The androgen receptor (AR) is a ligand-activated transcription factor that regulates gene expression in response to the steroids testosterone and dihydrotestosterone. AR-dependent gene expression is likely to play an important role in a number of receptor-associated disorders, such as prostate cancer, spinal bulbar muscular atrophy, male type baldness and hirsutism. The AR contains two transactivation domains, termed AF1 (activation function 1) located in the N-terminus and AF2 (activation function 2) in the C-terminal ligand-binding domain. AF2 exhibits weak transcriptional activity, whereas AF1 is a strong regulator of transcription. Transcriptional regulation by AF1 is thought to be modulated by a number of proteins that interact with this region, and by post-translational modifications. Our focus is on the N-terminal-interacting proteins and their regulation of transcription via interaction with the receptor. To better understand the mechanism of AR-AF1 action, we have reconstituted AR activity in HeLa nuclear extracts using a unique dual reporter gene assay. Multiple LexA-binding sites in the promoter allow transcription to be driven by a recombinant AR-AF1–Lex fusion protein. The findings from initial experiments suggest an increase in transcription initiation and elongation rates by AR-AF1–Lex. The role of protein–protein interactions involving co-activators and basal transcription factors and AR-AF1 activity are discussed.
The AR (androgen receptor) is a member of the steroid receptor superfamily, and is located on the long arm of the X chromosome, Xq11–12. It functions as a ligand-activated transcription factor, mediating the actions of the steroid hormones testosterone and dihydrotestosterone (reviewed in ). Androgens are involved in the development and differentiation of not only the male reproductive tract, but also of non-reproductive tissues such as brain, kidney, liver and muscle [2,3]. The protein has a similar domain structure as other members of the nuclear receptor superfamily, consisting of a highly variable N-terminal domain, and a relatively conserved central DNA-binding domain (DBD) and C-terminal ligand-binding domain (LBD) (Figure 1A). The AR binds as a homodimer to DNA sequences related to the consensus sequence 5′-AGA/TACA/TnnnT/AGTTCT-3′ (where n represents any nucleotide), located within the promoter and/or enhancer regions of androgen-regulated genes (, and references therein). The AR N-terminus is among the largest in the nuclear receptor superfamily, consisting of almost 550 amino acids, and is characterized by a number of amino acid repeat sequences, including polyglutamine (Q), polyglycine (G) and polyproline (P). This region of the receptor is involved in repression as well as activation of transcription. These two activities are thought to be mediated in turn by different N-terminal-interacting proteins (reviewed in ).
Reconstituting AR-AF1 activity in HeLa nuclear extracts
The AF1 (activation function 1) of the AR has a modular structure, and shows context-dependent activity . We have previously identified and characterized the interaction between the AR-AF1 and the general transcription factor TFIIF [5,6]. TFIIF is an α2β2 heterotetrameric component of the general transcriptional apparatus, and has been found to act at multiple steps during the transcription cycle. During transcription initiation, TFIIF plays an important role in assembly and stability of the pre-initiation complex, the wrapping of DNA around the RNA polymerase, open complex formation and promoter escape. Subsequently, TFIIF enhances transcription elongation by preventing pausing by the RNA polymerase II enzyme (see , and references therein). Recent structural information from cryo-electron microscopy suggests that TFIIF binding alters the conformation of the RNA polymerase II enzyme, and the protein shows an extended conformation over the surface of the polymerase in the region of the active-site cleft (reviewed in ). In protein–protein binding studies, the large subunit of TFIIF, RAP74, was identified as a target for the N-terminal transactivation function of the AR (amino acids 142–485) and was shown to specifically reverse AR-dependent squelching of basal transcription . More recently, we have mapped the AR-binding site to the C-terminal domain of RAP74 and have identified point mutations within AF-1 that selectively disrupt TFIIF binding [6,9].
In addition to TFIIF, a number of co-regulatory proteins bind to AF-1, including members of the p160 co-activator family, [6,10] and the co-repressor SMRT (silencing mediator for retinoic acid receptor and thyroid-hormone receptor) . From the mapping studies of Bevan et al. , and our own mutagenesis studies [6,9], the binding sites for p160 co-activators and TFIIF appear to be distinct. This allows for the possibility of both proteins forming a complex with the receptor at the same time (see ).
Squelching of transcription may be due to sequestering of TFIIF from the general transcriptional machinery, since adding back this factor or the large subunit rescues the transcriptional repression ([5,6]; A. Ball, M.A. Choudhry and I.J. McEwan, unpublished work). Another possible mechanism for the repressive effect could be steric hindrance in recruiting important transcriptional factors to the pre-initiation complex by AR in complex with TFIIF. These possibilities are not mutually exclusive, but do raise the question, ‘Would AR-AF1 bound to the promoter region behave differently?’
Reconstituting AR-AF1 activity under cell-free conditions
Despite the increase in our knowledge of potential AR-binding partners, the mechanism(s) of AR-dependent gene activation remain poorly understood. Transcriptional regulation in eukaryotic cells requires two general steps to be accomplished: (1) the remodelling of chromatin structure to open up regulatory regions and the promoter; and (2) the recruitment of the general transcription machinery to the promoter to enhance transcription initiation and/or elongation. The AR can potentially regulate both these steps, leading to an increase in the levels of target-gene mRNA.
In order to better understand the molecular details of the direct action of the AR on the transcription machinery, we have reconstituted receptor-dependent transcription in HeLa cell nuclear extracts. To investigate different steps in the transcription cycle we have used a modified dual reporter gene system originally described by Lee and Greenleaf . The plasmid pSLG407 (kindly given by A. Greenleaf, Duke University, Durham, NC, U.S.A.) contains two G-less reporter genes, whose expression is driven by the adenovirus major late promoter (Figure 1B). We have cloned multiple LexA response elements at a unique restriction site upstream of the TATA box, which destroys the binding site for USF (Figure 1B). The two reporter genes are separated by 368 bp, and yield different size transcripts after treatment of the mRNA with RNase T1. The short, promoter-proximal transcript (85 nt) gives a measure of initiation rates, whereas the relative levels of the larger, promoter-distal transcript (377 nt) to the short transcript gives an indication of elongation efficiency. The levels of each transcript can be measured directly by incorporating [32P]UTP, and then each transcript is resolved by denaturing urea/PAGE and quantified using a phosphoimager. A LexADBD fusion protein was used in order to investigate the AR-AF1 activity independently of other receptor functions (i.e. steroid binding, dimerization and DNA binding). The fusion protein was expressed in Escherichia coli and purified by affinity chromatography on Ni2+-nitrilotriacetate resin (Figure 1A).
Figure 1(C) shows activation of the reporter genes by AR-AF1 from either a single or multiple LexA binding sites. Only very weak signals were obtained with the single site. However, robust levels of both the short and long transcripts were observed with up to eight DNA response elements, strongly suggesting a role for synergistic activity between AF1 dimers in AR activation. In the absence of the AR-AF1–Lex polypeptide, there is no transcription from this promoter (results not shown). In contrast, increasing amounts of AR-AF1 alone, without DNA binding, leads to squelching of the basal transcription from the original pLSG407 plasmid (A. Ball, M.A. Choudhry and I.J. McEwan, unpublished work).
We have established a cell-free transcription system based on recombinant receptor proteins, HeLa nuclear extract and a unique dual G-less cassette reporter gene. This assay system will be useful to analyse further the mechanism(s) of AR-dependent gene activation under controlled conditions. Our findings to date can be summarized in the model shown in Figure 2. In the absence of a DBD, the AR-AF1 domain squelches or inhibits basal levels of transcription. This is thought to result from the sequestering of one or more components necessary for transcription. Introduction of point mutations in the AF1 domain that disrupt function or addition of recombinant TFIIF rescues this squelching phenotype (A. Ball, M.A. Choudhry and I.J. McEwan, unpublished work). It is unclear at present whether co-activators can also be sequestered by the AF-1 domain alone. When AR-AF1 is tethered to DNA via LexADBD, synergistic activation of transcription is observed. This is most likely to result from protein–protein interactions with both co-activator proteins as well as components of the general transcription machinery (Figure 2). Furthermore, comparison of the levels of both the small and large transcripts reveals that the AR-AF1 acts to increase initiation and elongation efficiency (A. Ball, M.A. Choudhry and I.J. McEwan, unpublished work). This is in good agreement with our previous studies identifying and characterizing the interaction with TFIIF, a basal factor important for both initiation and elongation events (see [5,6]). Our findings are also consistent with results from Chang and co-workers [14,15], describing interactions between the AR and subunits of TFIIH and the elongation factor P-TEFb, two other basal transcription factors involved in transcription initiation and the early stages of elongation respectively. The direct involvement of these factors in AR-dependent activation remains to be demonstrated.
Model of AR-AF1 function
Androgens are involved in both development and expression of the male phenotype and the development of prostate cancer, male type baldness and hirsutism. Androgenic steroids are also involved in morphological characteristics, such as increasing muscle mass , and have a role in bone metabolism [17,18]. All these androgen-mediated actions are attributed to the activity of AR, which carries out these activities in the presence of hormone and a number of receptor-interacting proteins that modulate the functional activity of AR. Thus the ultimate goal of research is to be able to describe the molecular details of AR action on different genes in the complex environment of target cells and tissues. As a step towards that aim, we have described the action of the AR-AF1 domain at different steps in the transcription cycle in an in vitro transcription model system.
Research Colloquia: Research Colloquia at BioScience2004, held at SECC Glasgow, U.K., 18–22 July 2004. Edited by M. Bouvier (Montreal, Canada), G. Milligan (Glasgow, U.K.), V. O'Donnell (Cardiff, U.K.), M. Brand (MRC-Dunn Human Nutrition Unit, Cambridge, U.K.), M. Schweizer (Heriot-Watt University, Edinburgh, U.K.), R. Insall (Birmingham, U.K.), A. Ridley (Ludwig Institute for Cancer Research, London, U.K.) and M. Sutcliffe (Leicester, U.K.). The first eight papers featured in this Section were presented as a part of the GPCR Regulation and Signalling Research Colloquium, incorporating the GPCR–Ion Channel Interactions Pfizer-Sponsored Research Colloquium.
This work was supported by Biotechnology and Biological Sciences Research Council (BBSRC) grant no. 1/C18001.