Morphological and molecular characterization of invasive Biomphalaria straminea in southern China

Abstract

Background

Schistosomiasis is a common parasitic disease designated as a neglected tropical disease by the World Health Organization. Schistosomiasis mansoni is a form of the disease that is caused by the digenean trematode Schistosoma mansoni, transmitted through Biomphalaria spp. as an intermediate host. Biomphalaria was introduced to Hong Kong, China in aquatic plants shipments coming from Brazil and the snail rapidly established its habitats in southern China. Earlier studies of Biomphalaria spp. introduced to southern China identified the snails as Biomphalaria straminea, one of the susceptible species implicated in S. mansoni transmission in South America. However, recent molecular investigations also indicated the presence of another South American species, B. kuhniana, which is refractory to infection. As such, it is important to identify accurately the species currently distributed in southern China, especially with emerging reports of active S. mansoni infections in Chinese workers returning from Africa.

Methods

We combined morphological and molecular taxonomy tools to precisely identify Biomphalaria spp. distributed in Guangdong Province, southern China. In order to clearly understand the molecular profile of the species, we constructed a phylogeny using mtDNA data (COI and 16S rRNA sequences) from six populations of Biomphalaria spp. from Shenzhen City in Guangdong Province. In addition, we examined the external morphology of the shell and internal anatomy of the reproductive organs.

Results

Both morphological and molecular evidences indicated a close affinity between Biomphalaria spp. populations from Guangdong and B. straminea from Brazil. The shell morphology was roughly identical in all the populations collected with rounded whorls on one side and subangulated on the other, a smooth periphery, an egg-shaped aperture bowed to one side, and a deep umbilicus. The shape and number of prostate diverticula (ranged from 11.67 to 17.67) in Guangdong populations supports its close affinity to B. straminea rather than B. kuhniana. Molecular analysis did not conflict with morphological analysis. Little genetic differentiation was observed within Biomphalaria populations collected. Phylogenetic analysis of COI and 16S rRNA haplotypes from snails collected and B. straminea sequences from Brazil and China using Bayesian inference revealed that Guangdong populations were clustered in one clade with B. straminea from Hong Kong of China and B. straminea from Brazil indicating their close affinity to each other.

Conclusions

Data obtained in the current study clearly show that the populations of Biomphalaria spp. investigated are B. straminea, and we assume that those snails were either introduced via passive dispersal from Hong Kong of China or as a result of multiple introduction routes from Brazil.

Multilingual abstract

Please see Additional file 1 for translations of the abstract into the five official working languages of the United Nations.

Background

Schistosomiasis is an endemic parasitic disease affecting almost 240 million people worldwide, and an additional 700 million people are at risk of infection. The infection is prevalent in seventy-eight developing, tropical and sub-tropical countries [1, 2]. Some species of the genus Biomphalaria act as obligatory intermediate hosts for Schistosoma mansoni Sambon, 1907, the parasite responsible for intestinal schistosomiasis in different parts of the world. In the absence of either the parasite or the snails, the life cycle cannot be maintained [3, 4]. Efforts are thus directed toward eradicating intermediate hosts using molluscicides and/or the parasite via treatment of infected populations using praziquantel (the drug of choice for schistosomiasis treatment) [5,6,7,8]. Biomphalaria species have a close association with freshwater habitats closely related to human activities [9, 10]. The natural histories of the various species of Biomphalaria snails have been studied extensively, although taxonomic difficulties and population-level variation are among the factors which have confounded this task [5]. A detailed knowledge of the geographic distribution of Biomphalaria snails is important for the control of intestinal schistosomiasis and its epidemiologic surveillance [11]. Biomphalaria snails are thought to have evolved in South America and are currently distributed throughout the continent as well as in central America, the Caribbean, Africa and the Middle East [12, 13]. In addition to South America and Africa, Biomphalaria spp. was also reported from Asia in Hong Kong, China 40 years ago [14], as a result of human-mediated accidental introduction, likely via aquatic plants imported from Brazil for commercial use [15]. The introduced snail was identified as B. straminea (Dunker, 1848), which is known by its invasiveness and colonization capabilities [16].

In China, in the years following its introduction, B. straminea dispersed through different watercourses in southern China. The snail was identified in numerous water bodies in Shenzhen City, Guangdong Province, China in 1981 [17], and in the next 2 years it showed a wider distribution in Shenzhen rivers, suggesting an origination from Hong Kong’s river system [18]. This distribution pattern confirms the ability of B. straminea to survive and form new colonies in southern China and its ability to spread to new water bodies [19]. The continuous rising in Earth’s surface temperature and climate change [20], as well as the continued rise in the mean annual temperature of China [21], may accelerate B. straminea spread due to its ability to adapt to different climate and ecological conditions [22]. In addition, the risks for B. straminea to be the source of S. mansoni transmission has been evaluated with the following concerns: (i) the involvement of China in numerous projects in Africa and the export of labor services to regions endemic with African schistosomiasis; (ii) S. mansoni infections have been reported in Chinese workers returning from Africa [23]; (iii) The presence of snails and infected humans in the same settings will increase the potential of S. mansoni transmission to be established in southern China in view of the role played by environmental changes such as climate change and increase in water manipulation and intensification of irrigation [23, 24].

A multidisciplinary approach for identification of species of medical importance combining morphological and molecular techniques is fundamental for our understanding of diseases epidemiology and surveillance. Proper identification of the intermediate host will help to map the potential schistosomiasis transmission foci in endemic areas as well as in areas without the disease but with abundant susceptible species [25]. For example, although Biomphalaria snails found in southern China were referred to as B. straminea, a new challenge was encountered. In 2015 a morphological and molecular study for seven Biomphalaria populations from Guangdong Province, China showed that, five of the populations are likely to be Biomphalaria kuhniana (Clessin, 1883) but not B. straminea [26]. The study suggested multiple colonization events and displacement of B. straminea by the closely related species, B. kuhniana. The latter species is also a neotropical snail described from Suriname, but reported in other South American countries [27]. Together, B. straminea, B. kuhniana, and B. intermedia (Paraense and Deslandes, 1962), are known as B. straminea complex [28]. The previous three species differ greatly in their susceptibility to infection with S. mansoni; for example, B. kuhniana is proven to be refractory to infection [27]. Thus, an urgent taxonomy of these species should be carefully examined through combined analysis of morphological and molecular variables, and their biomedical importance in the transmission of S. mansoni should be evaluated. Careful identification of Biomphalaria snails in southern China will lead to a better delineation based on biological profiles related to the potential transmission risk, and will help to develop an appropriate intervention for schistosomiasis surveillance or control in potential foci, particularly in areas of constrained resources.

Methods

Snail sampling and collection sites

Biomphalaria snails were collected from six locations in Shenzhen during April and May, 2015 (Fig. 1). Sampling was quantified, with two collectors inspecting a 50–100 m length of the water body for 30 min and collecting all Biomphalaria found. A snail dip net with long handle was used for snail collection. Sampling was carried out at different water levels [29]. The materials recovered by the nets were examined for Biomphalaria snails, which were carefully collected with long blunt forceps and kept in plastic jars containing water from the same watercourse. Species identifications in the field were made through shell morphology, based on published keys [5, 30, 31]. As presented in Table 1, other gastropods were identified at species level [26, 32] and their presence recorded.

Fig. 1

Map of the surveyed sites for Biomphalaria snails in the South of China

Table 1 Habitats, number of snails and GPS coordinates for snails’ localities

Each location was georeferenced with a handheld geographical positioning systems (GPS) device (Garmin GPSMap 60CS, Garmin Ltd., Kansas City, USA) and the coordinates were recorded. The in situ water temperature (°C), microconductivity (measured in micro-siemens, μS), and pH, all factors that have been implicated as potentially important factors influencing freshwater snail distribution, were recorded [5, 33]. These water chemistry variables were measured using a PH-200 handheld portable pH and temperature meter and AP-2 aquapro conductivity meter (HM Digital, Inc., Culver City, USA). The substrate type was also observed. Collected Biomphalaria snails were sorted per site after returning to the laboratory. The snails were placed in individual wells filled with 2 ml of dechlorinated water and exposed to artificial light for 3 h to encourage emergence of trematodes cercariae. After exposure, the wells were inspected with a dissecting microscope for the presence of infection.

Identification of collected Biomphalaria snails

Morphological identification

Ten snails from each site were allowed to relax overnight in an aqueous solution of menthol, followed by rinsing in water heated at 70 °C for 40 s. The snails were then removed from their shells individually and the head-foot was cut off and placed in 75% ethanol for molecular work and the remaining soft tissue was placed in a modified Raillet–Henry fluid (distilled water 930 ml, sodium chloride 6 g, 40% formalin 50 ml, and glacial acetic acid 20 ml) [34]. Shells were cleaned with 0.5% sodium hypochlorite (CLOROX), and photographed using an AxioCam attached to a stereomicroscope (Olympus SZX12, Japan). Two sets of morphological measurements were taken: conchological and internal anatomical measurements. The number of whorls was counted using a hand held magnifying lens. The shell height, shell diameter, diameter of the umbilicus, ratio between shell height and diameter were measured using digital calipers (accurate to 0.01 mm). The soft tissue preserved in Raillet–Henry fluid (5 per site) was dissected to reveal the internal anatomy of each snail, and particularly the reproductive organs. After dissection of the reproductive system under the stereoscopic microscope, three measurements were made: length of penis sheath, length of preputium and number of prostate diverticula. Mean ratios between shell height and shell diameter and length of penis sheath and length of preputium were compared by an analysis of variance (ANOVA) and Scheffé test [35]. Differences were considered significant at P < 0.05.

Molecular characterization of collected snails

Genomic DNA was extracted separately from 10 head-foot samples preserved in ethanol using DNeasy blood and tissue kit (Cat No. 69506, QIAGEN GmbH, Hilden, Germany) following the manufacturer instructions and stored at 4 °C. Polymerase chain reaction (PCR) amplifications were made using Bio-Rad C1000 thermal cycler (Bio-Rad, USA). Two regions of the mitochondrial genome; the cytochrome oxidase subunit I (COI) gene and the 16S sub-unit of ribosomal RNA (16S rRNA) were amplified. The general PCR cycle used was 2 min initial denaturing at 94 °C, followed by 35 cycles of 1 min at 94 °C, 45 s at the annealing temperature and 1 min extension at 72 °C; primer details and annealing temperatures can be found in Table 2 [13, 36, 37]. PCR products were visualized on a 2% agarose gel stained with Goldview (No. 141008, SBS, China). The PCR products were then sent to Sangong Biotech Company (Shanghai, China) for sequencing.

Table 2 Primer sets with sequences for amplification of cytochrome oxidase I (COI) and 16S rRNA

Sequences alignment and genetic analysis

Nucleotide sequences were visually edited using Bioedit software v 7.0 [38]. BLAST searches were performed for the obtained sequences at NCBI (http://www.ncbi.nlm.nih.gov/) against the GenBank database for initial identification of the collected snails. The sequences were aligned using Clustal X [39]. The COI and 16S rRNA data sets were phylogenetically analyzed separately and as a combined data set. DNA samples of B. straminea, B. kuhniana, and B. intermedia (B. straminea complex) from some Brazilian states and from Colombia were obtained from Medical Malacological Collection (Fiocruz-CMM) of René Rachou Institution. The 16S rRNA gene was sequenced from these samples and used in the phylogeny construction of the 16S rRNA tree. Unfortunately, the COI gene could not be sequenced as most of the DNA was lost during the shipment process and the remaining amount was only sufficient for testing one genetic region (16S rRNA). The alignments of all data sets were based on gap-free sequences. Phylogenetic relationships were estimated using Bayesian inference in MrBayes version 3.2.0 programs [40], using haplotypes from the current study and all known sequences of B. straminea and B. kuhniana from GenBank. Biomphalaria tenagophila (Orbigny, 1835) was chosen as an out-group. Prior to Bayesian inference, the best fit nucleotide substitution models (HKY for COI and for combined data set and TrN for 16S rRNA) were determined using a hierarchical likelihood ratio test in jMODELTEST version 0.1.1 [41]. The posterior probabilities were calculated via 6 000 000 generations using Markov chain Monte Carlo (MCMC) simulations, and the chains were sampled every 1000 generations. At the end of this run, the average standard deviation of split frequencies was below 0.01, and the potential scale reduction factor was reasonably close to 1.0 for all parameters. A consensus tree was summarized and visualized in FigTree version 1.4.3 (http://tree.bio.ed.ac.uk/software/figtree/). Evolutionary analyses were conducted in MEGA6 [42] using the Maximum Composite Likelihood model and codon positions included were 1st + 2nd + 3rd + Noncoding. All positions containing gaps and missing data were eliminated.

Results

Collection of snails and study area

Our malacological surveys included six separate locations in Shenzhen City (Fig. 1 and Table 1). All the locations were typical habitats for snails and all were positive for the presence of Biomphalaria with variable population densities with numbers ranging from 50 snails as in Hubin, Xinzhen River to 850 snails in Tuyang, Kuichong River. The largest numbers were found near the streams’ banks crawling on substrates or attached to aquatic plants. The depth of the habitats was qualitatively moderate (10–50 cm). No significant differences were observed in the physiochemical properties of water bodies harboring the snails. The water temperature ranged from 27 to 29 °C, pH from 7.4 to 8, and conductivity from 190 to 570 μS. The total dissolved solids were within the range of 0.18–0.28 parts per thousand (ppt). The type of the substratum was variable from one habitat to another. The substrate was formed of mud, sand, rock or a combination of these substrates. The most common freshwater snails co-localized in most of the habitats were Pomacea canaliculata (Lamarck, 1828) and Physella acuta (Draparnaud, 1805). The screening of collected snails for natural trematode infections revealed the absence of any trematode larvae.

Morphology

Shell characteristics were roughly the same in most of the collected Biomphalaria snails, with rounded whorls on one side and subangulated on the other, a smooth periphery, an egg-shaped aperture bowed to one side, and a deep umbilicus. Most of the snails shared the same number of whorls (4 whorls). Shell measurements were variable among the populations, with significant differences in overall length and height (P < 0.001 by ANOVA) among the different populations though no significant differences were observed in the length to height ratio (Fig. 2 and Table 3).

Fig. 2

Shells of Biomphalaria straminea. Collected from different localities in Guangdong Province. a: Donghu Park, b: Shahexi Road, c: Guihua Road, d: Hubin Road, e: Jinkui, f: Tuyang. Bar = 5 mm

Table 3 Measurements of shell and reproductive organs of Biomphalaria straminea collected from different localities in southern China (Mean ± SD)

Similar to the external morphology of the shell, the internal anatomical features of the collected snails, as inspected by careful and precise dissection, did not show significant variability among locations. The general feature of the reproductive system revealed many constant characteristics across the collected Biomphalaria populations. For example, the corrugation of vaginal wall is conspicuous; the ovotestis diverticula were short, unbranched; and the seminal vesicle had apparent diverticula. The spermatheca was found round or oval in shape. The prostate contained from 10 to 22 branched diverticula. The spermiduct wrinkle is very prominent. The penis sheath was longer than the preputium in all populations (Fig. 3). The ratio between penis sheath length to preputium length was the highest in Jinkui snail’s population with 1.41 ± 0.28 mm and lowest in Shenzhen reservoir snails which measured 1.24 ± 0.08 mm. The situation was reversed regarding the number of prostate diverticula where Shenzhen Reservoir populations significantly differed from Jinkui populations, with a mean number of 20.6 ± 2.3 and 10.8 ± 0.83 diverticula, respectively (Table 3). The prostate diverticula of Guangdong populations of Biomphalaria were very much similar to those recorded for B. straminea rather than B. kuhniana from Brazil (Fig. 4).

Fig. 3

Anatomy of the reproductive system of Biomphalaria straminea, Tuyang, Shenzhen, Guandong Province. AG, albumin gland; Ca, carrefour; NG, nidamental gland; Od, ovispermiduct; Ot, ovotestis; Ov, oviduct; Po, pouch of the oviduct; PP, preputium; Pr, prostate; PS, penis sheath; Sd, spermduct; Sp, spermatheca; SV, seminal vesicle; Ut, uterus; Va, vagina; Vd, vas deferens. Bar = 1 mm; insert = 500 μm

Fig. 4

Prostate diverticula of Biomphalaria straminea from Tuyang, Shenzhen, Guangdong, China (a) B. straminea from Tangará da Serra, Mato Grosso, Brazil (b) and B. kuhniana from Tucurui, Tocantins river, Pará, Brazil (c). Both (b) and (c) are modified from Paraense [28]. Bar = 200 μm in (a) and 1 mm in (b) & (c)

Molecular data

Few variations were observed in the sequences of COI and 16S rRNA mitochondrial DNA regions from the 10 samples from each location site. Analysis of each mitochondrial gene alone revealed four (GenBank accession numbers: MF491792 − MF491795, 557 bp) and three (GenBank accession numbers: MF491796 − MF491798, 358 bp) haplotypes of Chinese Biomphalaria, respectively for COI and 16S rRNA without including gaps. There were only five variable sites for COI and the haplotype diversity was 0.8333, while for 16S rRNA there were three variable sites with a haplotype diversity of 1.00. Using the combined data set, six haplotypes were identified from 60 sampled sequences (distribution of all haplotypes is presented in Table 4). The number of variable sites was seven and haplotypes diversity was 0.9333. Both COI and 16S rRNA haplotypes showed low genetic divergence. COI haplotypes showed 0.002–0.007 pairwise distance and the number of base substitutions per site based on averaging over all haplotypes sequence pairs was 0.005 ± 0.002 (as obtained by a bootstrap procedure of 1000 replicates). The nucleotide diversity was 0.004488 among COI haplotypes. For 16S rRNA, the pairwise distance was 0.001–0.005 and the overall pairwise distance was 0.003 ± 0.003 while the nucleotide diversity was 0.005587. The divergence between COI + 16S rRNA haplotypes obtained from analyzing the combined data set was 0.001–0.005 pairwise distance with an average of 0.003 ± 0.001. The mean corrected distance across the combined haplotypes data set as a whole was 0.3% divergence and the nucleotide diversity was 0.002794.

Table 4 Guangdong haplotypes of Biomphalaria straminea and their distribution across the collection sites

All the populations conformed to the assumptions of the neutral hypothesis, as tested by Tajima’s D and significance of Tajima’s test for neutrality based on the total number of mutations [43]. No population was found to deviate from the assumptions of normality using the three data sets. The D values were − 0.796844, − 0.897067 and − 1.390308, respectively, for COI, 16S rRNA and COI + 16S rRNA data sets.

The samples of B. straminea, B. kuhniana and B. intermedia (B. straminea complex) that were obtained from Brazil and Colombia were also analyzed for unique 16S rRNA haplotypes in order to be used in the phylogenetic analysis. These sequence haplotypes of B. straminea complex were deposited in GenBank (Table 5).

Table 5 Haplotypes of South American Biomphalaria straminea complex, their location of collection and GenBank accession numbers

Phylogenetic reconstructions of COI haplotypes and GenBank sequences for B. straminea from Brazil and China using Bayesian inference with B. tenagophila (AF199089.1) as an outgroup resulted in a consensus tree with major nodal support values obtained based on the Bayesian posterior probabilities. The Guangdong populations and B. straminea from Hong Kong formed one monophyletic clade with a high nodal support indicating their close affinity to each other. In the same clade four sequences for B. straminea from São Paulo were also clustered with Chinese haplotypes. Basal to this clade situated B. intermedia from São Paulo with 0.62 nodal support value (Fig. 5).

Fig. 5

Bayesian phylogram based on analysis of COI sequences of Guangdong populations of Biomphalaria straminea (haplotypes) and Brazilian sequences retrieved from GenBank using B. tenagophila as an outgroup. Support values for individual branches are given as posterior probabilities based on 6 000 000 generations in a Bayesian analysis. Bs, B. straminea; Bt, B. tenagophila; SZ, Shenzhen

The analysis of 16S rRNA data set of Guangdong populations (haplotypes), South American samples of B. straminea complex (haplotypes) and reference sequences retrieved from GenBank using B. tenagophila as an outgroup showed the clustering of Guangdong populations in one clade with B. straminea sequences from Hong Kong of China and Brazil (Distrito Federal, Belem Para, and Pernambuco). Sequences of B. straminea from Minas Gerais were basal to this clade with a nodal support of 0.82 posterior probabilities. Biomphalaria kuhniana sequences formed one monophyletic group comprising samples from Brazil, Dominica and Colombia. Both of B. straminea and B. kuhniana groups described above shared a common ancestral sequence for B. straminea from Pernambuco, Brazil. Biomphalaria intermedia sequences clustered together in two monophyletic clades that were distal from Chinese B. straminea (Fig. 6).

Fig. 6

Bayesian phylogram based on analysis of 16 s rRNA data set for Guangdong populations of Biomphalaria straminea (haplotypes), South American samples from B. straminea complex (haplotypes) and reference sequences retrieved from GenBank using B. tenagophila as an outgroup. Support values for individual branches are given as posterior probabilities based on 6 000 000 generations in a Bayesian analysis. Bi, B. intermedia; Bk, B. kuhniana; Bs, B. straminea; Bt, B. tenagophila; SZ, Shenzhen

The analysis of the combined data sets (COI + 16S rRNA) showed a tree topology relatively similar to that obtained with COI data set. Guangdong populations were clustered in the same clade with B. straminea from Hong Kong and those previously identified from Yantian, Guangdong [22]. Basal to this clade was B. straminea from Chacara Azaleia, São Paulo, Brazil (LBMSU434) with strong nodal support > 0.70. Together the Guangdong populations from China and Brazilian isolates shares a common ancestor B. straminea from Corrego do Pinheiro, São Paulo, Brazil (LBMSU430). B. tenagophila (LBMSU422) from Araraquara, Rio Ouro, São Paulo was used as an outgroup (Fig. 7).

Fig. 7

Bayesian phylogram based on analysis of combined COI and 16 s rRNA dataset for Guangdong populations of Biomphalaria straminea (haplotypes) and reference sequences retrieved from GenBank using B. tenagophila as an outgroup. Support values for individual branches are given as posterior probabilities based on 6 000 000 generations in a Bayesian analysis. Bs, B. straminea; Bt, B. tenagophila; SZ, Shenzhen

Discussion

The present study attempted to identify the species of Biomphalaria snails introduced to southern China using morphological and molecular identification methods. The early reports on Biomphalaria invasion to Hong Kong identified the snails as B. straminea [14, 32, 44]. Subsequent invasions and spread were reported from other water bodies in southern China such as Guanlan and Dasha rivers in Shenzhen City, Guangdong Province’s rivers [18]. This distribution scheme led to the hypothesis of multiple invasions and colonization including different roots of introduction. Moreover, recent molecular investigation of several Biomphalaria populations from Guangdong indicated that most of the species currently harboring water systems are B. kuhniana not B. straminea [26]. This conclusion merited further careful investigation because B. kuhniana has so far been shown to be refractory to S. mansoni infection [27, 45], whereas B. straminea is generally considered to be a susceptible species, albeit in some cases at a low level, that can nonetheless transmit schistosomiasis [46,47,48,49,50], and even sustain natural transmission in some settings [13]. It is worth noting, however, that the taxonomic confusion within Biomphalaria has led some to speculate that naturally-infected B. straminea found in Venezuela may have in fact been B. kuhniana [13]. The fact remains that based on current data, B. straminea is generally considered to be of higher biomedical concern.

The identification of Biomphalaria was performed through morphological study of five specimens from each location sites in Shenzhen, Guangdong Province. Shell diameters ranged from 7.59 ± 0.45 mm as in Shenzhen reservoir population to 9.70 ± 0.93 mm in Guanlan River population with significant differences among the populations. According to Paraense [28], the average shell of adult B. kuhniana is smaller than 7.5 mm while that of B. straminea is from 11 mm to 16.5 mm. Although almost all of the current measurements are higher than 7.5 mm (the average diameter of B. kuhniana), which may support the snails to be B. straminea, the shell diameter is of limited value in discrimination between B. straminea and B. kuhniana due to the difficulty in recognizing full-grown shells in these species of planorbids [28]. Also, studies by Dupouy et al. [51] on the African pulmonate snail Biomphalaria pfeifferi (Krauss, 1848) concluded that environmental components may influence shell morphology through phenotypic versatility and/or through impacts of local natural selection. The differences observed in the shell size of Biomphalaria collected in this study could be attributed to environmental factors as well as the snails’ ages. The data from internal anatomy are considered more informative than external features. For example, the number of prostate diverticula varies significantly between B. kuhniana and B. straminea, with 4 to 9 observed in B. kuhniana versus 9 to 18 in B. straminea [28]. In the present study, in all the snails collected, the number of prostate diverticula ranged from 10.8 ± 0.83 to 20.6 ± 2.3 and the diverticula were long and branched. Similarly, the anatomy of reproductive organs of Hong Kong snails of Biomphalaria revealed an average number of prostate diverticula ranged from 11.67 to 17.67 [52] and both are close to the numbers recorded for South American B. straminea (9–18 prostate diverticula) [28]. The corrugations of vaginal wall and of the spermiduct are very prominent, similar to B. straminea from Brazil. The shape and number of diverticula in Guangdong populations supports its close affinity to B. straminea rather than B. kuhniana (Fig. 4).

Molecular based taxonomy of snail intermediate hosts has been developed to overcome erroneous species identification resulting from morphometric measurements. Mitochondrial data are reliable and informative and its consideration in snails taxonomy stems from the fact that the COI mitochondrial gene proved to be an ideal barcode locus and default marker adopted by the Consortium for the Barcode of Life for all groups of organisms [53, 54] including Biomphalaria spp. [55,56,57,58]. In the current study, molecular evidence, retrieved from sequencing two mitochondrial loci (COI and 16S rRNA), were in harmony with anatomical evidence and suggested a close affinity between Guangdong Biomphalaria and B. straminea from Hong Kong of China and Brazil as showed in the cladograms of the data sets analyzed. The absence of existing COI data for B. kuhniana limits its use as an independent identifier although the analysis of this locus in combination with 16S rRNA gave a quite similar tree topology as obtained with COI alone. Moreover, COI analysis excluded the possibility of Guangdong populations of Biomphalaria to be B. intermedia. The haplotypes from the present study were clustered in one clade with B. straminea sequences from Hong Kong, Guangdong of China and Brazil. B. intermedia originated from Brazil (KF926179.1 [59]) was situated in a different clade basal to the previous one with a nodal support of 0.62 posterior probabilities (Fig. 5). The phylogenetic tree in Fig. 5 showed that the COI sequences of the samples collected were identical to those of B. straminea from Brazil and those previously deposited in GenBank for Chinese B. straminea (Fig. 5). It is worthy to mention that looking for similar sequences for COI sequences deposited by Atwood et al. [27] for B. kuhniana (KJ137284 − KJ137287) using the BLAST tool showed that those sequences are 98–100% identical to B. straminea isolates of China and Brazil indicating that the authors may misidentified the samples collected. For 16S rRNA, enough data were available in GenBank records in addition to the sequences obtained from samples of B. straminea complex obtained from Brazil and Colombia. Guangdong haplotypes were clustered in the same clade with Hong Kong isolates of B. straminea (KY697220.1, KY697223.1 and KY697236.1 [52]). Both Guangdong and Hong Kong isolates from China were clustered with Brazilian samples of B. straminea from Distrito Federal and Pernambuco and one GenBank isolate from Belem Para, Brazil (AY030213.1) with a major nodal support values. B. kuhniana sequences represented an in-group clade with a nodal support of 0.96 posterior probabilities. This tree construction indicates that the Guangdong populations are in close affinity with B. straminea present in Hong Kong and both share South American ancestors of B. straminea. This finding is in accordance with Zeng et al. [52] who identified two phenotypes of Biomphalaria snails (black- and red-colored shell Biomphalaria) in sites located at the border between Hong Kong and Shenzhen (few miles away from the samples collected in our study) and in the New Territories in Hong Kong, including places adjacent to the mainland China (30–40 miles away from our samples). The two phenotypes were genetically indistinguishable and were similar to B. straminea abundant in mainland China and South America and suggested that the B. straminea snails in Hong Kong are South American in origin.

In the present study, both Guangdong populations from China and Brazilian B. straminea formed a clade of closely related species with a minimal genetic divergence between them (Figs. 5, 6 and 7). Moreover, the sequence divergence estimates (0.3%) and strong support for the relationship between these populations (as demonstrated by Bayesian posterior probability values) suggest that these are the same species, and closely related populations, rather than a separate species. Independent analyses of COI and 16S rRNA sequences showed low genetic variability with four and three distinct haplotypes, respectively for both genes, out of 60 sequences representing six different populations of Biomphalaria snails collected from Shenzhen. The low nucleotide diversity observed within the Guangdong populations (COI, 0.004488; 16S, 0.005587; COI + 16S rRNA, 0.002794) is consistent with the literature; in a recent study by Attwood et al. [26], the authors found three distinct COI sequences, with a nucleotide diversity of 0.00279, and only one unique 16S rRNA sequence.

The observed levels of genetic diversity across the Guangdong populations of Biomphalaria, suggests a more likely recent connection between these populations in view of the fast divergence of mitochondrial genes in pulmonates [60,61,62] which is expected in this recently introduced snail species. According to Attwood et al. [26], the observed divergence is relatively high given the small area of sampling, and they suggest that the diversity is not due to mutations occurring in Guangdong populations from China but rather was exhibited in ancestor Brazilian populations, indicating that the current Guangdong populations could represent multiple, independent, introductions followed by subsequent dispersal and colonization. This multiple colonization hypothesis is also supported by a previous study of 19 electrophoretlcally detected enzyme loci in four Hong Kong populations of B. straminea that showed high levels of variability among the populations. This variability resulted from multiple introduction events between 1973 and 1981 [44]. Although the relatively high genetic diversity at the COI (0.004488) locus, in comparison with other pulmonates, assumes divergence and mutations in ancestors’ snails at the introduction source, it can also be attributed to the reproductive traits of B. straminea and their high selfing rates. The presence of the populations in the same cladogram with B. straminea from Hong Kong (Fig. 5) may support the hypothesis that part of the snails currently distributed in Guangdong could be present as a result of peripheral dispersal of descendant snails from Hong Kong populations via passive transport or container ship traffic, or naturally due to connections between the water systems of the two adjacent localities. Finally, the genetic similarities among the snail populations strongly support the idea that these are derived from the original colonist populations via dispersal [63].

Preliminary experimental infection of filed captured-B. straminea from Shenzhen and their first generation with a Puerto Rican strain of S. mansoni showed, at different ratios of miracidial exposure, no infection in the exposed snails [64]. Although this finding suggests that the invasive B. straminea snails are incompatible with S. mansoni and thus there is a low risk of S. mansoni transmission in mainland China, it may not be appropriate to consider this as a final conclusion because the compatibility was tested with only one strain of S. mansoni from South America and the experiments were undertaken with the first generation of laboratory bred B. straminea. Further studies are needed to test the compatibility of different isolates of both Chinese Biomphalaria and S. mansoni from a variety of locations especially African strains because most of the imported cases of human infection are from Africa and the number is anticipated to increase due to the expansion in Chinese projects in Africa [23].

Conclusions

The reassessment of the identity of Biomphalaria species present in southern China using morphological and molecular delimitation procedures suggests that B. straminea is the only Biomphalaria species abundant in southern China. However, to validate the current conclusion and in order to obtain a more reliable barcoding database for future studies, COI sequences from original reference snails (for B. straminea and B. kuhniana) from South America are required, and should be deposited in GenBank or other repositories. The current finding is important from a public health perspective due to the biomedical importance of B. straminea as an intermediate host for Schistosoma mansoni, and the potential risk of Schistosoma mansoni transmission to mainland China in view of the increase in labor services exchange with African countries where the parasite is endemic. More studies should be conducted to understand the biology of this snail and its competitive capacity with other local snail species and to describe its range of expansion. Moreover, the susceptibility of these local populations to S. mansoni should be tested. Better understanding of the population structure of this intermediate host snail is important to elucidate its potential role in schistosomiasis transmission, considering factors that may affect population structure such as environmental and demographical stability. Finally, it is widely encouraged to increase surveillance for all entry ports and checking for any exotic species, in order to reduce the likelihood of future introductions of potentially invasive species [64, 65].