Evasion of cytotoxic T lymphocyte (CTL) responses by nef-dependent induction of Fas ligand (CD95L) expression on simian immunodeficiency virus-infected cells

Abstract

Inoculation of macaques with live attenuated SIV strains has been shown to protect against subsequent challenge with wild-type SIV. The protective mechanism(s) remain obscure. To study the effect in more detail, we have investigated the role of virus-specific CTL responses in macaques infected with an attenuated SIV strain (pC8), which has a four-amino acid deletion in the nef gene, as compared with the wild-type SIVmac32H clone (pJ5). Cynomolgus macaques infected with pC8 were protected against subsequent challenge with pJ5 and did not develop any AIDS-like symptoms in the 12 months after infection. The pC8-induced protection was associated with high levels of virus-specific CTL responses to a variety of viral antigens. In contrast, pJ5-infected macaques had little, if any, detectable CTL response to the viral proteins after three months. The latter group of macaques also showed increased Fas expression and apoptotic cell death in both the CD4(+) and CD8(+) populations. In vitro, pJ5 but not pC8 leads to an increase in FasL expression on infected cells. Thus the expression of FasL may protect infected cells from CTL attack, killing viral-specific CTLs in the process, and providing a route for escaping the immune response, leading to the increased pathogenicity of pJ5. pC8, on the other hand does not induce FasL expression, allowing the development of a protective CTL response. Furthermore, interruption of the Fas-FasL interaction allows the regeneration of viral-specific CTL responses in pJ5-infected animals. This observation suggests an additional therapeutic approach to the treatment of AIDS.

Figure 1

Spontaneous DNA fragmentation of PBMCs from pC8-induced protected (N113-N114) or pJ5-infected (N174-N177) macaques. Fresh isolated PBMCs (1 × 10⁶) were cultured in medium containing 10% human AB serum for 4 or 16 h. Fragmented DNA was extracted as described in Materials and Methods. (A) DNA was analyzed by electrophoresis on 1.5% agarose gels; (B) Southern blot analysis of the DNA by hybridization with ³²P-labeled genomic macaque DNA; (C) The percentage of DNA fragmentation in B was calculated as dividing the total counts of each sample by the number of counts in the bottom 85–90% of the gel (below 23,000 bp). The percentage of DNA fragmentation of PBMCs from naive macaques was always <20% (data not shown). P value for the two groups <0.005 after 16 h.

Figure 2

Spontaneous DNA fragmentation of CD4 and CD8 subpopulations from pJ5-infected control (N174-N177) macaques. PBMCs were cultured in the medium for 16 h and CD4⁺ or CD8⁺ T cells was then purified by positive selection using anti-CD4 or CD8 magnetic beads, respectively. The purity of each subset was >95% as assessed by flow cytometry. After extracting the DNA, Southern blot was performed (A) and analyzed as described in Fig. 1 (B). P value for CD4 vs CD8 <0.05. These data are representative of three separate experiments.

Figure 3

Expression of Fas antigen on SIV-infected macaques. (A) Representative histograms of Fas expression on PBMCs. The percentage of Fas positive cells (mean fluorescent intensity >13 units): naive 6.2 ± 1.2% (n = 4), C8+J5-infected 17.6 ± 4.2% (n = 4), or pJ5-infected 36.9 ± 12.7% (n = 4) macaques (C8+J5 vs J5 P <0.05). (B) Fas expression on fractionated CD4⁺ or CD8⁺ T cells from pC8- and pJ5-infected groups (P = 0.08 for CD4⁺ cells and P <0.05 for the CD8⁺ cells).

Figure 4

SIV-infected cells kill Fas-sensitive target via FasL–Fas interaction. Naive monkey PBMCs (A) or CEM cells (B) were superinfected with SIV pC8 or pJ5 as indicated in Materials and Methods. SIV-infected cells were cocultured with ⁵¹Cr-labeled Fas-sensitive Jurkat cells in the presence or absence of Fas–Fc fusion proteins (10 μg/ml) for 12–16 h. Chromium release was determined by a β-plate counter. Specific lysis was calculated by subtracting the killing of Jurkat cells by mock-infected cells. Infectivity of SIV pC8 or pJ5 in CEM cells.

Figure 5

(A) Comparison of FasL induction by infection with pC8 or pJ5 as assessed by staining with the anti-FasL mAb. 24 h after infection cells were stained with biotinylated anti-human FasL mAb (NOK-1) plus PE-streptavidin (solid lines). Dotted lines indicate background staining with PE-streptavidin alone. (B) Staining of infected cells (solid lines) with anti-nef mAb demonstrates equal infectivity of pC8 and pJ5 infection. Dotted lines indicate background staining with control mAb.

Figure 5

(A) Comparison of FasL induction by infection with pC8 or pJ5 as assessed by staining with the anti-FasL mAb. 24 h after infection cells were stained with biotinylated anti-human FasL mAb (NOK-1) plus PE-streptavidin (solid lines). Dotted lines indicate background staining with PE-streptavidin alone. (B) Staining of infected cells (solid lines) with anti-nef mAb demonstrates equal infectivity of pC8 and pJ5 infection. Dotted lines indicate background staining with control mAb.

Figure 6

PBMCs from SIV pJ5-infected macaques kill Fas-sensitive target are CD4-dependent. PBMCs from pJ5-infected or uninfected macaques (n = 2 in each group) were stimulated with Con A (5 μg/ml) for 12 h. After washing three times with RPMI, CD4⁺ or CD8⁺ T cells were depleted by using anti-CD4 or CD8 magnetic beads, respectively. Cells were then cocultured with ⁵¹Cr-labeled Jurkat cells in the presence or absence of fusion proteins (10 μg/ml) or anti-human FasL mAbs (5 μg/ml) for 12–16 h. Specific lysis was assayed as described in Fig. 4.

Figure 7

Soluble Fas–Fc fusion protein regenerates CTL response from SIV pJ5-infected macaques. PBMCs were isolated from macaques (P93) 6 mo after infection and set up for bulk culture CTL in the presence or absence of either Fas–Fc fusion protein (10 μg/ml) or soluble CD4 (5 μg/ml) for 14 d. After washing the cells, the virus-specific CTL activities were determined as described in Materials and Methods. The data are representative of three seperate experiments.