**3. Discussion**

*G. duodenalis* is very commonly found in humans and domestic animals, as revealed by numerous prevalence studies across the globe [7,10,14]. The molecular epidemiology of *G. duodenalis* has been widely studied in livestock, which revealed its transmission dynamics and zoonotic significance in these animals. However, the parasite has not been extensively investigated in rodents; therefore, very little knowledge on the distribution, genetic diversity, and zoonotic potential of *Giardia* spp. in these animal hosts has been published [6,7,15,16]. Reports have demonstrated that *G. duodenalis* infections in rodents in Australia, Belgium, China, Croatia, Germany, Malaysia, Poland, Romania, Spain, Sweden, and the USA (Table 4) have prevalence rates ranging from 1.2% to 100% [12–34]. Herein, molecular analysis of 308 fecal samples collected from coypus in six provinces in China confirmed a *G. duodenalis* prevalence of 12.3% (38/308). To the best of our knowledge, this is the first molecular study of *G. duodenalis* infections in coypus, except for two studies in Italy and the USA that assessed *Giardia* spp. infection in coypus with the prevalence of 0% (0/153) and 73.3% (22/30), respectively, using an immunoenzymatic assay [12,13].

**Table 4.** *Giardia duodenalis* infection rates and genotypes in rodents worldwide.


N: number of positives for *G. duodenalis*; T: total of analyzed samples; "-" indicates not available.

Based on the current knowledge, eight *Giardia* species are considered valid, including *G. duodenalis*, *Giardia agilis*, *Giardia ardeae*, *Giardia psittaci*, *Giardia muris*, *Giardia microti*, *Giardia peramelis*, and *Giardia cricetidarum* [4,7]. In particular, two host-adapted species of *Giardia* spp. have been detected in rodents, including *G. muris* and *G. microti* [20,35]. To date, in China, *G. duodenalis* has been reported in rodents of several genera, including Chinchillas (*Chinchilla lanigera*), Asian house rats (*Rattus tanezumi*), brown rats (*Rattus norvegicus*), house mice (*Mus musculus*), chipmunks (*Eutamias asiaticus*), and bamboo rats (*Rhizomys sinensis*) [18,19,26,30]. In previous work, seven assemblages of *G. duodenalis* were found in rodents, including A, B, C, D, E, F, and G (Table 4). Both zoonotic and rodentspecific assemblages A, B, and G of *G. duodenalis* have been detected in rodents in China. Herein, the assemblages A and B were identified; assemblage B was the predominant genotype in coypus. It is worth noting that *G. duodenalis* assemblages A and B are important human pathogens; of them, assemblage B is more commonly reported in Asia, Oceania, Europe, and Africa than assemblage A [5,15,36]. We also reported mixed infections in the present study. Of note, the use of assemblage-specific primers in addition to MLST data is advocated in molecular epidemiological surveys, as mixed infections are likely to be underestimated [15,37]. Mixed infections are likely to be one of the main reasons for the adoption of multiple genetic markers in identifying distinct assemblages in the same sample [8,15]. Consequently, further studies on the molecular epidemiology of *G. duodenalis* is warranted to improve our understanding of the genetic diversity in rodents in China.

In recent studies, the MLG tool based on sequence analysis of the *gdh*, *bg*, and *tpi* loci represents a more informative approach for genotyping and elucidating the characteristics of *G. duodenalis* in different hosts, in addition to its zoonotic potential [9,15,19]. This has permitted the identification of sub-assemblages within assemblage A and B: AI to AIV and BI to BIV. Moreover, there are differences in the distribution of these sub-assemblages among hosts; for instance, human *Giardia* isolates mainly belong to sub-assemblage AII but also AI, whereas animals harbor AI, AII, and AIII [15]. In this study, 10 *G. duodenalis* isolates from coypus were successfully sequenced at all three gene loci; the sequence analysis of which resulted in one MLG in assemblage A and nine MLGs in assemblage B. Based on the phylogenetic analysis, the one MLG in assemblage A (AI–novel 1) was clustered with sub-assemblage AI-2, whereas the nine remaining MLGs in assemblage B (BIV–novel 1 to 7) were clustered with sub-assemblage BIV. Sub-assemblage AI was previously reported in various hosts in China, including humans, cattle, goats, sheep, dogs, cats, pigs, chinchillas, and chipmunks [6,10]. Other researchers also found sub-assemblage BIV in humans and non-human primates [6,10]. Additionally, compelling evidence on zoonotic transmission of *G. duodenalis* assemblage AI was revealed by epidemiological data showing a highly significant association between *Giardia* infection of schoolchildren and the presence of *Giardia*-positive dogs in the same household in Mexico [38]. Therefore, further studies into the epidemiology of *G. duodenalis* in handlers/workers on farms should be initiated to address the zoonotic potential of *G. duodenalis* in coypus.

#### **4. Materials and Methods**

#### *4.1. Ethics Statement*

This study was performed with strict adherence to the recommendations of the Guide for the Care and Use of Laboratory Animals of the Ministry of Health, China. The research protocol was reviewed and approved by the Research Ethics Committee of Tarim University (approval no. ECTU 2018-0026). Farm owners gave permission before we commenced fecal sample collection.

#### *4.2. Study Area and Sample Collection*

A total of 308 fecal samples were collected during autumn, winter, and spring of 2018 and 2019 from seven farms in Hebei Province (Baoding), Henan Province (Anyang and Kaifeng), Sichuan Province (Chengdu), Hunan Province (Yongzhou), Jiangxi Province (Ganzhou) and Guangxi Zhuang Autonomous Region (Laibin) in China (Table 1). The breeding scale of coypus was about 500–2000 in each farm, and the breeding conditions of each farm were basically the same. Approximately 10–15% of coypus representing each age group were investigated at each farm. All the fecal samples were collected immediately after excretion and placed in a plastic container; we recorded the date, site, age, and health condition at collection time. All samples were shipped to the laboratory in a cooler with ice packs within 48 h and stored at 4 ◦C.

#### *4.3. DNA Extraction and Nested PCR Analysis*

The genomic DNA sample was extracted from approximately 200 mg of each fecal specimen using a commercial E.Z.N.A® Stool DNA Kit (Omega Bio-Tek Inc., Norcross, GA, USA) following the protocol stipulated by the manufacturer. The quantity of nucleic acid in samples was photometrically estimated at OD260 and stored at −20 ◦C for subsequent molecular analysis.

Nested PCR amplification of the SSU rRNA gene was employed to screen *G. duodenalis* infection [39]. The *G. duodenalis*-positive samples were further analyzed via nested PCRs of gene loci *tpi*, *gdh*, and *bg* (see Table S1 and Figure S1) [8,22,40]. Samples positive for all four loci were used to assess the MLGs of *G. duodenalis*.

#### *4.4. Sequencing and Phylogenetic Analysis*

Bi-directional sequencing of all the secondary PCR products from SSU rRNA, *tpi*, *gdh*, and *bg* loci was performed at the biotechnology company, GENEWIZ (Suzhou, China). The sequence assembly and editing were performed with the DNAstar Lasergene Editseq 7.1.0. Multiple-sequence alignment analysis was employed on the obtained and GenBank reference sequences using software ClustalX 2.1 to ascertain the genotypes and subtypes of *G. duodenalis*. We concatenated sequences for each positive isolate to obtain a multi-locus sequence (*bg*, *tpi*, *gdh*) in accordance with a previous report [41]. Phylogenetic analyses of the concatenated MLG sequences were achieved using neighbor-joining methods based on the Kimura-2 parameter model in MEGA 7.0.
