The protocols (shvri-ra-2017070578) were approved by the Institutional Animal Care and Use Committee of the Shanghai Veterinary Research Institute, and authorized by the Animal Ethical Committee of Shanghai Veterinary Research Institute.

Ticks were obtained from a colony of R. haemaphysaloides maintained on the ears of New Zealand white rabbits at Shanghai Veterinary Research Institute, CAAS. Off-host ticks were kept in an incubator at 25°C and 95% humidity. To detect dynamic expression of RH36 and other proteins, midguts, salivary glands, ovaries, hemolymph, and fat bodies were dissected from unfed female ticks, 5-day-fed female ticks, 7-day-fed female ticks, engorging ticks and ticks on day 3 or 10 after engorgement. Furthermore, unfed female adult ticks (N = 80 females per group, three replicates) were microinjected with RH36 and luciferase double-stranded RNA (dsRNA), respectively, as described previously (Wang et al., 2017). Two individual five-day fed female ticks from each group were collected to characterize RH36 mRNA levels after RH36 RNA interference (RNAi). Blood-feeding ticks from RH36 or luciferase gene silencing groups (14 ticks per group) were removed from rabbits at 5 (indicated by RF5 and LF5, respectively) or 7 days (indicated by RF7 and LF7). Engorged ticks (14 ticks per group) were also collected from RH36 or luciferase gene silencing groups at day 0 (denoted by RE0 and LE0), day 3 (denoted by RE3 and LE3), day 7 (denoted by RE7 and LE7) and day 10 (denoted by RE10 and LE10) after engorgement. The ovaries, hemolymph, fat bodies and salivary glands were dissected from 7 ticks per group for RNA or protein extraction and washed in PBS to remove hemolymph -related cells.

Total RNA was extracted from ovaries using TRIZOL (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Total RNA quality was evaluated using Agilent 2100 Bioanalyzer RNA Nano Chip (Agilent Technologies, Santa Clara, CA, USA). Tick ovaries with SDT buffer (containing 4% SDS, 100 mM DTT and 100 mM Tris-HCl, pH 7.6) and quartz sand (another 1/4 inch ceramic bead MP 6540-424 for tissue samples) were homogenized by MP homogenizer (24 × 2, 6.0M/S, 60s, twice) and subsequently sonicated and boiled for 15 min. After it was centrifuged at 10,000 rpm for 20 min, the supernatant was filtered with 0.22 μm filters. Total protein was also extracted from tick salivary glands, fat bodies and hemolymph according to the above instruction, respectively. The filtrate was quantified with the BCA Protein Assay Kit (Bio-Rad, USA). These RNA and protein samples were stored at −80°C until use.

The B cell linear epitope of RH36 protein was predicted using the Immune Epitope Database (IEDB) of http://tools.iedb.org/bcell/result/. The peptide (LTDDYWKQGEHPSGEYPIS) was selected randomly and synthesized with peptide purify of at least 95% and coupled with KLH by GL Biochem (Shanghai, China), and emulsified with an equal volume of Freund's complete adjuvant (Sigma-Aldrich, St. Louis, MO, USA) used to intraperitoneally inject Balb/c mice. Two booster injections of RH36 peptide were given with the same procedure but using Freund's incomplete adjuvant. One week after the final immunization, sera were collected following centrifugation. Antibodies against RH36 peptides were used to detect expression levels by Western blot. According to standard techniques, total proteins were separated on a 12% SDS-PAGE gel and transferred to PVDF membranes (0.45 μm, Millipore, Billerica, MA, USA), and the membranes were immunoblotted with RH36 antibody (1:200). The monoclonal antibody of β-actin against mouse (Proteintech, Rosemont, USA, Cat. No. 60008-1-Ig) was used as the control. IgG goat anti-mouse or goat anti-rabbit antibodies conjugated with HRP were used for secondary antibodies (1:8,000, Bethyl Laboratories, Inc., Montgomery, Texas, USA). The membranes were visualized using automatic chemiluminescence image analysis system (Tanon-5200, Tanon Science & Technology Co., Ltd, Shanghai, China). Densitometry used to analyze indicated protein bands was determined by Image J (NIH), normalized with the control (Schneider et al., 2012).

Crude eggs were homogenized and subsequently centrifuged at 10,000 rpm for 20 min, the supernatant was applied onto a Sepharose CL-4B gel filtration column and diethyl-amino-ethyl (DEAE)-cellulose column in turn (Yang et al., 2015). Polyclonal antibody and monoclonal antibody against vitellin was generated from Balb/c mice (Wang et al., 2017). Splenocytes from immunized mice were fused to myeloma cells, SP2/0, using polyethylene glycol. Hybridomas were selected in HAT medium and the hybridoma culture supernatants were screened by enzyme-linked immunosorbent assay (ELISA) in 96-microtiter plates and Western blot (Yang et al., 2015).

Expression of vitellin was detected in the ovaries, hemolymph, fat bodies and salivary glands by Western blot according to the above description. The monoclonal antibodies of β-actin (Proteintech, Cat. No. 60008-1-Ig) and GAPDH (Proteintech, Cat. No. 60004-1-Ig) against mouse were used as the control. The bands were visualized using automatic chemiluminescence image analysis system (Tanon-5200, Tanon Science & Technology Co., Ltd., Shanghai, China). Heat shock protein 70 rabbit antibody (HSP70, Cell Signaling Technology, Boston, USA, Cat. No. 3177) was also used to detect the protein expression level in the different organs during different blood-feeding ticks by Western blot. The anti-GAPDH antibody against rabbit (Cell Signaling Technology, Cat. No. 2118) was used as the control.

For immunofluorescence, ovaries were obtained from adult female ticks at 10 days after engorgement as described above, fixed with 4% paraformaldehyde in 0.2 M sodium cacodylate buffer, and dehydrated in a graded series of ethanol and embedded in paraffin. Sections (4 μm) were prepared and mounted on glass slides. Paraffin was removed from the sections with xylene and the sections were hydrated by successive 5-min washes with a graded series of 100, 80, 75, and 65% ethanol. The slides were treated with EDTA antigen repair buffer for 30 min at 37°C, washed with PBST and incubated with 2% bovine serum albumin (BSA; Sigma-Aldrich) in PBST for 30 min at room temperature. The slides were then incubated for 14 h at 4°C with primary antibodies against vitellin diluted 1:5–1:200 in 2% BSA mixture and developed for 30 min with goat-anti-mouse IgG conjugated with FITC (Sigma-Aldrich) (diluted 1:500 in 2% BSA mixture) after 3 washes in PBST buffer. The slides were incubated for 20 min with DAPI (1:200, 2% BSA dilution) after 3 washes in PBST buffer and mounted in Fluoromount Aqueous Mounting Medium (Sigma-Aldrich). Samples were observed and photographed with 10X objective, 1.0X zoom and the excitation of 488 nm (FITC) and 405 nm (DAPI) under a Zeiss inverted fluorescence microscope (Zeiss, Germany).

Quantitative proteomic analysis by tandem mass tag (TMT) technology was performed by Shanghai Applied Protein Technology (APT, Shanghai, China). Each 100 μg peptide mixture digested with trypsin (Promega, Madison, WI, USA) was labeled using TMT reagent (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's instructions. The 6-plex TMT Labeling were fractionated into 15 strong cation exchange (SCX). LC-MS/MS analysis (1H) was performed on a Q Exactive mass spectrometer (Thermo Fisher Scientific) that was coupled to Easy nLC (Proxeon Biosystems, now Thermo Fisher Scientific) for 60 min. The mass spectrometer was operated in positive ion mode. MS data was acquired using a data-dependent top 10 method dynamically choosing the most abundant precursor ions from the survey scan (300–1,800 m/z) for HCD fragmentation. Automatic gain control (AGC) target was set to 3e6, and maximum inject time to 10 ms. Dynamic exclusion duration was 40.0 s. Survey scans were acquired at a resolution of 70,000 at m/z 200 and resolution for HCD spectra was set to 17,500 at m/z 200 and isolation width was 2 m/z. Normalized collision energy was 30 eV and the underfill ratio, which specifies the minimum percentage of the target value likely to be reached at maximum fill time, was defined as 0.1%. The instrument was run with peptide recognition mode enabled. And results were compared to a Uniprot database containing all sequences from Ixodidae (January 1, 2017) using MASCOT engine (Matrix Science, London, UK; version 2.2) embedded into Proteome Discoverer 1.4. The MASCOT identified parameters were set in Supplementary Table 1. Peptide identification was validated using false discovery rate (FDR ≤ 0.01). The standard of screening differentially expressed proteins was Fold change >1.2 and a P-value < 0.05 was considered statistically significant. Sequences of differentially expressed proteins were also searched using blastp tool against the online Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://geneontology.org/) and were subsequently mapped to pathways in KEGG. GO and KEGG pathway enrichment analysis were applied based on the Fisher's exact test. All assays were performed in three replicates.

After total RNA was treated with DNase I, mRNA was enriched by oligo (dT) with beads and fragmented into fragments of 100–400 bp. First-strand cDNA was synthesized from the mRNA fragments. And second-strand cDNA was then synthesized using DNA polymerase I and RNaseH. Short fragments were purified, end-repaired, connected with signal nucleotide A (adenine) addition and ligated with library-specific adaptors. The cDNA library was constructed and tested in qualification by Agilent 2100 Bioanalyzer and ABI StepOnePlus Real-Time PCR System, and then sequenced using Illumina HiSeq 4000.

Tick ovary transcriptome de novo assembly was performed by Beijing Genomics Institute (BGI, Shenzhen, China). After sequencing, low-quality, adaptor-polluted and high content of unknown base (N) reads were filtered to get clean reads. Clean reads were assembled by Trinity assembles software to obtain high-quality contigs (Yu et al., 2015). T gicl was used to perform clustering and eliminate redundant data in the assembled transcripts to obtain unique genes. Subsequently, expression analysis, heterozygous SNP detection, and functional annotation were performed. Differentially expression genes (DEGs) were further analyzed with gene ontology functional enrichment and KEGG pathway functional enrichment. The false discovery rate (FDR ≤ 0.001) was used to determine the threshold P value in multiple tests. Results were the mean of three replicates.

The dsRNA was synthesized using the T7 RiboMAX™ RNAi System (Promega, Madison, WI, USA) according to the manufacturer's recommendations. Oligonucleotide primers containing T7 promoter sequences (in italics at 5′-end) were listed in Supplementary Table 2. A dsRNA targeting luciferase used as a negative control was subjected to the same PCR amplification protocol using luciferase-specific primers (Yu et al., 2013). These dsRNA were maintained at −80°C until use. Unfed adult ticks (N = 50 females per group, two independent groups) were microinjected with 0.5 μL HSP70 dsRNA (about 1 μg RNA) at the base of the fourth right leg of the ventral surface of ticks. Control ticks were microinjected with unrelated luciferase dsRNA. After microinjection, ticks were maintained in an incubator at 25°C and 95% humidity for 24 h and then allowed to feed on rabbits' ears. Four female ticks per group were collected at 5 days of blood feeding for RNA and protein extraction to characterize gene knockdown by real-time quantitative PCR and Western blot with respect to luciferase control. Remaining ticks were allowed to feed until engorgement, which tick mortality, engorgement rate and engorgement weight was determined in individual engorged female ticks.

The primers and amplicon length of target genes used for real-time quantitative PCR are listed in Supplementary Table 3. The first-stranded cDNA was synthesized from 1 μg RNAs with the PrimeScript RT reagent kit (Perfect Real Time) (Takara, Shiga, Japan) in the following program: 37°C for 15 min, 85°C for 7 s, and finally 4°C. Quantitative real-time PCR was performed using SYBR Premix Ex Taq (Takara, Japan) with QuantStudio 5 (ABI, Waltham, USA) in 20 μL reaction mixtures including 0.4 μL ROX Dye II. The reactions were incubated for 30 s at 95°C, followed by 40 cycles at 95°C for 5 s, 60°C for 34 s, and analysis of a melting curve. Tick elongation factor 1α (ELF1α) was used as an internal control because of its stability in R. microplus and R. appendiculatus ticks (Nijhof et al., 2009). All reactions were performed in triplicate. The 2^−ΔCt method was used to calculate relative expression levels.

The body weights of engorged ticks were expressed as the mean ± standard error of mean (SEM) and analyzed using Student t-test. Differences in the attachment rate, engorgement rate, and death rate of ticks in different groups were tested using the Chi-square test. Significance of quantitative real-time PCR results was expressed as means ± SEM and determined with the one-way analysis of variance (ANOVA), and statistical analysis was conducted using software SPSS version 20.0. P < 0.05 and 0.01 were considered to be statistically significant (^*) and extremely significant (^**), respectively.

RH36 protein was expressed exclusively in the salivary glands, not in the midguts, ovaries or fat bodies of 5-day-fed ticks and engorged ticks (Figure 1A). Additionally, RH36 protein was expressed at higher levels in 7-day-fed and saturated ticks and subsequently decreased after engorgement, which was highly consistent with the time of tick ovarian development (Figure 1B). To confirm our hypothesis that RH36 modulating tick ovary development via vitellogenesis, four vitellogenin genes (named RHVg1, RHVg2, RHVg3, and RHVg4, respectively) were cloned from ovarian transcriptome from the post-engorgement R. haemaphysaloides female ticks (Supplementary Figure 1). These sequences are highly homologous in Haemaphysalis longicornis and Dermacentor variabilis (Supplementary Figure 2). RHVg1, RHVg2, RHVg3, RHVg4 sequences were available in the GenBank databases under the accession numbers: MK584634, MK584635, MK584636, MK584637, respectively. In addition, RHVg1 (Supplementary Figure 3A), RHVg2 (Supplementary Figure 3B) and RHVg3 (Supplementary Figure 3C) genes were widely distributed in the different development stages of ticks and RHVg4 (Supplementary Figure 3D) was detected only in engorged adults not larvae and nymphs by real-time quantitative PCR and semi-quantitative PCR (Supplementary Figure 3E). What's more, RHVg2 and RHVg4 was expressed in the fat body and ovary of engorged adult ticks, respectively (Supplementary Figure 3F).

Vitellin was purified from crude egg extraction by gel filtration (Sepharose CL-4B) and ion exchange chromatography on DEAE-cellulose and identified by possible polypeptide bands with molecular weights of 43.0, 47.3, 64.4, 72.6, 79, 85.9, 94, 102.3, 108.7, 136.5, and 161.5 kDa, respectively (Supplementary Figure 4A), in which two polypeptides (136.5, 72.7 kDa) were identified by LC-MS/MS (Supplementary Tables 4, 5). Monoclonal antibodies were generated and had specific immunological reaction with two polypeptides (136.5, 72.7 kDa) of vitellin (Supplementary Figure 4B). Vitellin proteins were highly expressed in fat body, hemolymph and ovary of adult ticks after engorgement (Figure 2).

After RH36 gene silencing, there was no difference for four vitellogenin genes in the fat body of adult ticks (Figures 3A,C,D). There was one exception: RHVg2 gene was up-regulated significantly at 7 days after engorgement (p < 0.01, Figure 3B). In the ovaries, although there was no difference in the transcript level of RHVg1 (Figure 4A), RHVg2 (Figure 4B) and RHVg3 (Figure 4C) genes from blood feeding to post-engorgement, RHVg4 gene was down-regulated significantly at 7 days or 10 days post engorgement (p < 0.01, Figure 4D). In addition, vitellin (the band of 136 kDa) protein decreased in the ovary of adult ticks at 3 or 10 days after engorgement, but not at 5 days of blood feeding (Figure 4E). Compared to mature oocytes from luciferase gene-silenced adult ticks, immunofluorescence showed that vitellin protein was less expressed in the cytoplasm of immature oocytes in response to RH36 gene-silencing (Figure 4F).

To identify which proteins regulate ovary development, ovaries from adult female ticks at 5 days blood feeding and 10 days post-engorgement were selected to conduct TMT proteomics analysis. Because of significant difference in total proteins found in ovaries between blood feeding and post-engorgement (Supplementary Figure 5), the TMT Mass-Tagging mass spectrometric (2, 6-plex) results were analyzed using MASCOT engine version 2.2. Ovary proteomics analysis resulted in the identification of 5,326 and 1,685 unique proteins in ticks at 5 days feeding and 10 days post-engorgement respectively. In the 5-day-fed ovaries, there were 44 proteins statistically significant (p < 0.05 and fold change > 1.2). Among these proteins, 25 and 19 proteins were up- and down-regulated, respectively (Figure 5A and Supplementary Table 6). In the day 10 post-engorgement ovaries, RH36 gene knockdown resulted in 208 proteins with statistically significant production differentiation (p < 0.05 and fold change > 1.2). Among these proteins, 74 were up-regulated and 134 were down-regulated (Figure 5A and Supplementary Table 7). According to GO and KEGG analyses, differentially expressed proteins mainly correlated with cellular process, developmental process and metabolic process with function of lipid transporter activity, catalytic activity and binding in the ovaries at 5 days feeding (Figure 5B) and 10 days post-engorgement (Figure 5C), in which vitellogenin (Supplementary Figure 6A) and HSP70 (Supplementary Figure 6B) involved in protein export and protein processing in endoplasmic reticulum, respectively (Figures 5D,E).

Since HSP70 protein was very important in tick blood-sucking and ovary maturation, we obtained HSP70 gene sequence (GenBank accession numbers: MK584638) from the ovary transcriptome (Table 1) and found that the open reading frame (ORF) of HSP70 was 1,980 bp long, encoding a deduced protein with 660 amino acids in length, a predicted molecular mass of 72.5 kDa and a pI of 5.32. It was also found that HSP70 protein included three signature sequences and the extreme C-terminal endoplasmic reticulum targeting sequence (KDEL, Supplementary Figure 7A). As shown in Supplementary Figure 7B, HSP70 from R. haemaphysaloides was highly conserved with over 83% identity in arthropods and mammals, in which it was highly homologous in Haemaphysalis flava and Ixodes scapularis with the identity of 96.21 and 91.06%, respectively. HSP70 protein was widely distributed in the hemolymph, ovary, salivary gland and fat body of post-engorgement ticks, with highest levels in the ovary at 7 days blood feeding and in the circulating hemolymph at 3 days post-engorgement (Figure 6).

After RH36 gene knocking down, there was no difference for HSP70 gene expression in the fat body, salivary gland, midgut, ovary and hemolymph of 5-day-fed and 7-day-fed adult ticks (Figures 7A,B, respectively). On the day of engorgement, HSP70 gene was down-regulated significantly in the ovary, midgut and salivary gland of adult ticks (p < 0.05, Figure 7C), coinciding with RH36 down-regulation expression. At 3 days post-engorgement, HSP70 gene also decreased markedly in the fat body, ovary, midgut, salivary gland, and hemolymph of adult ticks (p < 0.05, Figure 7D). In addition, HSP70 gene expression reduced observably in the ovary of adult ticks at 10 days post-engorgement (p < 0.01, Figure 7E). From tick blood-sucking to engorgement, HSP70 protein level decreased in the ovaries of adult ticks from RH36 RNAi groups by Western blot (Figure 7F). These results suggested that RH36 promoted expression of HSP70 protein in the post-engorged adult ticks.

HSP70 mRNA decreased by 73.86% (Figure 8A) and HSP70 protein was significantly down-regulated after microinjecting HSP70 dsRNA (p < 0.05, Figure 8B). After HSP70 gene silencing, mean of 48 h attachment rate and average engorgement rate was 52.22 and 2.56%, respectively. Meanwhile, these parameters showed 88.89 and 80.39% in the luciferase RNAi groups, respectively, which revealed that down-regulation of HSP70 gene decreased tick blood feeding and engorgement rate (Table 2). Although ticks attached to rabbit ears after HSP70 gene silencing (Figure 8C), the engorged body weight of ticks decreased and death rate increased compared to the luciferase RNAi groups (p < 0.05, Table 2 and Figure 8D), which shows that HSP70 gene facilitates abundance of tick blood feeding and protects ticks from death. After HSP70 gene knockdown, two vitellogenins (RHVg1 and RHVg2, Figure 9A) and vitellin (Figure 9B) were observably down-regulated (p < 0.05), indicating that HSP70 protein regulated expression of vitellogenin proteins via promoting tick blood feeding.

As shown in Figure 10, RH36 gene expressed-uniquely in salivary gland controlled tick blood feeding and modulated tick ovary development and oviposition by regulating the expression of vitellin in ovary, while HSP70 protein was down-regulated significantly in the ovaries of post-engorged ticks after RH36 gene silencing by real-time quantitative PCR and Western blot (p < 0.05). HSP70 gene positively modulated tick blood feeding and ovary development via nutrition uptake. Therefore, we speculated that RH36 gene silencing hindered tick blood feeding and oocytes maturity by inhibiting vitellin uptake mediated by HSP70 protein. In addition, HSP70 gene silencing considerably increased tick death rate during blood feeding.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.