A transcription factor DAF-5 functions in Haemonchus contortus development

Abnormal dauer formation gene (daf-5), located downstream of the DAF-7 signalling pathway, mainly functions in dauer formation and reproductive processes in the free-living nematode Caenorhabditis elegans. Although the structure and function of daf-5 have been clarified in C. elegans, they still remain totally unknown in Haemonchus contortus, a socio-economically important parasitic nematode of gastric ruminants. A homologue of daf-5, Hc-daf-5, and its inferred product (Hc-DAF-5) in H. contortus were identified and characterized in this study. Then the transcriptional profiles of Hc-daf-5 and the anatomical expression of Hc-DAF-5 in H. contortus were studied using an integrated molecular approach. RNA interference (RNAi) was performed to explore its function in transition from the exsheathed third-stage larvae (xL3s) to the fourth-stage larvae (L4s) in vitro. Finally, the interaction between Hc-DAF-5 and Hc-DAF-3 (a co-Smad) was detected by bimolecular fluorescence complementation (BiFc) in vitro. It was shown that Hc-DAF-5 was a member of the Sno/Ski superfamily. Hc-daf-5 was transcribed in all developmental stages of H. contortus, with significant upregulation in L3s. Native Hc-DAF-5 was localized in the reproductive organs, cuticle, and intestine via immunohistochemistry. RNAi revealed that specific small interfering RNAs (siRNAs) could retard xL3 development. In addition, the interaction between Hc-DAF-5 and Hc-DAF-3 indicated that the SDS box of Hc-DAF-5 was dispensable for the binding of Hc-DAF-5 to Hc-DAF-3, and the MH2 domain was the binding region between Hc-DAF-3 and Hc-DAF-5. In summary, these findings show that Hc-daf-5 functions in the developmental processes of H. contortus, and this study is the first attempt to characterize the daf-5 gene in parasitic nematodes.


Background
Transcription factors function in basal transcription by binding to particular DNA sequences in gene regulatory regions to control their transcription. Transcription factors are commonly classified into activator and repressor families based on their functions. Although the majority of transcription factors have a positive effect, a number of transcription factors exert an inhibitory effect on transcription [1].
For the transforming growth factor-β (TGF-β) signalling pathway in vertebrates, activation of receptors causes downstream R-Smad phosphorylation, which allows the latter to form an active Smad transcriptional complex with co-Smads. The function of this complex is affected positively or negatively by interaction with cofactor, co-activator, and co-repressor molecules, which ultimately regulates cellular processes such as proliferation, differentiation, and apoptosis [2]. As downstream components of the TGF-β signalling pathway, the Ski proto-oncoprotein family of co-repressors, including c-Ski and SnoN, play important roles in tightly controlled repression of Smad-mediated transcriptional activation, and antagonize TGF-β signalling by binding to the Smad transcriptional complex, thus preventing their interaction with co-activators, and recruiting co-repressors [3]. Additionally, Sno/Ski transcription is regulated by Smads. Thus, the degradation or accumulation of Sno/Ski protein occurs depending on the context of the TGF-β signalling [4].
DAF-5, the homologue of vertebrate Ski/Sno in the free-living nematode Caenorhabditis elegans, functions in the TGF-β signalling pathway (DAF-7 signalling pathway) for dauer regulation and egg-laying in this nematode, but not like a conventional Sno/Ski protein [5]. In brief, the DAF-7 signal is transduced by R-Smads. These molecules, when activated, function as a transcriptional complex and inhibit the functions of DAF-5. DAF-5 acts as a co-factor of co-Smad (DAF-3), rather than an antagonist of DAF-3. It also acts as a complex and is antagonized by upstream R-Smad components [5].
Although the functions of Sno/Ski and DAF-5 have been clearly identified in vertebrates and C. elegans, respectively, their roles in parasitic nematodes have not yet been elaborated. Elucidating the functions of DAF-5 in parasitic nematodes will help us understand how this molecule and the DAF-7 signalling pathway work in the parasites. Herein, a homologue of daf-5 was identified from Haemonchus contortus, a blood-sucking gastric parasitic nematode of small ruminants. Its temporal transcriptional profiles in different developmental stages and special expression patterns in adult worms were explored by real-time polymerase chain reaction (PCR) and immunohistochemistry, respectively. In addition, small interfering RNAs (siRNAs) were used to knock down Hc-daf-5 in exsheathed third-stage larvae (xL3s) by soaking. A bimolecular fluorescence complementation (BiFc) experiment was also performed to demonstrate the interaction between Hc-DAF-5 and Hc-DAF-3.

Worm maintenance
The H. contortus Haecon-5 strain was maintained by goats, which were infected orally with 5000 L3s. Free-living stages including eggs, first-stage larvae (L1s), secondstage larvae (L2s), and L3s were harvested from faeces of infected goats. Parasitic stages (L4s and adults) were collected from the abomasa of infected goats euthanized at 8 and 30 days, respectively, and these two developmental stages were washed extensively in 0.85% sodium chloride, and male and female worms were carefully separated under a microscope prior to storage at −80 °C.

Nucleic acid and protein preparation and gene cloning
Total RNA was extracted from individual developmental stages of H. contortus using TRIzol reagent (Simgen, China). All RNA samples were treated with DNase I to remove genomic DNA (gDNA), and then their integrity and yields were verified by electrophoresis and spectrophotometric analysis. Complementary DNA (cDNA) was synthesized from RNA (1 μg) using the PrimeScript RT reagent kit with gDNA Eraser (Takara, Japan) for coding sequence (CDS) amplification and real-time PCR. RNA was stored at −80 °C and DNA was stored at −20 °C until use.
The whole-worm proteins were prepared by grinding adult worms in phosphatase inhibitor and protein lysate (Roche Molecular Biochemicals, Switzerland), which were fractionated into soluble and insoluble materials after centrifugation at 10,000×g for 3 min at 4 °C. Then the soluble material was stored at −80 °C with protease inhibitor (Thermo Fisher, USA) added, followed by western blot analysis.
The sequences of full-length cDNA and gDNA, were retrieved from the transcriptomic and genomic datasets of H. contortus. The CDS was amplified from cDNA with the primer pair Hc-daf-5-cF/Hc-daf-5-cR (Additional file 1: Table S1) using the following protocol: 98 °C/10 min, then 98 °C/10 s, 55 °C/5 s and 72 °C/2 min for 35 cycles, and 72 °C/10 min. The PCR product was inserted into pMD-19T and sequenced in both directions directly by Tsingke Biological Technology, China.

Bioinformatics analyses
The CDS of Hc-daf-5 were conceptually translated into predicted amino acids using DNASTAR software (http:// www. dnast ar. com/). The predicted amino acid sequence of Hc-DAF-5 was compared with the sequences in nonredundant protein databases using BLASTP from the National Center for Biotechnology Information (NCBI) to confirm the homologous sequences.
The full-length protein sequences from eight species (C. briggsae, C. elegans, D. rerio, D. melanogaster, H. sapiens, M. musculus, N. brasiliensis, and T. canis) (Additional file 1: Table S2) were aligned and used for phylogenetic analyses by means of the neighbour-joining (NJ), maximum parsimony (MP), and maximum likelihood (ML) methods. Confidence limits were assessed using a bootstrap procedure employing 1000 pseudo-replicates in MEGA [21].

Transcriptional analyses of Hc-daf-5 in different stages of H. contortus by real-time PCR
Real-time PCR was carried out using the specific primers Hc-daf-5-qF/Hc-daf-5-qR (Additional file 1: Table S1) to determine the mRNA levels in different developmental stages of H. contortus including eggs, L1s, L2s, L3s, both sexes of L4s, and adult worms. Total RNA of each stage was isolated with TRIzol reagent according to the manufacturer's instructions and treated with DNase I to remove gDNA before synthesis of cDNA. The realtime PCR conditions were set as follows: one cycle at 50 °C/2 min and 95 °C/30 s, 40 cycles at 95 °C/15 s, 60 °C/15 s, and 72 °C/20 s, and one cycle at 60 °C/1 min, 95 °C/15 s, and 60 °C/15 s. Each sample was tested in triplicate, with β-tubulin of H. contortus (GenBank: M76493) as a reference gene (using specific primers Hctubulin-qF and Hctubulin-qR; Additional file 1: Table S1), and the average threshold (Ct) was taken to compare the relative quantities with Am (Am = 1) using the 2 −△△Ct method [22]. This assay was carried out three times. One-way analysis of variance (ANOVA) in GraphPad Prism 6 was adopted for statistical analysis. P-values were calculated using the Tukey post hoc test, and P < 0.05 was considered statistically significant.

Production of polyclonal antibody against recombinant Hc-DAF-5 and immunoblot analysis
A pair of specific primers Hc-daf-5-pF/Hc-daf-5-pR (Additional file 1: Table S1) containing double restriction sites was designed according to the CDS of Hc-daf-5, and it was employed to amplify the CDS of Hc-daf-5 through PCR. Then the amplified sequence was cloned into the prokaryotic expression vector to create the expression plasmid pET-28a-Hc-daf-5, which was transformed into BL21 (DE3) cells of Escherichia coli, followed by 1 mM isopropyl β-d-1-thiogalactopyranoside (IPTG) induction at 16 °C for 12 h to produce recombinant rHc-DAF-5. Next, the recombinant rHc-DAF-5 was purified using a Ni Sepharose column system (GE, USA) according to the manufacturer's protocol. Later, the purified rHc-DAF-5 was used in immunizing New Zealand white rabbit to produce anti-Hc-DAF-5 polyclonal antibody. Finally, the titer and specificity of anti-Hc-DAF-5 polyclonal antibody were determined by enzyme-linked immunosorbent assay (ELISA) and western blot.

Evaluation of protein expression in H. contortus via immunofluorescence assay
Fresh male and female adults of H. contortus were fixed in 4% paraformaldehyde at 4 °C for 24 h and then consecutively dehydrated in an ethanol series (75% for 4 h, 85% for 2 h, 90% for 2 h, 95% for 1 h, and 100% for 30 min twice). The dehydrated worms were then incubated in xylene/absolute ethanol (1:1) solution for 5 min and xylene for 10 min, and then embedded in paraffin. Next, 4 μm-thick paraffin-embedded sections were subjected to immunofluorescence staining. For immunofluorescence assay, the sections were treated with EDTA buffer at 100 °C for 10 min. After blocking with 5% BSA for 20 min, anti-Hc-DAF-5 polyclonal antibody and goat anti-rabbit IgG antibody diluted at 1:100 were sequentially added and incubated at 4 °C overnight and at 37 °C for 50 min. Then the sections were stained with 4′6-diamidino-2-phenylindole (DAPI) for 5 min at 37 °C in a dark place. After thorough washing with PBS, the sections were observed under a fluorescence microscope (Olympus CX21, Japan).

RNA interference (RNAi) in H. contortus by soaking in siRNA
The CDS of Hc-daf-5 was used to design the siRNA sequences with the siRNA Design Tools program, and siRNA oligos (Additional file 1: Table S3) were chemically synthesized by Shanghai GenePharma Co., Ltd.
For RNAi, L3s were exsheathed and washed five times in 0.9% NaCl solution, followed by centrifugation at 600×g with diethyl pyrocarbonate (DEPC)-treated water three times. In each silencing assay, 50 μl of nematode suspension (about 5000 larvae) was placed into a 96-well plate. Then 10 μl of Lipofectamine 2000 (Thermo Fisher, USA) was incubated with 5 μl of Earle's Balanced Salt Solution (EBSS) (pH 5.2) containing 2.5 μg/μl amphotericin, 100 μg/μl streptomycin, and 100 IU/ml penicillin (Gibco, USA) at 25 °C for 5 min. RNasin (0.2 U) and siRNA solutions were added for incubation for 15 min. Three siRNAs of Hc-daf-5 were mixed in equal amounts, and the final concentration of each siRNA was adjusted to 1 μM, while the final concentration of negative control siRNA was 3 μM, with nuclease-free water as the blank control.
The knockdown experiments were carried out as described previously [23]. In brief, three groups of xL3s (5000 in each group), which were kept in 80 μl of EBSS (pH 5.2) and supplemented with respective siRNAs, were soaked for 72 h. Approximately 100 larvae in each group were transferred to a fresh culture medium with EBSS for another 5 days to assess their development. The remaining larvae were subjected to RNA extraction.
All RNAs were extracted using TRIzol reagent according to the manufacturer's instructions. Then cDNA was synthesized using the PrimeScript RT reagent kit with gDNA Eraser (Takara, China). Each 10 μl of reaction was conducted with the same amount of cDNA from each sample. The 18S gene of H. contortus was taken as the endogenous control using the primer pair Hc-18S-qF/ Hc-18S-qR (Additional file 1: Table S1). Primers (Hcdaf-5-qF/Hc-daf-5-qR) used in detecting transcriptional profiles were also used here for the detection of transcriptional changes of Hc-daf-5 in worms treated with siRNA. The amplification efficiency of the primers was tested by a standard curve assay, and linear regression analysis showed similar slopes of all tested primers. The  [24].

Assessing the interaction between Hc-DAF-5 and Hc-DAF-3 by bimolecular fluorescence complementation (BiFC)
The BHK21 cells were grown and maintained in Dulbecco's modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/ streptomycin. The plasmids pbJun-HA-KN151 and pbFos-Myc-LC151 were used as original plasmids for plasmid construction. In brief, the CDS fragment of Hcdaf-3 was amplified with the primer pair Hc-daf-3-fF/ Hc-daf-3-fR (Additional file 1: Table S1) and cloned into pbJun-HA-KN151 between the NheI and XhoI sites to replace bJun, thus generating the pHcDAF3-HA-KN151 plasmid. Meanwhile, the CDS fragment of Hc-daf-5 was amplified with the primer pair Hc-daf-5-fF/Hc-daf-5-fR (Additional file 1: Table S1) and cloned into pbFos-Myc-LC151 between the NheI and PvuI sites to replace bFos, so as to generate the pHcDAF5-Myc-LC151 plasmid. The BHK21 cells were seeded onto six-well plates and grown to 70-80% confluence for transfection. To examine the interactions between Hc-DAF-3 and Hc-DAF-5, the plasmids pHcDAF3-HA-KN151 and pHcDAF5-Myc-LC151 were co-transfected using Lipofectamine 2000 (Thermo Fisher, USA). Then the cells were incubated at 37 °C with 5% CO 2 for a further 20 h, and fluorescence was subsequently detected at 580-680 nm and imaged.
A BLAST search with the Hc-DAF-5 protein sequences as bait against the NCBI GenBank databases revealed similarities to the SKI, SnoN, Skate, and Icy in H. sapiens and the Snowski and Iceskate in D. melanogaster (Additional file 1: Table S2) [25,26]. The predicted protein sequence of Hc-DAF-5 was then aligned with homologues from five nematodes and five other metazoans. Multiple sequence comparison revealed the highest identity of Hc-DAF-5 to SKI in N. brasiliensis (58.3%), but low identity ranging from 18.7 to 28.7% to its homologues from different metazoan species, including B. malayi The alignment of these sequences also indicated that Hc-DAF-5 had two functional domains including a Dach box (Dachshund-homology domain) and an SDS box (Smad binding domain) (Fig. 1a). Dach box is a domain shared by the Sno/Ski, Skate/Icy and Dachshund superfamily. Dachshund is a transcription regulator conserved throughout bilaterians. Alignment of the Dachshundhomology domain among selected species (Fig. 1b, above) revealed that Dach box of Hc-DAF-5 had higher similarity to that of Nb-SKI than to that of Ce-DAF-5 and Cb-DAF-5. In H. contortus, DAF-5 clearly possessed an SDS box in the Smad binding domain, which was not found in any other proteins in H. contortus (Fig. 1a). The SDS box of Hc-DAF-5 had higher similarity to that of the Snowski group than to that of the Iceskate group. In human SKI, the SDS box and about 20 amino acids on either side constituted the minimal region required for Smad4 binding, so this region was also used for alignment (Fig. 1b below). All members of this family had a conserved zinc finger in the SDS box. The SDS box of Hc-DAF-5 had higher similarity to that of parasitic nematodes (such as B. malayi) than to that of C. elegans (Fig. 1b, below). As with Ce-DAF-5, Hc-DAF-5 also had a predicted coiled coil structure at the C-terminus (Fig. 1a).
The results of phylogenetic analysis showed that the topologies of the MP, ML, and NJ trees were concordant. Hc-DAF-5 grouped with Nb-SKI with 100% bootstrap support, and then they grouped with DAF-5 from C. elegans and C. briggsae with 92% bootstrap support. This DAF-5 cluster grouped with Ski/Sno and Ice/Skate from H. sapiens, M. musculus, D. rerio, D. melanogaster, and T. canis, with 85% bootstrap support. The Dachshund subfamily from C. elegans, D. melanogaster, D. rerio, H. sapiens, and M. musculus constituted an independent cluster with 99% bootstrap support (Fig. 1c).

Transcription of Hc-daf-5 and expression of Hc-DAF-5 in H. contortus
The transcriptional profile of Hc-daf-5 was explored in eight developmental stages and sexes including egg, L1, L2, L3, L4 female, L4 male, adult female, and adult male throughout the life cycle of H. contortus. Hc-daf-5 was transcribed at a detectable level in all tested stages and sexes, with significant upregulation in L3 (ANOVA, F (7,16) = 12.03, L3 vs egg, L1, L2, L4f, L4m, and Af, P ≤ 0.0001. L3 vs Am, P = 0.001) (Fig. 2a). Truncated Hc-DAF-5 (108-297 aa) was expressed in E. coil BL21 (DE3) with a size of 27.5 kDa (Fig. 2b). The anti-rHc-DAF-5 polyclonal antibody was used in western blot, and a band of around 60 kDa in total protein extracted from adult worms was recognized, while no band was recognized by the negative serum (Fig. 2c). In further immunofluorescence assay, the anti-rHc-DAF-5 polyclonal antibody detected the endogenous Hc-DAF-5 expression in the cuticle and intestine of both female and male adult worms, as well as in the ovary of female adult worms and the cement gland of male adult worms. No fluorescence was observed in worm sections probed with the negative serum (Fig. 2d).

Effect of specific siRNA on xL3 development of H. contortus
As observed in previous studies, the development of H. contortus from xL3 to L4 underwent six stages of morphological changes, namely stage A to stage F [27,28]. In RNAi assay herein, the soaking of xL3s in Hc-daf-5-specific siRNA for 72 h resulted in a significant reduction in Hc-daf-5 transcript abundance, whereas no reduction in Hc-daf-5 transcript abundance was detected in the control siRNA-treated group or untreated group (ANOVA, F (2, 6) = 16.57, Hc-daf-5 siRNA-treated group versus control siRNA group, versus blank group, both P = 0.0060) (Fig. 3a). In addition, the proportion of xL3s developed to stage B and beyond in the Hc-daf-5 siRNA-treated group was significantly lower than that in the control siRNA-treated group and untreated group (ANOVA, (F (2, 6) = 23.65, Hc-daf-5 siRNA-treated group versus control siRNA group, versus blank group, P = 0.0054 and P = 0.0015, respectively) (Fig. 3b).

Hc-DAF-5 interacts with co-Smad Hc-DAF-3
BiFC was employed to detect the interaction between Hc-DAF-5 and Hc-DAF-3 in live mammalian cells. The cells co-transfected with plasmid constructs as the positive controls encoding bJun-HA-KN151 and bFos-Myc-LC151 showed bright red fluorescence, consistent with the result in a previous report (Fig. 3c) [29].

Discussion
In the present study, Hc-daf-5, a homologue of Ce-daf-5 and a transcription factor-encoding gene, was identified and characterized in H. contortus. Sequence analysis revealed that the predicted protein Hc-DAF-5 contained a Dach box, an SDS box, and a coiled coil motif. Approximately 2-3% of all protein residues form a coiled coil motif, and it functions by mediating oligomerization of a large number of protein subunits. It can participate in signal-transducing events or act as a molecular recognition system [30]. The conservation of the coiled coil motif between Hc-DAF-5 and Ce-DAF-5 indicates the potential function of Hc-DAF-5 in transcriptional regulation [30].
Recent research has shown that the insulin-like signalling pathway and dafachronic acid (DA) signalling pathway play a conserved role in larval development in both free-living and parasitic species [31][32][33][34]. In contrast, the role of the TGF-β signalling pathway does not seem to be conserved between free-living and parasitic species in larval development regulation [35]. In C. elegans, Cedaf-5 is transcribed in all developmental stages including dauer larvae. Its level peaks in L1s and decreases thereafter (https:// wormb ase. org). However, in Strongyloides stercoralis, the transcription of Ss-daf-5 is higher in infective third-stage larvae (L3i) [36]. In the present study, the transcription of Hc-daf-5 remained low in L1s and L2s, but peaked in L3i, which is different from that of Ce-daf-5 but similar to that of Ss-daf-5. In addition, the transcription pattern of Hc-daf-5 is consistent with that of genes of the DAF-7 signalling pathway in H. contortus, including Hc-daf-3 and Hc-tgfbr-2 [37,38]. The difference in transcriptional profiles of daf-5 between C. elegans and parasitic nematodes (including H. contortus in clade V and S. stercoralis in clade IV herein), together with that in previous studies [38], indicates divergent functions of the TGF-β signalling pathway in free-living and parasitic species, suggesting that parasitic nematodes utilize the existing signalling pathway for different purposes in evolution.
Considering the high transcription level of Hc-daf-5 in L3s, siRNA-mediated RNAi was performed to explore the role of Hc-daf-5 in transition from the free-living stage (xL3) to the parasitic stage (L4). In this study, silencing of Hc-daf-5 decreased the developmental rate of xL3. The experimental group treated with specific siR-NAs showed fewer developed worms because the buccal development failed to initiate. This is the same effect as described for the knockdown of Hc-daf-3 and Hc-akt-1 [38,39], and suggests that Hc-daf-5 functions in the development from xL3 to L4.
Based on the high abundance of Hc-daf-5 in L3s and its functions in xL3 development, Hc-DAF-5 localization in L3s was attempted but failed in this study. Therefore, protein localization was performed on adult worms. Herein, native Hc-DAF-5 protein was expressed prominently in the reproductive organs, which is consistent with Hc-DAF-3 [38]. Considering that Ce-DAF-5 is required for egg-laying [5], it was proposed that Hc-DAF-5 may also function in embryonic development and spermatogenesis. In addition to the gonad organs, Hc-DAF-5 was also expressed in the cuticle and intestine of adult worms. In H. contortus, the cuticle enables the worms to be resistant to harmful substances [40]. In summary, Hc-DAF-5 may also be associated with such processes as embryonic development and environmental resistance.
Phylogenetic analysis in this study revealed a relatively close relationship between DAF-5s in the free-living nematodes and parasitic species. In C. elegans, DAF-5 functions as a co-factor of DAF-3 (co-Smad), rather than an antagonist. However, its homologue functions as an antagonist of co-Smad in vertebrates [3]. The relatively close relationship between Hc-DAF-5 and Ce-DAF-5 suggested that Hc-DAF-5 probably also functions as a co-factor of Hc-DAF-3. In order to test this hypothesis, BiFc was used for verifying the interaction between Hc-DAF-5 and Hc-DAF-3. The results showed that the region downstream of the SDS box of Hc-DAF-5 was dispensable for Hc-DAF-3 binding, and the MH2 domain of Hc-DAF-3 was sufficient for its interaction with Hc-DAF-5, consistent with the findings in C. elegans and vertebrates [5], suggesting that Hc-DAF-3 and Hc-DAF-5 interact with each other and function as a transcriptional regulatory complex to further regulate the downstream transcription.