Generation and characterization of a defective HIV-1 Virus as an immunogen for a therapeutic vaccine

Abstract

Background: The generation of new immunogens able to elicit strong specific immune responses remains a major challenge in the attempts to obtain a prophylactic or therapeutic vaccine against HIV/AIDS. We designed and constructed a defective recombinant virus based on the HIV-1 genome generating infective but non-replicative virions able to elicit broad and strong cellular immune responses in HIV-1 seropositive individuals.

Results: Viral particles were generated through transient transfection in producer cells (293-T) of a full length HIV-1 DNA carrying a deletion of 892 base pairs (bp) in the pol gene encompassing the sequence that codes for the reverse transcriptase (NL4-3/ΔRT clone). The viral particles generated were able to enter target cells, but due to the absence of reverse transcriptase no replication was detected. The immunogenic capacity of these particles was assessed by ELISPOT to determine γ-interferon production in a cohort of 69 chronic asymptomatic HIV-1 seropositive individuals. Surprisingly, defective particles produced from NL4-3/ΔRT triggered stronger cellular responses than wild-type HIV-1 viruses inactivated with Aldrithiol-2 (AT-2) and in a larger proportion of individuals (55% versus 23% seropositive individuals tested). Electron microscopy showed that NL4-3/ΔRT virions display immature morphology. Interestingly, wild-type viruses treated with Amprenavir (APV) to induce defective core maturation also induced stronger responses than the same viral particles generated in the absence of protease inhibitors.

Conclusions: We propose that immature HIV-1 virions generated from NL4-3/ΔRT viral clones may represent new prototypes of immunogens with a safer profile and stronger capacity to induce cellular immune responses than wild-type inactivated viral particles.

Figure 1. NL4-3/ΔRT plasmid graphic scheme.

NL4-3 (WT) plasmid was used to generate the NL4-3/ΔRT plasmid by deleting an 892 bp fragment, which codes for the retrotranscriptase (RT) and is inside the pol gene. These plasmids were used to transfect 293-T cells in order to obtain viruses. Numbers refer to pNL4-3.

Figure 2. NL4-3/ΔRT plasmid renders virions infectious but non-replicative.

(A) PBMC infection was monitored using NL4-3 (WT) and NL4-3/ΔRT (ΔRT) virions labelled with a Gag fusion protein (Gag-EGFP). Samples were imaged using a Leica SP2 confocal microscope. We observed an accumulation of Gag-EGFP in the cells infected with NL4-3 and NL4-3/ΔRT (arrows) but not in cells infected with Gag-EGFP. Cellular nuclei are stained in blue (DAPI), F-actin in red (Phalloidine) and the fusion protein Gag in green (Gag-EGFP). (B) NL4-3 and NL4-3/ΔRT virion replication measurements by luciferase activity in TZM-bl cells. The NL4-3 virions showed a decrease in luciferase activity directly proportional to the p24 quantity, while for NL4-3/ΔRT virions, luciferase activity was under the background independently of the p24 quantity.

Figure 3. Analysis of virion morphologies

by electron microscopy. (A) Electron micrographic analysis of 293-T cells transfected with pNL4-3. Mature virions (black arrows) show their characteristic core (white arrow heads). Budding particles (black arrow heads) are also present. (B) Electron micrograph showing immature viral particles budding from 293-T cells transfected with pNL4-3/ΔRT (black arrows). (C) Electron micrographic analysis of pNL4-3 purified viruses showing mature, immature and eccentric virion forms. (D) Electron micrograph of pNL4-3/ΔRT purified virions in which only immature virion forms can be observed. (E and F) NL4-3 and NL4-3/ΔRT virion magnifications respectively are shown. (G) Virion form measurements of NL4-3 versus NL4-3/ΔRT. NL4-3/ΔRT virion conformation was mostly immature and no mature virions were observed. On the other hand, NL4-3 virions were mainly mature. Error bars represent standard deviations. Samples were imaged using a Tecnai Spirit Twin electron-microscope.

Figure 4. Protein profile of NL4-3 and NL4-3/ΔRT virions.

Virions purified by ultracentrifugation were used to extract proteins and separated in a gradient 4-12% Tris-Glycine gel. An anti-p24 monoclonal antibody was used to study the processing profile of Gag in NL4-3/ΔRT and NL4-3 virions. There is a weaker processing of Gag in NL4-3/ΔRT virions, represented by an increase in p55-gag and MA-CA p41 forms in detriment of CA p24. At the same, an increase in intermediate forms is also shown in NL4-3/ΔRT virions.

Figure 5. Immunogenicity of different viral constructions measured by ELISPOT responses.

(A) Cryopreserved PBMCs from asymptomatic HIV seropositive individuals were tested for specific IFN-γ secreting T cells by in vitro stimulation with different HIV virions (WT+AT-2, ΔRT and ΔRT+AT-2). PBMCs were pulsed with 200 ng/ml p24 equivalents in all cases. The response elicited by WT+AT-2, ΔRT and ΔRT+AT-2 virions (mean ± SEM) is shown. The positivity threshold for each construct or antigen was defined as at least 50 SFC/10⁶ PBMC and at least twice that of the control medium. The magnitude of response differed significantly between stimuli (***p<0.001). (B) Four responder individuals were tested with different concentrations of WT+AT-2 and ΔRT virions (from 200 ng/ml to 5 µg/ml). The titration showed that the difference found between both stimuli was maintained, independently of the amount of virions used.

Figure 6. Immature NL4-3 virions improved the elicited cellular immune response.

(A) Electron microscopy analysis of purified NL4-3 virions treated with AT-2 and APV. Different virion morphologies were observed; most were immature and eccentric. (B) 52 electron-micrographs were analysed for virus morphologies. The APV incubation reduces mature virions but increases immature and eccentric forms. Error bars represent standard deviaations. (C) Twenty-one cryopreserved PBMCs from HIV seropositive individuals were assessed by ELISPOT to detect IFN-γ production. PBMCs were pulsed with 200 ng/ml p24 equivalents in all cases. Magnitude response was significantly lower against WT inactivated virions (WT+AT-2) than against WT+APV+AT-2 (*p<0.05). Averaged values from duplicate wells normalized to SFC/10⁶ PBMCs are shown for the stimulation conditions indicated. Plots represent mean ± standard error of the mean (SEM). Positivity threshold for each construct or antigen was defined as at least 50 SFC/10⁶ PBMC and at least twice that of the control medium.