Simian immunodeficiency virus promoter exchange results in a highly attenuated strain that protects against uncloned challenge virus

Abstract

Among the many simian immunodeficiency virus (SIV) immunogens, only live attenuated viral vaccines have afforded strong protection to a natural pathogenic isolate. Since the promoter is crucial to the tempo of viral replication in general, it was reasoned that promoter exchange might confer a novel means of attenuating SIV. The core enhancer and promoter sequences of the SIV macaque 239nefstop strain (NF-kappaB/Sp1 region from -114 bp to mRNA start) have been exchanged for those of the human cytomegalovirus immediate-early promoter (CMV-IE; from -525 bp to mRNA start). During culture of the resulting virus, referred to as SIVmegalo, on CEMx174 or rhesus macaque peripheral blood mononuclear cells, deletions arose in distal regions of the CMV-IE sequences that stabilized after 1 or 2 months of culture. However, when the undeleted form of SIVmegalo was inoculated into rhesus macaques, animals showed highly controlled viremia during primary and persistent infection. Compared to parental virus infection in macaques, primary viremia was reduced by >1,000-fold to undetectable levels, with little sign of an increase of cycling cells in lymph nodes, CD4(+) depletion, or altered T-cell activation markers in peripheral blood. Moreover, in contrast to wild-type infection in most infected animals, the nef stop mutation did not revert to the wild-type codon, indicating yet again that replication was dramatically curtailed. Despite such drastic attenuation, antibody titers and enzyme-linked immunospot reactivity to SIV peptides, although slower to appear, were comparable to those seen in a parental virus infection. When animals were challenged intravenously at 4 or 6 months with the uncloned pathogenic SIVmac251 strain, viremia was curtailed by approximately 1,000-fold at peak height without any sign of hyperactivation in CD4(+)- or CD8(+)-T-cell compartment or increase in lymph node cell cycling. To date, there has been a general inverse correlation between attenuation and protection; however, these findings show that promoter exchange constitutes a novel means to highly attenuate SIV while retaining the capacity to protect against challenge virus.

FIG. 1.

Structure of the SIVmegalo promoter chimeras. The LTR structure of parental and SIVmegalo chimera is presented. The positions of transcription factor binding motifs (for a review, see reference 28) and TAR sequences are shown. The CMV-IE promoter and SIV sequences are fused such that the transcription start site of CMV coincides with that of SIV.

FIG. 2.

Replication kinetics and evolution of SIVmegalo promoter during replication on CEMx174 cells (left panels A, C, and E) and rPBMC (right panels B, D, and F). Five million cells were infected by 1 ng of RT activity of SIVmac239 or SIVmegalo. RT activity released in the supernatant was measured after infection of CEMx174 cells (A) or rPBMC (B). SIVmac239nefstop or SIVmegalo growth curves on CEMx174 cells are represented as the average of three separate experiments (verticals bars representing the standard deviation [SD]), whereas the growth curves on rPBMC were obtained from one donor (macaque 93035); comparable results were obtained from four other donors. (C and D) Genomic DNA was extracted from different time points, and a PCR was performed with primers spanning the recombined LTR region. The PCR product size was visualized under UV. The SIVmegalo amplicon was 750 bp, while that of SIVmac239 was 260 bp. (E and F) Nucleotide sequences obtained from PCR product 15 days after infection are indicated as horizontal bars relative to repeat sequences described in pCMV-IE. The frequencies of sequences are reported on the right.

FIG. 3.

Detail of the recombined LTR sequences obtained from in vitro CEMx174 or rPBMC infected cells or from in vivo lymph node infected cells (macaque 93035). CEMx174 or rPBMC (macaque 93035) were infected with SIVmegalo, the sequence of which is reported at the top of each lane. At 2 months after infection a unique viral sequence was obtained on CEMx174 cells. At 1 month after infection the two major viral sequences obtained from PBMC were also reported. At 100 days after infection of macaque 93035 with SIVmegalo, LNMC were collected and submitted to PCR amplification of the promoter region. All 10 sequences out of 10 showed the same deletion of 190 bp.

FIG. 4.

Activity of deleted chimeric LTRs in CAT and virus assays. (A) Structure of the HIV-1 Tat- and Rev-dependent CAT reporter construct. (B) CAT activity of deleted LTR compared to SIVmac239 and SIVmegalo reference clones. Δclone61 was derived from the CEMx174 culture at 60 days (Fig. 3). (C) Growth curve of Δclone61 virus compared to SIVmac239 and SIVmegalo controls.

FIG. 5.

Viral load after infection of Chinese rhesus macaques (Macacca mulatta) with SIVmac239nefstop or SIVmegalo. Individual serum viremia profiles determined by bDNA assay (Chiron) are shown on the top panel for 15 monkeys infected with SIVmegalo (A) and for five SIVmac239nefstop-infected monkeys (B). Given the large number of points and wide range, the bottom panel (C) shows median values ± the SD. The serum viremia cutoff was 10³ RNA copies/ml.

FIG. 6.

Dynamics of activated CD69⁺ CD4⁺ and CD69⁺ CD8⁺ cells in peripheral blood during primary SIVmegalo infection. CD69 markers were monitored in CD4⁺- and CD8⁺-T-cell subpopulations for seven SIVmegalo-infected animals and two SIVmac239nefstop-infected animals for 2 months. Insets show the profiles for samples taken almost daily for a single animal (96R0258) between days 4 and 11.

FIG. 7.

Serum antibody responses to SIVmegalo infection. (A) Reciprocal dilution titers for the 15 SIVmegalo-infected (A) and five SIVmac239nefstop (B) control macaques. The cutoff was a 100-fold dilution.

FIG. 8.

IFN-γ ELISPOT responses against pooled Gag (A) or Nef (B) peptides or heat-inactivated HIV-2 virions (C) in PBMC drawn from SIVmegalo-infected macaques during primary infection, along with two SIVmac239nefstop-infected animals as controls. Spot frequencies are given as the signal minus the background. A cocktail of 16 SIV Gag or 19 Nef 15-mer peptides was used. Note that the animal (96R0202) that failed to produce an antibody response to SIVmegalo showed ELISPOT responses comparable to some of the other antibody-positive animals.

FIG. 9.

Anti-Gag peptide responses in CD4⁺- and CD8⁺-T-cell compartment. Three months after infection with SIVmegalo, rPBMC from four SIVmegalo-infected animals were submitted to a CD4⁺- or CD8⁺-T-cell depletion (>90% depletion was achieved) and used for IFN-γ ELISPOT assay against pooled Gag peptides. Spot frequencies are given as the signal minus the background.

FIG. 10.

Serum viremia of a pilot challenge study with homologous virus. At 9 and 15 months after SIVmegalo inoculation, two macaques were challenged with 200 TCID₅₀ of homologous virus (SIVmac239nefstop). Two naive control animals were inoculated with the same dose of virus on the same day. The serum viremia cutoff was 10³ RNA copies/ml.

FIG. 11.

SIVmegalo infection protects against uncloned pathogenic SIVmac251 challenge virus. (A and B) Serum viremia for animals challenged at 4 and 6 months, respectively. (C) The data for six naive macaques infected by the same SIVmac251 challenge stock are also given. Panel D shows the median values ± the SD. The serum viremia cutoff was 10³ RNA copies/ml. The difference between the combined challenge groups (n = 8) and naive controls (n = 6) is statistically significant (P = 0.0019, Mann-Whitney test).

FIG. 12.

Anamnestic antibody response in SIVmegalo-challenged animals. Four animals were challenged at 4 (A) and 6 (B) months after SIVmegalo infection with a pathogenic SIVmac251 strain. The numbers under datum points indicate the numbers of coincident points.

FIG. 13.

SIVmegalo infection protects against uncloned pathogenic SIVmac251 challenge virus. Mean number of Ki67⁺ cells per mm² in lymph node biopsies taken at 60 to 70 days postinfection or postchallenge. Sample sizes are indicated by “n,” while the SD, maximum, and minimum values are also shown. The SIVmegalo-infected animals were macaques 960938, 960954, 970222, 970364, 97R0012, and 97R0276; those infected by SIVmegalo and challenged by SIVmac251 were macaques 97R0012, 96R0276, 970222, and 970364; the SIVmac251-infected animals were macaques 970034, 960976, 264, 9025, 92418, and 92428. All differences between SIVmac251 and the other categories are statistically significant (P < 0.01).