Characterization of rare migratory cattle and evaluation of its phylogeny using short-tandem-repeat-based markers

ABSTRACT

Bovine-specific microsatellite markers were used to assess the genetic variation in Belahi cattle. Both within-breed (Belahi) and between-breed (Sahiwal, Gir, Tharparkar, Rathi, Nagori, Mewati, Kankrej and Belahi breeds) diversity indices were estimated. Sixteen microsatellite loci were polymorphic with more than five alleles per locus. Mean allelic diversity was 9.31, with a total of 149 alleles. Polymorphism information content varied from 0.31 (TGLA227) to 0.87 (CSSM33) with an average of 0.71. Estimate of Nei’s gene diversity was 0.72 ± 0.15. The observed and expected heterozygosity at different loci were 0.69 ± 0.17 and 0.71 ± 0.15 and varied from 0.30 (TGLA227) to 0.88 (TGLA122) and 0.32 (TGLA227) to 0.88 (CSSM33), respectively. Within-population inbreeding estimates (F_IS) for Belahi cattle was 0.0337, indicating an average deficiency of 3.37%. Belahi cattle revealed the presence of genetic diversity within population and there was no significant heterozygosity excess indicating the absence of genetic bottleneck in the recent past. Global F_ST estimates demonstrated that 85% of the total variation was contributed by within-breed genetic differentiation, while 15% genetic variation was present across different breeds. Nei’s standard genetic distance estimates among the studied milch breeds varied from 0.27 (Tharparkar and Rathi) to 0.51 (Tharparkar and Sahiwal), whereas in studied dual-purpose breeds, it varied from 0.07 (Kankrej and Mewati) to 0.90 (Belahi and Mewati). Unweighted Pair Group Method with Arithmetic Mean based phylogenetic tree constructed from the Nei’s genetic distances revealed that Sahiwal and Belahi cattle were grouped together in one cluster. However, Kankrej and Mewati were found to be closely related breeds.

KEYWORDS:

• Microsatellite
• STR marker
• Belahi cattle
• genetic differentiation
• phylogeny
• diversity analysis

1. Introduction

Local cattle are integral contributor of food, agricultural power, agrarian culture and heritage and biodiversity as well. (FAO 2007)

Livestock genetic resources in India comprise 190.90 million cattle, 37.28% of the total livestock (BAHS 2014), of which 11.62% of indigenous cattle have been described and categorized into 44 different populations, including 39 distinct and registered cattle breeds. On the utility basis, these have been classified as milch, draught or dual purpose. Among the identified breeds, only 5 breeds are exclusively for dairy purpose and 10 are for milk and draught purpose. Of the total indigenous cattle genetic resource, hardly 6–7% are well defined. These defined breeds are well adapted across varied agro-climatic conditions of India. However, huge cattle genetic resource still remains untapped, which needs to be identified and utilized for milk production, especially the less-known and migratory cattle populations. Moreover, such cattle resources play a crucial role in securing livelihood and in the well-being of the traditional rural livestock keepers. One such newly registered cattle breed, Bos indicus, from North Himalayan foothills is Belahi (Breed Accession Number: INDIA_CATTLE_0532_BELAHI _03038). This rare migratory breed has about 20,000 animals and is well adapted to local environment and is an excellent dual-purpose (milk and draught) cattle genetic resource under extensive management system and provides livelihood security to Gujjar pastoralists of the region (Vohra et al. 2014). Belahi cattle is a distinct population based on body biometry traits, management practices and utility. Besides they possess geographically distinct breeding tract from other extant cattle breeds of the country (Vohra et al. 2012).

Genetic differentiation and diversity analysis of native breeds is an important step in genetic characterization, as it allows the evaluation of unexplored genetic variability. Assessment of genetic variability present in a breed is vital and it should be a basic component for working out conservation strategies and for designing genetic improvement programmes for a particular breed. Recently, microsatellite or short tandem repeat (STR)-based markers have been used to characterize the genetic variability, within and between populations. They are neutral markers and have high variability, widespread distribution throughout the genome and have shown a co-dominant inheritance. Microsatellites are best suited to evaluate genetic diversity (Metta et al. 2004; Li et al. 2005; Yoon et al. 2005; Sodhi et al. 2005, 2006; Pandey, Rekha, Singh, et al. 2006, Pandey, Rekha, Yatender, et al. 2006; Sharma et al. 2013) and relationships among cattle populations from Africa, Europe and Asia (Moazami-Goudarzi et al. 1997; Machugh et al. 1998; Martín-Burriel et al. 1999; Mommens et al. 1999; Beja-Pereira et al. 2003; Jordana et al. 2003). Using a set of bovine-specific microsatellite markers, Mukesh et al. (2004) studied the genetic diversity (both within and between) and relationship among the three important Indian cattle breeds, namely Sahiwal, Hariana and Deoni. Moreover, microsatellite markers had also been used as suitable tool for linkage mapping, identification of quantitative trait loci, parentage determination (Bruford & Wayne 1993). Although the genetic relationship between some indigenous cattle breeds of India have been reported earlier, the important migratory cattle breed, Belahi, was not studied before. The allelic diversity of Belahi breed and its genetic relationship with other indigenous cattle breeds is poorly understood. Hence, the present study was undertaken with the objectives of evaluating within-breed diversity of Belahi cattle and to assess between-breed diversity using multi-locus STR markers and to establish the phylogenetic relationship among the important Indian cattle breeds of India.

2. Material and methods

2.1. Populations studied

A total of 48 random animals were taken from different villages, covering the entire native tract of each breed. During sampling the care was taken that no two individuals are related, that is, all the individual have different sire and dam and are genetically unrelated as far as possible. Within-breed diversity indices were estimated in Belahi cattle and its between-breed diversity was compared with Sahiwal, Gir, Tharparkar, Rathi, Nagori, Mewati and Kankrej breeds of cattle. Belahi (a less-known migratory cattle) derived its name based on typical body colour pattern, which is a mix of reddish brown coat colour with white face, abdomen and extremities (Figure 1(h)). Utility of Belahi cattle is a milch-cum-draught breed (dual purpose). Animals of Belahi cattle are distributed in North Himalayan foothills with its breeding tract in North eastern parts of Haryana, Punjab and Himachal Pradesh (Vohra et al. 2012). Sahiwal cattle are distributed in North-western parts of Punjab and Rajasthan, border areas between India and Pakistan (Nivsarkar & Gupta 2000). Tharparker, Gir and Kankrej cattle are found in Rajasthan and Gujarat states, whereas Nagori and Rathi are the breeds of Rajasthan. Sahiwal is a heavy-built milch breed with brownish red coat colour (Figure 1(a)). Mewati animals are found in North-western parts of Haryana, eastern Rajasthan and Western Uttar Pradesh (Singh et al. 2007) and are usually white in colour with neck, shoulders and quarters of a darker shade (Figure 1(b)). Tharparkar has been developed as a typical breed adapted to the desert conditions and Tharparkar animals are white or light grey in colour (Figure 1(c)). Gir is a milch breed and animals of the breed are pure red/speckled red in colour with basic skin colour as white with red or black patches with typical convex head and pendulous ears (Figure 1(d)) (Singh et al. 2006). Kankrej is a dual-purpose migratory cattle. The animals of Kankrej breed have thick skin, broad chest, straight and powerful body with white, grey and iron grey coat colours (Figure 1(e)) (Vohra et al. 2011). Like Tharparker, Nagori are white to light grey in colour and have a proportionately built body (Figure 1(f)). Rathi is a medium-sized breed with symmetrical body. Rathi animals are usually brown with white patches all over the body (Figure 1(g)).

Figure 1. Different cattle breeds studied: (a) Sahiwal, (b) Mewati, (c) Tharparkar, (d) Gir, (e) Kankrej, (f) Nagori, (g) Rathi and (h) Belahi.

2.2. Sample collection and isolation of genomic DNA

Blood samples were collected aseptically in EDTA-coated tubes from all the animals under study. Genomic DNA was isolated from blood samples following phenol–chloroform extraction method, as described by Sambrook and Russel (2001). DNA concentration was quantified by spectrophotometric method at OD 260: 280.

2.3. Polymerase chain reaction amplification of microsatellite loci

Belahi cattle were genotyped using 16 microsatellite markers. The details of the primers are presented in Table 1. Forward primer of each microsatellite marker was 5′-labelled with FAM (Blue), VIC (Green), NED (Yellow) or PET (Red) fluorescence tags in order to perform fragment length analysis of the polymerase chain reaction (PCR) products with automated DNA sequencer (ABI 3100). For accurate sizing, primers producing fragment size >75 bp were preferred. The microsatellite loci with overlapping size range were labelled with different dyes to facilitate multiplexing. Each PCR was carried out in 15 μl reaction volume consisting of 200 µM of each dNTP, 5 pM of each primer, 1.5 mM MgCl₂ and 1.2U Taq polymerase (Invitrogen, CA). Amplification was performed using MASTERCYCLER EP (Eppendorf, Germany) with an initial denaturation at 95°C for 2 min followed by 30 cycles of 94°C for 60 s, respective annealing temperature (50–65°C) for 60 s and 72°C for 60 s, with a final extension for 10 min at 72°C. After completion of PCR cycle, amplification was confirmed by running a small aliquot of PCR product on 2% agarose gel. To maximize the throughput, products amplified by different primers with different dyes were pooled for one capillary injection. For fragment sizing, GeneScan-500 LIZ™ Size Standard (Applied Biosystems) was used as the internal standard. Internal lane size standard was run with every sample for accurate sizing. The samples were loaded on to ABI PRISM 3100 DNA analyser for genotyping using 36 cm array. For calculation of the molecular weight (size) of different alleles/DNA fragments, Gene Scan software was used.

Table 1. Microsatellite loci information, primer sequences with fluorescent tag and PCR amplification conditions used for analysing individual sample.

Marker	Primer sequences (5′–3′)	Fluorescent dye	Allele size	Annealing temp. (°C)	Genebank Acc. No.	Chromosome
BM1824	GAGCAAGGTGTTTTTCCAATC CATTCTCCAACTGCTTCCTTG	VIC	176–197	55–60	G18394	1
CSRM60	AAGATGTGATCCAAGAGAGAGGCA AGGACCAGATCGTGAAAGGCATAG	PET	79–115	55–65	AF232758	10
CSSM008	CTTGGTGTTACTAGCCCTGGG GATATATTTGCCAGAGATTCTGCA	VIC	179–193	55	U03796	Unknown
CSSM033	CACTGTGAATGCATGTGTGTGAGC CCCATGATAAGAGTGCAGATGACT	NED	154–175	65	U03805	17
CSSM66	ACACAAATCCTTTCTGCCAGCTGA AATTTAATGCACTGAGGAGCTTGG	FAM	171–209	55–65	Unknown	14
HEL1	CAACAGCTATTTAACAAGGA AGGCTACAGTCCATGGGATT	PET	99–119	54–57	X65202	15
ILSTS005	GGAAGCAATGAAATCTATAGCC TGTTCTGTGAGTTTGTAAGC	NED	188–214	58	L23481	7
ILSTS006	TGTCTGTATTTCTGCTGTGG ACACGGAAGCGATCTAAACG	FAM	283–295	55	L23482	7
ILSTS011	GCTTGCTACATGGAAAGTGC CTAAAATGCAGAGCCCTACC	NED	269–283	58	Unknown	14
INRA005	CAATCTGCATGAAGTATAAATAT CTTCAGGCATACCCTACACC	FAM	119–123	54.7	X63793	10
INRA035	ATCCTTTGCAGCCTCCACATTG TTGTGCTTTATGACACTATCCG	FAM	120–140	58	X68049	16
INRA063	ATTTGCACAAGCTAAATCTAACC AAACCACAGAAATGCTTGGAAG	PET	154–168	56	X71507	18
MM12	CAAGACAGGTGTTTCAATCT ATCGACTCTGGGGATGATGT	PET	101–145	50–55	Z30343	9
MM8	CCCAAGGACAGAAAAGACT CTCAAGATAAGACCACACC	NED	116–140	58	Unknown	2
TGLA122	CCCTCCTCCAGGTAAATCAGC AATCACATGGCAAATAAGTACATAC	PET	136–184	55–58	Unknown	21
TGLA227	CGAATTCCAAATCTGTTAATTTGCT ACAGACAGAAACTCAATGAAAGCA	PET	75–105	55–56	Unknown	18

2.4. Statistical analysis

Different within-breed diversity measures were estimated using POPGENE software (Yeh et al. 1999). Allele frequencies estimated were utilized for assessing polymorphism information content (PIC) values as per Botstein et al. (1980). To compute estimates of exact probabilities (p value) for departure from Hardy–Weinberg equilibrium among the microsatellite loci was investigated. The neutrality of all the markers employed in this study was checked with POPGENE software by applying Ewens–Watterson test. The test calculates the quantity F, which is equal to the sum of squared allele frequencies.

Genetic bottleneck tests are rooted in a wider class of population genetic methods aimed at detecting departures from expectations under mutation-drift equilibrium. Population genetic tests of mutation-drift equilibrium typically contrast two different indices of genetic diversity. One measure is expected to be only marginally affected by the underlying process causing deviations from mutation-drift equilibrium and represents a baseline against which the second, more sensitive, diversity index is compared. Examples of genetic diversity indices employed in mutation-drift equilibrium tests that are expected to be less affected by a population bottleneck initially are nucleotide diversity, heterozygosity and the variance and range in microsatellite allele size. In contrast, a greater reduction is expected in the number of alleles and segregating sites, particularly rare ones. In the present study, to estimate the bottleneck events in the investigated Indian cattle breed, two different approaches were followed. The first approach based on the heterozygosity excess consisted of three tests: sign test, standardized differences test and a Wilcoxon signed-rank test. These methods test for the departure from mutation-drift equilibrium based on heterozygosity excess or deficiency. The probability distribution was established using 1000 simulations under three models: infinite allele model (IAM), stepwise mutation model (SMM) and two-phase mutation model (TPM). The test was conducted using Bottleneck 1.2.01 software (Piry et al. 1999). Another test that was used to depict the bottleneck analysis was based on graphical representation of mode-shift equilibrium. This is based on the assumption that in bottlenecked populations one or more of the common allele classes have a higher number of alleles than the rare allele class. This test was re-scaled so that frequency distribution of the allele frequency classes in each population would be based on equal increments of 0.05.

For establishing genetic relationships of Belahi cattle with other cattle breeds, microsatellite data generated for Belahi cattle were compared with microsatellite data already generated in the lab for native breeds adapted to arid and semi-arid regions of north-western India (Sodhi et al. 2011). The cattle breeds and their respective sample sizes (given in parenthesis) analysed in the study are Sahiwal (48) from Punjab and National Dairy Research Institute, Karnal, Haryana; Gir (48) and Kankrej (32) from Gujarat; Tharparkar (47), Rathi (47), Nagori (40) and Mewati (48) from Rajasthan. Different F-statistics estimates – total inbreeding estimate (F), global inbreeding estimate (f) and measurement of population differentiation (θ) which are analogous to Wright’s (1951, 1978) F_IT, F_IS and F_ST, respectively – were estimated using the program FSTAT ver. 2.9.3.2 (Goudet 2002). Means and standard deviations of F-statistics parameters were obtained across breeds by using the Jackknifing procedure over all loci. Population differentiation was also analysed by exact test based on the principle that whether the allelic composition is independent of population assignment. Nei’s D_S and D_A-based methods (Nei 1972) were applied to calculate between-breed genetic distances using microsatellite allele frequency data.

3. Results and discussion

Present investigation is the first detailed genetic analysis of Belahi cattle using microsatellite markers. All microsatellite loci amplified well and exhibited sufficiently high levels of polymorphism, making them effective in detecting genetic variation in the studied cattle breed. Various variability parameters (i.e. observed and effective number of alleles and expected heterozygosity at different microsatellite markers) are presented in Table 2.

Table 2. Genetic heterogeneity estimates at each microsatellite locus in Belahi cattle.

Marker	Na	Ne	Ho	He	Nei’s He	PIC	I*	FIS	Chi value	df
BM1824	5	2.64	0.691	0.629	0.622	0.622	1.122	−0.1103	50934	10
CSRM60	9	3.51	0.810	0.724	0.715	0.715	1.588	−0.1315	30.04	36
CSSM08	6	2.75	0.542	0.643	0.637	0.637	1.305	0.1493	21.245	15
CSSM33	14	7.53	0.854	0.876	0.867	0.872	2.249	0.0150	96.434	91
CSSM66	12	6.85	0.792	0.863	0.854	0.854	2.099	0.0729	70.687	66
HEL1	15	6.68	0.792	0.859	0.850	0.860	2.283	0.0689	143.705**	105
ILST6	11	6.14	0.792	0.846	0.837	0.826	1.999	0.0542	127.552**	55
ILSTS05	9	5.88	0.766	0.839	0.830	0.830	1.933	0.0772	81.935**	36
ILSTS11	5	2.86	0.762	0.658	0.651	0.651	1.263	−0.1712	12.816	10
INRA05	9	3.88	0.708	0.750	0.742	0.742	1.592	0.0428	33.241	36
INRA35	9	4.65	0.787	0.793	0.785	0.785	1.771	−0.0029	60.169**	36
INRA63	6	2.77	0.667	0.646	0.639	0.639	1.243	−0.0428	11.309	15
MM12	10	3.18	0.500	0.693	0.685	0.685	1.556	0.2704	102.098**	45
MM8	6	1.96	0.375	0.496	0.491	0.491	0.969	0.2357	64.793**	15
TGLA122	11	7.47	0.875	0.875	0.866	0.866	2.134	−0.0103	41.767	55
TGLA227	12	1.47	0.313	0.323	0.319	0.319	0.871	0.0211	110.641**	66
Mean	9.31	4.39	0.69	0.72	0.71	0.71	1.62	0.0337	–	–
St. dev.	3.11	2.06	0.17	0.15	0.15	0.14	0.46	0.1197	–	–

Note: N_a, observed number of alleles; N_e, effective number of alleles; H_o, observed heterozygosity; H_e, expected heterozygosity; I*, Shannon’s Index; F_IS, within-population inbreeding estimate; PIC, polymorphism information content; Chi value, Hardy–Weinberg equilibrium; df, degree of freedom of H–W test.

** Significant at the 1% level.

3.1. Markers informativeness

Average PIC value indicates the informativeness of a genetic marker. The PIC estimates showed that most of the loci were highly informative with an overall mean of more than 0.71 ± 0.14 (Table 2) as usually more than 0.5 PIC is desirable for microsatellite markers for better efficiency of marker in analysing genetic diversity among native Indian cattle breed. Locus TGLA227 showed minimum PIC value (0.32), whereas highest PIC values were observed at locus CSSM08 (0.87). Rehman and Khan (2009) reported a little higher PIC (0.74) in Hariana cattle, whereas Pandey, Rekha, Yatender, et al. (2006) and Sodhi et al. (2006) reported a little lower PIC value of 0.66 and 0.60, respectively, in Kherigarh and Tharparkar cattle breeds; a probable reason for this could be the difference in number of microsatellite loci used to evaluate PIC. However, about 87.5% of the markers were observed to be highly informative (PIC > 0.50) and the remaining 12.5% (that is MM8 and TGLA227) were also reasonably informative (0.50 > PIC > 0.31) in Belahi cattle. Our findings are in agreement with Kumar et al. (2003), Metta et al. (2004), Mukesh et al. (2004), Sodhi et al. (2006) and Pandey, Rekha, Yatender, et al. (2006) in various indigenous cattle breeds.

3.2. Genetic diversity and within-breed genetic variation

In total 149 microsatellite alleles were observed across 16 loci surveyed in 48 animals of Belahi cattle with mean number of 9.31 alleles at each locus. The studied 16 microsatellite loci were found to be polymorphic with 5 or more than 5 alleles per locus. The most polymorphic marker was HEL1 with a total of 15 alleles and the least polymorphic markers were BM1824 with 5 alleles (Table 2). Most of the loci showed a similar trend in terms of allele numbers with the earlier characterized cattle breeds. The observed number of alleles in Belahi cattle was higher than observed in other Indian breeds (Mukesh et al. 2004; Pandey, Rekha, Singh, et al. 2006; Pandey, Rekha, Yatender, et al. 2006), namely Hariana (6.5), Sahiwal (5.2), Deoni (5.9), Kenkatha (5.95) and Kherigarh (3.77) cattle. Our findings had shown moderately higher number of alleles; this could be due to the presence of higher heterogeneity present in Belahi population. The effective number of alleles (N_e) across all loci was distinctively less (Table 2) than the observed values ranging from 1.47 (TGLA227) to 7.53 (CSSM33). On analysing the data for heterozygosity, the average expected gene diversity (H_exp) over all loci was sufficiently high and ranged from 0.32 (TGLA227) to 0.87 (CSSM08) with an overall mean of 0.72 ± 0.15. Only two loci (MM8 and TGLA227) exhibited H_obs values less than 0.400, whereas at 13 other loci, H_obs was more than 0.60. The mean values of H_obs were lower at most of the loci than the H_exp, which reflected less heterozygous status of population than it was expected. However, considerable high heterozygosity values indicated the existence of substantial genetic variability in the studied population. Similar high values for observed and expected heterozygosity have been observed for Kherigarh (0.71) cattle (Pandey, Rekha, Singh, et al. 2006) However, many workers reported comparatively lower mean expected heterozygosity, 0.67 in Tharparkar cattle (Sodhi et al. 2006) and Hariana cattle (Rehman & Khan 2009); 0.59 in Deoni cattle (Mukesh et al. 2004) and 0.68 in Kherigarh cattle (Pandey, Rekha, Yatender, et al. 2006). The mean observed heterozygosity values, though lower than the expected values, exhibited failure of significant differences suggesting random mating in Belahi cattle. Shannon’s Information Index, which measures the level of diversity, was sufficiently high in Belahi cattle with an overall mean of 1.62. Within-population inbreeding estimates (F_IS) for Belahi cattle is depicted in Table 2. The estimate ranged between −0.17 and 0.23 and was positive in most of the cases with an average F_IS value of 0.033. Thus, on an average deficiency of 3.37% of heterozygotes existed in Belahi cattle. Our results were similar to those reported by Sharma et al. (2013) in Bachaur, Shahbadi, Purnea and Siri cattle from Northeast of India. Slightly higher inbreeding estimates of 0.14 and 0.18 were reported in Tharparkar and Kherigarh cattle, respectively (Pandey, Rekha, Singh, et al. 2006; Sodhi et al. 2006). However, comparatively higher inbreeding estimates were reported in Sahiwal (0.32), Kenkatha (0.21) and Hariana (0.25) cattle (Mukesh et al. 2004; Pandey, Rekha, Yatender, et al. 2006; Rehman & Khan 2009). Our findings suggests that Belahi cattle has considerable genetic variability with very low inbreeding, compared to other indigenous cattle breeds of India; the probable reason could be the migratory nature and use of sires from diverse locations. All the 10 of 16 markers except INRA35, INRA63, BM1824, ILSTS11, CSRM60 and TGLA122 contributed to this observed heterozygote shortage. Heterozygote deficiency analysis revealed deviations from Hardy–Weinberg equilibrium (p < .01) at 7 of 16 loci in Belahi cattle. It is difficult to predict the exact basis of this departure, but the probable reasons could be the presence of null alleles (Callen et al. 1993), Wahlund effect, physical linkage and epistatic selection.

Ewens–Watterson neutrality test was performed for each of the 16 microsatellite loci. The observed sum of squared allele frequencies (Obs. F) was compared with 95% confidence intervals (CI) for expected sum of squared alleles using 5000 simulations. The overall test for neutrality showed that observed F values were within the lower and upper limits of the 95% CI region for most of the loci. Only for TGLA122 and TGLA227 microsatellite loci (Table 3), the observed F values were marginally outside the upper and lower limits of the 95% CI region. All other loci in were neutral with observed homozygosity values higher than the expected homozygosity (Min F), thus indicating the absence of selection in the population except for two regions which point towards the selection for some economic trait.

Table 3. The Ewens–Watterson test for neutrality at each microsatellite locus in Belahi cattle.

Marker	n	Na	Obs_F	Min F	Max F	Mean^a	SE^a	L95	U95
BM1824	84	5	0.3781	0.2000	0.9093	0.4769	0.0236	0.2605	0.8226
CSRM60	96	9	0.2846	0.1111	0.8277	0.2907	0.0100	0.1633	0.5575
CSSM33	96	14	0.1328	0.0714	0.7658	0.1881	0.0036	0.1139	0.3442
CSSM66	96	12	0.1461	0.0833	0.7971	0.2221	0.0055	0.1287	0.4132
CSSM8	96	6	0.3633	0.1667	0.9013	0.4256	0.0207	0.2305	0.7708
HEL1	96	15	0.1497	0.0667	0.7509	0.1737	0.0033	0.1057	0.3234
ILST6	96	11	0.1630	0.0909	0.8134	0.2450	0.0071	0.1404	0.4620
ILSTS05	94	9	0.1700	0.1111	0.8443	0.2937	0.0100	0.1661	0.5541
ILSTS11	84	5	0.3495	0.2000	0.9093	0.4780	0.0237	0.2633	0.8421
INRA05	96	9	0.2576	0.1111	0.8472	0.2973	0.0106	0.1680	0.5623
INRA35	94	9	0.2150	0.1111	0.8443	0.2936	0.0106	0.1668	0.5677
INRA63	96	6	0.3607	0.1667	0.9013	0.4215	0.0203	0.2303	0.7693
MM12	96	10	0.3147	0.1000	0.8301	0.2679	0.0086	0.1530	0.5163
MM8	96	6	0.5093	0.1667	0.9013	0.4245	0.0202	0.2296	0.7567
TGLA122	96	11	0.1339	0.0909	0.8134	0.2444	0.0067	0.1393	0.4607
TGLA227	96	12	0.6808	0.0833	0.7971	0.2223	0.0057	0.1293	0.4223

^aThese statistics were calculated using 1000 simulated samples.

3.3. Bottleneck analysis

Bottle neck hypothesis was explored in Belahi cattle population. According to this hypothesis, if the population that has experienced the recent reduction, effective population size exhibit a correlation in reduction of allele numbers and gene diversity. In a population at mutation-drift equilibrium (i.e. the effective size of which has remained constant in the recent past), there is approximately an equal probability that a locus shows a gene diversity excess or a gene diversity deficit. Populations that have experienced a recent reduction of their effective population size exhibit a correlative reduction of the allele numbers and gene diversity (H_e, or Hardy–Weinberg heterozygosity) at polymorphic loci. But the allele numbers is reduced faster than the gene diversity. Thus, in a recently bottlenecked population, the observed gene diversity is higher than the expected equilibrium gene diversity. However, the allele number is reduced faster than the gene diversity. Therefore, in a population experiencing bottleneck, the observed gene diversity is higher than the expected gene diversity which is computed from the observed number of alleles, under the assumption of a constant-size (equilibrium) population (Luikart et al. 1998). Three different tests, namely sign rank, standardized differences and Wilcoxon tests under all the three models (Cornuet et al. 1996) of microsatellite evolution (IAM, SMM and TPM) were employed to investigate whether Belahi cattle populations has undergone recent bottleneck (Table 4). Most often, gene diversity excess has been demonstrated only for loci evolving under the IAM. If the locus evolves under the strict SMM, there can be situations where this gene diversity excess is not observed (Cornuet & Luikart 1996). However, few loci follow the strict SMM and as soon as they depart slightly from this mutation model towards the IAM, they will exhibit a gene diversity excess as a consequence of a genetic bottleneck. Because few microsatellite loci follow the strict (one-step) SMM, it is recommend using the TPM test. The TPM is intermediate to the SMM and IAM. Most microsatellite data sets better fit the TPM than the SMM or IAM. However, TPM consists of mostly one-step mutations, with a small percentage (5–10%) of multi-step changes (Luikart et al. 1998). To check whether a population exhibits a significant number of loci with gene diversity excess, three tests (i.e. sign test, standardized differences test and Wilcoxon test) were conducted. Amongst the three tests, Wilcoxon’s test is considered best as it works well with as few as four polymorphic loci and any number of individuals, although 15–40 individuals and 10–15 polymorphic loci is generally recommended to get good results. Conversely, sign test suffers from low statistical power and standardized differences test requires at least 20 polymorphic loci (Luikart & Cornuet 1997). The values for three tests conducted for the Belahi cattle were significant and thereby null hypothesis of mutation-drift equilibrium was rejected. Our findings were similar to those reported by Deepika and Salar (2012), who reported absence of recent genetic bottleneck in grey-coloured cattle breeds (Hariana, Nagori, Mewati, Tharparkar and Kankrej) of India. Thus, Belahi cattle revealed moderate genetic diversity within population; although the population is deviating from mutation-drift equilibrium (except IAM), there was no significant heterozygosity excess indicating the absence of genetic bottleneck in the recent past.

Table 4. Different test for mutation-drift equilibrium in the 16 microsatellite loci in Belahi cattle.

Test	Parameters	IAM	TPM	SMM
Sign test	Observed no. of loci with He excess	13.00	8.00	2.00
	Expected no. of loci with He excess	9.60	9.48	9.43
	p-Value	0.064	0.305	0.000**
Standardized difference test	T2 Value	0.733	−3.343	−10.483
	p-Value	0.231	0.000**	0.000**
Wilcoxon signed-rank test	p-Value (one tail test for He excess)	0.022	0.609	0.999

**p < .01.

Another powerful test of qualitative graphical method based on mode-shift distortion was also utilized to visualize the allele frequency spectra as an indicator for genetic bottleneck. The qualitative descriptor of the allele frequency distribution (mode-shift indicator) which discriminates bottlenecked populations from stable populations (Luikart & Cornuet 1997). No mode-shift (Luikart & Cornuet 1997) was detected in the frequency distribution of alleles, and the alleles with the lowest frequencies (0.01–0.1) were found to be most abundant (Figure 2). The results indicated the absence of any recent bottlenecks in Belahi cattle.

Figure 2. Mode-shift test for bottleneck analysis in Belahi cattle.

3.4. Between-breed diversity and analysis of F-statistics and gene flow for microsatellite loci in different milch and dual-purpose cattle breed

The multi-locus data generated for Belahi cattle were compared with similar data on Sahiwal, Gir, Tharparkar, Rathi, Nagori, Mewati and Kankrej breeds of cattle. Population differentiation was examined by fixation indices F_IT, F_IS and F_ST using the methods of Weir and Cockerham (1984) for each of the 16 analysed loci. As presented in Table 5, mean estimates of F-statistics over loci were F(F_IT) = 0.1985 (total inbreeding estimate), f(F_IS) = 0.0506 (within-population inbreeding estimate) and θ(F_ST) = 0.1558 (population differentiation). The global deficit of heterozygotes across population (F_IT) amounted to 19.85% (p < .05). An overall significant deficit of heterozygotes (F_IS) of 5.1% (p < .05) occurred across the analysed loci. From bootstrap analysis, the true value of F_IS-statistic, with a 95% CI, ranged from −0.09 (CSRM60) to 0.253 (TGLA227). Except for four loci (CSSM08, CSRM60, ILSTS011 and HEL1), all other loci contributed significantly to the heterozygote deficit within population (F_IS). Three loci (TGLA227, MM12 and ILST006) contributed maximally for the observed deficit (Table 5), while all other loci except CSSM08, CSRM60, ILSTS011 and HEL1 significantly affected the global heterozygote deficit (F_IT). The value of breed differentiation obtained in our study was similar to those reported for other indigenous and exotic cattle breeds. For example, 11.3% in Hariana, Deoni and Sahiwal (Mukesh et al. 2004), 11.2% in 7 European cattle (Machugh et al. 1998), 9% in Swiss cattle (Schmid et al. 1999) and 10.7% in 20 North European cattle breeds (Kantanen et al. 2000).

Table 5. Global F-statistics (within-population and between-population inbreeding estimate, and measurement of population differentiation) and gene flow for each microsatellite locus in different milch and dual utility cattle breeds.

Marker	Sample size	FIS	FIT	FST	Nm*
BM1824	684	0.0396	0.1356	0.0999	2.2516
CSRM60	672	−0.0962	0.0736	0.1549	1.3640
CSSM08	576	−0.0136	0.2760	0.2857	0.6250
CSSM33	592	0.0312	0.2104	0.1849	1.1019
CSSM66	674	0.0201	0.0803	0.0614	3.8210
HEL1	422	−0.0014	0.5078	0.5085	0.2416
ILST06	684	0.1868	0.2620	0.0925	2.4524
ILSTS005	670	0.0191	0.1442	0.1276	1.7098
ILSTS11	664	−0.0026	0.2357	0.2377	0.8016
INRA05	680	0.0308	0.0892	0.0603	3.8978
INRA35	662	0.0353	0.1085	0.0759	3.0429
INRA63	664	0.0633	0.2173	0.1644	1.2707
MM12	680	0.1136	0.2253	0.1260	1.7337
MM8	666	0.0932	0.1380	0.0494	4.8120
TGLA122	680	0.0608	0.1293	0.0729	3.1775
TGLA227	620	0.2532	0.4031	0.2008	0.9950
Mean	643	0.0506	0.1985	0.1558	1.3542

Note: F_IS, within-population inbreeding estimate; F_IT, between-population inbreeding estimate; F_ST, measurement of population differentiation and N_m, gene flow, *estimated from F_ST = 0.25(1 − F_ST)/F_ST.

The genetic differentiation values (F_ST) per locus varied from 0.05 (MM8) to 0.51 (HEL1) with an average of 0.156 across all the loci. All the microsatellite loci (p < .05) significantly contributed to this differentiation. The F_ST values implied that overall genetic differentiation among breeds was moderate, but significantly different from zero. Average F_ST value indicated that 15.6% of the total genetic variation corresponded to breed differences while the remaining 84.4% corresponded to differentiation among individuals. The effective number of migrants (N_m) was estimated as 1.35 pointing to moderate gene flow and continuity among different cattle breeds. Our results were not in agreement with Deepika and Salar (2012), who reported lower F-statistic (F_ST = 0.07; F_IS = 0.03; F_ST = 0.07) and higher gene flow (N_m = 4.06) in 10 breeds of grey colour indigenous cattle of India. This could be due to admixture of Hariana cattle with other grey colour indigenous cattle, while it is not the case with Belahi cattle.

3.5. Genetic distance and phylogenetic relationship

Nei’s genetic distances among eight different milch cattle breeds are presented in Table 6. The least distance (0.06) was found between Mewati with Kankrej cattle. The highest (0.90) genetic distance was observed between Mewati and Belahi cattle. The distance among the milch breeds (Sahiwal, Gir and Tharparkar) varied from 0.40 (Gir and Sahiwal) to 0.51 (Tharparkar and Sahiwal), whereas the distance among the dual-purpose breed (Kankrej, Mewati, Rathi and Belahi) varied between 0.36 (Rathi and Belahi) and 0.90 (Belahi and Mewati). Thus, the genetic distance among dual-purpose cattle breeds was found to be higher than that from milch breeds studied.

Table 6. Nei’s genetic identity and genetic distance estimates among different milch and dual-purpose cattle breeds.

	Tharparkar	Rathi	Kankrej	Gir	Nagori	Mewati	Sahiwal	Belahi
Tharparkar		0.7661	0.6758	0.6353	0.6097	0.6498	0.5979	0.5315
Rathi	0.2664		0.6878	0.7319	0.5841	0.6865	0.7002	0.6962
Kankrej	0.3918	0.3742		0.7588	0.8594	0.9354	0.5049	0.4097
Gir	0.4536	0.3122	0.2761		0.7303	0.7221	0.6661	0.6000
Nagori	0.4948	0.5377	0.1515	0.3142		0.8779	0.4917	0.4150
Mewati	0.4311	0.3761	0.0668	0.3255	0.1303		0.4987	0.4043
Sahiwal	0.5143	0.3564	0.6834	0.4064	0.7098	0.7367		0.6312
Belahi	0.6321	0.3621	0.8924	0.5109	0.8796	0.9056	0.4602

Note: Nei’s genetic identity (above diagonal) and genetic distance (below diagonal).

Nei’s genetic distances were used to prepare genogram tree to delineate phylogenetic relationship among cattle populations and are presented in Figure 3. Kankrej and Mewati breeds showed closest relationship. Similarly, Tharparkar and Rathi grouped together in one cluster. Noteworthy finding is that Belahi cattle was clubbed with Sahiwal breed in a separate cluster and genetic closeness of Belahi cattle with Sahiwal (a good milch cattle) indicates the milk potential of this rare migratory cattle of North India. Probable reason for this similarity could be nearby geographical locations of these two breeds. Moreover, the stake holders of both Belahi and Sahiwal breeds of cattle are same (Gujjars) and it is hypothesized that during the course of time Belahi sub-group must have been separated Sahiwal cattle or migrated away towards foothills, leading to difference in climate. However, genetic divergence time and drift further needs to be studied for these cattle breeds.

Figure 3. Genogram showing the different clusters of milch and dual-purpose cattle from India.

4. Conclusion

The present work thus clearly demonstrates the effectiveness of studied STR markers in delineating the existing diversity levels, population structure, phylogeny and inferring the level of inbreeding among populations, even though the studied cattle breeds are evolutionary close populations with less between-breed differentiations. The present study revealed the first-hand information on analysis of genetic diversity in Belahi cattle, thus providing valuable insight to its existing genetic variability. Although Belahi cattle lacks the pan India distribution and has a lower population size (approx. 20,000), despite that, a comparatively higher degree of variability was realized, implying that sufficient genetic diversity exist in Belahi cattle breed. This fact, together with distinctive environmental adaptation (migratory cattle) of Belahi cattle, consolidates the significance of its conservation as a valuable pure breed and its utilization as dual-purpose cattle from North India. Results obtained from the present study will be helpful in designing breeding plan for genetic improvement, as well as for setting up priority in conservation for Belahi cattle.

Acknowledgement

Authors show sincere gratitude to all the livestock keepers who allowed sample collections from their animals. Assistance of Smt. Parvesh Kumari, Technical Officer and Miss Preeti, Senior Research Fellow, from ICAR-NBAGR is duly acknowledged.

Additional information

Funding

Authors wish to thank Director, ICAR-National Bureau of Animal Genetic Resources to provide financial and logistic support to carry out the work. This work was supported by the Indian Council of Agricultural Research, New Delhi [NBAGR_IRC_7.52].