Immunity against a ‘silent’ salivary antigen of the Lyme vector Ixodes scapularis impairs its ability to feed

Abstract

Ixodes scapularis ticks transmit the Lyme disease agent in the US. Although strong anti-tick immunity mediates tick rejection by certain vertebrates, only a few antigens have been molecularly characterized. We show that guinea pig vaccination against a secreted tick salivary immunomodulator—sialostatin L2—can lead to decreased feeding ability of I. scapularis nymphs. Increased rejection rate, prolonged feeding time and apparent signs of inflammation were observed for nymphs attached to vaccinated animals, indicating a protective host immune response. Interestingly, sialostatin L2 humoral recognition does not take place upon repeated tick exposure in control animals, but only in the vaccinated animals that neutralize sialostatin L2 action. Therefore, we demonstrate an essential sialostatin L2 role upon nymphal infestation that can be blocked by vertebrate immunity and we propose the discovery of similarly ‘silent’ antigens towards the development of a multicomponent vaccine that will protect against tick bites and the pathogens they transmit.

Introduction

Due to their small size and difficulty of detection, Ixodes scapularis nymphs are the key vector stage transmitting the Lyme agent Borrelia burgdorferi in both humans and animals in disease-endemic regions of the US. An alternative or complementary component of an integrated approach against ticks and/or the pathogens they transmit is the development of anti-tick vaccines. This idea is mainly supported by a long-known phenomenon (1): certain vertebrate hosts (e.g. guinea pigs) develop tick hypersensitivity upon repeated exposure to ticks, preventing ticks from taking a blood meal. This anti-tick immunity can, upon tick re-exposure, prevent Borrelia transmission, as well (2). This is not surprising, as intense work on the transcriptome and proteome of I. scapularis glands (3) has revealed numerous components of its saliva and, more importantly, their potential pharmacologic action on host coagulation(4), fibrinolysis (5), immunity (6–11), and angiogenesis (12). This action not only facilitates tick attachment to the vertebrate host and acquisition of the blood meal, but also creates a tick/vertebrate host interface advantageous for pathogen transmission. Such a pathogen transmission facilitation has already been demonstrated for Salp15, a tick salivary immunomodulator exploited during Lyme disease transmission (9). Therefore, vertebrate host immunity that blocks pharmacologic action of tick salivary constituents has the potential to affect tick feeding ability and transmission of tick-borne pathogens.

Additionally, as more salivary compounds are characterized, it becomes clear that some of them exert their detrimental action on the host in low nanomolar to picomolar concentration. This holds true for sialostatin L2 as well, a tick salivary cystatin that inhibits with high affinity cathepsins L and S -key enzymes in vertebrate immunity- serving as an immunomodulator upon tick infestation (8). Given the low amount of sialostatin L2 required for the inhibition of its enzymatic targets, we questioned whether sera coming from guinea pigs exposed to ticks for four successive periods recognize sialostatin L2 tick secretion. To our surprise, humoral recognition could not be detected and therefore we introduced the term ‘silent’ antigen for sialostatin L2 to describe the fact that although the protein is found in the tick-host interface, it can not be recognized by humoral vertebrate immunity upon repeated tick infestations, possibly due to its function or the amount of its secretion. Furthermore we show that vertebrate immunity can be pre-sensitized against sialostatin L2 (by vaccinating the animals with supraphysiological amounts of the respective protein) neutralizing its action and leading to impaired blood feeding by I. scapularis nymphs. In conclusion we show that a key salivary player for tick feeding success escapes from the ‘attention’ of vertebrate immunity, thus proposing a different approach for the discovery of similarly ‘silent’ tick salivary antigens that can confer protection from tick bites. Accordingly, this idea can possibly find applications in the attempt to develop vaccines against other parasitic diseases as well.

Materials and Methods

Unless otherwise indicated, the protocols followed standard procedures (13) and the experiments were performed at room temperature (25 ± 1°C). If not otherwise stated, all reagents were purchased from Sigma Chemical Corporation (St. Louis, MO).

Tick Exposure and Animal Handling

Pathogen-free nymphal I. scapularis ticks were obtained from colonies maintained at Oklahoma State University. Ticks were maintained at 24°C and 90% relative humidity under a 14:10 h photoperiod. Approximately 3-wks-old (250–300 g) outbred female albino Hartley guinea pigs were obtained from Charles River Laboratories (Wilmington, MA). Guinea pigs were maintained and subjected to intradermal vaccination four times in two wks intervals in the left and right front lateral edge of their back with 50 µg of pure recombinant sialostatin L2 (see below) on each side (100 µg total) according to the approved guidelines of the National Institutes of Health Animal Care and Use Committee (NIH/NIAID LMVR6, approved October 1^st 2006).

Two wks after the last vaccination and prior to tick placement, guinea pigs were sedated and the area between the ears was shaved. Twenty nymphal ticks were placed on the shaved area and allowed to attach. The head was completely covered with stockinet during sedation until all ticks were attached, verified by gentle pull using forceps. Ticks that did not attach by the time the guinea pig recovered from anesthesia were removed and subtracted from the total number of ticks placed. During tick attachment, guinea pigs were checked daily, and detached ticks were weighed and recorded. The animal cages were set in pans with water to ensure ticks could not escape, and nutrients provided ad libitum.

Sialostatin L2 Preparation and LPS Decontamination

Sialostatin L2 gene was overexpressed in Escherichia coli strain BL21(DE3)pLysS bacteria and the corresponding active protein was purified in 0.8 mM stock solution (10 g/l) as previously described (6). This stock solution was subjected to removal of any potential LPS contamination by Arvys Proteins Inc. (Stamford, CT) using a detergent-based method. Samples were subjected to decontamination treatment five times, and endotoxin presence by the end of the procedure was estimated as lower than 0.00004 endotoxin units/µg of protein (roughly, less than 3 × 10⁻¹⁴ g of endotoxin/µg of protein) with a sensitive fluorescence-based endotoxin assay (PyroGene recombinant Factor C endotoxin detection system; Lonza Biologics Inc., Portsmouth, NH). The protein concentration was then adjusted with sterile PBS in 1 g/l for vaccination of the animals.

ELISA and Enzymatic Assays

ELISA titer was determined under standard methods (13) as the dilution of the primary Ab that gives an absorbance read at 405 nm twice higher than that of the negative control (pre-immune serum from the same animal). The plates were coated with protein by overnight incubation with 50 µl of 1 µgr/ml of sialostatin L2 in 50mM sodium carbonate pH 9.5. As a secondary Ab, we used an alkaline phospatase-conjugated goat anti-guinea pig IgG diluted 5000 times in 50 mM Tris pH 8, 150 mM NaCl, 0.1 % Tween 20 (TTBS), and the 96-well plate was incubated with p-nitrophenyl phosphate liquid alkaline phosphatase substrate for 1 h at 37°C before absorbance reading in a THERMOmax colorimetric plate reader (Molecular Devices, Sunnyvale, CA).

Enzymatic assays for cathepsin L inhibition were performed as previously described (6). Total IgGs were purified from guinea pig sera using the Melon™ gel IgG purification kit (Pierce, Rockford, IL) according to manufacturer’s instructions and subsequently tested for neutralizing sialostatin L2 activity.

Statistical Analysis

Data are shown as mean ± SEM where applicable. Statistical differences were analyzed by analysis of variance (ANOVA), KS test (Kolmogorov-Smirnov comparison of two data sets), χ² test, and Student t-test depending on their applicability to the hypothesis tested. A P value of 0.05 or less was considered statistically significant.

Results

Guinea pigs, but not their natural hosts, develop immune-mediated tick hypersensitivity after repeated exposure to I. scapularis with a mechanism similar to humans (1). Although sialostatin L2 is a secreted salivary protein and its transcripts were detected in nymphal salivary glands (3), it was not possible to detect any sialostatin L2 recognition by ELISA assays or western blots using sera from guinea pigs exposed to I. scapularis nymphs for four successive periods in 2-wks intervals (data not shown). On the other hand, recognition of higher molecular weight salivary gland antigens could be detected by both ELISA and western blots using the same sera as a primary Ab and salivary gland homogenates as antigens (J. M. Anderson, personal communication). Perhaps the low amount of protein necessary to exert its action (8) upon tick infestation accounts for this unexpected result. This led us to test the immunogenicity of the protein by injecting supraphysiological amounts in guinea pigs. Therefore, four animals were vaccinated intradermally with sialostatin L2 protein in a much higher amount (100 µg) than nymphal salivary glands can secrete upon tick infestation. Two wks post vaccination, sera samples were prepared from the animals and tested by ELISA for sialostatin L2 recognition. An average antisialostatin L2 titer of 156±41.5×10³ (n=4) was observed, ranging betwen 8×10⁴ to 25.6×10⁴ in the sialostatin L2 vaccinated animals (Supplemental Figure 1).

Having presensitized the animals for a tick salivary gland antigen, we subsequently exposed them to I. scapularis nymphs by administering 20 ticks in each animal (see Material and Methods). All of the nymphs that were not attached to the animals within the first 3 hours of placement were removed using forceps. During the first three days of tick exposure all nymphs found on the bottom of the animal cage that had a body weight similar to that before attachment to the guinea pigs were collected and counted and were considered to be “unfed” as a consequence of early rejection (within the first 72 h of exposure). Figure 1A shows that increased early rejection was observed for the ticks attached to the sialostatin L2 vaccinated group. The average early rejection rate in the vaccine group (n=4) was 29 ± 5.6%, three times higher than that in the control group (n=4) (10 ± 1.1%; P = 0.016). We also observed higher early rejection percentage for the ticks attached to animals displaying higher antisialostatin L2 titer (Supplemental Figure 1).

Figure 1. Impaired feeding of I. scapularis nymphs when attached on sialostatin L2-vaccinated animals.

A) Increased rejection was observed for the ticks attached to sialostatin L2 vaccinated guinea pigs. The bar represents the mean percentage of early tick rejection (see results) from the control and the experimental group, while the lines represent the standard error of the mean. Each group consisted of four animals. The asterisk denotes a statistically significant difference between the means of the two experimental groups (P<0.05).

B) There was a statistically significant delay in drop off of ticks attached in the vaccine group compared with those in the control group between 4 and 6 days after initial attachment, as shown in the graph. The photo shows the difference in size of the ticks recovered from the two experimental groups 4 days post initial attachment.

C) Distribution of weight of nymphs recovered from the control and the vaccine groups. A statistically significant difference in mean nymphal weight is seen (1.9 mg for the nymphs attached to the vaccine group versus 2.8 mg for those attached to the control group, represented with a line in the distribution graph). The asterisk denotes a statistically significant difference between the means of the two experimental groups (P<0.05).

The remainder of the ticks—those that were not rejected during the first 72 h—managed to receive blood and consequently increased their body weight. Approximately 15% of the ticks that fed on the vaccinated animals (and not those that fed on the control animals), triggered apparent signs of inflammation in their feeding sites, i.e., increased redness and edema formation, 72 h post attachment (Supplemental Figure 2). Although this host reaction affected the feeding (and engorgement) of some ticks (Supplemental Figure 2, upper tick), other ticks appeared unaffected by this local reaction (Supplemental Figure 2, lower tick). This sign of apparent increased vertebrate host response to nymphal infestation was further supported by the delayed drop-off of ticks feeding on the vaccinated group (P = 0.03, χ² for days 4–6) (Fig. 1B, graph). Indeed, most of the engorged ticks (data not shown) dropped between days 4 and 5 in the control group, but between days 5 and 6 in the vaccine group (Fig. 1B, graph). The delay in blood feeding was even more obvious when comparing the body size and weight of ticks that dropped off on day 4 (Fig. 1B, photo). While ticks from the control group were almost fully engorged, those recovered from the vaccine group appeared (and weighed) partially engorged. Moreover, Figure 1C reveals that by the end of the nymphal feed, there was a statistically significant reduction (P = 0.009) in mean weight of ticks feeding on the vaccine group compared with those feeding on the control group (1.9 ± 0.2 mg and 2.8 ± 0.2 mg, respectively). A closer analysis of tick weight distribution (Supplemental Table I) reveals that a larger proportion of the nymphs attached on vaccinated animals fed poorly, while a smaller proportion of them fed well when compared with nymphs that infested control animals.

All of the above data collectively reveal that impaired I. scapularis nymphal feeding was observed upon attachment to sialostatin L2-vaccinated animals, which was the combined result of an early rejection of the nymphs and reduction in the ability of remaining nymphs to imbibe blood. Supplemental Figure 3 incorporates both mechanisms contributing to the observed feeding impairment; the graph represents the average weight per nymph initially attached to each animal (taking into consideration in our calculations the weight of the rejected nymphs in both groups). It is interesting that there is a greater effect on ticks attached to guinea pigs with higher antisialostatin L2 titer. This observation led us to investigate in vitro whether this Ab titer can also neutralize sialostatin L2 action, i.e., inhibition of cathepsins (8). Total IgGs were purified from animal #3 (sialostatin L2 vaccinated) and animal #5 (control) sera (see supplemental figures 1, 3), prior to their exposure to ticks. As a result of the IgG purification procedure there was a 5-fold reduction in the titer of animal #3, which was estimated as 4.8 × 10⁴ ± 0.6 × 10⁴. Subsequently, 5, 2, and 1 nM of sialostatin L2 were incubated in the presence or absence of 1 or 5 µl of purified IgG from animals #3 and #5 in 50 µl of cathepsin L assay buffer for 10 min (6). Subsequently, purified cathepsin L was added to the mix and incubated for another 10 min before addition of a fluorogenic substrate for estimation of cathepsin L enzymatic activity in triplicates (6). A statistically significant inhibition of sialostatin L2 activity on cathepsin L was observed upon addition of antisialostatin L2 IgG but not control IgG (Fig. 2A). The only exception was when incubating 1 µl of antisialostatin L2 IgGs with 5 nM of the protein that a statistically not-significant reduction in protein activity was observed (Fig. 2A). Although the inhibitory activity of sialostatin L2 was not completely neutralized in our assays, this could be due to the fact that the pH of cathepsin L assay buffer is 5.5, which is not optimal for binding of Abs to their respective antigens.

Figure 2. An insight into the mechanism underlying the observed feeding impairment of I. scapularis nymphs.

A) A statistically significant inhibition of sialostatin L2 action could be observed in in vitro assays for cathepsin L inhibition (6) upon addition of 1 or 5 µl of antisialostatin L2-purified IgGs (but not when adding purified IgGs from control animals) in 50 µl of reaction mix. Three different concentrations of sialostatin L2 were tested in the sera inhibition assays (5 nM, 2 nM, and 1 nM) in triplicate.

B) A boost in the mean titer of the vaccinated guinea pigs was observed two wks post tick infestation. The bar represents the mean from the sialostatin L2 vaccinated guinea pigs before and after exposure to ticks, while the lines represent the standard error of the mean. The group consisted of four animals. The asterisk denotes a statistically significant difference between the mean antisialostatin L2 titer before and after exposure to ticks (P<0.05).

Given the potential of the sera obtained from vaccinated animals to neutralize tick sialostatin L2 action, we next tested whether this neutralizing recognition of the tick inhibitor by the vertebrate immune system took place upon guinea pig infestation with I. scapularis nymphs. Sera were prepared 2 wks after removal of the last tick from the guinea pigs, and their sialostatin L2 titers prior to and after tick exposure were compared in the same ELISA plate. As shown in Figure 2B, a statistically significant ~2-fold boost in the titer of the animals was revealed (from mean antisialostatin L2 titer of 156±41.5×10³ before exposure to ticks to 286±71×10³ after tick exposure, paired t-test P value=0.04). Moreover a statistically insignificant fluctuation of the antisialostatin L2 titer was observed in similarly vaccinated animals that were not exposed to ticks (control group, data not shown) collectively suggesting that sialostatin L2 secretion was recognized by the vertebrate immune system upon tick infestation on the vaccinated guinea pigs.

Discussion

The increase in the number of cases of tick transmitted diseases in humans has enhanced research effort towards tick-host-pathogen interaction studies and more specifically to molecular dissection of tick salivary secretion, which has already been shown to significantly facilitate transmission of pathogens (14). The biochemical characterization of tick salivary constituents highlights their potential pharmacological action on the vertebrate host in nanomolar or picomolar concentration as it is the case for sialostatin L2 (8). Therefore we next questioned whether this salivary effector is recognized by vertebrate humoral immunity upon repeated exposure to ticks; the lack of recognition led us to introduce the term ‘silent’ antigens to cover all tick secreted salivary antigens that are not recognized by vertebrate immunity upon repeated exposure to ticks. Our results show that pursuing similarly ‘silent’ antigens could reveal novel anti-tick vaccine antigens that could complement the four tick salivary antigens whose vaccines were shown to partially protect from Ixodes ricinus or I. scapularis infestation or/and the pathogens they transmit (15–17).

We have previously shown (8) that reduction in the sialostatin L2 salivary transcripts by RNAi in adult ticks results impaired feeding in rabbits. We herein demonstrate that neutralizing immunity against sialostatin L2 secretion triggers a similar phenotype upon nymphal infestation in guinea pigs, supporting the notion for a pivotal immunomodulatory role of sialostatin secretion upon tick infestation. The observed boost in the anti-sialostatin L2 titers of the animals after exposure to ticks suggests that recognition of sialostatin L2 secretion from salivary glands, and possibly from other tissues (e.g. midgut), took place during tick feeding and contributed to the impairment of feeding.

Numerous research papers (including this one) reach the same conclusion: that ticks—and blood-feeding arthropods in general—are not mere syringes that transmit pathogens, but clever pharmacologists (18). Over their evolution, adaptation to the vertebrate host has refined the composition and amount of their saliva constituents through intense selection at the population level. The survivors are thus those that 'know well' what they will face upon infestation of the vertebrate host. Although the discovery of tick hypersensitivity on certain vertebrate animals upon re-exposure to ticks many decades ago (1) turned research efforts toward the antigenic molecules that appear to dictate this hypersensitivity, here we uncover a constituent of the salivary armamentarium of I. scapularis that goes undetected by the humoral immunity of these sensitized animals. Moreover, our results propose that by 'teaching' vertebrate immunity to recognize these pharmacologically active ‘silent’ antigens of tick saliva (by vaccinating higher amounts of protein than those usually secreted by ticks upon infestation) we can elicit a protective immunity beneficial for the host.