Long-term follow up of human T-cell responses to conserved HIV-1 regions elicited by DNA/simian adenovirus/MVA vaccine regimens

Abstract

Background: Durability of vaccine-elicited immune responses is one of the key determinants for vaccine success. Our aim is to develop a vaccination strategy against the human immunodeficiency virus type 1 (HIV-1), which induces protective and durable CD8+ T-cell responses. The central theorem of our approach is to focus T cells on highly conserved regions of the HIV-1 proteome and this is achieved through the use of the first-generation conserved vaccine immunogen HIVconsv. This immunogen vectored by plasmid DNA, simian adenovirus and poxvirus MVA was tested in healthy, HIV-1-negative adults in UK and induced high magnitudes of HIVconsv-specific plurifunctional CD8+ T cells capable of in vitro HIV-1 inhibition. Here, we assessed the durability of these responses.

Methods: Vaccine recipients in trial HIV-CORE 002 were invited to provide a blood sample at 1 and 2 years after vaccination. Their PBMCs were tested in IFN-γ ELISPOT, 25-analyte Luminex, CFSE proliferation and intracellular cytokine staining assays, the last enhanced by HLA-peptide dextramer analysis.

Results: 12/12 (1 year) and 8/8 (2 years) returning subjects had median (range) of 990 (150-2495) and 763 (70-1745) IFN-γ SFU/106 PBMC specific for HIVconsv, respectively, and recognized 5 (1-6) out of 6 peptide pools at 2 years. Over one-half of the HIVconsv-specific cells expressed at least 3 functions IFN-γ, TNF-α and CD107a, and were capable of proliferation. Among dextramer-reactive cells, naïve, transitional, effector and terminally differentiated memory subsets were similarly represented.

Conclusions: First generation HIVconsv vaccine induced human T cells, which were plurifunctional and persisted for at least 2 years.

Trial registration: ClinicalTrials.gov NCT01151319.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Fig 1. HIVconsv vaccine-induced human T-cell responses.

(A) Schematic representations of the HIVconsv immunogen and six pools of a total of 199 overlapping peptides used for the IFN-γ ELISPOT assay. HIVconsv is a chimeric protein assembled from 14 highly conserved regions of the HIV-1 proteome, the HIV-1 protein origins of which are colour-coded below. Each of the regions uses a consensus amino acid sequence of the HIV-1 clade indicated above the schematics. The C-terminal epitopes include Mamu-A*01- and H-2d-restricted immunodominant CTL epitopes and a Pk tag recognized by a monoclonal antibody, which together facilitate the quality control of the vaccines. (B) Fresh ex vivo net total (sum of six pools) IFN-γ ELISPOT assay frequencies of HIVconsv-specific T cells of returning healthy HIV-1-negative volunteers of trial HIV-CORE 002, who received either the CM (top) and DDDCM (bottom) vaccine regimen, are shown separately. Time points ‘S’ (screen) and 0–28 weeks were previously published [68] and are shown for completeness. ‘1y’ and ‘2y’ indicate 1 and 2 years after the last M (MVA.HIVconsv) vaccine administration at weeks 8 (CM) and 20 (DDDCM) indicated below the graphs. Volunteers’ ID numbers are shown on the graph legend. (C) Fresh ex vivo net total IFN-γ ELISPOT assay frequencies of HIVconsv-specific T cells after 1 and 2 years. The horizontal bars represent median frequencies for each regimen separately. The two time points and regimens were not statistically separable.

Fig 2. Long-term broad functional capacity of HIVconsv vaccine-elicited human T cells.

Cryopreserved and thawed PBMC samples from pre-vaccination (Pre), and 1 (1y) and 2 (2y) years after the last vaccine administration were re-stimulated with pools of stimulatory 15-mer peptides assembled for each returning volunteer individually for 48 hours and the tissue culture supernatants were analyzed in a 25-analyte Luminex assay. Individual values are shown with indicated median and boxed interquartile range. Volunteer 416 is depicted in blue. Only cytokines and chemokines with positive responses are shown. Results were analyzed using the Mann-Whitney U test for comparison between the long-term samples vs the pre-vaccination sample. Significant P values are indicated by asterisks, whereby: *—less that 0.05; **—less than 0.01; ***—less than 0.001; ****—less than 0.0001.

Fig 3. Proliferative capacity of long-term vaccine-elicited human T cells.

Cryopreserved and thawed PBMCs from pre-vaccination (Pre), 6 months (6m), and 1 (1y) and 2 (2y) years after the last vaccine administration were labelled with CFSE, stimulated with pools of stimulatory 15-mer peptides assembled for each returning volunteer for 5 days and analyzed using flow cytometry for dividing (decreased CFSE) CD4+ and CD8+ T cells. The CM and DDDCM vaccination regimens are indicated above. The gating strategy is given in S1 Fig. Individual values are shown with median as a horizontal line.

Fig 4. Functionality and memory subtypes of vaccine-elicited human CD4+ and CD8+ T cells.

Frozen and thawed PBMCs from 6 months (6m), and 1 (1y) and 2 (2y) years after the last vaccine administration were stimulated with personalized 15-mer peptide pools and subjected to ICS assay. The pie charts refer to the plurifunctionality of the responses defined by Boolean gating. The results from six individuals are shown. For the best responder 416, memory phenotypes were investigated by flow cytometry using IFN-⅟ release to identify antigen-specific cells. Memory subsets are defined as TN−naïve T cells (CD45RAhiCCR7hiCD27hi). TCM—central memory (CD45RAloCCR7hiCD27hi), TTM—transitional memory (CD45RAloCCR7loCD27hi), TEM—effector memory (CD45RAloCCR7loCD27lo), and TTD—terminally differentiated (CD45RAhiCCR7loCD27lo).

Fig 5. Dextramer-aided analysis of human YV9- and KV10-specific CD8+ T cells.

For HLA-A*02:01-positive vaccine recipients, frozen and thawed PBMCs from 6 months (6m), and 1 (1y) and 2 (2y) years after vaccination were incubated with HLA-A*02:01-YV9 (A and C) or HLA-A*02:01-KV10 (B and D) dextramers together with other cell-surface markers and analyzed using flow cytometry (S3 Fig). Alternatively, PBMCs of subject 416 were stimulated with personalized 15-mer peptide pools and reacted with a panel of functional mAbs together with HLA-A*02:01-YV9 (E and F) or HLA-A*02:01-KV10 (H and I) dextramers. The pie charts (F and I) show the plurifunctionality of dextramer positive CD8+ T cells. PBMCs from subject 416 reactive with the HLA-A*02:01-YV9 (G) or HLA-A*02:01-KV10 (J) dextramers were phenotyped for memory subsets defined as TN−naïve T cells (CD45RAhiCCR7hiCD27hi). TCM—central memory (CD45RAloCCR7hiCD27hi), TTM—transitional memory (CD45RAloCCR7loCD27hi), TEM—effector memory (CD45RAloCCR7loCD27lo), and TTD—terminally differentiated (CD45RAhiCCR7loCD27lo) (S3 Fig).