In Silico Characterization and Structural Modeling of Dermacentor andersoni p36 Immunosuppressive Protein

Ticks cause approximately $17–19 billion economic losses to the livestock industry globally. Development of recombinant antitick vaccine is greatly hindered by insufficient knowledge and understanding of proteins expressed by ticks. Ticks secrete immunosuppressant proteins that modulate the host's immune system during blood feeding; these molecules could be a target for antivector vaccine development. Recombinant p36, a 36 kDa immunosuppressor from the saliva of female Dermacentor andersoni, suppresses T-lymphocytes proliferation in vitro. To identify potential unique structural and dynamic properties responsible for the immunosuppressive function of p36 proteins, this study utilized bioinformatic tool to characterize and model structure of D. andersoni p36 protein. Evaluation of p36 protein family as suitable vaccine antigens predicted a p36 homolog in Rhipicephalus appendiculatus, the tick vector of East Coast fever, with an antigenicity score of 0.7701 that compares well with that of Bm86 (0.7681), the protein antigen that constitute commercial tick vaccine Tickgard™. Ab initio modeling of the D. andersoni p36 protein yielded a 3D structure that predicted conserved antigenic region, which has potential of binding immunomodulating ligands including glycerol and lactose, found located within exposed loop, suggesting a likely role in immunosuppressive function of tick p36 proteins. Laboratory confirmation of these preliminary results is necessary in future studies.


Introduction
Ticks are considered among the most important vectors of livestock diseases worldwide as well as major vectors of pet diseases [1]. In tropical Africa, ticks and the tick-transmissible diseases constitute a major obstacle to livestock development [2]. Like elsewhere in the world, chemical acaricides have been the mainstay of tick control in this region; however, increasing resistance to this group of insecticides threatens livestock production systems, especially small-holding sectors that rely on rearing of exotic cattle breeds that are more susceptible to tick infestation and tick-borne diseases [TBDs] [3]. Integrated tick control incorporating reduced acaricide use, breeding cattle for tick resistance, rotational grazing, and use of vaccines presents a sustainable and long-term strategy to the control of ticks and TBDs in the tropics [4].
Numerous studies have shown the potential of immunological methods to control tick infestation by targeting critical tick physiological processes. Existing antitick vaccines work by eliciting humoral and cellular responses against tick cell membrane antigens [5,6]. Vaccines capable of quelling both the arthropod vector and disease-causing pathogens are also under development [7]. Despite clear advantages of controlling ticks through vaccination, this strategy is presently hampered by antigenic sequence variations between geographically isolated tick populations and species causing 2 Advances in Bioinformatics vaccine resistance in some regions [8] and lack of efficacy in others [9]. These limitations necessitate search of alternative antigens for inclusion in the next-generation tick vaccines.
Proteins found in tick saliva play critical roles during blood meal acquisition [10]. The pharmacologically active components secreted in their saliva help ticks circumvent host defenses such as haemostatic and immune responses of the host, thereby enabling blood feeding in hematophagous arthropods [11]. One such class of biological compounds is immunosuppressant proteins, which modulate the host's immune system during tick's blood feeding [12], making them suitable target in the search of novel vaccines against arthropod-transmitted diseases [13]. Low molecular weight proteins 5-36 kDa from tick saliva proteins have been shown to inhibit T-lymphocytes proliferation in vitro [14]. Active immunization of mice with Salp15, a 15 kDa secreted salivary gland protein from I. scapularis, showed substantial protection (60%) from tick-borne Borrelia [15]. Tick subolesin (SUB), the ortholog of insect and vertebrate akirins (AKR), was discovered as a tick protective antigen in Ixodes scapularis [16]. Vaccines containing conserved SUB/AKR protective epitopes have been shown to protect against tick, mosquito, and sandfly infestations, thus suggesting the possibility of developing universal vaccines for the control of arthropod vector infestations [17].
Protein antigens conserved across vector species could be used in developing cross-protective vaccines against multiple arthropod vectors and their associated pathogens [17,18]. Alarcon-Chaidez et al. [19] cloned and characterized a 36 kDa immunosuppressive protein p36 from the salivary glands of partially engorged, female D. andersoni, that suppressed Con-A induced in vitro proliferation of normal murine T-lymphocytes by more than 90% [20]. Genes related to D. andersoni-derived p36 gene, such as Ra-p36, Av-p36, Hl-p36, and Rhp36, have been reported in A. variegatum [21], R. appendiculatus [22], H. longicornis [23], and R. haemaphysaloides [24]. Most proteins are, however, not sufficiently protective on their own suggesting the need for a multiantigen/chimeric vaccine that incorporates critical tick and pathogen antigenic epitopes [16,25] to elicit synergistic antipathogen and antitick immune responses. Computational characterization and 3D structure modeling of D. andersoni p36 protein undertaken by this study is an initial step in understanding molecular basis of immune recognition which is a challenge in vaccine development [26]. The p36 conserved antigenic region predicted by this study has binding residues for ligands like glycerol and lactose which are associated with an immunomodulatory role suggesting this site may have a role in suppression of select T-cell receptor induced signaling events of D. andersoni p36 and its related proteins.

Sequence Characterization of Tick p36
Proteins. All tick proteins deposited in National Centre for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov) protein database were retrieved and deposited in a standalone MySQL based database (https://www.mysql.com).
D. andersoni p36 protein was used as reference sequence in conducting homology searches, Blastp [27] and OrthoMCL [28,29], of tick proteins in the database.
The identified tick p36-related proteins were subjected to MEME tool search (http://meme-suite.org/tools/meme) to predict conserved motifs characteristic of p36 proteins. Motif search tool (http://www.genome.jp/tools/motif) then searched for function of identified common motifs in the database of known motifs. Tick p36 protein sequences were then aligned by a multiple sequence alignment tool, Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo). Phylogenetic tree construction was by maximum likelihood method [30] and evolutionary distance computed using Poisson correction method [31]. Bootstrap resampling (1000 replicates) assessed robustness of the groupings.

Structural Modeling of D. andersoni p36
Protein. Physicochemical properties of D. andersoni p36 protein were analyzed by ExPASyProtParam (https://www.expasy.org) server while its secondary structure was characterized by online tool Spider 2 (http://sparks-lab.org/yueyang/server/ SPIDER2). The crystal or NMR structure of tick p36 protein is currently not available in the protein data bank (PDB) (https://www.rcsb.org/pdb/). The 3D structure of D. andersoni p36 protein was developed by QUARK ab initio modeling [33] that builds 3D structure from "Scratch," based on physical principles rather than previously solved structures. 10 models, designated as 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10, were generated and validated by analyzing Verify 3D scores [34] of Ramachandran plots for each model. Based on the scores, models 2 and 9 were selected as likely 3D structures of D. andersoni p36 protein because they scored 81.41% and 88.44%, respectively, meeting Verify 3D validation tool limit of 80% of the amino acids residues scoring >=0.2 in the 3D/1D profile.

Identification and Phylogenetic Analysis of p36
Proteins from Ixodid Ticks. The study identified 32 homologs of D. andersoni p36 protein among 6 ixodid (hard) tick species (Table 1). These included p36 genes reported in earlier studies from R. appendiculatus, A. variegatum, H. longicornis, and Rhipicephalus haemaphysaloides tick species as well as those found by this study in Amblyomma sculptum and Amblyomma aureolatum. Among these homologs, 4 coorthologs which are potential orthologs of D. andersoni p36 protein were identified in R. appendiculatus species ( Table 1). Occurrence of p36 protein across a range of tick species may be related to a biological function for this protein in tick feeding [36]. Several p36 immunosuppressant protein sequences were found in a single tick species suggesting functional and structural redundancy in which a tick expresses multiple similar proteins in minute quantities during feeding [37]. Such redundancy may render saliva proteins less immunogenic, as reported with cystatins [38].
The tick p36 proteins have 3 potential common motifs designated as 1, 2, and 3 with motif 2 being the only one conserved among p36 proteins ( Figure 1). Motif 2 is located between amino acid positions "107-127" in the reference D. andersoni p36 protein. This motif 2 may be associated with a functional domain, possibly a role in immunomodulatory activity of tick p36 proteins [39]. The 3 common motifs were not found in the motif database and could be representing an orphan protein family [40].
Alignment of tick p36 proteins revealed a conserved region occurring between amino acid positions "107-115" ("IDKGMLSPF") in the reference D. andersoni p36 protein ( Figure 2). This region that coincides with location of conserved motif 2 has polar amino acid residue serine (S) and charged residues lysine (K) and aspartate (D), which are associated with potential active sites [41]. Phylogenetic tree ( Figure 3) showed that, among homologs with higher amino acid percentage similarity and E-scores, D. andersoni was closely related to homologs from R. haemophysaloides and R. appendiculatus as compared to homolog from A. variegatum indicating more recent ancestry between Dermacentor and Rhipicephalus than with Amblyomma genera as inferred by phylogeny [42].

Identification and Characterization of Antigenic Regions in the p36
Proteins. Most tick p36 proteins were predicted as secretory with signal peptide cleavage site at position 21-22 (Supplementary Table S1 and Figure S1). Secretory proteins are favoured candidates for vaccine development as they are easily accessible microbial antigens to the immune system [43]. D. andersoni p36 protein and most p36 variants were predicted as antigenic with several homologs having antigenicity score above 0.7000 (Table 2, Supplementary Table S1), surpassing the vaxijen tool threshold of 0.5000. JAP81944.1, a homolog in R. appendiculatus had antigenicity score of 0.7701, comparably higher than that of Bm86 (0.7681), the constituent antigen of Tickgard and Gavac6 commercial tick vaccines. Whether this theoretically predicted immunogenicity can confer protection against tick infestation there is need to be evaluated empirically through an immunization/tick challenge set up.
The potentially conserved motif 2 in p36 protein was predicted as a likely epitope-rich antigenic region with binding residues for glycerol and lactose ligands which are associated with an immunomodulatory role [44,45]. To facilitate tick feeding a single tick saliva protein ligand may bind receptors on several immune cell types in the vertebrate host; alternatively, multiple tick saliva proteins may bind to a common receptor [37].

3D Structure of D. andersoni p36
Protein. D. andersoni p36 protein has an instability index of 35.53 and GRAVY score of −0.324 classifying it as a stable, globular protein [46]. The protein's high aliphatic index of 86.41 is associated with increase in thermostability of globular protein [47]. The stable secondary structures alpha-helix ( ) and betasheets ( ) comprised approximately 55% of D. andersoni p36 protein amino acid sequence (Supplementary Figure S2). The predicted conserved immunogenic region "74-107" in processed secretory D. andersoni p36 protein had several segments within loop regions where epitopes are generally found [48]. The combination of -helixes and -structures through loops with specific geometric arrangements with respect to each is responsible in forming conserved structural motifs [49,50].
Based on Verify 3D [34] scores of the 10 models designated as 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 generated for D. andersoni p36 protein; models 2 and 9 were selected for further validation as they passed tool limit of 80% of the amino acids residues scoring >=0.2 in the 3D/1D profile (Supplementary Table S2). Comparison of validation scores of the selected models 2 and 9 ( Table 3, Supplementary  Figures S3 and S4) identified model 2 as the best 3D structure of D. andersoni p36 protein.
The predicted 3D structure of D. andersoni p36 protein (Figures 4(a) and 4(b)) is a ball-like structure comprised of 1 alpha-helix and several antiparallel beta-strands. The region predicted as a likely conserved antigenic region "74⋅ ⋅ ⋅ 107" in D. andersoni p36 protein is not only located in between the alpha-helix and beta-strands but also occurs within the potentially groove region of the predicted 3D structure and further has its loop exposed on the protein surface   (1)

ProQJPred
LGscore LG score > 1.5 fairly good model LG score > 2.5 very good model LG score > 4 extremely good model LGscore 2.914 LGscore  ( Figure 4(c)). Ligands bind in the largest cleft in over 83% of the proteins [51]; thus presence of the predicted conserved antigenic region within this potential groove may be associated with immunosuppressive function of D. andersoni p36 protein, as internal cavities in proteins are important structural elements that may produce functional motions such as ligand binding [52].
Potentially exposed loop region "87⋅ ⋅ ⋅ 94" (Figure 4(d)) in predicted 3D structure of D. andersoni p36 protein coincides with its likely conserved alignment region "107⋅ ⋅ ⋅ 115"   Figure 2: Multiple alignment of p36 homologous amino acid sequences showing likely conservation region. In the case of reference D. andersoni p36 protein, the conserved region "IDKGMLSPF" is located at positions "107-115." (:) and (.): marks conservation between groups of strongly or weakly similar properties, respectively. Note. Amino acids colour according to physicochemical properties: red is for small hydrophobic, blue for acidic, magenta for basic, and green for hydroxyl/sulfhydryl/amine amino acid residues. Highlighted region shows conservation in tick p36 proteins. after cleavage of signal peptide at amino acid position 21-22. This suggested loop region might be associated with binding site of D. andersoni p36 protein. The ligands predicted with potential to bind on this site include fatty acid glycerol and sugars like lactose. The hydroxyl group of polar amino acid residue serine (S), hydrophobic amino acid residue leucine (L), and charged amino acids lysine (K) and aspartic acid (D) found in this region could, respectively, have a role in binding of these ligands [53]. Immunomodulator ligands predicted with potential of binding at this site include fatty acid glycerol and sugars like lactose. There is need for future studies to evaluate whether immunomodulator ligands have a role in suppression of select T-cell receptor (TCR) induced signaling events in D. andersoni p36 protein mode of action [44,45].
Collectively results from this in silico study provide further insight into potential characters of p36 protein, which is vital in exploiting the proteins as targets for developing improved next-generation cross-protective tick control approaches. In an effort to determine exact role of these proteins in tick feeding process, it is necessary for future laboratory and animal studies to confirm these preliminary predictive findings.

Conclusion
The p36 immunosuppressive proteins from ticks exhibit antigen traits worth evaluating in future experimental in vitro and in vivo trials. This includes potential conservation across several tick species and presence of a likely conserved antigenic region that may be bound by immunomodulator ligands such as glycerol and lactose. A further study is necessary on suitability of this potentially conserved region in development of a multi/chimeric antitick vaccine that incorporates critical antigenic regions. The predicted 3D model of D. andersoni p36 protein may be used as a template to model structures of other orphan proteins related to p36. This work is a step towards developing cross-protective next-generation antitick vaccines, as the results expand our knowledge of p36 tick saliva protein and lay ground for future studies to determine their exact role in tick feeding process, which is useful in designing blockade approaches targeting these proteins.

Disclosure
This research did not receive any specific grant from funding agencies in the public, commercial, or non-profit sector.

Conflicts of Interest
The authors declare that there are no conflicts of interest.