Mitochondrial DNA D‐loop sequence analysis reveals high variation and multiple maternal origins of indigenous Tanzanian goat populations

Abstract The Small East African (SEA) goat are widely distributed in different agro‐ecological zones of Tanzania. We report the genetic diversity, maternal origin, and phylogenetic relationship among the 12 Tanzanian indigenous goat populations, namely Fipa, Songwe, Tanga, Pwani, Iringa, Newala, Lindi, Gogo, Pare, Maasai, Sukuma, and Ujiji, based on the mitochondrial DNA (mtDNA) D‐loop. High haplotype (H d = 0.9619–0.9945) and nucleotide (π = 0.0120–0.0162) diversities were observed from a total of 389 haplotypes. The majority of the haplotypes (n = 334) belonged to Haplogroup A which was consistent with the global scenario on the genetic pattern of maternal origin of all goat breeds in the world. Haplogroup G comprised of 45 haplotypes drawn from all populations except the Ujiji goat population while Haplogroup B with 10 haplotypes was dominated by Ujiji goats (41%). Tanzanian goats shared four haplotypes with the Kenyan goats and two with goats from South Africa, Namibia, and Mozambique. There was no sharing of haplotypes observed between individuals from Tanzanian goat populations with individuals from North or West Africa. The indigenous goats in Tanzania have high genetic diversity defined by 389 haplotypes and multiple maternal origins of haplogroup A, B, and G. There is a lot of intermixing and high genetic variation within populations which represent an abundant resource for selective breeding in the different agro‐ecological regions of the country.

not feasible. They are raised mainly for meat and manure, and as a source of income (Chenyambuga et al., 2012). Additionally, goats play different socio-cultural and traditional roles as gifts, dowry payments, and spiritual offerings. Despite their wide distribution in Tanzania, the indigenous goats have low productivity in terms of growth and milk production. Efforts to improve their productivity have mainly focused on crossbreeding with exotic germplasm which has proved to be unsustainable in the long run. Selective breeding utilizing the indigenous adapted animals would have a sustainable impact on the productivity of the animals (Syrstad & Ruane, 1998).
This requires the animals to be characterized to understand the level of genetic diversity and the relationship between different animal populations (FAO, 1992;FAO/UNEP, 1998). The limited information on the characteristics of indigenous goats in Tanzania is mostly based on the phenotypic features which can be subjective and dependent on the environment which makes it difficult to distinguish between populations (Falconer & Mackay, 1996). Previous efforts to study genetic diversity of indigenous goats in Tanzania have focused on only a few populations from few agro-ecological zones using microsatellite markers (Chenyambuga et al., 2012;Nguluma et al., 2018), making it difficult to draw conclusion on population structure of the indigenous goats of Tanzania in relation to agro-ecological zones of the country. Studying the genetic history of domestic animals can provide crucial clues about past events and main pathways used for commercial transport of the animals in historical times and therefore provide us with information about their genetic structure and relationship within and among populations. Information from such studies is needed in designing and implementing conservation and improvement programs for indigenous goats. This study, therefore, was designed to determine the genetic diversity, maternal origin, and phylogenetic relationship of 12 populations of the indigenous goats in Tanzania using the mitochondrial DNA (mtDNA) D-loop region.

| PCR amplification and sequencing
The primer pair 6807-F 5′-ACCAGAAAAGGAGAATAGCC-3′ and 8173-R 5′-GGTACACTCATCTAGGCATT-3′ flanking the mtDNA D-loop region were designed in this study to amplify the complete D-loop region. PCR amplification was performed in a total volume of 15 μl containing Phusion High-Fidelity PCR Master Mix (Thermo Fisher Scientific Inc.), 0.1 pM of each primer, 2% DMSO (Applied Biosystems), and 50 ng of template DNA. Amplification was carried out in a GeneAmp PCR System 9700 thermal cycler using the following cycling conditions: initial denaturation at 98ºC for 30 s, followed by 35 cycles at 98ºC for 10 s, 62ºC for 30 s, and 72°C for 1 min, with a final extension of 72°C for 10 min. The PCR products were fragment separated on a 1.5% agarose gel pre-stained with 0.25× GelRed (Biotium) and visualized under UV light. The PCR fragment sizes were estimated using O'Gene 100-bp DNA ladder (Thermo Fisher Scientific Inc.). The ExoSAP-IT™ PCR Product Cleanup Reagent (Thermo Fisher Scientific Inc.) was used to purify PCR products before Sanger sequencing.

| Data analysis
Sequences were manually edited and aligned using Clustal W program (Larkin et al., 2007) in CLC Workbench 8.0.3 (CLC Bio-Qiagen).
Twenty-two goat mtDNA reference sequences belonging to six known haplogroups/lineages (Naderi et al., 2007) were downloaded from GenBank and used for haplogroup identification. Sequences of 201 individuals belonging to 29 goat populations from nine other African countries (Table 1) were also downloaded from the GenBank and were included in the analysis. The phylogenetic relationship between individuals and populations was assessed based on the Tamura-Nei distance model (Tamura & Nei, 1993) using the neighbor-joining (NJ) algorithm implemented in MEGA6 (Tamura et al., 2013). The assessment was done using all the generated sequences, the 22 reference sequences and sequences of wild goat ancestors (capra aegagrus, Capra caucasica, Capra cylindricornis, Capra falconeri, and Capra sibirica) with the bootstrap percentage computed after 1000 replications. To confirm the maternal origin and relationship between the Tanzania goat populations and populations from other African regions, median-joining (Bandelt et al., 1999)  History and demographic dynamics were investigated through mismatch distribution patterns (Rogers & Harpending, 1992) complemented by Fu's Fs (Fu, 1997) and Tajima's D (Tajima, 1989) statistics calculated using the infinite sites model in Arlequin v3.5.

| mtDNA D-loop variation and genetic diversity
The complete goat mtDNA D-loop region analyzed in this study corresponds to nucleotide positions 15431 to 16643 of the C. hircus reference sequence (GenBank accession number GU295658.1). From the 627 sequences generated for the 12 Tanzanian goat populations, 276 polymorphic sites were identified of which 223 were substitutions (214 transitions, 9 transversions) and 69 were indels. The polymorphic sites defined 389 haplotypes in total and of these 308 were unique whereas 81 were shared between individuals from at least two different populations. Maternal genetic diversity parameters for Tanzanian goat populations are presented in Table 2. All populations showed high genetic diversity indicated by the haplotype diversity ranging between 0.9485 ± 0.011 in Newala to 0.9945 ± 0.001 in Sukuma goat population while haplotype proportion (number of haplotypes in relation to the sample size) was in the range of 59.2% in Newala to 95% in Fipa populations. Nucleotide diversity was the largest for Songwe (0.0162 ± 0.037) and lowest for Lindi (0.0120 ± 0.031) goat population. Twenty-six individuals had sequences with ambiguous nucleotides at various positions, and Ujiji goats had the highest number of individuals with ambiguous nucleotides (n = 14).
The ambiguous nucleotides were mainly found at the beginning and toward the end of the sequence reads which was lowly variable.
Analysis was done with and without the sequences with ambiguity,  (2014) and it was observed that they did not produce different results and therefore the sequences were retained during analysis.

| Population structure
The AMOVA results shown in Table 3 below revealed that a nonsignificant (p > .05) proportion (2.8%) of the total genetic variation occurred among the goat populations while a significant proportion (97.2%) of the total genetic variation was observed within the Tanzanian goat populations. On the continental level, a significant genetic variation was found within the regions (12%) and among the regions (87.57%).

| Population demographic history
Past population expansion events were inferred based on the patterns of the mismatch distributions ( Figure 5) and neutrality test estimates presented in Table 4.

| Genetic diversity and maternal origin
This study is the first assessment of the genetic diversity and pop- partly be attributed to reasons mentioned previously including high mutation rate of the control region, multiple maternal wild ancestors (Naderi et al., 2007), and capture of a large part of the wild diversity during domestication (Benjelloun et al., 2011). The overall ratio of transitions: transversions (28.7:1) revealed a heavy transition bias even higher than the 17:1 and 16.7 ratios reported before in domestic goats (Joshi et al., 2004;Luikart et al., 2001).
Using the available goat mtDNA haplogroup classification system (Naderi et al., 2007), three haplogroups (A, B, and G) were found among the indigenous Tanzanian goat populations. Predominance of haplogroup A in Tanzanian goat populations was expected as it is the most diverse and ancient (Naderi et al., 2007), and its wide distribution was consistent with the world scenario described in previous studies (Joshi et al., 2004;Liu et al., 2009;Sultana et al., 2003) in all goat breeds in the world. Results obtained in this study further support the concept of multiple maternal origins of domestic goats (Joshi et al., 2004;Luikart et al., 2001). Classification of the Tanzanian indigenous goats into three distinct haplogroups could be interpreted as evidence that they come from three separate genetically distinct

F I G U R E 4
Median-joining network based on the haplotypes of HV1 control region of indigenous Tanzanian goats and goats from nine different African countries. The different colors are related to the geographical origin, and the area of the circle is proportional to haplotype frequency  and Oman, respectively, from the origin of domestication which is considered to be southwest Asia (Payne & Wilson, 1999

| Population structure and phylogenetic relationship
An AMOVA carried out for Tanzania indigenous goat mtDNA revealed a comparative lack of genetic structure, supporting a geographic distribution trends observed in previous studies of mtDNA diversity in goats (Chen et al., 2005;Joshi et al., 2004;Luikart et al., 2001;Naderi et al., 2007Naderi et al., , 2008Sultana et al., 2003;Tarekegn et al., 2018), in worldwide dataset, the Indian subcontinent, China, and Africa. The AMOVA results revealed high variation within the Tanzanian goat populations but very low and insignificant among population variation. The AMOVA results are further supported by a median-joining network, in which the haplotype distribution pattern did not cluster according to population or agro-ecological zones.
These findings are somehow consistent with previous observations by Nguluma et al. (2018) who, based on microsatellite analysis, observed variation among four Tanzanian goat population to be insignificant though slightly higher (8%) than in this study. The low genetic variation observed between Tanzanian goat populations could be attributed to intermixing of animals across geographical regions due to pastoralism (Mwambene et al., 2014;Tenga et al., 2008) genotypes (Mwambene et al., 2014).
Significant population sub-structuring was observed when goat populations from different regions of the African continents were considered in the analysis consistent with what was reported for sub-Saharan African goat breeds with 14% of inter-population variation using microsatellites markers (Chenyambuga et al., 2002). This is due to physical distance and low level of interaction between some regions of Africa especially between East Africa and West Africa.

| Demographic history
The bimodal pattern of distribution observed for each population indicates that the populations were in equilibrium or stable (Hartl, 2004;Rogers & Harpending, 1992)

| CON CLUS ION
This is the first study that investigates the genetic diversity within and between the indigenous goat populations using samples from all the agro-ecological zones where goat production is practiced in Tanzania Information generated in this study provides a valuable tool for conservation strategies and the data herein indicate that for many of the populations, the inherent genetic diversity has been successfully maintained. The adaptive traits and other unique features in these populations need to be well studied, understood, and preserved in the breed improvement programs as a strategy for conservation of animal genetic resources. Authors greatly appreciate the owners of the animals used in this study and the district extension office in the sampling areas for supporting the sampling process.

CO N FLI C T O F I NTE R E S T
There are no conflicts of interest between authors, and permission from each author has been granted.