Characterization of Selected Zambian Rice (Oryza sativa) Accessions using Simple Sequence Repeat (SSR) Markers

The study was divided into the following phases:
 
Phase I: Experimental location
 
The experiments were conducted at the Zambia Agricultural Research Institute (ZARI), Mount Makulu Research Station in Chilanga, Lusaka Province that lies on latitude 15.547745 and longitude 28.249392.

Phase II: Processing of plant materials

Thirty rice accessions were collected from the major rice-growing regions in Zambia and curated at the National Plant Genetic Resource Centre (National Gene Bank) at ZARI in Chilanga. Fig 1 elucidates the processing of rice seedlings. Table 1 elucidates details and passport information of the accessions.

Fig 1: Processing of plant material.

Phase III: Genomic DNA extraction

All chemicals used for DNA extraction and nucleic acid amplification were of analytical grade and procured from HiMedia Laboratories, Mumbai-India. Genomic DNA was isolated from young leaves of 10-day old plants following the mini-modified DNA extraction method (Zheng et al., 1995), (Fig 2).

Fig 2: Genomic DNA Extraction.

Phase IV: Polymerase Chain Reaction (PCR)

PCR amplification was carried out using 10 SSR markers selected from 7 of the 12 chromosomes listed in Table 1.  Lyophilized primers were reconstituted in nuclease-free water to a stock solution of 100mM from which working solutions of 50ng/ìl were made. Template DNA extracted from 10 independent plants from each of the 30 accessions was amplified using the selected SSR primers.

Table 1: Rice SSR markers (primer pairs).

SSR markers reveal polymorphism in local rice varieties
 
The size of most intensely amplified fragments was determined by comparing the migration distance of amplified fragments relative to 100 bp molecular weight DNA marker. polymorphic information content (PIC) values were calculated for each SSR locus based on Anderson et al., 1993. The amplified bands were scored for each SSR primer pair based on the presence or absence of bands, generating a binary data matrix of 1, 2 and 0 for each marker system. The matrices were then analyzed using the Genstat statistical package 18th edition. Dendrograms displaying relationships among the 30 rice accessions were constructed using the Genstat statistical package (Fig 4).

PCR fragment analysis reveals distinct product band patterns

The markers were found to be polymorphic among the rice accessions and generated reproducible and informative allelic profiles. Fragment profile as a measure of polymorphism was observed in DNA samples amplified using RM5 and RM 13 for instance. For RM5, a DNA fragment of 110 base pairs was observed in all accessions with examples of ZM84338, ZM8316, ZM8328 and ZM8321 shown in Fig 3a. In the case of ZM8338 DNA extracted from seven independent plants amplified using RM5 showed variation with reference to PCR products. Three of these plants (36, 37 and 40) had amplification products whereas four (34, 35, 38 and 39) showed no amplification products. In the case of ZM8316 DNA from 10 independent plant samples was amplified to produce the expected fragment of 110 base pairs. In the case of ZM8328, the RM5 marker was able to amplify all the ten samples except for two (51 and 57), Fig 3a.

Fig 3a: DNA profile of four accessions with SSR marker RM 5.

Fig 3b: DNA profile of four accessions with SSR marker RM 13.

Fig 3c: DNA profile of four accessions with SSR marker RM 225.

Fig 4: A genstat clustering dendrogram showing the genetic relationships among 30 accessions on the alleles detected by 9 microsatellite markers.

Clustering of rice accessions

Cluster analysis was done to group the accessions by constructing a dendrogram illustrated in Fig 4 based on allelic information gathered from the genotype-marker interaction. The Genetic Similarity (GS) index ranged from a minimum of 80% to a maximum of 100%. The Genstat-dendrogram clustered the 30 genotypes into two major clusters (A and B). Cluster A comprised of only one accession, ZM8321 commonly called Ndelema from Chama district in Muchinga province. Cluster-B consisted of the 29 accessions which were subdivided into 10 sub-clusters that showed more genetic relatedness among the accessions with ZM8295 (Supa-Samfya district; Luapula Province) and ZM8313 (Kaputa District, Northern Province) demonstrating 100% relatedness.

The success of a crop improvement program depends on the magnitude of genetic variability and the extent to which the desirable characters are heritable (Ravi et al., 2003). The assessment of genetic diversity has been demonstrated to be essential in establishing relationships among different cultivars (Sivaranjani et al., 2010). Molecular characterization could reveal their phylogeny and this information would be quite useful in utilizing the rice germplasm in genetic improvement of the existing rice accessions. This research aimed at characterizing selected rice accessions in Zambia at the genetic level to create baseline information that could be useful in the genetic improvement of the crop. This suggested that the observed absence of amplification products were not polymerase chain reaction artifacts which could be interpreted as failed reactions. On the contrary, it showed that the seed accessions from which the DNA was isolated did not carry the targeted alleles. This also indicated that the observed presence or absence of PCR product bands revealed polymorphism between or even within what are currently considered to be pure accessions. The other possibility was that the PCR products were absent in some of the reactions due to the absence of the target DNA sequence in those rice accession, thereby contributing to the diversity of the accessions. The amplification reactions were repeated with similar results, ruling out reaction failure as the cause of the absence of band fragments. Matin et al., (2012) proposed PIC as an accurate indicator of allele diversity and frequency among accessions. The PIC values for the tested markers in this work ranged between 0.2 and 0.8. The SSR marker RM225 revealed the highest PIC value also backed by results from agarose gel electrophoresis analysis of PCR products obtained with this particular marker, Fig 1c in harmony with Matin et al., (2012) findings. The frequencies of null alleles were not included in the calculation of PIC values for each SSR locus. Cluster analysis via a dendrogram revealed that most of the accessions were closely related and clustered into two major groups designated A and B. Only one accession (ZM8321) appeared to be distinct and distantly related to all the other 29 accessions. Cluster B contained the rest of the 29 accessions which were subdivided into 8 sub-clusters. Of the 29, accessions ZM8295 from Samfya District and ZM8313 from Kaputa District showed 100% similarity (cluster K) meaning that they are the same accession despite the different names by which they are identified. Furthermore, despite the different names and accession numbers assigned to these two accessions, visual examinations revealed morphological similarities which confirm the genetic information obtained from the study. These were closely related to ZM8346 (Chandege-white and brown- Chama, Muchinga) than the rest of the accessions. MLK020 and MLK003 showed relatedness at almost 99% and are from the same agro-ecological zones but are known by different district’s local names. ZM 8338 (Senga from Chama, Muchinga) and ZM 8316 (Malawi faya from Nyimba, Eastern) in cluster E showed 97.5% relatedness, which was a very high level of relatedness despite being known by different names in two closely located provinces as well as an ecological zone in Zambia. Nonetheless, clusters F, G, H, J and K were closely related but quite distant from clusters A, B, C, D and E.

From the clustering, it was observed that a number of accessions were very similar to each other; for example, clusters E, H, J and K; this could mean these accessions paired together are the same despite being known by different local names and coming from distantly spaced growing regions. This information would help in decongesting the gene bank and also reduce the possibility of breeders dealing with the same accessions while thinking they are distantly related. For all the sub-clusters, few accessions where showed genetic differences with a different local name, agro-ecological zone from which it was collected and outstanding morphological distinctiveness. Some few accessions showed the exact opposite. They were observed to be quite distant in terms of relatedness with other accessions despite them being known by the same local name in different ecological zones e.g. Supa. This may mean that most of the accessions analyzed in this study were heterogeneous even within those collected from the same agro-ecological zones and believed to be homogeneous and treated as being the same accession. Therefore, it can be said that there is genetic distinctness among accessions selected in this study, but with a higher level of relatedness. It can also be seen that genotypes from far distinct cluster groups when crossed can give rise to a genetically wide range of off spring, giving a diversity of genotypes to select from.

The study also revealed that although two variants could be present in the same cluster and share only about 75% similarity at the genetic level. By convention, it is assumed that the two variants should have been the same and must be genetically similar, but our genetic diversity analysis contradicts this fact. Such a kind of cultivar identity crisis is often witnessed throughout the world (Matin et al., 2012). Such discrepancies emphasize the importance of molecular characterization of rice germplasm prior to their deployment in varietal improvement. The results revealed that all the primers showed distinct polymorphisms among the accessions studied, indicating the robust nature of microsatellites in revealing polymorphism. Therefore microsatellite marker-based molecular fingerprinting could serve as a potential basis in the identification of genetically distant genotypes as well as in sorting of duplication for morphologically close accession.