A total of 2,700 one-day old Saudi native chicks, representing six breeds (naked neck black, frizzled brown, black, black barred, brown and gray) (450 each) were kept in the poultry research station, Qassim University. The present breeds were collected and propagated under an extensive breeding program to purify and conserve native genetic recourses. All breeds were kept under similar management and environmental conditions. The care and handling of chickens were in accordance with regulations of animal care and ethics committee of Qassim University. The name, abbreviation, morphological appearance and photo for each population are summarized in Table 1. Also, geographical location of Saudi native chicken populations is shown in Figure 1.

A total of 144 blood samples were collected from the different chicken populations (24 each). Approximately one mL blood/bird from a wing vein was collected in ethylenediamine tetraacetic acid (EDTA) tubes and stored at −20°C. DNA was extracted from 0.5 mL of whole EDTA blood using ILLUSTRA blood mini spin kit (GE Life Sciences, Buckinghamshire, UK). Then, DNA samples were taken to determine the quantity and quality of DNA using Thermo Scientific Nano Drop 8000 UV-Vis Spectrophotometers. The selected microsatellites were amplified using ProFlex PCR System Applied Biosystems (Toronto, ON, Canada). The polymerase chain reaction (PCR) was performed for each locus in 10 μL reactions consisted of 2 μL of Genomic DNA (20 ng), 5 μL 2×PCR AmpliTag gold PCR Master mix (Applied Biosystems, Foster City, CA, USA), 0.4 μL primer mix (50 pmoles) and 2.6 μL DNase free water. The PCR program was carried out at 95°C for 5 min, followed by 30 cycles of 95°C for 30 s. Annealing temperature (ranged from 58°C up to 64°C depending on primers sequence) was determined for each primer for 30 s. and 72°C for 30 s, and final extension at 72°C for 10 min. Following the completion of the PCR cycles, 3 μLof the reaction products was mixed with 1 μL 6×gel loading dye and then loaded into each well of vertical 8% polyacrylamide gel made with 1×tris base, boric acid and EDTA buffer at 100 V for 60 to 90 min and stained with Ethidium bromide (1%). A 50 bp DNA ladder was used to estimate allele sizes in base pairs (bp). A reference bird was used to compare and correct allelic size in each gel. Twenty microsatellite markers distributed on 14 autosomal chromosomes and previously used by Fathi et al [4] and recommended by the International Society of Animal Genetics (ISAG)-FAO [1] were selected to assess the genetic diversity among Saudi native chicken breeds (Table 2).

Genetic variability was estimated per locus and across all loci for each population by allelic frequencies, observed heterozygosity, expected heterozygosity in all populations (as estimated from the pooled allele frequencies [H_T]), expected heterozygosity estimated within a population (H_S) and Hardy Weinberg equilibrium (HWE) using GENETIX program [17] and GENEPOP 4.1 software [18].

The F_IT (inbreeding coefficient of an individual relative to the total population), F_ST (the effect of subpopulations compared with the total populations), and F_IS (inbreeding coefficient of an individual relative to the subpopulation) and values for each breed were calculated using the FSTAT 2.9.3 [19] and GENEPOP 4.1 software [18]. The following formulas were used to calculate their values:

Where, HT, total gene diversity or expected heterozygosity in the total population as estimated from the pooled allele frequencies; HI, intra-population gene diversity or average observed heterozygosity in a group of populations; HS, average expected heterozygosity estimated from each subpopulation.

Genetic distance was computed between populations without a bias correction [20]. These genetic distances were developed based on different models of molecular evolution and/or different taxonomic units [21]. The POPTREE2 program V. 2.0 [22] was used to perform evolutionary analyses of allele frequency data to construct a phylogenetic tree for the studied breeds. A neighbor-joining (NJ) tree of the seven chicken populations was constructed.

Across the 20 microsatellite markers studied, a total of 172 alleles were identified in all native chicken breeds (Table 3). All studied loci were polymorphic in all chicken breeds. Overall, the lowest number of alleles per locus (5 alleles) was recorded for MCW0034 and MCW00165, while the highest number (15 alleles) was recorded for LEI0234. The average number of alleles per locus was 8.6. The highest mean number of alleles was recorded in the black breed (5.5), whereas, the lowest number (4.4) was recorded in gray breed. However, FAO has specified a minimum of 4 distinct alleles per locus for efficient judgment of genetic diversity between breeds. Generally, both black chicken genotypes (naked neck and normally feathered) exhibit a high genetic diversity in terms of number of alleles. This may be due to the fact that the black chickens are more popular and are distributed in a wide area. The lowest allelic diversity observed in gray breed may also suggest a high degree of inbreeding. The mean number of alleles of Saudi native populations was higher than that of Taiwanese conserved breeds [23] and Vietnamese domestic chicken populations [24]. Compared with our findings (8.6 alleles), Shahbazi et al [25] reported a mean number of alleles of 4.8 per locus in Iranian native chickens. While, Kaya and Yildiz [26] reported that the mean number of alleles among studied loci for Turkish native chickens was 7.5. These values were lower than those reported by Zhang et al [7], who stated that there were 9.3 alleles in Chinese native chicken breeds. However, the number of alleles at a single microsatellite locus in any single chicken population has ranged from one (monomorphic) up to several [27,28]. The average of alleles per locus observed ranged from 3.4 to 9.3 in non-commercial birds [25,26,29,30]. The higher values of alleles in non-commercial chickens in relation to the commercials are due to a strong pressure of artificial selection which occurred in the latter.

Fixation indices (F_IT, F_ST, and F_IS) estimated according to Weir and Cockerham [31] are listed in Table 4. Also, the estimates of observed heterozygosity (H_O) and expected heterozygosity (H_E) were recorded based on allele frequency data for each locus per breed (Table 4). The average of observed heterozygozity was quite high (0.76) with a range of 0.35 to 0.91. The MCW0123 marker recorded the lowest figure (0.35), while the MCW0183 marker recorded the highest one (0.91). Expected heterozygosity recorded an average of 0.71 across all loci, ranging from 0.44 (MCW0123) to 0.83 (MCW0165). Similar average was found (0.71) for total gene diversity (H_T) across all native chicken breeds. However, the mean of H_E recorded in the current study was lower than that reported by Zhang et al [7] in Chinese native chickens. On the other hand, H_E was higher than that of Hillel et al [32], who reported that the average gene diversity within 52 breeds across all 22 loci was 0.47. The variation of expected heterozygosity may be due to the differences in location, sample size, breed structure and microsatellite markers [33]. The highest value of total gene diversity (H_T) for all studied breeds was 0.85 (MCW016), which meant that this locus was a highly informative locus among all loci. On the other hand, the value of H_T for MCW012 was 0.48, indicating a slightly informative locus.

The F_IS represents a degree of nonrandom mating (deviation from HWE). A positive number for F_IS means deviation from HWE. Only MCW0069, MCW0165, MCW0103, MCW0123, and MCW0295 showed a positive number. Out of all used markers, 75% showed negative figures. Relatively, low average F_IS (−0.08) across all loci in the resent study was recorded, indicating non-random mating in the Saudi native chicken breeds. However, F_IS is used to obtain a deeper insight in appraising the degree of in-breeding and endangerment potentiality and is considered as an important tool to judge the conservation priority [34]. Accordingly, when F_IS is less than 0.05, the breeds are not in danger. On average, the genetic differentiation index, F_ST, among breeds was 0.02 (Table 4). About 2% of the total genetic variation corresponded to the differences between breeds and the remaining 98% was the result of variation among individuals within breed. Ramadan et al [34] found a 8% of genetic differentiation across 21 studied loci used among six Egyptian local strains. A higher estimated value of F_ST (0.357) owing to line differences was recorded in pure-bred commercial chicken [35].

The expected heterozygosity was lower than the observed heterozygosity for all chicken breeds (Table 5). All populations were characterized by a high degree of genetic diversity, with the lowest heterozygosity found in the brown breed (72%) and the greatest in the frizzled and black barred populations (78%). Higher estimates of expected heterozygosity (0.68) were found in both black breeds (normal and naked neck) compared to the other chicken populations. Moreover, the normally feathered brown breed recorded the lowest expected heterozygosity (0.62). In comparison with naked neck and frizzled chickens kept mainly by fanciers, normally feathered Saudi native chickens exhibited a lower degree of heterozygosity. However, native chickens carrying major genes are kept by fanciers as ornamental breeds in small size populations in many regions of central and south area of Saudi Arabia. This condition seems to result in a higher degree of diversity than the other breeds. Normally feathered black breed had the highest genetic variability in terms of expected heterozygosity (0.68) and number of alleles (5.45). This may be due to the fact that the black native chickens has been found in large number and distributed in a wide area of Saudi Arabia. BB and BR-Ff breeds recorded the highest difference between observed and expected heterozygosity. The H_O and H_E observed in Saudi native chicken breeds in the current study were similar to or slightly lower than those reported by Zhang et al [7] in Chinese native breeds.

All studied breeds showed no inbreeding within breed (negative inbreeding coefficient [F_IS]). The mean F_IS for Saudi native chicken breeds was lower than zero (−0.07). These values were similar to that reported by Tadano et al [36] for 12 commercial chicken lines (0.000 to 0.141) based on 40 microsatellite loci, while it was lower than that reported by Kaya and Yildiz [26] and Ding et al [37] for Turkish and Chinese native chicken breeds, respectively. To decrease F_IS, full- or half-sib mating should be avoided to prevent inbreeding depression. Among breeds, the mean F_IS values varied from −0.01 for naked neck black breed to −0.10 in both normally feathered brown and black barred. The mean F_ST value of 0.02 indicates that approximately 2% of the total genetic variation is caused by breed differences, whereas the remaining amount (98%)was due to differences among individuals within breeds. With respect to HWE, black chicken breeds (either normal or naked neck) and normal brown breed exhibited a deviation from Hardy–Weinberg equilibrium (dHWE). Out of all 120 HWE tests in 6 breeds tested, dHWE at the 5% level recorded 12.9%. Contrary with our results, Ding et al [37] reported that all breeds showed statistically significant deviation from the HWE at many loci.

Molecular information of genetic diversity is playing an important role in conservation of chicken resources. Genetic distance based on Nei’s unbiased values is presented in Table 6. The high genetic diversity in Saudi native chicken breeds is consistent with great phenotypic variation among them. The greatest genetic distance was found between gray and brown breeds (0.246).The smallest figure was recorded between the naked neck and normally feathered black populations (0.107). Both normally feathered and naked neck Black breeds (0.107) were most closely related to each other than that of normally feathered and frizzle brown populations (0.139). According to many reports, local chickens from Africa and Asia are characterized by extensive phenotypic variations within and between different populations [35,38–46]. Similarly, high microsatellite diversity was reported within chicken strains from Iran, China, Turkey, Korea, Sudan, Southeast African countries [25,26,35,47,48]. Uncontrolling breeding programs and traditional management systems may be contribute to higher variations within and between these populations. Consistent with our findings, Osman et al [49] and Tadano et al [36] reported that the use of 20 microsatellite markers and approximately 24 birds per breed might be adequate to obtain accurate results of genetic diversity in Japanese chicken breeds.

The unrooted NJ tree derived from the Nei’s standard genetic distance of six Saudi native chicken breeds is given in Figure 1. It could be observed that three main clusters were clearly recognized. Frizzled brown breed and normally feathered brown breed located in one cluster, while naked neck black breed with black-barred breed located in the second cluster. Finally, normal black breed and the gray breed were located in the third cluster. The black and gray breeds were placed to the cluster whose distance is far from naked neck black and black barred breeds. Additionally, naked neck black breed (BL-Nana) grouped with the black barred one (BB).