Echinococcosis, which is listed by the World Health Organization (WHO) as a neglected tropical disease, is a zoonotic infection caused by several cestode species from the genus Echinococcus (Eckert and Deplazes, Reference Eckert and Deplazes2004; Brunetti et al., Reference Brunetti, Kern and Vuitton2010). Adult worms inhabit the small intestine of a wild or domestic species of Canidae, which serves as a definitive host. Intermediate hosts (ungulates) and accidental hosts (humans) become infected after ingesting parasitic eggs, which subsequently develop into hydatid cysts in the liver, lungs or other organs. Furthermore, humans can also acquire the infection by direct contact with dogs (Eckert et al., Reference Eckert2001). Alveolar echinococcosis (AE) and cystic echinococcosis (CE) are regarded as neglected zoonoses, with a higher overall disease burden associated with CE due to its global distribution and high regional prevalence, while a higher pathogenicity and case fatality rate is associated with AE, especially in Asia (Eckert and Deplazes, Reference Eckert and Deplazes2004; Deplazes et al., Reference Deplazes2017).
Echinococcus granulosus sensu lato (s.l.) was taxonomically defined based on sequence comparisons of portions of the mitochondrial genes cytochrome c oxidase subunit 1 (cox1) and NADH dehydrogenase subunit 1 (nad1) (Bowles et al., Reference Bowles, Blair and McManus1992; Bowles and McManus, Reference Bowles and McManus1993; McManus, Reference McManus2006; Rostami Nejad et al., Reference Rostami Nejad2012; Alvarez Rojas et al., Reference Alvarez Rojas, Romig and Lightowlers2014). Moreover, E. granulosus s.l. can be further divided into E. granulosus sensu stricto (s.s.) G1 (identified in sheep), G2 (identified in Tasmanian sheep), or G3 (identified in buffalo); E. equinus G4; E. ortleppi G5; E. canadensis (G6–G10); and E. felidis (lion strain) (Kedra et al., Reference Kedra1999; Nakao et al., Reference Nakao2013a). Variations between E. granulosus species include differences in infectivity within different intermediate hosts, the survival rate within a definitive host, antigenicity, transmission dynamics, and pathology (Eckert et al., Reference Eckert1993; Thompson, Reference Thompson2008; Soriano et al., Reference Soriano2010; Siracusano et al., Reference Siracusano2012; Alvarez Rojas et al., Reference Alvarez Rojas, Gauci and Lightowlers2013; Cucher et al., Reference Cucher2013; McManus, Reference McManus2013; Chaâbane-Banaoues et al., Reference Chaâbane-Banaoues2016). Thus, molecular epidemiological surveys utilizing E. granulosus isolates are fundamental to understanding the spatio-temporal epidemiology in endemic areas and will aid in disease control and prevention (McManus, Reference McManus2006, Reference McManus2013; Thompson, Reference Thompson2008).
So far, human CE cases reported in China have been associated with the G1, G3, G6, G7 and G10 genotypes, with most cases caused by G1. Furthermore, in China, disease cases are most often reported in the western, north-western and northern regions (Zhang et al., Reference Zhang1998, Reference Zhang2014; Cardona and Carmena, Reference Cardona and Carmena2013; Yan et al., Reference Yan2013; Wang et al., Reference Wang2014; Yang et al., Reference Yang2015; Deplazes et al., Reference Deplazes2017; Ito and Budke, Reference Ito and Budke2017). To further investigate the population genetic patterns in these regions, surgically obtained hydatid fluids were examined and Echinococcus cox1 and nad1 genes were sequenced and analysed.
Materials and methods
Collection of patient samples
Hydatid fluids were collected from 23 clinically confirmed hydatidosis patients. Seven samples were collected from herdsmen in the Xilingol League Hospital, Inner Mongolia. The remaining 16 samples were obtained by the 302 Military Hospital during a one-year period (from December 2015 to November 2016), with two patients from Qinghai Province and 14 from the Tibet Autonomous Region. Approximately 1 ml of hydatid fluid was collected per patient and stored at −20°C. All patients provided informed written consent prior to surgery, but the hospital Medical Ethics Board deemed consent for molecular characterization of hydatid fluids exempt due to collection occurring post surgery and samples being examined anonymously. The study protocol was approved by the Medical Ethics Review Committee of Ordos Central Hospital, Inner Mongolia, China.
Genomic DNA was extracted from the hydatid fluid using a blood/cell/tissue genomic DNA extraction kit (Tiangen Biotech, Beijing, China) according to the manufacturer's instructions. DNA concentrations were determined using a spectrophotometer (Implen, Munich, Germany) and samples were stored at −20°C until further use.
Mitochondrial polymerase chain reaction
Two mitochondrial DNA fragments within the cox1 and nad1 genes were amplified from each individual DNA sample using primer pairs designed by analysing Echinococcus mitochondrial gene sequences in GenBank. The cox1 DNA fragments (pcox1) were amplified using COI-FOR (5′-GATCGTAATTTTTGTTCTGC-3′) and COI-REV (5′-ATAAGAACCTAACGACATAACATA-3′) primers, with an expected amplicon size of 528 bp. Echinococcus granulosus s.s nad1 DNA fragments (pnad1) were amplified using NDI-FOR1 (5′-GTGGTTGTTTTTGGGTTAGT-3′) and NDI-REV1 (5′-AGCAAACAGCCCTAAAATCAA-3′) primers, with an expected amplicon size of 882 bp. Echinococcus multilocularis pnad1 was amplified using specific NDI-FOR2 (5′-TTGAGTTTGCGTCTCGATGA-3′) and NDI-REV2 (5′-CTAATTAACAATAACAAAAAATC-3′) primers, with an expected amplicon size of 1001 bp. Echinococcus canadensis pnad1 was amplified using NDI-FOR3 (5′-ATGGTTGTTTTGGGATTAAT-3′) and NDI-REV3 (5′-TACAAAAAATCCTAAAATCAA-3′) primers, with an expected amplicon size of 882 bp. Polymerase chain reactions (PCR) were carried out in a total volume of 50 μl [4 μl DNA (50–100 ng), 25 μl 2× Ultra-pfu Master Mix (0.1 U/μl Ultra-Pfu DNA Polymerase), 4 mm MgSO4, and 0.4 mm of each dNTPs in 2× PCR reaction buffer and 20 pmol of each primer]. Negative controls, with sterilized water substituted for the DNA template, were included for each PCR experiment. PCR was performed in a thermocycler with the following programme: an initial denaturation step at 94°C for 3 minutes, then 35 cycles of denaturation at 94°C for 30 s, annealing at 46.5°C (for primer pairs COI-FOR/COI-REV, NDI-FOR1/NDI-REV1 or NDI-FOR3/NDI-REV3) or 50°C (for the NDI-FOR2/NDI-REV2 pair) for 40 s and extension at 72°C for 40 s, followed by a final extension at 72°C for 6 minutes. PCR products were evaluated via electrophoresis on a 1% agarose gel containing the nucleic acid stain Golden View (Biomed, Beijing, China).
DNA sequencing and phylogenetic analysis
The PCR products were purified using a TIANgel Midi Purification Kit (Tiangen Biotech, Beijing, China) according to the manufacturer's instructions and then subjected to bidirectional, automated sequencing (ABI 3730XL DNA Sequencer; Applied Biosystems, Foster City, CA, USA) using the same primers that were employed in the amplification process. Sequence comparisons were performed using two steps: (1) sequences were first queried against the BLAST database (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and (2) multiple sequence alignments were performed using the obtained amplicons and some reference sequences in Clustal W within the Lasergene software package (DNASTAR, Madison, WI, USA). Sequence editing and pcox1 and pnad1 concatenations were performed using the EditSeq program in DNASTAR. The concatenated sequences were then aligned using the Clustal W program in MEGA6 and a haplotype network was constructed using statistical parsimony in TCS1.21 (Clement et al., Reference Clement, Posada and Crandall2000).
Results and discussion
When examining the causative agent in hydatid samples, it is necessary to obtain a molecular characterization of population genetic patterns to gain a better understanding of the distribution and diversity of Echinococcus globally (Nakao et al., Reference Nakao2007). In the present study, etiological agents from 23 patients with hydatidosis that were located in epidemic areas were systematically examined by sequencing Echinococcus cox1 and nad1 genes. Two amplicons, pcox1 (528 bp) and pnad1 (822 bp), were successfully amplified from each specimen, with a single 1001 bp amplicon generated when using the NDI-FOR2/NDI-REV2 primers with the sample ‘Qinghai1’. The 46 DNA fragments obtained were then bidirectionally sequenced.
BLAST results showed that the 1001-bp-long amplicon contains a truncated 894 bp region (at position 108 to 1001) encoding an E. multilocularis nad1 gene. Furthermore, sequence alignments indicated that the truncated region exhibits a high sequence identity (99.6–99.9%) with representative reference nad1 genes from E. multilocularis (GenBank accession nos AB018440 and AJ237639). The pcox1 gene fragment amplified from the same isolate also shared a high sequence identity (< 99.8%) with E. multilocularis cox1 genes (GenBank accession nos AB018440 and M84668). These two sequences obtained from the ‘Qinghai1’ sample were submitted to the GenBank database under the accession numbers KY094609 (pnad1) and KY094608 (pcox1). These results suggest that the etiological agent in the ‘Qinghai1’ sample was E. multilocularis. Further detailed information about this isolate is available in our previous paper, where the isolate is designated as E6 (Li et al., Reference Li, Gao and Wang2017).
While E. canadensis has been considered to have a low infectivity in humans (Eckert et al., Reference Eckert2001), in recent years, reports from countries such as Mongolia, Austria, Poland, Argentina, USA and China have indicated that E. canadensis is an underestimated cause of human echinococcosis and may include variants with differing human pathogenicities (McManus et al., Reference McManus2002; Guarnera et al., Reference Guarnera2004; Schneider et al., Reference Schneider2010; Jabbar et al., Reference Jabbar2011; Dybicz et al., Reference Dybicz2013; Zhang et al., Reference Zhang2014). In one patient sample from Tibet (‘Tibet14’), pcox1 exhibited a 99.4% identity with those from E. canadensis G6 strains (accession nos AB208063 and M84666; both from camels). However, for pnad1, a 100% identity with the nad1 gene from E. canadensis G7 strain (accession no. AB235847; from a pig in Japan) was noted. The reference sequences for the nad1 gene in the G6 (accession no. AJ237637) and G7 (accession no. AJ237638) strains are very similar and differ only at three nucleotide positions (positions 10, 333 and 438; Bowles and McManus, Reference Bowles and McManus1993). When examining the pnad1 sequence from the ‘Tibet14’ sample, two of the three nucleotides were shared with the G7 reference sequence (AJ237638), while only one of the three was identical to the G6 reference sequence (AJ237638). Thus, it was difficult to determine the specific genotype of the causative agent for this patient and this sample was instead determined to possess a G6/7 strain as previously described (Mogoye et al., Reference Mogoye2013; Nakao et al., Reference Nakao2013b). The obtained sequences for the ‘Tibet14’ sample were submitted to the GenBank database under the accession numbers MH050619 (pcox1) and MH050631 (pnad1). Furthermore, the epidemiological data showed that this patient was a native Tibetan herdsman who had never lived in any camel-raising areas. Thus, we speculate that there may be a dog–goat or a dog–pig cycle that transmits the G6/7 strain in the Tibetan Plateau.
When examining pcox1 in the ‘Tibet5’ sample, a < 99.4% identity to some key reference sequences of cox1 in E. granulosus s.s. genotype G3 (accession nos M84663 and KJ559023) was noted. Furthermore, pnad1 shared 100% identity with nad1 genes from the G3 strain (accession nos AJ237634, from a buffalo in India, and KJ559023, from a patient in China). Therefore, the isolate was classified as E. granulosus s.s. G3 and the sequences were deposited in GenBank under accession numbers MH050618 (pcox1) and MH050630 (pnad1). As for all the remaining 20 isolates, their pcox1 and pnad1 gene fragments exhibited < 99.4% and < 99.7% identity with E. granulosus s.s. G1 mitochondrial complete genomes (accession nos AB786664 and AF297617).
In total, 11 haplotypes (labelled H1 to H11) were identified among the G1 and G3 isolates. All of the G1 isolates were classified as haplotypes H1 to H10, while only the G3 isolate was classified as H11. The constructed parsimony network showed a star-like expansion, with one common ancestral haplotype (H1) occupying the centre of the network. When examining only the H11 haplotype (‘Tibet5’, G3), seven mutations were found when comparing it to the common H1 haplotype, while only one or two mutations were detected between the other samples and their associated common haplotype (fig. 1). Following concatenation of the pcox1 and pnad1 sequences, a BLAST query of the 11 haplotypes against those previously deposited in GenBank was performed and showed that H1 shared 100% identity with many of the haplotypes for E. granulosus s.s. G1. Furthermore, the haplotypes were prevalent in many countries in both humans (accession nos MG672293 (Algeria), MG672257 (Kazakhstan), MG672254 (Mongolia) and MG672129 (Spain)) and animals (sheep in Iran (MG672243), Argentina (MG672253) and Italy (MG672280); cattle in Argentina (MG672252) and Turkey (MG672202); goats in Iran (MG672256) and Spain (MG672154); buffalo in India (MG672260); and a domestic pig in Mexico (MG672259)). For the remaining 10 haplotypes, no sequences with a high sequence identity were identified. Due to this study adopting new primer sets, the pcox1 and pnad1 amplicon lengths were longer than the classical reference sequences (Bowles et al., Reference Bowles, Blair and McManus1992; Bowles and McManus, Reference Bowles and McManus1993). This increased length also contributed to increased sequence diversity, which may explain the difficulty in finding identical sequences for 10 of the haplotypes. Moreover, the parsimony network showed that the genetic relationships among the haplotypes were not associated with geographical divisions, as the haplotypes were seen across all of the examined regions (fig. 1, table 1). Furthermore, epidemiological examination showed that all of the patients were native herdsmen from Inner Mongolia, Tibet or Qinghai, who presumably remained local to their respective regions, without cross mobility occurring between these three regions. Moreover, these findings indicated that the regional populations of E. granulosus s.s. in the investigated areas were not fully differentiated from each other. Interestingly, a similar genetic structure was previously found for E. granulosus when examining patients in the Tibetan Plateau, eastern Tibet and Xinjiang regions (Nakao et al., Reference Nakao2010; Yan et al., Reference Yan2013). While these results suggest that the population structure of E. granulosus s.s. may be highly uniform throughout China, a larger number of samples from both intermediate and definitive hosts in these regions should be collected and analysed for confirmation. The frequency distributions of each of the haplotypes within the 21 E. granulosus s.s. isolates and their relevant accession numbers are shown in table 1.
In summary, this study characterized the etiological agents in 23 hydatid patient samples from three provinces or autonomous regions. The 23 samples were found to possess E. multilocularis (n = 1), E. canadensis G6/7 (n = 1), E. granulosus G3 (n = 1) or E. granulosus G1 (n = 20). While this study provides further information about the genetic patterns of Echinococcus in western and northern China, future efforts will focus on the further surveillance of echinococcosis in these regions.
We would like to thank Dr Bai Bingke from the Biobank of the 302 Military Hospital of China for providing hydatid fluids, and Professor She Zhanfei from the Ordos Central Hospital for collecting specimens for this study. We also would like to thank Dr Liu Zhijie from Lanzhou Veterinary Research Institute for his careful modification of the manuscript. We would like to thank LetPub (www.letpub.com) for providing linguistic assistance during the preparation of this manuscript.
This work was supported financially by the Science and Technology Program of Ordos, China (Contract No. YY201510017), the Program for Entrepreneurial and Innovative Talent Team of Ordos, China (2016), the Natural Science Foundation of Inner Mongolia Autonomous Region of China (Contract No. 2016MS08113) and the National Natural Science Foundation of China (NSFC; Contract No. 81760375).
Conflict of interest
The study protocol was approved by the Medical Ethics Review Committee of Ordos Central Hospital, Inner Mongolia, China.