OUP user menu

HIV Tat, its TARgets and the control of viral gene expression

Claudio Brigati, Mauro Giacca, Douglas M. Noonan, Adriana Albini
DOI: http://dx.doi.org/10.1016/S0378-1097(03)00067-3 57-65 First published online: 1 March 2003


The human immunodeficiency virus (HIV-1) (transactivator of transcription (Tat)) protein is a pleiotropic factor that induces a broad range of biological effects in numerous cell types. At the HIV promoter, Tat is a powerful transactivator of gene expression, which acts by both inducing chromatin remodeling and by recruiting elongation-competent transcriptional complexes onto the viral LTR. Besides these transcriptional activities, Tat is released outside the cells and interacts with different cell membrane-associated receptors. Finally, extracellular Tat can be internalized by cells through an active endocytosis process. Here we discuss some of the molecular mechanisms involved in intracellular and extracellular Tat function.

  • Human immunodeficiency virus
  • Transactivator of transcription
  • Transactivation
  • Lambda phage
  • Pathogenesis

1 Overview of Tat functions

The human immunodeficiency virus (HIV) transactivator of transcription (Tat) is a small polypeptide (101 amino acids in most clinical HIV-1 isolates, 86 amino acids in the laboratory HIV-1hxb2 strain) essential for efficient transcription of viral genes and for viral replication. At the HIV promoter, the protein binds a structured RNA element (TAR, transactivation-responsive region) present at the 5′-end of viral leader mRNAs (nucleotide position +1 to +59) [1]. Through this interaction, Tat recruits to the viral promoter a diverse series of transcriptional complexes, including enzymes with histone and factor acetyl transferase (HAT and FAT respectively) activity, which modify chromatin conformation at the proviral integration site, and a protein complex (P-TEFb) that phosphorylates the carboxy-terminal domain of RNA polymerase II, thus promoting elongation of viral RNA transcription. Another mechanism of Tat transactivation involves the direct, TAR-independent activation of NF-κB [2], which, in unstimulated cells, is retained in the cytoplasm through its interaction with the inhibitor protein IκB-α. Protein members of the Rel/NF-κB family are involved in the transcriptional activation of several viral genes of T- and non-T-cells and bind to the enhancer element of the viral LTR [3,4]. These three transcriptional activities of Tat account for a tremendous (several hundred-fold) increase in proviral transcription rate.

Unique among transcriptional activators of any species, Tat protein is also released from Tat expressing cells into the extracellular milieu through a non-canonic, Golgi-independent pathway of secretion. Extracellular Tat binds to specific receptors on the cell membrane, triggering different signal transduction pathways. In addition, extracellular Tat, a heparin/heparan sulfate binding protein [5], can be readily internalized by most mammalian cell types through a pathway involving its interaction with cell surface proteoglycans containing heparan sulfate glycosaminoglycans [5,6].

Work conducted in the last years by several laboratories has substantially contributed to our understanding of the molecular mechanisms involved in these pleiotropic activities of the protein. Here we will discuss some of these mechanisms, with particular reference to those involved in control of transcriptional elongation and on the consequence of Tat function on cellular genes.

2 Tat–TAR interaction and the similarity to bacteriophage lambda transcriptional regulation

One of the most striking effects of Tat–TAR interaction is a remarkable increase in transcriptional processivity of the HIV provirus. Upon cell transfection with plasmids carrying the HIV LTR, the populations of transcripts which were obtained differed dramatically depending on the presence of Tat in the cell nucleus. In the absence of Tat, short, non-polyadenylated RNAs terminating at the end of the TAR stem predominated. In the presence of Tat, a remarkable increase in longer, polyadenylated transcripts was readily observed, with a consequent dramatic increase of gene expression [710].

Transcriptional control by the regulation of RNA elongation is also widely used in the microbial world. Phage lambda, for example, uses a mechanism called anti-termination to regulate transcription of early and delayed genes by the use of a single promoter. Obviously, this control is strictly dependent on the relative position of the genes on the genome, with the immediate early and delayed early genes being adjacent, separated only by a terminator. The terminator sequence is fairly typical, and similar to that present on many bacterial operons. The system is energetically economical, since it does not necessitate the synthesis of new enzymes or other factors, and it allows quick and coordinated expression of genes committed to a specific goal. Moreover, it is specific and does not cause generalized anti-termination on host genes.

In the case of phage lambda, a multiprotein complex centered on a protein known as the N protein, interacts with the transcribing RNA polymerase permitting it to continue the nascent chain elongation beyond the termination sites. Transcription through the Nut (N utilization) sites produces an RNA stem–loop structure bound directly by N; this allows the assembly of a complex composed by N and several host proteins, including NusA and NusB, forming a large complex with RNA polymerase. The polymerase is then enabled to overcome the termination activity of the Rho factor at distal sites [11].

In the HIV provirus, the cis-acting RNA target of Tat is TAR, a structured region located at the 5′-end of the viral RNAs and composed of a stem, a bulge and a loop. At a first glance, the TAR element has many features very similar to terminators or attenuator sequences found on phage and bacterial operons. As compared to lambda, TAR can be viewed as analogous to the Nut site, since the two entities share similar structural features typical of hairpin structures: high GC content, and dyad symmetry. The similarities between the two systems also extend beyond the nucleic acid target. For instance, the RNA binding domain of the N protein, similar to Tat, contains an arginine-rich region; moreover, Spt5, a component of the cyclin/CSK9 complex, bears a remarkable homology to NusG. Finally, Tat and cyclin T1 — the protein binding to the loop structure, see below — bind TAR RNA cooperatively, much like the interaction of lambda N protein and Escherichia coli NusA, NusB and S10 with Nut RNA (Fig. 1).

Figure 1

A comparison between the λ phage (upper panel) and HIV (lower panel) transcription control termination mechanisms. Diagrammed in the upper panel is a Rho-dependent site, although the anti-termination could also occur at Rho-independent sites.

The Tat:TAR interaction has been studied functionally and structurally by several means, including circular dichroism and absorption spectroscopy (reviewed in [12,13]). The critical area of TAR for transcriptional activation is the upper part of the structure which harbors a U-rich bulge near the apex of the stem. Accordingly, Tat:TAR interactions are mainly sensitive to mutations in the bulge area. As far as Tat is concerned, the TAR-interacting surface on Tat is within a region that is rich in basic amino acids and contains an arginine-rich motif (ARM). However, remarkable differences are noticeable in the different lentiviruses so far analyzed in the TAR-interacting domains [14]. For example, while in HIV-1 Tat arginine 52 makes the key contribution for binding, bovine immunodeficiency virus (BIV) Tat makes essential RNA contacts also with three glycines, one threonine and one isoleucine [15].

Mutations in the apical loop of HIV-1 TAR do not interfere with Tat binding; however, these mutations have a dramatic effect on Tat transactivation. On this basis, several studies had focused on a search for loop-specific binding factors, culminating in the identification of cyclin T1 as the factor that interacts with the activation domain of Tat and binds the TAR loop [16]. Also in this respect, notable variations are found in the different viruses so far analyzed. Whereas HIV and SIV Tat appear to require cyclin T1 for high affinity TAR binding, the bovine immunodeficiency virus (BIV) and the Jembrana disease virus (JDV) Tat proteins bind TAR in the absence of the cyclin. Moreover, JDV can specifically recognize two different TAR sequences, those from HIV and BIV, using two structural adaptations [17]. Generally, Tat appears to be a very flexible protein. For example, BIV Tat has been observed to switch from a looser structure allowing for several non-specific nucleic acid contacts to a defined β ribbon conformation only upon TAR binding [18]. Interestingly, this structural motif bears similarities to the 17-residue β ribbon structure used by the bacterial Met repressor for recognition of target DNA [19]. This structural flexibility is also observable in the bacteriophage N protein, which, either free in solution or as a complex with non-specific RNA, lacks recognizable structure and binds RNA sequences indiscriminately. In contrast, bound N adopts an ordered α-helical structure and binds its target structure a thousand-fold more tightly (with a Kd approximately 10−9 M).

The plasticity of Tat (and N) structure is paralleled by the plasticity of their target RNAs, which can adopt alternative structures allowing for mutual protein-RNA accommodation in search of an optimum interaction. Productive conformational modifications by HIV-1 TAR seem to be mainly confined to the U23 position in the bulge by one of arginines in the ARM of Tat. This is thought to create a suitable pocket and to twist critical phosphates on the TAR backbone in a way that they are recognized by additional Tat basic amino acids [20]. Notably, in phage lambda, Nut RNA looping and conformational flexibility is thought to contribute to the delivery of N to the transcriptional complex [21], conferring, as for Tat, a rather transient nature of nucleic acid binding.

Finally, several variations are found among lentiviruses in the mechanisms of Tat–TAR interaction and function. For instance, structural data on the equine infectious anemia virus (EIAV) Tat protein reveal a helix-loop-helix-turn-helix structure very similar to homeobox domains known to bind specifically to DNA. EIAV Tat, although containing a fairly typical arginine-rich peptide, is capable of specifically binding DNA at the long terminal repeat Pu.1 and AP-1 sites [22]. Thus, in other species Tat could have features of a DNA binding factor not apparent in humans, posing additional interesting evolutionary questions. In addition, the Tat proteins of other lentiviruses, such as the feline immunodeficiency virus (FIV) Orf-A protein, show little amino acid similarity to HIV-1 Tat, do not appear to bind to a cognate TAR sequence, and act as poor transactivators [23,24]. This raises additional interesting questions on their actual function and on the possible evolutionary relationship with HIV-1 Tat.

3 P-TEFb and the control of RNA polymerase II processivity

Originally the presence of short transcripts in cells lacking Tat was interpreted as the proof that TAR is a terminator sequence, forcing the detachment of the polymerase from the template [25], or imposing a pause site on transcription by determining polymerase stalling [26]. However, as yet there is no evidence that TAR acts as a true terminator for viral transcription, or that Tat works as an anti-terminator.

In contrast, the presence of short transcripts in the absence of Tat derives from the recruitment to the LTR promoter of transcriptional complexes containing a poorly processive RNA polymerase II. Control of the intrinsic processivity of RNA polymerase II in mammalian cells arises from a complex interplay between negative transcription elongation factors, such as DSIF and NELF, and kinase complexes that phosphorylate the carboxy-terminal domain (CTD) of the polymerase and regulate promoter clearance (reviewed in [27]. RNA Pol II with hypophosphorylated CTD initiates transcription on the HIV promoter but is bound to pause and stop after transcribing 30–50 bases (which correspond to the TAR sequence). Phosphorylation of the CTD prevents transcriptional pausing and promotes polymerase processivity [28].

One of the kinase complexes that phosphorylate the RNA Pol II CTD in mammalian cells is the positive transcription elongation factor b (P-TEFb), composed of a cyclin T and the CKD9 kinase. Recruitment of P-TEFb to TAR is an essential component in the mechanism by which HIV Tat transactivates viral gene expression. Tat has been originally shown to co-purify with a nuclear Tat-associated kinase (TAK) [29], which corresponds to the kinase subunit of the P-TEFb complex [30,31]. The catalytic subunit of this Cdc2-related kinase was identified as PITALRE/CDK9 [32]. Finally, Tat has been shown to specifically bind cyclin T1, the cyclin partner of the CDK9 kinase [16]. This interaction, in turn, strongly enhances the affinity and specificity of Tat:TAR binding. Among the different cyclins interacting with CDK9, including cyclin T1, cyclin T2a, cyclin T2b and cyclin K, Tat selectively binds cyclin T1 [33,34]. By fluorescent resonance energy transfer (FRET) studies in living cells, Tat was visualized to associate cyclin T1 in the nucleoplasm, and to direct the protein outside of the nuclear foci where it normally resides [35].

Recruitment of cyclin T1 to promote transcriptional elongation appears a very conserved feature among even very divergent lentiviral Tat proteins [36]. Support for the essential role of cyclin T1 recruitment by Tat also comes from the observation that Tat is a poor transactivator in rodent cells, in which cyclin T1 lacks a critical cysteine residue which, at position 261 in human cyclin T1, mediates the interaction with Tat [37,38].

The P-TEFb complex is regulated by its association with an antagonistic RNA known as 7SK [39,40]. This RNA was discovered as an abundant small nuclear RNA almost 30 years ago, but its role has remained elusive [41]. Recently it has been shown that 7SK acts as a negative regulator of P-TEFb, and that this activity is influenced by different stress response pathways. In contrast to the smaller P-TEFb complexes, which have a high kinase activity, the larger 7SK/P-TEFb complexes phosphorylate weakly. Inhibition of cellular transcription by chemical agents or ultraviolet irradiation triggers the complete disruption of the P-TEFb/7SK complex, and enhances CDK9 activity. The transcription-dependent interaction of P-TEFb with 7SK may therefore contribute to an important feedback loop modulating the activity of RNA Pol II.

Whether P-TEFb is the only protein kinase complex recruited by Tat to the LTR promoter to promote transcriptional processivity is still unclear. Other studies have shown that also the Cdk7-cyclin H-Mat1 (CAK) subunit of basal transcription factor TFIIH, an integral component of the RNP holoenzyme, biochemically associates Tat. To what extent the CDK9 and CDK7 kinase activities might be differentially required in vivo in different cells or under different activation circumstances is still a matter for investigation (reviewed in [42]).

Recently, Tat-mediated control of processivity has been linked to its capacity to stimulate TAR-independent cotranscriptional capping of HIV mRNA [43] during Sp5-induced stalling at promoter-proximal sites by the RNA Pol II. Thus the arrest and restart promoted by different complexes could be now interpreted to ensure proper cap formation in nuclear transcripts. This view of capping in an elongation checkpoint, and the precise role played by Tat, warrants further investigation.

4 Interaction of Tat with protein acetylating enzymes

Once integrated into the host cell genome, the HIV-1 provirus is packaged into chromatin and nucleosomes are deposited at specific positions within the promoter region (reviewed in [44]). In particular, the position of a nucleosome immediately downstream of the transcription start site correlates with a strong repression of transcriptional initiation and might account for the extremely weak promoter activity of the HIV-1 LTR in the absence of Tat transactivation. Consistently, when transcription is activated by Tat, the chromatin associated with sequences immediately downstream of the transcription start site becomes accessible to nucleases [45].

Tat-induced chromatin remodeling is mediated by the functional association of the protein with cellular enzymes able to acetylate histones. These HATs include the transcriptional co-activator p300/CBP [46,47], the p300/CBP-associated protein P/CAF [48], and the Tip60 [49] and hGCN5 [50] acetylases. Complexes containing HATs assist transcriptional activation by modulating nucleosomal repression of specific promoters through acetylation of the N-terminal tails of histones, followed by destabilization of histone–DNA interactions [5153]. Chromatin immunoprecipitation experiments at the HIV-1 promoter indicate that transcriptional activation by Tat induces recruitment of p300/CBP to the LTR and acetylation of histones in this region [46].

In addition to histones, HIV-1 Tat itself is a substrate for acetylation by p300/CBP and the associated protein P/CAF [5457], and by hGCN5 [50]. Lysines at positions 50 and 51 are major substrates for acetylation by p300 and hGCN5, while lysine 28 is also modified by P/CAF [56]. The possible role and timing of Tat acetylation inside the cell is still largely unclear. Lysines 50 and 51 lay in the arginine-rich, RNA binding motif of Tat, and their acetylation might be involved in either the regulation of binding of Tat to TAR, or the formation of a stable complex between Tat, TAR and cyclin T1, or in modulation of the interaction between Tat and other factors such as TBP, CBP and RNA polymerase II [5557]. In this respect, mutation of lysines 50 and 51 to alanines to prevent acetylation at these sites strongly reduced transactivation [55,56], while a more conservative substitution with arginine had a more modest effect. Molecular dynamics simulation on Tat variants at positions 50 and 51 indicates that alanine substitutions at these sites have a profound effect on overall protein folding, while mutation of lysines to arginines maintains protein structure, in agreement with the transactivation results [58]. Structure of the Tat variant with an acetylated lysine at position 50 — a modification that alters charge but preserves a long amino acid side chain — does not induce significant variations in folding as compared to the wild-type protein. Thus, it is very conceivable that the role of acetylation might be the specific modification of Tat affinity for other interacting proteins. Consistent with this conclusion are some recent experimental results that indicate that acetylation of Tat at lysine 50 establishes a novel protein–protein interaction domain at the surface of Tat that binds P/CAF; this interaction dissociates Tat from TAR and is necessary for Tat transactivation of the LTR promoter [59,60].

5 Transcellular transactivation

The basic domain of Tat (amino acids 49–57) plays a crucial role for a number of key functions. Besides binding to TAR RNA, this domain contains a nuclear localization signal [61] that directly binds importin β [62], and interacts with other cellular proteins including p300/CBP [46] and PKR [63,64]. In addition, this domain confers to Tat the unusual feature of exiting from cells in the absence of a signal peptide for secretion [6,6568]; reviewed in [69,70]. Most of extracellular Tat is not released in the extracellular environment, but remains associated to the cell membrane through the interaction of its basic domain with cell surface proteoglycans containing heparan sulfates (HSs) [71].

Tat released by infected cells is likely to exert autocrine and paracrine activities that are possibly beneficial for HIV survival and spread and might play a role in the pathogenesis of HIV disease [70]. Extracellular Tat promotes the production of cytokines [7277] and the expression of cytokine receptors [7880]; modulates the survival, proliferation and migration of different cell types [8185]; exerts angiogenic activity in vitro and in vivo [5,8587]; inhibits antigen-specific lymphocyte proliferation [8890]; and induces neurotoxicity in the central nervous system [9196].

Several of these effects might be mediated by the specific interaction of extracellular Tat with cell surface receptors. Tat binds and activates vascular endothelial growth factor receptors 1 and 2 (VEGFR-1/Flk-1 and VEGFR-2/Flt-1) in endothelial and other cell types [9799]; interacts with β-chemokine receptors CCR2 and CCR3 and acts as a potent chemoattractant for various leukocytes [100103]; binds chemokine receptor CXCR4 and competes with X4-HIV-1 virus infection on T-cells [104,105]; and induces cell adhesion through interaction of its RGD domain (located in its second exon) with integrin receptors α5β1 and αvβ3 [106108].

Besides its interaction with these cell surface receptors and the consequent activation of the respective intracellular signal transduction programs, several of the activities of extracellular Tat are most likely mediated by its unique property of being rapidly internalized by a variety of cell types through its basic domain, as originally shown over 10 years ago [109111]. The molecular mechanisms involved in extracellular Tat internalization have been recently started to be elucidated. In contrast to short peptides corresponding to the isolated basic domain, which directly cross cell membranes [112,113], internalization of full length Tat depends on its interaction with cell membrane HS proteoglycans followed by an active endocytosis process [6,114].

The uptake, internalization, and nuclear translocation of extracellular Tat through ubiquitously expressed HS proteoglycans can also be exploited as a biotechnological tool for intracellular protein delivery to a variety of mammalian cell types, both in vitro and in vivo. Chemical crosslinking of Tat peptides with heterologous proteins [115] or, more efficiently, production of recombinant proteins containing the Tat basic domain [116,117] facilitate the intracellular delivery of these proteins. Despite a variety of recent applications exploiting this unusual feature of the Tat basic domain [118122], the molecular mechanisms involved in endocytosis of the HS proteoglycans/Tat complexes and those mediating Tat release from the endosomes are still largely unexplored.

6 Conclusions

The overview of the molecular activities of Tat clearly indicates that, far beyond a canonical transcriptional transactivator, the protein acts as a pleiotropic factor for a surprising number of functions both inside and outside the cell. The identification of the cellular partners that bind Tat and mediate these activities now offers a whole new series of interactions for the development of novel drugs that might be beneficial for AIDS therapy. In this context, however, it must be emphasized that the real significance of most of these pleiotropic activities of Tat is still elusive. In particular, the finding of Tat as a protein associated to the cell surface through its interaction with cell membrane HS proteoglycans might suggest a direct role of Tat either in the modulation of HIV virion infectivity — possibly by facilitating virus to cell attachment — or in the pathogenesis of HIV disease — for example, by protecting the infected cells from immune recognition and destruction. Novel experiments specifically aimed at elucidating these possibilities are now clearly required.


These studies were supported by grants from the Istituto Superiore di Sanità– Progetto AIDS and Progetto Italia-USA sulla Terapia dei Tumori, the AIRC (Associazione Italiana per la Ricerca sul Cancro), the Ministero della Sanità– Progetto Finalizzato, and the Compagnia di San Paolo. We thank Dr. Anna Rapetti for expert secretarial assistance and Monica Barabino for data management.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
  52. [52].
  53. [53].
  54. [54].
  55. [55].
  56. [56].
  57. [57].
  58. [58].
  59. [59].
  60. [60].
  61. [61].
  62. [62].
  63. [63].
  64. [64].
  65. [65].
  66. [66].
  67. [67].
  68. [68].
  69. [69].
  70. [70].
  71. [71].
  72. [72].
  73. [73].
  74. [74].
  75. [75].
  76. [76].
  77. [77].
  78. [78].
  79. [79].
  80. [80].
  81. [81].
  82. [82].
  83. [83].
  84. [84].
  85. [85].
  86. [86].
  87. [87].
  88. [88].
  89. [89].
  90. [90].
  91. [91].
  92. [92].
  93. [93].
  94. [94].
  95. [95].
  96. [96].
  97. [97].
  98. [98].
  99. [99].
  100. [100].
  101. [101].
  102. [102].
  103. [103].
  104. [104].
  105. [105].
  106. [106].
  107. [107].
  108. [108].
  109. [109].
  110. [110].
  111. [111].
  112. [112].
  113. [113].
  114. [114].
  115. [115].
  116. [116].
  117. [117].
  118. [118].
  119. [119].
  120. [120].
  121. [121].
  122. [122].
View Abstract