Human immunodeficiency virus type 1 Tat increases the expression of cleavage and polyadenylation specificity factor 73-kilodalton subunit modulating cellular and viral expression

Abstract

The human immunodeficiency virus type 1 (HIV-1) Tat protein, which is essential for HIV gene expression and viral replication, is known to mediate pleiotropic effects on various cell functions. For instance, Tat protein is able to regulate the rate of transcription of host cellular genes and to interact with the signaling machinery, leading to cellular dysfunction. To study the effect that HIV-1 Tat exerts on the host cell, we identified several genes that were up- or down-regulated in tat-expressing cell lines by using the differential display method. HIV-1 Tat specifically increases the expression of the cleavage and polyadenylation specificity factor (CPSF) 73-kDa subunit (CPSF3) without affecting the expression of the 160- and 100-kDa subunits of the CPSF complex. This complex comprises four subunits and has a key function in the 3'-end processing of pre-mRNAs by a coordinated interaction with other factors. CPSF3 overexpression experiments and knockdown of the endogenous CPSF3 by mRNA interference have shown that this subunit of the complex is an important regulatory protein for both viral and cellular gene expression. In addition to the known CPSF3 function in RNA polyadenylation, we also present evidence that this protein exerts transcriptional activities by repressing the mdm2 gene promoter. Thus, HIV-1-Tat up-regulation of CPSF3 could represent a novel mechanism by which this virus increases mRNA processing, causing an increase in both cell and viral gene expression.

FIG. 1.

HIV-1 Tat protein increases the expression of CPSF3. (A) Identification of a 632-bp gene fragment up-regulated in K562-Tat cells. To carry out the differential display assays, we used four similar samples of the same cell to minimize artifactual amplification. Thus, two different extractions of mRNA (first extraction, lanes 1 and 3; second extraction, lanes 2 and 4) were subjected to two different RT-PCRs (lanes 1 and 2 and lanes 3 and 4). A differentially amplified 632-bp fragment (arrow) separated on 4.5% denaturing polyacrylamide gels was identified by autoradiography, excised from the gel, reamplified, cloned, and sequenced. (B) Tat increases the expression of CPSF3 in tat-transfected cells. Total RNA was extracted from HeLa (lanes 1), HeLa-Tat (lanes 2), Jhan (lanes 3), Jhan-Tat (lanes 4), K562 (lanes 5), and K562-Tat (lanes 6) cells, and the expression of CPSF3 and β-actin was detected by RT-PCR.

FIG. 2.

HIV-1 Tat protein specifically up-regulates the 73-kDa subunit of the CPSF complex in both K562 and HeLa cells. (A) Total RNA was extracted from K562-pcDNA₃, K562-Tat, HeLa, and HeLa-Tat cells, and the expression of three CPSF subunits was studied by semiquantitative RT-PCR. β-Actin amplification was carried out in parallel as a control. (B) Quantitative real-time RT-PCR of CPSF subunit expression in K562-pcDNA₃, K562-Tat, HeLa, and HeLa-Tat cells. Amplification was normalized against β-actin gene expression, and fold increases were calculated by using the comparative threshold cycle method for quantification. The results shown are representative of those from three experiments. Error bars indicate standard deviations.

FIG. 3.

CPSF3 protein induction by HIV-1 Tat. (A) Total cell extracts were obtained from K562-pcDNA₃, K562-Tat, HeLa, and HeLa-Tat cells, and CPSF3 protein expression was detected by Western blotting. (B) Cellular proteins were obtained from cycloheximide (CHX)-treated HeLa and HeLa-Tat cells at the indicated times, and the levels of CPSF3 protein were analyzed by Western blotting. (C) HeLa-On-Tat cells were stimulated with doxycycline (Dox) (1 μg/ml) for 1, 3, 6, and 12 h; total proteins were extracted; and the expression of CPSF3 was identified by immunoblotting. In all the cases the same membranes were reblotted with a monoclonal antibody recognizing α-tubulin. (D) HIV-LTR induction in HeLa-On-Tat cells by doxycycline. The cells were transfected with the HIV-LTR-Luc plasmid and 24 h later were stimulated with doxycycline as indicated. After this time, the cells were lysed and the luciferase activity was measured as described in Material and Methods. The results shown are representative of those from three experiments. RLU, relative light units. Error bars indicate standard deviations.

FIG. 4.

Effects of CPSF3 overexpression on viral and cellular gene expression. 293T and HeLa cells were transiently transfected with the pNL4-3.Luc.R-E, p21-Luc, or mdm2-Luc plasmid along with either the expression vector encoding CPSF3 (pcDNA₃-CPSF3) (gray bars) or the empty pcDNA₃ plasmid (black bars). After 48 h of transfection, luciferase activity was assayed. The results show mean fold induction ± standard deviation from three different experiments.

FIG. 5.

Specific inhibition of CPSF3 mRNA expression by siRNA. 293T cells were transiently transfected with either the pSI-CPSF3 plasmid (lanes 2) or the pSI-Control plasmid (lanes 1). Forty-eight hours later, total RNA was extracted and CPSF3 expression was studied by RT-PCR, using the β-actin gene expression as a control. The change in expression was analyzed by using image analysis software (Kodak digital science 1D, version 2.01) and expressed as relative units. Results from one representative experiment out of four independent experiments are shown. Error bars indicate standard deviations.

FIG. 6.

Effects of CPSF3 silencing on viral and cellular gene expression. 293T and HeLa cells were transiently cotransfected with the indicated luciferase reporter vectors and either the pSI-CPSF3 plasmid (gray bars) or the pSI-Control plasmid (black bars) (silencer/target plasmid ratio of 10:1; 1 μg of DNA total). After 48 h of transfection, luciferase activity was assayed. Each transfection was assayed in triplicate, and the results show mean percent activation ± standard deviation.

FIG. 7.

Long-term inhibition of CPSF3 subunit expression. (A) Total RNA was isolated from either 293T-siControl (lanes 1) or 293T-siCPSF3 cells (lanes 2), and the expression of CPSF1, -2, and -3 was analyzed by RT-PCR, using β-actin gene expression as a control. (B) Phenotype changes in 293T-siCPSF3 clones were monitored by light (A and B) and scanning electron (C and D) microscopy in comparison to the parental controls. Corresponding images were taken at the same exposure. Magnifications, ×400 (light microscopy [LM]) and ×1,600 (scanning electron microscopy [SEM]).

FIG. 8.

Effects of CPSF subunit specific inhibition on viral and cellular gene expression. 293T cells were transiently cotransfected with the indicated plasmids along with either each CPSF subunit silencer plasmid (pSI-CPSF1 [dark gray bars], pSI-CPSF2 [light gray bars], or pSI-CPSF3 [white bars]) or the pSI-Control plasmid (black bars) (silencer/target plasmid ratio of 10:1; 1 μg of DNA total). The luciferase activity was measured and expressed as percent activation. Values are means ± standard deviations from three independent experiments.