SDS-PAGE of saliva from female and male O. erraticus ticks
In the current study, we aimed to obtain and compare, qualitative and quantitatively, the proteome of the saliva of O. erraticus female and male ticks. For this, we prepared three replicated saliva samples from each sex (F1, F2, F3 from females and M1, M2, M3 from males), which were first examined by SDS-PAGE to check reproducibility and then subjected to LC-MS/MS and SWATH-MS analyses.
SDS-PAGE of saliva samples showed protein band patterns that were very similar within each sex and noticeably different between the sexes (Fig. 2). Both sexes displayed numerous bands with molecular weights ranging from more than 260 to 10 kDa. However, in female saliva, the more numerous and intense bands were in the range from 300 to 100 kDa, while the more numerous and intense bands in male saliva ranged from 100 to 10 kDa. These band patterns indicate good reproducibility among samples of the same sex and anticipate some differences in the saliva protein composition between the sexes.
Proteins identified by DDA LC–MS/MS
First, female and male saliva samples digested in-solution were analysed by LC-MS/MS in DDA mode in two separate pools. Up to 469 protein hits were detected in the male saliva pool (Table 1), but only 332 protein hits were detected in the equivalent female saliva pool (data not shown). This difference might be related to the different protein band patterns observed in SDS-PAGE between female and male saliva (Fig. 2). In females, proteins larger than 100 kDa seemed somewhat more abundant than the rest and could have masked the identification of the less abundant proteins (Fig. 2A). Thus, to achieve a better characterisation of the female saliva proteome, pooled female saliva samples were resolved by SDS-PAGE, fractionated in three slices (Fig. 2B) and each slice was analysed by LC-M/MS in DDA mode. This analysis yielded 470 female protein hits (Table 1). Because of the better results, we used the information obtained from the in-gel analysis for female protein annotation in the spectral library. Protein Pilot reports showing the spectrum, peptide and protein data in O. erraticus female and male ticks were generated and can be accessed in Additional file 1: Dataset S1 and Additional file 2: Dataset S2, respectively.
The information obtained from in-solution processed male saliva and from in-gel processed female saliva was combined to generate the reference spectral library for analysis of the SWATH-MS acquired data, which can be accessed in Additional file 3: Dataset S3.
The spectrometric and identification data extracted from these libraries are summarised in Table 1 and Fig. 3. Up to 470 and 469 protein hits were detected in female and male saliva, respectively. After eliminating the hits to non-annotated sequences in the O. erraticus sialotranscriptome database (102 in each sex, 21.7% of the identifications) and the redundant identifications (94 in females, 104 in males), we obtained two filtered lists of 274 and 263 non-redundant proteins from females and males, respectively. Of these, 152 proteins are common to both sexes, and 122 and 111 are unique to males and females, respectively, which equates to 385 unique salivary proteins (Fig. 3A). This result indicates that male and female saliva shows different protein composition, since apparently only 39.5% of the identified proteins are expressed by both sexes. Regarding the reference spectral library, up to 639 protein hits were obtained. After removing hits to non-annotated proteins (127, 19.9%) and redundant identifications (132), we obtained a list of 380 unique annotated proteins (Table 1). This library was used as a reference for the analysis of SWATH-acquired data. Merging these three identification lists resulted in 387 unique proteins identified in the saliva of O. erraticus adult ticks.
Additional file 4: Table S1 lists the 387 characterised proteins, including their quantification by SWATH-MS and their functional annotation and classification. The number of proteins identified herein in O. erraticus saliva is comparable to those found in recent proteomic studies of the saliva of other argasid ticks and reflects a similar level of complexity and functional redundancy [25, 28-29, 37]. These results represent a great improvement in the proteomic identifications of salivary proteins for this species, as only six salivary proteins of O. erraticus had been identified hitherto, in a former proteomics study using bi-dimensional gel electrophoresis and MALDI-TOF MS/MS [47].
Proteins detected by SWATH-MS
While DDA MS methods are based on the random selection and fragmentation of a fixed number of peptide precursors, which are generally the most intense peptide ions, in SWATH-MS data acquisition, all ionised peptides of a given sample that fall within a specified mass range (m/z) are fragmented in a systematic and unbiased fashion using rather large precursor isolation windows [42]. Several studies have shown that SWATH-MS may outperform DDA LC-MS/MS by increasing the sensitivity and reproducibility of protein and peptide identification across multiple replicates [25, 48-49]. Accordingly, we wanted to determine whether SWATH-MS might detect and quantify a higher number of proteins in O. erraticus male and female saliva than DDA LC-MS/MS, which would also allow protein expression levels to be compared between sexes.
Thus, individual female and male saliva samples were analysed by SWATH-MS and its performance was compared to that of DDA LC-MS/MS (Table 1, Fig. 3; Additional file 4: Table S1). In female saliva, 274 and 224 proteins were identified by DDA and quantified by SWATH, respectively (Table 1). Up to 173 of these proteins were detected by both methods, 101 exclusively by DDA and 51 exclusively by SWATH (Fig. 3B). In male saliva, 263 and 224 proteins were identified by DDA and quantified by SWATH, respectively (Table 1). Up to 187 male proteins were detected by both methods, 76 exclusively by DDA and 37 exclusively by SWATH (Fig. 3B).
Although SWATH-MS did not outperform DDA-MS in the total number of identified proteins (224 vs. 387), it is interesting to note that SWATH-MS detected more common proteins, i.e. expressed by both sexes, than DDA-MS did, namely 224 vs. 152 (Fig. 3A). This represents a 47.4% increase in the identification of common proteins and indicates that at least 57.9% (224 out of 387) of all proteins identified in O. erraticus saliva are expressed by both sexes, which reduces the qualitative differences observed by DDA-MS and rather suggests quantitative differences instead.
In the slow-feeding ixodid ticks, the difference in the saliva protein composition between females and males is well established [40, 50], and can be related to differences between sexes in their feeding behaviour, anatomy and functions of the salivary glands [51-53]. As opposed to ixodids, adult argasid ticks typically are fast-feeders, do not show anatomical or functional differences in their salivary glands between sexes and exhibit similar feeding behaviours as they ingest equivalent amounts of blood relative to their body weight during a similar time interval, usually less than one hour. This means that female and male argasid ticks face the same host defensive responses; therefore, it could be expected that both sexes would use the same repertoire of anti-defensive salivary proteins to complete feeding [43].
Despite the above hypothesis, the different band patterns shown in Fig. 2 and the different set of proteins identified by LC-MS/MS in female and male saliva suggest qualitative differences in the protein composition of O. erraticus saliva between both sexes. However, it must be noted that the O. erraticus sialotranscriptome used in the current study as a reference database for protein identification was obtained from female salivary glands only. Accordingly, it could be supposed that the majority of proteins identified in the present study, including those detected in males only, would also most likely be expressed in female saliva. Consequently, the differences observed between sexes might probably be due to quantitative differences in expression instead of the genuine absence/presence of concrete proteins in one or other sex. As previously suggested, and as we will see later, the quantitative results of SWATH-MS also lend support to this hypothesis.
Differences in saliva protein composition between the sexes in other argasid species were reported by Díaz-Martín et al. [43] for Ornithodoros moubata, which have been recently confirmed at the quantitative level by Oleaga et al. [25]. The reason behind these differences in argasids remains unknown, but it might be related to the post-feeding processing of the ingested blood and/or to attraction and mating [43, 54].
Functional annotation and classification of the proteins identified in the saliva of O. erraticus
The 387 unique proteins identified in the O. erraticus adult saliva were functionally annotated and characterised using the Gene Ontology (GO) terms and cross-references in the InterPro, Pfam and Panther databases associated with UniProt IDs (Additional file 4: Table S1).
Up to 223 proteins were assigned GO terms; these included 121 cellular components, 114 molecular functions and 208 biological processes, which were visualised using the Web Gene Ontology Annotation Plot (WEGO) [55]. Fig. 4 represents these proteins classified according to cellular component, molecular function and biological process, using level 2 GO terms. The cellular components were classified into 13 categories, of which the more abundantly represented were, jointly, cell and cell part (n = 136), organelle and organelle part (n = 59), membrane and membrane part (n = 58), and extracellular region and extracellular region part (n = 48). Classification by molecular function resulted in eight categories. The more abundantly represented by far were catalytic (n = 112) and binding activity (n = 107); the remaining categories were noticeably less represented and included molecular function regulator (n = 30), transporter activity (n = 15), antioxidant activity (n = 10), structural molecule (n = 1), molecular transducer (n = 1) and toxin (n = 1). The classification of biological processes resulted in 18 categories. The eight more abundant were cellular processes (n = 82), metabolic process (n = 78), localisation (n = 23), response to stimulus (n = 22), biological regulation (n = 16), cellular component organisation or biogenesis (n = 14), regulation of biological process (n = 12), and multi-organism process (n = 12). This GO distribution is similar to those recently reported for the sialotranscriptomes of O. erraticus and O. moubata females [24, 26].
The functional classification of the 387 proteins according to categories reported by Kim et al. [41] resulted in 24 functional groups and families (Table 2; Additional file 4: Table S1). Among them, the most numerous (i.e., with a higher number of proteins) were the proteins involved in metabolic processes (n = 71), proteases (n = 44), protease inhibitors (n = 32), lipocalins (n = 25), antioxidants (n = 22), and proteins with unknown function (n = 59), with the latter representing 15.2% of the proteins identified. Comparing between sexes shows that lipocalins, antioxidants and proteins involved in the metabolism of carbohydrates, energy and nucleic acids were more numerous in males than in females, while the remaining categories were either more numerous in females or equally numerous in both sexes (Table 2). Typically, these functional groups and families are also the most abundantly represented in the sialomes of the soft and hard tick species analysed to date [28-29, 37].
Quantification by SWATH-MS of the proteins identified in female and male saliva
As already noted, SWATH-MS is a type of DIA method of analysis used to evaluate quantitatively complex samples with high reproducibility [42].
Using this technique, we detected and quantified 224 proteins in the saliva of both female and male ticks, which were later classified into 23 functional groups and families (Table 2; Additional file 5: Table S2), and most of which coincide with the groups and families more abundantly represented in the sialotranscriptome of O. erraticus females [26]. The groups with the highest numbers of proteins quantified were proteins involved in metabolic processes (n = 43), protease inhibitors (n = 26), proteases (n = 24), lipocalins (n = 16), antioxidants (n = 16), transporters/receptors (n = 11), heme/iron binding (n = 10), cytoskeletal (n = 10), and those with an unknown function (n = 33).
The expression levels of these protein groups and families in the saliva of both sexes, calculated as the average spectral signal peak area for samples F1, F2 and F3 or M1, M2 and M3, are shown in Additional file 5: Table S2 and summarised in Fig. 5. The 33 proteins with unknown function have been excluded from the charts, and the groups containing six or fewer proteins (metabolism of nucleic acids and amino acids, protein modification, glycine rich, antimicrobial, extracellular matrix, nuclear regulation, signal transduction, proteasome machinery, transposon element and protein synthesis) have been merged into one group named “other”. Fig. 5 shows obvious differences in protein composition at the quantitative level between female and male saliva.
In female saliva, the most abundantly expressed functional category was heme/iron binding, as these proteins represented 22% of the protein mass in this fluid (Fig. 5A, 5B). The following more abundant functional categories were protease inhibitors (17.8%), proteases (14.7%), lipocalins (8.3%) and immune-related (6.7%). In contrast, the proteins involved in carbohydrate metabolism were the most abundant in male saliva (33.0%), followed by protease inhibitors (19.1%) and lipocalins (18.3%).
Heme/iron binding
The heme/iron binding group included 10 proteins: five vitellogenins (Vgs) and five hemelipoglyco-carrier proteins (CPs), with vitellogenins being the most abundant as they accounted for 73.6% of the protein mass of this group (Table 2; Fig. 5; Additional file 5: Table S2). The iron-containing heme group is required for normal tick physiology, including egg embryogenesis and reproduction. Since ticks do not have a heme biosynthetic pathway, they must obtain heme from host blood. However, as free heme is toxic to ticks, heme metabolism and management must be tightly regulated [56-57]. Heme-binding proteins, such as Vgs and CPs, take part in the removal and detoxification of free heme excess, as well as in lipid transport and storage [56-58]. Additionally, Vgs are precursors of vitellin, a protein that is essential for egg development and oviposition [59], which could explain the high abundance of vitellogenins in female saliva compared to male saliva, where these proteins represented only 0.45% of the total protein mass (Additional file 5: Table S2). Vgs were thought to be synthesised in the midgut and fat body only. However, Vg mRNA has been recently found in the salivary glands of Rhipicephalus bursa, O. moubata and O. erraticus, where it is up-regulated in response to blood feeding [24, 26, 31]. Moreover, the Vg protein has been abundantly detected in the saliva of A. americanum [41] and O. moubata [25], consistent with our current results. These findings indicate that heme-binding Vg-like proteins are indeed synthesised in the salivary glands and abundantly secreted into the host with tick saliva, where they likely play relevant functions in tick feeding; for instance, as anti-inflammatory agents, by reducing the concentration of free heme at the feeding lesion, which diminishes the heme potential to promote inflammation as well as the heme cytotoxicity [31, 41], or even as antioxidants and transporters of cholesterol, phospholipids and fatty acids [60]. The up-regulation of Vg-like proteins upon feeding, their abundance and functions in tick saliva, and the recognised high immunogenicity of tick Vgs [31], make this group of proteins interesting targets for tick vaccines.
Protease inhibitors
Most of the defensive responses that hosts deploy upon tick bites, including haemostasis, inflammation and immunity, are mediated by proteases, particularly serine and cysteine proteases [61-62]. Accordingly, tick saliva contains an abundance of protease inhibitors, mainly serine and cysteine protease inhibitors, which counteract host defences and facilitate blood ingestion [63-64].
Up to 26 protease inhibitors were detected and quantified by SWATH-MS in female and male saliva samples (Table 2; Additional file 5: Table S2). They represented 17.8% and 19.1% of the protein mass in female and male saliva, respectively (Fig. 5B, 5C), and were the second most abundant functional category in both sexes. Twenty-three of these protease inhibitors were serine protease inhibitors and the remaining three were cysteine protease inhibitors, or cystatins.
Among the serine protease inhibitors, we found Kunitz domain inhibitors, trypsin inhibitor-like cysteine rich domain (TIL) inhibitors and serpins. Kunitz type inhibitors are abundant in tick sialomes/saliva and include numerous anticoagulants that inhibit different proteases in the coagulation cascade, mainly thrombin and factor Xa [37, 40-41, 65]. TIL inhibitors form a family commonly found in blood-feeding insects and tick sialomes, which includes chymotrypsin, elastase and trypsin inhibitors. Some TIL inhibitors are known to act as anti-inflammatory agents and others behave as antimicrobial peptides [66-67]. Serpins are also abundantly detected in tick sialomes/saliva, where they play a role as immunomodulators, anti-inflammatory agents and inhibitors of platelet aggregation and blood coagulation [41, 63, 68]. Regarding cystatins, two types have been reported in ticks: Type 1 cystatins, which are intracellular and involved in the intracellular digestion of haemoglobin and developmental processes, and Type 2 cystatins, which are typically secreted to saliva, where they act as immunomodulators in the tick-host relationship [63].
Hemocytin was the most abundant protease inhibitor in female saliva as it represented 44.3% of the protein mass in this category (Additional file 5: Table S2). Hemocytin is a large modular glycoprotein, orthologue of the human von Willebrand factor, and contains several serine inhibitor TIL domains and several domains homologous to coagulation factor VIII. Hemocytin was first described in the silkworm Bombyx mori and functionally related to the insect immune response to invading pathogens through mechanisms such as haemolymph coagulation, haemocyte aggregation and nodule formation [69-70]. Hemocytin has been also detected in tick genomes [71], and although its functions in ticks remain to be elucidated, it could participate in the defence against pathogens ingested with blood.
On the other hand, a putative thyropin was the most abundantly detected protease inhibitor in male saliva (up to 20.5% of the protein mass in this category) and the second most abundant in female saliva (13.9%) (Additional file 5: Table S2). Thyropins are cysteine protease inhibitors that contain thyroglobulin type-1 (Thyr-1) domains and inhibit either cysteine or cation-dependent proteases [72]. Thyropins have been found in the sialomes and mialomes of several hard and soft ticks [37, 73-74]. Their function in ticks remains unknown, but it has been suggested they may act as immune modulators by regulating endosomal cathepsins, cysteine proteases involved in antigen processing by antigen presenting cells [73].
Proteases
Up to 24 proteases, including serine, cysteine, metallo- and aspartic proteases, were quantified by SWATH-MS in O. erraticus saliva samples, which represented 15% and 4% of the protein mass in female and male saliva, respectively (Table 2; Fig. 5). Serine proteases were the most abundant proteases in female saliva, as they accounted for 51.1% of the protein mass of this group, followed by cysteine (37%), metalloproteases (10%) and aspartic proteases (1.8%). In contrast, in male saliva, cysteine proteases were the most abundant ones (46.9%), followed by serine (40.1%) metalloproteases (12.9%) and aspartic proteases (0.2%) (Additional file 5: Table S2).
Serine proteases are commonly found in salivary glands and saliva from argasid and ixodid ticks [24, 26, 37, 41]. Salivary serine proteases of ticks are thought to regulate host defensive mechanisms at the tick bite site, such as blood clotting and fibrinolysis, matrix remodelling, inflammation and innate immunity, all of which may facilitate tick blood feeding [39, 73, 75].
Tick cysteine proteases, including L and B cathepsins and longipain, are mostly expressed in the midgut and involved in blood meal digestion, embryogenesis, and pathogen transmission [76-78]. However, the abundance of cysteine proteases found in O. erraticus saliva suggests that they might be also playing a role in tick feeding. Recently, a cathepsin L from Rhipicephalus microplus was shown to impair thrombin-induced fibrinogen clotting via a fibrinogenolytic activity, contributing to maintain blood fluidity of the ingested blood [62]. If the cysteine proteases found in O. erraticus saliva played a similar anti-clotting role, they might also contribute to maintaining blood fluidity, helping ticks to ingest the blood meal. Notably, the most abundant cysteine protease in O. erraticus saliva was a gamma-glutamyl hydrolase-like isoform X1 (XP_013781036). Gamma-glutamyl hydrolase (GGH) is a ubiquitously-expressed lysosomal enzyme that regulates intracellular folate metabolism for cell proliferation, DNA synthesis and repair [79]. GGH progressively removes gamma-glutamyl residues from poly-gamma-glutamyl forms of folic acid to yield folic acid and free glutamate. The abundance of this housekeeping enzyme in O. erraticus saliva suggests that it might play an additional extracellular function at the host-parasite interface, as has already been observed for other intracellular housekeeping enzymes, such as the salivary phospholipase A2 and an enolase of O. moubata, which act as anti-inflammatory and anti-haemostatic agents [80-82].
Tick metalloproteases are expressed in the midgut, ovary, salivary glands and saliva [37, 83-84]. Salivary metalloproteases are quite abundant and diverse, playing varied functions related to modulation of the host defensive responses that can facilitate blood feeding. For instance, salivary metalloproteases may contribute to formation of the feeding pool by degrading host extracellular matrix proteins at the tick bite site; metalloproteases also display anticlotting and anti-inflammatory activity by degrading fibrinogen, fibrin, and inflammatory mediators, and could even prevent host tissue repair because of their anti-angiogenic activity [40, 61, 74]. As a result of these activities, the immunoprotective potential of a Rhipicephalus microplus metalloprotease has been analysed and shown to confer 60% protection against tick infestation, highlighting this metalloprotease as a potential candidate for an anti-tick vaccine [83]. Metalloproteases were the most abundantly represented proteases in the O. erraticus sialotranscriptome [26] and in other tick sialomes [24, 37, 39-40, 74]. However, in the O. erraticus saliva proteome, metalloproteases are only the third most abundant protease family, suggesting some additional post-transcriptional regulation, with neprilysins being the most numerous and abundantly represented metalloprotease group (Additional file 5: Table S2).
Lipocalins
Lipocalins constitute a large and diverse family of secreted proteins that bind to and transport small hydrophobic molecules. Lipocalins are abundantly represented in tick sialomes, where they contribute to evading the haemostatic, inflammatory and innate immune responses of the host, mainly as scavengers of biogenic amines and eicosanoids [24, 26, 37, 41, 43, 74].
In this study, we found that lipocalins were the fourth most abundantly represented protein family in female saliva (8.3% of the protein mass) and the third in male saliva (18.3% of the protein mass) (Fig. 5). Up to 16 lipocalins were quantified by SWATH-MS in both sexes (Additional file 5: Table S2): six of them belonged to the biogenic amine (histamine and serotonin) binding clade of lipocalins [85] and another four belonged to the moubatin-like clade [86]. Among the remaining lipocalins, the one annotated as savicalin (ADI60053) was remarkable by its high abundance.
Amine-binding lipocalins were the most abundant in female saliva, as they represented up to 50.4% of the lipocalin mass versus the 8% represented by moubatin-like lipocalins (Additional file 5: Table S2). This strongly suggests that it is very important to prevent the inflammatory reaction induced by the release of histamine at the tick-feeding lesion for O. erraticus females be able to feed. On the contrary, moubatin-like lipocalins were the most abundant in male saliva (60.8% of lipocalin mass), far from the amine-binding lipocalins that accrued 18.0% of male lipocalin mass (Additional file 5: Table S2). The moubatin clade includes inhibitors of platelet and neutrophil aggregation, which act by scavenging thromboxane A2 (TXA2) and leukotriene B4 (LTB4), and inhibitors of complement activation, which sequester the C5 component [86-88]. The high abundance of moubatin-like lipocalins in male saliva suggest that blocking the host haemostasis and innate immunity would be more important for O. erraticus males be able to feed than preventing the histamine-mediated inflammatory reaction at the feeding lesion. Savicalin was also quite abundant in the O. erraticus saliva representing 32.9% and 11.5% of the female and male lipocalin mass, respectively. Savicalin was first described in haemocytes, midgut and ovaries of Ornithodoros kalaharensis [89] and more recently in the sialotranscriptome of O. erraticus [26]. Savicalin is up-regulated in the midgut and ovaries after feeding and in haemocytes after bacterial challenge, suggesting its involvement in tick development and anti-microbial defence [89]. The potential function of savicalin in tick salivary glands and saliva is currently unknown but it might be related to protection against pathogens acquired during feeding.
In contrast to our results, lipocalins were by far the most abundant proteins in the saliva proteome of O. moubata [25, 43] and Rhipicephalus microplus [90], but marginal in the proteome of salivary glands of Hyalomma dromedarii, where they only represented 0.8% of the secreted proteins in both sexes [40].
Immune related
Seven immune-related proteins were quantified by SWATH-MS, representing 6.7% and 4.2% of the protein mass in the female and male saliva, respectively (Table 2; Fig. 5). Among them, an immunoglobulin G binding protein A, a cysteine-rich venom protein, and two orthologues of savignygrin and ixodegrin were the most abundant (Additional file 5: Table S2).
Immunoglobulin binding proteins (IGBPs) are used by ticks to evade the host immune system and the damage caused by host antibodies that are ingested with blood. Intact host antibodies taken up into the gut by a feeding tick pass through into the haemolymph and can reach their antigen targets in internal organs. To prevent this immune mechanism, host immunoglobulins are bound, transported and finally excreted back to the tick-feeding site by IGBPs in haemolymph and salivary glands [91]. In ixodids, IGBPs are mainly expressed by males and secreted into the feeding site to help co-feeding females feed on blood and remove antibodies from the tick itself preventing antibody-mediated damage. Because of their function, IGBPs have been studied as potential vaccine candidates with partial success [92]. In argasids, IGBPs have been also detected in O. moubata male and female saliva [25], suggesting the conservation of these anti-defensive mechanisms among hard and soft ticks.
Cysteine-rich venom protein 1 belongs to the CAP superfamily (Cysteine-rich secretory proteins (CRISPs), Antigen 5 (Ag5), and Pathogenesis-related 1 (PR-1) proteins) of venomous proteins found in metazoans including ticks [93-94]. Snake venom CRISPs are the best-known CAPs and some of them have been functionally characterised; they inhibit a number of ion channels and the growth of new blood vessels, acting as antiangiogenic and vasodilator agents [95]. CRISPs in lamprey buccal gland secretions also act as inhibitors of ion channels and vasodilators, helping this hematophagous fish to feed [96]. This suggests an evolutionarily conserved function for CRISPs that would facilitate haematophagy in ticks.
Savignygrin and ixodegrins are disintegrin-like inhibitors of platelet aggregation discovered in O. kalaharensis [97] and Ixodes pacificus [98], respectively. Disintegrins are peptides that have a RGD or KGD domain, which can bind to integrins and impede interaction of integrins with their ligands blocking downstream events. In this way, the binding of disintegrins to integrin αIIbβ3 on activated platelets prevents fibrinogen-platelet interaction and inhibits platelet aggregation [99]. Platelet aggregation is the first step in the host haemostatic response and must be abrogated by ticks to feed; thus, anti-platelet aggregation agents, including disintegrins, are frequently found in tick sialomes [37]. Disintegrins can be interesting targets for tick vaccines and some studies have explored the value of the recently discovered O. moubata mougrin [100] and of an Ixodes ricinus ixodegrin [101] as antigen candidates for anti-tick and pathogen transmission-blocking vaccines.
Carbohydrate metabolism
Nineteen proteins involved in the metabolism of carbohydrates were quantified by SWATH-MS (Table 2; Additional file 5: Table S2). They comprised the most abundant functional category in male saliva, accounting for 33.0% of the protein mass in this sex, yet they only represented 6.3% of the female saliva protein mass (Fig. 5B, 5C). Most of them (n = 14) are enzymes of the glycolysis and tricarboxylic acid (TCA) cycle pathways. Cytoplasmic glycolysis transforms glucose to pyruvate, which is transported to the mitochondria, converted in acetyl-CoA and later metabolised in the TCA cycle yielding reducing equivalents that enter the oxidative phosphorylation chain to produce ATP [102].
These enzymes are considered housekeeping proteins, and the potential extracellular functions they might play in the tick saliva and the host-parasite interface remain unknown, except for some of them such as enolase. Salivary enolase of O. moubata can bind host plasminogen and stimulate its activation to plasmin at the feeding lesion, promoting fibrinolysis and contributing to the prevention of blood clot formation [82]. A similar function could be presumed for the salivary enolase of O. erraticus, although this needs to be experimentally demonstrated.
The reason why these glycolytic enzymes are much more abundant in male than in female saliva (Fig. 5) also remains unknown. It could be speculated that this difference would be a consequence of the additional functions that these enzymes might play in male saliva; these functions are currently unknown but perhaps be related to attraction, mating and spermatophore transfer [54, 103]. Interestingly, in another argasid species, O. moubata, the enzymes of the glycolysis pathway were also more abundantly expressed in male than female saliva [25, 43], suggesting that this could be a conserved tendency in the Ornithodoros genus or even in the Argasidae family.
Proteins differentially expressed between the sexes
Among the 224 common proteins quantified by SWATH-MS in female and male saliva, 97 were differentially expressed (p < 0.05) between the sexes (Table 3); thirty-seven proteins were over-expressed in females and 60 were overexpressed in males. The signal peak areas of the differentially expressed proteins in each of the samples analysed were shown using a heat map after z-score normalisation, using Euclidean distances. The heat map shows two main clusters comprising the F1–F3 samples and M1–M3 samples, which correspond to the saliva of females and males, respectively (Additional file 6: Figure S1).
The 97 differentially expressed proteins were classified into functional categories according to Kim et al. [41] and the average Log-fold change was calculated for each category and sex. Fig. 6 shows that 14 and 17 categories contained overexpressed proteins in females and males, respectively. Eleven of these categories contained overexpressed proteins in both sexes; three categories (heme/iron binding, signal transduction and glycine rich) contained proteins overexpressed only in females; and six categories (cytoskeletal, antimicrobial, nuclear regulation, and metabolism of amino acids, energy and carbohydrates) contained proteins overexpressed only in males. In females, the overexpressed categories showing the highest fold change were extracellular matrix and heme/iron binding, whereas the overexpressed categories showing the highest fold change in males were the metabolism of carbohydrates and lipocalins, in good agreement with the more abundant functional categories shown in Fig. 5.
Additional file 7: Figure S2 represents the top 10 proteins that are differentially (P < 0.05) overexpressed in the saliva of female or male ticks. In agreement with the above-reported results, the top 10 differentially overexpressed proteins in females were 3 heme/iron binding proteins (2 carrier proteins and 1 vitellogenin), 2 proteases (longipain and carboxipeptidase Q), 2 proteins of the extracellular matrix (mucin/peritrophin-like protein and laminin subunit beta-1), one apolipoprotein B-100 involved in lipid metabolism, one vitellogenin receptor, and one protein with unknown function. Regarding males, the top 10 differentially overexpressed proteins included 3 lipocalins (both of them moubatins), 2 glycolytic enzymes (pyruvate kinase and fructose-1,6-bisfosfato aldolase), one chymotrypsin inhibitor, one histone involved in nuclear regulation, a phosphatidylinositol transfer protein involved in lipid metabolism and two proteins of unknown function, one of them an acid tail salivary protein (Table 3).
Mucins are glycoproteins usually found in ixodid and argasid tick sialomes, but their function in ticks has not been studied [24-26, 37, 41, 74]. Mucins are secreted into tick saliva and may function in tick feeding by coating and protecting the chitinous tick mouthparts and by interacting with proteins of the host extracellular matrix [75]. Additionally, indirect evidence in humans suggest that mucins might participate in the antimicrobial defence, as human mucins were shown to encapsulate microbes [104].
Acid tail proteins, together with basic tail and tailless proteins, constitute a superfamily of tick-specific proteins abundantly found in the sialomes of both ixodid and argasid species [24-26, 37, 41, 74]. They are supposed to play important and specific roles at the tick-host feeding interface, but only a few members of this family have been functionally characterised as anti-coagulants [105-107] and specific complement inhibitors [108], while most of them have as yet unknown functions.
As a whole, these results of SWATH-MS show that at least 224 out of 387 of the salivary proteins identified in the current study are shared by both sexes, which significantly enhances the number of shared proteins identified by DDA LC-MS/MS (152 out of 387). However, these results show remarkable differences in the ratios of salivary proteins that males and females secrete in their saliva, which raises the question of the biological significance of these differences. It has been proposed that it could be related to the post-feeding blood processing or attraction and mating [25, 43, 54], but this question remains unsolved.