Discovery by organism based high-throughput screening of new multi-stage compounds affecting Schistosoma mansoni viability, egg formation and production

Schistosomiasis, one of the most prevalent neglected parasitic diseases affecting humans and animals, is caused by the Platyhelminthes of the genus Schistosoma. Schistosomes are the only trematodes to have evolved sexual dimorphism and the constant pairing with a male is essential for the sexual maturation of the female. Pairing is required for the full development of the two major female organs, ovary and vitellarium that are involved in the production of different cell types such as oocytes and vitellocytes, which represent the core elements of the whole egg machinery. Sexually mature females can produce a large number of eggs each day. Due to the importance of egg production for both life cycle and pathogenesis, there is significant interest in the search for new strategies and compounds not only affecting parasite viability but also egg production. Here we use a recently developed high-throughput organism-based approach, based on ATP quantitation in the schistosomula larval stage of Schistosoma mansoni for the screening of a large compound library, and describe a pharmacophore-based drug selection approach and phenotypic analyses to identify novel multi-stage schistosomicidal compounds. Interestingly, worm pairs treated with seven of the eight compounds identified show a phenotype characterized by defects in eggshell assemblage within the ootype and egg formation with degenerated oocytes and vitelline cells engulfment in the uterus and/or oviduct. We describe promising new molecules that not only impair the schistosomula larval stage but also impact juvenile and adult worm viability and egg formation and production in vitro.


Introduction
Parasitic trematodes of the genus Schistosoma cause schistosomiasis, a life-threatening infectious disease affecting both humans and animals. Schistosomiasis, one of the world's greatest neglected tropical diseases, contributes to the global morbidity with 4,026,000 DALYs (disability-adjusted life years) [1]. Among human parasitic diseases, schistosomiasis ranks second behind malaria in socio-economic terms, public health importance and prevalence in the developing world. More than 780 million people are at risk of infection and it is estimated that there are approximately 261 million infected people in 78 countries, of whom 85% reside in sub-Saharan Africa [2]. Three major species (S. mansoni, S. haematobium and S. japonicum) account for the majority of human infections. Similar to other trematodes, schistosomes have a complex life cycle consisting of both free-living and parasitic forms with several developmental stages [3]. Moreover, whereas most of the trematodes are hermaphrodites, schistosomes have evolved two separate sexes and the sexual maturation of female worms as well as the subsequent production of eggs are both dependent on the pairing status with males [4][5][6]. Sexually mature adult females can produce, a large number of eggs each day, formed within the ootype by one oocyte and 30-40 vitelline cells. The eggs are responsible for both parasite transmission and disease pathogenesis as part of the egg produced can be trapped into host's tissues causing granuloma formation and inflammatory processes which interfere with organs function [7]. To date praziquantel (PZQ) is the only drug recommended for the treatment of schistosomiasis being very effective against adult worms of all the medically important Schistosoma species (S. mansoni, S. haematobium and S. japonicum) [8,9]; however, PZQ is relatively ineffective against the juvenile and schistosomula larval stages both in vivo and in vitro [10][11][12][13] and it does not prevent re-infection [14,15]. Moreover its increasingly widespread use in mass chemotherapy campaigns and the identification of field [16][17][18][19] and laboratory isolates that exhibit significantly reduced susceptibility to PZQ [20][21][22][23][24] represents a serious concern for the development of drug-resistance strains. Consequently the search for novel schistosomicidal compounds is currently viewed as an urgent goal and there is great interest in new chemical compounds with demonstrated (i) ability to kill the parasites, possibly targeting different developmental stages, in order to lower worms burden and (ii) capacity to impair egg production, so that pathological effects are minimized, or even completely abolished. In this work we show how we succeeded in meeting both these requirements, and identified a number of new molecules having effects on both worm viability and egg production. To meet our aims, a recently developed high-throughput assay based on ATP quantitation in the larval stage schistosomula of S. mansoni [25] was used to screen a large compound library. In the follow up step a pharmacophore-based drug selection approach and phenotypic analyses were employed to identify novel multi-stage schistosomicidal compounds. We herein describe promising new hits for further chemical optimization. These hits are active on the schistosomula larval stage as well as on juvenile and adults worms. Interestingly, treatments of worm pairs with the majority of the new hits identified, induce a phenotype characterized by defects in eggshell assembly in the ootype and egg formation with degenerate oocytes and vitelline cells engulfment in the uterus and/or oviduct. Moreover, some compounds induce also gut dilation with detachment of the gastrodermis in adult worms.

Compound library composition
At the time the high throughput screening (HTS) was performed the compound collection consisted of around 40,000 small molecules from both commercial and non-commercial suppliers. In addition to FDA-and/or EMA-approved drugs the collection contained a structurally diverse range of chemotypes with average molecular weight 370 Da. Analysis of the collection revealed an attractive distribution of physicochemical properties (e.g. logP, sp3 character and hydrogen bond donor/acceptors) and good structural diversity (average Tanimoto [26] distance from the nearest neighbour of 0.38).

Ethics statement
Female ICR (CD-1) 4-7 week-old mice (Harlan Laboratories or EMMA) were housed under controlled conditions (22˚C; 65% relative humidity; 12/12 hours light/dark cycle; standard food and water ad libitum). Animals were subjected to experimental protocols (Authorization N. 25/2014-PR) approved by the National Research Council, Institute of Cell Biology and Neurobiology animal care and use committee and the public Veterinary Department of the Italian Ministry of Health, and experiments were conducted according to the ethical and safety rules and guidelines for the use of animals in biomedical research provided by the relevant Italian law and European Union Directive (Italian Legislative Decree 26/2014 and 2010/63/EU) and the International Guiding Principles for Biomedical Research involving animals (Council for the International Organizations of Medical Sciences, Geneva, CH). All adequate measures were taken to minimize animal pain or discomfort.

Maintenance of the S. mansoni life cycle
A Puerto Rican strain of S. mansoni was maintained by passage through albino Biomphalaria glabrata, as the intermediate host, and ICR (CD-1) outbred female mice as definitive host as previously described [25]. Mice infected 7-8 weeks previously with single sex or double sex cercariae were euthanized with intraperitoneal injections of Tiletamine/Zolazepam (800 mg/ kg) + Xylazine (100 mg/kg) and adult parasites were harvested by reversed perfusion of the hepatic portal system and mesenteric veins [27]. Juvenile worms were obtained from mice 28 days after infection.

Preparation of parasites, viability assays and egg counts
Cercariae were converted to newly transformed schistosomula by mechanical transformation using an optimized version of the protocol of Brink et al., 1977 [28], previously described by Protasio et al. [29] and adapted in our laboratory [25].
The schistosomula viability assay was carried out as previously described [25]. Briefly, compounds dissolved in DMSO were transferred to 384-well, black, tissue culture plates using the acoustic droplet ejection technology (ATS-100, EDC Biosystems, USA). DMSO alone and gambogic acid (10 μM) were used as high and low control in each plate. A suspension of schistosomula in complete DMEM medium (without phenol red) was transferred to assay plates by a multidrop dispenser (Thermo Fisher, USA) in the number of about 100 schistosomula per well in a final volume of 30 μl. After 24 hours incubation at 37˚C and 5% CO 2 , a volume of 30 μl of CellTiter-GLO reagent (CTG) (Promega, USA) was added resulting in the generation of a luminescence signal proportional to the amount of ATP present in the well. Sample luminescence levels (proportional to ATP levels) were detected 30 minutes after CTG addition and quantified as RLU (Relative Luminescence Unit) by a charge-coupled device (CCD)-based detector (ViewLux, PerkinElmer USA). Screening and potency results were evaluated by GraphPad (Prism, USA).
Worms recovered from infected mice at 28 days (juvenile worms) or 7-8 weeks (adult worms) post-infection were cultured in DMEM (with phenol red) complete tissue culture medium at 37˚C in 5% CO 2 atmosphere. For all treatments, 5-10 males or couples were incubated with selected compounds at the indicated concentrations and cultured in 3-5 ml of complete tissue culture medium for up to 7 days, unless otherwise stated as previously described [30]. The compound was given to parasites in vitro only once without medium addition and/ or replacement.
During the culture time, survival was monitored daily under a Leica MZ12 stereomicroscope and viability scored as previously described, based on phenotypical changes like motility and general appearance (tegumental damage and darkness, gut peristalsis and morphology, plate attachment) Briefly, the type and number of phenotypic responses were recorded into a 'severity score' ranging from 0 (severely compromised) to 3 (no effect) as previously described [25,30]. The following phenotype scoring criteria were used: 3 = plate-attached, good movements, clear; 2 = slower or diminished movements, darkening, minor tegumental damages; 1 = movements are heavily slowered, dark, tegument is heavily damaged; 0 = dead, lack of any movements. For each sample the following formula was used: S ðworm scoresÞ number of worms Data were expressed as % severity score (viability) relative to DMSO. All tests were repeated at least three times.
Images were recorded using with an Olympus BX41 microscope served by an Olympus SP-350 camera (for adult worms) or an Olympus AX70 fluorescence microscope supported by an Olympus XM10 camera with the Olympus CellSens Standard 1.8.1 software (for eggs, spermatozoa, oocytes and vitelline cells in the worm culture media).
The number of eggs produced by all worm couples was counted at day 3 using an inverted LEICA DM IL microscope.

Carmine-red staining
Carmine-red staining was performed essentially as previously described [30]. Briefly, adult worms were fixed for at least 24 hours in AFA (95% ethanol at 70%, 3% formaldehyde at 37%, 2% glacial acetic acid) at room temperature, stained for 30 minutes with 2.5% hydrochloric carmine-red and then de-stained by several washes in acidic alcohol (70% EtOH at 100%, 2.5% HCl at 37% and 27.5% double-distilled H 2 O) until no more color was released by the samples. Next, samples were dehydrated for 2 min in 90%, 30 sec in 100% ethanol and worms preserved in Canada balsam on glass slides. Images were taken on an Olympus FV1200 confocal laser scanning microscope using an UPlanFLN 40X immersion oil objective (NA = 1.30) with optical pinhole at 1AU and a multiline argon laser at 488 nm as excitation source. The images were collected as a single stack.

Statistical analysis
All statistical tests were performed using GraphPad Prism version 6.0c software (San Diego, CA, USA). The data are shown as mean ± standard error of the mean (SEM) or ± standard deviation (SD) as indicated. Differences observed in the in vitro assays were analyzed by Student's t-test. For all experiments, p-values < 0.05 were considered to be statistically significant.

Hit identification and confirmation
A compound collection comprising 38,811 molecules was screened at a single concentration of 10 μM using the schistosomula viability assay previously described [25]. The screening was carried out in multiple batches, according to the production of schistosomula, with the average number of plates per run being 10. In order to determine the quality of the results, the Z' value [31] was calculated for each tested plate using the vehicle (DMSO) dispensed wells as negative control and gambogic acid dispensed wells as positive control. All plate Z' was found to be above the 0.5 threshold which is commonly considered the lowest acceptable value for a robust assay ( Fig 1A). The activity of each tested compound was calculated as percentage of ATP reduction against the vehicle (0%) and gambogic acid (100%), thus representing the percentage of dead schistosomula. This normalization allowed compensation for the schistosomula batch-to-batch variations. The compound activity distribution was found to be Gaussian, as a consequence the positivity threshold was set to 60%, which is the average sample activity plus three times the standard deviation ( Fig 1B).
By using this threshold, 275 active compounds were identified (0.7% hit rate). It has to be noted that the average compound activity is greater than 0 and the activity distribution is rather broad. These aspects will be commented on in the discussion section. The 275 hits from the HTS were subjected to quality control by LC-MS in order to check compound identity and purity (acceptable purity criteria set to be > 90% peak area in the diode array trace). Sixteen compounds were discarded at this stage as they failed to pass the QC. Hit selection was performed using a clustering approach based on the Taylor Butina algorithm [32], a non-hierarchical clustering method that ensures that each cluster contains molecules with a certain cutoff (or threshold) distance from a central compound. Circular fingerprints with radius 2 and 2048 bits were generated using the RDKit software [33], with the purpose of generating a similarity matrix based on a Tanimoto index [26,33]. The effective number of neighbours for each molecule was calculated based on the Tanimoto level (0.8) used for clustering. This procedure gave a collection of 80 clusters of which 30 were singletons. Eighty centroids were selected together with 30 compounds picked randomly from the most populated clusters. The final set of 110 compounds was tested again in an independent schistosomula dose response curve assay (40 nM-50 μM) and from this screen 22 compounds were confirmed to be active with LD 50 ranging from 5 to 40 μM. After visual inspection eight compounds were selected to be further profiled. Fresh powders of these eight compounds were repurchased/resynthesized and retested on schistosomula (Fig 2). The previously found LD 50 's were confirmed for all eight compounds.

Identification of multi-stages schistosomicidal compounds
In order to investigate the efficacy of the eight selected compounds on other S. mansoni developmental stages, survival assays and bright-field microscopy analyses were performed on adult male worms. All compounds but SmI-8 and SmI-82 negatively influenced adult worms viability; SmI-1, SmI-10 (perhexiline maleate), and SmI-11 showed the highest effect being the most active compounds at 10 μM (Fig 3). Indeed SmI-1 and SmI-11 were able to induce complete parasites death 7 days after a single compound treatment, an effect similar to that previously reported for perhexiline maleate (SmI-10) [30] (Fig 3).
Importantly, SmI-1 and SmI-11 were highly effective also on juvenile S. mansoni worms (4 weeks old) ( Fig 4A) and mature paired parasites (7-8 weeks old) (Fig 4B). In particular at the concentration of 10 μM, SmI-1 and SmI-11 were able to induce parasites death in juvenile New multi-stage schistosomicidal compounds worms respectively 24 hours and 5 days post treatment ( Fig 4A) and in mature couples at 5 and 7 days after treatment using 5 μM concentration ( Fig 4B).
Such impact on all parasites stages viability has been previously reported also for the compound perhexiline maleate (SmI-10) [30].
Importantly at sub-lethal concentrations (1 and 2.5 μM), even though not significantly affecting viability or parasites pairing, both compounds caused marked reduction in the number of eggs laid by worm pairs 3 days upon treatment (Fig 4C).
A pharmacophore approach to identify multi-stage schistosomicidal compounds Ligand-based molecular modelling approaches are generally used when a reliable target structure is not available from which to derive the structural requirements important for activity. Pharmacophore modeling belongs to this kind of methodology and has been widely used in drug discovery [34]. To identify novel scaffolds by using the three identified hits, SmI-11, SmI-10 (perhexiline maleate), and SmI-1, 3D pharmacophore models were developed. A 3D structure for each compound was generated using energy minimization and the resultant minimized structures were used as starting points for subsequent conformational analysis using a multi-objective genetic algorithm implemented in BALLOON software [35,36]. Unique low energy conformations within 10 kcal/mole of the corresponding global energy minimum were collected for each molecule. Pharmacophore models were then generated using Align-it software [37]. On the basis of the structural characteristics of compounds SmI-11, SmI-10, and SmI-1 aromatic and lipophilic features, hydrogen bond donor, hydrogen bond acceptor and positive charge were considered for model development. The generated pharmacophore models were visually evaluated through the generation of molecular maps as described in Fig 5. A database search to find compounds matching the pharmacophores was conducted using an expanded version of IRBM's internal library consisting of 40,000 molecules. Results were ranked on the basis of the Tversky similarity index [38] and the number of features in the corresponding phamacophoric model. Top 130 candidates were selected and assayed on schistosomula: 11 resulted with a good LD 50 and were further profiled. The resulted compounds New multi-stage schistosomicidal compounds

Fig 4. Effects of SmI-1 and SmI-11 compounds on S. mansoni juvenile worms and adult pairs viability and egg production in vitro.
Viability curves of juvenile worms (4 weeks old) (A) and adult worms pairs (7-8 weeks old) (B) incubated with the SmI-1 and SmI-11 compounds. Worm viability was assessed as described in the methods section. Parasites were cultivated in presence of increasing concentrations of both compounds as follow: 2.5 μM (solid square), 5 μM (solid triangle), 10 μM (solid inverted triangle), and 20 μM (solid diamond). DMSO was used as negative control (solid circle). C) Total egg counts laid by parasites treated during 72 hours with different sub-lethal doses of SmI-1 and SmI-11 normalized to worm couples. The mean data ± SEM of three independent experiments are shown. The levels of statistical significance are indicated above bars; *p-value < 0.05, ****p-value < 0.0001, Student's t test.
Next, to test their ability to impact worms survival the new 11 compounds were all assayed on adult male worms at the concentrations of 10 and 20 μM with five of them (SmI-233, SmI-251, SmI-290, SmI-291, and SmI-308) being active. In particular at the concentration of 20 μM we found less than 50% survival rate in worms 72 h upon treatment and almost complete death occurring at day 7 (Fig 7).
Importantly all the selected compounds were also very active on both juvenile worms (4 weeks old) ( Fig 8A) and mature pairs (7-8 weeks old) (Fig 8B). In particular at the concentration of 20 μM all five compounds induced death of all juvenile parasites in 24 h and at the concentration of 10 μM less than 30% survival rate 72 h upon treatment with almost complete death occurring at day 5 (SmI-233, SmI-290, SmI-291, and SmI-308) or day 7 (SmI-251). The SmI-308 compound was also very effective on juvenile worms even at lower concentrations (2.5-5 μM). With respect to adult worm pairs, at the concentration of 20 μM all five compounds induced almost complete death at day 5 (SmI-233, SmI-290, SmI-291 and SmI-308) or day 7 (SmI-251). The compounds also demonstrated to be very active on pairs treated at the concentration of 10 μM at day 7.
Moreover using sub-lethal doses (5 or 2.5 μM) all the compounds, with the only exception of compound SmI-308, had a strong impact on egg production during 72 hours of treatment ( Fig 9A). Intriguingly, the compounds causing egg production impairment also induced the release of a number of oocytes, spermatozoa, vitelline cells, eggshell fragments, and abnormal eggs in the tissue culture medium while no effect was recorded for the compound SmI-308. An example of the tissue culture media observed 24 and 72 hours after treatment with compounds SmI-290 and SmI-308 is shown in Fig 9B. Compared to the other compounds a quite peculiar phenotype was observed in parasites treated with the SmI-308 compound. Despite the fact that parasites did not seem to be affected in terms of viability, pairing status, egg production, and germ cells release during the first 72 hours at sub-lethal doses, bright-field microscopy showed a general swelling of both male and female worms with a marked gut dilatation (S1 Fig). New multi-stage schistosomicidal compounds

Morphological alterations of compounds-treated parasites
In order to characterize the phenotypic alterations induced by compound treatment in more detail, carmine-red staining and confocal laser scanning microscopy analysis were performed. Remarkably, with the unique exception of compound SmI-308, all selected compound-treated worms showed impairment in egg formation (Fig 10A and 10B). Indeed, in the ootype of female parasites treated for 72 hours with SmI-1, SmI-11, SmI-233, SmI-251, SmI-290 (5 μM), and SmI-291 (2.5 μM) we did not detect any mature eggs. Moreover, the ootype and often also the uterus showed the presence of disorganized oocytes, vitelline cells and intense carmine-red positive elements thought likely to be eggshell components. A similar ootype phenotype was observed in worms treated with perhexiline maleate (5 μM) 72 hours after treatment (S2c Fig). On the contrary the ootype of worms treated with the SmI-308 compound or DMSO (vehicle) Fig 6. Activity on schistosomula of compounds selected by a pharmacophore approach. Compounds were grouped based on the parent compound. Compounds within the red frame were identified based on the SmI-10 pharmacophore, compounds within the orange frame were identified based on the SmI-1 pahrmacophore, and the compound in the black frame was identified by the SmI-11 pharmacophore. For each compound dose response curves on schistosomula viability assay are reported. The % ATP reduction (schistosomula death) plotted on the y-axis is normalized between DMSO (0%) and gambogic acid 10 μM (100%). Each point represents the average and standard deviations of three independent experiments. https://doi.org/10.1371/journal.pntd.0005994.g006 New multi-stage schistosomicidal compounds contained healthy eggs ready to be laid (Fig 10A and 10B). The overall structure of the ovaries seemed not to be affected thus retaining the normal organization with small immature oocytes (IO) in the anterior part and large mature oocytes (MO) in the posterior part. Also the vitellarium seemed to preserve its structure. However the ovaries of females treated with compounds SmI-233, SmI-251, SmI-290, and SmI-291 (5 μM) contained oocytes with alterations. In particular the ovary of worms treated with compounds SmI-233, SmI-290 and, to a lesser extent, with SmI-291 showed black spots in MO likely due to degeneration. In SmI-251-treated females aggregates of degenerate cells in the anterior part containing MO and a marked degeneration of IO were observed in the ovary. A strong degeneration of IO was also observed in parasites treated with SmI-233 and SmI-291 compounds treatment (Fig 10A and 10B). Moreover oviducts of parasites treated with compounds SmI-11, SmI-251, and SmI-290 showed a remarkable engulfment and defects of trafficking of MO towards the ootype. This phenotype was also sporadically observed in some worms treated with SmI-1, SmI-233, and SmI-291 compounds (Fig 10A and 10B) and in perhexiline maleate-treated samples (S2 Fig). On the contrary we did not detect oocytes in the oviduct of female parasites treated with compound SmI-308 or DMSO (vehicle) (Fig 10A and 10B). Finally female worms of pairs treated for 72 New multi-stage schistosomicidal compounds hours with SmI-233, SmI-290 (5 μM) and SmI-291 (2.5 μM) also showed an increased number of degenerated cells in vitelline follicles (Fig 10A and 10B).
In addition, although phenotype consistency was not always seen for all worms within a test sample, we observed that all compounds with the only exception of SmI-1 led to gut dilation with detachment of the gastrodermis and accumulation of carmine-red positive particle New multi-stage schistosomicidal compounds  Representative confocal laser microscopy images of S. mansoni female reproductive system (ootype, ovary, oviduct and vitellarium). Worm pairs were treated for 72 hours with the indicated compounds: A) SmI-1 and SmI-11 at 2.5 μM concentration; SmI-233 and SmI-251 at 5 μM concentration; B) SmI-290 and SmI-308 at 5 μM concentration and SmI-291 at 2.5 μM concentration (ootype, oviduct and vitellarium images) and 5 μM concentration (ovary). Worm pairs were stained with carmine-red as described in the methods section. In Figure 10A and 10B immature oocytes (io), mature oocytes (mo), uterus (ut), ootype (ot), vitelline duct (vd), oviduct (od), vitello-oviduct (vd/od). Arrows indicate degenerated cells and asterisk mature oocytes engulfment within the oviduct. Scale bars: 50 μm.
aggregates in the lumen (Fig 11). This alteration was especially notable for compounds SmI-290, SmI-291 and SmI-308 (Fig 11). The morphology and cellularity of the testicular lobes were similar in males of pairs treated with all selected compounds or DMSO with the exception of those treated with SmI-308 that showed cavities in the testicular lobes (S3 Fig), cavities also present in the ovary of females. Mature sperm was present in all seminal vesicles and receptacles of male and female treated parasites.

Discussion and conclusion
In modern drug discovery, the process of lead identification involves both the screening of a compound collection against a target or a complete organism and the validation of a set of hits having acceptable activity to fight the disease. To this aim small organisms such as schistosomula can be employed using in vitro assays to screen large set of compounds in an automated, objective and high-throughput manner [25]. The initial screening using schistosomula can offer both advantages and disadvantages: the use of such a small and handy stage represent an useful tool for increased throughput and improved automation but the screen workflow with schistosomula need to be validated in the other stages of the parasite.
Further phenotypic screen in the juvenile and adult parasite developmental stages as well as egg formation and production assessment are required in order to identify novel multi-stage schistosomicidal compounds with dual effects on both parasites viability and egg production impairment.
In an attempt to search for new chemical compounds for the treatment of schistosomiasis having these features, combination of high throughput library screening on schistosomula, a pharmacophore approach and phenotypic analyses on both juvenile, known to be less sensitive to PZQ compared to mature egg-laying adults [10][11][12][13], and adult parasites was successfully New multi-stage schistosomicidal compounds used for the identification of eight novel multi-stage schistosomicidal compounds, including perhexiline maleate that we previously described [30].
As reported in the results section, the HTS campaign was carried out in multiple batches, with compound throughput depending on the yield of each lot of the schistosomula production. On average ten 384-well plates were carried out per run. As a consequence we found extremely important to ensure consistent parasite production and minimize potential sources of contaminations. Once all HTS data were analysed they were found to be Gaussian distributed confirming that there is no bias in the compound collection or in the assay itself. The distribution of the results was rather broad likely due to random errors arising from both the influence of multiple assay runs and the limited number of parasites per well. In addition, the average of all HTS values was greater than zero suggesting that DMSO controls, which are located in a corner of the plate, may have slightly suffered from evaporation.
In follow up to the HTS screen our approach to profile 300 hit compounds was supported by computational chemistry. A combination of clustering and retesting identified 22 bona fide inhibitors from which eight hits were selected for further work. In macroscopic terms these eight compounds can be grouped into 3 categories: i) biphenyl analogs (SmI-1, SmI-8, SmI-12, SmI-82); ii) biaryl analogs/compounds containing two linked aryl rings (SmI-5, SmI-11, SmI-85); iii) non-aryl analogs (SmI-10). One compound from each of these categories (SmI-1, SmI-11 and SmI-10) was selected taking into account its effect on viability reduction of both juvenile and adult parasites as wells as egg production impairment. Computational analysis was used to generate more detailed pharmacophore models to identify the key elements of the structures responsible for biological activity. Use of these pharmacophores to identify further potential schistosomicidal compounds proved successful. From the 130 compounds that based on the pharmacophores were selected for assay on schistosomula, 11 confirmed hits were obtained (Fig 6). These included five analogs that were of interest and that retained activity against juvenile and adult worms.
Importantly, all selected compounds, with the only exception of SmI-308, were also able to interfere with the egg production process when used at sub-lethal doses. Along with survival data, the egg impairment caused by the selected compounds represents an outstanding aspect in fighting the spreading of schistosomisiasis as the egg production represents a key component for the transmission and immunopathology of the disease.
In addition, morphological studies by confocal microscopy analyses highlighted strong characteristic egg-associated phenotypes for seven out of eight hits. A lack of eggshell formation and an absence of mature eggs in the ootype associated to degeneration of immature and/ or mature oocytes and vitelline cells and a remarkable engulfment of MO and vitelline cells, appeared to be some common features among these seven compounds (Fig 10). Defects in the egg machinery with production of abnormal eggs were observed for all compound-treatments with the only exception of compound SmI-308. On the other hand, parasites treated with SmI-308 showed a completely different phenotype characterized by a remarkable gut enlargement with tegumental invaginations and oedema-like swellings of the body (S1 Fig) also detectable in the reproductive organs (ovary and testis) (S3 Fig) similar to the one recently described by others with arylmethylamino steroids compounds [39]. Defects in the ootype, with formation of dysplastic eggs have been shown before as consequences of knocking-down the activity of protein kinases [40][41][42][43] as well as treatment with derivatives of biarylalkyl carboxylic acid [44]. Alteration of the gastrodermis, that we describe here, has been detected with several other schistosomicidal compounds including mefloquine [45,46] or artemether [47] or derivatives of biarylalkyl carboxylic acid [44] or arylmethylamino steroids compounds [39].
Eggshell assemblage and egg formation occurring in schistosomes [48] is a complex process also epigenetically regulated [49]. Further understanding of their regulatory molecular mechanisms is valuable since disruption of these processes may provide leads for intervention with drugs for controlling schistosomiasis.
We can conclude that the pharmacophore approach can represent an important tool in new hits identification process. It is encouraging that we have been able to identify a number of compounds with IC50 <10 μM against all developmental stages of the parasite definitive host and with potency against juvenile worms higher than PZQ. It is interesting that some morphological alterations associated to the treatment seem to be related to defects in the egg machinery and even thought the potential targets are still unknown, the phenotypes observed with these novel schistosomicidal compounds could drive investigation toward some common mechanisms, such as on genes and signaling pathways involved in eggshell assemblage and eggs formation.