Spatial variation of life-history traits in the freshwater snail Bulinus truncatus, the intermediate host of human and cattle schistosomiasis, in relation to eld application of niclosamide in northern and central Côte d’Ivoire

Background: Control of intermediate host snails using molluscicides for the control and/or elimination of schistosomiasis is strategy in the line with WHO recommendations. Niclosamide is the main chemical molluscicide recognized by WHO. However, the extent of the application of molluscicide outside the immediate killing of the snail such as the impact on the evolution of life-history traits; in relation to recolonization of treated sites is less or not known. This study aimed to characterize the spatial variation of life-history traits in Bulinus truncatus populations in north and central Côte d’Ivoire, in relation to niclosamide spraying in the eld. From 2016 to 2018, we conducted a trial to control the intermediate host snails for interrupting seasonal transmission urinary schistosomiasis in northern and central Côte d’Ivoire, using niclosamide. The molluscicide was sprayed three times per year in habitats harboring the freshwater snail B. truncatus. Snails were collected before niclosamide application and 2-3 months after the sites were treated, and also from some untreated sites. Families from six natural populations of snails were monitored for several life-history traits, including growth, fecundity and survival, under laboratory conditions, over one generation (G 1 ). Results: Survival rate varied among populations with the highest rates observed in northern populations. No signicant difference was detected between populations before and after treatment, for this trait. Numbers of eggs and eggs per capsule at rst reproduction, fecundity and growth were signicantly lower in treated than untreated groups. Similar nding was observed for populations of before and after treatment. Egg production also varied across populations with the highest values found in northern populations. Within treated group, a signicant difference for survival rate was detected between northern and central populations. Almost all parameters of reproduction and growth varied signicantly, except for the number of egg capsules. Conclusions: Our study shows a spatial variation of life-history traits in B. truncatus snails. Lower values of these traits were observed in populations from recolonized sites after treatment with niclosamide. This trend was much more perceptible in populations from central Côte d’Ivoire. sites following the treatment (treated group). These same parameters were compared using the non-parametric Wilcoxon rank test between populations found before treatment (LBT) and after treatment (LAT) in Linguèbo. Survival rates were compared using the Fisher’s exact test. Variation of traits between populations and within each group was tested using the parametric One-way ANOVA test. The non-parametric Spearman’s rank correlation rho test was performed to assess the relationship between pairs of reproduction and growth parameters in each population. Snail groups were investigated using Principal Component Analysis (PCA). All statistical analyses were performed using the software R version 3.6.1. Analysis; GPS: Global Positioning System; pH: Potential Hydrogen; TDS: Total Dissolved Solids; DAT, SAT and LAT: Djémitédouo, Sambakaha and Linguèbo populations collected 2 to 3 months after niclosamide treatment, respectively; NWT and KWT: Noumousso and Kongobo populations collected in untreated sites with niclosamide, respectively; LBT: Linguèbo population collected before niclosamide application; p = p-value; SD: Standard deviation; Ppop: Population p-value; Ptgr: Treated group p-value; Pugr: Untreated group p-value.


Background
Schistosomiasis is the most important snail-borne disease, endemic in the tropical countries, with more than 90% of cases occurring in the WHO African Region and the highest proportion in sub-Saharan African [1][2][3]. Despite increasing efforts to control this parasitic disease, it still remains a public health concern in endemic countries, affecting approximately 260 million people [3] and leads to serious economic losses [4].
Transmission occurs in freshwater bodies, used intensively by both humans and livestock, where intermediate host snails and their associated schistosomes are present [5,6]. Hence, the distribution of genera and species of intermediate host snails compatible with the schistosome parasite in uence the distribution of the disease. Schistosome transmission cannot occur without compatible snails [7].
The World Health Assembly (WHA) recommended local schistosomiasis elimination "where feasible" to member countries such as Côte d'Ivoire, through resolution WHA65.21 [8]. Control intermediate host snails using molluscicides for the control and/or elimination of schistosomiasis is strategy in the line with WHO recommendations [9]. Niclosamide (Bayluscide®) is the main molluscicide registered and recommended by WHO [10]. Of note, schistosomiasis control projects conducted in Morocco, in China and in Egypt showed that snail control using this molluscicide could be an e cient approach for reducing or interrupting the transmission of schistosomes [11][12][13]. Moreover, recent systematic reviews and metaanalysis on the impact of the application of chemical molluscicides [14,15] demonstrate the importance of integrating snail control using niclosamide towards schistosomiasis elimination campaigns in endemic areas [7].
Host snail populations of different environments are known to vary in biological features at regional and local geographical scales [16][17][18]. E cient control of these snail populations therefore requires deep knowledge of their life cycle [19]. For that purpose, it is necessary to assess the life-history traits of individuals [20,21]. These traits include growth, survival and reproduction, with age at rst reproduction being one of these organizer parameters [22].
In this context, freshwater gastropods of the Hygrophila group offer excellent biological model. The snail Bulinus truncatus is one of these gastropods. In Africa and the Middle East, this snail acts as intermediate host for both Schistosoma haematobium and S. bovis, the agents of human and cattle schistosomiasis, respectively [23,24]. B. truncatus is a hermaphroditic sel ng species, with sel ng rates as high as 80% in some populations [25]. Life-history traits can vary within and among populations under the in uence of genetic and environmental factors [26][27][28].
Previous studies reported that the niclosamide is highly active at all stages of the freshwater snail life cycle; killing them within a few hours at low concentrations [7,10,14,15]. However, the extent of the application of molluscicide outside the immediate killing of the snail such as the impact on the evolution of their life-history traits is less or not known. Nonetheless, pesticides are known to impact several biological traits of organisms [29,30]. For instance, it has shown that chlorpyrifos and profenophos reduce survival rate, growth rate and egg production, and cause severe damages in the hermaphroditic gland cells of B. truncatus snails [31]. Therefore, niclosamide may in uence the evolution of biological traits. It is thus important to know how life-history traits vary in B. truncatus populations in relation to habitat treatment with the molluscicide. We monitored several traits in six natural populations of B. truncatus, over one generation, under laboratory conditions, using a family design, in order to characterize the spatial variation of life-history traits in B. truncatus populations, in relation to niclosamide application in the eld.
This study is part of a Schistosomiasis Consortium for Operational Research and Evaluation (SCORE) large-scale project [32] that aims to interrupt seasonal transmission of S. haematobium in northern and central regions of Côte d'Ivoire by combining chemical snail control using niclosamide and preventive chemotherapy with praziquantel [33].

Results
Survival rate of the populations There was a variation for survival rate among snail populations. Survival rates reached 100% within the rst three weeks, and decreased until week 12 ranging from 91.1% in DAT population to 48.9% in LBT population. The survival rate was higher in northern populations compared to the central populations from week 13 to week 18 (Fig. 1a). However, no signi cant difference was observed between untreated and treated groups (Fisher's exact test, p = 0.674). Within each group, no signi cant variation was detected among populations for untreated group (Fisher's exact test, p = 0.249), whereas for treated group the survival rate signi cantly varied among populations (Fisher's exact test, p < 0.001). There was no signi cant difference for survival rate between LBT and LAT populations (Fisher's exact test, p = 0.079).

Reproduction And Growth Of The Populations
Most of the parameters (8 over 13) of reproduction and growth signi cantly varied when comparing treated and untreated groups (Student's t-test, p < 0.05) ( Table 1). Similar variation was observed before and after treatment for populations of LAT and LBT (Wilcoxon rank test, p < 0.05) ( Table 2). Tables 1 and  2 show that the numbers of eggs and eggs per capsule at rst reproduction, the mean fecundity as well as the growth were signi cantly lower in the treated populations compared to the untreated ones. Variation of the reproduction and the growth parameters between populations and within each population group is given in supplementary le (Additional le 1: Table S1). There was a signi cant variation for all reproduction and growth parameters among populations (One-way ANOVA, p < 0.05). Within treated group, variation was detected for several parameters, except for the number of egg capsules (One-way ANOVA, p = 0.901). However, most populations in the North such as DAT and NWT laid eggs earlier than those of the Centre namely LAT and LBT. The dynamic of egg-laying also varied across the populations (Fig. 1b). An increase of egg production up to a peak at 42th days was observed in some populations as NWT, SAT, KWT and LAT. The rst peaks were observed in LBT and DAT populations at 28th and 35th days, respectively. At the end of 70 days, the number of eggs laid were higher in northern populations than in central populations; the highest value being found in NWT population and the lowest in LAT population.
Correlations between reproduction and growth parameters in each population are reported in supplementary le (Additional le 1: Table S2). A negative correlation was detected between age at rst reproduction and fecundity parameters in populations from untreated and treated sites, except for the LAT population. However, growth was positively correlated with number of eggs per capsule in the central populations. Principal component analysis (PCA) based on the most potential reproduction and growth traits of the six populations showed that the rst two components accounted for 70.12% of the total variation observed in the PCA (Additional le 1: Fig. S1). Age at rst reproduction was negatively correlated with fecundity and growth. The cluster analysis revealed three groups. The group I included individuals characterized by a slow growth, a low fecundity (numbers of egg capsules, eggs and eggs per capsule) and a late reproduction. The group II consisted of individuals with earlier reproduction, smaller size and larger number of egg capsules. This group was composed of individuals from Djemitedouo only.
The third group (group III) was characterized as follows: a larger size at rst reproduction, an important number of eggs and eggs per capsule. This group included individuals from Noumousso, Sambakaha and Kongobo.

Discussion
To our knowledge, this is the rst study assessing niclosamide impact on the evolution of life-history traits in B. truncatus populations following eld treatments.  [38]. Together with our study, these results indicate that Bulinus sp. is susceptible to environmental or seasonal temperature uctuations, and therefore to anthropogenic changes that may occur in their habitats. B. truncatus populations of the northern displayed a higher survival rate than those of the central from week 13 to week 18. This difference between areas might be explained by frequent uctuations in snail size within and between habitats resulting from a variability in environmental and genetic factors [16,23,34]. Indeed, these factors including temperature, sunlight, bottlenecks and founding events, more important in the North than in the Centre, might have made northern snails more robust and provide them with a longer survival time to the chemical treatments. Survival rate was signi cantly higher in northern population than in central population within treated group, whereas no difference was observed for those of the untreated group. These observations might indicate that treatments are more effective on central populations which have low survival rates.
However, no difference between population of before (LBT) and after (LAT) treatment was detected at rst generation, for this trait.
Variability in most of reproduction and growth parameters was observed between untreated and treated groups. Variation in several parameters was also detected between pre-and post-treatment populations. Interestingly, fecundity at rst reproduction, fecundity, growth of both treated group and LAT population were signi cantly lower than those of the untreated group and LBT population. The difference between these traits could be due to the niclosamide treatment, as a result of an elimination of most fertile and larger snails replaced by least fertile and smallest individuals. This might be considered as a good outcome of the mollusciciding, as larger host snails are known to produce more eggs per capsule and more cercariae (when infected) [39][40][41]. In addition, a high variability of reproduction and growth parameters was found within treated group compared to untreated one. These traits could be adaptive ones, allowing snails to recolonize treated sites [42]. One should pay more attention to these life-history traits. Most populations from the North laid eggs earlier than those from the Centre. Moreover, the dynamic of egg-laying varied across the populations, with numbers of eggs laid higher in northern populations than in central populations. These observations might indicate that egg production of northern snails recolonizing treated sites was less in uenced by niclosamide application.
Negative correlations were found between age at rst reproduction with mean numbers of egg capsules, egg and eggs per capsules. Hence, snails which reproduced earlier, laid more egg capsules, eggs and eggs per capsule. This might be considered as an advantageous feature in the recolonization process of habitats following treatments. We can infer that the niclosamide applications had no perceptible effect on age at reproduction, as no signi cant difference was detected for this trait between populations before and after niclosamide application.
On the other hand, principal component and cluster analyses revealed an in uence of the most varying traits among B. truncatus populations. Reproduction and growth parameters characterizing groups I and III could be due to population mixing as a result of snail migration [43][44][45][46]. Indeed, snail populations of these groups share the same watershed, that of the watershed of Bandama river.

Conclusions
Niclosamide application impact on B. truncatus snails appears satisfactory, on the whole. No major effect on survival, reproduction and growth was detected in snails recolonizing treated habitats. However, our study revealed an ability of snails to recolonize treated sites, and this was much more pronounced in central populations. Further investigations should be conducted over several snail generations for accurate assessment of niclosamide impact on life-history traits.

Study area
This study was carried out in the North (Bounkani and Tchologo regions) and in the Centre (Gbêkê region) of Côte d'Ivoire (Fig. 2). These areas are known to harbour B. truncatus [47][48][49] which is predominantly involved in the transmission of S. haematobium, S. bovis and the S. haematobium x S. bovis hybrid [50]. In those regions, the snail is mainly observed in man-made dam, which are subject to a marked seasonality characterized by a long dry season and a short rainy season [47]. In the northern areas, one control village namely Noumousso (10°06´25.67´´N latitude, 05°08´30.81´´W longitude) and two test villages Djémitédouo (09°17´05.87´´N, 02°58´08.81´´W) and Sambakaha (09°24´10.72´´N, 05°06´21.24´W ) were randomly chosen. In the central part, Kongobo (07°45´50.33´´N, 05°28´33.99´´W) and Linguèbo (07°30´16.36´´N, 05°42´22.60´´W) were selected as control and test villages, respectively.

Snail Sampling
In each study village, human-water contact sites were identi ed and georeferenced using a hand-held global positioning system (GPS; Garmin Sery GPS MAP 62, Olathe, KS, USA) device [33]. Snail surveys were carried out at each human-water contact in March 2017, June 2017, November 2017, March 2018 and June 2018. Before each snail sampling, physico-chemical parameters of the water were measured in situ using a pocket multi-parameter tester (HANNA® Instruments HI 98129 Combo, Woonsocket, RI, USA). The parameters measured were temperature, pH, conductivity and total dissolved solids (TDS). Snails were collected by the same two experienced collectors, using a long-handled scoop and/or forceps for a period of 15 min [51,52]. Shortly after each sampling, snails were morphologically identi ed on the basis of standard identi cation keys to the species level [23,53] and enumerated. Then, an appropriate volume of niclosamide solution (at concentration of 10 g/l) was applied along the bank of the test sites if B. truncatus was found (Fig. 3a).
The populations studied were composed of individuals from recolonized sites 2 to 3 months after treatment of Djemitedouo (DAT), Sambakaha (SAT) and Linguèbo (LAT); from untreated sites with niclosamide of Noumousso (NWT) and Kongobo (KWT); and individuals collected before treatment in Linguèbo site (LBT). The B. truncatus populations collected (G 0 generation) were transferred to the laboratory, placed between two layers of moistened cotton in a labelled petri dish [54] (Fig. 3b).

Experimental Design
According to the experimental protocol of the current study (Fig. 4), snails were reared following of a common garden approach, where individuals from different sites were raised together in the same controlled (laboratory) conditions [55]. Thus, the rearing room and water (30 mL of mineral water) were maintained at a temperature of 22 to 25°C and under a photoperiod of 12 L/12 D. During the rearing, G 0 snails were fed ad libitum with granules for aquarium sh and juveniles with boiled lettuce. Water was changed and food given twice a week in rearing boxes.
In the laboratory, G 0 snails from the same site were put together in a 1.5 L transparent plastic box for acclimatization. The following day, 20 G 0 individuals of each population were randomly chosen. Each G 0 snail was isolated in a 40 mL transparent plastic box lled with 30 mL of mineral water for G 1 offspring production. Two to three days later, egg-layings were observed synchronously in the boxes. After hatching, 7 to 10 days after the rst egg-laying, the offsprings were maintained with the parents in rearing boxes for approximately two weeks. Then, two to three G 1 offsprings including two in 15 G 0 boxes and three in ve boxes, aged 10 to 14 days were chosen at random in each box to reach a sample of 45 G 1 snails for each population. These G 1 individuals were isolated and reared in the same rearing conditions as their parents.

Assessment Of Life-history Traits
Life-history traits of G 1 individuals were monitored over 28 weeks, including survival, fecundity, growth and hatching rate. Size (shell height and width) was measured once a week using a graph paper under binocular magnifying glass. This allowed the estimation of the size at rst reproduction, as well as the growth. Individual survival was monitored three times a week by observation with the naked eye.
Fecundity of each individual snail was evaluated while checking egg-layings twice a week: number of egg capsules, eggs and eggs per capsule. The age and fecundity at rst reproduction were recorded. The numbers of egg capsules as well as eggs laid over 10 weeks were estimated using a binocular magnifying glass and recorded. Fecundity at rst reproduction was de ned as the number of egg capsules, eggs and eggs per capsule over seven days after the rst egg-laying day. The hatching of eggs was also monitored every two days over two weeks. At the end of each week, the egg capsules laid in each box were collected using a plastic spatula and put in a petri dish containing mineral water. Egg capsules were checked every two days over two weeks and the number of hatched individuals was recorded in order to calculate the hatching rate per week of each population.

Statistical analysis
A normality test was rst performed to assess the distribution and variance homogeneity of each quantitative parameter using the Kolmogorov-Smirnov test. Then, the parametric Student t-test was used to compare the reproduction and growth parameters between population groups from untreated sites with niclosamide (untreated group) and recolonized sites following the treatment (treated group). These same parameters were compared using the non-parametric Wilcoxon rank test between populations found before treatment (LBT) and after treatment (LAT) in Linguèbo. Survival rates were compared using the Fisher's exact test. Variation of traits between populations and within each group was tested using the parametric One-way ANOVA test. The non-parametric Spearman's rank correlation rho test was performed to assess the relationship between pairs of reproduction and growth parameters in each population. Snail groups were investigated using Principal Component Analysis (PCA). All statistical analyses were performed using the software R version 3.6.1. . The purpose of the study and procedures for eld sample collection were explained to village authorities and their verbal agreement was obtained before conducting eld surveys.

Consent for publication
Not applicable.

Availability of data and materials
The data used and / or analyzed during this study are available from the corresponding author on reasonable request. CKK, YNTTB and EKN conceived and designed the study. YNTTB and EKN supervised the study. CKK, YNTTB, NRD, MO, DS, AK and AKK followed up eld surveys and collected eld data. CKK collected laboratory data. YNTTB and SCG contributed to collecting laboratory data. CKK and YNTTB analyzed and interpreted the data. CKK wrote the rst draft. YNTTB, JTC, RKA, NRD, MAE and EKN revised the manuscript. All authors read and approved the nal manuscript at submission.