TAR RNA decoys inhibit Tat-activated HIV-1 transcription after preinitiation complex formation
TAR RNA decoys inhibit Tat-activated HIV-1 transcription after preinitiation complex formationPaul R. Bohjanen1,2, Yi Liu1,3 and Mariano A. Garcia-Blanco1,2,3,*
1Department of Pharmacology and Cancer Biology, 2Division of Infectious Diseases, Department of Medicine and 3Department of Microbiology, Levine Science Research Center, Duke University Medical Center, Durham, NC 27710, USA
Received August 8, 1997;Revised and Accepted October 2, 1997
ABSTRACT
The ability of the HIV-1 Tat protein to trans-activate HIV-1 transcription in vitro is specifically inhibited by a circular TAR RNA decoy. This inhibition is not overcome by adding an excess of Tat to the reaction but is partially overcome by adding Tat in combination with nuclear extract, suggesting that TAR RNA might function by interacting with a complex containing Tat and cellular factor(s). A cell-free transcription system involving immobilized DNA templates was used to further define the factor(s) that interact with TAR RNA. Preinitiation complexes formed in the presence or absence of Tat were purified on immobilized templates containing the HIV-1 promoter. After washing, nucleotides and radiolabelled UTP were added and transcription was measured. The presence of Tat during preinitiation complex formation resulted in an increase in the level of full-length HIV-1 transcripts. This Tat-activated increase in HIV-1 transcription was not inhibited by circular TAR decoys added during preinitiation complex formation but was inhibited by circular TAR decoys subsequently added during the transcription reaction. These results suggest that TAR decoys inhibit Tat-activated HIV-1 transcription after preinitiation complex formation, perhaps by interacting with components of transcription complexes.
INTRODUCTION
Human immunodeficiency virus type 1 (HIV-1) uses the host cell's transcription machinery to produce the RNA transcripts necessary for viral replication. Although the cellular transcription machinery can sustain a low basal level of HIV-1 transcription, the viral regulatory protein Tat is required for high level HIV-1 transcription and can activate HIV-1 transcription by >100-fold (1 -3 ). In the absence of Tat most of the RNA transcripts produced from the HIV-1 promoter are short transcripts resulting from early transcription termination, whereas in the presence of Tat the number of long full-length HIV-1 transcripts increases (4 -7 ). Several cellular proteins have been shown to bind to Tat in vitro, including several components of the transcriptional machinery (8 -12 ). Tat has also been shown to associate with the RNA polymerase II holoenzyme (13 ,14 ). The functional significance of these interactions is not clear, but it is possible that transcription complexes that contain Tat are modified such that they are capable of elongation competence. Perhaps Tat increases the elongation competence of transcription complexes by regulating phosphorylation of the C-terminal domain (CTD) of the RNA polymerase II large subunit (10 ,15 ).
Activation of HIV-1 transcription by Tat depends on the presence of an intact TAR RNA element. TAR is a 59 nt RNA stem-loop structure that forms the 5'-end of all HIV-1 transcripts. TAR consists of two stems, a 4 nt bulge and a 6 nt loop (16 -20 ; see Fig. 1 ). Tat binds to the TAR bulge region in vitro (21 -23 ) and may bind this same region in vivo, since mutations in the TAR bulge that inhibit Tat-binding in vitro also abolish Tat-mediated transcriptional activation in vivo (3 ,24 ,25 ). Interestingly, mutations in the TAR loop that have little or no effect on Tat binding in vitro abolish Tat-activated HIV-1 transcription in vivo (21 ,26 -28 ), suggesting that a cellular factor that recognizes the TAR loop is required for Tat function.
MATERIALS AND METHODS
Synthesis of circular TAR RNA decoys
The plasmids pTC, pTC-31/34 and pTC-BL (33 ) were used as templates for synthesis by T7 RNA polymerase of TAR circle RNA, 31/34 circle RNA and bulgeless circle RNA respectively. RNA synthesis and circularization were carried out using the self-splicing activity of a group I permuted intron-exon as described previously (33 ). Circular RNA was gel purified and quantified by measuring the optical density at 260 nm. The circular TAR RNA, 31/34 RNA and bulgeless RNA molecules are depicted in Figure 1 .
In vitro transcription in HeLa nuclear extracts
HeLa nuclear extracts were prepared as previously described (34 ). In vitro transcription reactions were carried out for 30 min at 30oC in a 25 µl volume containing 8-16 µl nuclear extract, 14 mM HEPES, pH 7.9, 14% glycerol, 68 mM potassium chloride, 15 mM sodium chloride, 7 mM magnesium chloride, 4 mM sodium citrate, 250 ng poly(I)-poly(C), 300 ng poly(dI)-poly(dC), 1 mM DTT, 10 mM creatine phosphate, 0.1 mM EDTA, 625 µM each ATP, CTP and GTP, 40 µM UTP and 10 µCi [[alpha]-32P]UTP (3000 Ci/mmol; New England Nuclear). Each reaction also contained 100 ng BamHI-cut plasmid pBC12/HIV/SEAP (35 ), which served as template for transcription from the HIV-1 promoter, and 250 ng AatII-cut plasmid pFLBH (36 ), which served as template for transcription from the adenovirus major late (AdML) promoter. Some reactions contained 50-200 ng Tat protein (37 ) and/or 2-100 pmol circular TAR RNA or mutant TAR RNA. The circular RNA was the last component added to the reactions. The reactions were stopped and newly transcribed RNA was isolated and separated by electrophoresis on 6% urea-polyacrylamide gels as previously described (37 ,38 ). Bands on the gel were visualized by autoradiography and quantified using a Molecular Dynamics phosphorimager.
Transcription from purified preinitiation complexes
Plasmid pBC12/HIV/SEAP (35 ) was amplified from positions -466 to +1125 relative to the start of HIV-1 transcription using PCR. The PCR product was digested with EcoRI at position -466 and then biotinylated by filling in the recessed ends with biotin-14-dATP using the Klenow fragment of DNA polymerase I. To immobilize the DNA template, 250 ng biotinylated DNA was suspended in 25 µl buffer containing PBS supplemented with 0.1% BSA and 0.1% (v/v) Tween 20 and was then added to each well of a streptavidin-coated panel (Labsystems). The panels were rotated on an orbital shaker for 1 h at room temperature and then each well was washed three times with 75 µl buffer D (20 mM HEPES-KOH, pH 7.9, 0.1 M KCl, 0.2 mM EDTA, pH 8.0, 0.5 mM DTT and 20% v/v glycerol). Preinitiation complexes were formed by adding to each well 25 µl pre-mixed sample containing 10 µl HeLa nuclear extract in buffer D supplemented with 4 mM sodium citrate. Some reactions contained 400 ng purified recombinant Tat protein and/or 50 pmol circular TAR decoys or mutant TAR decoys. The reactions were incubated at 30oC for 20 min and then the wells were washed three times with 75 µl buffer D supplemented with 0.1 mg/ml PMSF and 0.1% (v/v) Nonidet P-40. Transcription reactions were then performed by adding to each well 25 µl solution containing 14 mM HEPES, pH 7.9, 14% glycerol, 68 mM potassium chloride, 15 mM sodium chloride, 7 mM magnesium chloride, 4 mM sodium citrate, 250 ng poly(I)-poly(C), 300 ng poly(dI)-poly(dC), 1 mM DTT, 10 mM creatine phosphate, 0.1 mM EDTA, 625 µM each ATP, CTP and GTP, 40 µM UTP and 10 µCi [[alpha]-32P]UTP (3000 Ci/mmol; New England Nuclear). At this point, 50 pmol circular TAR decoy or mutant circular TAR decoy was added to some reactions. After a 9 min incubation at 30oC the reactions were stopped and radiolabelled RNA was isolated and separated by electrophoresis on 6% urea-polyacrylamide gels. Bands on the gel were visualized by autoradiography and quantified on a Molecular Dynamics phosphorimager or a Molecular Dynamics densitometer.
RESULTS AND DISCUSSION
TAR decoys inhibit Tat-activated HIV-1 transcription in an in vitro system involving HeLa nuclear extracts (32 ,33 ). In this system exogenous Tat induced an increase in HIV-1 transcription but had no effect on transcription from the adenovirus major late (AdML) promoter, which was used as an internal control (Fig. 2 , lanes 1 and 2). When an HIV-1 promoter containing the [Delta]TAR mutation (32 ) was used as an internal control the same result was seen; Tat induced an increase in transcription from the wild-type but not the mutant HIV-1 promoter (data not shown). This Tat-dependent increase in HIV-1 transcription was inhibited by circular TAR decoys (Fig. 2 , lanes 3-5) but not by circular TAR decoys containing the 31/34 loop mutation (Fig. 2 , lanes 6-8). The inhibitory activity of circular TAR decoys was specific, because TAR decoys did not inhibit transcription from the AdML promoter (Fig. 2 ) or from an HIV-1 promoter containing the [Delta]TAR mutation (data not shown). Circular rather than linear TAR decoys were used for these experiments because circular TAR decoys were found to be very stable in HeLa nuclear extracts, whereas linear TAR decoys were relatively unstable (33 ).
ACKNOWLEDGEMENTS
We thank Carlos Suñé for his critical review of this manuscript. The Keck Foundation is acknowledged for their generous support of the Levine Science Research Center at Duke University, where this work was performed. P.R.B. was supported by a Howard Hughes Medical Institute Postdoctoral Fellowship for Physicians. Y.L. was supported by a grant from the NIH to M.A.G.-B. This work was also supported by a grant from the VA to M.A.G.-B.
*To whom correspondence should be addressed at: Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, NC 27710, USA. Tel: +1 919 613 8632; Fax: +1 919 613 8646; Email: garci001@mc.duke.edu
