Assessment of Genetic Variation in Soybean (Glycine max) Accessions from International Gene Pools Using RAPD Markers: Comparison with the ISSR System

Soybean (Glycine max) is one of the most important 
crops in the world in terms of total production and usage. It is also among the 
least diverse species. The main objectives of the present study were to 1) 
assess the level of genetic variation among soybean (G. max) accessions from 
different countries using Random Amplified Polymorphic DNA (RAPD) markers and 
2) compare Inter Simple Sequence Repeats (ISSR) and RAPD marker systems in 
detecting polymorphic loci in soybeans (G. max). Genomic DNAs from 108 soybeans 
(G. max) accessions from 11 different gene pools were analyzed using 
several ISSR and RAPD primers. The average level of polymorphic loci detected 
with the RAPD primers was 35%. The soybean accessions from the China, 
Netherlands, and Canada gene pools were the least genetically variable with 25%, 
26%, and 30% of polymorphic loci, respectively. Accessions from Hungary (43%) 
and France (48%) showed the highest level of polymorphism based on the RAPD 
analysis. Overall, RAPD data revealed that the accessions from different 
countries are closely related with 64% genetic distance values below 0.40. The 
levels of polymorphic loci detected with the RAPD and ISSR marker systems were 
in general moderate and similar even if they target different regions of the 
genome. A combination of different marker systems that include RAPD/ISSR, 
microsatellites (SSR), and SNPs should provide the most accurate information on 
genetic variation of soybean (G. max) accessions.


Introduction
Soybean (Glycine max) is the most important grain legume in the world in terms of total production. Diversities in many crops such as soybean have been documented based on morphological and agronomical traits [1] [2]. Such characters are strongly influenced by environmental factors and the developmental stage of the plant. Since the early nineties, molecular markers have been developed as alternative methods and they are extremely effective in population genetics studies [3].
Protein or enzyme variation can be used to study genetic diversity of crop germplasm. However, the limited number of isozymes of proteins and enzymes can limit their usefulness. Polymorphic DNA markers can provide an ideal alternative method for evaluating genetic diversity in soybean germplasm [4] [5].
Random Amplified Polymorphic DNA (RAPD) markers have been shown to be a simple and effective means to evaluate variability in crop [9] [10] [11] [12]. Based on principal component analysis of RAPD data on 35 soybean lines, Thompson et al. [13] established a core set of RAPD primers with high polymorphism in soybean. These 35 core RAPD primers have been used in other studies for genetic diversity analysis in soybean [14]. They evaluated 18 U.S. soybean ancestors and 17 selected accessions from the USDA Soybean Germplasm Collection. The clusters defined by the RAPD data corresponded to known pedigrees, origins and maturity groups.
The first demonstration of simple sequence repeat (SSR) allelic variation and heritability in a plant species was in soybean, and SSRs have been shown to be highly polymorphic in soybean [15] [16]. Brown-Guedira et al. [14] observed a higher level of genetic diversity with the SSR system compared to RAPD markers. Akkaya et al. [15] employed SSRs to evaluate the diversity of 43 ancestral and commercial cultivars representing the U.S. gene pool. They determined that, in general, SSRs with AT core motifs are most polymorphic in soybean, followed by those with ATT cores. Rongwen et al. [17] detected 11 to 26 alleles at each of seven SSR loci in a diverse sample of soybean genotypes including U.S. cultivars, and introductions of G. max and G. soja, including Chinese landraces. Recent studies showed that SSR markers are still widely used to assess soybean genetic diversity [18] [19].
Inter Simple Sequence Repeats (ISSR) is a different marker system that has been successfully applied to genetic analysis of plants. The ISSR method provides an alternative choice to other systems for obtaining highly reproducible markers without any necessity for prior sequence information for various genet-ic analyses [12] [20] [21] [22]. ISSR method takes advantage of the ubiquitously distributed SSRs in the eukaryotic genomes. Because of those abundant and rapidly evolving SSR regions, ISSR amplification has the potential of revealing larger numbers of polymorphic fragments per primer than any other marker system used such. As the PCR reaction amplify the sequence between two SSRs, the PCR products generated reveal multilocus profiles which could be revealed on agarose or polyacrylamide gels.
The objectives of the present study were to 1) assess genetic variation in soybean accessions from 11 countries using RAPD markers and 2) compare the levels of polymorphic loci generated with the RAPD and ISRR marker systems.

Genetic Materials
G. max varieties were provided by the Plant Gene Resources of Canada (PGRC).
They were from 11 countries including Canada, China, Russia, Germany, Hungary, France, Netherlands, Sweden, South Korea, Japan, and Poland were used for the present study (Table 1). Seeds from these varieties were placed in clear Petri dishes lined with two layers of wet filter paper and kept in a growth chamber for a period of 14 to 21 days. The germination conditions consisted in a repeating cycle: 16 hours of daylight at 30˚C and 8 hours of darkness at 20˚C. Once the seedlings reached 5 to 10 cm, leaf samples were collected, frozen in liquid nitrogen and stored at −80˚ until the DNA was extracted.

DNA Extraction and Amplification
Total DNA was extracted from fresh frozen leaf material using the CTAB extraction protocol as described by Boyd et al. [23] and Moraefi et al. [22]. After extraction, DNA was stored in a freezer at −20˚C.
Twelve RAPD primers were chosen for preliminary amplification of DNA K. K. Nkongolo et al. from 108 G. max varieties. Five of these primers were selected for this genetic variation investigation (Table 2). To compare RAPD and ISSR markers systems, five ISSR primers that were previously evaluated were also used for DNA amplification [24]. They are described in Table 2. All primers were standardized to a 25 ng/ml stock solution. DNA amplification was performed using the procedure described by Nkongolo et al. [20] and Boyd et al. [23]. PCR (Polymerase Chain The amplified DNA was separated using a 2% agarose gel in 0.5X TrisBorate-EDTA buffer (TBE) containing 0.5 µg/ml ethidium bromide. The gels were documented using a Bio-Rad Chemidoc XRS system and analyzed with Discovery Series Quantity One 1D Analysis Software.

ISSR and RAPD Analysis
ISSR and RAPD primers that amplified consistent profiles across the populations were selected for the final analysis. ISSR and RAPD amplification products from each accession from the eleven countries were scored using POPGENE version 1.32 [17]. The presence or absence of fragments was scored as 1 or 0 for each band, in order to determine variation between accessions (within and between countries). Genetic distances were generated using FreeTree [25] [26].

RAPD Analysis
All DNA samples were tested to assess their quality. They all showed a large molecular weight band, indicating that they were not degraded and were deemed suitable for PCR amplification. A total of 12 RAPD primers were screened. Five primers were selected for further study based on their amplification and reproducibility to analyze the DNA samples from all the targeted countries. These primers include OPA 11, Pinus 23, UBC 377, UBC186, and Grasse 8 (Table 3).  (Table 4). Overall, there were no significant differences among primers for the detection of polymorphism using all G. max accessions.
The genetic distance values based on RAPD data are described in Table 5.
Accessions from Canada and Netherland were the most genetically closely related with accessions from Russia and South Korea being the most distant (      Detailed analysis of ISSR analysis generated with the same primer used in this study has been described elsewhere. For the present study, the level of polymorphism generated with ISSR and RAPD primers were compared. Data are summarized in figure 19 and 20. Overall, with the exception of data with Chinese accessions, they were no significant difference between ISSR and RAPD polymorphism data (Figure 3). This was confirmed when the accessions from all the countries were combined to compare ISSR and RAPD data (Figure 4).

Discussion
Assessment of genetic diversity in a crop species is a prerequisite to its improvement and helps to generate genetically diversified breeding populations.
Considerable variation has been recorded for morphological, physiological and agronomic traits in G. max crops [12].
The average level of genetic variation detected with the five RAPD primers was 35%. This level of polymorphism is lower than other studies [11] [14].   RAPD usually uses a 10 bp arbitrary primer. Although the sequences are arbitrary chosen, two basic criteria must be met: at least 50% Guanine-Cytosine content and the absence of palindromic sequences [32]. Primers as short as 5 nucleotides give more complex banding patterns requiring more sophisticated electrophoretic and staining procedures (acrylamide gels and silver staining). RAPD analysis results in the amplification of one locus and two kinds of polymorphism: the band may be present or absent, and the brightness of the band may be different. Band intensity differences could be due to low copy number or relative sequence abundance (Devos and Gale [33] and may serve to distinguish homozygote dominant individuals from heterozygotes, as less bright bands are expected for the latter. Ellsworth et al. [34] indicated that the fact that fainter bands are generally robust, varying degrees of primer mismatch may account for band intensity differences. As the source of the band intensity difference is uncertain most studies disregard scoring differences in band intensity [32]. The origin of the ISSR amplification products is known to be from the sequences between two simple-sequence repeat (also known as Microsatellite) primer sites where length variation does not necessarily reflect simple-sequence length polymorphism [35]. Microsatellite loci are dispersed throughout the genome and are hypervariable because of DNA slippage (Semagn et al., 2006).
ISSR marker system accesses variation in the numerous micro-satellite regions dispersed throughout the genome (Semagn et al., 2006) and circumvents the challenge of characterizing individual loci that other molecular approaches require. ISSR involves amplification of regions between adjacent, inversely oriented microsatellites, using a simple sequence repeat (SSR) motif containing primers anchored at 3' or 5' end by two or four arbitrary, often degenerate nucleotides [35]. Microsatellites are very short (usually 10 -20 bp) stretches of DNA that are hypervariable, expressed as different variants within populations and among different species. They are characterized by mono-, di-or tri-nucleotide repeats (AA, AG, CAG respectively) that have 4 -10 units side by side. ISSR marker system is based on the use of 15 -20 bp primers designed to be complementary to microsatellite sequences found throughout Eukaryotic genomes. Therefore, this PCR based technique involves the amplification of DNA segments present between two identical microsatellites that are oriented in opposite directions [36]. ISSRs specifically target the di-and tri-nucleotide repeats a type of microsatellite that is characteristic of the nuclear genome (mono nucleotide are found in chloroplast genome [37].
Most often ISSR detects more polymorphisms than RAPD primers because of the high levels of variability in microsatellite loci. The discrepancy between variations revealed by RAPD and ISSR result from different targeted genomic areas, which undergo a different evolutionary process due to selection forces [12] [31] K. K. Nkongolo et al. [38]. Different genetic information is generated when RAPD and ISSR molecular marker techniques are used to assess the inter-specific and intra-specific variability. The level of variation detected with each system greatly depends on the primer used therefore making comparisons for the levels of polymorphism generated with ISSR and RAPD marker systems inappropriate. In addition, both markers systems are cost efficient being PCR based with results from ISSR analyses more repeatable from lab to lab due to longer primers used compared to RAPD [20] [22] [23].
Among other molecular marker systems, microsatellites and AFLP have been widely used to assess genetic diversity among and within populations [3] [38] [39]. Microsatellites amplification is expected to produce a single marker since the microsatellite primers target a single locus [3]. This means that many reactions would be needed to properly determine a population's genetic variability.
AFLP has been gaining popularity over microsatellites in these types of studies. This is in part because AFLP assay is equivalent to the amplification of several microsatellite primers [40]. But, ISSR and RAPD primers are easier to use than AFLP and microsatellites [29] [40] [41].

Conclusion
The main objectives of the present study were to 1) assess the level of genetic variation among 108 soybean accessions from different countries using RAPD markers and 2) compare ISSR and RAPD marker systems in detecting polymorphic loci in soybeans. The average level of polymorphic loci detected with the RAPD primers was 35%. Overall, RAPD data revealed that the accessions from different countries are closely related with 64% genetic distance values below 0.40. The levels of polymorphism detected with the RAPD and ISSR marker systems were moderate and similar even if they target different region of the soybean genome. A combination of different marker systems that include RAPD/ISSR, SSR, and SNPs should provide the most accurate information on genetic variation of soybean accessions.