Interference with LTβR signaling by tick saliva facilitates transmission of Lyme disease spirochetes

Significance

Among vectors, ticks transmit the most diverse human and animal pathogens, leading to an increasing number of new challenges worldwide. In this study, we found that Ltbr knockout mice were more susceptible to Lyme disease spirochetes and showed that a 15-kDa protein (Ixodes persulcatus salivary protein [IpSAP]) functioned as an immunosuppressant to facilitate the transmission and infection of Lyme spirochetes. IpSAP immunization provided mice with significant protection against I. persulcatus–mediated Borrelia garinii infection and considerable cross-protection. The significance of this study is to provide potential protein candidates for broad-spectrum vaccine development for the control and prevention of Lyme disease.

Abstract

Lyme spirochetes have coevolved with ticks to optimize transmission to hosts using tick salivary molecules (TSMs) to counteract host defenses. TSMs modulate various molecular events at the tick–host interface. Lymphotoxin-beta receptor (LTβR) is a vital immune receptor and plays protective roles in host immunity against microbial infections. We found that Ltbr knockout mice were more susceptible to Lyme disease spirochetes, suggesting the involvement of LTβR signaling in tick-borne Borrelia infection. Further investigation showed that a 15-kDa TSM protein from Ixodes persulcatus (I. persulcatus salivary protein; IpSAP) functioned as an immunosuppressant to facilitate the transmission and infection of Lyme disease spirochetes. IpSAP directly interacts with LTβR to block its activation, thus inhibiting the downstream signaling and consequently suppressing immunity. IpSAP immunization provided mice with significant protection against I. persulcatus–mediated Borrelia garinii infection. Notably, the immunization showed considerable cross-protection against other Borrelia infections mediated by other ixodid ticks. One of the IpSAP homologs from other ixodid ticks showed similar effects on Lyme spirochete transmission. Together, our findings suggest that LTβR signaling plays an important role in blocking the transmission and pathogenesis of tick-borne Lyme disease spirochetes, and that IpSAP and its homologs are promising candidates for broad-spectrum vaccine development.

Lyme disease, caused by the most common tick-borne infection in many parts of North America, Europe, and Asia, can lead to clinical manifestations such as arthritis, carditis, and neurological disorders (1). Human activity and global climate change have impacted the geographical distribution of ticks and the incidence of Lyme disease (2, 3). The causative agents of Lyme disease are spirochetes belonging to the Borrelia burgdorferi sensu lato (s.l.) group, which are transmitted by bites from infected Ixodes ticks (4). Despite considerable long-term efforts to develop vaccines against Borrelia antigens, no effective vaccines are currently available for Lyme disease. Furthermore, although the safety and efficacy of the outer-surface protein A vaccine have been verified, further efforts for improving the vaccine for application are still needed (5, 6). In addition, DNA vaccinations for Ixodes scapularis proteins Salp15, tHRF, TSLPI, and Tix-5 have been shown to fail against Borrelia challenge in mice (7). These limitations indicate that vector-based recombinant and messenger RNA (mRNA) vaccines against tick feeding and Borrelia transmission might be needed (8).

In China, the dominant B. burgdorferi s.l. genotype is Borrelia garinii, which is present in the medically important tick Ixodes persulcatus (the taiga tick). However, the transmission mechanism of the Chinese native B. garinii strain remains unclear (9). The enzootic cycle of B. burgdorferi s.l. in nature encompasses transmission from an infected tick vector to a vertebrate host and acquisition from an infected vertebrate host to a tick vector (10). Tick vectors are required for Lyme disease spirochete pathogenicity and persistence in mammalian hosts, and Lyme disease resulting from infected tick bites mostly manifests as a dysregulated immune response (11). Salivary proteins facilitate feeding and help pathogens evade both innate and adaptive host immune responses through immunosuppressive or immunoablative mechanisms, leading to efficient transmission and persistent infection of tick-borne pathogens (12). To date, some tick salivary molecules (TSMs) have been identified and characterized. Salp15, one of the best-characterized TSMs, inhibits CD4⁺ T cell activation and proliferation by interacting with the coreceptor CD4 and by attenuating interleukin-2 (IL-2) cytokine production (4, 13–15). Moreover, Salp15 can bind to B. burgdorferi and augment its transmission to and survival within mammalian hosts by protecting against antibody- and complement-mediated killing (16, 17). Antibodies against Salp15 can protect mice from B. burgdorferi infection (18). Interestingly, two Salp15 homologs from I. persulcatus, namely IperSalp15-1 and IperSalp15-2, exhibit comparable functions in the transmission of Lyme disease spirochetes as Salp15 (19). Therefore, the properties of salivary proteins in ixodid ticks may favor the development of novel broad-spectrum tick antigen-based vaccines and therapeutic strategies to combat Lyme borreliosis.

Lymphotoxins play pivotal roles in lymph node development and form essential communication links between lymph nodes and the surrounding stromal microenvironment (20). Lymphotoxin-beta receptor (LTβR) signaling can be activated by two ligands, LTα₁β₂ and LIGHT. LTβR-deficient mice show similarly defective lymphoid organogenesis as genetically LTα₁β₂-deficient mice, indicating that LTβR signaling in response to LTα₁β₂ binding is crucial for its activation (20, 21). LTβR signaling regulates leukocyte migration via the constitutive and ligand-driven nuclear factor-kappa B (NF-κB) signaling pathway (22). LTβR signaling also plays protective roles in host defense against various bacterial, viral, and parasitic infections, and disruption of LTβR signaling is associated with exacerbation of Mycobacterium tuberculosis, Citrobacter rodentium, Listeria monocytogenes, mouse cytomegalovirus, herpes simplex virus 1, Leishmania, and Toxoplasma gondii infections (21, 23). We also previously showed LTβR signaling plays a profound role in mosquito-borne Zika virus infection (24).

In the current study, we found that the LTβR signaling pathway plays a critical role in inhibiting B. garinii infection. Transmission of B. garinii spirochetes occurs at the complex interface of bacteria, tick vectors, and mammalian hosts. LTβR-dependent cross-talk between epithelial and immune cells may be a potential target of TSMs. We identified a 15-kDa protein (I. persulcatus salivary protein; IpSAP) from the salivary glands of I. persulcatus ticks and found that IpSAP functions as an immunosuppressive factor to facilitate B. garinii transmission in mice and worsen infection and lesions. Manganese salts are reported to show good adjuvant effects by strongly promoting immune responses via both cGAS-STING and NLRP3 (25). Here, we found that IpSAP immunization in mice with a colloidal manganese salt (Mn jelly; MnJ) adjuvant provided considerable cross-protection against other Borrelia strain infections in salivary gland extract (SGE) models representing other tick species.

Results

Abolishing LTβR Signaling Promotes Infection of Lyme Spirochetes.

As the major causative agent of Lyme borreliosis in China, B. garinii is transmitted to the host during tick blood feeding. Previous studies have shown that spirochetes can be phagocytized and killed by antigen-presenting cells (26, 27). Here, we examined whether B. garinii affects LTβR signaling by infecting mouse bone marrow–derived macrophages (BMDMs) and THP-1–induced macrophages using a B. garinii strain (NMJW1) isolated from I. persulcatus. As shown in Fig. 1A, B. garinii infection in the macrophages resulted in an increase in the expression of genes encoding lymphotoxin-alpha (LTα). Thus, we further explored the role of LTβR signaling in B. garinii infection in vivo. Ltbr^−/− and wild-type (WT) control C3H/HeJ mice were challenged with an intradermal inoculation of 2 × 10³ B. garinii (9, 28). At 28 d postinfection, B. garinii burden in the heart and joints increased significantly in the Ltbr^−/− mice compared with the WT mice (Fig. 1B). Moreover, the Ltbr^−/− mice showed more aggravated arthritis than the WT mice (Fig. 1 C and D). Thus, abolishing LTβR signaling promoted B. garinii infectivity, indicating the potential role of LTβR in antiinfection of B. garinii.

Fig. 1.

Abolishing LTβR promotes infectivity and pathogenesis of B. garinii. (A) Expression analysis of LTA and Lta in THP-1 cells and mouse BMDMs, respectively, post B. garinii infection (MOI 1) for 12 h. Data represent mean ± SEM; ***P < 0.001 by unpaired t test. (B) Ltbr^−/− and WT control C3H/HeJ mice were challenged with an intradermal inoculation of 2 × 10³ B. garinii. B. garinii burden in joints was examined by genomic DNA qRT-PCR analysis at 28 d postinfection. Data represent mean ± SD (n = 8); ***P < 0.001 by unpaired t test. (C) Joint footpad swelling in B. garinii–infected mice at 28 d postinfection. Data represent mean ± SD (n = 5); **P < 0.01 by unpaired t test. (D) Histopathology of joints in mice (n = 5) at 28 d post B. garinii infection. Representative images and arrows show leukocyte infiltration. (Scale bar, 400 μm.) Arthritis index scores are shown (Right). Data represent mean ± SD; *P < 0.05 by unpaired t test. All data represent at least two independent experiments. BG, B. garinii; KO, knockout; NC, negative control.

Tick Salivary Protein IpSAP Blocks LTβR Signaling.

Tick salivary proteins contribute to pathogen transmission success by disrupting host homeostasis (8). Therefore, it is reasonable to speculate that some tick proteins may interfere with LTβR signaling. Here, we investigated potential tick salivary factors targeting LTβR. We first used LTβR-Ig (immunoglobulin), a soluble fusion protein, as a competitive inhibitor of LTβR signaling to test the existence of LTβR signaling inhibitors in tick SGEs. C3H/HeJ mice were intradermally injected with 2 × 10³ B. garinii mixed with LTβR-Ig or control IgG and SGE at the same site. Another group of mice were separately injected with the mixture and SGE as a control. As shown in SI Appendix, Fig. S1, the tick SGE significantly facilitated B. garinii infection, while LTβR-Ig administration largely counteracted the local effects. Immunoprecipitation (IP)–pull down assays were performed to identify LTβR inhibitors in the SGE of I. persulcatus ticks. A 15-kDa band was coprecipitated and subjected to tandem mass spectrometry (MS-MS) analysis after trypsinization (SI Appendix, Fig. S2A). Based on National Center for Biotechnology Information (NCBI) BLAST search results, the IpSAP protein, which shares a similar EF-hand calcium-binding domain with a putative tick protein, was identified. Prior to determining the role of IpSAP in pathogen transmission and host immune response, recombinant IpSAP was expressed and purified from Escherichia coli and antibodies were generated (SI Appendix, Fig. S2B). Western blot analysis of saliva indicated that IpSAP may be a secreted salivary protein and IpSAP protein levels were remarkably higher in salivary glands and midgut compared with other body tissues (SI Appendix, Fig. S2C).

Lyme spirochetes are maintained in the midgut and saliva of tick vectors and transmitted during tick feeding. Tick salivary proteins are secreted to aid engorgement during feeding and may also assist in pathogen transmission (4). Results showed that IpSAP contained 140 amino acid residues with an approximate molecular mass of 15.5 kDa. qRT-PCR revealed that IpSAP expression was two to three times higher in the salivary glands and midgut during blood feeding (Fig. 2A) than in unfed ticks. The physiological concentrations of IpSAP in the nymph and adult tick salivary glands and midgut were also determined by enzyme-linked immunosorbent assay (ELISA) (Fig. 2B). Interestingly, Borrelia infection did not affect IpSAP levels (Fig. 2B). Surface plasmon resonance (SPR) and confocal analyses confirmed direct interactions between IpSAP and LTβR, with a dissociation constant (K_D) value of 124.1 ± 22.6 nM (Fig. 2C). Colocalization of IpSAP and LTβR was also confirmed by confocal immunofluorescence staining analysis (Fig. 2D). These data confirm that the identified tick salivary protein IpSAP targets LTβR.

Fig. 2.

Tick protein IpSAP interacts with LTβR and blocks downstream signaling. (A) Expression analysis of IpSAP in salivary glands (SGs) and midguts (MGs) of unfed and blood-fed I. persulcatus ticks (n = 50) by qRT-PCR. (B) Physiological concentration of IpSAP in SGs and MGs of Borrelia free or B. garinii–infected I. persulcatus ticks (n = 4 groups, 300 to 400 ticks in each group). (C) Analysis of direct interaction between IpSAP and LTβR by SPR. (D) Confocal microscopy analysis of THP-1–induced macrophages stained with anti-IpSAP and anti-LTβR antibodies. (Scale bar, 20 μm.) (E) Immunoblot analysis of phosphorylation of NIK, IKKα/IKKβ and IκBα and protein levels of P52, P100, NIK, IKKα and IκBα in THP-1 derived macrophages post B. garinii stimulation (MOI=1) with or without IpSAP administration for 12 hours at indicated concentration. Equal amount of BSA was added as negative control. β-actin was used as the loading control. (F) Dimerization analysis of LTβR. HSFs were stimulated with 1 nM LTα₁β₂ and 1 nM IpSAP for 1 h and cross-linked for 15 min with disuccinimidyl suberate, followed by lysis and Western blot analysis. (G) Expression analysis of CXCL8, CXCL13, CCL2, and TNFA in HSFs post B. garinii stimulation (MOI 10) with or without IpSAP administration at the indicated concentration for 24 h. Data represent mean ± SEM; *P < 0.05 and ***P < 0.001 by one-way ANOVA. Data represent at least three independent experiments for D–G and two independent experiments for A–C. ns, not significant; RU, relative unit; UF, unfed.

The interaction between IpSAP and LTβR may interfere with the downstream signaling pathway, which mounts an efficient immune response against pathogenic infection (20). Therefore, to further determine the effects of IpSAP on B. garinii stimulation, we challenged THP-1–induced macrophages with B. garinii in the presence or absence of IpSAP. Phosphorylation of NIK, IKKα/IKKβ, and IκBα was significantly induced upon B. garinii stimulation, while IpSAP administration significantly inhibited their phosphorylation levels (Fig. 2E). In addition, the ratio of P52 to P100, a hallmark of noncanonical NF-κB activation, was dramatically reduced in the THP-1–induced macrophages stimulated with B. garinii in combination with IpSAP (Fig. 2E). Dimerization of LTβR is necessary and sufficient to activate the noncanonical and canonical NF-κB pathways and consequently induce the production of inflammatory cytokines and chemokines. Skin is composed of overlying epidermis and underlying dermis. Fibroblasts are representative cells in the skin with a relatively higher level of LTβR. We found that in the absence of LTαβ₂, LTβR was predominantly monomeric in the human THP-1–induced macrophages and primary human skin fibroblasts (HSFs), while LTαβ₂ stimulation induced dimerization (Fig. 2F). LTβR dimerization induced by LTαβ₂ was substantially inhibited by IpSAP (Fig. 2F), suggesting that IpSAP directly blocks LTβR activation in both cell types. LTβR signaling and downstream NF-κB signaling regulate the production of many proinflammatory cytokines and chemokines (29). Here, IpSAP administration decreased the expression levels of CXCL13, CCL2, CXCL8, and TNFA induced by B. garinii stimulation in both HSFs and THP-1 cells (Fig. 2G and SI Appendix, Fig. S3). These results indicate that IpSAP attenuates the downstream signaling of LTβR triggered by B. garinii infection.

IpSAP Exacerbates B. garinii Infection in Mice by Inhibiting Immune Responses.

Impaired antiinfectious immune responses may delay the recognition and elimination of spirochetes and elicit weaker responses in systemic organs (30). We therefore tested immune cells in skin around the injection site and found that the percentages of macrophages and neutrophils were remarkably higher at 1 d post B. garinii infection, while coadministration of IpSAP reduced the percentages of these cells (SI Appendix, Fig. S4A). This difference indicated an IpSAP-mediated local immunosuppressive status of the tick bite (infection) site. Immunoblotting analysis of the skin around the injection site indicated that IpSAP administration decreased B. garinii infection-induced VCAM-1 and ICAM-1 levels (Fig. 3A). Flow cytometry of cells prepared from skin around the injection site was performed. The percentages of VCAM-1⁺ and ICAM-1⁺ cells increased significantly at 1 d post B. garinii infection, while IpSAP coadministration reduced them in the skin (Fig. 3B). Cell-adhesion molecules, including VCAM-1 and ICAM-1, play important roles in stabilizing leukocyte and endothelial cell interactions (31). Reductions in VCAM-1⁺ and ICAM-1⁺ cells may impair immune responses and delay spirochete recognition. We also found that the inflammatory score of the injection-site skin was significantly reduced after IpSAP coadministration at 1 d post B. garinii infection (Fig. 3C). Infection with B. garinii in mice represents an experimental Lyme disease model that recapitulates the inflammatory arthritis found in Lyme disease patients (11). To determine whether IpSAP can facilitate B. garinii infection in mice, we assessed bacterial burden using qRT-PCR of genomic DNA of B. garinii in mouse tissues at 28 d postinfection. Hematoxylin and eosin (H&E) staining revealed that IpSAP and B. garinii treatment caused more severe Lyme arthritis and spleen swelling in mice (Fig. 3D). The number of live B. garinii in the joints after 2 d of cultivation in BSK-H medium was also counted (Fig. 3D). Unsurprisingly, IpSAP administration significantly increased B. garinii load in the joint (Fig. 3D). These data collectively suggest that IpSAP augments B. garinii infectivity in mice by inhibiting local immune responses at the early stage of infection.

Fig. 3.

IpSAP exacerbates B. garinii pathogenesis in vivo by diminishing local immune responses during the early phase of infection. C3H/HeJ mice were inoculated with 2 × 10³ B. garinii combined with 40 μg/kg BSA or IpSAP. (A) Immunoblot analysis of ICAM-1 and VCAM-1 in mouse skin (n = 4) at the inoculation site at 1 d post B. garinii infection. β-actin was used as loading control and signaling intensity was quantified. Data represent mean ± SD; **P < 0.01 by one-way ANOVA. (B) Flow cytometry analysis of ICAM-1⁺ and VCAM-1⁺ cells in mouse skin (n = 4) at the inoculation site. (C) Representative H&E-stained sections of mouse skin at the inoculation site (n = 5, ear root). Inflammatory scores are shown (Right). (Scale bar, 250 μm.) Data represent mean ± SD; **P < 0.01 by unpaired t test. (D) Histopathology of joints in mice (n = 5) at 28 d post B. garinii infection. Representative images and arrows show leukocyte infiltration. (Scale bar, 400 μm.) Arthritis index scores are shown (Right). B. garinii burden in joints was examined by genomic DNA qPCR analysis at 28 d postinfection (n = 5). Live B. garinii number in the joints was counted after 2-d cultivation in BSK-H medium. Tissues were separated and cultured in an equal ratio of BSK-H medium (20 μL/mg tissue) after rapid surface sterilization. Data represent mean ± SD; *P < 0.05 and **P < 0.01 by unpaired t test. Data represent at least two independent experiments.

IpSAP Immunization Reduces Efficiency of Tick-Mediated B. garinii Transmission.

Adjuvants have been used for decades to increase the immunogenicity of antigens. The recently developed manganese-containing adjuvant MnJ works as an immunopotentiator and delivery system (25). To further confirm the physiological roles of IpSAP and its application potential for vaccines, we actively immunized mice with the IpSAP protein and different adjuvants. As shown in Fig. 4A, the MnJ adjuvant and Freund’s complete adjuvant (FCA) induced strong IpSAP-specific antibodies after one injection. The mice were then infected with B. garinii through tick bite. Transmission of B. garinii mediated by nymph I. persulcatus ticks was impaired, with lower spirochete burdens in mouse skin 7 d postinfection (Fig. 4B). As bacterial DNA can persist in tissue for weeks after clearance, we counted live B. garinii in the skin after 3 d of cultivation in BSK-H medium. Consistently, IpSAP immunization significantly reduced live B. garinii numbers in the skin (Fig. 4C), with 60% of MnJ mice and 20% of FCA mice fully protected (Fig. 4C). We used SGE to simulate a very high dose tick bite to evaluate the protective effects of IpSAP-MnJ immunization. As expected, IpSAP immunization with MnJ showed remarkable protective efficacy and abolished the effects of tick SGE (Fig. 4C). Thus, IpSAP immunization not only reduces the efficiency of tick-mediated B. garinii transmission but also indicates that MnJ is a potent adjuvant for tick antigens.

Fig. 4.

IpSAP immunization with MnJ adjuvant reduces the efficiency of tick-mediated B. garinii transmission. C3H/HeJ mice were immunized intramuscularly with IpSAP (5 μg) and MnJ adjuvant (5 μg) or FCA (50 μL) on day 0. Serum was collected on days 7 and 14 to quantify IpSAP-specific IgG by ELISA (n = 5). (A) Anti-IpSAP IgG titers in sera from mice immunized with one injection. (B) B. garinii load in skin at day 7 post tick feeding, determined by qPCR (n = 5). IpSAP-immunized and control mice were bitten by 10 B. garinii–infected I. persulcatus nymph ticks. Data represent mean ± SD; **P < 0.01 by one-way ANOVA. (C) Live B. garinii number in the skin after 3-d cultivation in BSK-H medium. At day 7 post tick feeding, mouse skin (n = 5) was separated and cultured in an equal ratio of BSK-H medium (20 μL/mg tissue) after rapid surface sterilization. Data represent mean ± SD; ***P < 0.001 by one-way ANOVA. Representative results from two independent experiments are shown.

IpSAP-MnJ Immunization Provides Cross-Protection against Other Borrelia Infections Mediated by Other Ixodid Ticks.

The properties of tick salivary proteins may be important for the transmission of tick-borne pathogens, and salivary gland protein homologs have been identified in different ticks (32–35). As such, we wondered whether IpSAP homologs may exist in other ixodid ticks. Based on NCBI BLAST and sequence analysis, several IpSAP homologous proteins of ticks were identified with two conserved EF-hand domains (Fig. 5A). The causative species of human tick bites that cause Lyme disease include I. persulcatus and Ixodes ovatus in East Asia and I. scapularis in North America (36). Here, we established and verified an artificial model of tick-mediated spirochete transmission using SGE from I. persulcatus ticks. As shown in Fig. 5B, IpSAP immunization significantly suppressed SGE-exacerbated B. garinii infection in mice. Thus, we collected SGE from I. scapularis and I. ovatus ticks to test whether IpSAP immunization provides broad protection. Notably, mice immunized with IpSAP-MnJ were more resistant to SGE-induced B. burgdorferi B31-A3 infection (Fig. 5C). Thus, we hypothesize that the homologs of IpSAP in other ixodid ticks may also play key roles in Lyme spirochete transmission.

Fig. 5.

IpSAP immunization protects mice from simulated I. scapularis– and I. ovatus–mediated B. burgdorferi s.s. infection. (A) Partial sequence similarity of IpSAP to homolog proteins from other ticks (Ds, Dermacentor silvarum, XP_037577723.1; Ha, Hyalomma asiaticum, KAH6928784.1; Hl, Haemaphysalis longicornis, BAI99729.1; Ip, I. persulcatus, QIS68510.1; Is, I. scapularis, AAY66924.1; Rm, Rhipicephalus microplus, XP_037281547.1; Rs, R. sanguineus, KAH7947016.1). EF-hand domains are indicated. The panel was generated with ESPript. (B) C3H/HeJ mice were immunized intramuscularly with IpSAP (5 μg) and MnJ adjuvant (5 μg). IpSAP-immunized or adjuvant-only C3H/HeJ mice were challenged with intradermal inoculation of 2 × 10³ B. garinii with or without 30 μg of I. persulcatus tick SGE. B. garinii burden in joints and hearts, examined by genomic DNA qPCR analysis at 28 d postinfection (n = 5); two samples were collected from one mouse, and each point represents a sampling site. Data represent mean ± SD; *P < 0.05 by unpaired t test. (C) IpSAP-immunized or adjuvant-only C3H/HeJ mice (n = 6; except Io-SGE mice, n = 4) were challenged with an intradermal inoculation of 2 × 10³ B. burgdorferi B31-A3 with or without 30 μg of tick SGE. B. garinii burden in joints and hearts, examined by genomic DNA qPCR analysis at 28 d postinfection. Data represent mean ± SD; P values were determined by unpaired t test. Bb, B. burgdorferi; Io-SGE, I. ovatus SGE; Is-SGE, I. scapularis SGE. Representative results from two independent experiments are shown.

IpSAP Homolog (IsSAP) from I. scapularis Facilitates B. burgdorferi B31-A3 Infection.

We obtained recombinant protein samples of the IpSAP homolog in I. scapularis (IsSAP) to examine its role in B. burgdorferi sensu stricto (s.s.) transmission. Compared with unfed ticks, IsSAP expression was elevated after 72 h of feeding in the tick salivary gland (1.45-fold increase) and at 24 h of feeding in the midgut (2.1-fold increase), consistent with the IpSAP results (Fig. 6 A–C). Direct interaction between IsSAP and LTβR was also confirmed by immunocytofluorescence (Fig. 6D). To determine the function of IsSAP in promoting B. burgdorferi transmission in vivo, B. burgdorferi B31-A3 spirochetes were inoculated intradermally into mice in the presence or absence of IsSAP and the serum levels of proinflammatory cytokines, T cell cytokines, and chemokines were analyzed. The production of BLC, CCL5, and GCSF was impaired by IsSAP administration at day 1 post B. burgdorferi infection (SI Appendix, Fig. S5). We then investigated the serum levels of CCL5 and GCSF at 1, 3, 7, 14, and 28 d postinfection. As shown in Fig. 6E, the levels of CCL5 and GCSF were impaired by IsSAP administration at the early phase of B. burgdorferi infection instead of the late phage. These results suggest that the impaired early immune response caused by IsSAP may exacerbate Lyme borreliosis. Like IpSAP, IsSAP administration significantly increased B. burgdorferi load in the joint and caused more severe Lyme arthritis in mice at 28 d postinfection (Fig. 6F). These data indicate that ixodid ticks may have similar tactics to facilitate spirochete transmission by targeting LTβR signaling.

Fig. 6.

Homolog of IpSAP in I. scapularis facilitates B. burgdorferi s.s. infection by attenuating host immune response. (A) Physiological concentration of IsSAP in SGs and MGs of I. scapularis ticks (n = 300). (B and C) Expression analysis of IsSAP in SGs and MGs from unfed and blood-fed I. scapularis ticks (n = 300) by qRT-PCR. (D) Analysis of direct interactions between IsSAP and LTβR by confocal microscopy analysis of THP-1–induced macrophages stained with anti-IsSAP and anti-LTβR antibodies. (Scale bar, 20 μm.) (E) Concentrations of CCL5 and GCSF were analyzed by ELISA in serum from mice infected with B. burgdorferi combined with BSA or IsSAP at the indicated time points. *P < 0.05, **P < 0.01, and ***P < 0.001 by one-way ANOVA. (F) Histopathology of joints in mice (n = 6) at 28 d post B. burgdorferi B31-A3 infection. Representative images and arrows show leukocyte infiltration. (Scale bar, 200 μm.) Arthritis index scores are shown (Right). Bacterial burdens in joints were examined by genomic DNA qPCR analysis. Live B. garinii number in the joints was counted after 2-d cultivation in BSK-H medium. Tissues were separated and cultured in an equal ratio of BSK-H medium (20 μL/mg tissue) after rapid surface sterilization. Data represent mean ± SD; *P < 0.05 and ***P < 0.001 by unpaired t test. Representative results from two independent experiments are shown.

Discussion

Unlike other hematophagous arthropods, ticks can remain attached to the host’s skin and feed for days. A variety of bioactive molecules in tick saliva can modulate host defenses of pain, itch, hemostasis, inflammation, innate and adaptive immunity, and wound healing (37). Tick–host–pathogen interactions are important for the efficient transmission of Lyme spirochetes and development of Lyme borreliosis. Many bioactive molecules in tick saliva were found to play important roles in favoring the transmission of B. burgdorferi bacteria from the tick to the host (10, 12). Understanding the mechanisms by which specific salivary compounds facilitate pathogen transmission and dissemination should help develop strategies to combat arthropod-borne infections. Here, we identified a tick salivary protein, IpSAP, which was highly expressed in the tick salivary gland and midgut. IpSAP attenuates the immune responses triggered by B. garinii infection by specifically binding to LTβR and blocking its downstream signaling.

The pleiotropic functions of signaling through the LTβR pathway are crucial for lymphoid organ development, microenvironment maintenance, and host immunity (38). LTβR signaling in endothelial cells regulates the production of chemokines, cytokines, and cell-adhesion molecules through NF-κB pathways (39). Inhibition of the LTα₁β₂-LTβR pathway attenuates host defense in a wide range of experimental infectious diseases, including intracellular mycobacterial infections, infectious colitis, and various viral and parasitic infections (23). However, the effect of altering LTα₁β₂–LTβR interactions on immune cell migration remains relatively unexplored. Previous research has shown that blockade of LTβR activation is associated with increased bacterial and viral burdens, impaired antiviral cytotoxic T cell immune responses, increased immune cell infiltration to infectious sites, and lower iNOS activity during microorganism infections (21, 23). Our results showed that during B. garinii infection, IpSAP treatment decreased VCAM-1 and ICAM-1 expression and the percentages of VCAM-1⁺ and ICAM-1⁺ cells in the skin. These findings indicate a complex role of LTβR signaling in leukocyte trafficking, thus requiring further study. LTβR signaling in intestinal epithelial cells can modulate innate immune responses against mucosal bacterial infection by recruiting neutrophils to the infection site at the early stage via CXCL1 and CXCL2 chemokine production, which is essential for epithelial IL-23 production and epithelial injury protection (40, 41). Epithelial LTβR signaling is required for the regulation of the early innate response against pathogens, and LTβR is considered a critical molecule that coordinates the innate and adaptive immune responses during bacterial and viral infections. Tick salivary compounds play multiple roles in the modulation of host defense and pathogen transmission (42, 43). Using Ltbr^−/− mice of C3H/HeJ background, our results showed that the LTβR signaling pathway played a key role in B. garinii infection. There is a strong possibility that tick vectors may have evolved mechanisms to facilitate blood sucking which was exploited by Lyme disease spirochetes in their favor. B. burgdorferi s.l. is an extracellular pathogen capable of establishing relatively uncommon persistent infection in vertebrates even after antibiotic treatment (44). The immune response dysregulation has been attributed to the clinical symptoms of Lyme borreliosis. Here, inhibition of LTβR signaling by IpSAP administration leads to an impaired local immune response, resulting in reduced recruitment of neutrophils and macrophages. Therefore, this immunocompromised effect diminishes the clearance of B. garinii in early infection.

Although early-stage Lyme disease can be treated with antibiotics, millions of people still suffer from chronic infections. All approved vaccines against infectious disease target pathogen components directly so far. Given that the interactions between ticks and hosts are complex but somewhat conserved, discovery of their pathogenic transmission strategies may help develop measures against tick-borne disease. Although Lyme vaccines employ conventional recombinant vaccination strategies, more and more researchers have focused on DNA- and mRNA-based vaccines as alternative platforms (7, 8). However, DNA vaccination does not appear to induce adequate immune responses to tick antigens and subsequent protection against B. burgdorferi (7). In this study, we found that the MnJ adjuvant, which works as an immunopotentiator and delivery system, was a more promising approach for tick antigens (25). Notably, IpSAP immunization with MnJ largely abolished the effects of the tick SGE. The MnJ adjuvant induced strong IpSAP-specific antibodies with one injection and 60% of the immunized mice were fully protected against tick-mediated B. garinii infection. Ixodid ticks are vectors for most known tick-borne pathogens. It has been reported that homologs of tick salivary gland proteins have related TSM-like features (32–35). Here, IpSAP-MnJ immunization also showed considerable cross-protection against simulated I. scapularis– and I. ovatus–mediated B. burgdorferi B31-A3 infections. Importantly, the homolog of IpSAP from I. scapularis also interfered with LTβR signaling. Our study provides more strong evidence that ixodid ticks share multiple tactics to combat the host immune system.

While sucking blood, ticks create an immunosuppressed cutaneous environment that enhances the local establishment of Lyme disease spirochetes. In the current study, we identified the important role of LTβR in host defense against the infection of tick-borne Lyme spirochetes. We suspect that SAPs are vital countermeasures in ticks against host defense responses, resulting in an environment that facilitates successful feeding and establishment of spirochetes within the host. Thus, stimulating host responses by modulating LTβR activation at the tick bite site may provide an advantage for recognizing and eliminating spirochetes. Furthermore, our findings suggest that IpSAP is a tick salivary antigen that affects the early stages of B. burgdorferi s.l. infection in vertebrate hosts, and may be applicable for broad-spectrum vaccine development for Lyme disease.

Materials and Methods

Ticks and Tick Dissection.

I. persulcatus and I. ovatus ticks were collected and passed several times to obtain pathogen-free nymphs at the Kunming Institute of Zoology (45,46). I. scapularis ticks were obtained from Oklahoma State University’s tick-rearing facility and maintained by established methods (47). Before placing them on mice or sheep for feeding, these ticks were maintained at room temperature and 90% relative humidity under 14-h light and 10-h dark periods (48). For tick feeding, six cells were placed on the sheep’s back. In each cell, there were 45 female ticks and 25 male ticks as previously described (49). Unfed ticks were used as control ticks for temporal gene expression studies. Approximately four or five ticks were removed from the sheep at regular intervals during blood feeding.

Tick tissues were dissected within 4 h of their removal from the host. Fully engorged and dropped-off ticks were kept under the above-mentioned conditions for ovipositioning. Ice-cold 3-(N-morpholino)propanesulfonic acid (MOPS) buffer (100 mM) containing 20 mM ethylene glycol tetraacetic acid (pH 6.8) was used for dissecting tick tissues from nymphs and adult ticks (49). After dissection, tick tissues were washed gently in the same buffer. SGEs were lyophilized and stored at −80 °C. For RNA isolation, the dissected tissues were stored in RNAlater (Ambion) or TRIzol reagent (Invitrogen). Institutional Animal Care and Use Committee–approved protocols 10042001 and 08110401 were followed for all animal studies performed at the University of Southern Mississippi, and the experiments performed at the Kunming Institute of Zoology were conducted in accordance with the guidelines approved by the Kunming Institute of Zoology, Chinese Academy of Sciences Animal Care and Use Committee (SMKX2017021).

RNA Extraction, Complementary DNA Synthesis, and Gene Expression Analysis.

Total RNA was isolated from unfed and partially fed tick salivary gland and midgut tissues by using TRIzol reagent (Invitrogen). Isolated RNAs were either used directly for complementary DNA (cDNA) synthesis or stored at −80 °C. cDNAs were synthesized by using iScript Reverse Transcription Supermix (Bio-Rad).

To quantify the transcriptional gene expression of ticks, qPCR was performed on a Bio-Rad thermocycler that had a built-in CFX96 real-time fluorescence detection system. SYBR Green qPCR Master Mix was obtained from Bio-Rad, and instructions were followed as given by the manufacturer. All the samples were run in triplicate along with no-template controls for each primer.

For cells, total RNA was isolated by using TRIzol reagent (Invitrogen). cDNA was reverse-transcribed by using M-MLV reverse transcriptase (Promega). Real-time qPCR was performed by using the primers listed in SI Appendix, Table S1.

Expression of Recombinant Proteins.

DNA sequences encoding IpSAP and IsSAP were synthesized and cloned into the pET-32a (+) vector (Sangon Biotech). The designed cleavage site -DDDDK-, which is susceptible to enterokinase, was inserted between the 6 × His tag and the N-terminal of IpSAP. The IpSAP/pET-32a (+) construct was transformed into E. coli strain BL-21 (DE3) for expression. After hydrolysis by enterokinase, recombinant IpSAP was purified through one step of Sephadex G-50 (Superfine, Amersham Biosciences, 2.6 × 100 cm) gel filtration for preliminary fractionation and one step of C4 reversed-phase high-performance liquid chromatography (UniSil C4 column, 5-μm particle size, 10 × 250 mm) for final purification. IsSAP was expressed in the same way.

Polyclonal Antibody Preparation and ELISA.

According to our previous report (50), rabbit polyclonal antibodies against IpSAP and IsSAP were produced in 6-mo-old male rabbits (purchased from Kunming Medical University), which were primed by multiple subcutaneous injections of IpSAP or IsSAP (1 mg) mixed with 1 mL FCA (Sigma-Aldrich), respectively, followed by subcutaneous booster injections of the antigens (0.5 mg) mixed with incomplete Freund’s adjuvant (Sigma-Aldrich) every 2 wk. After the antibodies were produced, blood was collected from the heart of the rabbits with syringes, and the sera were isolated and stored at −80 °C for further use.

For ELISA, protein samples were absorbed to a 96-well flat-bottom immunoplate at 4 °C overnight. After washing, the plate was blocked for 1 h at 37 °C by using 1% bovine serum albumin (BSA) in PBST (phosphate-buffered saline with 0.1% Tween-20) and incubated with antibodies in PBST with 1% BSA for 2 h at 37 °C. The anti-rabbit IgG secondary antibody (1:2,000, horseradish peroxidase [HRP]–labeled; KPL) was then added. 3,3′,5,5′-Tetramethylbenzidine was used for color development, and the absorbance at 450 nm was measured on a plate reader (Epoch Etock, BioTek). Mouse CCL5 (CME0048, 4A Biotech) and GCSF (CME0030, 4A Biotech) in the samples were measured using ELISA kits according to the manufacturers’ instructions.

LTβR-Binding Specificity of IpSAP and IsSAP.

For SPR assays, following chip activation by 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (0.4 M) and N-hydroxysuccinimide (10 mM), an extracellular fragment of LTβR (629-LR-100, R&D Systems) was immobilized on the CM5 sensor chip (BR100012, GE) by amine coupling. The remaining activated sites on the CM5 sensor chip were blocked by 75 μL of ethanolamine (1 M, pH 8.5). Real-time detection was performed on a Biacore 3000 (GE Healthcare Life Sciences) with a constant flow rate of 30 μL/min. HBS-EP (10 mM Hepes, pH 7.4, 150 mM NaCl, 3.4 mM ethylenediaminetetraacetate, and 0.005% surfactant P20) running buffer was injected onto the immobilized LTβR for 5 min as a blank to exclude any nonspecific signal before sample injection and after chip regeneration. Serially diluted IpSAP and IsSAP samples were injected onto the immobilized LTβR for 5 min. Then, HBS-EP running buffer was introduced onto the sensor surface for 5 min to collect the dissociation data. The chip was regenerated by injecting 5 μL of 10 mM NaOH, followed by incubation with HBS-EP running buffer. The equilibrium K_D for binding was determined by the BIA evaluation program (GE Healthcare Life Sciences).

Lyme Spirochetes.

B. garinii strain NMJW1 and B. burgdorferi B31-A3 cultures were grown to mid- to late-log phase (∼10⁶ spirochete cells per milliliter) in Barbour–Stoenner–Kelly BSK-II medium (BSK-H medium, Complete sterile-filtered, with 6% rabbit serum [Sigma], 1% CO₂). B. garinii strain NMJW1 and B. burgdorferi B31-A3 cultures were grown at 30 or 34 °C, respectively (9, 51).

In Vitro Stimulation of Spirochetes.

THP-1 cells and primary HSFs were obtained from the Kunming Cell Bank, Kunming Institute of Zoology, Chinese Academy of Sciences. THP-1 cells were cultured in RPMI medium 1640 (Gibco) supplemented with 10% fetal bovine serum (FBS) in 5% CO₂ at 37 °C. Primary HSFs were cultured in Dulbecco’s modified Eagle’s medium/F12 medium (Gibco) supplemented with 15% FBS in 5% CO₂ at 37 °C. Mouse BMDMs were prepared as previously described (24).

For spirochete stimulation, cells in nonantibiotic medium were seeded in 12-well plates at ∼1 × 10⁶ per well. THP-1 cells were pretreated with 100 nM phorbol myristate acetate (Sigma-Aldrich) for 24 h. The nondifferentiated cells in the supernatant were washed with fresh medium. The spirochetes of B. garinii were grown to midlog phase and washed twice in the corresponding medium. After counting, the spirochetes were added to the plates to make the density of B. garinii reach 1 MOI (multiplicity of infection) for THP-1 and BMDMs or 10 MOI for HSFs, and the testing sample was dissolved in the same medium and added at an equal volume. The plates were horizontally centrifuged at 300 × g for 5 min to facilitate contraction and incubated as mentioned above (52).

In Vivo Artificial Infection.

Ltbr^−/− mice were previously described (40) and were kindly provided by Mingzhao Zhu, Institute of Biophysics, Chinese Academy of Sciences. LTβR-Ig for blocking the LTβR pathway (53) was kindly provided by Hua Peng, Institute of Biophysics, Chinese Academy of Sciences. Ltbr^−/− mice in the C3H/HeJ background were generated and kept under specific pathogen-free conditions in the Animal Resource Center at the Kunming Institute of Zoology, Chinese Academy of Sciences.

For the experiment, each group of five female mice were separately infected by intradermal inoculation with 2 × 10³ spirochetes mixed with different concentrations of IpSAP or IsSAP or the control protein BSA in 100 μL of 0.9% salt water. For LTβR-Ig treatment, each group of five female C3H/HeJ mice was injected with 2 × 10³ B. garinii mixed with 1 μg of LTβR-Ig or control IgG and 30 μg of tick SGE intradermally on the same site (left ear root) or separately injected with the mixture and SGE as control (right ear root). Serum samples of the mice were collected from their tail blood and stored at −80 °C. At the indicated time points, mice were killed and the serum, spleen, tibiotarsal joint, lymph node, heart, bladder, and skin at the injection site were removed aseptically. A part of the tissues was fixed in 10% formalin, and 5-μm sections were stained with H&E and examined. Other samples were kept at −80 °C until use. The animal experiments performed at the Kunming Institute of Zoology were conducted in accordance with the guidelines approved by the Kunming Institute of Zoology, Chinese Academy of Sciences Animal Care and Use Committee (SMKX2017021).

Inflammatory Bioarrays.

For semiquantitative analysis of cytokines, the serum of the mice was mixed equivalently for every group, and bioarrays were set up at 120 μL per group using mouse inflammation antibody arrays (AAM-INF-G1, RayBiotech). Following the user’s manual, briefly, the arrays were blocked and incubated with samples and detecting reagents at room temperature. The fluorescence intensity data were read on a scanner and normalized to positive control spots before analysis.

Quantification of Lyme Disease Spirochetes.

For the quantification of B. garinii and B. burgdorferi B31-A3 in vivo, the specific gene Flab was quantified and normalized to mouse β-actin as described previously (16). Total DNA of tissue samples was extracted by using a TIANamp Genomic DNA Kit (Tiangen). qPCR was performed in accordance with the manufacturer’s protocol with a StepOnePlus Real-Time PCR System (Thermo Fisher Scientific) using the total DNA as the template. The reaction mixtures (20 μL) included template DNA (1 μL), SYBR Green buffer mix (10 μL; TransStart Tip Green qPCR SuperMix, TransGen Biotech), 10 μM primer mix (1 μL), and ddH₂O (8 μL). For live spirochete quantification, mouse tissues were separated and cultured in equal ratios of BSK-H medium (20 μL/mg tissue) after rapid surface sterilization.

Immunoblot Analysis.

LTαβ₂ (8884-LY-025, R&D Systems) and anti-LTβR (ab10493, Abcam) were used for LTβR dimerization analysis as in our previous study (24). Total proteins extracted from cells or tissues were used for Western blotting analysis with primary antibodies. The protein concentrations of extracts from tick tissues and mouse skin were determined by a BCA protein assay kit (Thermo) and adjusted before loading. Primary antibodies were anti–ICAM-1 (ab179707, Abcam), anti–VCAM-1 (ab134047, Abcam), anti–p-IKKα (Ser176)/IKKβ (Ser177) (2078P, CST), anti-p100/p52 (4882T, CST), anti-LTβR (ab70063, ab10493, Abcam), anti–NF-κB–inducing kinase (NIK) (ab191591, Abcam), anti-IKKα (ab32041, Abcam), anti–p-NIK (Thr559) (sc-12957, Santa Cruz), anti–p-IκBα (2859L, CST), anti-IκBα (9242L, CST), and anti–β-actin (sc-69879, Santa Cruz). The secondary antibodies were HRP-labeled anti-rabbit, anti-mouse, and anti-goat antibodies (KPL). The total proteins were separated by 12% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and electrotransferred onto a polyvinylidenedifluoride (PVDF) membrane (Roche Diagnostics). The PVDF membrane was blocked with Tris buffered saline with Tween 20 (TBST) containing 5% BSA at room temperature for 1 h. After washing three times with TBST, the membrane was incubated overnight with primary antibodies at 4 °C and incubated with the secondary antibody for 1 h at room temperature. After washing with TBST, the membranes were developed with an enhanced chemiluminescence kit (Tiangen) on an ImageQuant LAS 4000 mini system (GE Healthcare Life Sciences).

Immunofluorescence Staining and Microscopy.

Cells were fixed in 4% paraformaldehyde for 15 min at room temperature and then permeabilized with 0.3% Triton X-100 (T0694, Amresco) for 20 min, washed with PBS, and blocked in 2% BSA for 1 h. Cells were stained with antibodies overnight at 4 °C. After washing, cells were stained with fluorescence-conjugated secondary antibody for 30 min and mounted with ProLong Gold Antifade Reagent with DAPI (8961S, CST). Immunostaining was detected by a FluoView 1000 confocal microscope (Olympus).

Flow Cytometry.

Single-cell suspensions from tissues were prepared as previously reported (54). Briefly, skin was removed from mice and digested by 0.2 mg/mL liberase and 100 μg/mL DNase I for 1 h at 37 °C. The tissues were further disrupted, and single-cell suspensions were filtered through 100-μm cell strainers. Cells were stained as described in our previous study (24). Fluorescence-activated cell sorter (FACS) analysis was performed on a BD FACSCalibur flow cytometer. Data analysis was performed using FlowJo v10.5.3 software.

Mouse Immunization and Tick Challenge.

Each group of 10 female C3H/HeJ mice was immunized with 20 μg of IpSAP protein and 20 μg BSA in FCA (Sigma-Aldrich) or MnJ adjuvant as previously reported (18). The serum samples of the mice were collected from their tail blood. An antibody titer test was performed accordingly (25).

Pathogen-free I. persulcatus nymphal ticks were naturally infected with B. garinii by feeding on infected mice. For tick-mediated B. garinii infection, up to 10 ticks were placed on each mouse and allowed to be fully engorged. The mice were kept in a relatively isolated clean room for further analysis as described above.

Statistical Analysis.

Statistical analysis is indicated in each figure legend. A P value ≤ 0.05 was considered significant. All data analyses were performed using Prism 7 (GraphPad Software).

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix.