Molecular and biological characterization of the Amblyomma americanum organic anion transporter polypeptide

Ticks are among the most successful globally distributed disease vectors with an extensive veterinary and public health impact(Sonenshine, 1993). They transmit the highest number of pathogens of all known vector arthropods and are considered second only to mosquitoes in terms of the importance of the vectored human pathogens they carry(Sonenshine, 1993). In the livestock industry, annual monetary losses due to ticks and tick-borne diseases amount to millions of dollars(Jongejan and Uilenberg, 2004; Guerrero et al., 2006). The control of ticks and tick-borne diseases has primarily been achieved by acaricide application, which has led to the selection of resistant tick populations, and contamination of the environment and of animal products(de la Fuente and Kocan, 2006). To forestall the possibility of failed tick control programs, development of a new generation of tick control methods has been advocated(Willadsen, 1990). The limiting step to this has been the lack of effective targets that, if disrupted alone or in a cocktail with others, will impede tick feeding success and/or pathogen acquisition and transmission. Our laboratory is investigating the molecular basis of how ticks begin to feed as a means of discovering new targets for rational design of novel tick control methods. We recently identified a putative Amblyomma americanum tick organic anion transporter polypeptide among genes that are induced or differentially upregulated in ticks that have attained appetence and/or been exposed to feeding stimuli (Mulenga et al.,2007).

The organic anion transporting polypeptides [abbreviated as `Oatp' in non-human organisms and `OATP' in humans(Meier-Abt et al., 2005; Konig et al., 2006; Hagenbuch and Meir, 2004)] are gene products of the recently recognized solute carrier gene superfamily[non-human, Slco; human, SLCO(Meier-Abt et al., 2005)]. Oatp/OATP proteins have been identified in a broad range of organisms,including rodents, humans, chickens, zebrafish, frog and insects(Hagenbuch and Meier, 2004). They are multi-specific Na⁺-independent transmembrane transporters that are found in a wide spectrum of endogenous and xenobiotic amphipathic substrates, such as bile acids, bilirubin, thyroid hormones, prostanoids(prostaglandins and leukotrienes), neutral steroids (such as cardiac glycosides, ouabain and digoxin) and numerous other drugs(Meier-Abt et al., 2004; Konig et al., 2006; Niemi, 2007; Pizzagalli et al., 2002). In humans, one of the important physiological functions of the Oatp/OATP proteins is their role as components of the detoxifying mechanism of the body(Kasuhara and Sugiyama, 2005; Naud et al., 2007). Liver-,kidney- or the blood–brain barrier-expressed Oatp/OATP proteins mediate uptake of potentially toxic amphipathic substances from blood circulation into cells of these organs, where they are inactivated and then excreted as harmless by-products (Kasuhara and Sugiyma, 2005; Naud et al.,2007). Similarly, a recent study(Torrie et al., 2004) suggests that Oatp proteins may be part of the detoxification mechanism in insects. These authors recently explained the paradox of why ouabain, a cardiac glycoside and potent inhibitor of Na⁺, K⁺ and ATPase,fails to inhibit fluid secretion by Malpighian tubules in many insects, by linking its clearance from the basolateral folding of Drosophilarenal tubules to expression of Oatp58Db.

The aim of this study was to characterize the full sequence of AamOatp and validate its importance in tick feeding. We have provided evidence showing that AamOatp is constitutively and ubiquitously expressed, and that changing physiological needs during the tick feeding process influence its transcription patterns. Our data also show that RNAi mediated suppression of the AamOatp caused ticks to obtain smaller blood meals and consequently to lay fewer eggs.

Amblyomma americanum L. ticks used in this study were obtained from a colony that is maintained on cattle in our department. Dissections of partially fed ticks were routinely carried out as previously published(Mulenga et al., 2003). Ticks partially fed for 24, 48, 72, 96, 120 and 168 hours were detached and rinsed in 70% ethanol and then washed in sterile 0.1% diethyl pyrocarbonate (DEPC)treated water. Washed ticks were placed onto a glass slide with a pair of soft tissue forceps and their edges were trimmed off using a sharp and sterile razor blade. After trimming, the tick was transferred onto the concavity of a hanging drop slide filled with 200 μl of DEPC treated water. Under a dissection microscope, the dorsal cuticle flap was lifted and tick organs:salivary glands (SG), midgut (MG) and ovary (OV) were teased out using an 18-gauge needle and a soft tissue forceps. All dissected tissues, including the carcass (CA), which represents the tick remnant after removal of SG, MG and OV, were homogenized in Trizol (Invitrogen, Carlsbad, CA, USA), the total RNA extraction reagent, and then stored at –80°C until needed. Prior to RNA extraction, whole ticks were rinsed in 70% ethanol, air dried,pulverized in liquid nitrogen and immediately transferred to the Trizol reagent. Total RNA was isolated by isopropanol precipitation and reconstituted in RNAse free water. Subsequently, mRNA was isolated from purified total RNA of unfed ticks using the Poly (A) Purist kit (Ambion, Austin, TX, USA)according to instructions provided by the manufacturer.

To clone the full-length AamOatp cDNA, 2 μg of mRNA was used to synthesize the 3′ and 5′ adapter-linked cDNA templates according to instructions in the RLM rapid amplification of cDNA ends (RACE) protocol(Ambion). 3′ and 5′ RACE PCR primers that were synthesized based on a previously cloned AamOATP cDNA fragment(Mulenga et al., 2007). Subsequently, these primers were used to amplify and clone the full-length cDNA using high-fidelity Taq polymerase (Applied Biosystems, Foster City, CA,USA) for long-distance PCR. PCR amplicons were routinely cloned in the TOPO PCR4 plasmid (Invitrogen) and sequencing was accomplished using the Bigdye terminator (Applied Biosystems), M13 forward and reverse primers, and internal primers were necessary. DNA sequence analysis was routinely carried out using the VectorNTI software (Invitrogen academic license).

To identify homologs in the Ixodes scapularis Say genome, the FASTA3 software version 3.4 (Pearson and Lipman, 1988) was used to perform a local BLASTX scan of supercontig downloads from vector base(Lawson et al., 2007)(http://iscapularis.vectorbase.org/GetData/Downloads) using the deduced AamOatp as query. To identify coding exons, the retrieved I. scapularis supercontigs were scanned against Drosophila sequences using BLASTX analysis. Subsequently,best matches from the BLASTX scan and the AamOatp deduced protein were aligned with I. scapularis supercontigs using GenomeScan(Burge and Karlin, 1998) or the NetGene2 (Hebsgaard et al.,1996) servers to delineate introns and exons regions boundaries. The predicted exon regions marked by the `AT' acceptor and `GT' donor splice sites were manually assembled and translated using VectorNTI (Invitrogen).

Identity and similarity comparison with other sequences was accomplished using the BLASTX analysis at the NCBI. the levels of identical amino acids between AamOatp and other Oatp/OATP sequences were determined by pairwise alignment analyses using the VectorNTI software. The guide phylogeny tree was constructed from the dataset of Oatp/OATP protein sequence download from GenBank and four tick Oatp sequences (one from this study and three from the I. scapularis genome project) using the `neighbor joining'method. Specifications were set for bootstrap values at 1000 replications with gaps proportionately distributed and correction for distance set to a `Poisson distribution'.

To determine spatial and temporal expression profiles, DNAse-treated total RNA from SG, MG, OV and CA dissected from partially fed ticks (24, 48, 96, 120 and 168 h) was subjected to two-step semi-quantitative RT-PCR. Approximately 5μg of total RNA was used to synthesize the first-strand cDNA template using the Verso cDNA kit (Thermo Fisher Scientific, Pittsburgh, PA, USA). 1 μl of the first strand cDNA was used as template in a PCR reaction with AamOatp specific primers. A 15 μl aliquot of the PCR product was electrophoresed on a 2% agarose gel containing 1 μg ethidium bromide.

RNAi gene silencing was carried out according to previously published methods (Nijhof et al., 2007; de la Fuente et al., 2006; Decrem et al., 2008; Hatta et al., 2007). Using the cloned AamOatp plasmid DNA as template and PCR primers (forward,5′TAA TAC GAC TCA CTA TAG GGG GAC CCG GCA AGC ACT TGC TAG TT3′; reverse, 5′TAA TAC GAC TCA CTA TAG GGT GTT CTT GGA GAC CGC GTC CTC C3′) with added T7 promoter sequence (in bold),were used to amplify the template for AamOatp double-stranded RNA(dsRNA). For negative control dsRNA, the green fluorescent protein (GFP) PCR primers with added T7 promoter sequence (forward, 5′TAA TAC GAC TCA CTA TAG GGTGGCCGCTTTACTTGTACAGC3′; reverse 5′TAA TAC GAC TCA CTA TAG GG ACCGGTCGCCACCATGGT3′) were used to amplify the amplicon. Subsequently, dsRNA was synthesized in vitro using the MegaScript RNAi kit according to instructions provided by the manufacturer(Ambion). After purification of dsRNA, 20 unfed female ticks were microinjected with ∼1 μl (∼3 μg μl^–1 in TE buffer, pH 8.0) of AamOatp or GFP dsRNA via the groove between the basis capituli and the scutum using half-inch 33-gauge needles attached to a 10μl gastight syringe (Hamilton, Reno, NV, USA). Injected ticks were kept overnight at 22°C to observe any mortality resulting from injection injury. Subsequently AamOatp- and GFP-dsRNA-injected and non-injected control ticks (N=20) were put in cells that were secured on the back of the calf using livestock identification cement (Nasco, Fort Atkinson, WI). Ticks were allowed to feed until detachment.

To validate the specificity of RNAi-mediated silencing of AamOatp,three ticks per treatment (AamOatp-dsRNA and GFP-dsRNA-injected) were detached 48 h post-attachment. These ticks were individually processed for total RNA extraction. Total RNA was treated with RQ1 DNAse (Promega, Madison,WI, USA) to remove genomic DNA (gDNA) contamination and then subjected to two-step semi-quantitative RT-PCR. To forestall the possibility of re-amplifying the injected AamOatp-dsRNA during validation of AamOatp silencing, PCR primers (forward,5′GACGAATTCGTGCAGCGACGGCTGCGACTG3′; reverse,5′CTGAAGCTTTCACTTGGTCACGTTGGCGAT3′) were targeted to cDNA regions flanking the domain that was used for dsRNA synthesis. Lack of amplification of the AamOatp transcript confirmed the silencing of AamOatp.

To assess the effect of silencing AamOatp on tick-feeding success,tick-feeding parameters, including numbers of attached ticks 24 h post-attachment, mortality, time of feeding to repletion, survival and engorgement masses (EM, measure of amount of blood imbibed) were evaluated. To assess the effects of AamOatp silencing on fecundity, the egg mass conversion ratios (EMCR) were compared between treatment and control ticks. This was accomplished by incubating engorged ticks at 25°C for 20 days to allow them to lay eggs and then the egg masses were weighed. The EMCR was calculated by mass of the egg mass as a fraction of the EM. A web-based unpaired Student's t-test (GraphPad t-test calculator, www.graphpad.com/quickcalcs/ttest1.cfm)was used to test the statistical significance of the differences that were observed between AamOatp-dsRNA-injected ticks and control groups.

Rapid amplification of cDNA ends was successfully used to clone a 2860 bp full-length cDNA (GenBank accession number FJ206228) that encode a 724 amino acid protein (Fig. 1). Provisional identification of AamOatp as a member of the Oatp/OATP protein family (Hagenbuch and Meier,2004) was based on its similarity to previously characterized proteins (data not shown). When scanned against known protein entries in GenBank, the top 20 best matches to AamOatp were arthropod, including fruit flies (Drosophila spp), the jewel wasp (Nansonia vetripennis), the red flower beetle (Tribolium castaneum),mosquitoes (Aedes and Anopheles spp) and honeybee (Apis melifera) (data not shown). Hydropathy and bioinformatics analyses,revealed that AamOatp contains the Kazal type serine proteinase inhibitor domain in addition to all typical features of Oatp/OATP proteins: 12 transmembrane (TM) domains, a 13 amino acid signature motif,D-X-RW-(I,V)-GAWW-X-G-(F,L replaced by Y)-L (replaced by I) and a large extracellular region (ER) between the ninth and tenth TM domains(Fig. 1). Consistent with other sequences, 11 consensus cysteine residues that are present in the large ER of all Oatp/OATP proteins (Meier-Abt et al.,2005; Hagenbuch and Meier,2003; Hagenbuch and Meier,2004) are 100% conserved in AamOatp(Fig. 1). Visual analysis of the AamOatp deduced amino acid sequence revealed that, unlike most known vertebrate Oatp/OATP protein sequences, which posses two putative N-glycosylation sites, in the second and fifth ER(Meier-Abt et al., 2005; Hagenbuch and Meier, 2004) the AamOatp protein posses one putative N-glycosylation site in its fifth ER (Fig. 1). It is interesting that the AamOatp protein also contains two additional putative N-glycosylation sites. However, these may not be functional as they are located within the second and fifth intracellular domains(Fig. 1).

When scanned against supercontigs of the I. scapularis genome, AamOatp produced hits to eight supercontigs (not shown), which were determined to contain 11 different I. scapularis (Isc) Oatp sequences. Because of gaps in supercontig sequences, eight of the 11 IscOatp sequences are partial and the rest could be assembled into complete open reading frames. Among the three complete sequences, AamOatp show 50/67% identity/similarity to IscOatp-1(Fig. 1), 26/41% to IscOatp-2 and 21/42% to IscOatp-3 (not shown). Consistent with our multiple sequence alignment data in Fig. 1, phylogenetic analysis revealed that AamOatp sequence was closely related to IscOatp-1 (Fig. 2). From the phylogeny tree in Fig. 2, it is interesting to note that IscOatp-2 is distantly related to other arthropod sequences, but it is closely related to human OATP4A1. At the time of preparing this manuscript, the annotation of the I. scapularis genome had not been completed. Thus accession numbers of the primary sequences that were used to assemble ORFs, ABJB010623253.1,ABJB010665419.1 and ABJB010386346.1 for IscOatp-1, ABJB011035923.1 for IscOatp-2 and ABJB010326688.1 for IscOatp-3 are indicated.

To assess temporal and spatial expression profiles of AamOatp mRNA in tick organs (SG, MG, OV and CA) during tick feeding, total RNA of organs from ticks that fed from 1–7 days, was subjected to semi-quantitative RT-PCR analysis (Fig. 3). Results summarized in Fig. 3Ashow that AamOatp mRNA is ubiquitously expressed, as revealed by its being amplified from cDNA of all tick organs. Analysis of the normalized transcript abundance shown in Fig. 3B revealed that from day 1 to 5 of feeding, AamOatp mRNA in the MG was ∼60–80-fold higher than levels found in SG, OV and CA. By contrast, by day 7 of feeding, the AamOatp mRNA was 60–80-fold more strongly expressed in the OV than the SG, MG and CA(Fig. 3B).

Qualitative two-step RT-PCR expression analysis summarized in Fig. 4 was used to validate the RNAi-mediated silencing of AamOatp mRNA. On the basis of detectable PCR bands amplified by 40 PCR cycles, complete silencing of AamOatp was confirmed in two out of the three ticks tested(Fig. 4). In order to assess the effect of silencing on tick-feeding efficiency, feeding parameters were compared between ticks that were injected with AamOatp-dsRNA and those injected with GFP-dsRNA. Silencing of AamOatp did not affect the ability of ticks to attach onto host skin, initiate and continue to feed(results not shown). We also found that silencing of AamOatp did not cause any mortality (results not shown).

To quantify the amount of blood taken in by a tick, we determined engorgement masses (EM, mass of ticks upon naturally detaching from the host,representatives are shown in Fig. 5A). The assumption was that the EMs represented the amount of blood taken in by a tick. EMs summarized in Fig. 5B were 532–940 mg(N=15, mean ± s.e.m.=730±37.58 mg) for non-injected controls, 437–774 mg (N=10, mean ±s.e.m.=620±31.95 mg) for GFP-dsRNA-injected controls, 283–647 mg(N=10, mean ± s.e.m.=482±31.21) for AamOatp-dsRNA-injected ticks. The unpaired Student's t-test comparison of mean EMs showed that blood meals obtained by AamOatp-dsRNA-injected ticks were significantly smaller than those obtained by both non-injected (P<0.0001, two-tailed test) and GFP-dsRNA-injected (P=0.0073, two-tailed test) controls ticks. Importantly, blood meals obtained by GFP-dsRNA-injected control ticks were also statistically significantly smaller than those obtained in the non-injected controls (P=0.0479, two-tailed test).

In order to assess the effect of silencing AamOatp on fecundity,we compared the ability of ticks to convert their blood meal into the egg mass. As was to be expected, AamOatp-dsRNA-injected ticks, which had obtained smaller blood meals (Fig. 5B), laid apparently fewer eggs (mean ±s.e.m.=205±14.95) than both the non-injected (mean ±s.e.m.=400±21.21) and the GFP-dsRNA-injected (mean ±s.e.m.=318±18.43) control groups. Comparison of EMCRs summarized in Fig. 6 show that silencing of AamOatp caused a significant reduction in the efficiency of the tick to convert blood meal nutrients into eggs (P<0.0001 versus non-injected controls) (P=0.0008, versus GFP-dsRNA-injected controls). By contrast, EMCR of GSP-dsRNA-injected control ticks did not statistically differ from the non-injected control (P=0.2199),although they laid significantly fewer eggs.

Several Oatp/OATP proteins have been identified in a broad range of organisms, including rodents, humans, chickens, zebrafish, frog and insects(Hagenbuch and Meier, 2004). This study describes the first report on biological and bioinformatics characterization of an Oatp protein from a blood-feeding organism. Based on data in this study, we are unable to determine conclusively whether or not the AamOatp protein is functional. However, based on our bioinformatics analyses, we believe that the AamOatp protein is likely to be a functional transporter because of the high conservation of sequence structures that are thought to be important for the transport function of Oatp/OATP proteins (Meier-Abt et al.,2005; Hänggi et al.,2006). In a recent study, Hanggi et al.(Hanggi et al., 2006) showed a single mutation among the 11 consensus cysteine residues in the large ER between TM9 and TM10 was enough to destroy the transport function of the OATP2B1 protein. Given that these cysteine residues are 100% conserved in the AamOatp, it is a good indicator that this protein is likely to be functional.

In humans, mice and rats, at least 39 Oatp/OATP proteins were described by 2005 (Meier-Abt et al., 2005). Given that the A. americanum genome has not been sequenced, the full repertoire of Oatp proteins in this tick will remain unknown. However, based on our analysis of supercontigs of the I. scapularis genome, ticks may encode at least 11 different Oatp proteins. In a recent study, Hagenbuch and Meir (Hagenbuch and Meir,2004) suggested a new classification system of Oatp/OATP proteins,where sequences that show ≥40% identity belonged into the same family and those showing ≥60% identity being further classified into a subfamily. On the basis of this classification system, the four tick Oatp proteins in this study (three IscOatp and the AamOatp proteins) may be classified into at least three families: AamOatp and IscOatp-1, which are 56% identical belonging to the same family; with IscOatp-2 and -3, which showed <40% identity to any of the sequences, belonging to distinct families. From our phylogenetic analysis data, it was interesting that IscOatp-3 was more closely related to the human OATP4A1 protein than to other arthropod sequences. OATP4A1 is a known transporter of prostaglandin, among other substrates(Gao et al., 2005). Ticks do synthesize prostaglandins (PGs), which are thought to play a role in facilitation of tick–host interactions(Bowman et al., 1996). The in vivo transport of tick PGs has not been studied. It will be interesting to investigate whether IscOatp-3, similar to OATP4A1, is also involved in transport of PGs.

An important component of this study was to gain insight on the potential of AamOatp as a druggable or an anti-tick vaccine target. Our RNAi data suggest that AamOatp represents an important anti-tick target,as its silencing caused ticks to obtain significantly smaller blood meals. Although, silencing of AamOatp did not prevent ticks from attaching and starting to feed, these findings suggest that AamOatp may play a role in the processing of host blood meal. It is noteworthy that, although silencing of AamOatp did not prevent egg laying, it reduced the ability of these ticks to convert their blood meals into eggs. A potential limitation to our approach is that we could not account for the actions of the pre-existing AamOatp protein prior to silencing. Our speculation is that, after silencing of the AamOatp mRNA, the pre-existing AamOatp protein will still be viable and thus the effects of silencing would not be immediate. This might explain the seemingly lack of impact of AamOatp silencing on early tick-feeding parameters,attachment onto host skin and beginning to feed.

Consistent with vertebrate Oatp/OATP genes that show multiple tissue expression patterns (St-Pierre et al.,2002; Gao et al.,2005; Choudhuri et al.,2003), our RT-PCR expression analyses data revealed that AamOatp mRNA is both constitutively and ubiquitously expressed, as it was it is expressed in all tested organs of ticks that fed for 1–7 days. Transcriptional regulation of tick genes has been shown in response to blood meal feeding, being either shut down or induced, up- or down regulated(Leboule et al., 2002a; Leboule et al., 2002b; Mulenga et al., 2001; Rudenko et al., 2005). In the case of AamOatp, the switch in expression pattern, from being predominant in the midgut during the first 5 days of feeding and the ovary, by day 7 of feeding suggest that AamOatp is transcriptionally regulated in response to changing physiological needs as tick feeding progresses. The first 5 days of feeding, when AamOatp is strongly expressed in the MG, correspond to the slow feeding phase (SFP) of the tick feeding cycle. This phase prepares the tick for the rapid feeding phase (RFP), during which it feeds to repletion (Sonenshine,1993; Bowman et al.,1996; Reuben,2007). Day 7 of tick feeding corresponds to the beginning of the RFP, and is when predominant expression of AamOatp switches to the ovary. Assuming that the AamOatp protein expression profiles are correlated with transcript production, we speculate that during the first 5 days of feeding, AamOatp is involved in transport of key nutrients(s) from host blood into tick hemolymph. Subsequently, as the tick feeds to repletion, AamOatp could be involved in transporting nutrients from hemolymph into the ovary for egg development. It is interesting that the thyroid hormone derivatives thyroxine and triidothyronine, which are transported across cell membranes by Oatp/OATP proteins(Hagenbuch and Meir, 2004; Pizzagalli et al., 2002), have been linked to embryo development in birds (Willson and McNabb, 1997). Whether or not this is the case in ticks is not known. However, if it were the case,it will be interesting to investigate whether the biological functions of the AamOatp protein are related to egg development. Whether or not the silencing of AamOatp reduces the efficiency of the tick to convert its blood meal to eggs is an indication of this possibility is not currently known.

Because of their ubiquitous distribution and their wide activity range as transporters of several drugs, the role of Oatp/OATP proteins in drug disposition, bio-distribution and absorption is emerging(van Montfoort et al., 2003; Faber et al., 2003; Mizuno et al., 2003; Gerloff, 2004; Niemi, 2007; Bachmakov et al., 2008). This study presents a framework for similar studies to be carried out in ticks. Although only a single study has linked the Oatp transmembrane transport function to the detoxification mechanisms in arthropods(Torrie et al., 2004), these proteins represent significant druggable or anti-tick vaccine targets as their blockade could prevent ticks from excreting important chemical compounds. As anti-tick vaccine targets, another attractive feature of these proteins is that they are mostly localized on cell surfaces(St-Pierre et al., 2002),which will make them accessible to host antibodies. The next phase of this research will be to characterize the transport functions of AamOatp and whether or they could be targeted for tick control.

Funding for this project was provided by start-up funds to A.M. from the Texas A&M AgriLife Research (formerly, College of Agriculture and Life Sciences), Department of Entomology and the Texas Agriculture Experiment Station.