• Open Access

Comparative analysis of genetic diversity of sesame (Sesamum indicum L.) from Vietnam and Cambodia using agro-morphological and molecular markers


Toan Duc Pham, Department of Plant breeding and Biotechnology, Swedish University of Agricultural Sciences (SLU), PO Box 101, SE–23053 Alnarp, Sweden. E-mail: phamductoan@hcmuaf.edu.vn


The purpose of this study was to comparatively analyze the genetic diversity of sesame (Sesamum indicum L.) using agro-morphological and molecular markers. Twelve sesame populations collected from three regions in Cambodia and Vietnam were used in this study. A high genetic variation was revealed both by agro-morphological and RAPD markers within and among the 12 sesame populations. The range of agro-morphological trait based average taxonomic distance among populations (0.02 to 0.47) was wider than that of RAPD based genetic distance (0.06 to 0.27). The mean distance revealed by agro-morphological markers (0.23) and RAPD markers (0.22) was similar. RAPD based analysis revealed a relatively higher genetic diversity in populations from South Vietnam as compared to the other two regions. Interestingly, populations from this region also had higher values for yield related traits such as number of capsules per plant, number of seeds per capsule, and seed yield per plant suggesting positive correlation between the extent of genetic variation within population and yield related traits in sesame. A highly significant positive correlation (r = 0.88, P < 0.001) was found between agro-morphological and RAPD markers in estimating the genetic distance between populations. Both methods suggested the existence of a substantial amount of genetic diversity both in the Vietnamese and Cambodian populations. Although both agro-morphological and RAPD markers were found to be useful in genetic diversity analysis in sesame, their combined use would give superior results.

Sesame (Sesamum indicum L., Pedaliaceae) is the most commonly cultivated edible oil crop species out of over 30 species in the genus Sesamum (Nayar and Mehra 1970; Kobayashi et al. 1990). The crop has been cultivated in various ecological regions of Vietnam and Cambodia for hundreds of years. Information on genetic diversity and relationships among populations is important for plant breeding programs as it helps to select the right genetic material to be used (Ganesh and Thangavelu 1995). Genetic diversity in crop species can be determined by using the agro-morphological as well as biochemical and molecular markers (Koornneef 1990; Reiter et al. 1993; Liu 1997; Geleta et al. 2007, 2008). Studies on sesame genetic diversity and divergence have been mainly based on agro-morphological traits. Several of these agro-morphological trait based studies have found a high genetic diversity in sesame populations (Bedigian and Harlan 1986; Ganesh and Thangavelu 1995; Bisht et al. 1998; Arriel et al. 2007). PCR-based techniques such as amplified fragment length polymorphism (AFLP), simple sequence repeats (SSR), inter simple sequence repeats (ISSR) and random amplified polymorphic DNA (RAPD) also have been widely used in genetic diversity studies in sesame (Bhat et al. 1999; Kim et al. 2002; Ercan et al. 2004; Laurentin and Karlovsky 2006; Salazar et al. 2006; Pham et al. 2009). Some of these techniques, such as RAPD do not require prior knowledge of DNA sequence and therefore arbitrary primers can be used.

The combined use of agro-morphological and molecular markers is the best choice to characterize germplasm as it gives the opportunity to comparatively analyze the phenotypes from field experiments with molecular phenotypes and genotypes from laboratory studies. Comparison of different methods in genetic studies provides researchers and plant breeders with more information in the screening and selection process. For example, comparative analysis of genetic diversity using agro-morphological and RAPD markers have been reported in Trifolium pratense L. (Greene et al. 2004), Momordica charantia L. (Dey et al. 2006), Andrographis paniculata (Burm. f.) Nees (Lattoo et al. 2008). In the present study, agro-morphological and RAPD markers were used for comparative analyses of the genetic diversity of Vietnamese and Cambodian sesame populations in order to generate information for the selection of divergent genotypes for breeding and for designing effective conservation strategies for sesame genetic resources in these countries.


Agro-morphological traits

Twelve sesame landrace populations collected from various regions in Vietnam and Cambodia were used for this study with North Vietnam, South Vietnam and Cambodia represented by four populations each (Fig. 1). Out of the 12 populations used in this study, one population from North Vietnam, all populations from South Vietnam and one population from Cambodia are brown-seeded whereas the remaining six populations are white-seeded. The codes for the populations used in this study correspond to the initials of the name of their collection sites (Fig. 1). The field experiment for agro-morphological study was carried out at the Cuu Long Rice Research Institute, Can Tho, South Vietnam (10°07′15N, 105°35′01E). During the study period, the temperature and rainfall ranged from 25.8–28.1°C and from 19–116 mm, respectively, with relative humidity of 80.2%. The field layout was a randomized complete block design with three replications. Data was collected from five random plants from each replicate of each population for nine agro-morphological traits. These traits are 50% flowering, 50% maturity, plant height, number of capsules per plant, number of seeds per capsule, 100-seed weight, yield per plant, internode length and capsule length (Table 1). Minitab statistical package (ver. 15.0) was used for analysis of variance (ANOVA) and cluster analysis.

Figure 1.

Map of Vietnam and Cambodia showing collection sites of sesame populations used in this study.

Table 1.  List of RAPD primers used and agro-morphological traits considered in this study.
Primer namePrimer sequence (5′– 3′)% PNBH'spHtTrait nameTrait descriptions
  1. aprimer K2 was modified from primer OPU-10 (Operon Biotechnologies); %P = percentage of polymorphic loci; NB = no. of bands; Hsp= species diversity; Ht= total gene diversity.

OPU-05TTGGCGGCCT85.770.540.36Days to 50% floweringno. of days to 50% plants having the first flower open from date of planting
K2aACCTCGCCAC62.580.520.34Days to 50% maturityno. of days to 50% plants having the first capsule filled seeds up from date of planting
OPU-18GAGGTCCACA85.7140.540.36Plant heightplant height in cm (mean of five random plants)
OPA-02TGCCGAGCTG71.4140.530.35Capsules plant−1no. of capsules per plant (mean of five plants)
OPA-03AGTCAGCCAC61.5130.500.32Seeds capsule−1no. of seeds per capsule (mean of seeds from five capsules per plant)
OPA-09GGGTAACGCC77.790.530.35100-seed weight100-seed weight in gram
OPA-18AGGTGACCGT50.0120.540.36Yield plant−1yield per plant in gram
OPM-06CTGGGCAACT100.070.490.33Internode lengthinternode length in cm
OPL-07AGGCGGGAAC76.9130.450.29Capsule lengthcapsule length in cm
Total  107    

Molecular markers

DNA extraction. The twelve populations used for agro-morphological study were further analyzed using random amplified polymorphic DNA (RAPD) molecular maker technique. Ten plants from each population were sampled for DNA extraction. DNA was extracted from leaves of three to four weeks old seedlings using a protocol described in Warwick and Gugel (2003) with minor modification described in Pham et al. (2009). DNA quality and concentration was measured using Nanodrop® ND-1000 spectrophotometer (Saveen Werner, Sweden).

PCR and electrophoresis. DNA amplification was performed in a volume of 25 µl containing 50 ng DNA template, 1 × PCR reaction buffer (10 mM Tris-HCl, pH 8.3 and 50 mM KCl), 0.2 mM dNTPs, 3 mM MgCl2, 1 U Taq DNA polymerase (Sigma), 0.2 µM RAPD primers (Operon Biotechnologies) following the protocol of Williams et al. (1990). The PCR reaction was carried out using a GeneAMP PCR system 9700 thermocycler (Applied Biosystems) with the following temperature profiles: 5 min denaturation at 94°C followed by 40 cycles of 1 min denaturation at 94°C, 1.5 min primer annealing at 33–40°C, and 3 min primer extension at 72°C. The cycles were followed by a final 10 min primer extension at 72°C. The annealing temperature was changed based on Tm values of each RAPD primer. The electrophoresis of PCR products was carried out on 1% agarose gel in 1 × TAE buffer (40 mM Tris-acetate, pH 8.0, 1 mM EDTA) at 80 V and 100 mA for three h. PstI digested λ DNA ladder with fragment size between 150 bp to 20 kb was used as a molecular size marker. After the electrophoresis, gel staining, visualization and photographing were done as described in Pham et al. (2009).

Data scoring and analysis

The RAPD band profiles were treated as dominant markers and each locus was considered as a bi-allelic locus with one amplifiable allele and one null-allele. Data was scored as 1 for the presence and 0 for the absence of a DNA band for each locus across the 120 individuals. The molecular size of each fragment was determined based on the molecular size markers. The population genetic analyses software, POPGENE ver. 1.31 (Yeh and Boyle 1997) was used for Shannon diversity index H’ (Shannon and Weaver 1949) and Nei's gene diversity estimate H (Nei 1973). The modifications to Nei's gene diversity proposed by Lynch and Milligan (1994) for dominant markers resulted in negligible differences with the results obtained using POPGENE and thus the results from POPGENE is presented in this paper. Shannon diversity index and Nei's gene diversity were estimated as inline image In pi and inline image, respectively, where k is the number of bands produced with the respective primer and pi is the frequency of ith fragment. The NTSYSpc program (Rohlf 2000) was used for cluster analysis. Free Tree-freeware program (Pavlicek et al. 1999) was used for bootstrap analysis whereas the TreeView (Win32) 1.6.6 program (Page 1996) was used to view the trees.


Agro-morphological analysis

The summary of the results of the nine agro-morphological traits recorded from the 12 sesame populations is presented in Table 2. Analysis of variance (ANOVA) revealed a significant variation among the 12 populations (P ≤ 0.05) for six of the nine traits. 100-seed weight, number of seeds capsule−1 and capsule length were not significantly different among populations. The distance between the 12 populations ranged from 0.02 to 0.47 with a mean of 0.23 (Fig. 2). The agro-morphological trait based cluster analysis grouped the 12 populations into three groups (two clusters and one solitary population). Cluster I comprised five brown-seeded populations (TN, AG, VINH, DT, SKD) being separated from the other populations at a mean distance of 0.18. Cluster III comprised six white-seeded populations (HD, EKD, ND, KPC, TH, KPT) being separated from the other populations at a mean distance of 0.21 whereas the solitary brown-seeded population (CT) was separated at a mean distance 0.23. The brown-seeded populations in cluster I showed higher mean values for most of the traits investigated especially for yield plant−1 (24.2 gram), plant height (139.3 cm) and number of capsules plant−1 (99); 50% days to flowering (41) and 50% days to maturity (53) in comparison with populations in other groups. On the other hand, these populations had low mean 100-seed weight (0.27). The population CT showed a moderate yield plant−1 but interestingly high 100-seed weight (0.31), and was appeared to be early flowering (36 days to 50% flowering) and early maturing (48 days to 50% maturity).

Table 2.  Mean values for nine agro-morphological traits for each populations and across the whole populations.
Population codeSeed color50% flowering (DAP)a50% maturity (DAP)aPlant height (cm)Capsules per plant(capsules)Seeds per capsule (seeds)100-seed weight (g)Yield per plant (g)Internode length (cm)Capsule length (cm)
  1. adays after planting; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001; NSnot significant; SE = standard error.

THwhite4050118.3 ± 1.655 ± 7.884 ± 1.80.291 ± 0.0113.67 ± 1.77.33 ± 0.62.97 ± 0.32
NDwhite3949120.0 ± 2.170 ± 1.284 ± 1.20.287 ± 0.0317.00 ± 1.68.00 ± 0.12.70 ± 0.05
HDwhite4051124.3 ± 4.370 ± 6.485 ± 0.30.281 ± 0.0216.83 ± 1.68.00 ± 0.22.70 ± 0.20
VINHbrown4052131.0 ± 6.692 ± 14.488 ± 4.10.29 ± 0.0123.67 ± 3.99.00 ± 1.03.13 ± 0.03
TNbrown4153136.7 ± 7.2112 ± 4.388 ± 5.90.266 ± 0.0126.33 ± 1.19.00 ± 1.02.63 ± 0.08
AGbrown4253136.7 ± 6.798 ± 15.890 ± 4.60.285 ± 0.0125.33 ± 4.39.00 ± 1.02.73 ± 0.13
DTbrown4254148.3 ± 4.499 ± 10.789 ± 6.10.266 ± 0.0223.33 ± 1.69.67 ± 0.32.63 ± 0.03
CTbrown364893.3 ± 3.362 ± 22.684 ± 3.20.31 ± 0.0216.00 ± 5.76.33 ± 0.43.07 ± 0.40
SKDbrown4253144.0 ± 5.892 ± 7.791 ± 4.60.265 ± 0.0122.50 ± 2.39.67 ± 0.92.70 ± 0.10
EKDwhite3949121.7 ± 4.469 ± 10.485 ± 0.40.281 ± 0.0116.67 ± 1.97.67 ± 0.32.67 ± 0.13
KPCwhite3849111.7 ± 6.060 ± 2.484 ± 0.70.285 ± 0.0114.50 ± 2.87.67 ± 0.72.60 ±0.11
KPTwhite3951123.3 ± 10.144 ± 2.185 ± 3.10.301 ± 0.0111.67 ± 1.67.67 ± 0.42.93 ± 0.18
Mean 4051125.877860.28418.968.252.79
Range 36–4248–5493–14844–11284–910.26–0.3111.7–26.36.3–9.62.6–3.13
SE 0.5970.8063.98310.4603.0020.0162.3550.5290.162
F-value 7.35(***)6.11(***)14.16(***)4.03(**)0.75(NS)0.72(NS)4.46(***)3.68(**)1.29(NS)
Figure 2.

Agro-morphological data based dendrogram generated using OBS Cluster Observation for the 12 sesame populations.

Molecular marker analysis

The ten RAPD primers used in this study produced 107 fragments across the 12 populations, of which 80 were polymorphic. The number of fragments per primer varied from seven to fourteen with an average of 10.7 fragments per primer. The percent polymorphism per primer ranged from 50% (OPA-18) to 100% (OPM-06) with an average of 74.1% (Table 1). The estimates of the overall Shannnon diversity (H'sp) and gene diversity (Ht) were 0.51 and 0.33, respectively. Genetic distance among populations ranged from 0.06 and 0.27 and averaged 0.22.

The genetic diversity of each population was estimated using both Shannon diversity and Nei's gene diversity as H'loci and HGD, respectively, which is the mean value across the whole loci. The within population genetic variation was the lowest in population KPT (%P = 23.4%, H'loci= 0.14, HGD= 0.09) and the highest in population CT (%P = 71%, H'loci= 0.41, HGD= 0.28). The average genetic variation within the 12 populations studied was 42.8% (%P), 0.25 (H'loci) and 0.17 (HGD) (Table 3).

Table 3.  Country/region of origin, percentage of polymorphism, mean Shannon diversity and mean Nei's gene diversity for the 12 populations.
IDPopulations codeCountry/region% polymorphismShannon diversity H'lociNei's diversity HGD
  1. Values in bold in each column represent the highest and lowest values of the corresponding parameter.

1THNorth Vietnam29.90.180.13
2NDNorth Vietnam29.90.180.13
3HDNorth Vietnam48.60.260.20
4VINHNorth Vietnam34.60.200.14
5TNSouth Vietnam42.30.250.17
6AGSouth Vietnam28.90.160.11
7DTSouth Vietnam56.10.350.24
8CTSouth Vietnam71.00.410.28

Nei's (1972) standard genetic distance based UPGMA cluster analysis generated four clusters (Fig. 3). The genetic distance among the 12 populations ranged from 0.06 to 0.27 with the mean of 0.22. Cluster I and II consisted only one brown-seeded and one white-seeded populations each at a genetic distance of 0.11 (VINH, ND) and 0.12 (SKD and KPT), respectively. Cluster III and IV comprised four populations each at a genetic distance of 0.20 (DT, TN, CT, AG; all brown-seeded) and 0.17 (HD, TH, EKD and KPC; all white-seeded), respectively (Fig. 3). Bootstrap analysis with 500 replications generated 258 different trees, with the most frequent tree and the original tree generated 24 and 17 times. The bootstrap tree generated using FreeTree 1.6.6 was similar to the Nei's (1972) genetic distance based UPGMA tree generated using NTSYSpc 2.1 program.

Figure 3.

RAPD based UPGMA dendrogram generated based on Nei's standard genetic distance for the 12 sesame populations.

Comparison of agro-morphological and RAPD markers

A relatively high correlation (r = 0.88, P < 0.001) was found between the agro-morphological and RAPD data in terms of relationship between populations (unpubl.). The two methods are in agreement in grouping 7 out of the 12 populations in to two groups (TN, AG, DT; all brown-seeded) and (HD, TH, EKD, KPC; all white-seeded). The clustering pattern of the remaining five populations was different between the two methods (Table 4, Fig. 2, 3). The range of agro-morphological data based genetic distance between pairs of populations was wider (0.02 to 0.47) as compared to that of RAPD based data (0.06 to 0.27). The cluster analyses of the two data set resulted in somewhat different clustering pattern of the populations at country and regional level. For example, the four Cambodian populations were placed under cluster I (SKD and KPT) and cluster IV (EKD and KPC) in pairs in RAPD based UPGMA dendrogram where as Agro-morphological data based cluster analysis placed SKD and KPT populations under cluster I and III (Fig. 2, 3).

Table 4.  Comparison of RAPD- and agro-morphology-based clusters for the sesame populations used in this study.
RAPDAgro-morphological traitsWithin cluster mean values for some traits
ClusterNPPopulation codeRegion of collectionMGDNPPopulation codeRegion of collectionMDYield plant−1 (g)Plant height (cm)Capsules per plant50% FL
  1. NP = no. of populations; MGD = mean genetic distance within clusters; MD = mean distance within clusters; 50% FL = 50% flowering.

I2VINH, NDNorth Vietnam0.1135TN, AG, VINH, DT, SKDSouth, North Vietnam and Cambodia0.1824.2139.39941
II2SKD, KPTCambodia0.1221CTSouth Vietnam0.2316.093.36236
III4DT, TN, CT, AGSouth Vietnam0.206HD, EKD, ND, KPC, TH, KPTNorth Vietnam and Cambodia0.2115.1119.96139
IV4HD, TH, EKD, KPCNorth Vietnam and Cambodia0.175    


Evaluation of genetic diversity in crop species is essential for breeders in order to have a good starting material for breeding purposes. Genetic diversity in crop species can be studied using various methods such as agro-morphological, biochemical and molecular markers. In the present study, agro-morphological analysis revealed a significant variation among most sesame populations (P < 0.05). RAPD based analysis of genetic diversity revealed a relatively higher genetic diversity in brown-seeded populations from South Vietnam as compared to the other two regions. Interestingly, populations from this region also had higher values for yield related traits such as number of capsules per plant, number of seeds per capsule, and seed yield per plant suggesting positive correlation between the extent of genetic variation within population and yield related traits in sesame. Thus, populations from this region should get priority for conservation. Agro-morphological data based genetic distance between populations ranged from 0.02 to 0.47 with a mean of 0.23 (Fig. 2). The result is in agreement with Arriel et al. (2007) who estimated the genetic divergence of 30 morphological and agronomic traits in 108 sesame genotypes through multivariate analysis. Several molecular marker based studies on sesame also revealed detailed information on its genetic diversity (Bhat et al. 1999; Kim et al. 2002; Ercan et al. 2004; Laurentin and Karlovsky 2006; Salazar et al. 2006; Pham et al. 2009), which is in agreement with the present study.

A significant positive correlation between different markers used in genetic diversity analyses has been reported in various crop species, such as maize (Zea mays L.) (Kantety et al. 1995; Pejic et al. 1998), barley (Hordeum vulgare L.) (Russell et al. 1997), soybean (Glycine max (L.) Merr.) (Powell et al. 1996). The correspondence between agro-morphological and RAPD markers also has been reported in species such as Andrographis paniculata (Burm.f.) Nees (Lattoo et al. 2008) and Trifolium pretense L. (Greene et al. 2004). A highly significant positive correlation between the agro-morphological and RAPD markers was found (r = 0.88, P < 0.001) in the present study. The two methods revealed a similar level of mean genetic distance between populations (0.23 and 0.22, in that order). However, the clustering of the populations based on the two data sets showed some discrepancies. The number of clusters was determined by applying the principle of an ‘acceptable number of clusters’ (i.e. where the within cluster genetic distance is less than the overall mean genetic distance and where the between-cluster distances are greater than the within-cluster distance of the two clusters involved, as described in Brown-Guedira et al. (2000). RAPD data was better in clustering the populations according to geographical origin whereas agro-morphological data was better in clustering populations according to their seed color. For example, all four populations from South Vietnam were clustered into Cluster III. Overall, the two data sets produced similar clustering pattern for seven out of the twelve populations used in this study: (1) DT, TN, AG, and (2) HD, TH, EKD, KPC (Table 4). There may be several reasons for the observed differences between the agro-morphological and RAPD markers. Differences might occur if agro-morphological similarity was due to different combinations of alleles producing similar phenotypes (Johns et al. 1997). Differences between agro-morphological and RAPD markers could also occur if a single or few genes controlled the expression of agro-morphological traits that RAPD markers fails to detect (Steiner and de los Santos 2001). Linhart and Grant (1996) indicated that the discordance might also be due to differences in evolutionary rates between agro-morphological characters and characters originating from selectively neutral, non-coding DNA region, especially if the agro-morphological characters have adaptive value and the molecular markers are selectively neutral. In addition, several other factors may affect the estimation of genetic diversity and relationship between individuals such as type and number of markers used, distribution of markers in the genome, and the nature of evolutionary mechanisms underlying the variance measured (Powell et al. 1996).

In genetic diversity analysis, the use of different methods could give relatively different estimates for various genetic diversity parameters. However, the use of two or more different methods helps to better understand the genetic diversity and relationship within and among populations of crop species, as shown in several studies in species, such as bitter gourd (Momordica charantia L., Dey et al. 2006), red clover (Trifolium pretense L., Greene et al. 2004) and andrographis (Andrographis paniculata (Burm. f.) Nees, Lattoo et al. 2008). In the present study, the RAPD markers revealed a high level of polymorphism (74.1%) across the sesame populations. Similarly, the agro-morphological markers showed a high variation in the nine traits studied. Both methods suggested the existence of a substantial amount of genetic diversity both in the Vietnamese and Cambodian populations. In conclusion, both agro-morphological and RAPD markers were found to be useful in genetic diversity evaluation of sesame.


This project was supported by the Swedish International Development Agency (SIDA/SAREC). The authors acknowledge Dr Nguyen Thi Lang at Cuu Long Rice Research Institute (CLRRI) for her support during the field trials, Ho Viet The at Nong Lam University for technical assistance during field trials. Mr Thanh and Mr Duong (CLRRI) were also acknowledged for taking care of the sesame populations. We are thankful to Susanne Hjerdin and Helen Lindgren (SLU, Alnarp) for their technical assistance and support.