Molecular Comparison between the Sicilian-Sardinian Sheep Populations in Tunisia using ISSR Markers

Sheep farming occupies an important place not only internationally but also in Tunisia. In fact, the activity of sheep farming has been always associated with social and ritual importance. The sheep sector is one of the pillars of the Tunisian national economy; in fact, it helps to create jobs in rural areas and therefore limits the rural exodus. Dairy sheep farming represents a micro-sector dominated by the Sicilian-Sardinian breed, which is the only specialized dairy sheep, breed in North Africa. Its workforce has been declining sharply in recent years. The use of molecular genetics for sheep genetic studies has dramatically increased recently in Tunisia (Ben Sassi-Zaidy et al., 2014; Kdidi et al., 2015; El Hentati et al., 2017). The genetic improvement program of this breed becomes the only escape route to remedy its current state characterized by the drastic decrease in the number of sheep from this breed since the 1990s (Aloulou et al., 2018).

Blood sampling
 
Blood samples were collected from Sicilian-Sardian breeds from three locations:
- The pilot farm of Fretissa (n=25), attached to the Office of Livestock and Pastures (governorate of Bizerte)
- The cooperative agricultural production unit of Gnadil (n=25), a state farm (governorate of Beja)
- Two farms belonging to two breeders in the region of Nagachia (Governorate of Beja) (n=25).

The selection scheme applied in Tunisia consists of recovering at the age of six months, from the herds registered for performance control, the best males likely to be future breeders based on their growth between 10 and 30 days and on their mothers’ milk production. These breeders are then distributed at the age of 18 months at the level of both the selection base and the rest of the population while taking care to minimize inbreeding. Consequently, the animals used are inevitably related but with a very low degree without being able to determine it.

A veterinarian took the blood from the jugular vein of animals on EDTA tubes. The samples were stored at -20°C until DNA extraction.
 
DNA extraction
 
Total genomic DNA was extracted using the iPrep PureLink gDNA Blood Kit from Invitrogen on the iPrep machine as described by the manufacturer.

The quality and quantity of the extracted DNA were evaluated by horizontal electrophoresis on an agarose gel (0.8%). DNA standards of known concentrations were used as a reference to determine the concentration of the DNA.
 
PCR amplification
 
All DNA samples were diluted to a concentration of 50 ng/μl. DNA amplification was performed using ISSR primers (UBC 811, UBC 834 and UBC 898) which are developed by the University of British Columbia (UBC, Vancouver, Canada). The PCR amplifications were carried out in a reaction volume of 25 μl containing 25 ng of genomic DNA, 0.4 μM of the ISSR primers, 100 μM of dNTP, 2 mM of MgCl2, 0.8 units of Taq DNA polymerase (Invitrogen) and five μl of Taq buffer (5X). In order to detect any contamination, control reactions without genomic DNA were carried out for each amplification. The ISSR-PCR reactions were conducted in a thermocycler (Kyratec) programmed to perform 45 cycles of 94°C for 40 seconds, TA (hybridization temperature of each primer) for 40 seconds, and 72°C for 90 seconds; an initial denaturation step of 5 minutes at 94°C and a final eight minutes extension step at 72°C was included in the first and last cycles, respectively. The amplification products were separated by 3% agarose gel electrophoresis containing ethidium bromide in Tris Borate EDTA buffer and visualized under ultraviolet light.
 
Statistical analysis
 
Nei’s gene diversity (H) (Nei, 1973) under Hardy-Weinberg equilibrium (Clark and Lanigan, 1993), Shannon index (I) (Lewontin, 1972), the percentage of polymorphic loci (P) and the Nei’s standard genetic distances (Nei, 1978) corrected for small samples were all calculated. In addition, an unweighted pair-group method with arithmetic average (UPGMA) dendrogram containing the three studied populations was constructed based on the matrix of genetic distances using Popgene (Population Genetic Analysis) version 1.32 software (Yeh et al., 1999).

One hundred and fifty-three bands with a gene size ranging from 55 bp to 2809 bp were obtained from the 75 animal samples collected for this study with three ISSR primers. Primers UBC834, UBC898 and UBC811 amplified 40, 51, and 62 ISSR fragments, respectively, with an average of 51 bands per primer. The percentage of overall polymorphism is 100. Fig 1,2 shows an example of profiles of DNA fragments amplified by the primer UBC811. (Zamani et al., 2011) reported percentages of ISSR primer polymorphism ranging from 69 to 77% by studying genetic diversity in the Mehraban Iranian sheep breed. Similarly (Mohammadabadi et al., 2017) found 80.6% and 85.7% ISSR polymorphic loci for two primers used for characterization of the Kermani sheep breed in Iran. The high level of polymorphism shows that the markers used in this study are very informative and allow an unbiased estimation of the genetic variability in the studied populations.





In order to evaluate the genetic diversity within the different studied populations, we have calculated for each one of them the number of polymorphic loci (NPL), the percentage of polymorphism (P), the genetic diversity of Nei (H) and the Shanon index (I) (Table 1,2).





The genetic diversity detected was significantly lower than expected heterozygosity values found in other genetic characterization studies in sheep using microsatellite markers namely 0.73 in Turkish sheep (Das et al., 2014), 0.77 in Chinese indigenous breeds (Bai et al., 2015), 0.76 in Indian Tiruchy Black sheep (Kavitha et al., 2015) and 0.66 in Saudi sheep populations (Mahmoud et al., 2018). However, the results found in this study are comparable to those found by (Zamani et al., 2011) in the Mehraban sheep breed in Iran using ISSR primers (H = 0.1258, I = 0.21). The low values of the calculated heterozygosity corroborate the results found by (El-Hentati et al., 2017) in Sicilian-Sardian sheep in the Nagachia region.

To better visualize the phylogenetic relationships between the different geographical populations, a UPGMA dendrogram was constructed from Nei’s genetic distances (1978). It shows that the population of Fretissa and the population of Nagachia were the closest while the population of Gnadil seems to be furthest away from the others. This shows that there is no geographical separation between the different populations knowing that the kilometric distances that separate the study sites are 50, 65 and 20 kilometers between Fretissa-Gnadil, Fretissa-Nagachia and Gnadil-Nagachia respectively.

The GST gene differentiation coefficient between all populations is 0.2, which shows that only 20% of the total genetic diversity is due to interpopulation variability.

Table 3 illustrates the gene differentiation coefficients (GST) and gene flows (Nm) between pairs of populations. It emerges that the migration of fertile individuals between the Nagachia-Fretissa populations is more important than between the rest of the populations. This confirms that, in this study, geographic distance does not affect gene exchange.

The highest genetic diversity is found in the population of Fretissa while the lowest is found in the population of Gnadil. Regarding gene differentiation (GST), the highest value was detected between the populations of Gnadil and Nagachia (0.2) while the lowest is between the populations of Nagachia and Fretissa (0.12). This shows that insulation is more important between farms than between regions. The main effect of gene flow is the homogenization of allelic frequencies between populations: the greater the gene flow between two populations, the more similar populations are expected (same alleles present, same allelic frequencies). Therefore, the highest Nm (3.65) is found between the least differentiated populations (Nagachia-Bizerte) and the lowest (1.99) between the Gnadil-Nagachia populations. The unbiased genetic distances of Nei between the different studied populations ranged from 0.03 to 0.049. These values indicate a closer phylogenetic relationship between the populations of Fretissa and Nagachia while it is more distant between the two other populations.

Referring to the set of statistical parameters calculated, it is concluded that the genetic diversity in the Sicilian-Sardinian sheep population is low and that there is no real geographical differentiation between the populations of this breed and that the isolation is larger between farms than between regions. We, therefore, advise breeders to exchange breeding males between farms.