Engineering a vector-based pan-Leishmania vaccine for humans: proof of principle

Abstract

Leishmaniasis is a spectrum of diseases transmitted by sand fly vectors that deposit Leishmania spp. parasites in the host skin during blood feeding. Currently, available treatment options are limited, associated with high toxicity and emerging resistance. Even though a vaccine for human leishmaniasis is considered an achievable goal, to date we still do not have one available, a consequence (amongst other factors) of a lack of pre-clinical to clinical translatability. Pre-exposure to uninfected sand fly bites or immunization with defined sand fly salivary proteins was shown to negatively impact infection. Still, cross-protection reports are rare and dependent on the phylogenetic proximity of the sand fly species, meaning that the applicability of a sand fly saliva-based vaccine will be limited to a defined geography, one parasite species and one form of leishmaniasis. As a proof of principle of a future vector saliva-based pan-Leishmania vaccine, we engineered through a reverse vaccinology approach that maximizes translation to humans, a fusion protein consisting of immunogenic portions of PdSP15 and LJL143, sand fly salivary proteins demonstrated as potential vaccine candidates against cutaneous and visceral leishmaniasis, respectively. The in silico analysis was validated ex vivo, through T cell proliferation experiments, proving that the fusion protein (administered as a DNA vaccine) maintained the immunogenicity of both PdSP15 and LJL143. Additionally, while no significant effect was detected in the context of L. major transmission by P. duboscqi, this DNA vaccine was defined as partially protective, in the context of L. major transmission by L. longipalpis sand flies. Importantly, a high IFNγ response alone was not enough to confer protection, that mainly correlated with low T cell mediated Leishmania-specific IL-4 and IL-10 responses, and consequently with high pro/anti-inflammatory cytokine ratios. Overall our immunogenicity data suggests that to design a potentially safe vector-based pan-Leishmania vaccine, without geographic restrictions and against all forms of leishmaniasis is an achievable goal. This is why we propose our approach as a proof-of principle, perhaps not only applicable to the anti-Leishmania vector-based vaccines' field, but also to other branches of knowledge that require the design of multi-epitope T cell vaccines with a higher potential for translation.

Figure 1

PdSP15 T cell epitope mapping: overall picture of the in silico analysis performed against murine and human MHC-I and MHC-II molecules. PdSP15 (GenBank Acc. No. ABI15933) T cell epitope mapping was performed using the IEDB Analysis Resource considering 3 murine and 27 human MHC-II alleles and 5 murine and 27 human MHC-I alleles. The translated results of the predictions of murine MHC-II restricted epitopes (A,A′), murine MHC-I restricted epitopes (B,B′), human MHC-II restricted epitopes (C,C′) and human MHC-I restricted epitopes (D,D′) are represented. Results are shown by allele. Each arrow represents one or more (contiguous) predicted epitopes: the top 1% hits for MHC-I molecules and the top 2.5% (very high affinity binders) or 10% hits for MHC-II human and murine alleles, respectively. The underlined protein residues represent the signal peptide sequence. The magnified protein residues are potentially important for protein biological activity. Dashed boxes represent the convergent analysis of the different in silico determinations that enabled the selection of two protein portions to be included in the final chimeric sand fly salivary antigen.

Figure 2

LJL-143T cell epitope mapping: overall picture of the in silico analysis performed against murine and human MHC-I and MHC-II molecules. LJL-143 (GenBank acc. No. AAS05319) T cell epitope mapping was performed using the IEDB Analysis Resource considering 3 murine and 27 human MHC-II alleles and 5 murine and 27 human MHC-I alleles. The translated results of the predictions of murine MHC-II restricted epitopes (A,A′), murine MHC-I restricted epitopes (B,B′), human MHC-II restricted epitopes (C,C′) and human MHC-I restricted epitopes (D,D′) are represented. Results are presented by allele. Each arrow represents one or more (contiguous) predicted epitopes: the top 1% hits for MHC-I molecules and the top 2.5% (very high affinity binders) or 10% hits for MHC-II human and murine alleles, respectively. The underlined protein residues represent the signal peptide sequence. The magnified protein residues are potentially important for protein biological activity. Dashed boxes represent the convergent analysis of the different in silico determinations that enabled the selection of two protein portions to be included in the final chimeric sandfly salivary antigen.

Figure 3

Engineering a multi-sand fly species based anti-Leishmania vaccine. A single DNA-based vaccine was engineered, considering a reverse vaccinology approach applied to the two salivary proteins this study focuses on: PdSP15 and LJL-143. The scheme of the final DNA vaccine, consisting of the VR2001-TOPO vector backbone and the fusion salivary antigen coding sequence, composed by the selected immunogenic portions of the two salivary proteins joined contiguously in an interpolated manner, a FLAG-tag, a HIS-tag and a stop codon is represented (A). The final sequence coding for the fusion salivary antigen (codon optimized for the mammalian codon usage bias), used in the cloning approach, as well as the respective translation, is shown (B). The underlined nucleotides code for the portions of PdSP15, while the non-underlined ones code for the portions of LJL-143. FLAG- and HIS-tags coding sequences are highlighted. The fusion protein amino-acid sequence was subjected to in silico predictions of antigenicity, solubility upon overexpression, presence of transmembrane domains, disulphide bonds formation and allergenic potential, whose quantitative and/or qualitative results are represented (C).

Figure 4

Vaccination with the sand fly-derived DNA chimeric vaccine elicits specific T cell responses against both P. duboscqi and L. longipalpis saliva. BALB/C mice were immunized intradermally in the right ear three times at 2 weeks intervals with 5 µg of VR-2001-SfSPChimera plasmid (crossed circles), or with either L. longipalpis (white circles) or P. duboscqi (crosses) SGH—the equivalent of 1 sand fly salivary gland pairs. Control animals received the same volume of the vehicle solution (PBS; black circles). One month after the last immunization, animals were euthanized, their spleens collected and processed to obtain CFSE-stained splenocytes suspensions. Frequencies of proliferating splenic T cells (total CD3 + and CD3 + /CD4 + or CD3 + /CD8 +) were determined by flow cytometry after 4 days of culture in the presence of BMDCs (5:1 ratio) pulsed with P. duboscqi (A) or L. longipalpis (B) SGH (final concentration of 3 sand fly salivary gland pair/ml). Results from three independent experiments are shown. Each dot represents one animal. Average and SD of the values within each group are shown. Statistical differences are properly identified (One-Way ANOVA with Tukey’s post hoc analysis: *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001 and ****p ≤ 0.0001).

Figure 5

Sand fly-derived DNA chimeric vaccine preliminary efficacy trial. C57BL/6 or BALB/C mice were immunized intradermally in the right ear three times at 2 weeks intervals with 5 µg of VR-2001-SfSPChimera plasmid (crossed circles), or with the empty vector (control animals; black circles). C57BL/6 mice were challenged 1 month later on the contralateral ear via sand fly bites—10 P. duboscqi sand flies carrying mature L. major infections were used per animal. Ear pathology was assessed weekly, and experiments were terminated at either 2 or 6 weeks post challenge for parasite burden evaluation (A). Ear thickness measurements are represented for both VR-2001-SfSPChimera immunized and control animals (B). Total ear parasite burdens at 2 and 6 weeks post-challenge are also shown (C). BALB/c mice were also challenged 1 month after the last immunization on the contralateral ear via sand fly bites, using a different species—10 L. longipalpis sand flies carrying mature L. major infections were used per animal (D). The follow up and endpoints used were the same as above-mentioned (D). Ear thickness measurements are once more represented for both VR-2001-SfSPChimera immunized and control animals (E), as are the determined parasite burdens at 2 and 6 weeks post-challenge (F). Results from at least two independent experiments are represented in each graph. Each dot represents one animal. Average and SEM or SD of the values within each group are shown for ear thickness or parasite burden measurements, respectively. Each group represented in the 2 weeks’ time-point of panel F was subdivided in two groups based on parasite burden values: below and above average. Statistical differences are properly identified (Mann–Whitney test: *p ≤ 0.05).

Figure 6

The protection against L. longipalpis sand fly-transmitted L. major infection in BALB/c mice is dependent on high IFN-γ/IL-10 and IFN-γ/IL-4 ratios. BALB/C mice were immunized intradermally in the right ear three times at 2 weeks intervals with 5 µg of VR-2001-SfSPChimera plasmid (crossed circles), or with the empty vector (control animals; black circles). One month after the last immunization mice were challenged on the contralateral ear via sand fly bites—10 L. longipalpis sand flies carrying mature L. major infections were used per animal. Two weeks later animals were euthanized and their ears collected and processed. Parasite burdens were determined by limiting dilution (Fig. 5F) and T cell cytokine secretion was analysed by flow cytometry after an overnight culture of total ear cells in the presence of L. major SLA, and additionally PMA, Ionomycin and Brefeldin A during the last 4 h of culture. For each animal, including controls, parasite burden values were plotted against the frequencies of CD4 + T cells secreting IFN-γ, IL-10 or IL-4, in a way to assess their correlation (A). These frequencies are represented for VR-2001-SfSPChimera immunized and control animals, sub-grouped (Fig. 5F) according to parasite burden values (B). IFN-γ/IL-10 and IFN-γ/IL-4 ratios were calculated and are represented using the same group division criteria (C). The correlation between these ratio values and the parasite burdens was also assessed, now only considering the animals immunized with VR-2001-SfSPChimera DNA vaccine (C). Results from two independent experiments are represented. Each dot represents one animal. Average and SD of the values within each group are shown. Statistical differences are properly identified (Mann–Whitney test: *p ≤ 0.05 and **p ≤ 0.01) and refer to differences between sub-groups (black bars), or differences between VR-2001-SfSPChimera vaccinated and control groups (grey bars). Pearson correlation coefficients (r) and significance are also represented.