Combinatorial Latency Reactivation for HIV-1 Subtypes and Variants▿ †

1. 

   View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   FIG. 1.

   Lentiviral latency system and HIV-1 LTR of various subtypes. (A) FACS sorting procedure for polyclonal and clonal populations of LGIT and LG virus infections. Jurkat cells were infected at a low MOI with LGIT (1a) or LG (1b) lentivirus. Six days postinfection, gene expression was strongly stimulated with exogenous Tat protein and HMBA for LGIT virus-infected cells (2a) or Tat protein, HMBA, and TNF-α for LG virus-infected cells (2b). Eighteen hours after stimulation, single GFP-positive (GFP⁺) cells were sorted from LG virus infections (3b) and cultured for 4 weeks to generate LG clones (4b). Similarly, 18 h after stimulation, polyclonal FACS isolation of GFP⁺ LGIT virus-infected cells removed uninfected cells (3a). After sorting, GFP⁺ LGIT-infected cells were cultured under normal conditions for 2 weeks, during which substantial fractions of GFP⁺ cells relaxed to the off expression mode (4a). FACS was again applied to isolate polyclonal fractions of “infected but off” (GFP⁻) cells, and these fractions are used as models for latent infections (5b). Likewise, single cells were sorted and expanded to generate LGIT clones, and after 4 weeks of culturing, phenotypic bifurcation (PheB) clones were identified (5a). Off sorts, polyclonal clones that sorted into the off expression mode. (B) Schematic of alignments and DNA-binding elements of U3 regions for subtypes in this study. Binding sites were identified using the Transcription Element Search System (TESS) (http://www.cbil.upenn.edu/cgi-bin/tess/tess). Gray ovals indicate deviations in Sp1 sequence that likely compromise the function of Sp1 site II (for subtypes A2, A, and A/G) and Sp1 site III (for subtypes D, F, and H). Full U3 subtype sequences are supplied in Fig. S6 in the supplemental material. Two distinct isolates of subtype C were analyzed throughout this investigation (C refers to the sequence with GenBank accession no. AF067157, and C′ refers to the sequence with GenBank accession no. AF067154).
2. 

   View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   FIG. 2.

   Latency reactivation for LGIT virus mutants and HIV-1 subtypes in the Jurkat cell system. (A) Infection and serial FACS sorting were performed to isolate the infected, off populations for variants of the LGIT virus. These include mutants of subtype B (mutI Sp1 [S1], mutII Sp1 [S2], mutIII Sp1 [S3], mutI NF-κB [N1]), and mutII NF-κB ([N2]) or variants with U3 regions isolated from the following subtypes: B, A2, A, A/G, B/C, C′, C, B/F, D, F, and H (as in Fig. 1B). One day after FACS sorting (day 22 postinfection [Fig. 1A, panels 5b and 6b]), polyclonal off sorts for WT LGIT mutant and HIV-1 subtype variants were treated with the following pharmacological agents to reactivate latent infections: NF-κB/PKC activators TNF-α (white bars), PMA (gray bars), or prostratin (black bars). Data represent the averages of three independent measurements for each drug perturbation, and error bars are standard deviations. For the LGIT mutants of subtype B, upward and downward arrowheads indicate statistically significant deviations from the wild-type subtype B LTR configuration of LGIT (P < 0.05). The broken gray line at 50% reactivation is drawn as a reference marker. (B) Same as in panel A for latency reactivation by TSA (white bars), SAHA (light gray bars), valproic acid (dark gray bars), or HMBA (black bars). (C) Same as in panel A for latency reactivation using the combination of prostratin and SAHA. Asterisks denote statistical synergism by the combination of drugs relative to the reactivation by either individual agent. For details on the quantitative treatment of synergy, see Materials and Methods.
3. 

   View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   FIG. 3.

   Isolation and infection of human primary CD4⁺ T cells. (A to D) Naïve CD4⁺ T cells were isolated from human whole blood from three donors (9.1, 9.2, and 9.3). Six days after isolation, the cells were analyzed by flow cytometry for expression of surface receptors CD4 (A), CD45RA (B), CD45RO (C), and CD27 (D). Histogram overlays include negative controls (shaded gray) and naïve CD4⁺ T cells (black outline). Fluorescence channels 6 (FL 6), 8, 4, and 9 are shown in panels A, B, C, and D, respectively. (E) Naïve CD4⁺ T cells were activated with CD3/CD28 antigen beads for 3 days after isolation to promote T-cell activation and expansion. At 7 days postisolation, cells from each donor were infected by one of seven different LGIT subtype variants at a low MOI. Cells were cultured in activating conditions until day 14, at which cells were transferred to minimal growth medium (1 ng/ml IL-7 and 10 U/ml IL-2).
4. 

   View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   FIG. 4.

   Reactivation of latent HIV-1 in primary memory CD4⁺ T cells. (A) After 14 days of culturing in quiescent conditions (and 28 days after CD4⁺ T-cell isolation), the cells were treated with antilatency drugs HMBA (H), prostratin (P), SAHA (S), or the combination of prostratin and SAHA (P+S). The data shown include each LGIT subtype for each of three independent donors. Asterisks indicate synergism of the prostratin-SAHA combination with respect to either drug alone. For details on the quantitative treatment of synergy, see Materials and Methods. (B to D) At day 21 postinfection (day 28 postisolation), primary cells were examined for the proviral LTR expression of GFP (x axis) and the surface expression (y axis) of CD4 (B), CD45RO (C), or CD27 (D). (E to G) At days 21 postinfection (same as panels B to D), primary cells were treated with prostratin and SAHA for 24 h and examined for proviral LTR expression of GFP (x axis) and surface expression (y axis) of CD4 (E), CD45RO (F), or CD27 (G).
5. 

   View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   FIG. 5.

   Latency reactivation of LG clones with subtype B LTR. (A) Five HIV-1 subtype B LTR-GFP (LG) clones (BB1, BC5, DA4, DD1, and DD2) were isolated to examine latency reactivation for the Tat-deficient lentivirus (Fig. 1A, panel 4b). Each Jurkat cell clonal population was subsequently infected with another lentivirus (OrIT) that constitutively expresses mOrange and HIV-1 Tat (subtype B) from the human ubiquitin promoter (MOI of 0.15 to 0.20). The cells were analyzed for expression of mOrange (x axis) and GFP (y axis), as shown in two-dimensional (2-D) histograms. Although cells are clonal with respect to the LG (GFP) infection, there are subpopulations that are either responsive (mOrange⁺/GFP⁺) or resistant (mOrange⁺/GFP⁻) to Tat transactivation. (B) LG clones, as isolated in Fig. 1A, panel 4b, were reactivated with either Tat (OrIT infection) or antilatency agents. As measured by flow cytometry, the percentage of GFP⁺ cells are indicated for the original LG clone (green bars), the LG population infected with the OrIT lentivirus (orange bars), and each LG clone stimulated with HMBA (yellow bars), prostratin (blue bars), SAHA (red bars), or the combination of prostratin and SAHA (black bars). The position of the off gate is set for uninfected Jurkat cells (GFP⁻), whereas the on gate indicates GFP⁺ cells. All data are averages of three biological replicate samples, and error bars are standard deviations. The small black circles indicate at least a 10% increase from the unperturbed samples, and asterisks indicate synergistic reactivation by the drug combination. Drug concentrations are provided in Materials and Methods, and the histograms are provided in Fig. S8 in the supplemental material. For details on the quantitative treatment of synergy, see Materials and Methods.
6. 

   View larger version:

   • In this window
   • In a new window

   • Download as PowerPoint Slide

   FIG. 6.

   Integration site analysis for LG clones after antilatency activation. (A) The integration site of LG clone BB1 was identified at chr16:29684467 (chromosome 16, position 29684467) outside a region of known transcripts and positioned between and in the opposite orientations of the reading frames for SPN (GeneID 6693) and QPRT (GeneID 23475). (B) LG clone BB1 was treated with DMSO (negative control), HMBA, prostratin, SAHA or the combination of prostratin and SAHA. After a 3-h incubation period, RT-PCR was performed on the untreated and treated cells to quantify the expression from the LTR (GFP), the nearest endogenous downstream gene (SPN), and the nearest upstream gene (QPRT). Using the ΔΔC_T method, all data were first normalized by the respective expression of β-actin and then by the relative expression of the unperturbed control. All control and drug perturbations were performed in three biological replicate samples, and data represent averages of three independent measurements. Error bars represent standard deviations. Upward and downward arrowheads indicate statistically significant deviations from the DMSO negative control (P < 0.05). Asterisks denote statistically significant synergism of the prostratin-SAHA combination with respect to the individual components. For details on the quantitative treatment of synergy, see Materials and Methods. (C) The integration position of LG clone BC5 was identified at chr16:28841942 inside the reading frame and in the opposite orientation of ATXNL2 (GeneID 11273). (D) Same as in panel B for LG clone BC5. RT-PCR was performed on the untreated and treated cells to quantify the expression from the LTR (GFP), and the mRNA from the ATXNL2 gene upstream (atxUS) and downstream (atxDS) of the LG integration position. (E) The integration position of LG clone DA4 was identified at chr13:20605604 inside the reading frame and in the orientation opposite that of ZMYM2 (GeneID 7750). (F) Same as in panel B for LG clone DA4. RT-PCR was performed on the untreated and treated cells to quantify the expression from the LTR (GFP), and the mRNA from the ZMYM2 gene upstream (zmUS) and downstream (zmDS1 and zmDS2) of the LG integration position. Primers for two different downstream sites of ZMYM2 were used to examine different potential splice variants.

• American Society for Microbiology