Genetic variability within and among Haemonchus contortus isolates from goats and sheep in China

Background Haemonchus contortus (order Strongylida) is a common parasitic nematode infecting small ruminants and causing significant economic losses worldwide. Knowledge of genetic variation within and among H. contortus populations can provide a foundation for understanding transmission patterns, the spread of drug resistance alleles and might assist in the control of haemonchosis. Methods 152 H. contortus individual adult worms were collected from seven different geographical regions in China. The second internal transcribed spacer (ITS-2) of the nuclear ribosomal DNA and mitochondrial nicotinamide dehydrogenase subunit 4 gene (nad4) were amplified by polymerase chain reaction (PCR) and sequenced directly. The sequence variations and population genetic diversities were determined. Results Nucleotide sequence analyses revealed 18 genotypes (ITS-2) and 142 haplotypes (nad4) among the 152 worms, with nucleotide diversities of 2.6% and 0.027, respectively, consistent with previous reports from other countries, including Australia, Brazil, Germany, Italy, Malaysia, Sweden, the USA and Yemen. Population genetic analyses revealed that 92.4% of nucleotide variation was partitioned within populations; there was no genetic differentiation but a high gene flow among Chinese populations; some degree of genetic differentiation was inferred between some specimens from China and those from other countries. Conclusions This is the first study of genetic variation within H. contortus in China. The results revealed high within-population variations, low genetic differentiation and high gene flow among different populations of H. contortus in China. The present results could have implications for studying the epidemiology and ecology of H. contortus in China.


Background
Haemonchus contortus is a trichostrongyloid nematode and one of the major pathogens affecting small ruminants worldwide [1]. The adult female of this species produces large numbers of eggs, which are excreted in hosts' faeces. The eggs hatch on pasture and continue to develop under moist conditions to third-stage larvae (L3s). L3s are then ingested by a suitable host animal and eventually establish as dioecious, adult worms [2]. The blood-feeding activity of adults causes anaemia, oedema, diarrhoea and even death [3], consequently causing serious production and economic losses, particularly in tropical and temperate regions of the world [4].
Population genetic studies of H. contortus in the USA have shown that this species exhibits high within population variation and low genetic differentiation within continuous geographical regions, likely ascribing to high gene flow influenced by host movement [5,6]. However, strong barriers to gene flow have been observed on a global scale, which appear to be attributed to poor dispersal ability of the parasite and restricted opportunities for host movement across continents [7]. Other research, focused on domestic and wild animals, has found high genetic variation and relatively low host specificity for H. contortus within Brazil and Italy [8,9].
Population genetic studies of H. contortus have been conducted in a wide range of geographical regions of the world, including Australia, Brazil, Europe, Malaysia, and the USA [5,[7][8][9][10][11][12]. However, surprisingly, nothing is known about genetic variability within H. contortus in China, in spite of its endemic status and economic impact in this country [13,14]. Therefore, in the present study, we explored genetic variation within and among seven populations of H. contortus from southwest, central and northeastern regions of China, employing the second internal transcribed spacer (ITS-2) of nuclear ribosomal DNA and the mitochondrial nicotinamide dehydrogenase subunit 4 (nad4) gene as markers.

Parasite material
In total, 152 individual adult specimens of H. contortus were collected from the abomasa of slaughtered sheep or goats from seven geographical locations in tropical to subtropical climate zones and six provinces of China ( Figure 1 and Table 1). Geographical locations were separated by distances of 370 to 4000 km. Samples from Liaoning and Heilongjiang were from sheep, whereas those from other regions were from goats. The adult specimens of H. contortus (17 to 24 per population) were washed extensively in physiological saline, stored in 70% ethanol and then sent to the College of Veterinary Medicine, Huazhong Agricultural University, Wuhan. Upon arrival, individual worms were identified morphologically, according to Lichtenfels et al. (1994) [15].

Isolation of genomic DNA
Total genomic DNA was isolated from individual worms using sodium dodecyl-sulfate/proteinase K treatment [16], followed by spin-column purification (Wizard™ DNA Clean-Up, Promega). DNA samples were stored at −20°C until use.
conditions: initial denaturation at 94°C for 5 min, followed by 30 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 30 s and extension at 72°C for 1 min, with a final extension of 72°C for 5 min.
All PCR products were examined on agarose gels (1.5%) to verify that they represented single bands, columnpurified (Wizard PCR-Preps, Promega) and then sequenced directly (BigDye Terminator v.3.1 cycle sequencing kit, Applied Biosystems) in an automated sequencer (PRISM3730, ABI) using appropriate forward and reverse primers (in separate reactions). Forward and reverse sequences were merged, consensus sequences determined and then deposited in the GenBank database.

Data analysis
Sequences were aligned over a consensus length (231 bp for ITS-2 and 412 bp for nad4) using the program Clustal W within MEGA v.5.0 [18]. Pairwise comparisons were made with previously published sequences, and identities (%) calculated using the program BioEdit [19].
The phylogenetic analysis was conducted using the neighbour-joining (NJ), maximum parsimony (MP) and maximum likelihood (ML) methods, respectively, based on the Tamura -Nei model [18]. Confidence limits were assessed using bootstrap procedure with 1000 pseudoreplicates for NJ, MP and ML trees, and other settings were obtained using the default values in MEGA v.5.0 [18]. A 50% cut-off value was implemented for the consensus tree. The Collapse program online (http:// sing.ei.uvigo.es/ALTER/) was used to define haplotype sequences. The diversity indices (Fst and Nst) were calculated using the program DnaSP5.1 [20] to evaluate degree of gene flow among populations. Tajima's D [21] and Fu's Fs [22] were also calculated to test neutrality using the same program (DnaSP 5.1) [20]. To estimate genetic diversity within and among populations (isolates), the hierarchical analysis of molecular variance (AMOVA) was performed using the Arlequin 3.1 package [23]. For this analysis, the data set was divided into three groups. Groups 1, 2 and 3 contained samples from Southwest (Guangxi, Yunnan), Central (Suizhou, Yidu, Shaanxi) and Northeast (Liaoning, Heilongjiang) China, respectively. The median joining (MJ) network [24] was drawn using the program Network v.4.6.1.1 to study of haplotype relationships. In addition, 176 nad4 sequences from previous studies [5,8,10] were retrieved from GenBank and used for comparisons. Information on geographical origins, accession numbers and parasite codes are given in Table 2.

Sequences analyses
ITS-2 sequences were determined from 152 worms from seven different geographical locations in China (see Table 1). After sequence editing and alignment, a consensus length of 231 bp was obtained for all specimens. The analysis of the 152 ITS-2 sequences revealed 18 distinct genotypes (sequence data are publicly available under accession numbers KC415117-KC415134 for ITS-2), with sequence identities ranging from 97.4% to 100%, when compared with each other or with two reference  A total of 176 nad4 sequences were retrieved from the GenBank database for comparison. The nucleotide diversity (π) within each population was also calculated.  Table 4; their low negative values implied no significant deviation from neutrality.

Phylogenetic analysis
Trees were constructed using the nad4 sequence data for H. contortus from China, using an nad4 sequence of H. placei (accession no. AF070825) as the outgroup. The tree revealed a relatively continuous variation in genetic distance among haplotypes (Additional file 2). There was a random distribution of sequences representing different locations, with weak support (< 50%) for some nodes. There were no obvious boundaries among individuals in the phylogenetic trees, except that some individuals from the Yunnan population grouped with moderate nodal support (72%), with one sequence from Guangxi and one cluster contained one sequence from Yunnan and three sequences from Liaoning with strong nodal support (90%). Consistent with the results obtained for the NJ tree, analysis of the nad4 sequences using MP and ML methods, respectively, produced phylogenetic trees with similar topologies (not shown).
Extending the analyses, to discern the relationships among 142 haplotypes, a parsimony network was constructed ( Figure 2). The network profile showed that there was no preferential or distinct grouping of specimens according to geographical region, except for several individuals from Yunnan. In order to genetically compare H. contortus from China with those from other parts of the world, according to the approach used by Troell et al. (2006) [7], ten nad4 sequences, chosen randomly from the Chinese sample set and from each of the four other H. contortus populations representing other countries (Italy, Malaysia, the USA and Yemen) were included (Figure 3). The three methods produced consensus trees, in which there was no clear grouping according to country (at pp > 80%). In contrast, on three occasions, two samples from the same country (two each from China, Malaysia and the USA) grouped together with strong nodal support (pp > 80%). In addition, five samples from Yemen grouped together with low support (50%).

Population genetic structure
The random distribution of 142 nad4 haplotypes from Chinese samples across the parsimony network ( Figure 2)   did not support any particular genetic structure among H. contortus specimens, which is corroborated by the low pairwise Fst/Nst values (   (Table 5).
To assess possible factors that might affect gene flow, an analysis of molecular variance (AMOVA) was computed among populations. With the introduction of three-levels in the analysis, it was shown that 92.4% of the variance was distributed within populations and only 7.6% among populations ( Table 6). The average Fst within seven populations (0.0759) was greatest compared with an Fst value among groups of 0.0143, and an Fst among populations within groups of 0.0655.
Fst values were also calculated between samples from China and those from each of the other four countries, including Italy, Malaysia, the USA and Yemen, for which nad4 sequence data were available. The results revealed that the highest level of genetic differentiation was recorded between China and the USA, with an Fst value of 0.41294 (Table 2). Moderate levels of genetic differentiation were recorded between China and Italy, or China and Yemen, with the Fst values of 0.14407 and 0.26710, respectively. The lowest level of differentiation was found between China and Malaysia, with an Fst value of 0.05841 (Table 2).

Discussion
In the present study, the ITS-2 and partial nad4 sequences were determined from 152 individual H.
contortus specimens from seven populations of China. The ITS-2 sequences from these worms confirmed their specific identity as H. contortus. Further analysis revealed sequence variation of 2.6% in ITS-2 among all 152 individual worms from all locations in China. This magnitude of variation is consistent with variation (2.6%) detected in H. contortus populations in various countries, such as Germany and Sweden/Kenya [25,26], but lower than that (5.2%) detected between or among populations from seven countries, including Australia, France, Germany, New Zealand, Switzerland, The Netherlands and the UK [11]. For nad4, the nucleotide diversities within each of the seven H. contortus populations in China ranged from 0.018 to 0.037, with an average nucleotide diversity of 0.027 (Tables 2 and 4), which is in accordance with previously published data for this mitochondrial gene in various countries including Italy (0.026-0.03), Malaysia (0.032-0.044), the USA (0.024-0.03) and Yemen (0.021-0.036) [5,8,10]. Similarly, the high degree of diversity (i.e. 142 distinct haplotypes representing 152 individuals) and a mean haplotype diversity of 0.996 were in accordance with previous studies of the trichostrongyloids, such as Haemonchus placei and Teladorsagia circumcincta [7,27,28].
The present phylogenetic analysis revealed that there was no clear grouping of H. contortus according to host species (Additional file 2: Figure S1), a finding supported by the lack of significant differences in pairwise Fst and Nst values between H. contortus populations from LN or HLJ (sheep) and those from SZ, YD, GX, YN and SX (goat) (see Table 5). This result is similar to findings of previous studies, showing a limited relationship of H. contortus populations with different ruminant species in Brazil, Italy, Malaysia and Yemen [8][9][10].
The genetic analysis of the nad4 gene in the present study showed that the majority (92.4%) of genetic diversity was partitioned within populations of H. contortus from China, with no clear phylo-geographic structuring (with the exception of a subset of seven specimens from Yunnan), suggesting a high gene flow without clear geographical barriers among populations from different  On the other hand, on a global scale, the highest Fst value between H. contortus from China and the USA indicated a high genetic differentiation and less gene flow between these populations from two distinct continents. In contrast, the lowest genetic differentiation of Chinese specimens from those from Malaysia was reflected in the lowest pairwise Fst value. These results support a previous proposal that the global population genetic structure of H. contortus is characterized by no or low genetic differentiation between specimens from abutting geographical regions or within continents with limited barriers to gene flow, but significant genetic differentiation on a global scale (across continents) with substantial barriers to gene flow [7].
Previous studies have attributed a high degree of sequence diversity in the nad4 within populations of H. contortus to a large effective population size (Ne), high biotic potential, rapid, direct life-cycle and a high mutation rate in this polymorphic nematode [6,9,29], which explains the population genetic characteristics in China. In this country, haemonchosis is widespread geographically, and the prevalence and intensity of H. contortus infection are relatively high. An appraisal of key studies published in Chinese [30][31][32][33][34] shows that H. contortus exists in at least 30 of the 32 provinces in China, and is the dominant intestinal parasitic nematode of sheep and goats, with prevalences ranging from 6.5% to 100% and infection intensities of at least 20-8000 worms per host. The movement of livestock throughout China is also common. Since 1978, the livestock industry has undergone an enormous transformation. By the end of 1996, the total numbers of goats and sheep reached 303 million head, and the number has increased by 78.5% since that time, with a high demand on the red meat market for lamb [35]. Taken together, these findings indicate that the population genetic characteristics of H. contortus in China appear to be similar to those in other countries and studies.
As there is no commercial vaccine against haemonchosis in China, like elsewhere, the control of H. contortus relies largely on anthelmintics [36]. Here, the most widely used anthelmintics are albendazole, mebendazole, levamisole and ivermectin [36]. In spite of the apparently excessive use of anthelmintics against trichostrongyloids [36], no precise information is available on the nature and extent of drug resistance in H. contortus in China. Nonetheless, there are a small number of studies [37][38][39] reporting drug resistance against benzimidazoles and ivermectin in H. contortus populations in China. As the level of gene flow among populations of H. contortus parasitizing goats and sheep in China was shown here to be very high, there is substantial opportunity for rare resistance alleles to spread, indicating an urgent need to conduct a nationwide investigation of drug resistance in H. contortus. Such a focus will inform the scientific and non-scientific communities about the prevalence and intensity of resistance, and should underpin future control efforts in China.

Conclusions
In conclusion, this is the first study of genetic variation within H. contortus in China. The results revealed high within-population variation, low population genetic differentiation and high gene flow of H. contortus in China. The results also indicated that on a global scale, there is no or a low level of genetic differentiation among H. contortus populations from the same continent, and significantly higher levels of genetic differentiation among those from different continents. These findings have important implications for studying the molecular epidemiology and controlling the spread of anthelmintic resistance against H. contortus in China.