Comparative study of excretory–secretory proteins released by Schistosoma mansoni-resistant, susceptible and naïve Biomphalaria glabrata

Background Schistosomiasis is a harmful neglected tropical disease caused by infection with Schistosoma spp., such as Schistosoma mansoni. Schistosoma must transition within a molluscan host to survive. Chemical analyses of schistosome-molluscan interactions indicate that host identification involves chemosensation, including naïve host preference. Proteomic technique advances enable sophisticated comparative analyses between infected and naïve snail host proteins. This study aimed to compare resistant, susceptible and naïve Biomphalaria glabrata snail-conditioned water (SCW) to identify potential attractants and deterrents. Methods Behavioural bioassays were performed on S. mansoni miracidia to compare the effects of susceptible, F1 resistant and naïve B. glabrata SCW. The F1 resistant and susceptible B. glabrata SCW excretory–secretory proteins (ESPs) were fractionated using SDS-PAGE, identified with LC-MS/MS and compared to naïve snail ESPs. Protein-protein interaction (PPI) analyses based on published studies (including experiments, co-expression, text-mining and gene fusion) identified S. mansoni and B. glabrata protein interaction. Data are available via ProteomeXchange with identifier PXD015129. Results A total of 291, 410 and 597 ESPs were detected in the susceptible, F1 resistant and naïve SCW, respectively. Less overlap in ESPs was identified between susceptible and naïve snails than F1 resistant and naïve snails. F1 resistant B. glabrata ESPs were predominately associated with anti-pathogen activity and detoxification, such as leukocyte elastase and peroxiredoxin. Susceptible B. glabrata several proteins correlated with immunity and anti-inflammation, such as glutathione S-transferase and zinc metalloproteinase, and S. mansoni sporocyst presence. PPI analyses found that uncharacterised S. mansoni protein Smp_142140.1 potentially interacts with numerous B. glabrata proteins. Conclusions This study identified ESPs released by F1 resistant, susceptible and naïve B. glabrata to explain S. mansoni miracidia interplay. Susceptible B. glabrata ESPs shed light on potential S. mansoni miracidia deterrents. Further targeted research on specific ESPs identified in this study could help inhibit B. glabrata and S. mansoni interactions and stop human schistosomiasis.


Background
Human schistosomiasis is caused by infection from digenetic trematodes of the genus Schistosoma and is one of the greatest threats to public health in the world [1,2]. The disease is currently endemic in 76 different countries and over 800 million people are at risk of infection [3,4]. Estimates claim that over 200,000 people die every year from the immunosuppressive and carcinogenic effects of the infection [5][6][7]. It decreases resistance to other harmful diseases including hepatitis B, HIV and malaria [8][9][10]. It also increases rates of seizures, infertility and anaemia [11,12].
The chemotherapeutic drug praziquantel is the most effective current method of dealing with human schistosomiasis [13]. While it has a low cost of production and few side effects less than 30% of those in need of preventative chemotherapy had access to it in 2015 [14,15]. Additionally, its decreased efficacy against immature schistosomes and reinfections necessitates the innovation of alternative methods for mitigating the spread of schistosomiasis [16,17]. To meet the World Health Organization's goal of eradicating the disease by 2025 the disruption of the parasites' lifecycles in their infective stages is an approach currently being investigated [18].
As a member of the class Trematoda, Schistosoma must infect an intermediate molluscan host [19]. Molluscan hosts are infected by Schistosoma miracidia, non-feeding infective stages that hatch from eggs released from mammalian host defecation within minutes of entering fresh water [7,20,21]. Following penetration, the miracidia transform into primary (or mother) sporocysts which give rise asexually to a second generation of secondary (or daughter) sporocysts [22]. Secondary sporocysts may produce thousands of cercariae, resulting in infection from one miracidium potentially resulting in the release of more than 100,000 cercariae [23,24]. Schistosoma cercariae are (a non-feeding infective stage) penetrate and reproduce within a host of the class Mammalia [20]. Among the species of Schistosoma which may infect humans, the most harmful are Schistosoma mansoni, Schistosoma japonicum and Schistosoma haematobium, which together comprise over 95% of human infections [25].
Miracidia can only survive in the environment for an average of 12 hours [21]. Analyses of the interactions between S. mansoni and one of its molluscan hosts, Biomphalaria glabrata, indicate that the miracidia locate the host through chemosensory signals [7]. The interactions between miracidia and their hosts' immune recognition receptors render the host susceptible to infection [19]. Haemocytes kill the schistosomes through hydrogen peroxide exposure upon identification [26]. Successful infections from schistosomes lead to chemical castration and decreased average lifespans in molluscan hosts [16,27].
Various haemocyte-related enzymes and receptors are upregulated within 5 hours after infection in more resistant strains (such as BS-90) while responses take several hours or days longer in susceptible strains of B. glabrata (such as NMRI) [28,29]. At this stage there is insufficient information to determine the precise relationship between haemocyte behaviour and snail excretory-secretory proteins (ESPs). Furthermore, it is unclear how specific ESPs are to certain species or strains. However, the availability of genomes for the S. mansoni and B. glabrata allow for in-depth genomic and proteomic studies [16,30]. Recently, studies have been conducted to identify the ESPs released by naïve B. glabrata [16,31]. There remain some gaps regarding our understanding of the differences and significance of ESPs released by susceptible and resistant B. glabrata.
In this study we performed behavioural bioassays in conjunction with video analyses using SCW derived from naïve, susceptible and F1 resistant B. glabrata (NMRI strain). ESPs identified by LC-MS/MS analysis were further assessed for their potential roles as attractants and defensive proteins in parasite infection. A protein-protein interaction (PPI) analysis was used to determine potential interactions between identified S. mansoni proteins and its entire proteome, or B. glabrata proteome, respectively. The findings of this study help elucidate the ESPs related to resistance mechanisms of B. glabrata. This information may facilitate the development of synthetic attractants or deterrents of miracidia, mitigating the spread of schistosomiasis.

Biomphalaria glabrata maintenance conditions
Biomphalaria glabrata snails of the NMRI strain (which reliably release cercariae in up to 95% of infection cases) [32], were maintained in an aerated tank of calcium carbonate conditioned-water (pH-neutral) at 27 °C in a 12 h alternating cycle of light and darkness. Their diet consisted of algae tablets and lettuce. Naïve B. glabrata were defined as those with no prior exposure to S. mansoni miracidia. The resistant snails were defined as the F1 progeny of B. glabrata that were exposed to S. mansoni miracidia (stock). These offspring were expected to maintain the resistance of their parent and therefore have a relatively high probability of also being resistant [33]. Susceptible snails were those previously exposed to S. mansoni miracidia and rendered infertile, a key indicator of reproductive dysfunction due to infection.

Snail conditioned water collection and semi-purification of biomolecules
The overall experimental procedure to map and annotate ESPs released by naïve, susceptible (14 days post-infection) and F1 resistant B. glabrata is outlined in Fig. 1. At QIMR Berghofer Medical Research Institute, B. glabrata snails (50 each) were washed four times with freshly prepared carbonate conditioned Milli Q water to remove any contaminants from the tank and separated into two 200 ml beakers, each containing 25 snails (Fig. 1). Snails were incubated in 20 ml of pH-neutral spring water at room temperature for 2 h. Snails were removed and returned to the aquarium, 20 ml of methanol was added to the water samples and mixed thoroughly. The mixture was filtered through a 0.45 µm Durapore PVDF filter (Bedford, MA, USA) to remove contaminants. Filtered samples were immediately frozen on dry ice until lyophilisation using a Savant SpeedVac Concentrator (Thermo Fisher Scientific, MA, USA).

Schistosoma mansoni miracidia isolation and behavioural bioassay
Schistosoma mansoni-infected Swiss mice were euthanised with CO 2 gas and their livers were perfused with chilled phosphate-buffered saline (PBS) to collect the eggs of S. mansoni. Two infected mouse livers were sliced with scalpel blades and blended to a smooth consistency in 50 ml PBS. The mixture was centrifuged (2000×g at 4 °C for 10 s), the supernatant was removed, and pellet resuspended in 50 ml chilled PBS. This step was repeated three times until the supernatant was transparent. The mixture was incubated in a measuring cylinder surrounded by black tape in pH-neutral water under a light for 2 h at room temperature. The top layer of the water was collected, and the average miracidia were counted under a microscope. The miracidia were concentrated through centrifuging the water at 5000×g for 15 min at 22 °C and the supernatant was removed. The method of behavioural bioassay has been described in detail Fig. 1 Overall workflow for B. glabrata SCW isolation, protein extraction and identification. SCW proteins were extensively fractionated by 1D SDS-PAGE followed by identification with high-accuracy uHPLC QTOF MS/MS. Biological triplicates were performed for each experimental condition elsewhere [31]. Briefly, miracidia water aliquots in 200 µl volumes were placed on a petri dish and monitored using an Olympus-CKX41 microscope (Olympus) equipped with an Olympus DPI Digital Microscope Camera DP22 (25 frames per second at 2.8-megapixel image quality). Miracidia behaviour was recorded and monitored for one minute, followed by one minute after the addition of 2 µl of SCW. This process was conducted nine times using naïve, susceptible (from B. glabrata exposed to miracidia 2 weeks prior) and F1 resistant B. glabrata SCW and one negative control (pH-neutral water used for incubating miracidia).
The susceptible, F1 resistant and control sample videos were analysed statistically using the method described previously [31]. Videos were split into pre-SCW and post-SCW segments and imported into ImageJ (fiji-win64). The miracidia were identified when they were within the field of view (FOV) and their velocity was calculated in pixel s −1 using the rolling mean subtraction method [34]. Employing the plugin for FIJI software [35], known as TrackMate [36], miracidia location was tracked in each frame along an x-y axis and the trajectories were interpolated. The MTrackJ plugin [37] was used to determine the average velocity, tortuosity (the track length to maximum displacement ratio), duration of presence and sum of tracks per min of miracidia presence for the pre-SCW and post-SCW segments. Due to constant overlapping of miracidia pathways in naïve SCW videos, heatmaps (showing the distribution density of miracidia) were constructed to compare the effects of susceptible, resistant and naïve SCW. The protocol for heatmap generation has been described elsewhere [34].
A two-way ANOVA test was used to calculate P-values and evaluate the significance of behavioural modifications in response to pH-neutral water, susceptible and F1 resistant SCW. The behavioural changes in swimming speed (velocity), tortuosity, number of miracidia entering the FOV and the time of duration of miracidia staying in the FOV within the defined duration were compared. A change was considered significant if the P-value < 0.05.

SDS-PAGE, Coomassie staining and in-gel trypsin digestion
The protein concentrations of the lyophilised samples were measured by Nanodrop 2000c (Thermo Fisher Scientific, MA, USA) before being resuspended in 100 µl of 6 M urea and mixed with 100 µl of sample buffer [95% of 2× Laemmli buffer (BioRad Laboratories, Hercules, CA, USA) and 5% of β-mercaptoethanol]. The mixture was heated at 95 °C for 5 min and loaded onto a preconditioned 4-15% Mini-PROTEAN ® TGX ™ Precast Protein Gels (Bio-RAD). The gel was run in a Mini-PROTEAN ® Dodeca Cell for 60 min under 200 V. The gel was stained using Coomassie brilliant blue G-250 for 1 h, rinsed in water for 30 min and placed in the fridge at 4 °C overnight. The image of the gel was scanned with the wavelengths of 700 nm for 40 min using an Odyssey CLx (Li-Cor) and visualised with Image Studio 4.0 (Li-Cor). The naïve B. glabrata gel was collected in an earlier study [16].
The entire gel lanes were excised into pieces using a scalpel blade and subjected to in-gel trypsin digestion as described elsewhere [38]. Briefly, the gel pieces were transferred to Eppendorf tubes and repeatedly washed in 500 µl of 50 mM NH 4 HCO 3 , incubated for 5 min and removed. A 500 µl volume of 50 mM NH 4 HCO 3 in 30% acetonitrile was added to remove Coomassie stain. Pieces were incubated in a sonicating water bath for 15 min and centrifuged using pulse centrifugation before the excess liquid was extracted. A 200 µl volume of acetonitrile was added to each tube to shrink the gel pieces, incubated for 15 min and spun down using pulse centrifugation before the liquid was removed. Samples were vacuumcentrifuged for 10 min. Pieces were swelled with 50 µl of 10 mM dithiothreitol in 100 mM NH 4 HCO 3 before being incubated at 56 °C for 1 h. Samples were spun down using pulse centrifugation after cooling to room temperature and the excess liquid was extracted. A volume of 200 µl of acetonitrile was added to each sample, incubated for 15 min and pulse-centrifuged before the liquid was removed. A 50 µl volume of 55 mM iodoacetamide in 100 mM NH 4 HCO 3 was added and the pieces were incubated in the dark for 45 min. The solution was removed, 100 µl of 5 mM NH 4 HCO 3 was added and incubated for 10 min before being removed. The pieces were shrunk using 200 µl of acetonitrile and incubated from 15 min before pulse centrifugation. The liquid was removed, and the gel pieces were vacuum centrifuged for 10 min. A 10 µl aliquot of 10 ng/µl trypsin in 5 mM NH 4 HCO 3 was added to each tube and incubated overnight (~16 h) at 37 °C.
The tryptic peptides were extracted from gel pieces by sonication in a water bath for 15 min after adding 20 µl of 50% acetonitrile containing 1% formic acid. The samples were spun down by pulse centrifugation and the excess liquid was transferred into the final sample tube. The gel pieces were shrunk by 50 µl of 100% acetonitrile and the liquid was collected into the corresponding tubes after 15 min sonicating in a water bath. The volume of the liquid in each tube was reduced to about 1 µl with Speed-Vac and transferred into the final sample tube. This was reconstituted in 5 µl of 30% acetonitrile, 0.1% formic acid and stored at − 20 °C for LC-MS/MS analysis (see Fig. 1).

uHPLC tandem QTOF MS/MS analyses
Tryptic peptides were resuspended in 25 μl of 1% formic acid in MilliQ water and analysed by LC-MS/MS attached to an ExionLC liquid chromatography system (AB SCIEX, Concord, Canada) and a QTOF X500R mass spectrometer (AB SCIEX, Concord, Canada) equipped with an electrospray ion source. A 20 µl sample of each of the B. glabrata fractions was injected into a 100 mm × 1.7 μm Aeris PEPTIDE XB-C18 100 uHPLC column (Phenomenex, Sydney, Australia) equipped with a Secu-rityGuard column for mass spectrometry analysis. Linear gradients of 5-35% solvent B over a 10-min period at a flow rate of 400 µl/min, followed by a gradient from 35% to 80% solvent B over 2 min and 80% to 95% solvent B in 1 min were used for peptide elution. Solvent B remained at 95% for a 1 min period for washing the column after which it was decreased to 5% for equilibration prior to the injection of the subsequent sample. Solvent A consisted of 0.1% formic acid in MilliQ water while solvent B contained 0.1% formic acid in 100% acetonitrile. The ion spray voltage was set to 5500 V, the declustering potential was set to 100 V, the curtain gas flow was set at 30, ion source gas 1 was set at 40, the ion source gas 2 was set at 50 and spray temperature was set at 450 °C. The mass spectrometer acquired the mass spectral data in an Information Dependant Acquisition, IDA mode. Full scan TOFMS data was acquired over the mass range of 350-1400 and for product ion ms/ms of 50-1800. Ions observed in the TOF-MS scan exceeding a threshold of 100 cps and a charge state of + 2 to + 5 were set to trigger the acquisition of product ion. The data were acquired and processed using SCIEX OS software (AB SCIEX, Concord, Canada).

Protein identification
LC-MS/MS data was imported to PEAKS studio (Bioinformatics Solutions Inc., Waterloo, ON, Canada, version 7.0) with the assistance of MSConvert module of Prote-oWizard (3.0.1) [39]. The ESPs of naïve B. glabrata have been analysed using a similar protocol with a previous version the genome (Ver 1.2) [16]. For the present study, the proteomic data were reanalysed with the most upto-date database (BglaB1.6) (see Additional file 1: Database S1) to provide a better comparison between naïve B. glabrata ESPs and those released by the susceptible and resistant snails (https ://www.vecto rbase .org/organ isms/ biomp halar ia-glabr ata) [16]. Meanwhile, MS/MS spectra of proteins extracted from susceptible B. glabrata conditioned water were analysed with reference to the S. mansoni protein database (https ://paras ite.wormb ase.org/ Schis tosom a_manso ni_prjea 36577 /Info/Index ). De novo sequencing of peptides, database search and characterising specific PTMs were used to analyse the raw data; false discovery rate (FDR) was set to ≤ 1%, and [− 10*log(p)] was calculated accordingly where p was the probability that an observed match was a random event. The PEAKS used the following parameters: (i) precursor ion mass tolerance, 0.1 Da; (ii) fragment ion mass tolerance, 0.1 Da (the error tolerance); (iii) tryptic enzyme specificity with two missed cleavages allowed; (iv) monoisotopic precursor mass and fragment ion mass; (v) a fixed modification of cysteine carbamidomethylation; and (vi) variable modifications including lysine acetylation, deamidation on asparagine and glutamine, oxidation of methionine and conversion of glutamic acid and glutamine to pyroglutamate. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE [40] partner repository with the dataset identifier PXD015129.

Prediction of secreted proteins, gene ontology and KEGG pathway analysis
Identified proteins were subjected to BLASTp using non-redundant protein sequences of NCBI. Protein N-terminal signal sequences were predicted using Sign-laP 4.1 [41] and Predisi [42], with the transmembrane domains predicted by TMHMM [43]. For SignalP predictions, positive identifications were made when both neural network and hidden Markov model algorithms gave coincident estimations; D-cut-off values were set to 0.34 (to increase sensitivity) for both SignalP-noTM and TM networks. Herein, a protein was designated as secreted only when it met the criteria of both SignalP and Predisi and did not have a transmembrane domain predicted by TMHMM.
BLAST results were combined and imported to BLAST2GO [44] (version 5.1), to perform gene ontology (GO) and KEGG pathway analysis. Fisher's extract test was carried out to evaluate the enrichment of GO terms in ESPs of susceptible and resistant snails with reference to entire proteome of B. glabrata [45]. Susceptible B. glabrata SCW was also referenced with respect to the S. mansoni proteome. The significant GO terms with P < 0.01 were considered as over-represented, and FDRs were calculated from p-values using the Benjamini-Hochberg procedure [46].

Protein-protein interaction (PPI) network
We investigated the PPI maps following a similar procedure reported elsewhere [47]. Both domain-domain interaction and gene ontology annotations were used. Briefly, HMMER [48] was first used to annotate all the known protein domains based on the Pfam database (Release 32.0) [49], then the high confidence domaindomain interactions from the DOMINE database [50] was exported based on annotations. Proteins with at least three domain-domain interaction supports were included in the final network. The PPI between annotated S. mansoni proteins identified in susceptible snail SCW and its entire proteome was further validated with STRING [51]. STRING integrates protein-protein interactions from multiple resources, including direct (physical) as well as indirect (functional) associations. All resources were selected to generate the network and 'confidence' was used as the meaning of network edges. The first shell was set to show no more than 20 interactors, while the second shell was not considered in this study. Proteins without any interaction with other proteins were excluded from the network of this study. Topological analyses were performed to explore the potential functions in our constructed network using the Network Analyzer plugin in Cytoscape 3.7.1 [52]. The final network visualization was performed using Cytoscape [52].

Schistosoma mansoni miracidia behavioural assays
We have previously shown that SCW of naïve B. glabrata stimulates significant behavioural changes in S. mansoni miracidia, including the elevation of swimming speed (velocity), tortuosity, number of miracidia entering the FOV and the time of duration of miracidia staying in the FOV within a defined time period [31]. In this study, we further quantified the changes in S. mansoni miracidia behaviour in response to pH-neutral water, susceptible and resistant B. glabrata SCW using behavioural bioassays. Figure 2 provides comparative data for the behavioural modifications monitored in the bioassay, with the statistical analysis results shown in Additional file 2: Table S1. Figure 2a displays there are more abundant red and yellow regions in pre-addition heatmaps, indicating relatively slower moving miracidia. The postaddition heatmap of naïve SCW depicts a significantly fewer linear motions and higher proportion of soft blue lines which suggest more tracks in the FOV and quicker circular movements. The post-addition heatmaps of susceptible and F1 resistant SCW only show quicker circular movements, but the changes in the amount of blue lines seems insignificant. The velocity of movement (swimming) of miracidia in three treatments was assessed (Fig. 2b), where no significant difference could be determined between the treatments (i.e. pH-neutral water versus susceptible versus resistant), or with one treatment (i.e. pre-versus post-addition within 1 min time frame) (Additional file 2: Table S1). In terms of tortuosity, the behavioural change was also determined to be insignificant within one treatment or between treatments, although the mean value post-treatment of susceptible SCW increased (Fig. 2c). The number of miracidia entering the FOV significantly increased within 1 min postaddition of susceptible SCW, but not after addition of pH-neutral water or resistant SCW, which also showed a higher average number compared to that observed preaddition (Fig. 2d). The duration of miracidia staying in the FOV was only found to be significantly elevated after the addition of susceptible SCW, while the resistant SCW presented an insignificantly increased mean duration (Fig. 2e).

Biomphalaria glabrata ESP proteomic analysis
ESPs from susceptible, resistant and naïve snails were fractionated using 1D SDS-PAGE and visualised by Coomassie Blue staining. Figure 3a provides a representative gel lane of the different SCW conditions showing visibly distinct differences that suggest the contents and abundance of the proteins released vary considerably. Naïve SCW presented a relatively higher abundance of ESPs at high molecular weight (>~100 kDa). Conversely, ESPs of susceptible snails are mainly distributed in middle-to-low molecular weight regions (~30-6 kDa) and the distribution of intense bands was lower compared to those of naïve and resistant SCW. Of the two intense bands observed in resistant SCW ESPs (~ 207 kDa), the lower band was more prominent. This differed from the naïve and susceptible SCW ESPs. ESPs were also more common in the middle molecular weight region (~60-23 kDa) in resistant compared to susceptible SCW ESPs (Fig. 3a).
All samples were subjected to high-accuracy mass spectrometry and the raw data were rigorously analysed using available informatics tools. Details of all 981 ESPs identified, including ID number, best BLAST match and MS peptide matches, are provided in Additional files 3, 4, 5: Tables S2-S4. Naïve SCW provided the highest number of ESPs (597; Fig. 3b), which represented almost 61% of all identified proteins. SCW obtained from resistant and susceptible snails revealed 410 and 291 ESPs, respectively (Fig. 3b). A comparison of ESPs identified in resistant, susceptible and naïve B. glabrata revealed that 61 (approximately 6%) of the 981 proteins identified are shared. There was most overlap between the proteins expressed by the resistant and naïve snails (188) and less than half that between the susceptible and naïve snails (85). The ESPs uniquely identified within the three conditions were 162, 178 and 385 from susceptible, F1 resistant and naïve SCW, respectively.

Biomphalaria glabrata ESP annotation and gene ontology analysis
The identified ESPs were annotated using BLAST against the reference database of NCBI. Several ESPs identified in naïve SCW with high confidence MS/MS spectra were enzymes, such as superoxide dismutase (SOD), leukocyte elastase and dipeptidase (Additional file 3: Table S2). Several others were associated with the dermis, including microtubule-related proteins and lamin derivatives. Multiple ESPs supported by high confidence MS/MS spectra  Table 1). The full list of proteins is provided in Additional file 4: Table S3.
ESPs of [-10lgP] value greater than 30 identified in SCW of susceptible snails are detailed in Table 2. This included 23 non-redundant characterised ESPs, considerably fewer than those identified in naïve (132) or F1 resistant SCW (42) with similar cut-off P-value. Many high confidence ESPs were enzymes with extremely low e-values. Structural proteins, including collagen alpha-3(VI) chain isoform X1, microtubule-actin crosslinking isoform and tropomyosin, were present. Some uncertainty exists in the identification of B. glabrata glyceraldehyde-3-phosphate dehydrogenase (G3PDH), since the only supporting peptide detected also matches   Figure S1). This is highlighted in Table 2 and was not included in the functional analysis due to this uncertainty. A complete list of proteins identified in the susceptible snails is provided in Additional file 5: Table S4. Figure 4 provides comparative data for GO enrichment of resistant and susceptible SCW ESPs against the whole B. glabrata proteome. More than 20% of resistant ESPs potentially have the molecular function of ionbinding (Fig. 4a), while 'catalytic activity' corresponds to nearly 30% of susceptible ESPs (Fig. 4b). Other enriched GO terms in resistant ESPs include metabolic process of cellular amino acids, carbohydrates, cofactors, chitin, sulphur compounds and nucleobase-containing compounds, cytoskeletal and certain enzyme activities (Fig. 4a). For susceptible ESPs, hydrolase, peptidase, catalytic and extracellular activity were also enriched significantly (Fig. 4b). The differences in ESP GOs and their enrichment in comparison to whole proteome are an indicator that different types of processes were activated between resistant and susceptible snails.

Schistosoma mansoni proteins identified from susceptible SCW of B. glabrata
The mass spectral data of susceptible B. glabrata SCW proteins were analysed using the S. mansoni protein database to identify S. mansoni proteins. Thirteen nonredundant S. mansoni proteins were supported by at least one unique peptide with high confidence ( Table 3). The MS/MS spectra of supporting peptides of each protein are shown in Additional file 7: Figure S2 which showed at least five consecutive b-or y-ions, confirming the high confidence of these spectra for identification. These S. mansoni proteins include egg protein CP391B-like, G3PDH, putative nicotinate phosphoribosyltransferase and transcription factor TFIIF-alpha (Additional file 8: Table S5). Of the S. mansoni proteins two uncharacterised proteins (Smp_179420.1 and Smp_202190.1) had three and two respective unique peptides. A SignalP analysis predicted that Smp_179420.1 and Smp_093980.1 contain signal peptides (Additional file 8: Table S5).
Protein-protein interaction (PPI) networks were constructed between identified S. mansoni proteins (see Table 3) and the B. glabrata genome-derived proteome (Fig. 5a), with node proteins shown in Additional file 9: Table S6. Smp_142140.1 was identified to interact with numerous B. glabrata proteins, of which major nodes include the calcium-independent protein kinase C (PKC), rho-associated protein kinase 2 and serine threonine-kinase MRCK alpha-like proteins. These three proteins were noted to each interact with S. mansoni neuropathy target esterase/Swiss cheese-related protein  Table 3 and Additional file 7: Figure S2), though its presence from the B. glabrata database is unconfirmed This protein interacts with seven other B. glabrata proteins, including two PKCs, two serine threonine-kinase and three ribosomal proteins. There are 70 B. glabrata proteins that interact exclusively with Smp_142140.1, such as 1-phosphatidylinositol 4,5-bisphosphate phosphodiesterase, calpain, mitogen-activated kinase and spermine oxidase (Additional file 9: Table S6). The BLASTp annotation of Smp_142140.1 predicts the  Figure 5b indicates that several identified S. mansoni proteins may interact with many other proteins of the S. mansoni proteome. G3PDH interacts with several enzymes playing important roles in metabolism of S. mansoni, including phosphoglycerate kinase, glucose-6-phosphate isomerase and phosphoglycerate mutase. Another highly connected node, TFIIF-alpha, interacts with various RNA polymerase subunits and transcription initiation factor proteins. Many of these interactions were supported by the co-expressions of their corresponding genes within S. mansoni and/or other model organisms suggested by STRING (Additional file 10: Figure S3). The enrichment analysis of this PPI identifies 107 edges among 33 nodes, which shows significantly more interactions (P-value < 1.0e-16). Enriched biological processes includes a few metabolic processes, such as carbohydrate metabolic, pyruvate metabolic, glycolytic, nucleotide catabolic and glucose metabolic processes (false discovery rate, FDR < 0.0001), associated KEGG pathways (FDR < 0.0001) are carbon metabolism, glycolysis/gluconeogenesis, biosynthesis of amino acids, pentose phosphate pathway and basal transcription factors (Additional file 9: Table S6).

Discussion
This study aimed to compare the ESPs of Schistosomaresistant, susceptible and naïve B. glabrata, to shed light on the changes of ESPs that may lead to significantly different behavioural modifications of S. mansoni miracidia. This involved the collection, behavioural bioassay analysis, fractionation and comparison of SCW ESPs from these three groups. Additionally, PPIs were conducted to provide further information into the potential functions and significance of several of these proteins regarding infection.
In the presence of SCW, miracidia tend to increase their angular velocity while slightly decreasing linear velocity [31,53]. We had previously shown that naïve B. glabrata SCW significantly reduced their velocity and elevated the tortuosity by approximately 20% and 70%, respectively [31]. As shown in Fig. 2b, c, none of the pH-neutral water, susceptible or F1 resistant B. glabrata SCW produced any significant change in miracidia linear velocity or tortuosity. The variation of velocity among post-SCW susceptible or F1 resistant samples is wider than those of pre-SCW, suggesting that SCW influences miracidia behaviour differently to naïve SCW [31]. The response of miracidia to pH-neutral water is consistent with expectations, as this water had not been exposed to any B. glabrata. The quantity of miracidia present following addition of susceptible SCW increased significantly when compared to pre-SCW or post-pH-neutral water addition, indicating a possible attraction effect (Fig. 2d). However, the increase in activity was remarkably weaker than that of naïve SCW, which produced an increasing magnitude of about 4-fold [31]. This suggests that susceptible B. glabrata at two-week post-infection might still release attractant(s) yet at a much lower concentration. A  Our GO enrichment analysis of resistant B. glabrata ESPs revealed that some noticeable activities were related to oxidoreductase activities (see Fig. 5a). This includes SOD, which catabolises superoxide radicals to hydrogen peroxide that haemocytes commonly employ to kill schistosomes [54,55]. Highly resistant strains, such as 13-16-R1 B. glabrata, tend to express higher concentrations of SOD than less resistant snails [56]. Some identified detoxifying agents for hydrogen peroxide include glutathione peroxidase and peroxiredoxin [57,58]. Their functions involve the oxidation of hydrogen peroxide to water to prevent phospholipid peroxidation [59,60]. Peroxiredoxin 1 and 2 have been identified to maintain molluscan health through catalysing the interaction between thioredoxin and hydrogen peroxide and are also expressed earlier in resistant strains than susceptible ones following infection [60,61]. Hydrogen peroxidedetoxifying protein concentrations positively correlate to the snail's resistance [60]. The detection of several different redox reactants and enzymes in the resistant snail ESPs is predominately consistent with the existing literature on intramolluscan defence mechanisms.
Other identified ESPs involved in parasite defence or immune protection include leukocyte elastase, tyrosinase, heat shock proteins and adenosine deaminase. The presence of leukocyte elastase is expected in resistant B. glabrata, as elastase has been identified to work in conjunction with hydrogen peroxide in terminating schistosome invaders [62]. Similarly, tyrosinase is suspected to be involved in anti-pathogen activity [63]. The upregulation of heat-shock protein is a common response in B. glabrata subjected to stress, such as molluscicide exposure and infection [64]. Abundance of adenosine deaminase has been noted in definitive hosts after taking praziquantel, suggesting anti-pathogen functions [65].
Carboxypeptidase's presence suggests that resistant B. glabrata are healthier due to its functions as a digestive enzyme [66]. Identifying collagen may not be relevant because although it is present within proximity to haemocytes, it is not consistently affected by S. mansoni infection [67]. It may be an indicator of increased cell sloughing. Rho-GDP dissociation inhibitor has been associated with cytoskeletal formation [68]. Its direct immunological significance in B. glabrata is unknown. Haemoglobin has been suspected to play some role in pathogen resistance, though its function in B. glabrata immunology is still in need of further investigation [69].
Superoxide dismutase, peroxidase and collagen were also identified in naïve SCW (see Additional file 3: Table S2). The significance of the excretion of typically endogenous proteins must be left open to future enquiry. Given S. mansoni miracidia's ability to differentiate between highly and less infected snails it may be possible that the release of these proteins attracts miracidia [70]. Future research, such as bioassays with specific proteins, should identify whether these proteins or their comprising peptides act as attractants for miracidia.
Our GO enrichment analysis of susceptible B. glabrata ESPs indicated greatest enrichment of proteins involved in hydrolase, catalytic and extracellular activity (see Fig. 4b and Table 2). Susceptible B. glabrata release zinc metalloproteinase, which is involved in mitigating tissue damage and inflammation [71]. Its presence may suggest the molluscan host is attempting to minimise S. mansoni damage. It is difficult to determine the significance of mucin-like proteins involved. The immunological role of mucins in B. glabrata has been subject to little investigation [72,73]. Some notable digestive hydrolases detected in the susceptible SCW include cysteine peptidases [66,74]. Notable cysteine peptidases include cathepsin B and cathepsin L, the latter of which has been identified in susceptible B. glabrata SCW [75]. While cathepsin B correlates with resistance to infection, previous studies have identified cathepsin L-like gene upregulation in B. glabrata susceptible to infection from the intestinal fluke Echinostoma caproni [76]. Similar responses to S. mansoni infection have not been confirmed. Phospholipase is a defensive enzyme vital to superoxide production and oxidase activation [76].
The presence of phospholipase is not necessarily inconsistent with susceptibility. An analysis from another species of Biomphalaria, Biomphalaria pfeifferi, has suggested that phospholipase activity is most prevalent after three days of infection [77]. This indicates correlation with relatively prolonged presence of S. mansoni. Glutathione S-transferase is an antioxidant and given its presence in S. japonicum endogenous ESPs it appears to be relatively ubiquitous [26,55]. Analyses of E. caproniinfected B. glabrata have identified an almost three-fold increase in mRNA for detoxifying enzymes such as glutathione S-transferase two days post-infection. It is estimated at this stage the parasite has been encapsulated and the enzyme is trying to diminish oxidative stress [78].
Little has been published about the other proteins identified in susceptible B. glabrata. Fewer proteins identified as immunologically significant were identified within the susceptible snails, consistent with expectations.
The analysis of susceptible B. glabrata ESPs using a reference S. mansoni proteome database revealed several proteins, such as G3PDH (see Table 3), which is prominent in host-parasite interactions [26,79]. Combined with cysteine peptidases, G3PDH is essential in the protection of S. mansoni due to its involvement in gene expression [80]. It is a key component of the glycolytic pathway and has been focussed on as a vaccine candidate [81,82]. Schistosoma mansoni G3PDH was supported by three peptides in this study, one of which shares the same sequence (i.e. K.LTGMAFR.V) with B. glabrata G3PDH. While the presence of S. mansoni G3PDH was confirmed, the presence of B. glabrata G3PDH can only be speculated. Egg protein CP391B has been identified exclusively in the sporocyst stage of S. japonicum and indicates successful infection [83]. The potential immunosuppressive properties of this protein have not been studied yet. Phosphoribosyltransferase has been identified as essential to nucleotide metabolism, though nicotinate phosphoribosyltransferase has yet to be subject to much analysis [84,85]. The specific role of proline-serine-threonine phosphatase interacting protein in S. mansoni is still unknown. The seven other identified proteins were all novel.
Several PKCs from B. glabrata were identified to be interacting with neuropathy targeted esterase and Smp_142140.1 from S. mansoni. PKCs are involved in such processes as reparation of damaged tissue and parasite termination through the regulation of hydrogen peroxide [28,86]. PKC receptors are upregulated most rapidly in resistant snails within 5-10 hours of exposure to miracidia as opposed to susceptible snails, which can take several days to respond [28], suggesting the need for upregulation of PKCs in the development of resistance. These two S. mansoni proteins could be targets of B. glabrata PKC. Serine threonine kinases have been identified in the RNA of haemocytes in B. glabrata; however, besides that little is known about its function in the interaction with S. mansoni [87]. Both PKC and serine threonine protein kinase are activated by diacylglycerol [88], which indicates the potential association of these two proteins in response to infection. There are gaps in our understanding of the immunological significance or functions of the other kinases in B. glabrata observable in Fig. 5a.
Several proteins present in the PPI conducted with reference to the S. mansoni proteome (see Fig. 5b) have been identified in previous studies where proteins isolated from miracidia developing into sporocysts were fractioned by SDS-PAGE. These include fructose-bisphosphate aldolase, transketolase, triosephosphate isomerase, enolase and glucose-6-phosphate isomerase [22,89]. Furthermore, nicotinamide has also been identified in the sporocysts of S. mansoni [90]. Some of these proteins, such as fructose-biphosphate aldolase, triosephosphate dehydrogenase and glucose-6-phosphate dehydrogenase have been identified as glycolytic enzymes, similar to G3PDH, which is also involved in the transition of miracidia to sporocysts [22,91,92]. This indicates that these proteins, or other uncharacterised proteins, may function as markers of a successful infection for other miracidia to detect, possibly decreasing their likelihood of further infection.

Conclusions
In this study, fractionated SCW ESPs from susceptible, F1 resistant and naïve B. glabrata were analysed using LC-MS/MS to identify proteins significant to B. glabrata and S. mansoni interactions. The significant modifications of miracidia behaviour were only observed following the addition of naïve and susceptible (although less prominent) B. glabrata SCW. F1 resistant B. glabrata SCW displayed ESPs corresponding with immunological activity while susceptible B. glabrata SCW contained fewer defensive-type enzymes, potentially conferring a weaker resistance to parasite infection. While several ESPs identified with reference to the S. mansoni database have well-documented functions in snails or other species, many remain uncharacterised. Our PPI analysis indicated potential proteins relevant to the response of susceptible snails to miracidia and proteins corresponding to sporocyst development. This suggests that they may be acting as deterrents to miracidia. This study identified several protein candidates to further investigate to reveal the interactions between B. glabrata and S. mansoni. This may facilitate future innovations into preventing the infection of B. glabrata snails and inform research into other molluscan hosts.