Plant materials

A total of twenty-five (25) pepper genotypes (Capsicum annum L.) were used for this study, which comprises of fifteen (15) accessions and ten (10) improved varieties (Table 1). Seed samples were obtained from Melkassa Agricultural Research Center (MARC).

Genotyping

Seeds from each genotype were sown in seedling raising tray in the greenhouse at National Agricultural Biotechnology Research Center (NABRC), Holeta. Young healthy leaves from a single seedling at the 2 to 3 leaf stage were collected in an Eppendorf tube and immediately dried using liquid nitrogen. The dried leaves were then pulverized using a Geno grinder (MM-200, Retsch) at 25 rpm for 3 min. Genomic DNA was extracted following plant DNA extraction protocol (DARTs, 2000)with minor modifications. The quality and quantity of the isolated DNA were checked by gel-electrophoresis using 0.8% agarose at 100 constant voltages for 45 min. The gels were visualized under UV light and photographed with a camera mounted on the UV Transilluminator. The quality and concentration of the DNA were further confirmed by a spectrophotometer (8 pedestal, Nano drop) at 260/280 nm wavelength absorbance. Good-quality DNA from each sample was used for PCR analysis after normalization to approximately 50 ng/µl (the normalization was carried out based on the concentration of each sample from Nano drop result).

A set of 16 SSR markers previously reported by Nagy et al. (2007), Dhaliwal et al. (2014) and Sharmin et al. (2018)were obtained and used for the final genotyping of pepper collections (Table 2). Prior to whole sample amplification, gradient PCR was applied to each primer pair on BIO-RAD T100^TM thermal cycler to get an optimum annealing temperature and other PCR setup. The polymerase chain reaction (PCR) was then carried out in a 12.5 µl final reaction volume containing 6.25 µl of Taq DNA polymerase, 0.5 µl of each forward and reverse primers, 3.25 µl of nuclease-free water, and finally 2 µl of gDNA. The PCR condition was adjusted at initial denaturation of 95°C for 5 min, 36 cycles of denaturation 94°C for 45 s, annealing varied with the primers (Table 2) for 45 s, extension 72°C for 90 s, and final extension 72°C for 10 min.

Gene diversity and markers polymorphism

16 SSR markers were successfully implemented in 25 pepper genotypes to evaluate the gene diversity and levels of polymorphism within and among the accessions and improved varieties. The gene diversity indices varied across the entire markers in tested genotypes. The 16 SSR markers generated a total of 67 alleles, ranging from 9 for marker AVRDC PP-67 to 2 for markers AVRDC PP-167, GPMS6, and GPMS112 with a mean of 4.19 alleles per marker (Table 3).

Various results have been reported from similar studies on varying number of genotypes and markers. For instance, Rabuma et al. (2020)reported a lower mean number of alleles (2.2) in 32 Ethiopian and Indian accessions using 14 SSR markers. In contrast, a higher number of alleles were reported by Woubit et al. (2021)that identified a mean of 8.54 alleles in 75 Ethiopian pepper germplasms using 13 SSR markers. Similarly, a mean number of alleles of 6.9 are reported from a large Capsicum annum collection in 179 individuals from six countries, other than Ethiopian origins using 21 SSR markers (Christov et al., 2021). A relatively lower (2.7) mean number of alleles were reported in 64 Indian pepper accessions using 27 polymorphic SSR markers. Obviously, the number of alleles detected in the germplasm or population is influenced by the species' genetic backgrounds and the molecular markers difference deployed.

The major allele frequency (MAF) ranged from 30 to 95% with a mean of 54%. A mean value of 0.20 was obtained for observed heterozygosity (Ho) with the highest record (1.00) was attained by marker AVRDC PP-167 and no (0.00) observed heterozygosity by the ten markers (Table 3). The lowest Ho observed in most of the markers in our study is directly correlated with the fact that the majority of improved varieties and accessions are being homozygous. This can also be attributable to the high level of inbreeding that improved varieties are expected to exhibit. A variation in the levels of markers polymorphism has been observed (Figure 1), with a PIC value ranging from 8 to 80% and a mean of 52% (Table 3).

Molecular markers with PIC values of >0.50 (50%) are considered to be highly informative for genotyping studies (Botstein et al., 1980). Except for the two markers (GPMS6 and GPMS112), the others showed high PIC values, implying the presence of immense genetic diversity among studied genotypes and the very suitability of the markers for molecular characterization of pepper genotypes. Similar results were obtained with a mean PIC value of 0.57 using different SSR markers (Rabuma et al., 2020).

A gene diversity index is one of the most important measures of genetic divergence and is useful for determining the amount of diversity in the genotypes. The gene diversity (GD) in this study ranged from 0.09 to 0.82 with an average of 0.57 (Table 3), indicating the presence of high genetic diversity among Ethiopian pepper genotypes. The highest GD value (GD=0.82) was recorded for marker AVRDC PP-67 while the lowest (GD=0.09) was for marker GPMS6.

Genetic dissimilarity and phylogenetic relationship

The pairwise genetic dissimilarity coefficient determines the genetic relatedness among the genotypes. The highest genetic distance value of 1.00 was observed between PBC-731 and Acc-22 and the lowest (0.25) was between Acc-13 and Acc-11 genotypes (Table 4). The dissimilarity coefficient value of 1.00 indicates that the two genotypes are genetically different; while, the the value 0.25 indicates that the two genotypes have a higher genetic similarity. In general, most of the pairwise dissimilarities obsereved were higher across the 25 genotypes implying a broad range of genetic variability basis among the tested pepper genotypes.

Phylogenetic analysis is useful not only for estimating the genetic distance of genotype collections but also for selecting crossing parental lines. Varieties with a greater genetic distance are generally recommended as parents to produce  progeny with a heterosis effect. Indeed, the wide range of diversity in our tested genotypes could be important for broadening the genetic base because it enhances the chances of discovering more unique genes. Moreover, it provides the opportunity of hybridization between distant genotypes which helps in the production of heterozygous individuals with desirable traits.

A two-dimensional display of principal coordinate analysis (PCoA) was also performed to further investigate the genetic relationship of pepper genotypes and the result showed the first three principal coordinate axes explained 44.33% of the total genotypic variation in the studied genotypes. The first and the second explained 17.01 and 16.06% of the total variation, respectively. The PCoA displayed a scatter plot with a wider dispersion of the genotypes in all the quadrants without forming a clear cluster, and some genotypes like PBC-731, Acc-45, Acc-9, and Acc-22 are plotted far from the central axis, indicating the individuals' genetic distance among pepper genotypes and such kind of genotypes are highly recommended for future pepper breeding (Figure 3). In most cases, even though the genotypes are displayed scattered across all quadrants in the PCoA, the majority of the accessions and improved varieties are somewhat separated and formed three clusters (C1, C2, and C3) based on their genetic background, as similar as the dendrogram.

Population structure and analysis of molecular variance

The population structure analysis was inferred on the 25 genotypes (15 accessions and 10 improved varieties).

The highest value of delta K (?K) was obtained for K=4, revealing the existence of four genetic groups (Group I, Group II, Group III and Group IV) of pepper genotypes (Figure 4A and B). Group I comprised 9 genotypes, of which 5 were improved varieties and 4 were accessions. Group II and Group III consists of 8 and 1 accessions, respectively. Whereas Group IV constituted 7 genotypes, 5 of which are improved varieties while 2 are accessions. However, this structure analysis displayed that the group I and IV, and genotypes from the predicted genetic groups had a high level of admixture (Figure 4C). We speculate that the reason for this is maybe pepper genotypes were acquired from the different gene pools with a high-level of mixture, as reflected by a high level of gene flow (Nm=2.4) and low genetic differentiation between groups (Table 5). The grouping at K=4 showed less concordance with that of the dendrogram and PCoA, this is because the very few distantly related genotypes could have contributed for the less concordance. Reports from population structure analysis in other pepper diversity panel indicated the existence of well-differentiated population groups (Solomon et al., 2019; Rabuma et al., 2020). The same is true in the present study except we used relatively small number of genotypes.

Analysis of molecular variance (AMOVA) was used to quantify the genetic divergence within and among groups. We partitioned the total molecular variance based on the K=4 from STRUCTURE result and the priory grouping information into accessions and improved varieties (Table 5A and B). As a result, the total variation was partitioned in to 9% among the four groups, and 91% within groups.

Apparently, the high genetic differentiation has been demonstrated among individuals (Fis=1.00) as it is evident from Table 5. Almost the same result is obtained from the AMOVA based on the grouping with priory information in to improved varieties and accessions (Table 5B). Except for the negligible difference in estimated variance and F-statistics, exactly identical % of variations were found for among population (9%) and among individuals (91%) sources. This result may suggest that the ?K based grouping of the genotypes is somehow related to grouping in to the accessions and improved varieties which are genetical grouping as well.

Several variable results have been reported from previous studies conducted by various authors. Woubit et al. (2021) reported a partitioning of the total molecular variance in to 7% among eight-geographic groups, 63% among individuals, and 26% within individuals of Ethiopian pepper accessions. Similarly, in 32 Ethiopian and Indian accessions grouped in 9 populations by SSR markers, Rabuma et al. (2020)reported 32% of the total variation among populations and 68% within populations of the total variation. In another study conducted by SSR markers on Mexican pepper populations, of the total molecular variance among the population, 10% was among wild, landrace, and hybrids, 15% was among individuals within populations, and 74% was within individuals in the populations (Pacheco-Olvera et al., 2012). In a nut shell, levels of molecular variations explained by sources of variations in diversity studies are a function of the grouping compositions (varieties, landrace accessions, hybrids, advanced breeding lines etc.) the number of individuals, polymorphic power of the markers used etc.