HIV-1 T cell epitopes targeted to Rhesus macaque CD40 and DCIR: A comparative study of prototype dendritic cell targeting therapeutic vaccine candidates

Abstract

HIV-1 infection can be controlled by anti-retroviral drug therapy, but this is a lifetime treatment and the virus remains latent and rapidly rebounds if therapy is stopped. HIV-1-infected individuals under this drug regimen have increased rates of cancers, cardiovascular diseases, and autoimmunity due to compromised immunity. A therapeutic vaccine boosting cellular immunity against HIV-1 is therefore desirable and, possibly combined with other immune modulating agents, could obviate the need for long-term drug therapies. An approach to elicit strong T cell-based immunity is to direct virus protein antigens specifically to dendritic cells (DCs), which are the key cell type for controlling immune responses. For eliciting therapeutic cellular immunity in HIV-1-infected individuals, we developed vaccines comprised of five T cell epitope-rich regions of HIV-1 Gag, Nef, and Pol (HIV5pep) fused to monoclonal antibodies that bind either, the antigen presenting cell activating receptor CD40, or the endocytic dendritic cell immunoreceptor DCIR. The study aimed to demonstrate vaccine safety, establish efficacy for broad T cell responses in both primed and naïve settings, and identify one candidate vaccine for human therapeutic development. The vaccines were administered to Rhesus macaques by intradermal injection with poly-ICLC adjuvant. The animals were either i) naïve or, ii) previously primed with modified vaccinia Ankara vector (MVA) encoding HIV-1 Gag, Pol, and Nef (MVA GagPolNef). In the MVA-primed groups, both DC-targeting vaccinations boosted HIV5pep-specific blood CD4+ T cells producing multiple cytokines, but did not affect the MVA-elicited CD8+ T cell responses. In the naive groups, both DC-targeting vaccines elicited antigen-specific polyfunctional CD4+ and CD8+ T cell responses to multiple epitopes and these responses were unchanged by a subsequent MVA GagPolNef boost. In both settings, the T cell responses elicited via the CD40-targeting vaccine were more robust and were detectable in all the animals, favoring further development of the CD40-targeting vaccine for therapeutic vaccination of HIV-1-infected individuals.

Fig 1. Overview of structure and biochemical analysis of αDCIR.HIV5pep and αCD40.HIV5pep vaccines.

(A) Cartoon of the five HIV-1 peptide regions fused to the C-termini of either the heavy chain C_H3 or light chain C_L regions of the targeting antibody and short flexible glycosylated peptide linkers are indicated as squiggles. V_L and V_H indicate, respectively, the L and H chain variable regions. (B) Reducing SDS PAGE analysis of αCD40.HIV5pep (Lanes 2, 3) and αDCIR.HIV5pep (Lanes 4, 5) vaccines. The dark upper arrow on the right indicates the position of the full heavy chains and the dark lower arrow indicates the position of light chains. The lighter grey arrows in between indicate minor breakdown or glycosylation variant products of the heavy chain. Molecular weight markers are shown in Lane 1 with mass (kDa) indicated beside each protein. Staining was with Coomassie Brilliant Blue.

Fig 2. T cell responses to αDCIR.HIV5pep and αCD40.HIV5pep vaccines in MVA-primed Rhesus macaques.

PBMCs were harvested at the indicated time points and analyzed by IFNγ ELISPOT using pools of overlapping Gag Pol and Nef peptides corresponding to sequences carried by the DC-targeting vector (HIV5 peptides, shown in light grey), or pools of overlapping Gag, Pol and Nef peptides corresponding to sequences not carried by the DC-targeting vaccines but carried by MVA GagPolNef (non-HIV5 pep, shown in dark grey). (A) IFNγ ELISPOT data from animals primed with two administrations of MVA GagPolNef, then vaccinated with 250 μg αDCIR.HIV5pep (G1) or (B) αCD40.HIV5pep (G2) plus 1 mg poly-ICLC according to the schedule shown in Table 1 and as indicated by the gray and black arrows below the timeline shown in weeks. (C) and (D) IFNγ ELISPOT data for individual Gag, Pol and Nef peptide stimulations corresponding to sequences carried by the DC-targeting vectors (individual HIV5pep peptides are indicated by colors). Responses for each peptide are the average response per sample point. The X-axis shows sampling time in weeks. S1 Table and S2 Table show data corresponding to this figure. Supporting information S1 File contains the primary data.

Fig 3. T cell responses to αDCIR.HIV5pep and αCD40.HIV5pep vaccines in naïve Rhesus macaques.

PBMCs were harvested at the indicated time points and analyzed by IFNγ ELISPOT using pools of overlapping Gag Pol and Nef peptides corresponding to sequences carried by the DC-targeting vector (HIV5 peptides, shown in light grey), or pools of overlapping Gag Pol and Nef peptides corresponding to sequences not carried by the DC-targeting vaccines but carried by MVA GagPolNef (non-HIV5 pep, shown in dark grey). (A) IFNγ ELISPOT data from animals vaccinated with three administrations of 250 μg αDCIR.HIV5pep (G3) or (B) αCD40.HIV5pep (G4) plus 1 mg poly-ICLC followed by a boost with MVA GagPolNef. Responses from individual animals in the indicated groups are shown as points and were summed for these two pool sets. The mid-line of the box denotes the median, and the ends of the box denote the 25^th and 75^th percentiles. The whiskers are the minimum/maximum value higher/lower than 1.5* Inter-Quartile Interval. Values after the second and third αDCIR.HIV5pep vaccinations at week 0 compared to weeks 6 or 14 were not significantly different (G3; ns; respectively, p = 0.03, p = 0.1). Values after the second and third αCD40.HIV5pep vaccinations at week 0 compared to weeks 6 or 14 were not significantly different (G4; *; respectively, p = 0.15, p = 0.03). (C) and (D) IFNγ ELISPOT data for individual Gag, Pol and Nef peptides stimulations as described above. The X-axis shows sampling time in weeks. S6 Table and S3 Table and S1 File show data corresponding to this figure.

Fig 4. Analysis of HIV-1 epitope-specific CD4⁺ and CD8⁺ T cell responses elicited in MVA-primed NHPs by αDCIR.HIV5pep and αCD40.HIV5pep vaccines.

PBMCs were collected from individual animals at week 10 (two weeks post MVA) and at peak response times of week 26 (2 weeks post DC-targeting vaccination) for G1 MVA αDCIR and G2 MVA αCD40. PBMC were stimulated in the presence of Brefeldin A for 6 h with pools of HIV-1 peptides corresponding to the four indicated gene regions and analyzed by flow cytometry. Each dot is the background-subtracted value for individual animals of (A) CD154⁺ CD4⁺ or (B) CD8⁺ T cells secreting IFNγ, TNFα, IL-2, or combinations thereof when stimulated with Gag p17, Gag p24, Nef and Pol peptides. Negative background subtracted values were set to zero. Responses from individual animals in the indicated groups are presented. The mid-line of the box denotes the median, and the ends of the box denote the 25^th and 75^th percentiles. The whiskers are the minimum/maximum value higher/lower than 1.5* Inter-Quartile Interval. One animal had a value (0.9%) outside the plotted scale for the CD4⁺ T cell response to Pol peptides. S4 Table shows the data corresponding to this figure.

Fig 5. Analysis of HIV-1 epitope-specific CD4⁺ and CD8⁺ T cell responses elicited in naïve NHPs by αDCIR.HIV5pep and αCD40.HIV5pep vaccines.

Intracellular cytokine staining analysis of HIV-1 antigen-specific CD154⁺ CD4⁺ and CD8⁺ T cells elicited by αDCIR.HIV5pep and αCD40.HIV5pep vaccines. PBMCs were collected from individual animals at week 14 (two weeks post DC-targeting vaccines) and week 24 (two weeks post MVA boost) for G3 αDCIR MVA and G4 αCD40 MVA. PBMC were stimulated in the presence of Brefeldin A for 6 h with pools of HIV-1 peptides corresponding to the indicated gene regions and analyzed by flow cytometry. Each dot is the background-subtracted value for individual animals of (A) CD154⁺ CD4⁺ and (B) CD8⁺ T cells secreting IFNγ, TNFα, IL-2, or combinations thereof when stimulated with Gag p17, Gag p24, Nef and Pol peptides. Negative background subtracted values were set to zero. Responses from individual animals in the indicated groups are presented. The mid-line of the box denotes the median, and the ends of the box denote the 25^th and 75^th percentiles. The whiskers are the minimum/maximum value higher/lower than 1.5* Inter-Quartile Interval. S4 Table shows the data corresponding to this figure.

Fig 6. Serum HIV5pep-specific IgG responses elicited by αDCIR.HIV5pep or αCD40.HIV5pep vaccines.

(A) Animals primed with two administrations of MVA GagPolNef and vaccinated with αDCIR.HIV5pep (G1) or αCD40.HIV5pep (G2) plus poly-ICLC according to the schedule shown in Table 1 and as indicated by the gray and black arrows below the timeline shown in weeks. Serum HIV5pep-specific IgG levels detected by ELISA using a mixture of fusion proteins for each of the five HIV-1 peptide regions are shown for individual animals. Bars indicate the median for the group at each sample time. 1/EC₅₀ values lower than 10 are not shown, but were set to a baseline of 10 for graphical simplicity. There was no significant difference in overall magnitude in weeks 14, 18, and 26 between antibody responses evoked by αDCIR.HIV5pep versus αCD40.HIV5pep (ns; p = 0.36). (B) Similar analysis for NHPs vaccinated with αDCIR.HIV5pep (G3) or αCD40.HIV5pep (G4) plus poly-ICLC according to the schedule shown in Table 1 and as indicated by the arrows below the timeline shown in weeks. They were boosted at week 22 with MVA GagPolNef. S5 Table shows the data corresponding to the figure.