Histone deacetylase 8 interacts with the GTPase SmRho1 in Schistosoma mansoni

Background Schistosoma mansoni histone deacetylase 8 (SmHDAC8) has elicited considerable interest as a target for drug discovery. Invalidation of its transcripts by RNAi leads to impaired survival of the worms in infected mice and its inhibition causes cell apoptosis and death. To determine why it is a promising therapeutic target the study of the currently unknown cellular signaling pathways involving this enzyme is essential. Protein partners of SmHDAC8 were previously identified by yeast two-hybrid (Y2H) cDNA library screening and by mass spectrometry (MS) analysis. Among these partners we characterized SmRho1, the schistosome orthologue of human RhoA GTPase, which is involved in the regulation of the cytoskeleton. In this work, we validated the interaction between SmHDAC8 and SmRho1 and explored the role of the lysine deacetylase in cytoskeletal regulation. Methodology/principal findings We characterized two isoforms of SmRho1, SmRho1.1 and SmRho1.2. Co- immunoprecipitation (Co-IP)/Mass Spectrometry (MS) analysis identified SmRho1 partner proteins and we used two heterologous expression systems (Y2H assay and Xenopus laevis oocytes) to study interactions between SmHDAC8 and SmRho1 isoforms. To confirm SmHDAC8 and SmRho1 interaction in adult worms and schistosomula, we performed Co-IP experiments and additionally demonstrated SmRho1 acetylation using a Nano LC-MS/MS approach. A major impact of SmHDAC8 in cytoskeleton organization was documented by treating adult worms and schistosomula with a selective SmHDAC8 inhibitor or using RNAi followed by confocal microscopy. Conclusions/significance Our results suggest that SmHDAC8 is involved in cytoskeleton organization via its interaction with the SmRho1.1 isoform. The SmRho1.2 isoform failed to interact with SmHDAC8, but did specifically interact with SmDia suggesting the existence of two distinct signaling pathways regulating S. mansoni cytoskeleton organization via the two SmRho1 isoforms. A specific interaction between SmHDAC8 and the C-terminal moiety of SmRho1.1 was demonstrated, and we showed that SmRho1 is acetylated on K136. SmHDAC8 inhibition or knockdown using RNAi caused extensive disruption of schistosomula actin cytoskeleton.


Background
Schistosoma mansoni histone deacetylase 8 (SmHDAC8) has elicited considerable interest as a target for drug discovery. Invalidation of its transcripts by RNAi leads to impaired survival of the worms in infected mice and its inhibition causes cell apoptosis and death. To determine why it is a promising therapeutic target the study of the currently unknown cellular signaling pathways involving this enzyme is essential. Protein partners of SmHDAC8 were previously identified by yeast two-hybrid (Y2H) cDNA library screening and by mass spectrometry (MS) analysis. Among these partners we characterized SmRho1, the schistosome orthologue of human RhoA GTPase, which is involved in the regulation of the cytoskeleton. In this work, we validated the interaction between SmHDAC8 and SmRho1 and explored the role of the lysine deacetylase in cytoskeletal regulation.

Methodology/principal findings
We characterized two isoforms of SmRho1, SmRho1.1 and SmRho1.2. Co-immunoprecipitation (Co-IP)/Mass Spectrometry (MS) analysis identified SmRho1 partner proteins and we used two heterologous expression systems (Y2H assay and Xenopus laevis oocytes) to study interactions between SmHDAC8 and SmRho1 isoforms.
To confirm SmHDAC8 and SmRho1 interaction in adult worms and schistosomula, we performed Co-IP experiments and additionally demonstrated SmRho1 acetylation using a Nano LC-MS/MS approach. A major impact of SmHDAC8 in cytoskeleton organization was documented by treating adult worms and schistosomula with a selective SmHDAC8 inhibitor or using RNAi followed by confocal microscopy.

Conclusions/significance
Our results suggest that SmHDAC8 is involved in cytoskeleton organization via its interaction with the SmRho1. 1

Introduction
Schistosomiasis is a Neglected Tropical Disease and represents the second most important human parasitic disease after malaria [1,2]. It is caused by flatworm parasites of the genus Schistosoma and more than 200 million people are infected in 76 countries [3]. Treatment and control of the disease depends on the only available drug, Praziquantel (PZQ). PZQ is effective specifically on adult worms and against the three major species of schistosomes infecting humans (S. mansoni, S. haematobium and S. japonicum [4]). However, its massive administration in endemic areas, in monotherapy, has promoted the emergence of PZQ-tolerant parasites, rendering more likely the risk of resistance [5][6][7][8], and the development of new therapeutic agents imperative. Lysine deacetylases (KDACs) also called Histone deacetylases (HDACs) form a family of enzymes that are conserved in metazoans [9]. They are attractive therapeutic targets because they are involved in the regulation of gene transcription and are already actively studied as drug targets in other pathologies, particularly cancer [10]. Our previous studies identified and characterized three class I HDACs in S. mansoni: HDAC 1, 3 and 8 [11] and we have shown that Trichostatin A (TSA), a pan-inhibitor of HDACs, induces hyperacetylation of histones, deregulates gene expression and causes the death of schistosome larvae and adult worms in culture [12]. Moreover, reduction of SmHDAC8 transcripts by RNAi led to the impaired survival of the worms after the infection of mice, showing that it is a valid therapeutic target [13]. The analysis of the 3D structure of SmHDAC8 by X-ray crystallography [13] showed that the catalytic domain structure diverged significantly from that of the human HDAC8 and this was exploited to identify selective inhibitors that induce apoptosis and death of worms and are thus lead compounds for the development of novel anti-schistosomal drugs [14]. However, the precise biological roles of schistosomal HDAC8 were unknown and in order to determine why SmHDAC8 knockdown or inhibition causes cell apoptosis and worm death, it was essential to study the cellular signaling pathways involving SmHDAC8. Potential binding partners of SmHDAC8 have been characterized by screening a yeast two-hybrid cDNA library and mass spectrometry analysis [15]. Among the potential partners identified, the schistosome orthologue of the human GTPase RhoA, SmRho1, indicated a potential role for SmHDAC8 in cytoskeleton organization [15].
Rho GTPases (Ras homologous) belong to the superfamily of small Ras (Rat Sarcoma) monomeric G proteins that are extremely conserved in eukaryotes [16]. They are able to create a switch between an active GTP-bound conformation and an inactive GDP-bound conformation. The "ON/OFF" activity of Rho GTPase is controlled by various regulatory proteins: the guanine nucleotide exchange factors (GEFs) induce the exchange of GDP for GTP; the GTPase-activating proteins (GAPs) promote the hydrolysis of GTP to GDP and the GDP dissociation inhibitors (GDIs) inhibit the dissociation of GDP from the GTPase [17]. Activation of Rho GTPase by GEFs transduces signals to various effector molecules, while remaining in the GTP-bound form, hence regulating various cell functions through reorganization of the actin cytoskeleton such as the formation of stress fibers and focal adhesions [18]. In S. mansoni, SmRho1 was identified and presents 71-75% identity to human RhoA, B, and C GTPases. It was also able to complement a Rho1 null mutation in the budding yeast Saccharomyces cerevisiae [19] and could play a role in actin cytoskeleton regulation in the gonads of adult worms [20]. Although HDAC8 is known to interact with specific cytoskeleton components [21], a direct link between HDAC8 and a Rho GTPase has never been shown in humans. Moreover, RhoA was not identified by Olson et al. as a partner of human HDAC8 [22]. Interestingly, RhoA does not seem to be acetylated in humans, although its orthologue was found to be acetylated in S. japonicum [23].
In the present study, two closely related isoforms of SmRho1, which we named SmRho1.1 and SmRho1.2, were characterized at the molecular and functional levels. Based on these data, we focused on the interaction between SmRho1 isoforms and SmHDAC8. Studies in the parasite showed that SmHDAC8 interacts with SmRho1. Using mass spectrometry analysis, we identified an acetylated lysine on SmRho1 in adult worms. Co-expression of SmRho1 isoforms and SmHDAC8 in X. laevis oocytes and in yeast shows a specific interaction between SmRho1.1 and SmHDAC8 that implicates the SmRho1.1 C-terminal domain. In the same way, we showed that SmRho1.2 co-immunoprecipitates with a Diaphanous homolog SmDia [24] and not with SmHDAC8. Finally, the use of selective inhibitors and RNAi followed by confocal microscopy revealed that SmHDAC8 is involved in the regulation of the actin cytoskeleton organization in adult worms and in schistosomula. Hence, here we demonstrate, for the first time, the role of SmHDAC8 in modulating the organization of the schistosome cytoskeleton, possibly via the SmRho GTPase signaling pathway.

Parasite material
The NMRI strain of S. mansoni is maintained in the laboratory using the intermediate snail host Biomphalaria glabrata and the definitive golden hamster host Mesocricetus auratus. S. mansoni adult worms were obtained by hepatic portal perfusion of hamsters infected six weeks before. Cercariae were released from infected snails, harvested on ice as described in [25] and schistosomula were obtained in vitro by mechanical transformation [25].

Frog and oocytes handling
X. laevis females, obtained from the CRB-University of Rennes (France), were anesthetized with tricaine methane sulfonate (MS222, Sandoz) at 1 gL -1 for 45 min. After surgical removal of ovaries, stage VI oocytes were harvested by using a 1 h collagenase A treatment (1 mg mL 1 , Boehringer Mannheim) for 45 minutes followed by manual dissociation in ND96 medium (96 mM NaCl, 2 mM KCl, 1 mM MgCl 2 , 1.8 mM CaCl2, 5 mM HEPES, adjusted to pH 7.5 with NaOH). Oocytes were kept at 19˚C for 2h.

Molecular cloning of SmRho1 isoforms
Total RNA from adult worms was isolated using the RiboPure RNA Purification Kit (Thermofisher Scientific). The cDNA was prepared using the GeneRacer kit with SuperScript III reverse transcriptase (Invitrogen) following the manufacturer's instructions. The 5 'and 3' ends of SmRho1.1 and SmRho1.2 were determined by RACE PCR using the primers listed in S1 Table (Rho GTPase 5'1 / 5'2 and 3'1 / 3'2) generated from the cDNA sequences obtained by Y2H screening (15) and amplified products were subcloned into the vector pCR4-TOPO (Thermofisher Scientific) and sequenced (Eurofins Genomics). The full length SmRho1 isoform sequences were amplified using FLRho1 5' and 3' primers and inserted into the pCR4-TOPO vector. A further PCR experiment was then carried out, using primers containing the NdeI and BamHI restriction sites respectively and the obtained fragment was again inserted into the pCR4-TOPO vector.

Phylogenetic analysis and protein modeling
Phylogenetic analysis of eukaryotic Rho GTPases was performed using protein sequences from vertebrates, helminths (nematodes, cestodes, turbellariates), insects, molluscs and yeasts (S2 Table). The sequences were aligned using the BioEdit program using the ClustalW method [26]. The phylogenetic tree was generated by the MEGAX software under the JTT + I + G model with 1 000 bootstraps [27] and the maximum likelihood method. The modeling of the two isoforms of SmRho1 was performed using the I-Packer server (https://zhanglab.ccmb. med.umich.edu/I-TASSER/) and compared to the structure of human RhoA [28]. The characteristic domains and the different residues between SmRho1.1, SmRho1.2 and Homo Sapiens RhoA (HsRhoA) were highlighted using PYMOL software [29].

Expression and purification of the SmRho1.1 and SmRho1.2 recombinant proteins
SmRho1.1-pGADT7 and SmRho1.2-pGADT7 constructs were cut by NdeI and BamHI. Sequences were inserted in frame into the pnEA-tH plasmid [13] (a kind gift from Dr. M. Marek and Dr. C. Romier, IGBMC, Strasbourg, France, which codes for a polyhistidine tag in the N-terminal position followed by a thrombin site), using T4 DNA Ligase (Invitrogen).

Production of anti-SmRho1 antibodies
Purified recombinant SmRho1.1 was used to generate mouse polyclonal antiserum. BALB/c mice were injected i.p. with 50 μg of SmRho1.1 and alum adjuvant in a total volume of 500 μL, three times at two-week intervals. The mice were bled two weeks after the final injection.  Table).
In order to further establish the specificity of the anti-SmRho1 antibodies, we performed an immunodepletion assay (S2D Fig). Briefly, 100 μg of SmRho1.1 recombinant protein were incubated with 10 μL of SmRho1 antiserum for 2h at room temperature on a rotating wheel. After TRIzol extraction (Invitrogen), 20 μg of protein extracts from adult worms and schistosomula were separated on a 10% (v/v) SDS-polyacrylamide gel and blotted on to a nitrocellulose membrane. Blots were developed with a mouse polyclonal anti-SmRho1 antibody (1/ 1000) or mouse anti-SmRho1 depleted serum and peroxidase coupled anti-mouse secondary antibody (1/50 000; Invitrogen). An antibody against human RhoA (Rabbit anti-HA protein tech group cat: PTG10749-1-Au, 1/1000) was used as a negative control.
Detection was carried out by chemiluminescence using SuperSignal West Dura Extended Duration Substrate (Thermo Scientific) and ImageQuant LAS 4000 imager (GE Healthcare). Results showed that only the major band at 22 kDa was depleted in the assay, supporting the recognition of SmRho1 by the mouse antiserum.

CoIP for Nano-LC-MS/MS analysis
For mass spectrometry analysis, two independent experiments were performed. Adult worms (100 couples) were suspended in 500 μL of lysis buffer (20 mM Tris-HCl pH 7.4, 50 mM NaCl, 5 mM EDTA, 1% Triton and protease inhibitors (20 μM leupeptin (Sigma), 2 μg mL -1 aprotinin (Sigma), 200 μM phenylmethylsulfonyl fluoride (Sigma), crushed with a Dounce homogenizer and sonicated ten times for 30 s (maximum power, Bioruptorplus, Diagenode). After centrifugation at 1000 x G for 10 min at 4˚C, immunoprecipitation of SmRho1 was performed using the Pierce Crosslink Immunoprecipitation Kit (Thermo Scientific) according to the manufacturer's instructions. Briefly, the protein lysate (500 μL) was pre-cleared by incubation with 20 μL of IgG from rat serum crosslinked to protein-L Agarose beads (5 mg of beads, Thermo Scientific) for 2h at 4˚C on a rotator. Then, pre-cleared lysate was collected after centrifugation, at 1000 x G for 1 min at 4˚C, and incubated overnight at 4˚C on a rotator, with 10 μL of anti-SmRho1 antibodies or 10 μL of IgG from mouse serum as a control, bound to protein-L Agarose beads (5 mg of beads, Thermo Scientific).

Mass-spectrometry proteomic analysis
Protein samples were denatured at 100˚C in 5% SDS, 5% β-mercaptoethanol, 1 mM EDTA, 10% glycerol, and 10 mM Tris pH 8 buffer for 3 min, and subsequently fractionated on a 10% acrylamide SDS-PAGE gel. The gel was stained briefly with Coomassie Blue. Five bands, containing the whole sample, were cut out. Digestion of proteins in the gel slices was performed as previously described [30]. Separation of the protein digests was carried out using an UltiMate 3000 RSLCnano System (Thermo Fisher Scientific). Peptides were automatically fractionated onto a commercial C18 reversed phase column (75 μm × 150 mm, 2 μm particle, PepMap100 RSLC column, Thermo Fisher Scientific, temperature 35˚C). Trapping was performed during 4 min at 5 μL min -1 , with solvent A (98% H 2 O, 2% ACN (acetonitrile) and 0.1% FA (Formic Acid)). Elution was carried out using two solvents, A (0.1% FA in water) and B (0.1% FA in ACN) at a flow rate of 0,3 mL/min. Gradient separation was 3 min at 5% B, 37 min from 5% B to 30% B, 5 min to 80% B, and maintained for 5 min. The column was equilibrated for 10 min with 5% buffer B prior to the next sample analysis. Peptides eluted from the C18 column were analyzed by Q-Exactive instruments (Thermo Fisher Scientific) using an electrospray voltage of 1.9 kV, and a capillary temperature of 275˚C. Full MS scans were acquired in the Orbitrap mass analyzer over the m/z 300-1200 range with a resolution of 35 000 (m/z 200) and a target value of 5.00E + 05. The ten most intense peaks with charge state between 2 and 4 were fragmented in the HCD collision cell with normalized collision energy of 27%, and tandem mass spectra were acquired in the Orbitrap mass analyzer with resolution 17,500 at m/z 200 and a target value of 1.00E+05. The ion selection threshold was 5.0E+04 counts, and the maximum allowed ion accumulation times were 250 ms for full MS scans and 100 ms for tandem mass spectrum. Dynamic exclusion was set to 30 s. Raw data collected during nanoLC MS/MS analyses were processed and converted into � .mgf peak list format with Proteome Discoverer 1.4 (Thermo Fisher Scientific). MS/MS data were interpreted using search engine Mascot (version 2.4.0, Matrix Science, London, UK) installed on a local server. Searches were performed with a tolerance on mass measurement of 0.2 Da for precursor and 0.2 Da for fragment ions, against a composite target decoy database (25 970 total entries) built with the S. mansoni Uniprot database (taxonomy id 6183, 12 861 entries) fused with the sequences of recombinant trypsin and a list of classical contaminants (124 entries). Up to one trypsin missed cleavage was allowed. For each sample, peptides were filtered out according to the cut-off set for protein hits with one or more peptides longer than nine residues, an ion score >30, an identity score >6, leading to a protein false positive rate of 0.8%.

IP and WB analyses of proteins expressed in adult S. mansoni
Adult worms (100 couples) and schistosomula (1000 parasites) were suspended in lysis buffer (20 mM Tris-HCl pH 7.4, 50 mM NaCl, 5 mM EDTA, 1% Triton and protease inhibitors: (20 μM leupeptin (Sigma), 2μg mL -1 aprotinin (Sigma), 200 μM PMSF (Sigma) crushed with a Dounce homogenizer and sonicated ten times for 30 s (maximum power, Bioruptoplus, Diagenode). After centrifugation, at 10 000 x G for 10 min at 4˚C, immunoprecipitation of SmHDAC8 and SmRho1 was performed using the Pierce Crosslink Immunoprecipitation Kit (Thermo Scientific) according to the manufacturer's instructions. Briefly, protein lysate (500 μL) was incubated with 10 μL of mouse polyclonal anti-SmHDAC8 or mouse polyclonal antiSmRho1 crosslinked to protein-L Agarose beads (Thermo Scientific) overnight at 4˚C on a rotating wheel. As a negative control, a Co-IP with a mouse IgG antibody alone was performed. Proteins were separated on a 10% (v/v) SDS-polyacrylamide gel and blotted on to a nitrocellulose membrane. Blots were developed with a mouse polyclonal anti-SmHDAC8 antibody (1/1000) or mouse polyclonal anti-SmRho1 antibody (1/1000) and peroxidase coupled anti-mouse secondary antibody (1/50 000; Invitrogen). Detection was carried out by chemiluminescence using SuperSignal West Dura Extended Duration Substrate (Thermo Scientific) and ImageQuant LAS 4000 imager (GE Healthcare).

Yeast two hybrid assay
The S. cerevisiae Y187 strain was transformed with the SmHDAC8-pGBKT7 bait construct and mated overnight with the AH109 strain transformed with the SmRho1 construct. After incubation, diploid yeasts were plated on selective medium lacking leucine and tryptophan and then on selective medium lacking adenine, histidine, leucine and tryptophan and the plates were incubated at 30˚C.

RNAi-mediated knockdown of SmHDAC8 and SmRho1
The SmHDAC8 specific PCR primers containing the T7 promoter-tail amplified~500 bp fragments (S1 Table). Similarly, the SmRho1 specific PCR primers containing the T7 promoter tail amplified a~500 bp fragment (S1 Table). A luciferase nonspecific~500 bp control was used (S1 Table). Double-stranded RNAs (dsRNAs) were synthesized in vitro from adult worm cDNA using the Megascript RNAi kit (Ambion) according to the manufacturer's instructions and concentrations were determined spectrophotometrically (NanoVue Plus, GE Healthcare). dsRNAs were also analyzed by 1% agarose electrophoresis to check the integrity and annealing of the dsRNA transcripts. To deliver the dsRNA, 10 adult couples/group in 100 μL M199 medium containing 25 μg dsRNA, were electroporated in a 4 mm cuvette by applying a square wave with a single 20 ms impulse, at 125 V and at room temperature [31]. Adult worms were then transferred to 4 mL complete M199. After two days in culture, 2 mL of medium was removed and 2 mL of fresh complete M199 culture medium was added. Gene knockdown was monitored by qRT-PCR 5 days after dsRNA treatment. For RNAi experiments on schistosomula, dsRNA delivery was performed using the soaking method. 10 μg of dsRNA was added to 2000 parasites in 4 ml complete M199 medium and after two days in culture, gene knockdown was monitored by qRT-PCR. Microscopic examination of RNAi-treated worms was carried out exactly as described below.

Quantitative RT-PCR
Complementary DNAs were obtained by reverse transcription of total RNA using the NucleoSpin RNA/Protein kit (Macherey-Nagel) and used as templates for PCR amplification using Brilliant III Ultra-Fast QPCR Master Mix (Agilent) and QuantStudio 3 Real-Time PCR System (Applied Biosystems). Primers specific for S. mansoni HDAC8, Rho1.1 and Rho1.2 were designed by the Primer BLAST tool (NCBI) and used for amplication in duplicate assays (S1 Table). Measurements of real time PCR efficiency for each primer pair allowed the ratios of expression to be calculated using the 2 -DDCt ratio with S. mansoni PSMB7 (Smp_073410) as the reference transcript [32]. Each sample was run in duplicate and the experiment was performed once. The results were expressed as means ± S.D.
After treatment, parasites were fixed in 8% PFA (paraformaldehyde)-CB buffer solution Cytoskeletal Buffer: 10 mM Hepes, 150 mM NaCl, 5 mM EGTA, 5 mM glucose, 5 mM MgCl2, pH 6,1), for 1h at room temperature. The fixed parasites were then incubated with CB buffer containing 0.05% saponin, NH4Cl (50 mM) and phalloidin (Alexa Fluor 488, Thermo Fischer Scientific, at 1/1000 dilution) overnight at room temperature to stain actin filaments. Three washes were performed in CB buffer and DAPI (1/1000 dilution, Thermo Fischer Scientific) was added during the third wash for 10 min at room temperature. Mowiol (Calbiochem) was used as mounting solution. Samples were pictured with a Zeiss LSM 880 confocal line scanning microscope (Zeiss microscopy GmbH, Germany) using an Airyscan detector and 63x oil immersion lens to obtain high-resolution images (voxel 0.35x0.35x0.30 μm 3 ). Images were processed using Zen software (Version2, Zeiss, France) for Airyscan processing. The visualization of the samples in depth were obtained by maximum intensity projection on an adapted selection of Z planes of the Z-stack and using orthogonal views created with FIJI (version: 2.1.1/1.53c).

Identification and characterization of the two SmRho1 isoforms
Among the Rho family of small proteins (~21 kDa) the Ras homolog family, member A, (RhoA), Ras-related C3 botulinum toxin substrate 1 (Rac1) and Cell Division Control protein 42 homolog (Cdc42) are the most studied members. Santos and coworkers [34] characterized Rho1 in S. mansoni, showing that it is related to the vertebrate RhoA, B, C subfamily and that SmRho1 complemented an S. cerevisiae Rho1 null strain. The SmRho family was further characterized by Vermeire and coworkers [19] and they identified S. mansoni orthologs of RhoA, CDC42 and Rac1. Interestingly, using the yeast two-hybrid methodology and mass spectrometry analysis, we recently identified the S. mansoni Ras homolog 1 (SmRho1), as a SmHDAC8 partner protein [15]. Moreover, after a more detailed reconstruction of the SmRho1 transcript using RT-PCR, we discovered that two distinct genes encode Rho1 isoforms: Rho1.1 and 1.2. These two genes code for two proteins of 193 residues organized into several domains, as for all Rho GTPases (Fig 1A). ClustalW alignment shows 96.37% identity between the two encoded proteins that differ only in 7 aa, and they share respectively 72% and 73% identity with the Homo sapiens RhoA (HsRhoA). The Rho GTPases contain five GTPase domains named "G boxes" (G1-G5) (in blue) and two functional elements, Switch I and II (boxed in green and yellow, respectively), which can interact with many regulators (GEFs, GAPs and GDIs) and effectors ( Fig 1A). SmRho1.1 and SmRho1.2 have an identical sequence in the switch I domain (residues 29-38 in RhoA) except for the positions 33 where the glutamic acid in SmRho1.2 is replaced by a lysine residue in SmRho1.1 (Fig 1A and 1B). As expected, the G domains and Switch II domain are completely conserved. At their C-termini both SmRho1.1 and SmRho1.2 contain a conserved prenylation (CDLI) CAAX motif (C represents cysteine, A is an aliphatic aa and X is a terminal aa) sequence. This motif is preceded by a run of basic aa (KKKRRR) which determine cellular localization for each protein [35], also typically found in Rho GTPases.
Using the threading method to predict SmRho1 isoform structures, we observed a conserved folding between SmRho1 isoforms and human RhoA consisting in a six-stranded βsheet surrounded by α-helices connected with loops, as found in human RhoA and other related small GTPases (Fig 1B and 1C). The β-sheet is formed by the anti-parallel association of two extended β-strands (B2 and B3) and the parallel association of five extended β-strands (B3, B1, B4, B5, B6) (Fig 1C). SmRho1.1 contains six α-helices (A1, A2, A3, A4, A5 and A6) 2 contain five GTPase domains named "G boxes" (G1-G5) (in blue) and two functional elements, Switch I and II (boxed in green and yellow, respectively), which can interact with many regulators (GEFs, GAPs and GDIs) and effectors. The insert region (in red) is essential for Rho GTPase functions [37]. The C-terminal part presents a hypervariable domain and a prenylation motif CAAX (C = cysteine residue, A = aliphatic residue, X = terminal aa). (B) Sequence alignment between the S. mansoni Rho1 isoforms, S. mansoni Rho2 (Uniprot G4V9A8) and Rho3 (Uniprot Q8I8A0), S. japonicum Rho1 (Uniprot Q8MUI8), RhoA of S. haematobium (Uniprot A0A094ZFT0), human RhoA (Uniprot P61586) and mouse RhoA (Uniprot Q9QUI0). Sequences were aligned using ClustalW Multiple Alignment (SnapGene). The identical and semi-conserved aa are highlighted in black and gray respectively. Residues not conserved between SmRho1.1 and SmRho1.2 are boxed in orange. SmRho1.1 and SmRho1.2 both share high sequence similarity with HsRhoA (72% et 73% respectively) The secondary structure elements of SmRho1 isoforms and HsRhoA are indicated at the top and below the aligned sequences, respectively with A for α helices, B for β sheets and H for 3 10 helices. (C) Modeling of the tertiary structure of SmRho1 isoforms and HsRhoA [28]. Shown is a ribbon representation of SmRho1 isoforms and HsRhoA with β-strands (red), α-helices (cyan), and 310-helices (green). (D) Surface representation of proteins structures with Switch I and II (green and yellow, respectively), G boxes (in blue). For SmRho1 isoforms, the CAAX prenylated site is shown in purple and the acetylated lysine in cyan. The aa sequence differences between SmRho1 isoforms and HsRhoA are shown in red.
https://doi.org/10.1371/journal.pntd.0009503.g001 and two 3 10 -helices (H1-H2) whereas SmRho1.2 contains only one 3 10 helix (H1) like human RhoA (Fig 1C). Three-dimensional models of the structures of SmRho1.1 and SmRho1.2 indicate that mutated aa (in red) are located on the surface of the proteins and form potential interaction domains for their partners (Fig 1D). The comparison with the HsRhoA threedimensional model shows that differences in aa sequences do not affect the overall structure of this potential interaction surface (Fig 1D).
Several studies have resulted in the characterization of a number of Rho members subdivided into 8 subfamilies: Rho, Rac, Cdc42, RhoD/Rif, Rnd, Wrch-1/Chp, RhoH and RhoBTB [36] (Fig 2). In order to understand the phylogenetic relationships within the schistosome Rho family and within metazoans, we constructed a phylogenetic tree including the subfamilies of the Rho GTPases and Ras superfamily with mitochondrial Rho GTPases as an outgroup. Aa sequences from vertebrates, insects, nematodes, cestodes, trematodes and ascomycetes were included in this analysis. Phylogenetic analysis showed that schistosomes have a low number of orthologs of the main mammalian Rho subfamilies. We initially identified 7 Rho-like proteins in S. mansoni, but only 4 in S. haematobium and in S. japonicum. We found that schistosome Rho, Rac and Cdc42 clustered together with all Rho, Rac and Cdc42 orthologues (Fig 2, Red, green and purple clusters).
Maximum Likelihood (ML) analysis was also performed to characterize the phylogenetic positions of the recently discovered SmRho1.1 and SmRho1.2 Rho proteins. These proteins, cluster together with high fidelity (bootstrap = 99) inside the RhoA, B, C subfamily, indicating that SmRho1.1 and SmRho1.2 are probably paralogous genes that are orthologs to the human RhoA, B, C family. S. haematobium and S. japonicum do not seem to have undergone the same duplication, suggesting that the SmRho1 duplication is recent. Moreover, the phylogenetic analysis indicates probable conserved functions between the vertebrate and platyhelminth proteins (Fig 2, Red cluster).

Biological functions of the two SmRho1 isoforms
Two independent Co-IP experiments were performed (named IP1 and IP2), using an anti-SmRho1 antibody which we produced in house in mice (S2 Fig) with pre-immune sera as control. MS analysis of the Co-IP proteins identified 1,000 different proteins (S3 Table). As expected, we demonstrated that our mouse anti-SmRho1 antibodies were not able to discriminate between the two SmRho1 isoforms (S1 Fig). We also noted a high degree of variation between identified proteins of each Co-IP experiment that can be explained by difficulties in trapping dynamic protein complexes, which potentially depend on post-translational modifications and GTP/GDP levels. In addition, S. mansoni is a complex multicellular parasite and protein quantities can vary between the different cellular types within a given worm as well as between different parasites. Finally, it should be noted that the two parasite protein extracts were each obtained from a pool of S. mansoni adult worms couples. Of the 1,000 proteins for which peptides were detected, we retained 86 and 32 proteins from IP1 and 2 respectively. Proteins that completed the three following criteria were considered: (i) at least three peptides in the Co-IP experiment, (ii) with no more than two peptides in the control and (iii) with a spectral count ratio between Co-IP SmRho1 and control of greater than 3.
Several regulatory partners were identified in co-IP with SmRho1, for instance the RhoGDI protein (Rho GDP-dissociation inhibitor-related, Accession number G4VK76; Smp_045610.1), a negative regulator of Rho GTPase activity [38], suggesting that this regulatory mechanism for Rho GTPase activity is conserved in S. mansoni. Other regulatory proteins like GEFs or GAPs were absent from the SmRho1 interactome in our experiments. However, we did detect an interaction between SmRho1 and SmRab6 (Putative Rab6, Accession number G4LXF1; Smp_163580) that could suggest the existence of a connection between the signaling of Rab proteins with SmRho1 during membrane trafficking, probably via the GEF regulator ARHGEF10, which in humans interacts with both Rab6A and Rab8A [39].
We have also identified Smkinesin (Kinesin-like protein, Accession number G4V5R8; Smp_001040), Smspectrin (Putative spectrin beta chain, Accession number G4VDE6; Smp_143470.1) and Smankyrin (Putative ankyrin, Accession number G4VKA7; Smp_145700) as SmRho1 partners. The cytoskeletal element, Smα-actinin (Putative alpha-actinin, Accession number G4VBW4; Smp_014780) was found in the SmRho1 interactome. α-actinins form a family of cytoskeletal actin-binding proteins playing crucial roles in organizing the framework of the cytoskeleton through crosslinking the actin filaments, as well as in focal adhesion maturation. In conclusion, our results show that SmRho1 is involved in cytosolic processes regulating cytoskeleton function, as is the human ortholog, HsRhoA. Despite the fact that we did not identify all the known protein partners of human RhoA, the identification of the different components of the cytoskeleton such as Smspectrin, Smα-actinin or Smcofilin suggests that signaling pathways involving SmRho1 are conserved in S. mansoni. However, since we cannot discriminate between the two SmRho1 isoforms using our MS analysis, no conclusions can be drawn concerning the specific roles of each SmRho1 isoform.

SmRho1 interacts with SmHDAC8 and is acetylated on lysine K136 in S. mansoni
To confirm in vivo the direct interaction between SmHDAC8 and SmRho1 shown by Y2H screening (15), we performed CoIP experiments on adult worms ( Fig 4A) and schistosomula ( Fig 4B). Using anti-SmHDAC8 and SmRho1 antibodies, endogenous proteins were reciprocally and mutually immunoprecipitated and identified by Western Blotting. This result is consistent with protein-protein interaction between SmHDAC8 and SmRho1 in S. mansoni parasites (Fig 4). Nano LC-MS/MS analysis showing a mass increment of 42 Da corresponding to the presence of an acetyl group on lysine 136 of the TKQLPVTFNEGK peptide of SmRho1 was observed (S3 Fig). This may suggest that SmRho1 could be a substrate for SmHDAC8, but this remains to be investigated.

SmRho1.1 C-terminal moiety binds SmHDAC8
To investigate the SmHDAC8-SmRho1isoforms interaction in more detail, we used the Y2H system in yeast and X. laevis oocytes as heterologous expression systems (Fig 5).
In Fig 5 (right panel), we show that only diploid yeasts which expressed SmHDAC8 and SmRho 1.1 could grow on the selective medium SD -Leu/-Trp/-His/-Ade, confirming an interaction between these two proteins. Moreover, the absence of growth for diploids expressing SmHDAC8 and SmRho1.2 suggested that the interaction between SmHDAC8 and SmRho1.1 was specific. We next co-expressed SmHDAC8 and SmRho1 isoforms in X. laevis oocytes ( Fig  5B) to confirm the specific interaction between SmHDAC8 and SmRho1.1 (Fig 5A). We showed that Myc-tagged SmHDAC8 binds HA-tagged SmRho1.1 but not HA-tagged SmRho1.2.
Based on the specificity of interaction between SmHDAC8 and SmRho1.1 and in order to determine the segments of SmRho1.1 responsible for this interaction, various SmRho1 mutant proteins were produced by site-directed mutagenesis. Protein sequence alignments of SmRho1.1 and SmRho1.2 showed that there were only 7 different aa between the two isoforms ( Fig 1B). These mutations are mainly located in the C-terminal part of the protein. In the Nterminal part, only the glutamic acid at position 33 of SmRho1.1 is substituted by a lysine in SmRho1.2. Two N-terminal fragments of different sizes were therefore produced, by insertion of premature stop codon, using SmRho1.1 in order to identify the domain binding to SmHDAC8 (Fig 5C). In parallel, two other mutant proteins were produced using both SmRho1.1 and SmRho1.2 as templates. The glutamic acid at position 33 of SmRho1.1 was substituted by a lysine (SmRho1.1 E33K). Similarly, the lysine at position 33 of SmRho1.2 was replaced by a glutamic acid (SmRho1.2 K33E) (Fig 5C). We co-expressed SmHDAC8 and SmRho1 mutants in X. laevis oocytes (Fig 5D) and showed that Myc-tagged SmHDAC8

PLOS NEGLECTED TROPICAL DISEASES
interacts with HA-tagged SmRho1.1E33K but not with SmRho1.2 K33E. These substitutions therefore failed to change the binding behavior, compared to the SmRho1 wild type isoforms.
In addition, we showed that HA-tagged SmRho1 N-terminal fragments cannot interact with Myc-tagged SmHDAC8 (Fig 5D) suggesting that the SmRho1.1 C-terminal moiety is involved in the binding to SmHDAC8. However, it is also possible that the SmRho1.1 fragments could be misfolded, inducing a loss of protein function and their ability to interact with SmHDAC8. In consequence, we performed point mutations using SmRho1.1 and SmRho1.2 to identify specific residues involved in the interaction with SmHDAC8. Using site-directed mutagenesis, the glutamine Q147 and the valine V148 of SmRho1.1 were substituted by a glutamic acid and a methionine (SmRho1.1 EM) and then the lysine K151 and the serine S153 by two asparagines (SmRho1.1 EMNN) (S6A Fig). We also produced SmRho1.2 QV and SmRho1.2 QVKS mutants by site-directed mutagenesis using the SmRho1.2 protein as template (S6A Fig). We then co-expressed the resulting mutant proteins in X. laevis oocytes. CoIP experiments revealed that none of the SmRho1.1 and SmRho1.2 mutants (HA-tagged) were able to bind SmHDAC8 (Myc-tagged) (S6B Fig). In conclusion, the mutations we carried out to SmRho1.2 were incapable of restoring the interaction between SmHDAC8 and SmRho1.2, suggesting that all seven aa differentiating the two isoforms are potentially involved in the interaction with SmHDAC8.

SmRho1.2 interacts specifically with SmDia
Because we have shown a specific interaction between SmHDAC8 and SmRho1.1 in yeast and X. laevis oocytes suggesting that each isoform could have specific partners, we investigated the interaction between SmRho1 isoforms and the Rho Binding Domain of SmDia (SmDia-RBD) (24) using the Y2H system in yeast and X. laevis oocytes.
In Fig 6A, we show that only diploid yeasts that express SmRho1.2 and SmDia-RBD grow on the selective medium SD -Leu/-Trp/-His/-Ade, indicating an interaction between these two proteins. In order to confirm the specific interaction between SmRho1.2 and SmDia-RDB ( Fig  6B), we co-expressed SmRho1 isoforms and SmDia-RBD in Xenopus oocytes. This shows that Myc-tagged SmDia-RBD specifically binds HA-tagged SmRho1.2.

SmHDAC8 inhibition or knockdown cause disruption of the parasite actin cytoskeleton
Finally, to elucidate the role of SmHDAC8 in the regulation of cytoskeleton dynamics, we examined the impact of SmHDAC8 inhibition on the organization of the actin network of the parasite tegument using both RNAi and selective inhibitors (Figs 7 and 8).
Adult worms and schistosomula were first treated with a selective SmHDAC8 inhibitor, TH65 (33), then fixed and Actin F was stained with Alexa488 conjugated phalloidin (green) and nuclei with DAPI (blue) (Figs 7 and 8, S4 and S5 Figs). As a reference, we used parasites treated with Trichostatin A (TSA) which inhibits both class I and II mammalian histone deacetylases (Figs 7A and 8A and S4 Fig). In parallel, we used an RNAi complementary approach to target transcripts encoding SmHDAC8 (Figs 7B and 8C and S5 Fig).
In adult worms, the TH65 inhibitor did not induce a significant disorganization of tegumental actin (Fig 7A and S4 Fig). Indeed, phalloidin, which binds to actin filaments, was detected in spines and tegumental cells and also in subtegumental muscle fibers. Actin filaments appeared as horizontal and vertical straight lines stretching across the whole thickness of the tegumental syncytium ( Fig 7A and S4 Fig). We could, however, observe that TH65 seems to impact the structure of the spines. (Fig 7A and S4 Fig). Moreover, no significant effects were observed in adult worms after treatment with TSA ( Fig 7A and S4 Fig) or Smhdac8 interference (Fig 7B and S5 Fig).
On the contrary, schistosomula treated with TH65 and TSA were highly affected at various levels and, for some of the parasites, we observed a strong phenotype with defective muscular actin organization (Fig 8A). In control parasites (Fig 8A: DMSO section, top and bottom panel), phalloidin staining revealed higher-order actin structures, forming a network which seems to correspond to the muscle fibers of the subtegumental muscle layer. In treated parasites (Fig 9A and 9B, TH65 and TSA section, top and bottom panel), we observed that the actin network structure was disrupted after inhibitor treatment with a modification of actin filament structure or absence of F-actin.
While dsRNA promoted only 22% silencing of the SmHDAC8 gene, in schistosomula this nevertheless led to a significant effect on the actin cytoskeleton organization (Fig 8B and 8C), similar to that observed after inhibitor treatment (Fig 8B). This suggests that the effects of TH65 could be due to selective inhibition of SmHDAC8.
In order to confirm the involvement of SmHDAC8 in the SmRho1 signaling pathway, we treated parasites with Rho inhibitor I (Figs 7A and 8A), used to selectively inactivate the human GTPases RhoA, RhoB, and RhoC by ADP-ribosylation on asparagine 41, which is conserved in SmRho1 isoforms, and we used RNAi to knock down SmRho1 (Figs 7B, 7C and 8B). In adult worms, inhibition and silencing of SmRho1 did not significantly affect actin network organization, but in schistosomula we observed a very similar phenotype to that obtained after inhibition and silencing of SmHDAC8.
Taken together, these findings suggest that SmHDAC8 is involved in regulation of actin cytoskeleton organization in S. mansoni, more evidently in schistosomula. However, the robust silencing and inhibition of SmHDAC8 in adult worms did not result in any significant phenotypic changes. This may be due to a more active turnover of the actin cytoskeleton in the larvae compared to adult worms. It should be noted that TH65 causes inhibition of the deacetylation activity of SmHDAC8 enzyme, but does not affect protein expression.

Discussion
In the present study, we have confirmed SmRho1 as a partner and potential substrate of SmHDAC8 and provided evidence that the latter is involved in the regulation of the actin cytoskeleton in S. mansoni. RNA interference was carried out by S. mansoni worms with dsRNA for SmHDAC8, SmRho1 (positive control) or luciferase (negative control) as described in the Methods section. Actin-F was revealed with phalloidin staining and the nuclei were stained with DAPI (blue). Microscopic examination was carried out 5 days after RNAi treatment. Scale bar represents 20 μm, magnification, x630. Experiments were performed three times on adult worms obtained by three different hepatic portal perfusions of hamsters. (C) RT-qPCR results of RNAi treatment with dsRNA for SmHDAC8 (dsSmHDAC8), SmRho1 (dsSmRho1) or luciferase (dsSmLuc) and analyses of relative transcript levels of SmHDAC8, SmRho1.1 or SmRho1.2 in adult worms. SmPSMB7 was used as an internal reference gene (32). The results were analyzed using the 2 −ΔΔCT method and experiments were carried out once in duplicate. https://doi.org/10.1371/journal.pntd.0009503.g007

PLOS NEGLECTED TROPICAL DISEASES
Overall, the MS approach led to the detection of a significant number of proteins orthologous to human proteins that are involved in the regulation of the organization of the cytoskeleton. The identification of SmRho2 as a partner of SmRho1 is unexpected (S3 Table). It is possible that the presence of SmRho2 in the SmRho1 interactome can be explained by the IP of this isoform using anti-SmRho1 antibodies, probably due to the significant sequence identity between SmRho1 and SmRho2 (72%) (19).
We did not identify SmROCK, GEFs and GAPs, and more surprisingly, SmHDAC8 as SmRho1-interacting proteins. As we have already seen, the dynamics of protein-protein interaction networks or absence of post-translational modifications could explain the absence of the histone deacetylase. Moreover, some proteins identified may be members of immunoprecipitated complexes and are not direct partners and it is possible that our polyclonal serum RNA interference was carried out by schistosomula with dsRNA for SmHDAC8, SmRho1 (positive control) or luciferase (negative control) as described in the Methods section. Actin-F was revealed with phalloidin staining and the nuclei were stained with DAPI (Blue). Microscopic examination was carried out 2 days after RNAi treatment. Scale bar represents 20 μm, magnification, x630 (top panel). Airyscan images with orthogonal views of S. mansoni schistosomula are also shown. Results shown are from one experiment. Scale bars represent 20μm, magnification, x630 (bottom panel). (C) RT-qPCR results of RNAi treatment with dsRNA for SmHDAC8 (dsSmHDAC8), SmRho1 dsSmRho1) or luciferase (dsSmLuc) and analyses of relative transcript levels of SmHDAC8, SmRho1.1 or SmRho1.2 in schistosomula. SmPSMB7 was used as an internal reference gene (32). The results were analyzed using the 2−ΔΔCT method and experiments were carried out once in duplicate.
https://doi.org/10.1371/journal.pntd.0009503.g008 masked the binding sites with some of them. In support of this hypothesis, Karolczak-Bayatti and coworkers have shown that human HDAC8 co-immunoprecipitates with cofilin [41]. Human cofilin is involved in the stabilization of actin filaments and forms a protein complex with RhoA and ROCK [37]. Although ROCK was not found as a partner of SmRho1, it is comforting to find cofilin as one of the members of a potential parasite protein complex formed by SmRho1, SmROCK1, SmLIMK and the Myosin light chain (MLC, Accession number G4VBS3; Smp_045220) also identified as an SmRho1 partner. We can assume that SmHDAC8 and SmRho1 form a multiprotein complex, especially with cofilin, which we have identified as SmRho1 partner. Moreover, in humans ROCK1 is the effector of GTP-RhoA and after its activation, the downstream effectors, myosin light chain (MLC) and cofilin are activated by phosphorylation. Phosphorylated MLC stimulates the binding of myosin to actin to regulate stress fiber formation and actomyosin contractility [42]. It was also shown that the activity of RhoA and its effector ROCK is inhibited by Rap1 signaling and PKA to regulate coronary artery relaxation [43]. The connection between SmRho1 and SmRap1 (Putative Rap1, Accession number G4VE67) illustrates once again the involvement of the GTPase in the regulation of cytoskeletal events.
The presence of actin binding proteins and regulators of the actin cytoskeleton like spectrin or ankyrin in the SmRho1 interactome, provides additional clues concerning a signaling pathway dependent on SmRho1. Intriguingly, some, such as α-actinin, and spectrin interact directly or indirectly with NMDA receptors [44]. Moreover, p250GAP is the first Rho family GAP GTPase shown to be enriched in the NMDA receptor complex, suggesting that p250GAP links the NMDA receptor to actin reorganization via RhoA [45].
Kinesin was also found in the SmRho1 interactome. Kinesins belongs to a class of motor proteins that move along microtubules. Pan and coworkers found that BNIP-2, a BCH domain containing protein binds RhoA, RhoGEF and kinesin-1 to regulate microtubule dynamics [46]. In addition, it was shown that RhoA-GTP binds a protein inserted in endoplasmic reticulum membranes named kinectin or KTN1 that interacts with the cargo-binding site of kinesin, thus activating its microtubule-stimulated ATPase activity, which is required for vesicle motility [47]. Kinesin also interacts with spectrin and ankyrin to regulate intracellular organelle transport [48]. In eukaryotic cells, spectrin is a cytoskeletal protein that lines the intracellular side of the plasma membrane. It forms pentagonal or hexagonal arrangements, forming a scaffold that plays an essential role in plasma membrane integrity and cytoskeletal structure. Ankyrins form a family of proteins that mediate the attachment of membrane proteins to the spectrin-actin cytoskeleton. Thus, the ankyrin-spectrin assembly provides mechanical stability to the lipid bilayer in addition to organization of membrane proteins. Moreover, in X. laevis, Cho and coworkers demonstrated that cytoskeletal organization, cell adhesion and ectodermal integrity are biological processes regulated by a catenin-spectrin-ankyrin-p190RhoGAP complex via RhoA activity [49].
To support our data of SmRho1interactome, we used single-cell RNAseq data of Collin's lab as additional support (http://www.collinslab.org/schistocyte/) and we have noted that some of SmRho1 partners share the same compartments/organs for expression. We found that SmRho1 (Smp_330190) is co-expressed with SmROCK1 (Smp_139800) in parenchyma and muscles. In S. mansoni tegument, Smkinesin (Smp_001040) and Smα-actinin (Smp_014780) are expressed with SmRho1. Moreover, expression levels of these genes appear to be influenced by pairing and they are highly expressed in tegument, parenchyma and muscles of females, and specially particularly in mature females. Interestingly, in each compartment in males or females, we identified genes encoding for specific RhoGEFs and RhoGAPs which are coexpressed with SmRho1. For instance, in parenchyma of matures females, we found genes encoding for two RhoGAPS (Smp_138150, Smp_122680) and one for a RhoGEF (Smp_ 316340).
Another element to be kept in mind is that SmRho1 may be a substrate of SmHDAC8, leading to an ephemeral interaction. Moreover, SmHDAC8 is not highly expressed in adult worms [11]. CoIP followed by Western Blotting, which is more sensitive, did allow us to detect SmHDAC8 bound to SmRho1. In addition, this interaction was confirmed in vitro using the proteins overexpressed in yeast and X. laevis oocytes.
The use of selective inhibitors of SmHDAC8 and SmRho, as well as knockdown of their transcripts using RNAi, strongly suggests that both proteins are involved in maintaining the integrity of the actin cytoskeleton. Because SmHDAC8 has multiple interactants, we can nevertheless not conclude that the observed phenotypes are the consequence of deregulation of a signaling pathway jointly mediated by SmRho1 and SmHDAC8. Indeed, an identified HsHDAC8 substrate is cortactin which contributes to the organization of the F-actin cytoskeleton. It was shown that cortactin-actin interaction is regulated by (de)acetylation and HsHDAC8 seems to influence smooth muscle contraction [21,50]. Further to this point, we cannot rule out an effect of inhibition/knockdown of SmHDAC8 directly on the muscle cells themselves, leading to subsequent disruption of the cytoskeleton. This point merits further investigation.
In our experiments inhibition or transcript knockdown of SmHDAC8 or SmRho1 the effects observed were much more apparent in schistosomula than in adult worms. This result apparently conflicts with our previous results showing the toxicity of both the pan-HDAC inhibitor TSA [12] and the selective SmHDAC8 inhibitor TH65 [14] for adult worms as well as for schistosomula. In the present study we developed a strategy designed to detect the early effects of inhibition or transcript knockdown on the parasite, whilst minimizing overt necrosis/apoptosis. The low level of effects on the adult worms, limited to the structure of the spines may be due to a relatively reduced access of the inhibitors to the worms compared to the larvae. However, the high level of transcript knockdown obtained for adult worms did not give rise to overt effects on the actin cytoskeleton, whereas a lower level of knockdown did cause extensive disruption of the cytoskeleton in schistosomula. This may suggest that schistosomula are particularly susceptible to interference into the maintenance of the cytoskeleton.
In future work we will consider determining the impact of inhibition of SmHDAC8 on the microtubular network and perform staining experiments after SmHDAC8 inhibition and KO using TH65 and RNAi. In 2011, Yamauchi et al. showed that the interference of transcripts encoding human HDAC8 using siRNA disrupted the microtubule network of cells [51].
Following our initial identification of SmRho1 as a potential partner for SmHDAC8, we have consolidated the demonstration of this interaction, both in vitro and within the parasite using co-immunoprecipitation studies. Our data demonstrated that SmHDAC8-SmRho1 interaction involves the C-terminal domain of SmRho1.1. However, our attempts to "transform" SmRho1.2 into SmRho1.1 via limited site-directed mutagenesis were unsuccessful, and suggest that the entire interaction interface formed by the aa that differ between the two isoforms is responsible for this difference in interaction.
To confirm this specificity of interaction it will be necessary in a future study to determine the interaction between SmHDAC8 and SmRho1 isoforms after IP with SmHDAC8 using adult worms and schistosomula protein extracts. The SmRho1 protein thus isolated will be analyzed in MS to identify which SmRho1 isoform is present. The observation that only SmRho1.1 can bind SmHDAC8, and that only SmRho1.2 can bind SmDia, suggests that these isoforms have developed distinct functions. This implies that there are two very distinct pathways that participate in the organization of the cytoskeleton via the two isoforms of SmRho1. Thus, only the SmRho1.1 isoform should interact with SmROCK for example, but this requires further investigation.
Moreover, although our results argue for a direct interaction between the SmRho1 isoforms and its partners, it should be borne in mind that the effects of inhibition and RNAi observed in the parasite may reflect the involvement of multiprotein complexes. Nevertheless, the detection of the acetylation of SmRho1 raises the possibility that it is a substrate of SmHDAC8 and that acetylation, could be involved in the modulation of the properties of SmRho1.
Although human RhoA and other Rho GTPases are not acetylated, Hong and coworkers identified an acetylation of S. japonicum Rho1 on lysine 141 [23]. This residue K141 is absent in SmRho1, and another target residue K136 in SmRho1 is apparently not acetylated in SjRho1. It should also be borne in mind that only one of the two isoforms of SmRho1 may be acetylated. Interestingly, there are regulatory proteins that could be described as "atypical regulators" and SmHDAC8 could be one of these. The Memo protein, for example, interacts with RhoA and appears to promote its membrane localization and therefore its activation, within a Memo-RhoA-mDia1 multiprotein complex [52]. Hence SmHDAC8 could also be considered as an atypical regulator of SmRho1 either via direct binding or via deacetylation. In addition, if only one of the two isoforms of SmRho1 is acetylated, the specific acetylation of SmRho1.1 and the potential deacetylation by SmHDAC8 could constitute a regulatory mechanism of the parasitespecific SmRho1-mediated signaling pathway. However, the lysine that we have identified as an acetylation site on SmRho1 is conserved in both isoforms suggesting that SmRho1 isoforms could be acetylated and deacetylated by the same KDAC. Tools like CRISPR-Cas9, would allow obtaining conditional KO or KO parasites in order to study the specific role of SmRho1 genes at different parasitic stages. Although this is not currently possible, projects concerning the development of CRISPR-Cas9 technology in schistosome are underway [53]. Finally, although SmDia interacts specifically with SmRho1.2, we have no knowledge of how the SmRho1.2-Sm-Dia complex could participate in the regulation of the polymerization of actin filaments. However, in 2009, Quack and coworkers [24] demonstrated that SmDia is able to interact directly with the SmTK3 protein (Src -like Tyrosine-Kinase). Moreover, in humans, the RhoA-GTP/ mDia/Scr complex is known to regulate the formation of actin filaments [54,55]. Using site-directed mutagenesis, the glutamine Q147 and the valine V148 of SmRho1.1 were substituted by a glutamic acid and a methionine (SmRho1.1 EM) and then the lysine K151 and the serine S153 by two asparagines (SmRho1.1 EMNN). SmRho1.2 QV and SmRho1.2 QVKS mutants were produced by site-directed mutagenesis using SmRho1.2 protein. First, the glutamic acid E147 and the methionine M148 were substituted by a glutamine and a valine and then, the two asparagines N151 and N153 were replaced by a lysine and a serine. (B) Co-immunoprecipitation and WB experiments performed in X. laevis oocytes revealed that SmRho1.2 mutants (HA-tagged) are not able to bind SmHDAC8 (Myc-tagged). cRNAs encoding HA-tagged SmRho1 isoforms, SmRho1.1 mutants or SmRho1.2 mutants were co-injected in X. laevis oocytes with cRNA encoding Myc-tagged SmHDAC8. Oocytes were incubated in ND96 medium and lysed. Proteins from soluble extracts were immunoprecipitated (IP) by anti-HA or anti-Myc antibodies and analyzed by WB to detect SmHDAC8 (50 kDa), SmRho1 isoforms (22 kDa) or SmRho1 mutants (22 kDa) with anti-Myc or anti-HA antibodies. (TIF) S1  Table. SmRho1 partners from Co-IP/MS analysis. Sheet "IP1-IP2 full list" contains the 1,672 different proteins identified from the two independent Co-IP/MS experiments IP1, IP2 respectively. Sheets "IP1 SmRho1-selected protein" and "IP2 SmRho1selected proteins" contain the 86 and 32 different proteins obtained after the selection step (cf. manuscript for details) for IP1 and IP2 respectively. (XLSX)