HIV-1 latency in actively dividing human T cell lines

Abstract

Background: Eradication of HIV-1 from an infected individual cannot be achieved by current drug regimens. Viral reservoirs established early during the infection remain unaffected by anti-retroviral therapy and are able to replenish systemic infection upon interruption of the treatment. Therapeutic targeting of viral latency will require a better understanding of the basic mechanisms underlying the establishment and long-term maintenance of HIV-1 in resting memory CD4 T cells, the most prominent reservoir of transcriptional silent provirus. However, the molecular mechanisms that permit long-term transcriptional control of proviral gene expression in these cells are still not well understood. Exploring the molecular details of viral latency will provide new insights for eventual future therapeutics that aim at viral eradication.

Results: We set out to develop a new in vitro HIV-1 latency model system using the doxycycline (dox)-inducible HIV-rtTA variant. Stable cell clones were generated with a silent HIV-1 provirus, which can subsequently be activated by dox-addition. Surprisingly, only a minority of the cells was able to induce viral gene expression and a spreading infection, eventhough these experiments were performed with the actively dividing SupT1 T cell line. These latent proviruses are responsive to TNFalpha treatment and alteration of the DNA methylation status with 5-Azacytidine or genistein, but not responsive to the regular T cell activators PMA and IL2. Follow-up experiments in several T cell lines and with wild-type HIV-1 support these findings.

Conclusion: We describe the development of a new in vitro model for HIV-1 latency and discuss the advantages of this system. The data suggest that HIV-1 proviral latency is not restricted to resting T cells, but rather an intrinsic property of the virus.

Figure 1

Schematic overview of HIV-1 variants used in this study. In the HIV-rtTA variant, the rtTA gene is inserted in place of the nef gene and TetO binding sites are inserted in the HIV-1 promoter. Inactivation of the Tat-TAR axis is obtained by mutations in the R region and the Tat protein. The shNef expression cassette is inserted in the U3 region. The Tat inactivating mutation (indicted by X) in HIV-rtTA has been restored in the HIV-rtTA Tat^wt variant.

Figure 2

Northern blot analysis of shRNA production in rtTA-shNef SupT1 cell lines. RNAs from the different HIV-rtTA-shNef SupT1 clones were isolated and separated on an agarose gel. After blotting the membrane was probed with a LNA probe against the shNef. Lane M, marker; lane 1, shNef clone B9; lane 2, clone C10 (a negative control cell line without a provirus); lane 3, shNef clone D2; lane 4, shNef clone D11; lane 5, shNef clone F7; lane 6, shNef clone F8; lane 7 and 8 contain positive controls from a transfection with the F-shNef^- and F-shNef⁺ plasmid [31]; lane 9 contains a negative empty-vector control; lane 10 in vitro produced shNef RNA.

Figure 3

Silencing and reactivation in HIV-rtTA SupT1 cell lines. The HIV-rtTA SupT1 cell lines F7 and B9 were induced with dox or with dox in combination with the indicated activators. DMSO is the solvent for genistein and therefore used as an additional control. Statistical significance was determined with a two-tailed student's T test for each combination versus dox alone (GraphPad Prism). Significant changes (P-value < 0.05) are indicated with asterisks.

Figure 4

Latent HIV-1 infection in SupT1 T cells. TNFα-induced reactivation of silent HIV-1 proviruses in SupT1 cells was analyzed by determining the percentage of CA-p24 producing cells by intracellular FACS analysis. (A) After a single round infection with the LAI isolate, the culture was split and one culture activated for 24 h with TNFα and the other used as a control. (B) The control culture was maintained for one week and split again into a TNFα treated and control culture. Statistical significance was determined with a two-tailed student's T test (GraphPad Prism).

Figure 5

Latent HIV-1 and HIV-rtTA infection in SupT1 T cells. TNFα-induced reactivation of silent HIV-1 proviruses in SupT1 cells was analyzed by determining the percentage of CA-p24 producing cells by intracellular FACS analysis. After a single round infection with HIV-1 (LAI isolate) or the HIV-rtTA-Tat^wt, each culture was split and either activated for 24 h with TNFα or not. The fold TNFα activation is the percentage CA-p24 positive cells in the culture with TNFα divided by the percentage of CA-p24 positive cells in the control culture.

Figure 6

Latent HIV-1 infection in various T cell lines. TNFα-induced reactivation of silent HIV-1 proviruses in the indicated T cell lines was done as described in Figure 4. The fold TNFα activation is the percentage CA-p24 positive cells in the culture with TNFα divided by the percentage of CA-p24 positive cells in the control culture.