Abstract: Genetic variation within and between five Iranian sheep populations; Sanjabi (SAN), Kordi Kordistan (KKO), Kordi Khorasan (KKH), Mehraban (MEH) and Moghani (MOG) was assessed using 6 microsatellite markers (McMA2, McMA26, MAF64, OarAE64, OarCP26 and OarFCB304). There are significant deviation from Hardy-Weinberg equilibrium in all population-locus combinations except the locus McMA2 in the MOG population (p<0.005). The lowest DA genetic distance (0.234) was obtained between KKH and KKO and the highest (0.388) between SAN and MOG. D based on DA distances, neighbor-joining revealed one cluster containing KKO, KKH and SAN and another cluster with MEH and MOG. The average expected heterozygosity within populations ranged from 0.744 to 0.847 for KKH and MEH, respectively. The estimated time of divergence for two Kordi populations (KKO and KKH) was 445 years that has accordance with historical evidence. This study showed that microsatellite loci can be an useful tool for evaluating variation evolutionary relationships among local sheep populations.
INTRODUCTION
Iranian sheep are about 60 million and are considered as the most important domestic animal which mainly used to produce meat. There are different indigenous sheep in different locations of Iran without well definition as distinct breeds. However, they are considered as geographically defined populations. Study on genetic variation among Iranian sheep populations is a critical necessity.
Microsatellites are valuable genetic markers due to their dense distribution in the genome, great variation, co-dominant inheritance and easy genotyping. In recent years, they have been extensively used in parentage testing, linkage analyses, population genetics and genetic studies (Goldstein and Pollock, 1997). They have greatly used to investigate on genetic structure in local sheep populations. For example, Arranz et al. (2001) were genotyped nineteen microsatellite loci in five indigenouse Spanish breeds of sheep to determine genetic relationships among them. Stahlberger-Saitbekova et al. (2001) estimated genetic relationships between Swiss sheep breeds on basis of microsatellite analysis. Buchanan et al. (1994) determined evolutionary relationships among six British sheep breeds and both Australian and New Zealand Merino using eight ovine microsatellite loci. They have also calculated the time of divergence between the Australian and the New Zealand Merino. Farid et al. (2000) were assessed genetic variability of ten sheep breeds using ten microsatellite loci.
The present study is the first research on genetic variation within and between five Iranian sheep populations using microsatellite loci. Another objective of the study was to investigate accordance variation resulted from microsatellites with historical and geographical evidences.
MATERIALS AND METHODS
Genetic variation at 6 microsatellite loci including McMA2, McMA26, MAF64, OarAE64, OarCP26 and OarFCB304 were analyzed for five Iranian sheep consisting of sanjabi (SAN), kordi kordistan (KKO), kordi khorasan (KKH), mehraban (MEH) and moghani (MOG). Whole blood samples were collected from spreading areas of these sheep (Fig. 1). The number of DNA samples were 35, 32, 25, 25 and 24 for SAN, MOG, KKO, KKH and MEH, respectively. The properties of 6 loci are given at Table 1.
Genomic DNA was extracted by the salting-out method (Miller et al., 1988) with some modifications. All PCR reactions contained the following components: 200 μM dNTPs, 3.5 mM MgCl2, 0.25 μM each of primer, 1U Taq polymerase, 100-200 ng DNA. The final volume was 15 μL.
| Table 1: | Characteristics of the 6 sheep microsatellites used in the study |
| * Primer sequences were cited from Ede et al. (1995). ** These temperatures were set up for this study | |
| Fig. 1: | The spreading regions of five Iranian sheep populations |
Reactions were run on a thermal cycler(Biometra) using two thermal cycling programs: (a) For McMA2 and McMa26 loci, one cycle of denaturation at 95°C (2.5 min), 29 cycles of denaturation at 95°C (30 sec), annealing 52°C (30 sec), extension 72°C (30 sec) and one cycle of extension 72°C (2.5 min) (Maddox et al., 2000) and (b) For other loci, seven cycles of denaturation at 95°C (30 sec) and annealing as in Table 1 (1min) followed by 20 cycles of denaturation at 90°C (30 sec) and annealing as in Table 1 (1 min). No extension step was used (Ede et al., 1995). The products were electrophoresed on 8% nondenaturing polyacrylamide gels and bands visualized by rapid silver staining (Sanguinetti et al., 1994).
The allele and genotypic frequencies were directly estimated from the gel. Hardy-Weinberg equilibrium (HWE) based on likelihood ratio (Weir, 1996) were tested for different locus-population combinations by POPGENE software (Yeh et al., 1999). Nei (1972) standard genetic distance and DA genetic distance (Nei et al., 1983) matrices were calculated by MICROSAT software (Minch et al., 1995) and dendrograms were constructed using neighbor-joining (NJ) (Saitou and Nei, 1987) by POPTREE software (Takezaki, 2000) with 1000 bootstrap replications. The unbiased average expected heterozygosity (He) (Nei, 1978) was calculated by POPTREE software. Polymorphism criteria such as polymorphic information content (PIC) (Buchanan and Thue, 1998) and the number of observed and effective alleles (Hedrick, 1999) were also estimated by HET (Ott, 1989) and POPGENE software, respectively. The time of divergence between two kordi populations (KKH and KKO) was estimated using D = 2αt equation (Nei, 1976) where t and D are the divergence time and Nei (1978) unbiased genetic distance, respectively. In order to estimate t, we assumed α = 4.5*10-4 (observed spontaneous mutations per locus per gamete in humans) as the rate of change in repeat number. The estimated time was compared with historical evidence for divergence about 400 years ago. Mutation rates were then calculated for each locus and per all loci using the estimated time of divergence as Buchanan et al. (1994) have described.
RESULTS AND DISCUSSION
PCR reactions were successfully done with all primers excepted to OarAE64. The possible explanations for this failure are mutation at the primer site or a mistake during primer synthesis. A few alleles were found in Iranian sheep that haven’t been previously reported in the same loci for other sheep (Farid et al., 2000; Buchanan et al.,1994; Buchanan and Crawford, 1993; Arranz et al., 2001; Stahlberger-Saitbekova et al., 2001; Diez-Tascon et al., 2000; Maddox et al., 2000). There are significant deviation from Hardy-Weinberg equilibrium in all population-locus combinations except the locus McMA2 in the MOG population (p<0.005).
Table 2 shows genetic distance matrices based on DA (upper diagonal matrix) and DS (lower diagonal matrix) using 1000 bootstrap replications.
| Table 2: | Genetic distance matrices based on DA (upper diagonal matrix) and DS (lower diagonal matrix) using 1000 bootstrap replications |
| Fig. 2: | Dendrograms showing evolutionary relationships among five
Iranian sheep based on DA (a) and DS (b) using NJ
algorithms with 1000 bootstrap replications The scale shows arbitrary genetic
distance. The numbers at nodes represent robustness of nodes |
Due to long geographic distance and natural barriers (mountains and sea) between MOG and other populations, we considered MOG as an outgroup. So, it is expected to find high genetic distance between MOG and others. Longtime usage of MEH rams for crossing with other sheep may be occurred low distance between MEH and MOG and intermediate distance MEH with three other populations.
Both dendrograms (Fig. 2) have the same topology that reflects the geographical locations of these five Iranian sheep. Support of nodes is are relatively weak (except KKO and KKH branch), although only 5 microsatellites were used. There are two separate clusters on both dendrograms. One includes KKO, KKH in a branch and then SAN. Another consists MEH and MOG. A possible explanation for this topology was expressed above.
PIC values and interpopulation diversities (Unbiased average expected heterozygosity) are given in Table 3. The values of He are relatively high and have wide range (0.506 to 0.915). However values per population (average of all loci) had narrow range from 0.743 (KKH) to 0.847 (MEH). PIC and heterozygosity values were very close to other studies (Farid et al., 2000; Buchanan et al.,1994; Buchanan and Crawford, 1993; Arranz et al., 2001; Stahlberger-Saitbekova et al., 2001; Maddox et al., 2000). Small population size and short time for variability may be a possible reason for the low He in KKH. Large population size and gene flow could be explain the relatively high He for MEH and KKO. PIC per population (average of all loci) indicating variability in each population had narrow range (from 0.692 to 0.796 for KKH and KKO, respectively). PIC for each locus had also narrow range (from 0.761 to 0.901 for OarFCB304 and OarCP26, respectively). However, these values for different locus-population combinations had wide range (0.4679-0.8595). The number of observed alleles per locus varied from 7 at OarFCB304 to 13 at McMA2 and OarCP26. The highest and lowest number of effective alleles were 10.9 and 4.7 for OarCP16 and OarFCB304 loci, respectively. Although sample sizes were small, polymorphism criteria such as PIC values and number of alleles indicate high polymorphism at studied loci and sheep. These loci will be useful for further studies of population relationships, demographic history and the traceability of animals and food products.
The divergence time of two kordi populations (KKH and KKO) was estimated 445 years old (with generation interval equal 3.5 years old). This time is very similar to historical evidences (about 400 years old). Mutation rates per locus were calculated using the estimated time and then compared with proposed mutation rate. Table 4 shows Nei`s unbiased genetic distances (D) and mutation rates (α).
The range of mutation rate is very wide (from 0.97*10-4 to 12.52*10-4). The mutation rates of McMA2 and for all loci were very near to the assumed rate. The correlation between two distance criteria was very high (0.903). Since DA has been found more useful to obtaining of correct topology (Takezaki and Nei, 1996), we present only the results of DA. The lowest DA was between KKH and KKO (0.234) and between SAN and KKO (0.246). These distances are rational due to co-descendant of two kordi sheep, short time passed from their separation (about 400 years old) and neighboring geographic distributions. Their phenotypic similarity also agree with these distances. The highest DA and DS were between SAN and MOG (DA = 0.388, DS= 0.559).
| Table 3: | He and PIC values at locus-population combinations, per population (average of loci) and per locus |
| Table 4: | Mutation rates (α) and Nei`s unbiased genetic distances (D) |
CONCLUSIONS
This research showed high variation within and between studied Iranian sheep populations for five microsatellite loci. This study showed that microsatellite loci can be an useful tool for evaluating variation evolutionary relationships among local sheep populations. Microsatellite-based estimates of population relationships were consistent with known demographic history and geographic distances.
ACKNOWLEDGMENTS
This research was carried out at Animal Science Research Institution of Iran (Department of Biotechnology). Authors thank G. Luikart and A. Beja-Pereira for their comments on our manuscript and also B. Ghareyazi and L. Dolati for technical assistance. Finally, we thank N. Takezaki for kindly providing the POPTREE program.