Orchestration of salivary secretion mediated by two different dopamine receptors in the blacklegged tick Ixodes scapularis

Ticks are obligatory ectoparasites of many vertebrates and transmit pathogens that cause numerous diseases, including Lyme disease, Rocky Mountain spotted fever (RMSF), and southern tick-associated rash illness (STARI). Lyme disease transmitted by Ixodes scapularis Say 1821 is the most commonly reported vector-borne disease in the United States. Ticks generally remain attached to the host for up to ~14 days depending on species for blood meals, unlike hematophagous insects, such as mosquitoes, which feed for ~3 min. Female ticks can normally increase their body weight up to 100 times compared with unfed ticks (Kaufman, 2007). A successful long tick feeding duration, which may be the major reason that they are such efficient disease vectors, requires that the tick be capable of overcoming host defensive responses, mainly through salivation.

Tick salivary secretion contains various bioactive components, such as anti-haemostatic molecules, anti-inflammatory compounds and immunomodulators that compromise the host immune responses (Ribeiro, 1987; Ribeiro, 1989; Nuttall et al., 2006; Francischetti et al., 2009). Salivation also functions to maintain homeostasis by removing excessive water and ions that are processed from large blood meals (Tatchell, 1967; Kaufman and Phillips, 1973a). A pair of salivary glands is located at the anterolateral regions of the tick, and they resemble a cluster of grapes, which consist of three types of acini, types I, II and III. Based on the cellular structure, each type is thought to have different functions: type I for secretion of hyperosmotic fluid for uptake of atmospheric water vapour in free-living unfed ticks; type II for secretion of bioactive proteins on the host; and type III for excretion of hyposmotic saliva on the host (Rudolph and Knülle, 1974; Gaede and Knülle, 1997).

In on-host osmoregulation that removes excess water and ions during blood feeding, the majority of water and sodium ions were secreted back into the host via salivary secretion (Tatchell, 1967; Kaufman and Phillips, 1973a). Homogenates of synganglia (tick brain) and dopamine induced fluid secretion in isolated salivary glands (Kaufman, 1976; Kaufman, 1977). Octopamine or cyclic AMP (cAMP) demonstrated low activities on salivary secretion (Kaufman, 1976; Needham and Pannabecker, 1983; Pannabecker and Needham, 1985), whereas acetylcholine, serotonin and glutamate did not show activity from isolated salivary glands (Kaufman and Phillips, 1973b; Kaufman, 1976).

Earlier studies found that the activity of dopamine on salivary secretion is dependent on extracellular Ca²⁺ (Kaufman, 1976; Needham and Sauer, 1979). In addition, cAMP was also involved in dopamine-mediated salivary secretion (Schmidt et al., 1981; Krolak et al., 1983; Hume et al., 1984). Recently, we characterised two dopamine receptors: dopamine receptor (D1) and invertebrate specific D1-like dopamine receptor (InvD1L) in both acini types II and III (Šimo et al., 2011; Šimo et al., 2014). Receptor localisation and downstream coupling in the heterologous expression system suggested that the D1 receptor is probably coupled to cAMP elevation in epithelial cells, and the InvD1L receptor is coupled to Ca²⁺ mobilisation in myoepithelial cells (Šimo et al., 2011; Šimo et al., 2014). Therefore, two different dopamine actions mediated through different acini cell types were proposed, although experimental verifications of the suggested downstream cellular actions and physiological roles of each receptor were not possible because RNA interference targeting those receptors were not practicable in those studies.

In this study, we identified pharmacological agents that distinguish D1 and InvD1L receptors in heterologous receptor expression. The agents discriminating the two dopamine receptors successfully revealed different physiological actions in salivary gland acini in vitro.

To identify receptor-specific agonists and antagonists of two dopamine receptors, we expressed the receptors in Chinese hamster ovary cell (CHO) for receptor-mediated calcium mobilisation reporter assays as described previously (Šimo et al., 2011; Šimo et al., 2014). Among candidate chemicals that were well known to be active on mammalian D1 receptors, we determined that SKF82958 and apomorphine specifically activated the D1 receptor but not the InvD1L receptor at 10 μmol l⁻¹ (Fig. 1A). SKF82958 and apomorphine activated only the D1 receptor by 80 and 60% of dopamine (10 μmol l⁻¹) activity, respectively, while they had no effect on the InvD1L receptor, showing 0.02 and 3% of dopamine (10 μmol l⁻¹) activity, respectively (Fig. 1A). Another agonist, (±)-2-amino-6,7-dihydroxy-1,2,3,4-tetrahydronaphthalene hydrobromide (6,7-ADTN, 10 μmol l⁻¹), activated both D1 and InvD1L receptors by 93 and 89% of dopamine activity. We also tested sulfate-conjugated dopamine, dopamine-3-O-sulfate (DA-3S) and dopamine-4-O-sulfate (DA-4S), which were highly abundant in the tick salivary gland in our previous study (Koči et al., 2014). However, both DA-3S and DA-4S did not show any activity on either the D1 or the InvD1L receptor in our assay system.

In search of receptor-specific antagonists, which measured the reduction in dopamine activity on the receptor after 15 min pre-incubation with the test compound (10 μmol l⁻¹), we identified four antagonists that were specific only to the InvD1L receptor: fluphenazine, acepromazine, clozapine and (+)-butaclamol, while (−)-butaclamol, sulpiride and haloperidol had no or low antagonistic activities on either receptor. We also found a common antagonist (SCH23390) that showed 77 and 69% antagonism on D1 and InvD1L, respectively (Fig. 1B), which was also previously described for dopamine receptors of mosquito and tick (Meyer et al., 2011; Meyer et al., 2012).

Taken together, we identified three groups of pharmacological discriminators: D1-specific agonists, InvD1L-specific antagonists, and a common antagonist. However, there was no InvD1L-specific agonist and D1-specific antagonist among the ligands assessed in this study. In subsequent in vitro experiments on isolated salivary glands, we utilised only SKF82958 as D1-specific agonist, fluphenazine as InvD1L-specific antagonist, and SCH23390 as a common antagonist (Fig. 1C).

To investigate the responses of salivary glands to pharmacological discriminators, we quantified the volumes of secreted saliva from isolated salivary glands by developing the assay method modified from Ramsay's original assay (Fig. 2) (Ramsay, 1954). The salivary glands that were pre-incubated with buffer only did not show any secretory activity. Dopamine (1, 10 and 100 μmol l⁻¹) induced significant salivary secretion in a dose-dependent manner (Fig. 3A,E). The rate of salivary secretion was increased immediately after agonist treatment, peaked at 10 or 15 min, and then gradually decreased until 30 min, which was the end of our observation (Fig. 3A–D). The maximal total salivary secretion volume for 30 min was ~1.2 μl at 100 μmol l⁻¹ dopamine (Fig. 3E).

SKF82958, a D1-specific agonist activating only the D1 but not the InvD1L receptor, also showed significant, but slightly lower activities than dopamine. At SKF82958 doses of 10 and 100 μmol l⁻¹, the secretory activity was 44 and 76% of dopamine at the same respective concentrations (Fig. 3E,F). The general patterns of secretory activities over time in the post-treatments were similar to the patterns observed in the dopamine treatments.

Pre-incubation of the isolated salivary gland with the InvD1L-specific antagonist fluphenazine (10 μmol l⁻¹) followed by co-incubation of fluphenazine with dopamine or SKF82958 significantly suppressed secretory activity, while the salivary secretion pattern peak at ~10 min was similar to agonist treatment alone (Fig. 3A–D). We observed ~30% reductions at both 1 and 10 μmol l⁻¹ dopamine treatment (Fig. 3E). The reduction in SKF82958 treatment was also significant at 10 μmol l⁻¹ but not at 100 μmol l⁻¹ (Fig. 3F).

Co-incubation of the isolated salivary gland with SCH23390 (10 μmol l⁻¹), a common antagonist for both D1 and InvD1L, and dopamine (10 μmol l⁻¹) almost completely abolished secretory activity (Fig. 4). The antagonistic activity of SCH23390 was reverted after replacing it with dopamine only (100 μmol l⁻¹) without the antagonist (Fig. 4).

Microscopic observation of each acinus after pharmacological agent treatment revealed dynamic changes occurring in each acinus. In the initial observations, we found that the acini sizes dynamically changed, especially when they were treated with dopamine. The changes in acini sizes were driven by the changes in luminal acini volumes but not by the changes in sizes of the cells composing the acini (Fig. 5A). Therefore changes in acini sizes were measured and calculated for volume and percentage volume change. Based on the volume calculated, each acinus luminal space increased up to ~4 nl by dopamine treatment.

Patterns of the changes over time were used to infer physiological changes of the acini: increased sizes were interpreted as water-solute influx into the acinar lumen and decreased sizes as pumping or gating for saliva discharge through the acinar duct, respectively (Fig. 5). Following influx of water-solute, rapid or slow reductions in acini sizes occurred and were referred to as pumping or gating, respectively, and together as pumping/gating. We set the threshold for counting the pumping/gating category based on the statistics of samples with mock treatments and a rate of reduction higher than 3.5% per minute (see further details in Materials and Methods). However, based on our close observations, this method was unavoidable from the false-negative cases, i.e. the cases where true pumping/gating were not counted because the fluid expulsion rate was occasionally masked by equal or higher fluid influx rates, although the positive pumping/gating counts were accurate.

As an example of dynamic changes in acini, dopamine triggered changes in acini sizes (Fig. 5A; supplementary material Movie 1) and volume (Fig. 5B). Typical pumping is shown by rapid reduction of the acini size that appeared, such as a squeezing muscular contraction of the acini (supplementary material Movie 2) (Coons et al., 1994). In some cases, slow reduction in size occurred over a relatively long duration. This case was considered to be gating because subtle movements of the acinar duct valve in concurrence with slow volume reduction were observed in some cases (supplementary material Movie 3).

The maximum sizes of the acini that were activated by dopamine at the four different doses (100 nmol l⁻¹ to 100 μmol l⁻¹) demonstrated a typical dose–response of size increase up to 1 μmol l⁻¹ but slight reductions at the higher concentrations of 10 and 100 μmol l⁻¹ (Fig. 6A). Similarly, the D1-specific agonist SKF82958 also showed significant activities for the increased acini size at all doses but with slightly lower activities than dopamine. Similar to the dopamine treatments, the higher doses induced stronger responses up to 10 μmol l⁻¹, but the highest dose (100 μmol l⁻¹) showed a slightly reduced degree of response. The pumping/gating frequency was highest in the 1 μmol l⁻¹ dopamine treatment (42%) but was progressively reduced in the higher dopamine doses (Fig. 6C). In the SKF82958 treatments, pumping/gating was much less frequent than that of dopamine with a maximum count of 18% of acini at the 10 μmol l⁻¹ dose (Fig. 6D).

The activities of dopamine or SKF82958 for increased acini volume were completely abolished by SCH23390, the common antagonist of both D1 and InvD1L (Fig. 7A). Consequently, the acini that were pre-treated with SCH23390 did not show any pumping/gating activity (Fig. 7B).

Pre-treatments of acini with the InvD1L-specific antagonist fluphenazine followed by dopamine or SKF82958 treatments did not significantly reduce the maximum acini sizes (Fig. 8A). However, reduced frequencies of pumping/gating by the pre-treatments were apparent in both dopamine and SKF82958 treatments (Fig. 8B). The frequency of dopamine-induced pumping/gating was reduced from 28 to 17% by fluphenazine pre-treatment. Pre-treatment eliminated SKF82958 treatment-mediated pumping/gating (from 18% to 0%) (Fig. 8B).

In dopamine-mediated osmoregulatory salivary secretion by the tick, we determined that two combinatory physiological actions occur in the type III salivary gland acini: inward fluid transport and release of the luminal content. Dopamine treatment immediately activated fluid influx as indicated by increased luminal volume (Fig. 5). In contrast to the description made in Ixodes ricinus, where increased luminal size accommodated the fluid influx without increased acini size (Bowman and Sauer, 2004), we observed rapid acini expansion up to approximately five times because of increased luminal volume (Fig. 5, supplementary material Movie 1). Visually detectable levels of fluid efflux through the acinar duct in this study found mainly two main categories of efflux patterns in different acini: acini with slowly reduced sizes over time (gating) and acini with rapid collapse for expelling fluid (pumping). On rare occasions, we observed subtle movements of acinar valve flaps in the gating acini (supplementary material Movie 3). In contrast, the pumping acini, when it occurred in the intermediate size before the acini was fully expanded, appeared to be driven by a squeezing contraction of the acini (supplementary material Movie 2). These were difficult to distinguish when the acini were fully expanded because rapid size reduction could be partly caused by release of high internal pressure in the absence of the contraction. Summing up the observations, two components of salivary secretion are epithelial cell fluid transport for water-solute influx and myoepithelial cell contraction (Coons et al., 1994) for generating internal pressure with valve opening allowing saliva efflux through the acinar duct. The full salivary secretion rate required both processes based on the experiment using pharmacological tools (Fig. 3 and see below). Observation of the salivary glands in real time provided physiological events that were involved in salivary secretion, although we do not know whether the observed levels of acini expansion occur in an in vivo system that is enclosed in the rigid body wall.

Use of receptor-specific pharmacological tools in this study suggests that the downstream physiological events are mediated through different dopamine receptors D1 and InvD1L, which were previously characterised in tick salivary glands (Šimo et al., 2011; Šimo et al., 2014). The results of this study support the function of D1 for water-solute influx and InvD1L for pumping/gating in type III acini. The most pronounced result was demonstrated in the acini that were treated with the combination of D1-specific agonist SKF82958 and InvD1L-specific antagonist fluphenazine, abolishing the pumping/gating with normal acini size increase (Fig. 8) and significantly diminishing fluid secretion by whole salivary glands (Fig. 3F). Treatment with dopamine and fluphenazine in combination and SKF82958 alone, however, reduced but did not completely abolish the pumping/gating frequencies and fluid secretion, although fluphenazine and SKF82958 clearly distinguished recombinant receptors in the heterologous system at the 10 μmol l⁻¹ dose. This leaky activity in suppression of pumping/gating suggests the presence of other signalling pathways that interact with dopamine signalling for pumping/gating. Such candidates could be the neuropeptides myoinhibitory peptide (MIP) and SIFamide, which were found in the neural projections reaching the basal region of acini (Šimo et al., 2009; Šimo et al., 2013). Alternatively, an unidentified signalling system acting through the octopamine receptor, which is probably cross-activated by dopamine, could have also been involved in this process.

Previous studies in dopamine signalling pathways (Schmidt et al., 1981; Hume et al., 1984; Sauer et al., 2000) and in characterising the D1 and InvD1L receptors in tick salivary glands (Šimo et al., 2011; Šimo et al., 2014) and the present study are all congruent and provide further insight into the mechanisms of tick salivary gland control. Dopamine activates crucial Ca²⁺ (Kaufman, 1976; Needham and Sauer, 1979) and cAMP (Hume et al., 1984) downstream signalling pathways in tick salivary secretion. Heterologous receptor expression determined that D1 has higher dopamine sensitivity and was mainly coupled to cAMP elevation, while InvD1L was coupled to Ca²⁺ elevation with lower dopamine sensitivity (Šimo et al., 2011; Šimo et al., 2014). Immunohistochemistry revealed that D1 is located in the acini epithelial cells and InvD1L is located in axonal projections reaching to the basal region of acini and in another projection further extending to the apical region of acini (Šimo et al., 2014; Šimo et al., 2011). These characteristics of D1 and InvD1L suggest that an initial low dose of dopamine activates D1 in epithelial cells leading the cAMP-mediated fluid transport. High dopamine doses acting on InvD1L trigger (or modulate) Ca²⁺-mediated myoepithelial cell contraction and/or acinar valve flap for saliva efflux through the salivary duct.

While this study focused on dopamine actions in type III acini, the roles of the two distinct dopamine receptors in type II acini are thought to be similar. Although type II acini are thought to function for secretion of the precursor of cement and bioactive molecules to manipulate host immune responses (Binnington, 1978; Walker et al., 1985), they also contain epithelial cell types with similar immunoreactive patterns for D1 and InvD1L receptors. Visual investigation of fluid flow through type II acini was not possible due to lack of dopamine-induced acinar expansion as in acini type III. Employing additional experimental tools, such as electrophysiology, will be required to understand the physiological mechanism underlying dopamine and its receptors in type II acini.

The autocrine/paracrine dopamine (Koči et al., 2014) and downstream systems built in the salivary glands also probably interact with neural commands. The neural projections reaching to the basal region of acini containing MIP/SIFamide are also positive with InvD1L antibodies (Šimo et al., 2014). The co-localisation suggests that dopamine action includes modulation of MIP/SIFamide neuronal secretory activities at axon terminals. Candidate functions of MIP and SIFamide were previously suggested as controls of acinar valves, myoepithelial cells and dopamine secretion, based on the anatomy of the axon terminal in acini (Šimo et al., 2009; Šimo et al., 2013). Therefore we conclude that osmoregulatory salivary secretion in ticks involves the orchestration of interactions of multiple components for dynamic controls.

The full open reading frame (ORF) for each D1 and InvD1L were inserted into the pcDNA 3.1 (+) expression vector (Invitrogen, Carlsbad, CA, USA). Expression plasmids were prepared for high purity and concentrations (>1 μg μl⁻¹) and were then sequenced. Functional G-protein-coupled receptor expression was performed based on a previous description with modifications (Park et al., 2003; Šimo et al., 2011; Jiang et al., 2013; Šimo et al., 2014). A one-to-one mixture of the receptor-containing plasmid and the codon-optimised human apoaequorin-containing plasmid (Vernon and Printen, 2002) were used for transfections of Chinese Hamster Ovary (CHO) cells using TransIT®-LT1 (Mirus, Madison, WI, USA). Approximately 24–48 h after the transfection, cells were harvested with trypsin-EDTA (Gibco, Grand Island, NY, USA) and resuspended in 3 ml Dulbecco's modified Eagle medium: nutrient mixture F-12 (DMEM/F12) with 0.1% bovine serum albumin (BSA; Amresco, Solon, OH, USA). Coelenterazine h (final concentration 5 μmol l⁻¹; AAT Bioquest, Sunnyvale, CA, USA) was added to the resuspended cells. After a 3 h incubation, cells were additionally incubated for 30 min followed by addition of 27 ml DMEM/F12 media, including 0.1% BSA (Amresco). For the agonist assay, various dopamine agonists (final concentration 10 μmol l⁻¹) were prepared in a 96-well plate, upon which transfected cells (~15,000 cells in 50 μl) were applied. For the antagonist assay, the cells were pre-incubated with various dopamine antagonists (final concentration 10 μmol l⁻¹) for 15 min, followed by 10 μmol l⁻¹ dopamine treatment. Luminescence, the indicator of calcium mobilisation, was monitored for 20 s (100 ms exposure in every 100 ms) immediately following the application of either transfected cells (for the agonist assay) or dopamine (for the antagonist assay). Luminescence values were integrated over time for 20 s and normalised to the largest positive control response and to the negative control background values.

The following chemicals were used for the agonist and antagonist assays. Dopamine agonists included SKF82958, 6,7-ADTN [(±)-2-amino-6,7-dihydroxy-1,2,3,4-tetrahydronaphthalene hydrobromide], apomorphine (Sigma-Aldrich, St Louis, MO, USA), dopamine-3-O-sulfate and dopamine-4-O-sulfate (RTI International, Research Triangle Park, NC, USA). Dopamine antagonists included acepromazine, fluphenazine, clozapine, sulpiride, haloperidol, SCH23390, (+)-butaclamol and (−)-butaclamol (Sigma-Aldrich).

The blacklegged tick I. scapularis was purchased from the Tick Rearing Center at Oklahoma State University (Stillwater, OK, USA). Partially engorged female ticks (weighing 12–19 mg, 4–5 days blood-fed) were prepared from an artificial feeding system that was modified from the method developed by Kröber and Guerin (Kröber and Guerin, 2007). Tick feeding chambers were maintained at 37°C with a 16 h:8 h light:dark photoperiod (Kröber and Guerin, 2007). The defibrinated bovine blood (Cedarlane, Burlington, NC, USA) contained 10 μmol l⁻¹ gentamycin solution (Sigma-Aldrich) and 100 μmol l⁻¹ ATP (Sigma-Aldrich), and we replaced blood meals every 12 h. Salivary glands of partially engorged female ticks were dissected and pre-incubated in Hanks' saline (Sigma-Aldrich). For microscopic video recording of individual acini, whole salivary glands were cut into small pieces, each containing ~10 acini per group. For the in vitro fluid secretion assay, whole salivary glands were examined following pre-incubation in Hanks' saline for 1 h.

A modified Ramsay assay (Ramsay, 1954) was established for tick salivary glands that have much shorter ducts than insect Malpighian tubules. All experiments in this study were performed with female salivary glands. Salivary glands were placed in a droplet (30 μl) of Hanks' saline on a 60 mm × 15 mm Petri dish coated with Sylgard® (World Precision Instruments, Inc., Sarasota, FL, USA) after pre-incubation of isolated intact salivary glands in Hanks' saline for 1 h. The main duct was pulled out from the Hanks' saline droplet crossing over a narrow dam (0.5–1 mm) made of high vacuum grease (Dow Corning Corporation, Midland, MI, USA) and immobilised on the Sylgard Petri dish surface. Heavy mineral oil (Fisher Scientific, Fair Lawn, NJ, USA) covered both the Hanks' saline and the main salivary gland duct to collect secreted saliva (Fig. 2A). To establish basal levels of excretion induced by Hanks' saline, measurements were taken in the first 5 min after preparation of the fluid secretion assay. A micro-injector (Nanoliter 2000, World Precision Instruments) controlled by a micro-syringe pump controller (Micro4, World Precision Instruments) was used for withdrawing the secretion formed at the tip of the main duct in the heavy mineral oil, and the volume withdrawn was recorded every 5 min for 30 min. For the agonist assay, we measured secreted saliva for 30 min after drug treatment. For the antagonist assay, salivary glands were pre-incubated in antagonist for 25 min, and then agonists were administered to salivary glands in the presence of the antagonists (Fig. 2B).

For the case of acini showing both pumping and gating, the acini were counted as pumping action because the criteria for pumping action are generally inclusive of the gating action. In results, pumping and gating were combined for the expression of frequency of discharging the fluid. For the antagonist assay, salivary glands were pre-incubated in antagonist for 25 min followed by administration of agonist to salivary glands in the presence of antagonists.

This paper is contribution no. 15-114-J from the Kansas Agricultural Experiment Station.