Background

Schistosomiasis, caused by Schistosoma japonicum, remains one of the most prevalent parasitic diseases and effects severe socio-economic and public health losses in China [1, 2]. Oncomelania hupensis is the unique intermediate host of S. japonicum, which plays a critical role in the transmission of Schistosomiasis japonica [1, 3]. The geographical distribution of O. hupensis coincides with the endemic area of S. japonicum [23–25]. To deepen our knowledge on the genetic diversity of the intermediate host snail, we developed a novel multiplex PCR method to screen and analyze the genetic diversity of O. hupensis using microsatellites loci among the four various ecological landscape populations in mainland China.

Methods

Snail sampling

A total of 24 populations of O. hupensis were sampled from four ecological landscape populations in mainland China covering: (1) the region of swamps and lakes in the middle and lower reaches of the Yangtze River, (2) the mountainous region of the Sichuan and Yunnan provinces, (3) the littoral hill part of the Fujian province and (4) the karst landscape of Guangxi autonomous region (Fig. 1, Table 1).

Fig. 1
figure 1

Illustration of geographical location of O. hupensis collection sites

Full size image
Table 1 Location of O. hupensis collection
Full size table

DNA preparation

Ten to 20 O. hupensis samples were randomly chosen from each site, fed for 1 week and identified as infected or non-infected with S. japonicum by observation of cercariae emerging from the snails. Only non-infected snails were used in this study. After removal of the gut and digestive glands from the soft parts of the snails, the 30 mg muscle tissues from the pleopod of a single snail were digested for 3 hours at 56 °C with proteinase K (Amresco Inc. Solon, OH, USA) followed by the standard DNA extraction procedure [26] using mollusc DNA Kit (Omega, USA).

PCR amplification and detection of PCR products

The microsatellite DNA polymorphic loci were selected and evaluated from previous microsatellite loci library [22]. Two rounds of multiplex PCR reaction were developed including four microsatellite loci in each one, which were identified by different lengths and fluorescence peaks of 6-FAM, VIC, NED and PET labeled by (Sigma-aldrich London, UK). Primer sequences and information are summarized in Table 2.

Table 2 Primers of the 8 microsatellite loci in O. hupensis
Full size table

The multiplex PCRs were developed using the Type-it Microsatellite PCR Kit (QiaGen, London, UK) with a 25 μl reaction system, including 2x Type-it Multiplex PCR Master Mix 12.5 μl, 10x primer mix 2 μl including four primers in each mix, template DNA 2 μl with less than 200 ng then add RNase-free water to 25 μl. The reaction conditions for PCR amplification were as follows: 95 °C, 5 min; 95 °C, 30 s, 60 °C, 60 s; 72 °C, 30 s, 30 cycles; 65 °C, 30 min for final extension. 1 μl of the PCR product was mixed with 0.6 μl of ROX and 8.4 μl ultrapure Hi-Di formamide, denatured at 95 °C for 5 min and detected using automatic genetic analyzer (3730XL, ABI, USA).

Analysis of microsatellite diversity

The accurate length of amplified fragments of microsatellite DNA loci were determined using Geneious software(Version 7.0.6) and subsequently exported as an Excel table. The raw data in the table were converted into a recognized format by Arlequin and Genepop using the toolkit of the Excel microsatellite toolkit. The data format which fits for Popgene were acquired by DataTrans 1.0. Various parameters of genetic difference within populations include: number of alleles (Na), number of efficient alleles (Ne), inbreeding coefficient (Fis), expected heterozygosity (He) and observed heterozygosity (Ho) were calculated. The degree of Hardy-Weinberg equilibrium (HWE) and linkage disequilibrium (LD) were tested with Genepop 4.1.10. The frequency of null alleles within every population was calculated in Genepop. The index of genetic variation between populations (Fst), gene flow (Nm) and genetic distance [Fst/ (1-Fst)] were determined using Arlequin [27]. The correlation between genetic distance and geographical distance were tested with Mantel regression. Analysis of molecular variance (AMOVA) was processed through Popgene software, clustering analysis was determined by unweighted pair group method with arithmetic means (UPGMA) and the phylogenetic tree was modified with TreeView [28]. The polymorphism information content (PIC) was calculated according to the formula previously described [28].

Results

Gene scan

From the 24 populations of O. hupensis sampled, 396 specimens were scanned at the genetic level across eight polymorphic loci of microsatellite DNA. The lengths of amplified fragments for a total of 6,196 microsatellite DNA loci were obtained.

Genetic differences within populations

Results obtained from the analysis of the 24 populations of O. hupensis showed that the number of alleles per locus ranged from 29 to 70 with an average of 45.625, and that of effective alleles were 18.5 to 45.8 with an average of 27.4. The GXYZ and HNHS populations had the minimum and maximum average Na values, respectively. The average He within populations ranged from 0.888 to 0.974, and the average Ho ranged from 0.331 to 0.57. The populations with the highest and lowest Ho values were HNHS and GXYZ, respectively. The average PIC for all populations of O. hupensis was 0.940 (Tables 3, 4 and 5).

Table 3 Coefficients of genetic diversity of O. hupensis at different loci (the populations of landscape of swamps and lakes)
Full size table
Table 4 Coefficients of genetic diversity of O. hupensis at different loci (the populations of landscape of mountains)
Full size table
Table 5 Coefficients of genetic diversity of O. hupensis at different loci (the populations of landscape of karst and coastal hills)
Full size table

Significant deviation from Hardy-Weinberg equilibrium (HWE) was observed: 47 out of 192 (24.48 %) possible single exact locus tests (P < 0.01).No significant linkage disequilibrium was found between all pairs of the eight loci examined (P < 0.01), which indicated the independent behaviour of all loci. Analysis with Genepop software showed the possible occurrence of null alleles, which may lead to deviations from HWE and result in exaggerated levels of genetic differentiation [26, 29, 30]. Null alleles may be due to flank sequence variation decreasing primer annealing efficiency, allele drop out or DNA quality [23, 31].

Genetic differences among individuals

Fit and Fst values were 54.95 and 37.62 %, respectively. This suggests that genetic differences mainly exist within populations rather than among those with unbalanced differentiation degrees (Table 6).

Table 6 F-Statistics and gene flow for all loci
Full size table

Mantel’s test of regression showed that the correlation (41.97 %) between geographic distance and genetic distance among populations is positive (R2 = 0.1011, P < 0.05) and genetic distribution of all populations accorded with the Isolation-by-distance Model (Fig. 2, Tables 7 and 8).

Fig. 2
figure 2

Analysis on the relationship between genetic distance and geographic distance

Full size image
Table 7 FST and geographic distance among paired O. hupensis populations of landscape of swamps and lakes
Full size table
Table 8 FST and geographic distance among paired O. hupensis populations of landscape of mountains, karst and Coastal hills
Full size table

Genetic parameters of the four groups from different landscapes (i.e. lakes and marshes, high mountains, Karst and coastal Hills) showed that Na ranged from 2.063 to 11.452, He from 0.465 to 0.852 and Ho from 0.274 to 0.492. The group from the Karst landscape had the lowest value in all three indices, which indicated its low differentiation degree. AMOVA displayed that variations of O. hupensis mainly exists among individuals, which accounted for 60.58 % of total variations, and that of among individuals within populations, among populations within groups and among groups were only 26.60, 8.04 and 4.78 %, respectively (Table 9). This suggests that there is no significant genetic differentiation among groups.

Table 9 Analysis of molecular variance (AMOVA) for the Oncomelania hupensis
Full size table

UPGMA cluster analysis for the 24 O. hupensis populations based genetic distance showed that the populations spread in the landscape of middle and lower reaches of Yangtze River (HBWH, JSYZ, JXNC, HNHS, JXJJ, AHWW, HBJL, JXDC, HNNX, JSYZJZ, ZJJH, AHNG and AHWJ) clustered together first and then gathered with the populations of high mountains (SCMS, SCYA, SCPJ, YNEY, SCLS, YNWS and SCXC), coastal hills (FJFQ and FJFZ) and Karst land form (GXBS and GXYZ) successively (Fig. 3).

Fig. 3
figure 3

UPGMA cluster analysis of 24 O. hupensis populations

Full size image

Discussion

Oncomelania hupensis is the sole intermediate host for transmitting Schistosoma japonicum in mainland China [32], and it is widely distributed in the southern region of the Yangtze River valley. Significant genetic variations have developed in O. hupensis from different geographic populations due to their distribution range, complexity of breeding environment and geographical location.

In this research, The genetic differentiation of four different landscape groups of O. hupensis were studied through eight screened polymorphic microsatellite DNA loci. This information is pertinent because it further improve our understanding on the effect of genetic diversities on the distribution of O. hupensis. This will ultimately help boost our surveillance activities and also strengthen the control of schistosomiasis transmission in China. genetic indices were tested aross eight microsatellite DNA loci. The mean Fis value for the 24 populations examined was 0.272, indicating a deficiency of heterozygotes and frequent inbreeding within populations, which is likely due to the small range of activity of O. hupensis. A total of 47 microsatellite DNA loci deviated from the Hardy Weinberg Equilibrium demonstrating a serious lack of heterozygotes. Possible explanations that may account for this include: activities of migration and inbreeding, drug pressure, gene mutation and null alleles. However, it is currently unclear which one is the dominant factor contributing to this phenomenon [33]. No significant linkage disequilibrium was found between all pairs of the eight loci, clearly showing the independent behaviour of all loci. Null alleles were found at all eight polymorphic loci. This may be due to: 1) mismatching of primer pairs: mutations in microsatellite DNA sites critical for binding with primers leads to abnormal amplification 2) losses of large alleles: the superiority of short alleles restrict amplification of long fragments or 3) differences in DNA quality: unevenness of templates character obstruct amplification in some loci [26, 31, 34]. Null alleles could implicate genetic diversity parameters for populations such as excess of homozygote individuals, reduction of Ho and He and increase of genetic distance and Fis; moreover, it leads to inaccuracy of parent analysis [30–37].

The abundance of the number of heterozygotes and the amount of genetic information in a population is directly proportional to the PIC value [38, 39]. Result shows that PIC was greater than 0.5 at every locus, and the mean value (0.947) from all populations was higher than (0.764) obtained from previous result [23]. This signifies that all the eight loci screened were highly polymorphic.

Furthermore, this study reveals that the average Fst for all loci was 0.376, which means that 37.6 % of genetic variation was among populations and 72.4 % was among individuals within populations. The analysis of AMOVA displayed that genetic variation among individuals (60.58 %) were far higher than that within populations (26.60 %), while among populations and among groups are (8.04 %) and (4.78 %) respectively. This implies that, genetic diversity is strongly derived from among-individuals rather than among-populations. However, the average Fst (0.376) and genetic variation among populations (8.04 %) were higher than values obtained from the previous results (0.048 and 4.8 %) respectively, revealing genetic variation among populations increased along with geographical distance [23]. The Mantel test demonstrated an apparent positive correlation between genetic distance and geographical distance. The genetic structure between geographical populations is embodied with some degree of independence. For example, the geographical distance between the HBWH and JSYZ populations located in the lake region was far, but with low degree of variation. This could possibly be related to the genetic differentiation principally being among individuals within populations rather than among geographic locations for the populations in Lakes and Marshes landscape.

The phylogenetic tree constructed by UPGMA also showed that populations in neighboring geographical locations generally cluster together, which was consistent with the Mantel test results. The cluster sequence of geographical populations showed us that the population from the karst landscape of Guangxi autonomous region maybe the most original one, then the population from the littoral hill part of the Fujian province, the population from the mountainous region of the Sichuan and Yunnan provinces and the population from the region of swamps and lakes in the middle and lower reaches of the Yangtze River, respectively. Regarding as the largest population spread throughout the middle and lower reaches of the Yangtze River [7], the populations from different provinces also crossed cluster, these include, between Hubei and Jiangsu, Hunan and Jiangxi, and Zhejiang and Anhui, which may be as a result of O. hupensis spreading along the river within the large population, or gene drifting for surged water flow in the lakes and marshes landscape [34]. Then this branch clustered with the populations of Sichuan and Yunnan province successively. Furthermore, the major branch clustered with the populations of Fujian and Guangxi province in turn, this agrees with the conclusion of four landscape populations relationships from previous studies using SSR-PCR [40] and DNA sequence markers [7, 41, 42].

Conclusion

This study has shown that the genetic diversity of O. hupensis, an important snail intermediate host of S. japonicum in China mainly originates from among-individuals rather than among-populations. It also reveals that the populations within subspecies have closer consanguinity than between subspecies in the mass, nevertheless, genetic variations exist within subspecies. These findings further provide important information on genetic structure of O. hupensis and strengthen our knowledge about diffusion trend and tracking to the source of Oncomelania in mainland China. Ultimately, these findings will help us develop more effective guidelines for controlling the spread and distribution of Oncomelania and consequently prevent the transmission of Schistosomiasis in China. Our data offers a better understanding of the genetic differentiation of Oncomelania hupensis, enhancing our ability to effective and efficient surveillance of Schistosomiasis.