Biomphalaria camerunensis as a viable alternative intermediate host for Schistosoma mansoni in southern Cameroon

Background Intestinal schistosomiasis due to Schistosoma mansoni was mapped in Cameroon in the 1990s and preventive chemotherapy launched since 2005. A situation analysis conducted in 2011 revealed an increase in schistosomiasis transmission, especially in the equatorial part of the country, despite the fact that Biomphalaria pfeifferi, the main intermediate host of this parasite, is now scarce in many foci. Biomphalaria camerunensis, restricted to the equatorial part of the country, is considered as a less suitable host for S. mansoni due to it resistance to the parasite, although exhibiting a better survival than B. pfeifferi. In a context where human migration is quite frequent as a consequence of terrorism, war in neighboring countries, as well as development of hydraulic projects, it seems appropriate to evaluate the current epidemiological role of B. camerunensis to estimate the risk of extension of S. mansoni distribution in Cameroon. To do this, the susceptibility of three B. pfeifferi and five B. camerunensis populations to a strain of S. mansoni was assessed. Juvenile snails (G1) of each population were infected with S. mansoni miracidia, and prepatent period, infection and survival rates of infected snails, as well as cercarial production were recorded and compared between snail species and populations. Results Compatibility tests were performed on a total of 827 snails: 344 B. pfeifferi and 483 B. camerunensis. Infection rates were quite heterogeneous, higher in B. pfeifferi (61.5%) as compared to B. camerunensis (7.8%) (Chi-square test: χ2 = 258.88, df = 1, P < 0.0001). All the three B. pfeifferi-infected populations were susceptible to S. mansoni. However, among the five B. camerunensis populations tested, four were susceptible to S. mansoni, with 21.9% as the highest infection rate. The prepatent period was, on average, shorter in B. pfeifferi than in B. camerunensis (P < 0.0001), but the cercarial production was significantly higher in B. camerunensis as compared to B. pfeifferi (P < 0.001). Conclusion These findings indicate that B. camerunensis populations might substantially contribute to S. mansoni transmission and dissemination in Cameroon, their low susceptibility being compensated by their high cercariae production. More attention and surveillance towards this species are required to achieve intestinal schistosomiasis elimination in Cameroon. Electronic supplementary material The online version of this article (10.1186/s13071-018-2763-2) contains supplementary material, which is available to authorized users.


Background
Schistosomiasis has been known to be endemic in Cameroon for more than three decades [1,2]. Control activities started with the creation of the National Program for the Control of Schistosomiasis and Soil Transmitted Helminthiasis (PNLSHI) in 2003, and have been rapidly scaled up. In addition to the mass administration of praziquantel to school-aged children (5-14 years) since 2005, the schistosomiasis control strategy in Cameroon also relies on education of communities about the disease and risks of infection, and the promotion of hygiene and environmental sanitation. These actions have contributed to significantly reducing the prevalence of this disease in several foci [3,4]. Despite these control measures, the current epidemiological situation reveals the persistence of schistosomiasis in some foci, and the emergence of new foci of the disease [3][4][5]. The establishment of new foci could be explained by the migration of parasitized individuals to disease free sites where sanitation is inadequate and intermediate hosts present [6]. To efficiently interrupt schistosomiasis transmission, the WHO recommends a synergy of actions, among which the fight against intermediate hosts holds a prominent place [7].
Intestinal schistosomiasis due to Schistosoma mansoni is the most largely distributed form of schistosomiasis in equatorial zones of Cameroon. S. mansoni is transmitted by two snails species of the genus Biomphalaria (B. pfeifferi and B. camerunensis), which present a wide geographical distribution though S. mansoni remains focalized. It was shown that B. pfeifferi preferentially reproduced by self-fertilization [8,9], thus exhibiting low genetic diversity within populations, and low survival of the offspring [9]. It is the main S. mansoni host in sub-Saharan Africa [10,11], including Cameroon where the most active S. mansoni foci are maintained by this snail species. This intermediate host is, however, becoming scarcer in many sites in Cameroon where it used to be abundant, likely due to climatic and environmental changes operating as a consequence of global warming. B. camerunensis species is expected to present a more polymorphic structure and a better offspring survival as an advantage of its preferential outbreeding reproduction system [9,12]. This species is found in the equatorial bioclime zone of the country, and was previously described as poorly susceptible to S. mansoni [11]. Two B. camerunensis sub-species are known to prevail in Central Africa, B. camerunensis camerunensis found in many localities in Cameroon, and B. camerunensis manzadica described in Bas-Congo in Democratic Republic of Congo (DRC) [13]. The latter species was shown to be involved in the creation and maintenance of S. mansoni active foci in DRC [14], and according to Frandsen [10], some B. camerunensis populations are as susceptible to S. mansoni as B. pfeifferi populations.
In Cameroon, human migration is quite frequent, mainly as a consequence of job seeking and instability related to wars and terrorism. The most important migration flow is originated from the Far North Region and from neighboring countries (Nigeria, Chad, Central African Republic) where S. mansoni is highly endemic, toward the South Region where many structuring projects (dams, ports and road construction) are ongoing. Indeed, a relatively high prevalence of onchocerciasis has been found in migrants escaping civil war from Central African Republic and who have now settled in the East Region of Cameroon), raising the concern of crossborder transmission of this filarial infection (Nana Djeunga et al., personal communication).
The aim of this study is to investigate the susceptibility of different B. camerunensis populations to S. mansoni, in order to assess the risk of establishment of new intestinal schistosomiasis foci as a corollary of potential introduction of foreign parasite in areas where this snail species is largely distributed.

Snail collection and maintenance
Snails were collected in eight sites (Fig. 1), using either a stainless steel sieve mounted on a 1.5 m wooden handle to comb aquatic vegetation, or by hand picking from the mud. Biomphalaria pfeifferi snails were collected in three sites, namely Ekorezok (3°53'327"N, 11°27'412"E) in the Center Region, Gounougou (9°4'33"N, 13°42'25"E) in the North Region, and Mokolo (10°44'00"N, Once in the laboratory, snails collected were sorted (based both on their shell morphology [15] and polymerase chain reaction restriction fragment length polymorphism analysis (PCR-RFLP) of the ITS2 region [16]) and distributed individually in 100 ml well labeled plastic boxes containing spring water. A piece of polystyrene was added in each rearing box for eggs laying. Overall, 15 sexually mature wild snails were isolated from each of the eight populations. One to two series of egg-laying were collected from each individual snail. After hatching, offspring were reared separately for one month [9,17]. This protocol was therefore important to monitor the number of wild snails contributing to the constitution of offspring samples to be exposed to miracidia.

Acquisition of schistosomes
A parasitological survey was performed in the Nkolbisson quarter (Ekorezok site), located in Yaounde (Cameroon City Capital), at about 7 km from the central town. Eligible individuals were those living near the Afeme River and who have been regularly in contact with the infested water of the river as a consequence of their professional and household activities. Stool samples were collected from each volunteer and analyzed in the laboratory using the Kato-Katz technique [18]. A total of nine stool samples with an intensity of infection ranging between 200 and 1700 eggs per gram of stool (epg) were used for parasite acquisition. Eggs were obtained after a series of washing and sedimentation of feces using saline solution, then spring water. Isolated eggs hatched under the action of both osmotic and thermal shock. Indeed, the pellet obtained after sedimentation was exposed to an artificial source of light for 30 min. After hatching, these eggs released miracidia necessary for the infection of the juvenile snails.

Snails' infection and follow-up of compatibility metrics
For each population, two to six young snails (2-4 mm), originating from each of the 15 wild parents, were individually put in the wells of a microtiter plate containing spring water. Five miracidia were then pipetted under a binocular dissecting microscope, and transferred in a well containing a young snail. This exposition of young snails to parasites (miracidia) was performed at room temperature (26 ± 1°C) for 4 h. A total of 827 snails (344 and 483 of B. pfeifferi and B. camerunensis, respectively) were exposed to the parasite. Each potentially infected offspring was returned back to its rearing box. Between 20 and 60 days post-infection, infected snails were followed up daily for cercarial shedding. The prepatent period (the duration of the larval development up to the first cercarial shedding) and the number of snails surviving at day 60 post-infection were recorded. The infection rate of each population was estimated as the ratio of the number of living individuals producing cercariae on the number of exposed living individuals.
Cercarial production was assessed from 40 days postinfection, when most of the potentially infected snails were susceptible to start shedding cercariae. Cercarial production was followed up for five consecutive days, for 16 and 8 infected individuals belonging to B. pfeifferi (from Nkolbisson) and B. camerunensis (from Petponoun), respectively. Snails were kept individually at room temperature (26 ± 1°C) for 24 h in 20 ml of water, and cercariae were counted under a dissecting microscope in a checkered plate containing 5 ml of this water and one or two drops of Lugol acting as a fixative for the cercariae.

Data analysis
Relevant data were entered into a purpose-built Microsoft Office Excel dataset, and analyzed using both the PASW Statistics Version 18 (SPSS Inc., Chicago, IL, USA) software, and the online program Vassarstat statistical computational web site [19], the latter being particularly useful for the calculation of the 95% confidence intervals. The mortality and infection rates were calculated and then compared between the different snail populations using the Chi-square test; Yates correction for continuity and Fisher's exact tests were used for small sample sizes (calculated or expected value Although a normal distribution was not expected for count data, the normality of distributions was tested and nonparametric tests chosen to analyze quantitative data. The non-parametric Mann-Whitney rank test was used to compare the duration of the prepatent period between species as well as the cercarial production between the different days of follow-up, and the two species. The non-parametric Kruskall-Wallis test was used to compare the duration of the prepatent period between snails populations, and pairwise comparisons performed when the difference was significant. The threshold for significance was set at 5% for all tests.

Results
A total of 344 B. pfeifferi and 483 B. camerunensis snails were exposed and followed up to investigate the compatibility between Biomphalaria spp. and S. mansoni.
Pairwise comparisons of infection rates among populations, for both B. pfeifferi and B. camerunensis, are given as supplementary material (Additional file 1: Table S1).
During the patent period, a high mortality rate was observed in the different populations. Although remaining similar between the two snail species (Chi-square test: χ 2 = 0.359, df = 1, P = 0.549), the mortality rates observed were significantly different between both B. pfeifferi populations (Chi-square test: χ 2 = 128.18, df = 2, P < 0.0001) and B. camerunensis populations (Chi-square with Yates correction for continuity: χ 2 = 41.78, df = 4, P < 0.0001) ( Table 1). The mortality rate among S. mansoni-infected individuals was significantly higher in allopatric populations (Mokolo and Petponoun) as compared to their rates in the sympatric population (Nkolbisson) to the parasite. Pairwise comparisons of mortality rates among   Table S2.

Duration of the prepatent period
The duration of the prepatent period varied from 21 to 40 days (Mean: 22.51, SD: 1.945) and from 28 to 40 days (Mean: 29.65, SD: 3.442) for B. pfeifferi and B. camerunensis populations, respectively (Table 1). It was significantly different both between the two snail species Cercarial production Table 2 summarizes the cercarial production follow-up for the two Biomphalaria species. The cercarial production was similar in both snail species on day 1 (Mann-Whitney U-test: U = 41.00, Z = -1.41, P = 0.158), but significantly more important in B. camerunensis than in B. pfeifferi from day 2 onwards (Mann-Whitney U-test: U = 7.00, Z = -2.67, P < 0.008) ( Table 2). Some of the infected B. pfeifferi survived more than eight months postexposure while continuing to shed cercariae. The mean value of daily cercarial production in B. camerunensis/S. mansoni combination was 1968.38 cercariae. Globally, cercarial production over five days of follow-up was significantly more important in B. camerunensis than in B. pfeifferi (Mann-Whitney U-test: U = 9.00, Z = -3.37, P = 0.001).

Discussion
This study was conducted in Cameroon in a context where S. mansoni transmission is persisting, with the creation of new foci [3,4]. This situation might have been driven by a number of reasons [18], and adaptation of a 'foreign' parasite to an intermediate host appears essential to the creation of new foci for disease expansion.

Compatibility
The infection rate was quite heterogeneous in the present study, with differences observed both between the two species and within the same species, thus confirming the important polymorphism of the susceptibility of Biomphalaria species [10,[20][21][22]. These results likely indicate an important genetic diversity among snail populations as was previously observed in Cameroon [12] and Madagascar [23]. The highest infection rate was observed in the sympatric pair B. pfeifferi/ S. mansoni of Nkolbisson, suggesting that this parasite is best adapted to its local snail population as previously described [15,22], though local adaptation is not always the rule [24]. This parasite strain seems to be less adapted to the population of northern Cameroon (Gounougou and Mokolo) as supported by the significantly high mortality rate in S. mansoni-infected snails belonging to these populations. This could be explained by the quite different conditions (climatic and phytogeographical) prevailing in these areas compared to that where the parasite is originated from (Fig. 1). Most of the B. camerunensis populations exhibited low infection rates (one being totally refractory) to the S. mansoni strain from Nkolbisson. This observation might suggest that these populations are resistant to the penetration and development of S. mansoni as a consequence of genetic predisposition of these snails to destroy the parasite as a foreign entity. A similar result was obtained by Simões et al. [6] in populations of Biomphalaria taenagophila and B. straminea, intermediate hosts of S. mansoni in Brazil. However, the Petponoun population exhibited a relatively high susceptibility rate (21.9%), quite similar to that of other B. pfeifferi [10], B. glabrata (the main host of intestinal schistosomiasis in Brazil) [25] and B. taenagophila [26] populations. This result suggests that some B. camerunensis populations could be as susceptible to S. mansoni as their 'appropriate' intermediate hosts, contrary to the until now accepted idea that B. camerunensis is a poor host with almost no role in the transmission of S. mansoni. Despite this high susceptibility, the significantly high mortality rate observed in infected individuals from Petponoun (Table  1) reflects the adverse effect of parasitic development on the survival of these infected individuals, and suggests that adaptation of the Petponoun population to S. mansoni from Nkolbisson is far from perfect. Since susceptibility to S. mansoni is easier to obtain over generations than resistance [26], and that an increase in temperature can induce the inversion of resistance to S. mansoni [27], the susceptibility of B. camerunensis populations to S. mansoni may increase as a consequence of global warming. Indeed, in Cameroon, the snail Bulinus camerunensis which has been formerly known to be resistant to S. haematobium strain of Barrombi-kotto is now susceptible to that strain [28]. Also, in southern Corsica (France), the susceptibility of Bulinus truncatus to a hybrid schistosome (S. haematobium/S. bovis) has contributed to the re-emergence of schistosomiasis [29][30][31].
The duration of the prepatent period was negatively correlated with the infection rate and was quite different between snail populations (either B. pfeifferi or B. camerunensis), suggesting that a short prepatent period reveals a better compatibility to the parasite and that our populations are genetically different. Interestingly, the cercarial production was two-fold higher in B. camerunensis than in B. pfeifferi, though this observation was done only in one B. camerunensis population among the five investigated. Indeed, Bennike et al. [32] also reported a high and constant cercarial production in susceptible B. camerunensis manzadica individuals, as compared to susceptible B. pfeifferi individuals in the Democratic Republic of Congo (DRC). The mean value of daily cercarial production in the B. camerunensis/S. mansoni combination (1968.38 cercariae) was higher than in the allopatric combination B. pfeifferi from Senegal -S. mansoni from Cameroun (386 cercariae) [33], and some sympatric combinations B. pfeifferi-S. mansoni from Dhofar in (Oman) (464.04 cercariae) [34] and B. pfeifferi-S. mansoni from Senegal (670 cercariae) [35]. Regarding the transmission of the infection, this ability to shed large numbers of cercariae could balance the disadvantage of a small number of susceptible S. mansoni individuals.
The mortality rate is known to be associated with several factors (pathogenicity of the parasite, intrinsic conditions of the snail, experimental conditions) [15]. In our study, the similarity observed in the mortality rates during prepatent and patent periods in both species likely suggests that the parasite is not the only covariate associated with the mortality of snails. The variability of the observed mortality rates between the different populations of both snail species at the end of the experiment would reflect the genetic differences between these populations, and implies a different fitness of the individuals.

Epidemiological implications
The significantly high infection rates observed in B. pfeifferi populations and the long-term survival of S. mansoni-infected individuals (> 8 months postinfection) with continued cercarial shedding confirms the predominant role of this species in the transmission of this intestinal parasite in Cameroon. This adaptation of S. mansoni to B. pfeifferi is likely the result of a good compatibility between the genes of this parasite and its preferential host [22]. Moreover, the low genetic diversity within B. pfeifferi populations resulting from the selection of self-fertilization as their main mating system would promote the transmission, maintenance and fixation of susceptibility genes in these populations. The variability of the susceptibility observed in B. pfeifferi could explain, at least partly, the differences in prevalence observed in Cameroon. Although the number of infected snails is an important indicator of transmission of infection, the cercarial production by these infested individuals is even more important [36] while evaluating the risk of transmission and extension of schistosomiasis. Thus, infected B. camerunensis individuals producing a large number of cercariae might therefore represent a potential risk of increasing S. mansoni transmission in Cameroon. In addition, the fact that B. pfeifferi snails are becoming scarce in many historical foci could exert a selection pressure enabling a viable adaptation of S. mansoni miracidia to B. camerunensis. Therefore, disadvantageous environmental changes for B. pfeifferi could increase the importance of B. camerunensis in the transmission of S. mansoni, and thus the incidence of the infection as a consequence of the evolution of the relationship between Biomphalaria camerunensis and S. mansoni.

Conclusions
This study confirmed the known predominant role of B. pfeifferi in S. mansoni transmission in Cameroon and revealed the ability of B. camerunensis to efficiently transmit and disseminate this parasite. The 'vector' role of this species should therefore be taken into account while assessing schistosomiasis transmission. Special attention and surveillance should be paid to this host since (i) some B. camerunensis populations would increase the risk of dissemination of schistosomiasis in Cameroon, and (ii) this intermediate host could represent an important research model to study the immune defense mechanisms against S. mansoni. In a context where major hydraulic projects are ongoing in southern Cameroon, together with the important migration flow of humans seeking jobs, malacological studies are highly needed to determine the current distribution of this intermediate hosts and identify susceptible populations to S. mansoni throughout the country.

Additional files
Additional file 1: Table S1. Pairwise comparison of the infection rate among snail populations. (DOCX 13 kb) Additional file 2: Table S2