Skip to main content
PMC home page
PLOS ONE logoLink to PLOS ONE
. 2013 Feb 20;8(2):e56552. doi: 10.1371/journal.pone.0056552

Phylogeography and Domestication of Chinese Swamp Buffalo

Xiang-Peng Yue 1,#, Ran Li 1,#, Wen-Mei Xie 2, Ping Xu 1, Ti-Cheng Chang 3, Li Liu 4, Feng Cheng 5, Run-Feng Zhang 6, Xian-Yong Lan 1, Hong Chen 1, Chu-Zhao Lei 1,*
Editor: Alessandro Achilli7
PMCID: PMC3577850  PMID: 23437167

Abstract

To further probe into whether swamp buffaloes were domesticated once or multiple times in China, this survey examined the mitochondrial DNA (mtDNA) Control Region (D-loop) diversity of 471 individuals representing 22 populations of 455 Chinese swamp buffaloes and 16 river buffaloes. Phylogenetic analysis revealed that Chinese swamp buffaloes could be divided into two distinct lineages, A and B, which were defined previously. Of the two lineages, lineage A was predominant across all populations. For predominant lineage A, Southwestern buffalo populations possess the highest genetic diversity among the three hypothesized domestication centers (Southeastern, Central, and Southwestern China), suggesting Southwestern China as the most likely location for the domestication of lineage A. However, a complex pattern of diversity is detected for the lineage B, preventing the unambiguous pinpointing of the exact place of domestication center and suggesting the presence of a long-term, strong gene flow among swamp buffalo populations caused by extensive migrations of buffaloes and frequent human movements along the Yangtze River throughout history. Our current study suggests that Southwestern China is the most likely domestication center for lineage A, and may have been a primary center of swamp buffalo domestication. More archaeological and genetic evidence is needed to show the process of domestication.

Introduction

Water buffaloes are widespread throughout central and southern China, with the total number of buffaloes in China in 2003 reaching almost 23 million. It is the third largest population of buffalo in the world, representing 17.37% of the total bovine populations in China [1]. Domestic water buffaloes are employed as draught animals, especially in the rice planting regions of Southern China, and they also have a tremendous economic importance due to an increasing demand for their milk, meat, horns, and skin in recent years [2]. The domestic water buffalo in Asia is generally classified into two major subspecies based on body size, outward appearance, biological characteristics, and chromosome karyotype: the river buffalo (2n = 50) and the swamp buffalo (2n = 48) [3]. Previous studies on the water buffalo mtDNA Control Region (D-loop) sequence analysis showed obvious genetic differentiation between the swamp and the river buffalo [4]–[14]. It was proposed that the river buffalo was first domesticated in the western part of South Asia [3], [11], [12], while the swamp buffalo was likely domesticated in Southeastern Asia [15] and China [14], [16]. Based on Cytb gene sequences of the mtDNA, the number of transitions in swamp buffaloes was higher than that in river buffaloes (average number: 7.8 vs 2.4) [17]. A large variety of genetic resources of swamp buffalo, especially in Southwestern, Southeastern and Central China along the Yangtze River, have been described in terms of adaptation to environment and distribution [18]. At present, the Chinese swamp buffaloes have been divided into 14 local types and many populations based mainly on regional distribution [18]. Though there are many indigenous wild swamp buffalo populations in widespread area of China in prehistoric times [19]–[21], studies on genomics and domestication of Chinese swamp buffalo were limited [13], [14], [22]–[25]. Two mtDNA lineages A and B have been determined in the Chinese, Brazilian/Italian and South-East Asian/Australian swamp buffalo [14]. However, due to a restricted sampling from Central and Southern China, the patterns of mtDNA sequence diversity and geographical partitioning of lineage A and lineage B have not been fully resolved. In order to unveil the domestication of Chinese swamp buffalo, a deep phylogeographic analysis with more individuals and populations is required. The current study obtained swamp buffalo mtDNA Control Region sequences from 15 provinces across China, which allow us to systematically investigate the phylogeographic structure of Chinese swamp buffaloes.

Materials and Methods

Specimen Collection and Buffaloes Control Region Sequences Mining

A total of 239 fresh blood samples in this study were collected from different villages covering the entire native tract of each population and the farmers were interviewed in detail to ensure unrelatedness among the sampled individuals. We also wrote a script to retrieve all the swamp buffalo sequences deposited in NCBI database, including 119 sequences of Chinese swamp buffaloes from our previous publication (GenBank Accession Nos. DQ364160-DQ364189 and DQ658051-DQ658139) [14], and 113 D-loop sequences from different geographic regions in China available from GenBank with Accession Nos: AY702618, EF597573-EF597662 and EF053531-EF053552 (Table S1), which are in total 471 sequences of mtDNA Control Region (915 bp), representing 22 local swamp buffalo populations from 15 provinces in China (Fig. 1a, Table S1). We also obtained 5 swamp buffalo mtDNA Control Region sequences from Thailand with 853 bp long (DQ995708-DQ995712), 10 from Italy and Brazil (AY195596-AY195599, AF197218-AF197223) with 924–962 bp long, 8 from Philippines (FJ873676-FJ873683) with 494–497 bp long (Table S1).

Figure 1. Geographic distribution and lineage composition of 22 Chinese swamp buffalo populations.

Figure 1

Open in a new tab

(a). The area of the circle is proportional to the sample size. And the median-joining networks for three different population group defined by Fig. 1a (b).

Amplification and Sequencing

DNA was isolated from fresh blood using a standard extraction method of phenol/chloroform [26]. The leucocyte pellets were prepared from the whole blood samples and subjected to cell lysis in a solution containing SDS and proteinase K. Protein precipitation was performed with phenol/chloroform, and genomic DNA was precipitated with isopropanol and ethanol. 915 bp mtDNA Control region sequences of swamp buffaloes were amplified with primers CB1∶5′-TAGTGCTAATACCAACGGCC-3′ and CB2∶5′-AGGCATTTTCAGTGCCTTGC-3′ [10]. A PCR reaction was performed in 50 µl reaction volume with 20 ng genomic DNA, 10 pM each primer, 0.25 mM dNTPs, 1×PCR buffer with 2.5 mM MgCl2 and 2.0 U Taq DNA polymerase (MBI). A Progene thermal cycler (PE 9600) was used for PCR amplification with the following conditions: initial denaturation at 94°C for 30 sec, followed by 35 cycles at 94°C for 1 min, 56.5°C for 1 min, and 72°C for 1 min and the final extension at 72°C for 4 min. The amplified products were purified using Wizard PCR Preps DNA purification kit (Promega) according to the manufacturer’s instructions. PCR products were sequenced using four internal primers: CB3: (5′- CCA TCA ACA CAC CTG ACC -3′), CB4: (5′- GCG AGG ACG GAT TTG ACT -3′), CB5: (5′- CAT AAC ATT AAT GTA ATA AGG GC -3′) and CB6: (5′- CCA TTC GGA GTA GTA GGG TC -3′). The sequencing reactions were carried out on an ABI 3730 automated sequencer. All new sequences produced in this study were deposited in the GenBank (accession nos GQ260217–260455).

Data Analysis

All the mtDNA Control Region sequences were edited using the DNASTAR 5.0 package (DNASTAR, Madison, WI). The ClustalX package [27] was used for multiple alignments. All mtDNA insertions/deletions in the alignment were excluded from the analyses. Identical sequences were considered as the same haplotype. Haplotype diversity (h), nucleotide diversity (π), the Fst and AMOVA were estimated using the Arlequin 3.5 package [28]. To demonstrate the phylogenetic clusters, the MEGA5.0 [29] was applied to construct Maximum likelihood (ML) trees, using Hasegawa-Kishino-Yano model with additional parameter of 1000 bootstrapping replicates, Gamma distribution (+G) with 8 rate categories, and evolutionarily invariable (+I).

In the ML trees, reliability percentages (RP) above 50% can be considered as strong supported. The Bayesian phylogenetic tree was also constructed using TOPALi 2.5 [30]. A complete mtDNA sequence (GenBank accession no: NC_006853) of cattle (Bos taurus) was used as an out-group in the phylogenetic tree construction. Then, median-joining network (MJ) was generated using the NETWORK 4.1 program [31]. All the gaps were excluded in constructing median-joining networks and parameters were set to a weight of two and threshold value of one. The network calculations were run with different values for ε [31], however the best results were obtained with ε = 0. Frequencies of haplotypes were converted into proportional areas in the figures. The pairwise mismatch distribution between buffalo sequences were generated using Dnasp 5.0 program and the pairwise FST values were displayed by multidimensional scaling (MDS) using SPSS11.0 [12]. Divergence time and most recent common ancestor were calculated using MEGA 5.0. The mutation rate was assumed to be 32%/million years [32].

Results

Phylogenetic Tree Construction and Haplotype Diversity of Chinese Buffaloes

The comparison of all 471 mtDNA Control Region sequences (915 bp) from 22 Chinese buffalo populations showed 157 haplotypes. The ML tree and Bayesian tree constructed with the 157 haplotypes clearly showed two main branches of swamp and river buffalo, respectively (Fig. 2, Fig. 3). The river buffalo branch included 9 haplotypes, representing 16 individuals from the Guangxi population, indicating the introgression of the river buffalo into the swamp buffalo due to importing the river buffalo to improve the dairy production of local swamp buffaloes. Whereas 148 swamp buffalo haplotypes can be grouped into two distinct mtDNA lineages A and B defined by Lei et al. [13], [14]. The comparison of 148 haplotypes representing 454 swamp buffalo sequences revealed 114 polymorphic nucleotide sites, including 99 transitions, 10 transversions and 5 coexistences of transition and transversion (Table S2). It is much higher than the 64 sites in the research by Lei et al. [14]. The 9 river buffalo haplotypes indicated 31 polymorphic nucleotide sites with 30 transitions and 1 transversion (data not shown). The ratios of transitions versus transversions in both swamp and river types were high, revealing a strong transition bias in buffaloes. For the swamp buffalo branch, five haplotypes were found to be the main haplotypes present in more than 10 samples. The predominant haplotype HA9 (lineage A) was observed 164 times with the highest frequency (36.04%, 164/455). The haplotypes GZ34, DH11 (lineage A), FA15 and HA4 (lineage B) were observed 22, 21, 31 and 13 times, respectively. Of the total 148 swamp buffalo haplotypes detected, 98 fall within the lineage A (352 sequences) and 50 are encountered within lineage B (103 sequences) (Table S3). Among above 148 haplotypes, 108 haplotypes are unique, only observed once. Meanwhile, 66 unique haplotypes (66/98 = 67.34%) belong to lineage A, and 42 unique (42/50 = 84.00%) haplotypes belong to lineage B. In contrast, 5 out of 9 (5/9 = 55.56%) haplotypes in river lineages occurred only once. The estimated divergence time of river and swamp type was 34998 years ago. In the swamp type, two separate maternal lineages A and B have been identified with an estimated divergence time of 17924 years.

Figure 2. ML tree of 157 Chinese buffalo haplotype.

Figure 2

Open in a new tab

Figure 3. Bayesian tree based on 157 haplotypes this study.

Figure 3

Open in a new tab

In order to focus on the phylogenetics among Chinese swamp buffaloes, the median-joining network of 455 Chinese swamp buffalo samples was constructed after removing the 16 river buffaloes from the GX population (Fig. 4). As expected, the network clearly identified lineages A and B which were separated by 26 mutation steps. Interestingly, two small lineages with haplotype DH11 representing 21 individuals and GZ34 representing 22 individuals can be found in lineage A, and there are 6 and 2 mutation steps connecting with the main haplotype HA9 (164 individuals, 46.59% of the total of lineage A), respectively. Lineage A showed an obvious star-like phylogeny and was detected in 22 Chinese swamp populations (Table 1, Fig. 4). However, lineage B comprised two sublineages in the network, named B1 and B2, which are not supported in the ML tree, because of low bootstrap value (Fig. S1). The sublineages B1 and B2 also showed a star-like phylogeny with centers being HA4 (representing 13 individuals) and FA15 (representing 31 individuals), respectively. There are 20 steps between the two centers. The sublineage B1 (representing 43 individuals, 41.75% of the total of lineage B) and B2 (60 individuals, 58.25% of the total of lineage B) included 27 and 23 haplotypes, respectively (Fig. 3, Table S3). Overall, the mean number of haplotype diversity for the lineage B (0.8932±0.0253) is higher than that for the lineage A (0.7752±0.0239).

Figure 4. MJ network of Chinese 454 swamp buffaloes.

Figure 4

Open in a new tab

The area of the circle is proportional to sample sizes.

Table 1. Source and genetic diversity index of 455 Chinese swamp buffalo samples.

Population Code Geographic distribution N K Lineageobserved Haplotype diversity(SE) Nucleotide diversity(SE)
Haizi HA Southeastern region (SE), Jiangsu 25 17 A, B1, B2 0.8800±0.0637 0.0183±0.0094
Shanqu SQ Southeastern region (SE), Jiangsu 5 5 A, B1 1.0000±0.1265 0.0157±0.0099
Dongliu DL Southeastern region (SE), Anhui 23 14 A, B1 0.9012±0.0512 0.0189±0.0097
Wenzhou WZ Southeastern region (SE), Zhejiang 30 14 A,B1, B2 0.8345±0.0638 0.0159±0.0081
Xinglong XL Southeastern region (SE), Jiangxi 35 19 A, B1, B2 0.8420±0.0614 0.0102±0.0053
Fu’an FA Southeastern region (SE), Fujian 22 9 A, B2 0.8268±0.0611 0.0202±0.0104
Binhu BH Central region (CR), Hunan 12 9 A, B1, B2 0.9091±0.0795 0.0204±0.0109
Jianghan JH Central region (CR), Hubei 36 25 A, B1, B2 0.9349±0.0327 0.0162±0.0083
Enshi ES Central region (CR), Hubei 20 9 A, B2 0.8579±0.0537 0.0163±0.0085
Xiajiang XJ Central region (CR), Jiangxi 15 10 A, B1 0.8952±0.0704 0.0068±0.0039
Xinfeng XF Central region (CR), Jiangxi 5 2 A, B1 0.4000±0.2373 0.0149±0.0095
Xinyang XY Central region (CR), Henan 29 12 A, B1, B2 0.6626±0.1010 0.0086±0.0046
Hanzhong HZ Central region (CR), Shaanxi 27 12 A, B1, B2 0.8006±0.0763 0.0144±0.0075
Guizhou GZ Southwestern region(SW), Guizhou 29 16 A, B1, B2 0.8966±0.0469 0.0191±0.0097
Dechang DC Southwestern region (SW), Sichuan 12 9 A, B2 0.9545±0.0467 0.0148±0.0081
Fuling FL Southwestern region (SW), Chongqing 27 14 A, B2 0.8803±0.0527 0.0194±0.0099
Guangxi GX Southwestern region (SW), Guangxi 13 9 A, B1, B2 0.9231±0.0572 0.0223±0.0118
Fuzhong FZ Southwestern region (SW), Guangxi 5 5 A, B1 1.0000±0.1265 0.0159±0.0100
Xilin XN Southwestern region (SW),Guangxi 20 12 A, B1, B2 0.8737±0.0653 0.0188±0.0098
Diandongnan DDN Southwestern region (SW), Yunnan 29 4 A, B2 0.9000±0.1610 0.0168±0.0106
Yanjin YNJ Southwestern region (SW), Yunnan 5 5 A, B2 1.0000±0.1265 0.0206±0.0129
Dehong DH Southwestern region (SW), Yunnan 31 17 A, B1, B2 0.8796±0.0481 0.0132±0.0068
Open in a new tab

Note: n = sample size; k = number of haplotypes, all 22 populations of China were grouped into three regional groups.

Population Structure

To further determine the location of the probable center for domestication of the two swamp buffalo mtDNA lineages among the three main hypothetical regions of swamp buffalo domestication, we performed detailed comparative analyses of the diversity for each lineage. We separately performed several AMOVA analyses at different hierarchical levels (Table 2),and analyzed nucleotide and haplotype diversity within lineage A and B for the three defined regional groups (Table 3), including Southeastern (DL, HA, SQ, XL, WZ, FA), Central (HZ, XY, JH, BH, ES, XJ, XF) and Southwestern China (DN, XN, GX, GZ, DC, YJ, DH, FZ, FL). FST values were calculated for all the possible population pairs. In lineage A, the Matrix of significant FST P values showed significant differences between Southwestern and Southeastern (P<0.01), Southwestern and Central (P<0.01)and no significant differences existed between Southeastern and Central (P>0.05) (Table 4). Notably, increased haplotype and nucleotide diversity was observed in lineage A of swamp buffaloes in Southwestern China (Table 3). The nucleotide diversity of the Southwestern group (0.0033±0.0019) is higher than those in the remaining two regional groups (Southeastern: 0.0022±0.0014, Central: 0.0017±0.0013) (Table 3). These analyses allowed us to hypothesize Southwestern China as the most likely location for the domestication of lineage A, and hence, a strong candidate for the primary center of swamp buffalo domestication. Indeed, lineage A is by far the most frequent and was observed in all populations in this study (Table 1).

Table 2. Analyses at different hierarchical levels between Southwestern group and remaining groups.

Grouping Within populations (%) Among populations withingroups (%) Among groups (%) FST
No grouping in Chineseswamp buffalo 99.43 0.57 (P = 0.35) 0.006(P = 0.29)
Southwestern versus remaining 99.16 (P = 0.35) −0.06 (P = 0.36) 0.90 (P = 0.09)
Southwestern (no DH)versus remaining 98.91 (P = 0.37) −0.23 (P = 0.33) 1.32 (P = 0.05)
Southwestern (+FA) versus remaining 98.65 (P = 0.37) −0.60 (P = 0.35) 1.95 (P = 0.02)
Open in a new tab

Table 3. Diversity comparison among the three hypothesized domestication centers in China.

Lineage Region N K (k) Haplotypediversity(SE) Nucleotidediversity(SE) Fu’s FS test Tajima’s D test
Lineage A Southeastern 106 38(29) 0.9518±0.0101 0.0022±0.0014 −27.528 (P = 0.00) −2.353(P = 0.001)
Central 117 43(31) 0.9565±0.0096 0.0017±0.0013 −28.250 (P = 0.00) −2.296(P = 0.00)
Southwestern 129 49(38) 0.9721±0.0059 0.0033±0.0019 −26.516 (P = 0.00) −2.278(P = 0.00)
Lineage B Southeastern 34 20(16) 0.9768±0.0167 0.0071±0.0038 −17.772 (P = 0.00) 0.248(P = 0.66)
Central 27 14(11) 0.9601±0.0259 0.0066±0.0036 −9.717 (P = 0.001) 0.865(P = 0.83)
Southwestern 42 22(17) 0.9826±0.0093 0.0069±0.0037 −21.166 (P = 0.00) −0.867(P = 0.21)
Open in a new tab

Note: N = sample size, K (k) = the number of haplotypes (unique haplotype).

Table 4. Population pairwise FST, matrix of significant Fst P values.

Lineage Region Southeastern Central Southwestern
Lineage A Southeastern 0.000
Central −0.001 0.000
Southwestern 0.035** 0.042** 0.000
Lineage B Southeastern 0.000
Central 0.003 0.000
Southwestern 0.100** 0.023 0.000
Open in a new tab

Note: **stand for extremely significant level = 0.01.

However, a more complex pattern of diversity was detected in lineage B using the similar analysis (Table 3 and Table 4), which prevent the unambiguous pinpointing of a single region for the domestication of the lineage B.

Population Expansions

The star-like phylogeny of lineages A and B (sublineages B1 and B2) in the network (Fig. 4) is a sign of population expansions. The sign of a population expansion of lineages A and B was also demonstrated by the bell-shaped curve of the mismatch pairwise distributions (Fig. 5). There was a large negative Fu’s Fs value (−21.80, P<0.005) for the complete dataset of Chinese swamp buffaloes, showing that there were two major peaks at around 1 and 32. These results suggested at least two expansion events in the demographic history of Chinese swamp buffalo populations. Lineages A and B yielded significantly large negative Fu’s Fs test values of −26.32 (P<0.005) and −24.73 (P<0.005), respectively. For sublineages B1 and B2, the Fu’s Fs values are −21.09 (P<0.005) and −13.27 (P<0.005), respectively (Table S3). These analyses uniformly indicated that lineages A and B had undergone three population expansion events.

Figure 5. Mismatch distribution and test of neutrality for three hypothesized domestication centers of Chinese buffalo (Southeastern, Central and Southwestern China).

Figure 5

Open in a new tab

Phylogenetic Analysis of Chinese Buffaloes and Swamp Buffalos Outside China

In order to investigate the relationship of swamp buffaloes between within and outside China, we compared our data with published mtDNA Control Region sequences of swamp buffaloes, raised in Thailand, Philippines, Brazil and Italy. For a complete alignment we then truncated our sequences to 495 bp. ML tree revealed that all the buffalo sequences outside China are incorporated in lineage A, no sequences fell into lineage B (Fig. 6).

Figure 6. ML tree of Chinese buffalo and Brazil, Italy, Philippine and Thailand buffalo.

Figure 6

Open in a new tab

All the buffalo haplotype outside China are named with NCBI accession number.

Discussion

Genetic Evidence

Previous studies observed that genetic diversity of modern population generally decreases with increasing distance from the domesticated center [33], [34]. In our present study, the genetic diversity comparison of Southwestern Chinese buffalos (DN, XN, GX, GZ, DC, YJ, DH, FZ and FL) with other hypothesized domestication centers (Southeastern and Central China) indicated that Southwestern China may be the most likely location for the domestication of the lineage A, and, thus, a strong candidate for the primary center of swamp buffalo domestication.

As the animals from domestication center can comprise the backbone of the network structure, show lower model value of pairwise difference, and harbor larger amount of unique haplotypes, Chen and his colleagues [35] use these three genetic indexes to support Indus valley as the primary domestication center of domesticated zebu. In this study, the same approaches were used to figure out where is the potential domestication center of swamp buffalo. First, the Southwestern populations revealed a complete backbone structure for the network of the lineage A (Fig. 1b and Fig. 3); second, the modal value [35] of the mtDNA mismatch distribution for the Southwestern China region is higher than the modal value generated for the lineage A of the Central and Southeastern China regions, suggesting that the expansion of the lineage A in Southwestern China region predates that within the Central and Southeastern China regions (Fig. 5); last but not least, the Southwestern China region displays a larger number of unique haplotypes for this lineage A compared with both the Central and Southeastern China regions (Table 3).

It is also worth noticing that three populations (FL, GX, and YJN) within six swamp buffalo populations (BH, FA, GZ, FL, GX, and YJN) with extremely high nucleotide diversity (>0.0190) among 22 Chinese swamp buffalo populations belonged to Southwestern group (Table 1). In fact, though BH and FA belonged to Central and Southeastern group respectively, BH population is native to Northern Hunan province which is adjacent to Guangxi region [18] (Fig. 1). Therefore, it is reasonable to propose that Southwestern region in China appears to be the most likely area where the first domestic Chinese swamp buffalo appeared. To further prove the assumption, all 22 swamp buffalo populations were grouped into two geographical regions (Southwestern versus Central+Southeastern), 0.90% (P = 0.09) of variation was among groups. In the previous microsatellite makers study, DH population revealed very high genetic diversity. Therefore, we do not think the DH samples observed in this study can stand for really pure DH swamp buffalo population [24]. So, the nucleotide diversity value in the DH population was definitely underestimated. If the DH population was removed from the Southwestern group, two geographical regions (Southwestern versus Central+Southeastern), 1.32% (P = 0.05) of variation was among groups. The new result also supported that the Southwestern region in China seems to be the most likely origin site of the first Chinese domestic swamp buffaloes. As expected, two geographical regions (Southwestern+FA population versus Central+Southeastern), 1.95% (P = 0.02) of variation was among groups (Table 2). The result further supported that the Southwestern China is most likely to have first witnessed the domestication of Chinese swamp buffaloes.

However, when similar analyses were performed for the lineage B, a more complex pattern of diversity was detected, whereby some haplotypes derived from the presumed ancestral lineage B haplotypes were observed at high frequencies, whereas other derivative haplotypes were observed only in specific locales. The most notable case is the lineage B haplotype observed to have two sublineages, B1 and B2 in the network, were not supported by ML and Bayesian tree for the low bootstrap value (Fig. 2 and 3). Although phylogenetic trees produced on the basis of sequenced genes or genomic data can infer evolutionary insight, they have important limitations. They do not necessarily accurately represent the species evolutionary history [36]. Phylogenetic trees also have trouble depicting micro-evolutionary events. Phylogenetic networks are used when bifurcating trees are not suitable, due to these complications which suggest a more reticulate evolutionary history of the organisms sampled [37]. In addition, our population expansion analysis indicated that lineage B had undergone two population expansion events. For the above case, we are inclined to believe that the lineage B has two sublineages, B1 and B2, shown up in the network.

Our phylogeographic analysis revealed a weak geographical structure of mtDNA variation in Chinese swamp buffalo types. This phenomenon was also detected in goats including Chinese domestic goats [38]–[40], Chinese sheep [41], Chinese donkeys [42] and Chinese yaks [43]. A possible explanation is that there was strong gene flow among Chinese swamp buffalo populations along the Yangtze River valley and adjacent rivers throughout history. As the economy develops and modernizes in China, swamp buffaloes are seldom herded in the developed regions in Southern China. So our samples were collected from remote mountainous villages of geographical distribution locations. We do not think that the very recent transportation of buffaloes among different regions in Southern China leads to any significantly geographical structure in Chinese swamp buffalo populations. It would be reasonable to conclude that frequent trade and transportation of buffaloes along the Yangtze valley happened throughout history. The fact that Chinese scientists classified all Chinese swamp buffaloes as one breed based on the similarity of body size, outward appearance, coat and biological characteristics is appropriate. Chromosome karyotype [18] also supports the weak geographical structure in Chinese swamp buffalo populations.

The genetic history of the modern Chinese swamp buffalo was likely to have been associated with inter-regional cultural interactions and population movements in the past. According to the historical records, several immigration upsurges of the human population occurred throughout over 3000 years of Chinese history due to warfare [44]. The immigration movements of human population especially in Southern China caused a parallel flow and merging of swamp buffaloes. It is reasonable that people in Southern China would prefer to take along buffaloes during their migratory movements, as buffaloes have the ability to survive in the similarly humid climate of Southern China and contribute immensely to the agricultural economy through milk, meat, hides and draught power. Buffaloes have become an indispensable part of the property, possession and agricultural tool of rural farmers for centuries [2]. In addition, frequent trade of buffaloes which play a crucial role in rice cultivation in the Southern China further improved the homogeneity of swamp buffaloes in China. Then, all these events contributed greatly to the formation and development of a weak phylogeographical structure of swamp buffaloes. However, even though extensive gene flow existed, we also detected the weak differentiation among 22 Chinese swamp buffalo populations (Fig. 7). Our result was consistent with that based on microsatellite markers [24].

Figure 7. Multidimensional scaling plot of pairwise FST values between 22 Chinese swamp buffalo populations.

Figure 7

Open in a new tab

Archaeological and Ethnographic Evidence

The wild ancestor of B. bubalis was B. arnee. The present distribution of B. arnee is in northeastern India, Nepal, Thailand and Sri Lanka [45], [46], but its distribution in prehistory was much broader, stretching from northwestern South Asia to Mainland Southeastern Asia, based on archaeological remains [47], [48]. Zooarchaeological studies indicate that B. bubalis was first domesticated in northwestern part of South Asia before the second millennium BC [48], [49]. These buffalo were likely the river type, as suggested by the modern DNA study [11], [12].

A DNA study of modern swamp buffalo from Southeastern Asia suggests that Thailand appears to have been the center of domestication for this species [15]. Nonetheless, the previous study [15] analyzed shorter Control Region (158 bp) of the much smaller sample size (80 animals) than our current study, which may lose useful genetic information. Furthermore, this study did not incorporate swamp buffaloes outside Southeast Asia in phylogenetic analysis. However, our ML tree construction of Chinese buffaloes and Southeastern Asian buffaloes mt DNA sequences revealed that all the buffalo sequences outside China are incorporated in lineage A, no sequences fell into lineage B. That means China has higher probability as the domestication center of swamp buffaloes; however, more buffalo samples from Southeastern Asia should be collected to confirm this conclusion. The earliest dates for the appearances of domestic swamp buffalo remains in Southeastern Asia fall into the mid first millennium BC, exemplified by Ban Tamyae in central Thailand (600–200 BC) [50], Ban Chiang in north Thailand (around 300 BC) [51], and Phum Snay in northwestern Cambodia (200 BC–AD 240) [52]. This animal appears to have been used for ploughing [51], burial offering, and consumed as food [52].

In China, no B. arnee has been found, and no securely identified B. bubalis remains dating to the Neolithic or Bronze Age have been reported in archaeological record [19]. The buffalo remains uncovered from Neolithic contexts belong to B. mephistopheles whenever identifiable to the species. When B. bubalis was mentioned in the archaeological reports, it was normally used as a general term for Bubalus sp.

If swamp buffalo was first domesticated in Southwestern China, as suggested by the DNA data discussed above, the most crucial areas for finding the archaeological evidence are Yunnan and Guanxi provinces. There are five prehistoric sites in these regions reporting the discovery of buffalo remains (Table S4). The buffalo from Tangzigou in Baoshan, Yunnan (ca. 6000 BP) was identified as B. mephistopheles, based on the morphology of a skull [53]. The B. bubalis remains were reported as being unearthed at Haimenkou in Jianchuan, northwest Yunnan, dating to the late part of the first millennium BC [54]. However, the dates are controversial and the deposits of the site may have been disturbed [55]. Therefore, the data from Haimenkou are not reliable. Buffalo bones from three other sites were assigned to Bubalus sp. mainly because the absence of horncorns in the faunal assemblages which otherwise would help us identify buffalo species. The lack of well identified buffalo remains from Southwestern China prevents us from using current zooarchaeological data to either support or reject the hypothesis regarding the domestication of buffalo based on DNA research.

Furthermore, there are plenty of records from ancient texts and art presentations for the first appearances of domestic buffalo in Southwestern China. As recorded in “Nanman Xinanyi zhuan” in Hou Han Shu, during the Eastern Han period (AD 25–220), Yongchangjun (Yunnan region today) “paid elephant ivory, water buffalo and zebu as tribute” to the Han court. This indicates that Yunnan was known, at latest, by the 1st century AD as a region which produced domestic buffalo. Bovid images, some identifiable as wild and domestic buffalo, were depicted on rock paintings in Cangyuan, Southwestern Yunnan, dating to AD 1st–6th Centuries. Sculptures of domestic buffalo, depicted as drought and mounted animal, are seen on bronze drums (AD 3rd –6th Century) from Guangxi. Clay models of buffalo-plough-rice paddy complex have been found in Guangdong, dated to the Western Jin dynasty (AD 265–316) [20]. These data suggest that domestic buffalo appeared in Yunnan probably by the first century AD and gradually spread to other parts of southern China. More archaeological research is needed to understand whether or not this animal went through a process of domestication before the 1st century AD in Southwestern China.

Buffalo sacrifice has been reported as an important ritual practice among several ethnic groups in Southwestern China, particularly in southern part of Yunnan. In the Wa people, for example, buffalo sacrifice is conducted for major ritual events and festivals [56]. It is notable that the ancient rock art with buffalo images was found in the same area where the Wa people live today. Similar buffalo sacrifice ritual has been recorded from many areas in Southeastern Asia [57], suggesting a close cultural tradition associated with buffalo in the region including Southwestern China.

Cultural interactions between Southwestern China and Southeastern Asia existed in ancient times, through many trade networks [58], including the Southwestern Silk Road connecting Sichuan via Yunnan and Burma with South Asia [59]. The material items transported through these routes include precious stones, salt, animals and tea [59]. These inter-regional interactions, likely also including migrations of human populations, may have facilitated the movement of water buffalo between Southeastern Asia and Southwestern China. It is unclear, however, which directions that buffalo was moved in this broad region. Based on the current archaeological information we cannot exclude the possibility that buffalo was first domesticated in Southwestern China and spread to other areas.

Conclusion

Previous studies have shown that China may have been the domestication center of swamp buffalo [11], [12], [23], [25]. This survey examined the mtDNA Control Region diversity of 455 individuals from 22 Chinese swamp buffalo populations and 16 river buffaloes from Chinese Guangxi population. Phylogenetic analyses revealed that there were two previously defined lineages A and B in Chinese swamp buffalo populations, showing multiple maternal origins of swamp buffalo. Based on genetic evidence, the results of our current study support that Southwestern China is most likely to be the center where Chinese domestic swamp buffaloes first appeared. Since there is no evidence in the archaeological record suggesting that the process of buffalo domestication took place in China, and domestic buffalo occurred in Southeastern Asia (by the 2nd century BC) earlier than it did in China (by the 1st century AD), it is still premature to claim that Southwestern China was the original center of buffalo domestication. There are three possible scenarios: (1) swamp buffalo was domesticated in multiple centers, including Southeastern Asian and Southwestern China; (2) swamp buffalo was first domesticated in Southeastern Asia, and then introduced to Southwestern China before spreading to the adjacent regions; and (3) swamp buffalo was first domesticated in Southwestern China, but archaeologists have not discovered the early buffalo remains. More DNA tests on buffalo from Southeastern Asia, particularly Myanmar, Laos, Vietnam, and Thailand, need to be done in order to compare with the data from China. The multidimensional display of population pairwise FST values and AMOVA analysis showed weak geographical structuring and weak genetic differentiation because of long-term strong gene flow among swamp buffalo populations caused by extensive transportation of buffaloes and frequent human movements along the Yangtze River in history. The lineage of river buffalo detected in Guangxi swamp buffaloes suggested that the imported river buffalo breeds led to the mixture of Chinese local swamp buffaloes.

Supporting Information

Figure S1

ML tree of Chinese swamp buffalo without cutting off reliability percentages (RP) below 50%.

(TIF)

Click here for additional data file. (1MB, tif)
Table S1

Swamp buffalo D-Loop sequences retrieve from NCBI database.

(XLSX)

Click here for additional data file. (18.9KB, xlsx)
Table S2

Polymorphic nucleotide sites of 148 swamp buffalo haplotypes.

(DOC)

Click here for additional data file. (149.5KB, doc)
Table S3

Total genetic diversity indexes in Chinese water buffaloes.

(DOC)

Click here for additional data file. (34.5KB, doc)
Table S4

Archaeological sites associated with buffalo remains dating to the Holocene period.

(DOCX)

Click here for additional data file. (20.5KB, docx)

Funding Statement

This work was supported by the National Natural Science Foundation of China (Grant Nos. 31272399, 31072001) and Agricultural Science & Technology Project of Shaanxi Province (No.2012K02-02). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

  • 1.FAO (2003) www.fao.org/dad.it.
  • 2. Nanda AS, Nakao T (2003) Role of buffalo in the socioeconomic development of rural Asia, current status and future prospectus. Animal Science Journal 74: 443–455. [Google Scholar]
  • 3. Cockrill WR (1981) The water buffalo, a review. British Veterinary Journal 137: 8–16. [DOI] [PubMed] [Google Scholar]
  • 4. Amano T, Miyakoshi Y, Takada T, Kikkawa Y, Suzuki H (1994) Genetic variants of ribosomal DNA and mitochondrial DNA between swamp and river buffaloes. Animal Genetics 25 (Suppl 1)29–36. [DOI] [PubMed] [Google Scholar]
  • 5. Tanaka K, Yamagata T, Masangkay JS, Faruque MO, Vu-Binh D, et al. (1995) Nucleotide diversity of mitochondrial DNAs between the swamp and the river types of domestic water buffaloes, Bubalus bubalis, based on restriction endonuclease cleavage patterns. Biochemical Genetics 33: 137–148. [DOI] [PubMed] [Google Scholar]
  • 6. Tanaka K, Solis CD, Masangkay JS, Maeda K, Kawamoto Y, et al. (1996) Phylogenetic relationship among all living species of the genus Bubalus based on DNA sequences of the cytochrome b gene. Biochemical Genetics 34: 443–452. [DOI] [PubMed] [Google Scholar]
  • 7. Barker JSF, Moore SS, Hetzel DJS, Evans D, Tan SG, et al. (1997) Genetic diversity of Asian water buffalo (Bubalus bubalis), microsatellite variation and a comparison with protein-coding loci. Animal Genetics 28: 103–115. [DOI] [PubMed] [Google Scholar]
  • 8. Barker JSF, Tan SG, Selvaraj OS, Mukherjee TK (1997) Genetic variation within and relationships among populations of Asian water buffalo (Bubalus bubalis). Animal Genetics 28: 1–13. [DOI] [PubMed] [Google Scholar]
  • 9. Navani N, Jain PK, Gupta S, Sisodiaand BS, Kumar S (2002) A set of cattle microsatellite DNA markers for genome analysis of riverine buffalo (Bubalus bubalis). Animal Genetics 33: 149–154. [DOI] [PubMed] [Google Scholar]
  • 10. Kierstein G, Vallinoto M, Silva A, Schneider MP, Iannuzzi L, et al. (2004) Analysis of mitochondrial D-loop region casts new light on domestic water buffalo (Bubalus bubalis) phylogeny. Molecular Phylogenetics Evolution 30: 308–324. [DOI] [PubMed] [Google Scholar]
  • 11. Kumar S, Nagarajan M, Sandhu JS, Kumar N, Behl V (2007a) Phylogeography and domestication of Indian river buffalo. BMC Evolutionary Biology 7: 186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Kumar S, Nagarajan M, Sandhu JS, Kumar N, Behl V, et al. (2007b) Mitochondrial DNA analyses of Indian water buffalo support a distinct genetic origin of river and swamp buffalo. Animal Genetics 38: 227–232. [DOI] [PubMed] [Google Scholar]
  • 13. Lei CZ, Zhang W, Chen H, Lu F, Ge QL, et al. (2007a) Two maternal lineages revealed by mtDNA D-loop sequences in Chinese native water buffaloes (Bubalus bubalis). Asian-Australasian Journal of Animal Sciences 20(4): 471–476. [Google Scholar]
  • 14. Lei CZ, Zhang W, Chen H, Lu F, Liu RY (2007b) Independent maternal origin of Chinese swamp buffalo (Bubalus bubalis). Animal Genetics 38: 97–102. [DOI] [PubMed] [Google Scholar]
  • 15. Lau CH, Drinkwater RD, Yusoff K, Tan SG, Hetzel DJS, et al. (1998) Genetic diversity of Asian water buffalo (Bubalus bubalis), mitochondrial DNA D-loop and cytochrome b sequence variation. Animal Genetics 29: 253–264. [DOI] [PubMed] [Google Scholar]
  • 16. Chen YC, Li XH (1989) New evidence of the origin and domestication of the Chinese swamp buffalo (Bubalus bubalis). Buffalo Journal 1: 51–55. [Google Scholar]
  • 17. Kikkawa Y, Yonekawa H, Suzuki H, Amano T (1997) Analysis of genetic diversity of domestic water buffaloes and anoas based on variations in the mitochondrial gene for cytochrome b. Animal Genetics 28: 195–201. [Google Scholar]
  • 18.Qiu H (1986) Bovine Breeds in China. Shanghai Scientific and Technical Publishing House, Shanghai. [Google Scholar]
  • 19. Liu L, Chen X, Jiang L (2004) A study of Neolithic water buffalo remains from Zhejiang, China. Bulletin of the Indo-Pacific Prehistory Association: The Taipei Papers 24 (2): 113–120. [Google Scholar]
  • 20. Liu L, Yang D, Chen X (2006) Zhongguo jiayang shuiniu de qiyuan (The origins of Chinese domestic buffalo). Kaogu Xuebao 2: 141–178. [Google Scholar]
  • 21. Xue X, Li X (2000) Shaanxi shuiniu huashi ji Zhongguo huashi shuiniu de dili fenbu he zhongxi fasheng (Fossil Bubalus from Shaanxi, distribution and phylogeny of fossil Bubalus in China). Gujizhui Dongwu Xuebao 38 (3): 218–231. [Google Scholar]
  • 22. Hu WP, Xu BM, Lian LS (1997) Polymorphism of mitochondrial DNAs of Yunnan domestic water buffaloes, Bubalus bubalis, in China, based on restriction endonuclease cleavage patterns. Biochemical Genetics 35: 225–231. [DOI] [PubMed] [Google Scholar]
  • 23. Lei CZ, Zhang CM, Song WN, Campana MG, Bower MA, et al. (2011) Genetic diversity of mitochondrial cytochrome b gene in Chinese native buffalo. Animal Genetics 42: 432–436. [DOI] [PubMed] [Google Scholar]
  • 24. Zhang Y, Su D, Yu Y, Zhang Y (2007) Genetic diversity and differentiation of Chinese domestic buffalo based on 30 microsatellite markers. Animal Genetics 38: 569–575. [DOI] [PubMed] [Google Scholar]
  • 25. Zhang Y, Vankan D, Zhang Y, Barker JSF (2011) Genetic differentiation of water buffalo (Bubalus bubalis) populations in China, Nepal and south-east Asia, inferences on the region of domestication of the swamp buffalo. Animal Genetics 42: 366–377. [DOI] [PubMed] [Google Scholar]
  • 26.Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Press, New York, USA. [Google Scholar]
  • 27. Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG (1997) The Clustal-X windows interface, Flexible strategies for multiple sequences alignment aided by quality analysis tools. Nucleic Acids Research 25: 4876–4882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Excoffier L, Lischer HEL (2010) Arlequin suite ver 3.5: A new series of programs to perform population genetics analyses under Linux and Windows. Molecular Ecology Resources 10: 564–567. [DOI] [PubMed] [Google Scholar]
  • 29. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5: Molecular Evolutionary Genetics Analysis Using Maximum Likelihood, Evolutionary Distance, and Maximum Parsimony Methods. Molecular Biology and Evolution 28 (10): 2731–2739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Bandelt HJ, Forster P, Röhl A (1999) Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution 16: 37–48. [DOI] [PubMed] [Google Scholar]
  • 31. Milne I, Wright F, Rowe G, Marshal DF, Husmeier D, et al. (2004) TOPALi: Software for Automatic Identification of Recombinant Sequences within DNA Multiple Alignments. Bioinformatics 20 (11): 1806–1807. [DOI] [PubMed] [Google Scholar]
  • 32. Troy CS, MacHugh DE, Bailey JF, Magee DA, Loftus RT, et al. (2001) Genetic evidence for Near-Eastern origins of European cattle. Nature 410: 1088–1091. [DOI] [PubMed] [Google Scholar]
  • 33. Shapiro B, Drummond AJ, Rambaut A, Wilson MC, Matheus PE, et al. (2004) Rise and fall of the beringian steppe bison. Science 306: 1561–1565. [DOI] [PubMed] [Google Scholar]
  • 34. Beja-Pereira A, England PR, Ferrand N, Jordan S, Bakhiet AO, et al. (2004) African origins of the domestic donkey. Science 304 (5678): 1781. [DOI] [PubMed] [Google Scholar]
  • 35. Chen SY, Lin BZ, Baig M, Mitra B, Lopes RJ, et al. (2010) Zebu cattle are an exclusive legacy of the South Asia Neolithic. Molecular Biology and Evolution 27 (1): 1–6. [DOI] [PubMed] [Google Scholar]
  • 36. Legendre P, Makarenkov V (2002) Reconstruction of biogeographic and evolutionary networks using reticulograms. Systematic Biology 51 (2): 199–216. [DOI] [PubMed] [Google Scholar]
  • 37. Woese C (2002) On the evolution of cell. PNAS 99 (13): 8742–8747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Chen SY, Su YH, Wu SF, Sha T, Zhang YP (2005) Mitochondrial diversity and phylogeographic structure of Chinese domestic goats. Molecular Biology and Evolution 37 (3): 804–814. [DOI] [PubMed] [Google Scholar]
  • 39. Liu RY, Lei CZ, Liu SH, Yang GS (2007) Genetic diversity and origin of Chinese domestic goats as revealed by complete mtDNA D-loop sequence variation. Asian-Australasian Journal of Animal Sciences 20 (2): 178–183. [Google Scholar]
  • 40. Luikart G, Gielly L, Excoffier L, Vigne JD, Bouvet J, et al. (2001) Multiple maternal origins and weak phylogeographic structure in domestic goats. PNAS 98 (10): 5927–5932. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Chen SY, Duan ZY, Sha T, Xiangyu J, Wu SF, et al. (2006) Origin, genetic diversity, and population structure of Chinese domestic sheep. Gene 376 (2): 216–223. [DOI] [PubMed] [Google Scholar]
  • 42. Lei CZ, Ge QL, Zhang HC, Liu RY, Zhang W, et al. (2007) African maternal origin and genetic diversity of Chinese domestic donkeys. Asian-Australasian Journal of Animal Sciences 20: 645–652. [Google Scholar]
  • 43. Guo S, Savolainen P, Su J, Zhang Q, Qi D, et al. (2006) Origin of mitochondrial DNA diversity of domestic yaks. BMC Evolutionary Biology 6: 73–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Ge JX, Wu SD, Chao SJ (1997) The migration history of China. Fujian People Press, Fuzhou. [Google Scholar]
  • 45. Groves CP (1996) The taxonomy of the Asian wild buffalo from the Asian mainland. International Journal of Mammalian Biology 61: 327–338. [Google Scholar]
  • 46.Mason IL (1974) Species types and breeds. In Cockrill, W. R, editor, The Husbandry and Health of the Domestic Buffalo. Food and Agriculture Organization of the United Nations, Rome, 1–47. [Google Scholar]
  • 47. Mudar K, Anderson D (2007) New evidence for Southeast Asian Pleistocene foraging economies, Faunal remains from the early levels of Lang Rongrien rockshelter, Krabi, Thailand. Asian Perspectives 46: 298–334. [Google Scholar]
  • 48. Patel A, Meadow R (1998) The exploitation of wild and domestic water buffalo in prehistoric northwestern South Asia. In Buitenhuis H, Barto siewicz L, Choyke AM, editors, Archaeozoology of the Near East III, Groningen, ARC - Publicaties 18: 180–198. [Google Scholar]
  • 49.Patel A (1997) The pastoral economy of Dholavira, a first look at animals and urban life in third millennium Kutch. In Allchin R and Allchin B, editors, South Asian Archaeology 1995, Porceedings of the 13th Conference of the European Association of South Asian Archaeologists, New Delhi, Science Publishers: 101–113. [Google Scholar]
  • 50.Higham C (1989) The Archaeology of Mainland Southeast Asia. Cambridge University Press, Cambridge. [Google Scholar]
  • 51. Higham C, Manly B, Kihingam A, Moore SJE (1981) The Bovid third phalanx and prehistoric ploughing. Journal of Archaeological Science 8: 353–365. [Google Scholar]
  • 52. O’Reilly DJW, Driesch AVD, Voeun V (2006) Archaeology and archaeozoology of Phum Snay, A late prehistoric cemetery in northwestern Cambodia. Asian Perspectives 45: 188–211. [Google Scholar]
  • 53. Zong G, Huang XS (1985) Yunnan Baoshan Pupiao quanxinshi zaoqi wenhua yiwu ji buru dongwu de yicun (Cultural and animal remains from Pupiao, Yunnan). Shiqian Yanjiu 4: 46–50. [Google Scholar]
  • 54. Yunnan Museum (1995) Yunnan sheng Jianchuan Haimenkou qingtong shidai zaoqi yizhi (Early Bronze Age site at Haimenkou in Jianchuan, Yunnan province). Kaogu 9: 775–787. [Google Scholar]
  • 55. Li X, Han R (2006) Yunnan Jianchuanxian Haimenkou yizhi chutu tongqi de jishu fenxi jiqi niandai. Kaogu 7: 80–85. [Google Scholar]
  • 56.Zhu FY, Li L, Jun L (2005) Jiemi Washan (Decoding the Wa Mountains). Kunming, Yunnan Meishu Press. [Google Scholar]
  • 57. De Jong, De Josselin PE (1965) An interpretation of agricultural rites in Southeast Asia, with a demonstration of use of data from both continental and insular areas. The Journal of Asian Studies 24 (2): 283–291. [Google Scholar]
  • 58. Tong E (1983) Shitan gudai Sichuan yu Dongnanya wenming de huanxi (Relitionship between Sichuan and Southeastern Asia in ancient times). Wenwu 9: 73–81. [Google Scholar]
  • 59.Jiang Y, editor (1995) Gudia Xinan Sichou Zilu Yanjiu (2). Sichuan Daxue Press, Chengdu. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1

ML tree of Chinese swamp buffalo without cutting off reliability percentages (RP) below 50%.

(TIF)

Click here for additional data file. (1MB, tif)
Table S1

Swamp buffalo D-Loop sequences retrieve from NCBI database.

(XLSX)

Click here for additional data file. (18.9KB, xlsx)
Table S2

Polymorphic nucleotide sites of 148 swamp buffalo haplotypes.

(DOC)

Click here for additional data file. (149.5KB, doc)
Table S3

Total genetic diversity indexes in Chinese water buffaloes.

(DOC)

Click here for additional data file. (34.5KB, doc)
Table S4

Archaeological sites associated with buffalo remains dating to the Holocene period.

(DOCX)

Click here for additional data file. (20.5KB, docx)

Articles from PLoS ONE are provided here courtesy of PLOS

RESOURCES