The SCoT markers used to analyze the genetic diversity of 78 hazelnut genotypes from in the natural forests of Fandoguloo in Ardabil province of Iran exhibited substantial polymorphism, with six polymorphic bands belonging to SC9 and 14 to SC21, SC12, and SC5 markers out of 165 polymorphic bands at the population level. Due to a large number of genes in different loci of the genome, for each primer, various numbers of bands were obtained per genotype (Collard and Mackill 2009). This has the advantage over other primers, which either act randomly or amplify nongenic loci. The marker index (MI) varied from 2.6 to 6.1, indicating its high efficiency to detect the amount of polymorphism across the examined populations. The SC21, SC12, and SC5 primers possessed the highest MI values (28). The resolving power index (RP) associated with the SCoT primers showed their efficiency in detecting the differentiation of the populations. It has been mentioned by Bhattacharyya et al. (2013), that evaluated the genetic diversity of wild orchid and supported the efficiency of 15 SCoT markers. Mulpuri and colleagues (2013) in their survey using SCoT marker, revealed highe polymorphism in Jatropha curcas and confimed application of this marker in distinguishing the toxic and edible genotypes (Mulpuri et al. 2013).
The highest Ne (0.02–1.69), Nei (0.01–0.388), and I (0.02–0.564) corresponded to the Essygran (Ardabil) subpopulation, and the lowest one (0.02–1.141) was estimated for the subpopulation from Fandoghloo (Ardabil), indicating the high allelic frequency and genetic diversity among the subpopulations. According to the calculated indices for the primers, SC3 had the highest PIC (0.45) and RP (0.35) values. These findings also suggest that SC3 outperforms other primers in terms of assessing genetic diversity. The analysis of SCoT markers revealed 100% polymorphism at the population level, and the SC12, SC5, and SC21 primers had the highest polymorphism with 14 bands, but the lowest uniformity when compared to other primers.
SC9 was allocated the lowest band number, with six polymorph bands reflecting the best population-level consistency. Further, the greatest Nm (0.7) and lowest Fst (0.554) values across the groups corresponded to intra populations. This illustrates that, despite relatively large Nm values, dispersion among populations can be observed to some extent that possibly influenced by glacial-interglacial climatic changes in the area. Similar pattern of climatic-affected differentiation in the Italian hazel populations was also reported by Palme´and Vendramin (2002).
The evaluation of genetic diversity among geographic locations due to factors such as linkage, inbreeding, migration, and differences associated with samples is of great importance (Mohammadi and Prasanna 2003). In other words, the validity of such analyses highly depends on the number of individuals examined, the number of allelic loci, the allelic and genotypic characteristics of the specimens, sample size, and the type of the intercross (Strauss et al. 1992). The Ht and Hs parameters were calculated to be 0.395 and 0.237, respectively. According to the population analysis, the amount of heterozygosity within subpopulations (Hs = 0.237) was higher than between subpopulations (Ds = 0.158). Genetic differentiation may result from differences in genetic features caused by environmental changes and natural selection (Bossdorf et al. 2005), as well as random events such as mutation, migration, and the degree of population divergence, which could be evaluated using several parameters (Slatkin 1995). In the case of this study, our results suggest that the isolation of hazel populations might be due to past climatic fluctuations during glacial periods that led to fragmented populations in the studied area. This pattern has been also reported by several authors for different entities in this area (Rajaei Sh et al. 2013; Ahmadzadeh et al. 2013; Rezazadeh et al. 2020).
Based on the combination of SCoT markers used, the variance within and among the subpopulations was estimated to be 29.5 and 6.43, respectively, implying considerable demographic differentiation in the studied area. The highest genetic distance (0.204) and the lowest similarity (0.816) among the subpopulations corresponded to the Fandoghloo and Essygran subpopulations. Due to the non-random feature of the Jaccard indices, they have no definite statistical distribution, which, in turn, makes it difficult to calculate sampling variance and, consequently, confidence intervals (Lombard et al. 2000). However, bootstrapping techniques could tackle the problem (Tivang et al. 1994).
Simple matching, Dice, and Jaccard were used to categorize the genotypes, with similar cophenetic indices indicating the concordance between the grouping outcomes. The population grouping was found to be in harmony with geographic locations, and this result was in accordance with the grouping based on the PCoA analysis; the examination of the genetic structure of the examined populations devided to three main groups with distinct genetic structures; 1- Ardabil (Fandoghloo, Essygran and Abibigloo subpopulations with less geographic distances), 2- Miyaneh and 3- Arasbaran.
The rate of total heterozygosity and heterozygosity within the subpopulations was estimated to be 0.395 and 0.237, respectively. However, the inter-subpopulation heterozygosity was estimated to be 0.157, showing that there is less genetic variation among subpopulations than within subpopulations. This was determined by estimating the inter and intra-subpopulation variances using the SCoT data from the Fandoghloo (Ardabil) and Arasbaran subpopulations. The Nei distance between the subpopulations of Fandoghloo and Arasbaran was 0.204. The current geographic isolation of these two subpopulations related to probable difficulty in gene flow, because these forests distributed in mountainous areas, and southern Caucasus range and Sabalan mountains may act as natural barriers for pollination.
This work is first survey on the genetic diversity of C. avellana in Iran by applying SCoT marker as a powerful and sensitive technique to identify low levels of genetic variations. The present study revealed higher genetic diversity within-populations but moderate genetic diversity among-populations. It can be concluded that identification, conservation and breeding programs of such genetically diverse wild germplasm from in the natural forests of Fandoguloo in Ardabil province of Iran would be necessary for this economically important taxon. Our results revealed a general trend for isolation by distance plus genetic differentiation between hazelnut populations of different geographic areas. It may relate to existing of natural geographic barriers between isolated populations as well as habitat degradation due to antropogenic activities, like livestock overgrazing and unsustainable utilization of forests. The remarkable diversity was found within the populations in this study (specially Fandoghloo forest) provides essential insights into maintanence and breeding programs of this valuable crop.