Novel pseudo-aspartic peptidase from the midgut of the tick Rhipicephalus microplus

The characterization of Rhipicephalus microplus tick physiology can support efforts to develop and improve the efficiency of control methods. A sequence containing a domain with similarity to one derived from the aspartic peptidase family was isolated from the midgut of engorged female R. microplus. The lack of the second catalytic aspartic acid residue suggest that it may be a pseudo-aspartic peptidase, and it was named RmPAP. In this work we confirm the lack of proteolytic activity of RmPAP and investigate it’s non-proteolytic interaction with bovine hemoglobin by Surface Plasmon Resonance and phage display. Moreover we carried out RNAi interference and artificial feeding of ticks with anti-RmPAP antibodies to assess it’s possible biological role, although no changes were observed in the biological parameters evaluated. Overall, we hypothesize that RmPAP may act as a carrier of hemoglobin/heme between the tick midgut and the ovaries.


Results
Amplification and cloning of the RmPAP ORF. The complete nucleotide sequence (Sup. Figure 1) of the R. microplus pseudo-aspartic peptidase was amplified from the midgut of the engorged females. The amino acid sequence derived from the translation of RmPAP mRNA revealed the presence of a putative signal peptide (M 1 -A 20 ) and the lack of a second catalytic Asp residue (Fig. 1). The mature protein (R 21 -K 361 ) had a theoretical pI of 5.76 and a molecular weight of 37.3 kDa. A mutant form (Pro 242 > Asp 242 ) was generated to restore the proteolytic activity (Sup. Figure 2).

Expression and localization of RmPAP in R. microplus tissues. RmPAP expression was observed
mainly in the midgut of partially ( Fig. 2A) and fully fed females (Fig. 2B). The comparison of the levels of expression between partially and fully fed females demonstrated that RmPAP expression was up-regulated in three tissues that were analyzed, including the midgut (30-fold greater), ovary (35-fold greater) and salivary glands (8-fold greater) (Fig. 2C). Western blot assays using purified anti-RmPAP antibodies (Sup. Figure 3) revealed the presence of a minor 25 kDa product in the midgut and a major product of approximately 40 kDa in the ovaries of engorged ticks (Fig. 2D).

Expression and purification of recombinant RmPAP WT and RmPAP MUT . Protein expression was
tested in different bacterial strains with a wide range of temperatures, IPTG concentrations and induction times, but at all conditions tested, both recombinant proteins were obtained in insoluble form and become soluble only in the presence of 8.0 M urea (data not shown). After protein purification ( Fig. 3A and B), a major protein product of 36 kDa was observed (Fig. 3C), and after refolding RmPAP WT was observed to have a mass of 36 kDa while RmPAP MUT was observed to have a mass of 32 kDa (Fig. 3C).
Interaction of RmPAP with bovine hemoglobin. After refolding, wild-type RmPAP showed no proteolytic activity towards bovine hemoglobin (Fig. 4A), while the site-directed mutation (Asp 242 ) was demonstrated to restore proteolytic activity (Sup. Figure 4). Preliminary data from native-PAGE using RmPAP WT and  bovine hemoglobin revealed a possible interaction between the two molecules (Sup. Figure 5A). To further investigate this interaction, SPR experiments were conducted, which found a strong affinity (K D = 3.35 × 10 −8 M) of RmPAP WT for bovine hemoglobin (Fig. 4B). To assess the specificity of rRmPAP, a screening against a hexapeptide library was performed using phage display. After three rounds of selection against rRmPAP (Sup. Table 2), the phages were sequenced and a high prevalence (21%) of the V-V-K-G/E-Q peptide was found (Fig. 4C).

Determination of the biological effects of RNA interference and anti-RmPAP antibodies.
To verify the possible role of RmPAP in ovary physiology, the inhibition of its activity was carried out during artificial feeding using both RNA interference and anti-RmPAP antibodies. Gene silencing was confirmed by qPCR in both the midgut (60% reduction) and ovaries (80% reduction) 48 hours post-injection (hpi) (Fig. 5A and B),   although no difference in egg mass was observed (Fig. 5C). Likewise, the biological parameters that were assessed post-artificial feeding, such as weight gain (Fig. 6A), total egg mass ( Fig. 6B) and egg hatching (Fig. 6C), also presented no significant differences.

Discussion
Aspartic peptidases are characterized by the presence of two aspartic acid residues that are required for their catalytic activity 6 and, in ticks, they are typically located in the midgut 29,30 and involved in protein degradation 7,11 . The advances in sequencing technologies shred light into a new class of molecules known as pseudoenzymes. Pseudoenzymes are proteins with high similarity to functioning enzymes but that lack residues that are key to their catalytic activity 18,31 . Despite the absence of catalytic activity, pseudoenzymes have emerged as important regulators of different physiological processes. In this report, we describe the characterization of the first pseudo-aspartic peptidase from the tick R. microplus and its possible functioning as a hemoglobin carrier. The domain analysis of the deduced RmPAP amino acid sequence reveals its similarity to other conserved aspartic peptidases, while BLASTp revealed its similarities to other aspartic peptidases in ticks (Fig. 1). However, RmPAP lacks the second catalytic aspartic acid residue suggesting the absence of proteolytic activity.
RmPAP is a major transcript in the midgut of both partially and fully fed females and appears to be up-regulated as digestion progresses, although the native protein was observed only in the ovaries of fully engorged females. A similar expression profile was observed for other aspartic peptidases, such as BYC and THAP. Both of the enzymes were shown to be expressed in extra-ovarian tissues (midgut and fat body), but the native protein was detected only in the hemolymph and ovaries 9,10,12 . It suggests that the enzymes may be produced and secreted to the hemolymph and carried to the ovaries, where they are accumulated. Since the same pattern was observed for RmPAP, it's tempting to suggest the same mechanism and is thus more relevant to ovary physiology than blood digestion.
Pseudoenzymes are a newly discovered group of molecules that have been identified with advances in sequencing techniques. These molecules' sequences usually resemble that of an archetypical enzyme but lack enzymatic activity 32 due to mutations that disrupt or occlude the original catalytic site 18,31 . It's important to note that mutation of the catalytic residues is not proof of the absence of enzymatic activity. In Plasmodium falciparum, a protein containing an aspartic peptidase domain that also lacked the second catalytic Asp residue showed proteolytic activity towards hemoglobin, and a His residue was found to play an important role in this activity 33,34 .  The aspartic peptidase from R. microplus, BYC, also lacks one of the Asp residues but is still able to process bovine hemoglobin 14 , indicating the existence of a-yet-to-be-discovered enzymatic mechanism. Mature RmPAP was obtained from bacterial inclusion bodies and, after purification and refolding, it maintained its 37 kDa protein size (Fig. 3 -Lane 2). Moreover, when incubated with bovine hemoglobin, RmPAP WT did not display proteolytic activity (Fig. 4A). To verify if the lack of RmPAP WT activity was due to the absence of the second Asp residue, a mutant (Pro 242 > Asp 242 ) was generated and processed using the same conditions. After refolding, RmPAP MUT appeared as a 32 kDa protein band, suggesting that N-terminal processing by auto-activation occur and incubation with bovine hemoglobin resulted in its degradation (Sup. Figure 4). Overall, these data strongly suggest that RmPAP WT lacks the expected proteolytic activity and can be considered a pseudoenzyme. It is believed that pseudoenzymes originate from gene duplications of active enzymes followed the accumulation of mutations that render them inactive; despite this, they maintain some of the functional characteristics from their ancestors. Preliminary analysis using native-PAGE (Sup. Figure 5A) suggested that RmPAP WT could alter the migration profile of bovine hemoglobin, and SPR data revealed a strong affinity (K D = 3.35 × 10 −8 M) (Fig. 4B) between the two molecules. This verified that RmPAP could bind, in a non-proteolytic fashion, to bovine hemoglobin, most likely due the presence of a conserved structural feature typical of aspartic peptidases. Since the post-refolding protein yield was inadequate for structure studies, as well as the feasibility of high-throughput screening with phage display, we decided to investigate the possible binding site(s) and specificity of RmPAP WT using a hexapeptide phage display library 35 . After two rounds of selection, an enrichment of 34-fold was observed (Sup. Table 2), indicating that specific phages were selected in the presence of RmPAP WT . The analysis of the selected phages reveals that a high proportion of the peptide V-V-K-G-V/E-Q (Fig. 4C). Interestingly, the VKG peptide can also be found in an exposed segment of the bovine hemoglobin chain A α-helix (Sup. Figure 5B), which could serve as a possible binding site for RmPAP WT .
To assess the possible role of RmPAP in ovary physiology, both gene silencing through RNA interference (Fig. 5) and the artificial feeding of partially-fed females with anti-RmPAP WT antibodies (Fig. 6) was conducted. No differences were observed in the biological parameters that were evaluated. It suggests that the level of RmPAP was expressed before RNAi treatment and it was enough to maintain its biological functioning and/or that it has a low protein turnover rate, which would work to conceal the effects of RNAi silencing that were observed in other RNA interference experiments 36 . Moreover, bioinformatics analysis of the R. microplus transcriptome (SRA accession numbers: SRX484287, SRX484284, SRX484280 and SRX484277) revealed the presence of other contigs that are similar to RmPAP and lack the second aspartic catalytic residue (data not shown). Therefore, it's tempting to hypothesize that the physiological activity of RmPAP may be compensated by these other molecules and that the lack of any one of these molecules may not suffice to produce a distinctive phenotype. Nevertheless, further studies are necessary to characterize the physiological roles of RmPAP.
In this study, we investigated a novel pseudo-aspartic peptidase from R. microplus females and found that RmPAP was able to bind to bovine hemoglobin in a non-proteolytic fashion with its possible binding site characterized. The high expression of RmPAP was observed in the midgut of both partially and fully fed females, although the protein appeared to accumulate in the ovary. Bearing those data in mind, we hypothesized that RmPAP may act as hemoglobin/heme carrier between the midgut and the ovaries and thereby contribute to ovary maturation. RNA extraction and cDNA synthesis. The ticks were washed with 70% ethanol followed by ultrapure water and dissected. The midgut, ovary, salivary glands and hemocytes were collected and added to Trizol (Invitrogen, CA, USA). RNA exctration was conducted according to the manufacturer's instructions. The RNA was treated with DNAse I (Fermentas, Vilnius, LT) for 1 h at 37 °C and 1 μg was used for cDNA synthesis using the Improm-II Reverse Transcription System (Promega, WI, USA).

Amplification and cloning of the RmPAP ORF. Primers (Sup. Table 1) containing the Xho I (sense)
and Bpu1102I (anti-sense) restriction sites were designed based on the RmPAP WT (Rhipicephalus microplus pseudo-aspartic peptidase wild type) nucleotide sequence (Sup. Figure 1), obtained from a R. microplus transcritpome (SRA SRX484287, SRX484284, SRX484280 and SRX484277). PCR was performed using 1 μL of midgut cDNA, 100 μM dNTPs, 1.5 mM MgCl 2 , 5 U Taq DNA polymerase (Sinapse, SP, BR) and 25 pmol of each primer. The reactions were subject to an initial denaturation at 94 °C for 10 min followed by 25 cycles of 94 °C -30 s, 55 °C -60 s, and 72 °C -60 s with a final extension at 72 °C for 10 min. The PCR products were analyzed using a 1% agarose gel and purified with the QIAEXII extraction kit (QIAGEN, Hilden, DE). The purified PCR products were later cloned into a pET14b vector containing a N-terminal His tag.
Primary structure analysis. A domain search was conducted using PFAM (https://pfam.xfam.org/) 37 and the signal peptide was identified using SignalP 4.1 (http://www.cbs.dtu.dk/services/SignalP/) 38 . The theoretical molecular weight and pI were estimated by the Compute pI/MW tool (https://web.expasy.org/compute_pi/) 39 . The sequence alignment was performed using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/) 40 Table 1). The PCR product was analyzed in 1% agarose gel and purified with the QIAEX II extraction kit (QIAGEN, Hilden, DE). The purified DNAs were mixed (1:1) and used as the template for a second PCR using the RmPAP.FW and RmPAP.RV primers. The resulting PCR product was purified and cloned into the pET14b vector.
RmPAP expression and purification. The recombinant RmPAP WT and RmPAP MUT proteins were expressed in the Escherichia coli BL21 plys S at 37 °C with IPTG (1 mM). After 16 hours of induction, the culture was centrifuged (10 min, 3000 × g, 4 °C) and the cells were resuspended in 50 mM Tris-HCl (pH 8.0). Bacterial lysis was conducted by 3 cycles of French press. The samples were centrifuged (20 min, 12.000 × g, 4 °C), the supernatant collected, and the pellet washed with 50 mM Tris-HCl (pH 8.0) containing urea (2,4,6 Western blot analysis. Protein was extracted from the midgut and ovaries of engorged R. microplus females using Trizol reagent (Invitrogen, CA, USA) according to the manufacturer's instructions, and 10 µg of the total protein was separated using SDS-PAGE (12%). Proteins were transfer to a PVDF membrane using a Mini Trans-Blot Cell system (BioRad) for 1 h at 15 V. After transfer, the membrane was incubated for 2 h in blocking solution (PBS containing 0.1% Tween -PBS-T -and 5% skim milk) at room temperature, followed by incubation with purified anti-RmPAP antibody, diluted 1:10 in blocking solution, overnight at 4 °C (Sup. Method 1). The PVDF membrane was then washed 3 times with a PBS-T 0.1% and incubated with anti-rabbit IgG conjugated with peroxidase (1:5000) in blocking solution. After 2 h of incubation, the SuperSignal West Pico Chemiluminescent substrate (Pierce, IL, USA) was added and the membrane incubated for 10 min at room temperature. Imaging was performed using the MR-ChemBis 3.2 (DNR Bio-imaging System) by exposing the membrane to UV light for 3 min.

Determination of RmPAP activity towards bovine hemoglobin. Refolded RmPAP WT and RmPAP MUT
(2.0 µg) were incubated with bovine hemoglobin (5 µg) in 50 mM phosphate-citrate buffer (pH 2.5-6.0) for 4 h at 37 °C and analyzed using SDS-PAGE (15%). The binding of RmPAP WT to bovine hemoglobin was measured by surface plasmon resonance (SPR) using a Biacore T-200 system. Bovine hemoglobin (2000 RFU) was immobilized on a CM5 series chip (FC 2) in acetate buffer (pH 5.5), while BSA was immobilized in FC 1. SPR experiments were conducted by injecting increasing concentrations of RmPAP WT (10 mM phosphate-citrate buffer, pH 4, with 0.15 M NaCl) at 20 µL/min, with association and dissociation times of 300 sec and 900 sec, respectively. The equilibrium constant was determined by plotting the intensity of the steady-state response (FC2 -FC1) against the RmPAP concentration using the Biacore T200 evaluation software (GE Healthcare).
Peptide library screening using the phage display system. A hexapeptide library 35 was used to determine recombinant RmPAP WT specificity. E. coli TG1-transformed cells were grown in 2YT medium containing ampicillin (200 μg/mL) and 2% glucose until the OD 550 reached 0.5-0.7. The helper phage M13K07 was added at a multiplicity of infection of 50 and the medium replaced with 2YT containing ampicillin (200 μg/mL) and kanamycin (50 μg/mL). After 16 h of incubation at 37 °C, the fusion phage particles were screened with recombinant RmPAP WT . RmPAP WT was adsorbed to a 96-well plate overnight at room temperature following blocking for 2 h at room temperature with PBS-T 0.005% (pH 7.4) and 2% BSA. Entry phages, pre-incubated with blocking solution (1:1), were added and incubated for 1.5 h at 30 °C following 10 washes with PBS-T 0.1% (pH 7.4). The elution was performed with 0.2 M KCl (pH 2.0) followed by neutralization with 1.0 M Tris-HCl (pH 8.0). The eluted phages were then used for E. coli TG1 transfection and subsequent amplification and titration. After 3 rounds of selection, 40 phagemids were randomly selected and sequenced. The translated peptides were represented using the WebLogo tool 43 . Silencing of the RmPAP gene via RNA interference. Double-stranded RmPAP RNA (dsRmPAP) was synthesized using the T7 Ribomax Express System (Promega, WI, USA). Engorged R. microplus females were injected with 4.0 μg of dsRmPAP or dsGFP and dissected 48 h post-injection (10 ticks). The midgut and ovaries of dissected ticks were added in Trizol reagent (Invitrogen, CA, USA) for RNA extraction and cDNA synthesis, while 15 ticks were used for egg laying analysis. RmPAP knockdown was confirmed by qPCR.
In vivo effects of the ingestion of antibodies against RmPAP in partially-fed R. microplus females. Partially fed R. microplus adult females were recovered from calves 20-21 days after larvae infestation. Groups of 30 ticks weighing 25-50 mg were trapped and artificially fed with capillary tubes filled with 50 µL of bovine blood every 2 h for 18 h 44 . The first two feeding cycles contained purified anti-RmPAP or antibodies from non-immunized rabbits (final concentration of 3.5 mg/mL). The biological parameters analyzed were weight gain (initial weight/post-feeding weight), egg production (weight of eggs/initial weight of ticks) and egg hatching (larvae mass/egg mass). Statistical analysis. The comparison of RmPAP expression among different tick tissues was performed using the Kruskal-Wallis test with Bonferroni's multiple comparison post hoc test 45 . The comparison between partially and fully fed ticks was performed using the Mann-Whitney test. RmPAP knockdown was analyzed with Mann-Whitney test and the biological parameters were analyzed using Student's two-tailed t-test. Analyses were conducted with the Graph Pad Prism 6.0 software (GraphPad Software, Inc.), and differences were considered to be statistically significant when p < 0.05.