A novel cell-free method to culture Schistosoma mansoni from cercariae to juvenile worm stages for in vitro drug testing

Background The arsenal in anthelminthic treatment against schistosomiasis is limited and relies almost exclusively on a single drug, praziquantel (PZQ). Thus, resistance to PZQ could constitute a major threat. Even though PZQ is potent in killing adult worms, it has been shown to be limited in its activity against earlier developmental stages. Current in vitro screening strategies for new drugs depend on newly transformed schistosomulae (NTS) for initial hit identification, thereby limiting sensitivity to new compounds predominantly active in later developmental stages. Therefore, the aim of this study was to establish a highly standardized, straightforward and reliable culture method to generate and maintain advanced larval stages in vitro. We present here how this method can be a valuable tool to test drug efficacy at each discrete intermediate larval stage, reducing the reliance on animal use (3Rs). Methodology/principal findings Cercariae were mechanically transformed into skin-stage (SkS) schistosomulae and successfully cultured under serum-free and cell-independent conditions for up to four weeks with no loss in viability. Under these conditions, larval development halted at the lung-stage (LuS). Addition of human serum (HSe) propelled further development into juvenile worms within eight weeks. Skin and lung stages, as well as juvenile worms, were submitted to 96-well format drug screening assays using known anti-schistosomal compounds such as PZQ, oxamniquine (OXM), mefloquine (MFQ) and artemether (ART). Our findings showed stage-dependent differences in larval susceptibility. Conclusion With this robust and highly standardized in vitro assay, important developmental stages of S. mansoni up to juvenile worms can be generated and maintained over prolonged periods of time. The phenotype of juvenile worms when exposed to reference drugs was comparable to previously published works for ex vivo harvested adult worms. Therefore, this in vitro assay can help reduce reliance on animal experiments in the search for new anti-schistosomal drugs. Author Summary Schistosomiasis remains a major health threat, predominantly in developing countries. Even though there has been some progress in search of new drugs, praziquantel remains the only available drug. Probably the most important advance in the search for new drugs was in vitro transformation of cercariae and their subsequent culture. However, hit identification in compound screenings is exclusively tested in skin stage parasites and is only confirmed for more mature worms in a subsequent step. This is in part due to the lack of an easy culture system for advanced-stage parasites. We present here a reliable and highly standardized way to generate juvenile worms in vitro in a cell-free culture system. The inclusion of in vitro drug tests on advanced-stage parasites in initial hit identification will help to identify compounds that might otherwise be overlooked. Furthermore, the ability to continuously observe the parasite’s development in vitro will provide an important platform for a better understanding of its maturation in the human host. Taken together, this opens up new avenues to investigate the influence of specific cell types or host proteins on the development of Schistosoma mansoni and provides an additional tool to reduce animal use in future drug discovery efforts (3Rs).


Introduction
Schistosomiasis, a chronic and debilitating helminthic disease, is one of the most important 66 neglected tropical diseases (NTD). The WHO estimates that currently more than 206 million 67 people are infected and in need of preventive chemotherapy world-wide [1] and over 200,000 68 people die each year due to the sequelae of the disease [2,3]. Among the parasitic diseases, 69 schistosomiasis is often considered only second in importance to malaria [2] and thus a major 70 public health menace. Therefore, the WHO aims to eliminate schistosomiasis as a public health 71 problem globally by 2025 [4,5]. Implementation of safe water, sanitation and hygiene (WASH) 72 strategies [6], intensified case management, veterinary public health, vector control and mass 73 drug administrations (MDAs) are all crucial to reduce the disease burden [5]. Of all these 74 approaches, MDA dominates national control programs thanks to the excellent safety and 75 efficacy profile of praziquantel (PZQ), the only currently available drug [7,8], as well as its 76 low cost per treated individual [9]. However, the reliance on PZQ also raises concerns about 77 emerging resistance should the drug pressure increase. Resistance against PZQ has already been 78 observed in experimental models [10] while a decrease in drug efficacy has been observed in 79 the field [7,11,12]. To be prepared for the emergence of resistant strains of schistosomes and  The current cultivation protocols for in vitro generated larval stages such as the schistosomulae 86 rely on the supplementation of fetal calf serum (FCS ) for short term culture [14,15]

98
To establish a robust in vitro assay, it is important to continuously imitate the parasites in vivo 99 development. This development within the final host is quite complex and occurs over a period 100 of seven weeks. After penetration of the skin by the cercariae, the schistosomulae can be found 101 for approx. three days within the skin before they start migrating through the host's vasculature. 102 This skin stage (SkS) is followed by traversing the capillaries of the lung where the majority of 103 the parasites can be found on day 7 after infection. To facilitate their migration, the parasites 104 become longer, slenderer and more active. The lung stage (LuS) schistosomulae continue their 105 journey to the portal and mesenteric veins following the bloodstream. There they initially 106 undergo several morphological changes. The gut is formed, the parasites initiate feeding and 107 they start growing. This early liver stage (LiS) is followed by the late LiS characterized by a 108 drastic increase in length and prominent oral and ventral suckers. These juvenile worms (late  Treatment with PZQ, although initially with good efficacy, does not diminish the high 112 reinfection rates encountered in the field [18]. This is partly due to the inability of PZQ to 113 efficiently target early larval stages of the parasite [19,20]. In the current search for new anti-7 114 schistosomal drugs, a two-step strategy is used. Firstly, SkS schistosomulae are tested 115 immediately following their transformation from cercariae and then, once an active compound 116 has been identified, it is tested mainly on ex vivo cultured adult worms that have been isolated 117 from infected hamsters or mice [14,[21][22][23]. Subsequent larval stages like the LuS, early and 118 late LiS are omitted as potential drug targets. Therefore, future compounds with an activity 119 predominantly directed against juvenile and adult stages, such as PZQ, might be overlooked.

120
Thus, a highly standardized and robust way to generate advanced larval stages of Schistosoma 121 mansoni provides an opportunity to incorporate initial advanced-stage schistosomula testing 122 into the current drug screening strategies to meet the desired compound target profile (TCP). 123 We disclose here a new serum-and cell-free cultivation method of newly transformed 124 schistosomulae (NTS) up to the LuS and a cell-free cultivation method up to late LiS juvenile 125 worms of S. mansoni. Our culture system allows the detection of stage-dependent differences 126 in the activity of drugs with known anti-schistosomal properties. We used PZQ, the only 127 currently available drug; Oxamniquine (OXM), the standard drug to treat schistosomiasis 128 caused by S. mansoni prior to the advent of PZQ, and two antimalarial drugs that have recently 129 been described to have anti-schistosomal properties, Mefloquine (MFQ) and Artemether (ART) 130 [16,24]. Therefore, the aim of this study was to establish a highly standardized, straightforward 131 and reliable culture method as the basis for integrating drug screenings of advanced larval stages 132 in initial hit identification. For verification of this cell-free cultivation assay, we tested 133 compounds with already known anti-schistosomal properties against SkS schistosomulae. We 134 hereby demonstrate that more advanced developmental stages such as LuS parasites and 135 juvenile worms (LiS) can also be cultured and tested in vitro under cell-free conditions.

139
Cercariae of a Brazilian strain of S. mansoni were harvested from infected Biomphalaria 140 glabrata snails and used for mechanical transformation into NTS as described before [25].

141
Briefly, cercariae were incubated 30 min on ice then centrifuged at 1932 x g for 3 min at 4°C.  Scoring criteria for the viability score 156 Viability of NTS was scored using an Axiovert10 microscope (Zeiss, Germany). The scoring morphology and granularity. The score was applied ranging from 0 (dead parasites, no 160 movement, heavy granulation, blurred outline, rough outer tegument and blebs) to 1 (very 9 161 reduced motility, rough outer tegument with some blebs) to 2 (reduced motility or increased 162 uncoordinated activity, slight granularity, intact tegument with slight deformations) and finally       (Fig 3B-D).
299 Surprisingly, however, at a concentration of 50% HSe, the survival rate (69.1% or 44.7 ± 10.1 300 dead out of 144.7 ± 21.2 total) as well as viability (2.2 ± 0.6) started to decline around week 8 301 (Fig 3D, E). Nevertheless, the final viability score and survival rate was still higher than in the 302 control (viability score of 0.3 ± 0.0 and survival rate of 1.8% or 160.3 ± 7.8 dead out of 163.3 303 ± 8.5 total) (Fig 3A, E). However, development of NTS up to early and late LiS was only 304 observed at higher concentrations (single early LiS starting in 5% HSe and late in 20% HSe) of 305 HSe (Fig 3C, D and S4 Fig) of PZQ (Fig 4A, B and S5 Fig A) OXM (Fig 4C, D and S5 Fig B), MFQ (Fig 4E, F  concentrations. In particular, MFQ had a clear and strong anti-schistosomal activity at 100 348 µg/ml. In contrast to MFQ, OXM was also potent at 1 µg/ml. even more so than at 10 µg/ml 349 (Fig 4C, E). MFQ showed similar activity on the SkS and LuS compared with the LiS (Fig 4E, (Fig 4G, H), but a slight reduction in viability of the 364 SkS at 100 µg/ml (S5 Fig D). On day 3 a.t., we could not detect a drop in viability or any 365 morphological changes in any of the tested stages in HSe-supplemented medium (S6 Fig) or in 366 serum-free medium; however, we could detect a clearly schistosomicidal effect at 100µg/ml, 367 with the death following an initial paralysis of the parasites. However, upon addition of the drug 368 to serum free-culture, the previously solved drug precipitated to a small extent (Fig 5D). Taken  human blood cells to generate advanced larval and juvenile worm stages [16,38]. Viability 421 scoring relies on visual assessment of the larvae in culture by bright field microscopy, which 422 still is the gold standard for drug efficacy tests [23,24]. While the method has its merits, and 423 can be highly effective for dedicated drug efficacy testing, it is very labor intensive. At the same 424 time, microphotographic-based automated analysis [39] using algorithms is complicated by the 425 presence of large numbers of RBCs and/or PBMCs that overlay or colocalize with larvae, 426 making reliable assessment of tegument damage, for example, something for the "trained eye" 427 only. Moreover, the repeated addition of fresh cells from human donors is a factor that is 428 difficult or impossible to fully standardize.

429
In this study, we generated a novel cell-free in vitro assay that allows the development and with other in vitro assays [16,24] something that is also reflected in the limited increase in toxicity we observed in non-serum 23 497 supplemented cultures. Importantly, our assay allows researchers to continually observe 498 toxicity effects and inhibition of maturation from early developmental stages towards mature 499 larval and juvenile worms in settings ranging from compound or drug testing, but also to screen 500 for natural factors from non-permissive hosts or, reversely, identify growth-promoting 501 compounds from the host-adapted, parasite friendly environment.

502
Taken together, juvenile worms generated in our cell-free assay behave and respond similarly 503 to ex vivo harvested worms from infected laboratory animals in drug screening assays [24]. 504 However, in contrast to the limited numbers of labor-intensive, in vivo generated parasites, our 505 in vitro culture method allows for the generation of good numbers of parasites at all stages of 506 development for large-scale drug screening assays. Also, the independency of host blood cells 507 facilitates automated assessment of larval viability in large-scale assays, due to a lack of visual 508 interference by host cells. In addition, the high level of standardization and minimization of 509 host-to-host variability in the assay, will allow researchers to use it to investigate and identify 510 components within HSe that are exploited by the parasite for its development in the dominant 511 definite human host and define mechanisms that underlie the host-specificity of this parasite.

512
Ultimately, such understanding will pave the way for the identification of new drug and 513 vaccination targets. Acknowledgements 516 We would like to thank Prof. Ulm for advice concerning the statistical analysis of the data, 517 Nermina Vejzagic, PhD for thorough reading and helpful advice concerning the manuscript and 518 Laura Hunt for careful proof-reading of the manuscript.

537
Photomicrographs were taken at indicated time points. For skin, lung and day 21 liver stage, a 538 10x magnification was used and for the day 35 liver stage, a 5x magnification. 26 542 NTS were transformed as described before and kept in HM alone or supplemented with 1%, 543 5%, 10%, 20% and 50% of human serum. Photomicrographs were taken with a 10x 544 magnification on day 35 post transformation. Arrowheads indicate dead NTS. Arrows indicate 545 LiS (early LiS (5% and 10%) and late liver stage (20% and 50%)).