Virus Gene Expression - NCBI

Vol. 66, No. 9

JOURNAL OF VIROLOGY, Sept. 1992, p. 5594-5597

0022-538X/92/095594-04$02.00/0 Copyright © 1992, American Society for Microbiology

Contribution of the TATA Motif to Tat-Mediated Transcriptional Activation of Human Immunodeficiency Virus Gene Expression HENRIK S. OLSEN AND CRAIG A. ROSEN* Department of Gene Regulation, Roche Institute of Molecular Biology, 340 Kingsland Street, Nutley, New Jersey 07110-1199 Received 4 March 1992/Accepted 8 June 1992

Tat-mediated transcriptional activation of human immunodeficiency virus (HIV) gene expression requires the presence of the cis-acting Tat-responsive element, TAR, and a functional enhancer-promoter element. The ability of Tat to function with heterologous enhancer sequences led us to examine the role of the minimal basal promoter for trans activation. Substitution of HIV TATA sequences (nucleotides -20 to -35) with TATA elements derived from other promoters had little effect on the basal level of transcription or the ability to activate the HIV long terminal repeat upon stimulation through upstream activation sequences. In contrast, minimal alterations within the TATA motif had a profound effect on trans activation, as demonstrated by the 3- to 10-fold reduction in activation following expression of Tat. Our findings suggest that minor changes in the TATA motif affect the composition of the initiation-elongation complex and that the composition of this complex is critical for Tat-dependent activation of gene expression. In addition to the structural genes common to all retrovithe human immunodeficiency virus (HIV) genome encodes two essential trans-acting regulatory proteins, Tat (1, 39) and Rev (12, 37), which serve as positive activators of virus gene expression. The Rev protein functions at the posttranscriptional level (7, 28) to regulate the export of structural gene mRNAs from the nucleus to the cytoplasm (10, 15, 16, 23). The mechanism of Tat function appears to be more complex, with transcriptional (18, 19, 21, 26, 34) and posttranscriptional (27, 29) components being considered. Most recent results obtained from in vitro Tat-dependent transcription systems suggest that Tat functions to enhance elongation (11, 20, 22, 24). However, these studies do not rule out an effect at the level of initiation as well. The binding site for Tat, known as TAR (17, 19, 38), comprises an RNA stem-loop structure that encompasses HIV nucleotides (nt) +14 to +45 (33). The demonstration that Tat-mediated trans activation is maintained in the absence of specific HIV upstream activation elements (i.e., NF-KB or SP1 sites) (27, 30) yet markedly reduced when TAR is linked to heterologous basal promoter elements (3, 7, 30) led us to examine the role of the minimal HIV promoter element in Tat function. We focused on the importance of nt -20 to -35, which encompass the TATA motif. The TATA box motif, which is found in many eukaryotic class II promoters (5), is located approximately 25 bp upstream of the transcription initiation site (5). It is believed that in promoters that contain this motif, the TATA element is required for transcriptional initiation as well as for correct positioning of the RNA start site (4, 5, 14, 43). To analyze the importance of the TATA box motif for Tat-dependent trans activation of HIV gene expression, a series of TATA box mutations, schematically illustrated in Fig. 1, were created by using a previously described polymerase chain reaction (PCR)-based mutagenesis procedure (9). The region of the HIV long terminal repeat (LTR) (nt -167 to +81) generated this way includes ruses,

*

Corresponding author. 5594

the basal promoter element, the CAAT box structure, the TATA box motif, and the Tat-responsive element, TAR. This fragment also contains the SP1 and NF-KB binding sites and has previously been shown to be functionally equivalent to the full-length HIV LTR with respect to basal transcription and trans activation (30). The alterations of the TATA motif included TATA motifs found in other viruses, as well as substitution of the HIV TATA element with other basal promoter elements, such as the CAAT box motif (5), the initiator element (36), or the GC box, commonly found in promoters that lack the TATA motif (32). Also included was a consensus TATA motif (5) and a nonsense mutation. The effects of the TATA box mutations were first analyzed in transient gene expression experiments by measuring the basal levels of chloramphenicol acetyltransferase (CAT) gene expression following transfection of the individual CAT reporter plasmids into COS-7 cells. Analysis of CAT activity directed from the different TATA elements (Table 1) revealed comparable levels of basal activity, with variations of no more than twofold. In contrast, the level of CAT activity following cotransfection with a Tat-expressing plasmid, pSVTat, was highly dependent upon the nucleotide sequence of the TATA motif (Table 1 and Fig. 2). The mutations giving rise to the lowest levels of Tat trans activation were the constructs in which the HIV TATA box was replaced with other basal promoter elements, including the initiator element (pH-INR), the CAAT box motif (pHMUT/2), the GC box (pH-GC), and the nonsense mutation (pH-MUT/1). It is most interesting that the mutation that creates a consensus TATA box sequence (pH-Con) also resulted in a low level of Tat-dependent trans activation (fivefold reduction). This sequence contains an A-to-G change 5' to the TATA sequence, as well as a 2-nt change that altered GC to AA. The effect of this mutation was studied in greater detail by creating a TATA motif containing only the 5' point mutation that changed the A to a G nucleotide (pH-G). This mutation alone resulted in a low level of trans activation, comparable to that obtained with the consensus mutation. Plasmids containing alterations that

VOL. 66, 1992

NOTES

-167

+1

NFkB NFkB

SP1 SPI SP1

TATAA

+81

5595

TABLE 1. Transcriptional activity of TATA motif mutationsa

TARA Plasmid

p-167

Plasaid

Origin

p-167

wildtype

pH-Con

Consensus

tgcgtataaaaag

pH-INR

Initiator

tgcactcaatctg

pH-MUT/1

nonsense

tgcagcgcagcag

pH-GC

GC box

tgcagggcggcag

pH-MUT/2

CAAT box

tgcacaatagcag

pH-SV

SV40

tgcatatttatag

pH-AD

Adenovirus

tgcatatagaaag

pH-MuLV/1

MuLV

tgcatataaaaag

pH-MuLV/2

MuLV

tgccaataaaaag

pH-G

point mutation

tgcgtataagcag

pH-CMV

CMV

tctatataagcag

pH-H2

HIV-2

tctgtataaatat

-20 -35 cagatcctgcatataagcagctgc

FIG. 1. Schematic illustration of plasmids used to analyze the effects of TATA motif mutations on HIV-1 promoter function. HIV-1 LTR mutants were generated by PCR (9). Four overlapping oligonucleotides, spanning nt -167 to +81, were used in a PCR reaction. The reaction product from the first round of PCR was used as a template in a second PCR reaction together with 5' and 3' flanking PCR primers containing restriction sites to facilitate cloning of the PCR-generated fragment into CAT reporter BLCATpA. Plasmid BLCATpA contains a multiple restriction cloning site 5' of the CAT gene and an SV40 polyadenylation site 3' to CAT. BLCATpA also contains the SV40 origin of replication. The HIV LTR fragment contains the NF-KB responsive elements, the SP1 binding sites, the TATA box motif, and the Tat-responsive element, TAR. The nucleotide changes in the TATA motif relative to the HIV TATA sequence are indicated by bold lettering. MuLV, murine leukemia virus; CMV, cytomegalovirus. generate TATA box sequences common to other viruses showed various levels of Tat-dependent trans activation

(Table 1). Exchange of the HIV TATA element for the cytomegalovirus (pH-CMV), simian virus 40 (SV40) (pHSV), HIV type 2 (HIV-2) (pH-H2), and adenovirus (pH-AD) TATA box motifs demonstrated reduced levels of Tatmediated trans activation relative to that obtained with the -167 to +81 authentic HIV promoter (p-167). The level of trans activation obtained with the two mutations resembling the TATA box motif found in the murine leukemia viruses (pH-MuLV/1 and pH-MuLV/2) was reduced to approximately 10% of wild-type activity. The above results indicate that Tat-mediated trans activation is highly dependent upon the nucleotide sequence surrounding the TATA motif. This suggested that Tat mediates its effect directly or in concert with additional cellular proteins, through interaction with the initiation-elongation complex that binds to the DNA sequence that encompasses the TATA motif. Alternatively, as trans activation is highly dependent upon the localization of TAR (i.e., TAR must be present at the extreme 5' terminus of the transcript) (33), the effects observed could reflect changes in the site of RNA initiation. To examine this possibility, the transcriptional start site of RNA directed from each of the TATA mutations was examined by primer extension analysis following transfection of the individual plasmid DNAs into COS cells (Fig. 3). RNA was harvested from cells at 48 h posttransfection and hybridized to an end-labelled DNA probe complemen-

pH-Con pH-INR pH-MUT/1 pH-GC pH-MUT/2 pH-SV pH-AD pH-MuLV/1 pH-MuLV/2 pH-G

pH-CMV pH-H2

Basal transcription

Tat inductionb

21.2 12.0 15.0 17.4 12.4 14.0 22.0 35.0 17.2 18.3 31.0 36.6 26.2

125.7 (1.0) 26.0 (0.21) 11.0 (0.09) 18.0 (0.14) 17.7 (0.14) 19.8 (0.16) 68.7 (0.55) 37.2 (0.30) 15.0 (0.12) 19.8 (0.15) 10.2 (0.08) 91.5 (0.73) 94.5 (0.74)

RNA startc

NF-KB induction'

+

5.5 9.6 ND ND 7.4 6.6 4.0 11.0 5.5 4.5 5.6 5.8 3.1

-

+ + + + + + + +

a COS cells were seeded at a concentration of 2.5 x 105/60-mm-diameter dish and transfected on the following day with 0.5 ,ug of plasmid DNA by using the DEAE dextran protocol (8). At 48 h posttransfection, cells were harvested and CAT activity was determined (13). Percent conversion of chloramphenicol to acetylated products in a 90-min reaction denotes the level of basal transcription. b Tat induction values were determined from CAT assays prepared from cells cotransfected with the TATA motif reporter plasmid and Tat-expressing plasmid pSVTat. Fold induction was calculated by measuring the increase in CAT activity obtained in the presence of Tat over that obtained in its absence (basal transcription). The values in parentheses represent fold trans activation relative to that obtained with the authentic HIV promoter (p-167). c The RNA start site was determined as described in the legend to Fig. 3. Plus and minus signs indicate correct and incorrect RNA start sites, respectively. I Fold NF-KB induction was determined by CAT assays prepared from cells cotransfected with chimeric NF-KB-expressing plasmid p5O-p65 (31) and the TATA motif CAT reporters. Fold induction was calculated by measuring the increase in CAT activity obtained in the presence of the NF-KB reporter over that obtained in its absence (1, 2). ND, not done.

tary to nt +50 to +81 of TAR. After extension with reverse transcriptase, labelled products were analyzed on a denaturing gel (Fig. 3). Surprisingly, several of the plasmid DNAs produced RNAs that initiated incorrectly (i.e., +20 nt from the correct initiation site). As it has previously been shown that HIV messages that initiate incorrectly are poorly transactivated, the reduction of trans activation experienced with these plasmids (i.e., pH-INR, pH-Con, pH-MUT/1, and pH-GC) may reflect the altered start site and as such may only be an indirect effect of the TATA substitution. Therefore, as it is impossible to discriminate between these two possibilities, those plasmids which initiated RNA incorrectly

subjected to further analysis. Many of the mutants gave rise to an end-labelled product of the expected size (Fig. 3). These include the CAAT box

were not

mutant (pH-MUT/2), the single-point mutation (pH-G), and the two mutants resembling murine leukemia virus TATA

boxes (pH-MuLV/1 and pH-MuLV/2) (42), all of which exhibited a substantial reduction in the level of trans activation. The level of RNA transcripts generated from these plasmids was comparable to that obtained following transfection with wild-type DNA, which contains the authentic HIV TATA motif. We also note that some of the TATA mutants gave rise to reproducibly higher levels of RNA relative to the wild type and yet were poorly responsive to Tat. Thus, the low level of transactivation observed with these mutants cannot be attributed to improper RNA initiation or a low level of transcription. To determine whether the alterations in the HIV TATA region influenced other aspects of HIV promoter-enhancer function, the ability to stimulate gene expression through the upstream NF-KB motif was examined. For this study, Jurkat

5596

J. VIROL.

NOTES

aL.

1.0 0

' :0.8

0.4 %°X0.6

00

0.0 0 ~ co

I:

I:

I

~ >

~

~

~ I

t

~

~~ 9

I

0. 0.

FIG. 2. Tat responsiveness of TATA motif mutations. COS cells were seeded at a density of 2.5 x 105/35-mm-diameter dish and transfected on the following day with 0.5 ,ug of a CAT reporter plasmid and 0.5 p.g of Tat-expressing vector pSVTat. At 48 h posttransfection, cells were harvested and the percentage of chloramphenicol converted to acetylated products was determined. The activity of the TATA motif mutants is relative to the activity obtained with the wild-type construct, which was assigned an arbitrary value of 1.

T lymphocytes were cotransfected with the indicator plasmids and a vector expressing a chimeric fusion protein containing the DNA-binding domain of the p50 subunit and the transcriptional activator or the p65 subunit of NF-KB (31). To further enhance the transcriptional activity through the NF-KB motif, the cells were stimulated with phorbol 12-myristate 13 acetate at 24 h after transfection. Expression of the chimeric p50-p65 protein in these cells resulted in 5- to 10-fold stimulation of CAT activity (Table 1). No significant differences in activation between the wild-type LTR and those mutants that demonstrated diminished responsiveness

1

2

3

4

5

6

7

8

9

10

FIG. 3. Mapping of the transcriptional start site for the TATA motif mutants. COS cells were seeded at a density of 5 x 106/150mm-diameter plate and transfected on the following day with a 5-,ug sample of plasmid DNA harboring the wild-type HIV TATA sequence or the TATA motif mutants. RNA was harvested at 48 h

posttransfection by usingRNAzolB (Cinna Biotex), and 20 ,g was hybridized at 45°C to a32P-end-labelled primer complementary to nt +50 to +81 in the HIV LTR. Following hybridization, the primer was extended with reverse transcriptase and the products were analyzed on a 10% denaturing polyacrylamide gel. The results shown reflect primer extension analysis of RNAs from mocktransfected cells (lane 1) and from cells transfected with p-167 (lane 2), pH-MUT/2 (lane 3), pH-SV (lane 4), pH-AD (lane 5), pHMuLV/1 (lane 6), pH-MuLV/2 (lane 7), pH-G (lane 8), pH-CMV (lane 9), or pH-H2 (lane 10). Also shown is a sequencing reaction of the HIV LTR, with an arrow denoting the position of the three G nucleotides that encompass the transcriptional start site.

to Tat were observed. Therefore, it appears that NF-KBmediated activation is independent of the exact nucleotide sequence of the TATA element. The aim of the present study was to examine the requirement for a specific nucleotide sequence encompassing the TATA element for Tat-mediated trans activation of the HIV-1 promoter. Our findings demonstrate that minor alterations of the nucleotide sequence around the TATA motif can elicit a significant reduction in the level of Tat-mediated trans activation. Analysis of the RNA transcribed from the altered HIV-1 promoter elements argues against the possibility that the reduced level of trans activation reflects a low level of RNA expression or generation of an aberrant transcriptional start site. We favor the possibility that alterations of the HIV TATA element result in the formation of different transcriptional activation complexes at the promoter. This model predicts that Tat-mediated trans activation is at least partly a result of an interaction of Tat or a Tat-associated factor with the transcriptional activation complex and that trans activation is highly dependent upon the precise composition of this complex. Consistent with this hypothesis, recent studies demonstrate that different TATA box elements from eukaryotic class II promoters are functionally distinct (2, 35, 41). Substitution of the sequence TATAAA with the SV40 TATA element results in loss of ElA inducibility of the HSP70 promoter (35), and substitution of the sequence TATAAA with the SV40 TATA element results in loss of tissue specificity of the muscle enhancer element (41). These studies suggest that the TATA box motifs are functionally heterogeneous and that the heterogeneity is likely due to differential binding of distinct TATA box binding factors in eukaryotic cells. This hypothesis has been supported by biochemical characterization of protein fractions participating in binding to the TATA box element (6, 25, 40). For example, it has been shown that a family of proteins binds cooperatively to TFIID and that this group of proteins differs from TFIIA and competes with TFIIA for binding to TFIID, suggesting that the preinitiation complex formed at the TATA box motif is indeed heterogeneous in composition (25). In a recent study by Berkhout and Jeang, it was shown that a minimal fragment of the HIV-1 promoter spanning nt -43 to +80 is sufficient for basal-level expression and the presence of additional enhancer elements, such as SP1 and NF-KB, is necessary for Tat-induced trans activation (3). These investigators also observed that alterations of sequences near and including the TATA box motif within the context of the HIV promoter reduced Tat activation significantly. In agreement with their findings, we have shown that minor alterations (i.e., 1 or 2 nt) in only the TATA motif that do not alter the adjacent sequence or the context of the TATA element have profound effects on the ability of Tat to transactivate the LTR. The recent biochemical characterization of proteins involved in the formation of the transcriptional initiation complex (6, 25, 40) and the establishment of in vitro Tat-dependent transcription systems (20, 24) should aid in elucidating the role of cellular factors in Tat function. We thank P. Dillon for the BLCATpA plasmid and T. Rose for preparation of the manuscript. This study was supported in part by a National Cooperative Drug Discovery Award to C.A.R. REFERENCES 1. Arya, S. K., C. Guo, S. F. Josephs, and F. Wong-Staal. 1985. Trans-activator gene of human T-lymphotrophic virus type III (HTLV-III). Science 229:69-73.

VOL. 66, 1992 2. Babiss, L., and L. Vales. 1991. Promoter of the adenovirus polypeptide IX gene: similarity to E1B and inactivation by substitution of the simian virus 40 TATA element. J. Virol. 65:598-605. 3. Berkhout, B., and K.-T. Jeang. 1992. Functional roles for the TATA promoter and enhancers in basal and Tat-induced expression of the human immunodeficiency virus type 1 long terminal repeat. J. Virol. 66:139-149. 4. Boyer, T., J. Krug, and L. Maquat. 1989. Transcriptional regulatory sequences of the housekeeping gene for human triosephosphate isomerase. J. Biol. Chem. 264:5177-5187. 5. Breathnach, R., and P. Chambon. 1981. Organization and expression of eucaryotic split genes coding for proteins. Annu. Rev. Biochem. 50:349-383. 6. Cortes, P., 0. Flores, and D. Reinberg. 1992. Factors involved in specific transcription by mammalian RNA polymerase II: purification and analysis of transcription factor IIA and identification of transcription factor IIJ. Mol. Cell. Biol. 12:413-421. 7. Cullen, B. R. 1986. Trans-activation of human immunodeficiency virus occurs via a bimodal mechanism. Cell 46:973-982. 8. Cullen, B. R. 1988. Use of eukaryotic expression technology in the functional analysis of cloned genes. Methods Enzymol. 152:684-704. 9. Dillon, P. J., and C. A. Rosen. 1990. A rapid method for the construction of synthetic genes by the polymerase chain reaction. BioTechniques 9:298-300. 10. Emerman, M., R. Vazeux, and K. Peden. 1989. The rev gene product of the human immunodeficiency virus affects envelopespecific RNA localization. Cell 57:1155-1165. 11. Feinberg, M. B., D. Baltimore, and A. D. Frankel. 1991. The role of Tat in the human immunodeficiency virus life cycle indicates a primary effect on transcriptional elongation. Proc. Natl. Acad. Sci. USA 88:4045-4049. 12. Feinberg, M. B., R. F. Jarrett, A. Aldovini, R. C. Gallo, and F. Wong-Staal. 1986. HTLV-III expression and production involve complex regulation at the levels of splicing and translation of viral RNA. Cell 46:807-817. 13. Gorman, C. M., L. F. Moffat, and B. H. Howard. 1982. Recombinant genomes which express chloramphenicol acetyltransferase in mammalian cells. Mol. Cell. Biol. 2:1044-1051. 14. Grosveld, G. C., C. K. Shewmaker, P. Jat, and R. A. Flavell. 1981. Localization of DNA sequences necessary for transcription of the rabbit P-globin gene in vitro. Cell 25:215-226. 15. Hadzopoulou-Cladaras, M., B. K. Felber, C. Cladaras, A. Athanassopoulos, A. Tse, and G. N. Pavlakis. 1989. The rev (trs/art) protein of human immunodeficiency virus type 1 affects viral mRNA and protein expression via a cis-acting sequence in the env region. J. Virol. 63:1265-1274. 16. Hammarskjold, M. L., J. Heimer, B. Hammarsljdld, I. Sangwan, L. Albert, and D. Rekosh. 1989. Regulation of human immunodeficiency virus env expression by the rev gene product. J. Virol. 63:1959-1966. 17. Hauber, J., and B. Cullen. 1988. Mutational analysis of the transactivation-responsive region of the human immunodeficiency virus type 1 long terminal repeat. J. Virol. 62:673-679. 18. Hauber, J., A. Perkins, E. P. Heimer, and B. R. Cullen. 1987. Trans-activation of human immunodeficiency virus gene expression is mediated by nuclear events. Proc. Natl. Acad. Sci. USA

84:6364-6368. 19. Jakobovits, A., D. H. Smith, E. B. Jakobovits, and D. J. Capon. 1988. A discrete element 3' of human immunodeficiency virus 1 (HIV-1) and HIV-2 mRNA initiation sites mediates transcriptional activation by an HIV trans activator. Mol. Cell. Biol. 8:2555-2561. 20. Kato, H., H. Sumimoto, C.-H. Chen, C. Rosen, and R. Roeder. 1992. HIV-1 Tat acts as a processivity factor in vitro: implications for the mechanism. Genes Dev. 6:655-666. 21. Laspia, M., A. Rice, and M. Mathews. 1989. HIV-1 tat protein increases transcriptional initiation and stabilizes elongation. Cell 59:283-292. 22. Laspia, M. F., A. P. Rice, and M. B. Mathews. 1990. Synergy between HIV-1 Tat and adenovirus ElA is principally due to stabilization of transcriptional elongation. Proc. Natl. Acad.

NOTES

5597

Sci. USA 4:2397-2408. 23. Malim, M. H., J. Hauber, S.-Y. Le, J. V. Maizel, and B. R. Cullen. 1989. The HIV-1 rev transactivator acts through a structured target sequence to activate nuclear export of unspliced viral mRNA. Nature (London) 338:254-257. 24. Marciniak, R., B. Calnan, A. Frankel, and P. Sharp. 1990. HIV-1 Tat protein trans-activates transcription in vitro. Cell 63:791-802. 25. Meisterernst, M., and R. Roeder. 1991. Family of proteins that interact with TFIID and regulate promoter activity. Cell 67:557567. 26. Muesing, M. A., D. H. Smith, and D. J. Capon. 1987. Regulation of mRNA accumulation by a human immunodeficiency virus trans-activator protein. Cell 48:691-701. 27. Peterlin, B. M., P. A. Luciw, P. J. Barr, and M. D. Walker. 1986. Elevated levels of mRNA can account for the transactivation of human immunodeficiency virus. Proc. Natl. Acad. Sci. USA 83:9734-9738. 28. Rosen, C. 1991. Regulation of HIV gene expression by RNAprotein interactions. Trends Genet. 7:9-14. 29. Rosen, C. A., J. G. Sodroski, W. C. Goh, A. L. Dayton, J. Lippke, and W. A. Haseltine. 1986. Post-transcriptional regulation accounts for the trans-activation of the human T-lymphotropic virus type III. Nature (London) 319:555-559. 30. Rosen, C. A., J. G. Sodroski, and W. A. Haseltine. 1985. The location of cis-acting regulatory sequences in the human T cell lymphotropic virus type III (HTLV-III/LAV) long terminal repeat. Cell 41:813-823. 31. Ruben, S. M., R. Narayanan, J. F. Klement, C.-H. Chen, and C. Rosen. 1992. Functional characterization of the NF-KB p65 transcriptional activator and an alternatively spliced derivative. Mol. Cell. Biol. 12:444 454. 32. Sehgal, A., N. Patil, and M. Chao. 1988. A constitutive promoter directs expression of the nerve growth factor receptor gene. Mol. Cell. Biol. 8:3160-3167. 33. Selby, M. J., E. S. Bain, P. A. Luciw, and B. M. Peterlin. 1989. Structure, sequence, and position of the stem-loop structure in tar determine transcriptional elongation by tat through the HIV-1 long terminal repeat. Genes Dev. 3:547-558. 34. Sharp, P., and R. Marciniak. 1989. HIV TAR: an RNA enhancer? Cell 59:229-230. 35. Simon, M. C., T. Fisch, B. Benecke, J. Nevins, and N. Heintz. 1988. Definition of multiple, functionally distinct TATA elements, one of which is a target in the hsp7o promoter for ElA regulation. Cell 52:723-729. 36. Smale, S., M. Schmidt, A. Berk, and D. Baltimore. 1990. Transcriptional activation by Spl as directed through TATA or initiator: specific requirement for mammalian transcription factor IID. Proc. Natl. Acad. Sci. USA 87:4509-4513. 37. Sodroski, J., W. C. Goh, C. Rosen, A. Dayton, E. Terwilliger, and W. A. Haseltine. 1986. A second post-transcriptional transactivator gene required for HTLV-III replication. Nature (London) 321:412-417. 38. Sodroski, J., R. Patarca, C. Rosen, F. Wong-Staal, and W. Haseltine. 1985. Location of the trans-activating region on the genome of human T-cell lymphotropic virus type III. Science 229:74-77. 39. Sodroski, J., C. A. Rosen, and W. A. Haseltine. 1984. Trans-acting transcriptional activation of the long terminal repeat of human T lymphotropic viruses in infected cells. Science 225:381-385. 40. Timmers, H. T., and P. Sharp. 1991. The mammalian TFIID protein is present in two functionally distinct complexes. Genes Dev. 5:1946-1956. 41. Wefald, F., B. Devlin, and R. S. Williams. 1990. Functional heterogeneity of mammalian TATA-box sequences revealed by interaction with a cell-specific enhancer. Nature (London) 344: 260-262. 42. Weiss, R., N. Teich, H. Varmus, and J. Coffin (ed.). 1985. RNA tumor viruses. Molecular biology of tumor viruses, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 43. Wu, L., D. Rosser, M. Schmidt, and A. Berk. 1987. A TATA box implicated in ElA transcriptional activation of a simple adenovirus 2 promoter. Nature (London) 326:512.