Research ArticleInfectious Disease

Identification of broadly conserved cross-species protective Leishmania antigen and its responding CD4+ T cells

See allHide authors and affiliations

Science Translational Medicine  21 Oct 2015:
Vol. 7, Issue 310, pp. 310ra167
DOI: 10.1126/scitranslmed.aac5477

Committing Leishmania vaccine to memory

Leishmaniasis is a potentially fatal disease caused by a protozoal parasite transmitted through sand fly bites. There is currently no vaccine, but affected individuals are resistant to further infection, suggesting vaccination is possible. Now, Mou et al. have found that vaccination with an immunodominant antigen—phosphoenolpyruvate carboxykinase (PEPCK)—protects against leishmaniasis. The authors identified PEPCK by examining peptides that could elicit memory T cell responses from healed but not uninfected animals. PEPCK was conserved in all pathogenic Leishmania and induced immune responses in both infected mice and human cells. Protection in mice was effective across species and was durable, supporting testing of a PEPCK-based vaccine in humans.

Abstract

There is currently no clinically effective vaccine against leishmaniasis because of poor understanding of the antigens that elicit dominant T cell immunity. Using proteomics and cellular immunology, we identified a dominant naturally processed peptide (PEPCK335–351) derived from Leishmania glycosomal phosphoenolpyruvate carboxykinase (PEPCK). PEPCK was conserved in all pathogenic Leishmania, expressed in glycosomes of promastigotes and amastigotes, and elicited strong CD4+ T cell responses in infected mice and humans. I-Ab–PEPCK335–351 tetramer identified protective Leishmania-specific CD4+ T cells at a clonal level, which comprised ~20% of all Leishmania-reactive CD4+ T cells at the peak of infection. PEPCK335–351–specific CD4+ T cells were oligoclonal in their T cell receptor usage, produced polyfunctional cytokines (interleukin-2, interferon-γ, and tumor necrosis factor), and underwent expansion, effector activities, contraction, and stable maintenance after lesion resolution. Vaccination with PEPCK peptide, DNA expressing full-length PEPCK, or rPEPCK induced strong durable cross-species protection in both resistant and susceptible mice. The effectiveness and durability of protection in vaccinated mice support the development of a broadly cross-species protective vaccine against different forms of leishmaniasis by targeting PEPCK.

INTRODUCTION

Leishmaniasis is endemic in 88 countries and affects more than 12 million people, and an estimated 1.5 million to 2 million new cases and 70,000 deaths occur each year (1, 2). Although there is currently no effective vaccine for preventing human cutaneous leishmaniasis, the observation that most individuals who recover from clinical disease are resistant to subsequent reinfection strongly suggests that vaccination is feasible. Many experimental vaccines comprising heat-killed or recombinant Leishmania antigens can induce effector T helper 1 (TH1) response and transient protection to secondary Leishmania major challenge, but meaningful, long-lasting protection is rarely obtained (3, 4).

As in humans, recovery from experimental L. major infection in C57BL/6 mice is associated with long-term immunity to rechallenge infections. This infection-induced immunity is mediated by interferon-γ (IFN-γ)–producing CD4+ T cells (1, 35), but the antigens that induce and/or maintain this immunity and the fine-specificity of the protective CD4+ T cells are not known. The lack of effective vaccines against cutaneous leishmaniasis could also be attributed in part to the experimental approaches currently used in defining potential vaccine candidates, including using highly susceptible Balb/c mice to test in silico–predicted antigens (6, 7). Because Balb/c mice are easily manipulated to become resistant, this model falls short of the infection outcome in the resistant C57BL/6 mice. Although effector and central memory–like cells are induced in C57BL/6 mice after L. major infection and are thought to contribute to protection against secondary challenge (4, 5, 8), no study has identified Leishmania-specific memory T cells at the clonal level, and nothing is known about the antigens that induce their activation. Here, we used cellular immunology and proteomics approaches to identify a highly immunogenic and conserved peptide of Leishmania and its specific CD4+ T cells at the clonal level. We extensively characterized the expansion, contraction, T cell receptor (TCR) fine-specificity, effector activities, and functions (proliferation, cytokine production, and adoptive transfer of protection) of these clonal CD4+ T cells. We show that the parent protein, which is expressed at both the promastigote and amastigote stages of Leishmania, elicits strong cellular immune response in infected mice and human patients and demonstrates its potential as a promising cross-species vaccine candidate against two major forms of leishmaniasis.

RESULTS

Tandem mass spectrometry identifies naturally processed L. major antigenic peptides displayed on dendritic cell major histocompatibility complex class II molecules

To determine the antigens that elicit protective immunity and their responding CD4+ T cell at a clonal level, we identified naturally processed major histocompatibility complex (MHC) class II–restricted L. major peptides by a proteomics-based approach from infected bone marrow–derived dendritic cells (BMDCs). Infected BMDCs up-regulated expression of MHC class II and other costimulatory molecules (fig. S1). Infected BMDCs stimulated CD4+ T cells (~30%) from healed (but not naïve) mice to proliferate, and ~6.5% (corresponding to ~22% of proliferating cells) produced IFN-γ (Fig. 1A), suggesting that Leishmania-reactive CD4+ T cells from healed mice recognize L. major–derived peptides presented by MHC class II molecules on infected BMDCs.

Fig. 1. Identification of peptides presented by MHC class II molecules on infected dendritic cells (DCs).

(A) L. major–infected BMDCs induce proliferation and IFN-γ production by CD4+ T cells from healed mice. CFSE (carboxyfluorescein diacetate succinimidyl ester)–labeled splenic CD4+ T cells from naïve or healed mice were cocultured with L. major–infected BMDCs, and proliferation and IFN-γ production were assessed by flow cytometry. (B) Venn plot showing tandem mass spectrometric results of peptides isolated from lysates of L. major–infected BMDCs as identified by GPM and ProteinPilot software. aa, amino acid. (C) Amino acid sequence of peptides selected for in vitro restimulation of CD4+ T cells. (D to I) Synthetic (PEPCK335–351) peptide stimulates proliferation and IFN-γ production by CD4+ T cells from healed mice. CFSE-labeled CD4+ T cells from healed mice were stimulated with synthetic peptides (20 μM) using irradiated syngeneic naïve splenocytes as antigen-presenting cells (APCs). After 7 days, the cells were restimulated with peptides and rIL-2 (10 U/ml) for another 7 days, and proliferation (D) and IFN-γ production (E) were determined by flow cytometry. The culture supernatant fluids were assayed for IFN-γ by ELISA (F). P3-stimulated cultures were rested for 14 days and restimulated for 24 hours with P3 or other (P1, P2, or P4) peptides, and IFN-γ secretion was measured by ELISA (G). Splenocytes from healed mice were stimulated with different peptides (P1 to P4) or freeze-thawed L. major antigen (FT-Ag) overnight, and IFN-γ production was assessed by ELISpot assay (H and I). Results presented are representative of two to three independent experiments with similar results (P values shown, two-tailed t tests, n = 3).

Next, we purified MHC class II–peptide complexes using anti–I-Ab affinity chromatography (9, 10), eluted the bound peptides, and analyzed them by tandem mass spectrometry. Automatically generated peaks were searched against L. major and mouse genome databases using the Global Proteome Machine (GPM) and ProteinPilot software. We identified 24 and 55 Leishmania-derived peptide sequences from infected (but not uninfected) BMDCs, using GPM and ProteinPilot, respectively, of which 13 were co-identified by the two software (Fig. 1B). Characteristic of MHC class II binding peptides, the sequences range from 11 to 29 (average 17) amino acids and 71% of the peptides were found within nested sets (table S1).

Synthetic L. major peptides activate Leishmania-reactive CD4+ T cells from healed mice

Healing of primary L. major infection is associated with strong CD4+ T cell recall responses (proliferation and IFN-γ production) in vitro and in vivo (4, 11). Therefore, we synthesized four peptides (P1 to P4) that were predicted by computer modeling to stably interact with I-Ab with low binding energy (Fig. 1C and fig. S2) and assessed their ability to recall CD4+ T cells from healed mice in vitro. P3 (PEPCK335–351), derived from Leishmania glycosomal PEPCK (gene: LmjF27.1805) that is highly conserved in all Leishmania (fig. S3), induced robust proliferation (Fig. 1D) and IFN-γ production (Fig. 1E) by CD4+ T cells from healed mice. We further validated IFN-γ induction by PEPCK335–351 using enzyme-linked immunosorbent assay (ELISA) (Fig. 1F). To assess specificity of PEPCK335–351 response, we rested P3-stimulated CD4+ T cells for 2 weeks and then restimulated them with different peptides (P1 to P4). Only P3-restimulated cells produced measurable amounts of IFN-γ, suggesting that the responding CD4+ T cells were specific to P3 (Fig. 1G). We also validated the stimulatory ability of P3 to activate Leishmania-reactive CD4+ T cells by enzyme-linked immunospot (ELISpot) assay (Fig.1, H and I).

I-Ab–PEPCK335–351 tetramer detects PEPCK-specific CD4+ T cells in blood and tissues of L. major–infected mice

We produced I-Ab–PEPCK335–351 tetramer to study PEPCK-specific CD4+ T cell responses at the clonal level. We detected PEPCK-specific CD4+ T cells at a frequency of ~10−4 to 10−5 in spleens but not in draining lymph nodes (dLNs) or blood of uninfected mice, and these cells displayed a naïve phenotype (that is, >99% were CD44lo; Fig. 2A). An irrelevant I-Ab–CLIP tetramer did not give significant positive staining at any point after infection (Fig. 2B, 5 weeks shown). After infection, the numbers of PEPCK-specific CD4+ T cells markedly increased (>500-fold), representing ~6, 1, and 12% of total CD4+ T cells in the spleens, dLNs, and blood, respectively, at the peak of infection (Fig. 2, A and C to G). In addition, virtually all (>98%) PEPCK-specific CD4+ T cells were CD44hi, consistent with an activated phenotype (Fig. 2A). As healing occurs, the numbers of PEPCK-specific CD4+ T cells rapidly declined but were maintained in these tissues for more than 60 weeks at significantly higher frequency than in naïve mice (Fig. 2, C to G). Thus, PEPCK-specific CD4+ T cells underwent expansion, contraction, and maintenance phases of T cell response, characteristics of antigen-specific activation. We also assessed whether PEPCK-specific CD4+ T cells home to the infection site. As shown in Fig. 2H, PEPCK-specific CD4+ T cells comprised about 12% of all CD4+ T cells in the infected footpads, and all were CD44hi. Given that PEPCK335–351 is conserved in all Leishmania, we assessed whether I-Ab–PEPCK335–351 tetramer could detect PEPCK-specific CD4+ T cells in tissues from mice infected with other Leishmania spp. I-Ab–PEPCK335–351 tetramer robustly detected PEPCK-specific CD4+ T cells in the spleens and livers of Leishmania donovani–infected (Fig. 2I) and dLNs of Leishmania mexicana–infected mice (Fig. 2J), suggesting that PEPCK335–351 is a conserved immunogenic epitope.

Fig. 2. Expansion, effector activities, and contraction of PEPCK-specific CD4+ T cells in L. major–infected mice are followed by stable maintenance after lesion resolution.

(A and C to G) Infected C57BL/6 mice were sacrificed at indicated times, and the percentages (A and E to G) and absolute numbers (C and D) of PEPCK335–351–specific CD4+ T cells in spleens, popliteal lymph nodes (pLN), and blood were assessed by flow cytometry. (B) Representative I-Ab–CLIP control tetramer staining of splenocytes from 5 week–infected mice. (H) Total mononuclear cells isolated from 5 week–infected footpads were stained with I-Ab–PEPCK335–351 tetramer and assessed by flow cytometry. (I and J) C57BL/6 mice were infected with L. donovani (I) or L. mexicana (J), and after 3 weeks, and the frequency of PEPCK335–351–specific CD4+ T cells in the livers and spleens (I) or pLN (J) was determined. Results are representative of three (A to H) and two (I and J) independent experiments (n = 4 to 5 mice in each group per experiment) with similar results.

PEPCK-specific CD4+ T cells are oligoclonal in their TCR gene usage

Next, we determined the TCR gene usage of tetramer-binding CD4+ T cells. Using a panel of antibodies specific for different TCRβ genes, we found that ~55 and ~35% of PEPCK-specific CD4+ cells preferentially use vβ8 and vβ6 genes, respectively (Fig. 3, A and B). To validate this result, we sorted single (n = 100) tetramer-binding cells from the spleens of infected mice and amplified their TCRβ and TCRα complementary DNA (cDNA) pairs using the single-cell reverse transcription polymerase chain reaction (RT-PCR) from each cell. We analyzed the sequences of the obtained TCRβ and TCRα cDNA by direct sequencing and transduced the TCRβ and TCRα cDNA pair from a single cell into TG40 cell line or primary CD4+ T cells to confirm their antigen specificity. Most of the transduced TG40 (>90%) and CD4+ T cells (>60%) that expressed the cloned TCR pair strongly bound the I-Ab–PEPCK335–351 tetramer (Fig. 3, C and D), and the transduced CD4+ T cells proliferated and produced IFN-γ in response to PEPCK335–351 stimulation in vitro (Fig. 3D). In few cases where the transfected TCR did not bind the tetramer, we recloned another TCRα gene from the cells (because some T cells often express dual TCRα) (12) and confirmed their antigen specificity. Figure 3 (E to G) shows the TCR repertoire that bound I-Ab–PEPCK335–351 tetramer. We found that ~76 and ~17% of tetramer-positive cells used TCR Vβ8 (TRBV13) and Vβ6 (TRBV19) genes, respectively (Fig. 3E), suggesting that the diversity of PEPCK-specific CD4+ T cells was highly restricted. Figure 3F shows the TCRα gene usage. With the exception of TRAV16 (1%), all the TCRα-transfected TG40 that bound the tetramer were from the TRAV6 gene group: TRAV6-3 (25%), TRAV6-4 (3%), TRAV6-5 (8%), and TRAV6-7 (63%). The framework amino acid sequences of TRAV6-3, TRAV6-4, TRAV6-5, and TRAV6-7 were similar (fig. S4), whereas that of TRAV16 is distinct. The results again indicate (as for TCRβ) a highly restricted TCRα repertoire usage. Figure 3G shows the pair of TRAV gene and TRBV gene with the CDR3 amino acid sequences, showing the diversity in amino acid sequence within the CDR3 of the specific V genes. Two Leishmania-specific T cell populations (indicated by *) expressed the same TCRα but expressed TCRβ with different CDR3β. Collectively, these results show that although the repertoire of PEPCK-specific CD4+ T cells is highly restricted, these cells are oligoclonal in nature.

Fig. 3. Analysis of PEPCK-specific CD4+ T cell TCRαβ gene usage.

(A) Representative flow cytometry analysis of PEPCK-specific CD4+ T cell TCRβ after staining with a panel of mAb (monoclonal antibody) specific for different TCR Vβ gene products. (B) Mean ± SE of the percentage of cells positive for the different TCR Vβ gene products from three independent experiments. (C to G) Determination and confirmation of TCR gene repertoire of specific T cells from L. major–infected mice by single-cell RT-PCR. (C and D) Determination of TCR antigen specificity by staining TCRβ and TCRα cDNA–transduced TG40 cells (C) and primary CD4+ T cells with I-Ab–PEPCK335–351 reagent and cell proliferation and IFN-γ production by transduced primary cells responding to PEPCK335–351 (D). (E) TCR repertoire analysis of PEPCK-specific CD4+ T cell TCRβ from sorted single cells. Numbers of analyzed T cells that expressed the indicated TRBV are shown. (F) TCR repertoire analysis of PEPCK-specific CD4+ T cell TCRα from sorted single cells. Numbers of analyzed T cells that expressed the indicated TRAV are shown. (G) The diversity in amino acid sequence within CDR3 of the different TCR Vβ and Vα genes are shown. Blue, Vβ6; red, Vβ8; green, Vβ14. “*” indicates two T cell clones that expressed the same TCRα but expressed TCRβ with different CDR3β genes.

PEPCK-specific CD4+ T cells are polyfunctional effector cytokine producers in vitro and ex vivo

Next, we assessed effector activities of PEPCK-specific CD4+ T cells including proliferation and production of IFN-γ and tumor necrosis factor (TNF), key cytokines that mediate anti-Leishmania immunity (13). Greater than 90% of PEPCK-specific CD4+ T cells in the dLNs (Fig. 4A) and spleens (fig. S5A) produced IFN-γ and TNF directly ex vivo. In contrast, only 28% of PEPCK-specific CD4+ T cells produced interleukin-2 (IL-2) and none produced IL-10 and IL-17 (Fig. 4A and fig. S5A). Most of PEPCK-specific CD4+ T cells in the dLNs (>85%) and spleens (~70%) coproduced IFN-γ and TNF directly ex vivo (Fig. 4B and fig. S5B), suggesting that they have polyfunctional properties. In addition, more than 70% of lesion-derived and 90% of blood-derived PEPCK-specific CD4+ T cells produce IFN-γ directly ex vivo (Fig. 4C). PEPCK-specific CD4+ T cells represent ~17% of recalled Leishmania-reactive CD4+ T cells, and >90% of Tet+ cells proliferated and produced IFN-γ in response to L. major stimulation in vitro (Fig. 4D).

Fig. 4. PEPCK-specific CD4+ T cells are polyfunctional effector cytokine producers.

(A) dLN cells from 5 week–infected mice were stimulated with PMA (phorbol 12-myristate 13-acetate), ionomycin, and BFA (brefeldin A) for 4 hours, and IFN-γ, TNF, IL-2, IL-17, and IL-10 production by PEPCK335–351–specific (upper panels) and nonspecific (lower panels) CD4+ T cells was analyzed directly ex vivo by flow ctyometry. (B) In addition, the coexpression of IFN-γ and TNF, IL-2, IL-17, and IL-10 was also assessed. (C) Total mononuclear cells isolated from footpads or blood of infected mice were stimulated as above, and IFN-γ secretion by PEPCK335–351–specific CD4+ T cells was assessed. (D) CFSE-labeled splenocytes from healed mice (>14 weeks after infection) were stimulated with L. major–infected BMDCs for 5 days, and proliferation and IFN-γ production by PEPCK335–351–specific and nonspecific CD4+ T cells were analyzed. Results are representative of two to three independent experiments (n = 4 to 5 mice in each group per experiment) with similar results.

PEPCK-specific CD4+ T cells from healed mice display memory characteristics and protect against L. major challenge

Next, we assessed PEPCK-specific CD4+ T cells for expression of markers associated with memory cells. At the peak of infection (6 weeks after infection), <4 and >80% of antigen-experienced (CD44hi) PEPCK-specific CD4+ T cells were CD62Lhi and Ly6Chi, respectively, which is consistent with these cells having an effector phenotype (Fig. 5A). After disease resolution (>14 weeks after infection), a significant percentage of PEPCK-specific CD4+ T cells expressed CD62Lhi and down-regulated their expression of Ly6C, suggesting that some may have differentiated into memory cells (Fig. 5A). In contrast, the phenotype of nonspecific (tetramer-negative) CD4+ T cells did not significantly change over time (fig. S6). When mice that healed their primary infection were challenged with L. major, PEPCK-specific CD4+ T cells underwent massive expansion (17- to 19-fold over primary response) in the blood and spleens (Fig. 5, B to E) within 7 days after challenge. Adoptive transfer of a few thousand (2 × 104) highly purified PEPCK-specific CD4+ T cells (Fig. 5F) induced protection comparable to 4 million nonspecific CD4+ T cells from healed mice (Fig. 5G). Collectively, these results indicate that some PEPCK-specific CD4+ T cells differentiated into memory cells after resolution of infection.

Fig. 5. PEPCK335–351–specific CD4+ T cells from healed mice display memory markers and enhanced expansion after secondary L. major challenge.

(A) Infected C57BL/6 mice were assessed ex vivo at indicated times for the expression of CD62L and Ly6C by CD44+CD4+ PEPCK335–351–specific T cells in the spleens and dLNs by flow cytometry. (B to E) Fourteen weeks after infection, some mice (secondary) and age-matched naïve controls (primary) were rechallenged with 5 × 106 L. major and sacrificed at indicated times, and the frequency of PEPCK335–351–specific CD4+ T cells in the blood (B and C) and spleens (D and E) was determined directly ex vivo by flow cytometry. (F) PEPCK-specific (Tet+) and non–PEPCK-specific (Tet) CD4+ T cells were purified from the spleens of L. major–infected mice by cell sorting. Tet+ (2 × 104), Tet (2 × 105 or 4 × 106), and naïve total CD4+ T cells (controls) (4 × 106) were adoptively transferred intravenously into naïve C57BL/6 mice that were then challenged with L. major the next day. (G) After 4 weeks, mice were sacrificed to determine parasite burden. Results presented are representative of two independent experiments with similar results (P values shown, two-tailed t tests, n = 4).

PEPCK is expressed in both promastigotes and amastigotes and induces strong immune responses in different hosts infected with L. major

We cloned L. major PEPCK gene, expressed the recombinant protein in Escherichia coli (Fig. 6A), and raised rabbit polyclonal antibodies against it. Western blot analysis showed that in addition to recognizing rPEPCK, the polyclonal antibody also recognized only one specific protein band from the cell lysate of L. major promastigotes and amastigotes (Fig. 6B) with molecular weight corresponding to L. major PEPCK (58 kD). Confocal microscopy confirmed the expression of PEPCK in the promastigotes and amastigotes and further showed that PEPCK was mainly expressed in the cytosol (Fig. 6C). To validate the strong stimulatory property of P3 peptide, we restimulated purified CD4+ T cells from healed mice with irradiated naïve splenocytes pulsed with rPEPCK and assessed anti-PEPCK recall response by several parameters. As shown in Fig. 6 (D to G), rPEPCK induced robust T cell proliferation and IFN-γ production as assessed by flow cytometry (Fig. 6D), ELISA (Fig. 6E), and ELISpot (Fig. 6, F and G) assays.

Fig. 6. Glycosomal phosphoenolpyruvate carboxykinase (PEPCK) is expressed by promastigotes and amastigotes and elicits strong cellular immune response in L. major–infected mice and humans.

(A) SDS–polyacrylamide gel electrophoresis showing the purity of recombinant PEPCK expressed in E. coli. (B) Western blot showing expression of PEPCK in lysates of L. major promastigotes and amastigotes. rPEPCK was included as a positive control. (C) Immunofluorescence showing expression of PEPCK in promastigotes and lesion-derived amastigotes. (D to J) Cellular immune responses against PEPCK in mice and humans infected with L. major. CFSE-labeled CD4+ from the spleens of healed mice were stimulated with soluble Leishmania antigen (SLA; 50 μg/ml) or rPEPCK (0.5 μg/ml) for 5 days, and proliferation and IFN-γ secretion were assessed by flow cytometry (D). The production of IFN-γ in cultures was determined by ELISA (E) and ELISpot (F and G) assays. PBMCs from patients that healed ZCL (n = 28) or unexposed controls (n = 7) were stimulated with rPEPCK (20 μg/ml) or SLA (10 μg/m) for 5 days and pulsed with [3H]thymidine (2 μCi per well) for 6 hours, and the uptake of [3H]thymidine was measured by scintillation counter (H). The levels of IFN-γ (I) and granzyme B (Grz B) (J) in the supernatant fluids were also determined by ELISA. Results presented are representative of three (D to G, n = 4 to 5 mice in each group per experiment) independent experiments with similar results (P values shown, two-tailed t tests, n = 3 to 28).

Next, we assessed the cellular immune responses in humans against rPEPCK by measuring cell proliferation, IFN-γ, and granzyme B production by peripheral blood mononuclear cells (PBMCs) from 7 healthy subjects and 28 individuals who recovered from zoonotic cutaneous leishmaniasis (ZCL) caused by L. major. The indexes of proliferation obtained in healed ZCL individuals varied from 0.44 to 9.43 (mean ± SD, 2.03 ± 2.1), compared with (0.76 to 2.38) in healthy subjects (1.28 ± 0.54) (Fig. 6H). In addition, rPEPCK induced significantly (P < 0.05) higher IFN-γ production by PBMCs from healed individuals (3.75 ± 4.69 ng/ml) than in those from healthy controls (0.82 ± 1.06 ng/ml) (Fig. 6I). Moreover, IFN-γ levels in the supernatant fluids of rPEPCK-stimulated PBMCs from healed ZCL individuals were significantly (P < 0.001) higher than those in unstimulated conditions (data not shown). rPEPCK also induced significantly (P < 0.01) higher granzyme B production in PBMCs from healed ZCL patients (Fig. 6J). Collectively, these results indicate that L. major infection induces activation of PEPCK-specific T cells in both mice and human patients who are maintained after disease resolution and are easily recalled in vitro.

Vaccination with PEPCK induces strong protective immunity against virulent L. major challenge

The preceding observations strongly suggest that vaccination with PEPCK might induce protective immunity against Leishmania spp. Therefore, we assessed the outcome of infection in mice vaccinated with either P3 peptide, DNA expressing full-length PEPCK, or rPEPCK in CpG adjuvant. We used C57BL/6 mice because experimental infection in this strain more closely mimics the cutaneous disease in humans. Vaccinated mice mounted strong CD4+ T cell proliferative and IFN-γ responses against L. major in both the local dLNs and spleens (fig. S7, A to L). Vaccinated mice were strongly protected against virulent L. major challenge as evidenced by markedly reduced lesion sizes (Fig. 7, A, C, and E) and parasite burden (Fig. 7, B, D, and F) in the challenged footpads. The marked reduction in lesion size and parasite burden was associated with significantly higher proliferation and IFN-γ production by CD4+ T cells from the dLNs of vaccinated mice (fig. S8). Balb/c mice immunized with rPEPCK also displayed significantly smaller lesion size and reduced parasite burden compared to CpG controls (Fig. 7, G and H). The protection induced by vaccination with PEPCK in C57BL/6 mice was long-lasting as evidenced by significantly reduced lesion size (Fig. 7I) and parasite burden (Fig. 7J) in mice that were challenged at 12 weeks after vaccination. In addition, vaccination with DNA expressing full-length PEPCK also induced cross-protection against visceral leishmaniasis caused by L. donovani in mice (Fig. 7, K and L). Collectively, these observations show that immunity induced by PEPCK is independent of mouse genetic background and parasite species.

Fig. 7. Vaccination with the synthetic peptide (PEPCK335–351), DNA expressing PEPCK, and recombinant PEPCK protein induces cross-species protection in mice.

(A to L) C57BL/6 (A to F, I, and J) or BALB/c (G, H, K, and L) mice were immunized subcutaneously in the footpads with 50 nmol PEPCK335–351 peptide (A and B), 100 μg of plasmid expressing full-length PEPCK DNA or empty vector plasmid (C, D, and I to L), or 20 μg of rPEPCK (E to H) three times at weekly intervals. Immunized C57BL/6 mice were challenged with 2 × 106 L. major or 5 × 107 L. donovani (K and L only) promastigotes in the contralateral footpads or intravenously (K and L) at 4 weeks (A to H, K, and L) and 12 weeks (I and J) after the last immunization, and lesion development and progression were monitored weekly (A, C, E, G, and I). Six weeks after challenge, mice were sacrificed, and parasite burden in the challenged footpads (B, D, F, H, and J), spleens (K), and livers (L) was determined by limiting dilution assay. Results are representative of two independent experiments with similar results (P values shown, two-tailed t tests, n = 4 to 5).

DISCUSSION

The major objective of this study was to identify protective Leishmania antigens that induce memory CD4+ T cell responses and characterize the repertoire of these cells at the clonal level. We identified naturally processed MHC class II–associated L. major peptide that induces strong recall T cell responses in healed mice. We further showed that the parent protein, PEPCK, is highly immunogenic in mice and humans and induces cross-species protection in vaccinated animals. We generated I-Ab–PEPCK335–351 tetramer and demonstrated at the clonal level that L. major–specific CD4+ T cells undergo the classical expansion, effector function, contraction, and stable maintenance phases of T cell activation and immunity. One caveat with this approach is that the peptides we identified are restricted to I-Ab and therefore would not be presented by other MHC class II alleles, thus minimizing the potential application in humans. However, the fact that the parent protein PEPCK induced strong immune responses in humans and dogs and vaccination with the recombinant protein or DNA expressing full-length protein elicited strong protection in immunized BALB/c mice against virulent L. major challenge suggest that PEPCK contains epitopes that are presented by diverse MHC alleles. Another caveat relates to the use of CpG as the adjuvant because it is unlikely that CpG will be approved as an adjuvant for human vaccines. Nevertheless, the results of the study provide strong proof-of-concept evidence for future studies using conventional adjuvants such as alum and MPL (monophosphoryl lipid A).

Although several Leishmania proteins have been shown to be immunogenic and induce T cell responses, only a few Leishmania-specific T cell epitopes have been mapped or predicted using in silico MHC-peptide binding algorithms (6, 1417). Most of these have focused on CD8+ T cells (6, 14, 17) with little or nothing shown for CD4+ T cells. This is because CD4+ T cell epitope prediction has a high failure rate because the peptide motifs that bind to MHC class II molecules are poorly defined (18) and most of the predicted epitopes may not be naturally processed by APCs. The genome sequence of L. major (19) provides a “blueprint” of possible gene products, making it possible to identify L. major proteins by mass spectrometry (20, 21). Mass spectrometry provides direct analysis and identification of potential processed and presented T cell epitopes through sampling of numerous MHC-associated peptides isolated from L. major–infected APCs. We identified L. major peptides from infected (but not uninfected) BMDCs. Most of these peptides contained nested peptide sets and vary in length between 11 and 29 residues, which is consistent with previous reports on the length of class II–associated peptides (10, 22, 23).

P3 (PEPCK335–351), derived from PEPCK, showed a strong ability to induce CD4+ T cells from healed mice to proliferate and produce IFN-γ and TNF, two key effector cytokines that mediate resistance in leishmaniasis. PEPCK is a widely distributed enzyme that plays an essential role in glycolysis by catalyzing the reversible decarboxylation and phosphorylation of oxaloacetate to yield phosphoenolpyruvate and carbon dioxide (24). On the basis of substrate specificity, two major subgroups of PEPCK are known: ATP (adenosine 5′-triphosphate)–dependent family, found in bacteria, yeast, higher plants, and trypanosomatids, and GTP (guanosine 5′-triphosphate)–dependent family, found in mollusks, insects, fungi, and vertebrates. Although some residues are conserved in all PEPCKs, no statistically significant homology is found between PEPCKs of the two subgroups (<20% identity) (24). The sequence identity between L. major and human/mouse PEPCKs is <17%, and no identity exists within the immunodominant epitope (PEPCK335–351) reported here. There is more than 90% amino acid homology among PEPCKs from pathogenic Leishmania species, and PEPCK335–351 is 100% conserved in all Leishmania (fig. S3), suggesting that targeting this protein could be a viable vaccination strategy against different forms of leishmaniasis. Indeed, we found strong cross-reactive anti-PEPCK B cell responses in humans and dogs infected with L. donovani and Leishmania infantum, respectively (fig. S9), and I-Ab–PEPCK335–351 tetramer reagent identified Leishmania-specific CD4+ T cells in mice infected with L. major, L. donovani, and L. mexicana. Furthermore, vaccination with DNA expressing full-length PEPCK significantly protected mice against L. donovani challenge. Recent studies have shown that Schistosoma and Mycobacterium tuberculosis PEPCK induce strong T cell response in mice (2527). PEPCK has also been associated with virulence in Mycobacterium bovis BCG (Bacille Calmette-Guérin) (26), is essential for the differentiation of Trypanosoma brucei from bloodstream to procyclic (insect) forms (28), and plays a crucial role in metabolic activities in kinetoplastids (24, 29). Collectively, these reports indicate that PEPCK might be a potential drug target or vaccine candidate.

Because of lack of reliable peptide–MHC class II tetramers, no study has convincingly characterized the expansion, effector function, contraction, and maintenance of Leishmania-specific CD4+ T cells after L. major infection at the clonal level. The immune response to Leishmania homolog for activated C kinase (LACK) has been studied, and LACK is thought to contribute to susceptibility in mice (3032). The immunodominant peptide from LACK has been determined using overlapping synthetic peptides and a panel of T cell hybridomas generated from BALB/c mice immunized with the recombinant protein (16). However, the long-term impact of LACK on the immune response in the resistant mice is unknown, and whether LACK induces T cell response in human patients has not been demonstrated. Recently, adoptive transfer of naïve ovalbumin (OVA) TCR transgenic CD4+ T cells followed by infection with OVA-expressing L. major (Lm-OVA) revealed early proliferation in the dLNs and generation of both central and effector memory cells (33). However, the transfer of large, nonphysiologic numbers of T cells with the same specificity can inhibit endogenous response and could substantially alter the kinetics, proliferative expansion, phenotype, and efficiency of memory T cell generation (34). Another study used Leishmania expressing 2W peptide, which is a variant of peptide 55–68 derived from the α chain of mouse I-Ed MHC class II molecule (35), to attempt to demonstrate expansion and contraction of CD4+ T cells after L. major infection. However, 2W peptide is a foreign antigen (alloantigen) to C57BL/6 mice, and its presentation leads to nonphysiologic and unusually high recognition by a large population of naïve CD4+ T cells (36). The frequency of CD4+ T cells specific for this peptide in infected mice was very low (below 0.1%) at the peak of immune response, and hence, 2W-specific cells have to be selectively enriched before any meaningful analysis could be performed (37). We identified a naturally processed and protective L. major peptide (PEPCK335–351), generated PEPCK335–351 tetramer, and tracked the expansion of PEPCK335–351–specific CD4+ T cells during infection. After infection, PEPCK-specific CD4+ T cells became activated (all were CD44hi) and underwent massive expansion (>500-fold) comprising ~6 and ~12% of all CD4+ T cells in the spleens and blood of infected mice (without enrichment) at the peak of response. PEPCK335–351–specific CD4+ T cells displayed strong effector function (polyfunctional cytokine production), which was followed by contraction and stable maintenance (>60 weeks) in blood, spleens, and dLNs, consistent with previous reports on T cell activation dynamics (32, 33, 37). These observations suggest that some PEPCK-specific CD4+ T cells differentiated into memory T cells. Indeed, some activated (CD44+) PEPCK-specific CD4+ T cells increased their expression of CD62L and down-regulated Ly6C (that is, changed from CD26LloLy6Chi to CD26LhiLy6Clo), which are markers associated with memory cell phenotype. In addition, when healed mice were rechallenged with L. major, PEPCK-specific CD4+ T cells underwent massive expansion (>20-fold over primary infection), and adoptive transfer of few numbers of these cells induced protection against L. major challenge.

We detected more PEPCK-specific CD4+ T cells in the spleens than in the dLN after resolution of infection, indicating that this tissue is the major reservoir of parasite-specific memory CD4+ T cells, which is consistent with previous observations (37). PEPCK-specific CD4+ T cells from lymphoid organs and infected footpads coproduced IFN-γ and TNF (key cytokines that are critical for resistance to cutaneous leishmaniasis) directly ex vivo and constitute ~17% of all Leishmania-reactive CD4+ T cells at the peak of immune response in L. major–infected mice. In contrast, PEPCK-specific CD4+ T cells did not produce IL-10 or express Foxp3 (data not shown), suggesting that they may not have regulatory properties. In line with this, adoptive transfer of a few thousand PEPCK-specific CD4+ T cells induced robust protection against virulent L. major challenge. As in mice, we also observed strong proliferative and recall cytokine responses (IFN-γ and granzyme B production) against PEPCK in PBMCs of patients who recovered from ZCL.

The repertoire of PEPCK-specific CD4+ T cells TCRs was highly restricted, with vβ8 and vβ6 genes constituting more than 90% of TCRβ usage. Similarly, 99% of tetramer-binding (PEPCK-specific) CD4+ T cells use TRAV6 TCRα gene. The analysis was reproducible, because we obtained similar results for TCRβ genes with antibody staining and single-cell PCR technique. A previous report that analyzed total TCR gene usage in L. major–infected C57BL/6 mice showed preferential expansion of vβ4 clones and a slight increase in vβ8 and vβ11 clones (38). Similar TCR usage was observed in the susceptible BALB/c mice, suggesting that the assay may not properly discriminate between protective and nonprotective T cell clones. In addition, the study assessed all T cells (CD4+ and CD8+) and did not focus on Leishmania-reactive/specific T cells as we did here. Thus, our study is the first to reproducibly analyze the clonality of TCR usage of protective Leishmania-specific T cells in infected mice.

In conclusion, we identified a highly conserved immunodominant antigen of L. major that induces robust immune response in animals and patients, and vaccination with this protein induces protection against different species of Leishmania. The durability of the protection in vaccinated mice is remarkable and clearly more superior than any vaccine evaluated in this mouse strain. Thus, our study provides critical insights into the immunologic mechanisms that regulate resistance to leishmaniasis and rekindles hope for the potential and real possibility of developing a cross-species protective vaccine against the disease. They also provide new perspective and approach to assessing and identifying novel vaccine candidates and information that would aid vaccine designs and vaccination strategies against leishmaniasis.

MATERIALS AND METHODS

Study design

The objective of this study was to identify naturally processed immunoprotective Leishmania peptides and fully characterize their responding CD4+ T cells at the clonal level. We used proteomics and cellular immunologic techniques to identify Leishmania-derived peptides presented by the MHC class II of infected DCs, particularly those that are capable of restimulating proliferation and IFN-γ production by CD4+ T cells from healed (immune) mice. After identifying the naturally processed L. major antigenic peptides, we used in silico computer modeling to predict those with the best binding affinity to MHC class II I-Ab. We generated I-Ab–PEPCK335–351 tetramer, used it to quantify PEPCK-specific CD4+ T cell response during L. major infection, and elucidated the TCR diversity, protective activity, and memory characteristics of PEPCK-specific cells. We cloned PEPCK, determined its expression in different species of Leishmania, and demonstrated PEPCK-induced immune responses in different hosts infected with different Leishmania spp. Vaccination with PEPCK induced strong protective immunity against L. major and L. donovani challenges in mice.

Isolation of I-Ab–associated peptides and liquid chromatography–tandem mass spectrometry analysis

L. major parasites (MHOM/80/Fredlin) were grown in M199 culture medium (Sigma) supplemented with 20% heat-inactivated fetal bovine serum. Peptide isolation was performed mainly as described previously (9, 10). Briefly, 3 × 109 uninfected or L. major–infected BMDCs were lysed, and MHC class II–peptide complexes were purified by affinity chromatography using anti–I-Ab antibodies (Y3-P) bound to protein A Sepharose beads (GE Healthcare Life Sciences), and bound class II peptides were eluted with 0.2 M acetic acid. Peptides were separated from MHC molecules by filtration through a 10-kD cutoff Ultrafree-MC filter (Millipore). Extracted peptides were lyophilized and stored at −80°C until used for mass spectrometry.

Lyophilized peptides were separated by nano-flow reversed-phase high-performance liquid chromatography. Online liquid chromatography–tandem mass spectrometry (LC-MS/MS) analyses were performed with QStar Pulsar I (Applied Biosystems/MDS-SCIEX) equipped with a Protana nanospray source. MS/MS spectra were acquired in a data-dependent mode. Automatically generated peak lists were submitted to the GPM (http://thegpm.org) and ProteinPilot software and searched against the mouse and L. major database. Peptide mass tolerance of 0.4 dalton and fragment ion mass tolerance of 100 ppm (parts per million) were used to restrict the searches for accurate identifications. The protein identifications are presented as ENSEMBL entries. All proteins reported have an expectation value of log (e) −3 or less.

Generation of L. major peptide–specific CD4+ T cell lines

Purified CD4+ T cells from healed C57BL/6 mice were cultured in 96-well round-bottom plates (2 × 105 per well) and stimulated with synthetic peptides (20 μM) in the presence of irradiated syngeneic naïve spleen cells as APCs. The cells were repetitively restimulated with peptide, fresh APCs, and rIL-2 (10 U/ml) at 2-week intervals. For some experiments, CD4+ T cells were labeled with CFSE and stimulated with synthetic peptides, and cell proliferation and IFN-γ production were determined by flow cytometry. To assess specific peptide recognition, the cells were rested for 2 weeks and then restimulated overnight with fresh APCs in the presence of P3 or other peptides.

Cellular immune responses against PEPCK in humans

Peripheral blood samples were from 28 healed individuals from zoonotic human cutaneous leishmaniasis (ZCL) caused by L. major who live in Sidi Bouzid, an endemic region located in the center of Tunisia, and 7 healthy donors were from a non-endemic ZCL region in Tunis. The study was approved by the ethics committee of Pasteur Institute, Tunis. All donors provided written informed consent for the collection of samples and subsequent analysis. PBMCs were cultured in 96-well plates in the presence or absence of rPEPCK (20 μg/ml) or SLA (10 μg/ml, positive control) for 5 days. For proliferation studies, the cells were pulsed with [3H]thymidine (2 μCi per well) in the last 6 hours and harvested, and radioactivity was counted in a scintillation counter (MicroBeta2 LumiJET, PerkinElmer). Results were expressed as a proliferation index: mean counts of triplicates in antigen-stimulated cultures/mean counts of triplicates in unstimulated cultures. IFN-γ and granzyme B levels in cell culture supernatants were determined by ELISA.

Generation of I-Ab–PEPCK335–351 tetramer and staining protocol

I-Ab–PEPCK335–351 tetramer and the corresponding negative control tetramer (I-Ab–CLIP87–101) were generated at the NIH Tetramer Core Facility, Emory University, Atlanta, GA. Spleen and popliteal lymph node cells or PBMCs collected at different times from infected mice were stained with tetramer I-Ab–PEPCK335–351 (2 μg/ml) or I-Ab–CLIP87–101 control for 30 min at 37°C. The cells were then stained routinely for surface markers or intracellular cytokines. The frequency of PEPCK335–351–specific CD4+ T cells at the site of infection was determined as previously described (39).

Single-cell RT-PCR and determination of the antigen specificity of cloned TCRs

Single-cell sorting and RT-PCR were performed mainly as described previously (40) and modified as described in Supplementary Materials and Methods. Briefly, tetramer-binding cells were single-cell–sorted into 96-well PCR plate by using a FACSAria II cell sorter (BD Biosciences) and were stored at −80°C until RT-PCR. The PCR products of TCRα and TCRβ were then analyzed by direct sequencing using the mAC-3 and mBC-3 primers for TCRα and TCRβ, respectively. The TCR repertoire was analyzed with the IMGT/V-QUEST tool (http://imgt.org/).

Construction of the retroviral vector for expressing TCRα and TCRβ proteins was described in Supplementary Materials and Methods. The constructed plasmid vector, pMXs-TCRβ-P2A-TCRα-IRES-GFP, was then transfected into PLAT-E cells. The culture supernatant was collected 72 hours after the transfection. Murine CD4+ TG40 or primary CD4+ T cells were infected with the recombinant retroviruses using retronectin (TaKaRa Bio) according to the manufacturer’s instruction. The antigen specificity of the cloned TCRαβ pairs was analyzed using tetramer staining. Transduced primary CD4+ T cells were stimulated with splenic DCs plus 20 μM PEPCK335–351 for 3 days, EdU was added 18 hours before assay, and cell proliferation and IFN-γ production were assessed by flow cytometry.

Assessing the protective ability of PEPCK-specific CD4+ T cells by adoptive transfer

Infected C57BL/6 mice were sacrificed in 5 weeks after infection, and PEPCK-specific (Tet+) and nonspecific (Tet-) CD4+ T cells were purified from spleens by cell sorting (FACSAria III, BD Biosciences). The indicated numbers of PEPCK-specific or non–PEPCK-specific CD4+ T cells were adoptively transferred into naïve C57BL/6 mice intravenously. The mice were then challenged with 2 × 106 L. major the next day and sacrificed after 4 weeks to assess parasite burden.

Immunization and challenge

Mice were immunized in the footpads three times at weekly intervals with either PEPCK335–351 peptides (50 nmol), rPEPCK (20 μg), or plasmid DNA (100 μg) containing full-length L. major PEPCK gene with or without CpG (50 μg). Immunized mice were challenged with 2 × 106 L. major (footpad) or 5 × 107 L. donovani (intravenously) at 4 or 12 weeks after the last immunization. Lesion size was monitored weekly by measuring footpad swelling with calipers. At 4 (L. donovani) or 6 (L. major) weeks after challenge, mice were sacrificed to assess immune response and parasite burden by limiting dilution (41).

Statistical analysis

Two-tailed Student’s t test was used to compare means and SEM between groups using GraphPad Prism software. Differences were considered significant at P < 0.05. Data are presented as means ± SEM.

SUPPLEMENTARY MATERIALS

www.sciencetranslationalmedicine.org/cgi/content/full/7/310/310ra167/DC1

Materials and Methods

Fig. S1. Bone marrow cells from C57BL/6 mice were differentiated to dendritic cells (BMDCs) with GM-CSF in vitro and infected with L. major promastigotes.

Fig. S2. Computer models showing binding of P1 (upper panel) and P3 (PEPCK335–351, lower panel) to MHC class II (I-Ab) molecule and their respective binding energies.

Fig. S3. Multiple amino acid sequence alignment of PEPCK from various Leishmania genus.

Fig. S4. Comparison of amino acid sequences of Leishmania-specific TCRα chain.

Fig. S5. Direct ex vivo cytokine profile of PEPCK335–351–specific CD4+ T cells from the spleens of L. major–infected mice.

Fig. S6. Expression of memory markers on non–PEPCK335–351–specific CD4+ T cells from L. major–infected mice.

Fig. S7. Prechallenge immune response in the draining LNs and spleens in mice immunized with PEPCK335–351 (P3) peptide.

Fig. S8. Postchallenge immune response in dLNs of mice immunized with the PEPCK335–351 (P3) peptide.

Fig. S9. Antibody responses against PEPCK in different hosts infected with Leishmania.

Table S1. Amino acid sequences and the corresponding source proteins of some key peptides identified by both GPM and ProteinPilot software.

Table S2. DNA sequence of primers and artificial DNAs.

Source data

References (4244)

REFERENCES AND NOTES

  1. Acknowledgments: We acknowledge the NIH Tetramer Core Facility (contract HHSN272201300006C) for provision of I-Ab–PEPCK335–351 tetramers. We thank Y. Cao (Third Medical Military University, Chongqing, China) for Western blot analysis, S. Hirota (Department of Immunology, University of Toyama, Toyama, Japan) for performing plasmid DNA and single-cell TCR gene sequencing, and T. Watts (The University of Toronto) for critically reading the manuscript and making insightful comments and suggestions. T. Kitamura (University of Tokyo) provided PLAT-E cell line, and T. Saito (Riken) provided TG40 cell line. Funding: This study was funded by the Canadian Institutes for Health Research (MOP 114923) and Research Manitoba (to J.E.U.) and the National Natural Science Foundation of China (no. 30872466 to Z.M.). Z.M. was also supported by the Manitoba Institute of Child Health postdoctoral fellowship. Author contributions: Z.M. and J.E.U. designed the experiments. Z.M. performed the experiments unless otherwise specified. J.L. cloned the PEPCK gene and constructed PEPCK expressing DNA vaccine vector. T.B. and H.L. performed cellular responses of human patient PBMCs stimulated with rPEPCK. H.K., H.H., A.M., and K.S. designed primers, performed single-cell PCR experiments, and analyzed PCR data. P.E. and J.A.W. performed MS/MS analyses. C.H. and W.Y. made recombinant PEPCK and polyclonal antibody against rPEPCK. D.L. performed peptide vaccination experiments. F.K. measured antibody level in sera of L. major–infected patients. I.O. optimized tetramer staining. P.J. measured antibody level in sera of L. infantum–infected dogs. S.W. did computer modeling of peptides binding to MHC class II molecule. C.P. provided sera of L. infantum–infected dogs. S.R. provided sera from L. major–infected patients. M.N. and J.C. provided sera from L. donovani–infected patients. Z.M., H.K., J.A.W., and J.E.U. wrote, revised, and edited the manuscript. Competing interests: The University of Manitoba holds the rights to a patent (USSN61/937795; Identification of Leishmania major antigenic peptides and protein that elicit protective T cell response: Implications for vaccination against Leishmaniasis) in which J.U. and Z.M. are the co-inventors. The other authors declare that they have no competing interests.
View Abstract

Navigate This Article