A recombinant herpesviral vector containing a near-full-length SIVmac239 genome produces SIV particles and elicits immune responses to all nine SIV gene products

Abstract

The properties of the human immunodeficiency virus (HIV) pose serious difficulties for the development of an effective prophylactic vaccine. Here we describe the construction and characterization of recombinant (r), replication-competent forms of rhesus monkey rhadinovirus (RRV), a gamma-2 herpesvirus, containing a near-full-length (nfl) genome of the simian immunodeficiency virus (SIV). A 306-nucleotide deletion in the pol gene rendered this nfl genome replication-incompetent as a consequence of deletion of the active site of the essential reverse transcriptase enzyme. Three variations were constructed to drive expression of the SIV proteins: one with SIV's own promoter region, one with a cytomegalovirus (cmv) immediate-early promoter/enhancer region, and one with an RRV dual promoter (p26 plus PAN). Following infection of rhesus fibroblasts in culture with these rRRV vectors, synthesis of the early protein Nef and the late structural proteins Gag and Env could be demonstrated. Expression levels of the SIV proteins were highest with the rRRV-SIVcmv-nfl construct. Electron microscopic examination of rhesus fibroblasts infected with rRRV-SIVcmv-nfl revealed numerous budding and mature SIV particles and these infected cells released impressive levels of p27 Gag protein (>150 ng/ml) into the cell-free supernatant. The released SIV particles were shown to be incompetent for replication. Monkeys inoculated with rRRV-SIVcmv-nfl became persistently infected, made readily-detectable antibodies against SIV, and developed T-cell responses against all nine SIV gene products. Thus, rRRV expressing a near-full-length SIV genome mimics live-attenuated strains of SIV in several important respects: the infection is persistent; >95% of the SIV proteome is naturally expressed; SIV particles are formed; and CD8+ T-cell responses are maintained indefinitely in an effector-differentiated state. Although the magnitude of anti-SIV immune responses in monkeys infected with rRRV-SIVcmv-nfl falls short of what is seen with live-attenuated SIV infection, further experimentation seems warranted.

Fig 1. Position of the SIVcmv-nfl expression cassette in recombinant RRV.

A near full-length genome (nfl) sequence of SIVmac239 with a 520 bp deletion (red X) in the 5’ Long Terminal Repeat (LTR) region, a deletion in the pol region spanning 306 bp, corresponding to the active site of the reverse transcriptase, and a 414 bp deletion in the 3’ LTR region, was generated via Gibson cloning. Numbering system corresponds to that of Regier and Desrosiers [16]. Subsequently, utilizing overlaps containing PmeI restriction sites, the cytomegalovirus immediate-early enhancer and promoter (pCMV), the SIV-nfl sequence containing a C-terminal V5-tagged nef open reading frame (ORF), and the bovine growth hormone (BGH) polyA signal were inserted between the left terminal repeats (TR) and the first ORF R1 of RRV via Gibson cloning. Full length recombinant RRV was obtained from subsequent transfection of overlapping cosmid clones [21].

Fig 2. Kinetics of recombinant RRV-mediated SIV-nfl expression.

Early passage rhesus fibroblast (RF) cells were grown in 6-well culture plates and the cells in each well were infected with the indicated strains of recombinant rRRV-SIV-nfl. Cells were cultured for the indicated number of days post infection prior to harvest. Subsequently, cells were lysed, and the lysate was subjected to immunoblotting, measuring the expression levels of SIV239 Env, Gag, or Nef. The two bands indicated with arrowheads in the immunoblotting of Gag correspond to p55 and p27. The major Env protein band is gp120; the Nef protein is 34 kDa. For rRRV-SIVltr-nfl the pCMV was replaced with an intact 5’ SIVmac239 LTR. For rRRV-SIVdual-nfl the pCMV was replaced with a promoter construct consisting of the RRV ORF26 promoter (p26) and the RRV Poly Adenylated Nuclear RNA promoter (PAN).

Fig 3. Production of replication-defective SIV particles by rRRV-SIV-nfl.

A) Early passage rhesus fibroblast (RF) cells were grown in 6-well culture plates and subsequently infected with 3.5x10⁹ genome copies of the indicated recombinant SIV-nfl. The RF cell culture supernatants were collected at the indicated time points post infection and subjected to an antigen capture assay to measure the levels of SIV Gag p27. Supernatant of uninfected RF cells served as a negative control. B) Cell culture supernatant was harvested from early passage rhesus fibroblast (RF) cells infected with the respective recombinant RRV-SIV-nfl strains. Subsequently, CEMx174 cells were infected with these supernatants containing 20ng of p27 as measured by an antigen capture assay. An equal amount (20ng of p27) of SIVmac239 wild-type virus produced in HEK239T-cells served as a positive control. The supernatants from the CEMx174 cell cultures were collected at the indicated days post infection and were used in an antigen capture assays measuring p27 levels. Supernatant of uninfected RF cells served as a negative control. C) Early passage rhesus fibroblast (RF) cells grown in T75 culture flasks were infected with rRRV-SIVcmv-nfl and harvested at days 4–6 post infection. Subsequently, the cells were fixed with 2.5% glutaraldehyde in sodium cacodylate buffer. Transmission electron microscopic (TEM) images were taken at The Core Electron Microscopy Facility, University of Massachusetts. Shown are SIV particles budding from cell membrane (filled arrowheads). The bar in the lower left corner represents 500 nm D) TEM image showing free SIV particles including mature SIV particles (filled arrowheads) containing cylindrical, rod-shaped nucleoid as well as immature particles lacking any nucleoid structures (empty arrowheads). The bar in the lower left corner represents 200 nm. In the inset on the upper right, a TEM image of recombinant RRV derived from a separate image is shown.

Fig 4. Humoral immune responses following infection of rhesus monkeys with rRRV-SIVcmv-nfl.

Sera obtained at the indicated time points following rRRV-SIVcmv-nfl inoculation were diluted 1:20 and tested for the presence of A) anti-RRV responses via an ELISA utilizing ELISA plates coated with purified wild type RRV lysate. Reactive antibodies were detected with a horseradish-peroxidase-conjugated anti-rhesus IgG antibody. B) Similarly, the sera were tested for the presence of anti-gp140 antibodies via an ELISA using recombinant SIVmac239 gp140 to coat the ELISA plates. Reactive antibodies were detected with a horseradish-peroxidase-conjugated anti-rhesus IgG antibody. C) The obtained sera were tested for the presence of anti-gp120 antibodies via an ELISA using recombinant SIVmac239 gp120 to coat the ELISA plates. Reactive antibodies were detected with a horseradish-peroxidase-conjugated anti-rhesus IgG antibody.

Fig 5. SIV gp120/140 reactivity at week 20 post rRRV-SIVcmv-nfl inoculation.

A) Sera obtained at week 20 following rRRV-SIVcmv-nfl inoculation were serially diluted and tested for the reactivity to gp120 via an ELISA using recombinant SIVmac239 gp120 to coat the ELISA plates. Subsequently, reactive antibodies were detected with a horseradish-peroxidase-conjugated anti-rhesus IgG antibody. Furthermore, sera of two SIVmac239Δnef-infected animals (week 18 post infection) and sera obtained from two monkeys that had received rRRV expressing a codon-modified version of SIVmac239 gp160 (week 19 post inoculation) were tested [24]. B) Similarly, the monkey sera were serially diluted and tested for reactivity to gp140 by ELISA, using recombinant SIVmac239 gp140 to coat the ELISA plates. Reactive antibodies were detected with a horseradish-peroxidase-conjugated anti-rhesus IgG antibody.

Fig 6. SIVmac316 neutralization titers at week 20 post rRRV-SIVcmv-nfl inoculation in rhesus monkeys.

Sera from the rRRV-SIVcmv-nfl-inoculated monkeys taken at week 20 post inoculation were serially diluted and subjected to neutralization assays against SIVmac316 utilizing TZM-bl cells. A pool of sera from SIVmac239-infected monkeys served as a positive control and pooled sera from specific pathogen free (SPF) monkeys as a negative control. The dashed line indicates 50% relative light units representing 50% neutralization activity.

Fig 7. rRRV-SIVcmv-nfl vaccination elicits Gag CM9- and Tat SL8-specific CD8+ T-cell responses in Mamu-A*01+ macaques.

Fluorescently-labeled Mamu-A*01 tetramers folded with peptides corresponding to the Gag CM9 (A) or Tat SL8 (B) epitopes were used to monitor the ontogeny of vaccine-induced CD8+ T-cell responses in the two Mamu-A*01+ rRRV-SIVcmv-nfl-inoculated monkeys (r11089 and r11099).

Fig 8. Memory phenotype of vaccine-induced CD8+ T-cell responses in Mamu-A*01+ rRRV-SIVcmv-nfl vaccinees.

The frequencies of Mamu-A*01/Gag CM9 and Mamu-A*01/Tat SL8 tetramer+ CD8+ T-cells in PBMC at week 10 post rRRV-SIVcmv-nfl inoculation are shown in the left panels for r11089 (A) and r11099 (B). The middle panels show the delineation of memory subsets within the tetramer+ gate based on the differential expression of CD28 and CCR7. Three subsets were identified: central memory (T_CM; CD28+CCR7+), transitional memory (T_EM1; CD28+CCR7-), and effector memory (T_EM2; CD28-CCR7-). The histograms in the right panels show the levels of granzyme B expressed by tetramer+ CD8+ T-cells (blue lines). The red lines correspond to tetramer+ CD8+ T-cells stained with an isotype-matched control monoclonal antibody.

Fig 9. Intracellular cytokine staining analysis of vaccine-induced SIV-specific T-cell responses in rRRV-SIVcmv-nfl-inoculated macaques.

The magnitude and specificity of vaccine-elicited CD8+ (A) and CD4+ (B) T-cell responses were measured in PBMC by ICS using pools of peptides (15mers overlapping by 11 amino acids) spanning the appropriate SIVmac239 proteins. In the week 10 assay (A & C), Vpx and Vpr peptides were grouped in a single pool and so were the Rev and Tat peptides. Pol peptides were omitted in the week 10 assay. In the week 18 assay (B & D), 1–3 peptide pools corresponding to individual SIVmac239 proteins were used as stimuli. The percentages of responding CD8+ T-cells or CD4+ T-cells displayed were calculated by adding the frequencies of positive responses producing any combination of three immunological functions (IFN-γ, TNF-α, and CD107a). Lines represent medians and each symbol corresponds to one vaccinee.