Sand fly salivary proteins induce strong cellular immunity in a natural reservoir of visceral leishmaniasis with adverse consequences for Leishmania

Abstract

Immunity to a sand fly salivary protein protects against visceral leishmaniasis (VL) in hamsters. This protection was associated with the development of cellular immunity in the form of a delayed-type hypersensitivity response and the presence of IFN-gamma at the site of sand fly bites. To date, there are no data available regarding the cellular immune response to sand fly saliva in dogs, the main reservoirs of VL in Latin America, and its role in protection from this fatal disease. Two of 35 salivary proteins from the vector sand fly Lutzomyia longipalpis, identified using a novel approach termed reverse antigen screening, elicited strong cellular immunity in dogs. Immunization with either molecule induced high IgG(2) antibody levels and significant IFN-gamma production following in vitro stimulation of PBMC with salivary gland homogenate (SGH). Upon challenge with uninfected or infected flies, immunized dogs developed a cellular response at the bite site characterized by lymphocytic infiltration and IFN-gamma and IL-12 expression. Additionally, SGH-stimulated lymphocytes from immunized dogs efficiently killed Leishmania infantum chagasi within autologous macrophages. Certain sand fly salivary proteins are potent immunogens obligatorily co-deposited with Leishmania parasites during transmission. Their inclusion in an anti-Leishmania vaccine would exploit anti-saliva immunity following an infective sand fly bite and set the stage for a protective anti-Leishmania immune response.

Figure 1. Dogs develop a strong humoral and cellular immune response to bites of Lu. longipalpis sand flies.

Dogs (n = 9) were exposed for 10 min to bites of 20 sand flies three times at one week intervals (first exposure, E1; second exposure, E2; third exposure, E3). (A) Weekly measurement of IgG, IgG₁ and IgG₂ antibody levels in dogs exposed to sand flies. (B) Induration score in a representative dog 48 h after each of three sand fly exposures. The induration score is an arbitrary scale corresponding to the area of induration and redness resulting from multiple bites where 1≤1 cm², 2≤5 cm², 3≤10 cm² and 4>10 cm². (C) Representative H&E staining of biopsies taken from sand fly bite sites prior to exposure (E0) and 48 h after each of three sand fly exposures (E1–E3). Note marked cellular infiltrate within dermis and thickening of epidermis in E3. (D) Immunohistochemical labeling of tissue sections from E3 demonstrating the presence of abundant CD3+T cells (CD3), macrophages (Mac387) and eosinophil granules (Luna stain).

Figure 2. Identification of salivary proteins from Lu. longipalpis that produce a cellular immune response in dogs.

(A) A schematic representation of the reverse antigen screening approach based on the intradermal injection of DNA plasmids in dogs previously exposed to sand fly bites (first exposure, E1; second exposure, E2; third exposure, E3). (B–F) Dogs pre-exposed to sand fly bites were challenged intradermally with DNA plasmids and one pair of salivary gland homogenate (SGH) and PBS (positive and negative controls, respectively) and investigated 48 h post-injection. (B) The number of dogs showing local induration and/or erythema at the site of injection for 35 DNA plasmids coding for secreted salivary molecules. Yellow bars highlight the response of dogs to LJM17 and LJL143. (C) Photograph to demonstrate specificity of the cellular reaction to DNA plasmids and SGH. (D) The diameter of erythema in the absence (◊) or presence (♦) of induration for each dog at the site of injection of SGH, PBS, LJL143 and LJM17 (reactive plasmids) and LJL04 and LJM111 (intermediate and non-reactive plasmids, respectively). (E–F) Skin biopsies (6mm) obtained from injection sites were cut in half and processed for histology and RNA extraction. (E) Representative H&E staining and immunohistochemical labeling of dermal T cells (anti-CD3) and macrophages (Mac387) at the injection sites of SGH, LJL143, LJM17 and LJM111. Note marked dermal infiltrates of inflammatory cells characterized as CD3+ T cells and scattered macrophages (Mac387) in the SGH, LJL143 and LJM17. There is no inflammation with LJM111. (F) Reverse-transcriptase quantitative PCR showing the expression levels of IFN-γ, IL-12, IL-4 and TGF-β for LJL143, LJM17, a pair of SGH and control (a mix of PBS and empty plasmid) 48 h post-injection. Error bars represent means±S.E.

Figure 3. Sand fly salivary recombinant proteins produce a DTH response in dogs previously exposed to sand flies.

(A) Purity of the recombinant salivary proteins produced by HEK-293F mammalian cells and purified by a HPLC nickel trap column. (B) The diameter of erythema in the absence (◊) or presence (♦) of induration for each dog at the site of injection 48 h after challenge with salivary gland homogenate (SGH), PBS, recombinant proteins rLJL143 and rLJM17 (reactive), rLJM111 (non-reactive) and a non-related tick recombinant protein TB179. (C) Representative H&E staining and immunohistochemical labeling of T cells (anti-CD3) and macrophages (Mac387) at the injection sites of rLJL143, rLJM17 and rLJM111. Note marked dermal infiltrates of inflammatory cells characterized as CD3⁺ T cells and scattered macrophages (Mac387) with rLJL143 and rLJM17; rLJM111 is negative.

Figure 4. Dogs immunized with the Lu. longipalpis salivary molecules LJM17 or LJL143 develop strong and specific humoral and cellular immune responses.

(A) Total IgG and (B) IgG₁ and IgG₂ antibody levels up to day 224 in dogs immunized with either LJM17 (n = 5), LJL143 (n = 5) or the empty plasmid (n = 5). LJL143- and LJM17-immunized dogs were tested using the appropriate recombinant proteins (LJL143 and LJM17). Dogs immunized with the empty vector (control dogs) were tested against both recombinant proteins for IgG, (LJL143-control, LJM17-control), IgG₁ (IgG₁-control) and IgG₂ (IgG₂-control). (C) In vitro IFN-γ production by PBMC from LJL143- and LJM17-immunized and control dogs stimulated with media (Med), ConcavalinA (ConA), salivary gland homogenate (SGH), rLJM17 or rLJL143 two weeks after the final vaccination. Error bars represent means±S.E.

Figure 5. Bites of Lu. longipalpis sand flies induce a strong focal and systemic adaptive cellular immune response in dogs immunized with LJL143 or LJM17.

(A–C) Dogs were exposed to uninfected and infected sand flies for 10 min one month after the final immunization with either LJM17, LJL143 or the empty plasmid (control). (A–C) Skin biopsies (6mm) obtained from bite sites 48 h post challenge with 20 and five uninfected and 10 infected sand flies were cut in half and processed for histology and RNA extraction. (A) Representative H&E staining and immunohistochemical labeling of T cells (anti-CD3) and macrophages (Mac387) at the bite sites of 20 uninfected sand flies in LJL143- and LJM17-immunized and control dogs. (B) Reverse-transcriptase quantitative PCR showing the expression levels of IFN-γ, IL-12, IL-4 and TGF-β at the bite sites of 20 or five uninfected sand flies in LJL143- and LJM17-immunized and control dogs (for control dogs RNA was combined from sites of 20 and 5 uninfected sand fly bites). (C) Same as (B) using 10 infected sand flies. Histological sections from bite sites of five uninfected and 10 infected sand flies are provided as Figure S1 and Figure S2, respectively. (D–E) PBMC from LJL143- and LJM17-immunized and control dogs obtained one week after exposure to sand flies. (D) Frequency and mean fluorescence intensity (MFI) of CD3⁺ T cells following stimulation with medium, rLJL143 or rLJM17. (E) Frequency of CD4⁺ and CD8⁺ T cells expressing IFN-γ in PBMC from LJL143- and LJM17-immunized dogs. Error bars represent means±S.E. * P<0.05, ** P<0.01.

Figure 6. Macrophages efficiently kill L. i. chagasi in vitro following the addition of autologous lymphocytes from LJL143- and LJM17-immunized dogs stimulated with SGH.

Percent of infected macrophages 72 h after the addition of autologous lymphocytes alone (Med), or together with ConA or SGH from PBMC of LJL143- and LJM17-immunized and control dogs. The percentage of amastigote-infected macrophages was evaluated by microscopic examination of Giemsa-stained preparations. Error bars represent means±S.E. * P<0.0001.