Cyclin T1-dependent genes in activated CD4 T and macrophage cell lines appear enriched in HIV-1 co-factors

Abstract

HIV-1 is dependent upon cellular co-factors to mediate its replication cycle in CD4(+) T cells and macrophages, the two major cell types infected by the virus in vivo. One critical co-factor is Cyclin T1, a subunit of a general RNA polymerase II elongation factor known as P-TEFb. Cyclin T1 is targeted directly by the viral Tat protein to activate proviral transcription. Cyclin T1 is up-regulated when resting CD4(+) T cells are activated and during macrophage differentiation or activation, conditions that are also necessary for high levels of HIV-1 replication. Because Cyclin T1 is a subunit of a transcription factor, the up-regulation of Cyclin T1 in these cells results in the induction of cellular genes, some of which might be HIV-1 co-factors. Using shRNA depletions of Cyclin T1 and transcriptional profiling, we identified 54 cellular mRNAs that appear to be Cyclin T1-dependent for their induction in activated CD4(+) T Jurkat T cells and during differentiation and activation of MM6 cells, a human monocytic cell line. The promoters for these Cyclin T1-dependent genes (CTDGs) are over-represented in two transcription factor binding sites, SREBP1 and ARP1. Notably, 10 of these CTDGs have been reported to be involved in HIV-1 replication, a significant over-representation of such genes when compared to randomly generated lists of 54 genes (p value<0.00021). The results of siRNA depletion and dominant-negative protein experiments with two CTDGs identified here, CDK11 and Casein kinase 1 gamma 1, suggest that these genes are involved either directly or indirectly in HIV-1 replication. It is likely that the 54 CTDGs identified here include novel HIV-1 co-factors. The presence of CTDGs in the protein space that was available for HIV-1 to sample during its evolution and acquisition of Tat function may provide an explanation for why CTDGs are enriched in viral co-factors.

Figure 1

A) MM6 cells: non-infected or MM6 cells infected at an m.o.i. of five with the indicated lentiviral vector were cultured for five days. Cells were then treated with LPS for 24 hours as indicated. Cell extracts were prepared and immunoblots performed to measure levels of Cyclin T1, CDK9, and β-actin. B) Jurkat cells: non-infected or Jurkat cells infected at an m.o.i. of five with the indicated lentiviral vector were cultured for five days. Cells were then treated with PMA+ ionomycin for 24 hours as indicated. Cell extracts were prepared to measure expression of Cyclin T1, CDK9, and β-actin.

Figure 2. CTDGs.

The Venn diagram represents the number of genes whose mRNAs were reduced >1.5-fold (p value<0.05) by Cyclin T1 depletions. Cyclin T1-dependency was observed for 965 of the PMA-inducible genes in MM6 cells, 778 of the LPS-inducible genes in MM6 cells, and 644 of the PMA/ionomycin-inducible genes in Jurkat cells. The number of genes in intersections of gene sets is indicated.

Figure 3. SREBP1 binding sites in CTDGs.

The 1.5 kb sequences upstream of the first exon of genes are indicated (both strands). Predicted binding sites on both DNA strands are indicated.

Figure 4. CDK11 depletion reduces protein expression of CDK9 and HEXIM1 and Tat transactivation of HIV-1 provirus.

A. HeLa cells were transfected with siRNAs against CDK11 siRNA or control siRNAs. Cells were harvested 72 h post-transfection and cell lysates were analyzed for expression of CDK11, CDK9, HEXIM1, and CDK8 by immunoblots. B. HeLa cells were infected with a lentiviral DHFR-Renilla Luciferase virus with GFP marker. Cells sorted for GFP were then infected with either HIV-1 NL4-3-firefly Luciferase virus (wild type) or with HIV-1 NL4-3 firefly Luciferase virus encoding a mutant Tat (Tat−) for 48 h. Cultures were transfected with the indicated siRNAs. Cell extracts were prepared 48 h post-transfection and assayed for Luciferase expression. Results are presented as the ratio of firefly to Renilla Luciferase expression.

Figure 5. CSNK1G1 depletions.

As illustrated, 293T cells were transfected with either control siRNAs or siRNAs against CSNK1G1. One day later, cells were co-transfected with the pNL4-3-Luc and pVSV-G plasmids for virus production. Supernatants were collected 48 hours later and p24 levels were measured by an ELISA assay. Infectivity of supernatants containing reporter the NL-4-3-Luc reporter viruses were measured by infecting HeLa cells with equal volume of supernatants, and Luciferase expression was measured 48 hours post-infection. Luciferase expression was normalized to the amount of p24 in the input supernatants.

Figure 6. Effects of over-expression of wild type and dn-CSNK1G1 proteins on Gag processing.

Cultures of 293T cells were transfected with 0.5 or 1.0 µg wild type CSNK1G1, dn-CSNK1G1, or parental vector (pcDNA 3.0) expression plasmids. One day later, cultures were transfected with pNL4-3-GFP proviral reporter plasmid and cell extracts were prepared one day later. Gag processing in virions associated with cells were examined in an immunoblot using a p24 monoclonal antibody. The ratio of p24 to the 55 kDa Gag precursor is indicated at the bottom of the immunoblot and it was and it was quantified by scanning the film and software analysis.