymorphic information content (PIC). The term polymorphic information content (PIC) was originally introduced into human genetics by Botstein et al. (1980). It refers to the value of a marker for detecting polymorphism within a population, depending on the number of detectable alleles and the distribution of their frequency (Ni et al. 2002). The PIC value of a marker could be calculated according to a simplified version after Anderson et al. (1993).

Where P_ij is the frequency of the j^th allele for the i^th marker, and summed over n allele. The present (1) and absent (0) of alleles for each marker were recorded for all accessions and then converted to a genetic distance matrix. Genetic distances, GD (Nei et al. 1979) between two entries were computed as

 GD = 1− [2N_ij / (N_i + N_j)]

Where N_ij is the number of shared bands and N_i and N_j are the total number of bands for entries i and j. A cluster dendrogram was constructed based on these genetic distance matrices by the unweighted pair group method using arithmetic averages (UPGMA) algorithm method with the help of SAHN-clustering and TREE programs from NTSYSpc. V.2.0 (Rohlf, 1998).

Population genetic structure analysis

For the analysis of population structure and identification of ancestral and hybrid forms, the model-based software program, STRUCTURE (Pritchard et al. 2000) was used. In this model, a number of populations (K) are assumed to be present, each of which is characterized by a set of allele frequencies at each locus. Individuals in the sample are assigned to populations (groups), or jointly to more populations if their genotypes indicate that they are admixed. All loci are assumed to be independent, and K population is assumed to follow Hardy-Weinberg equilibrium. The posterior probabilities were estimated using a Markov Chain Monte Carlo (MCMC) method. The MCMC chains were run at burn-in period lengths 100,000 at fixed iterations of 5 for each fixed number of population (K, 1-10). The final results were based on a burn-in period length of 100,000 and 5 iterations of this chain using a model allowing for admixed and correlated allele frequencies. An individual having more than 75% membership or more than 0.750 of its genome fraction value was assigned to a group.

RESULT AND DISCUSSION

Genetic diversity in Meedon rice

The variability at each SSR locus for 154 accessions of Meedon rice germplasm was measured in terms of the number of observed alleles, observed heterozygosity, expected heterozygosity and polymorphic information content. A summary of measures of genetic variability at nine loci in the test germplasm was given in Table 1. All loci were producing a total of 86 alleles among 154 rice accessions. The loci varied in the number of detected alleles from 5 to 17, with a mean of 9.6 alleles per locus which was relatively high and all the loci used in this study were found to be polymorphic.

The sizes of alleles ranged from 112 bp to 219 bp. The locus RM 263 produced the highest number of alleles (17 alleles). The lowest number of alleles (5) was observed at RM 258. The major allele frequency ranged from 0.3322 (RM 202) to 0.8028 (RM 309) with a mean value of 0.5943. The observed heterozygosity (H_o) values ranged between 0 (RM 263 and RM 258) and 0.3041 (RM 10) with a mean value of 0.0745, whereas that of expected heterozygosity (H_E) ranged between 0.3355 (RM 309) and 0.7827 (RM 202) with a mean value of 0.5774. The polymorphic information content (PIC) ranged between 0.3096 (RM 309) and 0.7528 (RM 202) with the mean values of 0.5496. The locus RM 202 was the most informative marker, whereas the locus RM 309 was the least informative marker among nine loci. The result of mean values of H_E and PIC, 0.5774 and 0.5496 indicated the relatively high genetic variation among the accessions, and also indicated the great difference of alleles among loci, ranging from 5 to 17. However, no correlation was observed between the number of alleles and H_E and PIC in this study. This suggests the degree of polymorphism detected by these markers could not be correlated with H_E and PIC values. Certain loci producing similar number of alleles were not significantly different in their H_E or PIC values. For example, 6 alleles were detected at each of the two loci, RM168 and RM309, 10 alleles were at RM225 and RM202, and 12 alleles were at RM6 and RM335, but no significant difference was detected in their H_E or PIC values. Of the 86 alleles found, 53 alleles (61.6%) were rare (frequencies < 0.05), 26 alleles (30.2%) were common with frequencies between 0.05 and 0.5, and 7 alleles (8.2%) were abundant (frequencies > 0.5) alleles showing uneven distribution of the frequencies of rare, common and abundant alleles (Table 2 and Table 3). Marker RM263 provided the highest number of 14 rare alleles, followed by RM335 (8), RM6 (7), RM225 and RM202 (each 6), RM10 and RM309 (each 4), and RM168 (3). RM258, the lowest provider of rare allele, provided only one (Table 2).

Table 2. Frequency distribution of different alleles among 154 accessions of Meedon rice germplasm at nine SSR markers.

Table 3. Distribution of alleles according to frequency range.

High frequency of rare alleles was observed among rice accessions. It is important to include rare alleles for maximizing the genetic variation in Meedon rice germplasm and to utilize them in varietal improvement. Among rare alleles, a total of 35 unique (accession-specific) alleles were observed at eight different SSR loci (Table 4), of which, eight were observed at locus RM10, and each five unique alleles were observed at three loci (RM6, RM225 and RM202). Each three unique alleles were observed at four loci (RM263, RM168, RM335 and RM309). None of the accession-specific allele, however, was detected at RM258. 

Table 4. Number of unique (accession-specific) alleles at eight SSR loci among Meedon rice germplasm.

With regard to accession-specific allele, 5 accessions of Paw San subgroup were specific to four loci (RM263, RM6, RM225 and RM202) while 3 accessions of Paw San Bay Kyar to three loci (RM10, RM6 and RM225), 2 accessions of Paw San Taung Pyan to two loci (RM10 and RM309), 4accessions of Nga Kywe to three loci (RM263, RM309 and RM225), and 21 accessions of Non-Meedon rice to seven loci (RM263, RM168, RM335, RM10, RM309, RM6 and RM202). However, none of the accession from Nga Kywe Taung Pyan was specific to all loci (Table 4). Unique alleles are also important because they may be diagnostic of a particular type of genotype for identification for breeding purposes. The occurrence of a relatively high number of rare and unique alleles in Paw San and Nga Kywe indicated the potentiality as a reservoir of novel alleles for varietal improvement for Meedon rice.

In comparative study of genetic diversity among the tested germplasm showed that the accessions of non-Meedon subgroup (42 accessions) possessed the highest genetic diversity (H_E = 0.7479) followed by Nga Kywe of 20 accessions (H_E = 0.3873), Paw San Taung Pyan of 10 accessions (H_E = 0.2496), Paw San of 32 accessions (H_E = 0.2417), Paw San Bay Kyar of 27 accessions (H_E = 0.1604) and Nga Kywe Taung Pyan of 23 accessions (H_E = 0.0863). 

Cluster analysis

Based on the analysis of nine SSR loci among 154 accessions of Meedon rice germplasm, genetic distance values for all accessions of rice germplasm were calculated. The genetic distance values ranged from 0 to 1.0, and these values were used to construct the UPGMA dendrogram. As results of NTSYS cluster analysis based on nine SSR markers, the test accessions were clustered into two main groups, Group I and Group II (Fig. 1). Forty two accessions (26.6%) together with check varieties, Khao Dawk Mali, Milyang and Nipponbare were clustered into Group II which represented the non-Meedon rice accessions. Majority of the accessions (112 accessions, 73.4%) were mainly clustered into Group I which could be further divided into two subgroups, Group I (a) and Group I (b). The sub-cluster Group I (b) was consisted of the non-Meedon rice, and sub-cluster Group I (a) was composed with Meedon rice accessions. When sub-cluster Group I (a) was observed, Paw San and Paw San Bay Kyar were clustered together in the same subcluster. Some accessions of Nga Kywe were separately clustered in different lower sub-cluster. Accessions of Paw San Taung Pyan and Nga Kywe Taung Pyan were grouped in the same sub-cluster. Meedon rice accessions were mainly clustered into Group I whereas non-Medon rice accessions were clustered in Group II indicating the different genotypes by the same names. When subgroup Group I (a) was observed, Paw San, Paw San Bay Kyar and Nga Kywe were clustered together in the same subgroup. Some accessions of Nga Kywe were separately clustered into different subgroups. Among Meedon rice, Paw San and Paw San Bay Kyar were genetically closer relationship than other subgroups. Paw San Taung Pyan and Nga Kywe Taung Pyan possess long sterile lemma grouped together in the same sub-cluster, and interestingly, they were not discriminated from each other. They could be further discriminated by morphological characters. Some accessions which were not discriminated from each other seemed to be duplicate accessions, and it may be repeated sampling of the same genotypes from a certain site by different collecting missions.

Fig. 1. NTSYS dendrogram among 154 accessions of Meedon group rice germplasm generated by cluster analysis of UPGMA method, based on 9 SSR markers.

Population genetic structure

Meedon rice germplasm was assessed to analyze the genetic structure and detection of heterozygous forms among 154 accessions using a model-based approach (Prit-chard et al., 2000). Five runs of STRUCTURE were done by setting the number of population (K) from 1 to 10. For each run, burn-in time and replication number were both set to 100,000. K=2 was found to converge well and showed comparable or higher likelihoods than K = 3 ~ 10 among runs of the program (Fig. 2). An individual having more than 75% membership or more than 0.750 of its genome fraction value was assigned to a group. Among the germplasm, 148 accessions (96.1%) shared > 0.750 membership with one of the two genetic groups and these accessions were classified as members of that genetic group or population. In addition to the accessions that were clearly assigned to a single genetic group, 6 accessions (3.9%) which possess genome fraction value < 0.750 were categorized as admixtures with varying levels of membership shared among the two populations. The first genetic group (Population I), the second genetic group (Population II) and the admixed group consist of 105 accessions (68.2%), 43 accessions (27.9%) and 6 accessions (3.9%) (Table 5).

Fig. 2. Magnitude of ΔK as a function of K, and the modal value of this distribution is the true K or the uppermost level of structure.

Table 5. Distribution of Meedon rice accessions among model-based genetic groups using microsatellite data.

The first genetic group Population I included 105 accessions with majority of them (103 accessions, 98.1%) having genome fraction value > 0.9 membership of this population and showed their memberships between 0.827 and 0.997. Population II was represented by 43 accessions, all of which showed their memberships between 0.772 and 0.997, and majority of them (37 accessions, 86.1%) having genome fraction value > 0.900 membership of this population. Further analysis of distribution patterns of all the accessions belonging to different populations revealed some interesting observations. The distribution of rice accessions between the two populations was different. Rice accessions belong to Population I were Paw San, Paw San Bay Kyar, Paw San Taung Pyan, Nga Kywe, and Nga Kywe Taung Pyan (93.8%, 96.3%, 90.0%, 80.0% and 95.5% of the total accessions of individual subgroups). On the contrary, accessions belong to Population II were predominant to non-Meedon rice (88.4% of this subgroup). Among 154 accessions, only six accessions (3.9%) were categorized as admixed group.

According to the reports of Evanno et al. (2005) and Petit et al. (2001), population structure determination was usually based upon geographical origin of samples or phenotypes (Gwag & J. G., 2008). The model-based method

utilizes a Bayesian clustering approach to probabilistically assign individuals to populations based on their genotypes and attempts to find population structure in which the population is in Hardy-Weinberg equilibrium (Barkley et al. 2006). Population genetics deal with the variation of allele frequency between and within populations. With this analysis, 148 accessions (96.1%) out of 154 accessions were classified into two distinct genetic populations shared more than 0.750 of its genome fraction value (75% membership) with one of the two genetic groups. The distribution of rice accessions between genetic groups is quite different according to varietal subgroups. The entries belong to admixed population (hybrid forms of accessions) probably had a complex breeding history involving intercrossing and introgression between germplasm from diverse background, overlaid with strong selection pressure for agronomic and quality characteristics (Gwag & J. G., 2008). 

Nine loci of Microsatellite have proved to be useful markers for detecting genetic variation in Meedon rice for germplasm management and utilization purposes. The data could also be used to establish core collection for Meedon rice germplasm. By observing the result, the three genetic groups consist of 105 accessions (68.2%), 43 accessions (27.9%) and 6 accessions (3.9%) indicating the inclusion of different genotypes (non-Meedon rice accessions) by means of the same names to Meedon rice, which was about one-third of the germplasm (49 accessions, 31.8%), and it was a large proportion. The result also pointed out that this approach of analysis using small number of microsatellite markers as suggested by Ni et al. (2002) is possible for discriminating between Meedon and non-Meedon rice accessions.

CONCLUSION

Meedon, local quality rice in Myanmar is important for its grain quality, local adaptability, premium price and market availability. However, little is known about its genetic diversity and population structure. In this study, a total of 86 alleles were detected among 154 accessions with an average of 9.6 alleles per locus, which was relatively high. All the mirosatellite loci were found to be polymorphic, and there were relatively high genetic variations among the accessions. The distribution of the frequencies of rare, common and abundant alleles was uneven. However, the occurrence of a considerably high number of rare and unique alleles indicated their potentiality as a reservoir of rare genotypes.

Population genetics deal with the variation of allele frequency between and within populations. Two distinct genetic population groups with an admixture could be classified. Two major groups, Meedon and non-Meedon were clearly classified. Among Meedon rice, Paw San and Paw San Bay Kyar were genetically closer relationship than other subgroups. Paw San Taung Pyan and Nga Kywe Taung Pyan with long sterile lemma were grouped together. Many accessions in Meedon group were genetically similar that seemed to be duplicate accessions. They showed close genetic relationships, and this observation is indicative of the conservation of duplicate accessions in Meedon rice germplasm. In this case, the combined use of molecular markers and morphological characters may allow further correct discrimination. On the other hand, different genotypes accounted by the same name may also exist in this germplasm since the accessions shared the same name are classified into different groups. Moreover, it could be concluded that indica, japonica or intermediate types and aromatic accessions are existing in the germplasm.

In summary, nine SSR loci have proved to be useful markers for detecting genetic variation in Meedon rice for germplasm management and utilization purposes. The small number of SSR markers is possible for discriminating Meedon and non-Meedon rice. The data are useful for establishing core collection for Meedon rice. The occurrence of a considerably high number of rare and unique alleles in Meedon germplasm indicates their potentiality as a reservoir of rare genotypes for use. Unique alleles are also important because they may be diagnostic of a particular type of genotype for identification for plant varietal protection and for breeding purpose. 

ACKNOWLEDGEMENT

The first author wishes to thank Ministry of Agriculture and Irrigation, Myanmar for permission to study this program. Special thanks are due to the Regional Office for Asia, Pacific and Oceania of Bioversity International, Malaysia and Rural Development Administration (RDA), Republic of Korea for support to attend the 6^th  International Symposium of Rice Functional Genomics, November 10- 12, Jeju, to study genebank work, and to conduct molecular analysis in RDA genebank. The authors would like to thank the staff of RDA genebank for their helps for DNA analysis of Meedon rice germplasm (PJ00862302).