Mutation of Histidine 874 in the Androgen Receptor Ligand-Binding ...

0888-8809/05/$15.00/0 Printed in U.S.A.

Molecular Endocrinology 19(12):2943–2954 Copyright © 2005 by The Endocrine Society doi: 10.1210/me.2005-0231

Mutation of Histidine 874 in the Androgen Receptor Ligand-Binding Domain Leads to Promiscuous Ligand Activation and Altered p160 Coactivator Interactions Jennifer Duff and Iain J. McEwan School of Medical Sciences, College of Life Sciences and Medicine, University of Aberdeen, Aberdeen, Scotland, United Kingdom AB25 2ZD The androgen receptor (AR) signaling pathway is a major therapeutic target in the treatment of prostate cancer. The AR functions as a ligand-activated transcription factor in the presence of the cognate hormone ligands testosterone and dihydrotestosterone (DHT). We have characterized a highly conserved sequence at the C-terminal end of helix 10/11 in the ligand-binding domain (LBD), which is prone to receptor point mutations in prostate cancer. This sequence includes threonine 877 that is involved in hydrogen bonding to the D ring of the steroid molecule and leads to promiscuous ligand activation of the AR when mutated to alanine or serine. A second mutation in this region, H874Y, also results in a receptor protein that has broadened ligand-binding specificity, but retains an af-

finity for DHT (Kd ⴝ 0.77 nM) similar to that of the wild-type receptor. The structure of the mutant LBD, expressed in Escherichia coli, is not dramatically altered compared with the wild-type AR-LBD in the presence of DHT, but shows a modestly increased sensitivity to protease digestion in the absence of hormone. This mutant AR showed wildtype AR-LBD/N-terminal domain interactions, but significantly enhanced binding and transactivation activity with all three members of the p160 family of coactivator proteins. Together, these phenotypic changes are likely to confer a selective advantage for tumor cells in a low androgen environment resulting from hormone therapy. (Molecular Endocrinology 19: 2943–2954, 2005)

C

(5–8), mutations of the receptor protein (reviewed in Refs. 2–4 and see below), receptor signaling through cross-talk with growth factor stimulated pathways (reviewed in Refs. 9 and 10), and/or alterations in receptor coregulatory protein activity or levels (11, 12). In addition, the loss of the AR from the surrounding stroma tissue has been implicated in prostate cancer disease progression (13). It is thought that the alterations in receptor levels, through amplification of the AR gene, and point mutations in the receptor protein are likely to confer selective advantage for the tumor cells in the low androgen milieu associated with androgen ablation therapy (3, 4). The AR is a member of the nuclear receptor superfamily and mediates the actions of the androgens testosterone and dihydrotestosterone (DHT) at the level of target gene expression. The receptor protein consists of a DNA-binding domain (DBD) flanked by a C-terminal ligand binding domain (LBD) and a structurally distinct N-terminal domain (NTD) important for transactivation (4, 14). The occurrence of point mutations in the AR and their role in prostate cancer tissue have been areas of intense research and the subject of some controversy. However, the picture that is emerging from the literature is that there is a very low incidence of receptor mutations in untreated cancer and that AR mutations do not play a role in the early stages of the disease (15–19). Although the identification of AR point mutations in prostate cancer patients has varied widely from 0–50% of tumors, there does ap-

ANCER OF THE prostate gland arises in the epithelium cells and represents a major cause of cancer-related death in men in the Western world: for example, in 2000 it was the most commonly diagnosed cancer in men in the United Kingdom (1–3). Androgen signaling is necessary for normal growth and differentiation of the prostate gland and is also important for tumor growth (reviewed in Refs. 2–4). Thus, treatment primarily involves targeting the androgen-signaling pathway by blocking the production of testicular androgens and inhibiting androgen receptor (AR) function. Although initially such treatments are successful in managing the disease, the period of remission is variable, and inevitably the tumor escapes this androgen ablation therapy and progresses to a hormone-independent state with concomitantly poor prognosis (2, 4). However, the hormone-resistant tumors still retain a functional AR, and progression may be associated with amplification of the receptor gene First Published Online August 4, 2005 Abbreviations: AF, Activation function; AIS, androgen insensitivity syndrome; AR, androgen receptor; DBD, DNAbinding domain; DHEA, dehydroepiandrosterone; DHT, dihydrotestosterone; GRE, glucocorticoid response element; LBD, ligand-binding domain; NTD, N-terminal domain; SRC1a, steroid receptor coactivator-1a; TIF-2, transcriptional intermediary factor-2. Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

2943

2944 Mol Endocrinol, December 2005, 19(12):2943–2954

pear to be a good correlation with antiandrogen treatment and the advance stages of the disease and changes in the receptor protein (16–26). An exception to the above is the report of 44% AR mutations in 25 advanced primary prostate tumors before any hormone therapy (27). The majority of AR point mutations have been mapped to mutational hot spots in the LBD: amino acids 670–678, 701–730, and 874–919 (28). Significantly, these mutational hot spots are distinct from those regions associated with mutations found in the androgen insensitivity syndrome (AIS) (28). The latter represents a spectrum of disorders that range from complete to mild androgen resistance and defects in male development. The lack of overlap of mutational hot spots is consistent with the idea that mutation in prostate cancer are likely to be gain of function changes, whereas those in AIS result in loss of function. Characterization of the molecular phenotype of the AR point mutations in prostate cancer could have important implications for a patient’s response to hormone therapy and the effectiveness of various treatment strategies. To better understand the role of point mutations in AR function and prostate cancer disease progression, we have introduced selected mutations into the isolated AR-LBD or full-length receptor to investigate receptor-structure function relationships. In this study we present a detailed analysis of the exon 8 mutation H874Y, which is located within helix 10/11 of the LBD. The results indicate that a combination of

Duff and McEwan • AR Mutations in Prostate Cancer

altered ligand-binding properties and enhanced binding to p160 coactivators may be responsible for the gain of function of this point mutation and may suggest mechanisms leading to hormone refractory disease. RESULTS Analysis of the amino acid sequence of the AR from a wide range of species currently available in the database reveals that histidine 874 is highly conserved from primates through fish (Fig. 1). The notable exceptions are trout and bullfrog AR, where glutamine and valine are found respectively. Histidine 874 is located within the C terminus of helix 10/11 and is part of a highly conserved eight-amino acid sequence, 873 LHQFTFDL880, that includes threonine 877 and aspartic acid 879, residues also mutated in prostate cancer (21, 23). The conserved nature of all three residues strongly suggests an important structural and/or functional role. Indeed T877A, which was originally identified in the LNCaP metastatic prostate cell line (29), has been shown in the crystal structure of the AR-LBD to play a role in the discrimination of the steroid D ring (30, 31). In contrast, histidine 874 and aspartic acid 879 are not directly implicated in steroid binding from the LBD crystal structures. However, the AR with the H874Y mutation has been reported to be activated by estradiol and progesterone (21).

Fig. 1. Alignment of AR-LBD Helix 10/11 from Different Species The amino acid sequences of the AR from a number of mammals, amphibians, and fish species, corresponding to the C terminus of helix 10/11, have been aligned using Clustal (*, amino acid identity; :, amino acid homology). The residues H874, T877, and D879 (numbering for the human AR) have been highlighted.


To better understand the molecular phenotype of H874Y, we have introduced this mutation, together with T877A and D879G, into the full-length AR. Because the original report of the H874Y mutation had shown a response to both estradiol and progesterone (21), we used a glucocorticoid response element (GRE)2-driven luciferase reporter gene to examine the abilities of a range of natural steroids and nonsteroidal antiandrogens to activate the wild-type and mutant receptor proteins. Luciferase activity was measured after cotransfection of the receptor cDNAs and the reporter gene in COS-7 cells. In the presence of testosterone or DHT, the wild-type receptor robustly activated transcription 10.5- and 8-fold, respectively, compared with the vehicle control (Fig. 2A). The H874Y and T877A mutant receptors also responded strongly with these androgenic steroids: 12.3- and 11.6-fold with DHT and 7.3- and 4.9-fold with testos-

Fig. 2. Reporter Gene Activation by Wild-Type (WT) and Mutant AR Proteins A, COS-7 cells were transfected with plasmids expressing the wild-type or mutant AR together with a (GRE)2TATAluciferase reporter gene. Cells were incubated with vehicle control (ethanol) or 100 nM DHT testosterone (Test.) and 10 nM DHEA, estradiol (E2), and progesterone (Prog.). The cells were harvested, luciferase activity was measured, and protein concentration was determined in cell extracts. Experiments were carried out in duplicate, and the results are presented (mean ⫾ SD) for at least two independent experiments reported. The activity of the wild-type receptor in the absence of ligand (ethanol) was set at 1. B, As in A, except the cells were treated with 100 nM of the nonsteroidal ligands, flutamide and nilutamide, and 10 nM bicalutamide. Experiments were carried out in duplicate, and the results are presented (mean ⫾ SD) for at least two independent experiments reported. The activity of the wild-type receptor in the absence of ligand (ethanol) was set at 1.

Mol Endocrinol, December 2005, 19(12):2943–2954 2945

terone, respectively. However, the D879G mutant receptor showed considerably less activity: 5- and 3-fold with DHT and testosterone, respectively (data not shown). In contrast to the wild-type AR, the mutant receptors H874Y and T877A showed significant (3- to 5-fold) levels of activity in the absence of ligand (Fig. 2A). The wild-type receptor, as expected, showed little or no activity in response to the adrenal androgen dehydroepiandrosterone DHEA, the steroid hormones estradiol and progesterone, or the antiandrogens flutamide and nilutamide (Fig. 2). In contrast, the H874Y mutant receptor showed significant activity with DHEA (6.5-fold) and progesterone (9.8-fold) and modest activity with estradiol compared with the wild-type control (Fig. 2A). As expected, the T877A mutation was also activated by these steroids and showed an 8-fold response to estradiol. Interestingly, the H874Y mutant receptor was also activated by nilutamide (7.3-fold), but not by bicalutamide or flutamide (Fig. 2B). Indeed, treatment with the antiandrogen flutamide reduced the ligand-independent activity of this mutation. T877A was activated by all three nonsteroidal antiandrogens tested, with bicalutamide giving the greatest response (18-fold; Fig. 2B). Interestingly, the wild-type receptor showed a weak agonist response with bicalutamide (4-fold; Fig. 2B). Mutation of D879G showed no activation by DHEA, estradiol, progesterone, or the nonsteroidal antiandrogens (data not shown). Western blot analysis indicated comparable levels of wild-type and mutant AR proteins under different activating conditions (data not shown). Taken together, these data suggest that H874Y confers ligand-independent and promiscuous ligand-binding activity on the AR protein to a wide range of steroid and nonsteroidal ligands. To investigate the ligand-binding properties of the H874Y mutation, the affinity for DHT was measured after transfection of the full-length receptor into COS cells. Figure 3 shows that a Kd of 0.77 nM was measured for both wild-type and mutant receptor proteins. However, the levels of ligand-bound H874Y receptor were only 20% those of the wild type. Table 1 summarizes the Kd measurements for all three mutant receptors and indicates no dramatic change in binding affinity for DHT. Interestingly, from the Western analysis, the levels of wild-type and H874Y mutant receptor are comparable, which contrasts with the data from Scatchard analysis and suggests that a significant population of mutant receptor does not bind ligand stably in vivo. To investigate possible structural alterations due to mutating these three residues, the point mutations were introduced into a construct consisting of the AR-DBD-LBD (amino acids 529–910) and expressed in Escherichia coli (Fig. 4A). Figure 4B shows the expression and purification of AR-DBD-LBD proteins in the absence and presence of DHT. In the absence of steroid, the AR polypeptide is associated with equimolar amounts of the bacterial chaperone protein DnaK, a homologue of 70-kDa heat shock protein (32). The identity of DnaK was confirmed by Western blot analysis (data not shown). In the presence of 50 ␮M DHT,

2946

Mol Endocrinol, December 2005, 19(12):2943–2954


Fig. 3. Binding Affinity of Wild-Type (WT) and Mutant AR Proteins for DHT COS-7 cells were transfected with expression plasmids for wild-type or mutant receptor proteins. The cells were incubated with 0.25–50 nM [3H]DHT in the absence (total binding) or presence (nonspecific binding) of a 200-fold excess of nonlabeled DHT for 2 h at 37 C. Cell extracts were prepared, and the amount of bound steroid was measured by scintillation counting. Specific binding (Œ) was determined by total (⽧) ⫺ nonspecific (〫; insets), and the affinity constant (Kd) was measured from the Scatchard plot of bound/free steroid (Bs/Fs) against bound steroid (Bs). Data for wild-type and H874Y mutant are shown.

the levels of DnaK were significantly reduced, and fewer breakdown products or contaminating proteins were observed for both wild-type and mutant receptors (Fig. 4B). Partial digestion with the protease trypsin, which cleaves after lysine and arginine residues, has proved a useful tool for investigating the conformation of the receptor LBD (33, 34). Thus, to compare the global structure of the wild-type and mutant ARLBD, the recombinant proteins were treated with trypsin for up to 12 min. Figure 5 shows that in the absence of hormone, digestion of wild-type AR-DBD-LBD resulted in two major fragments of 29 and 35 kDa. Similar fragments were observed with the H874Y, T877A, and D879G mutant proteins, although the 35-kDa fragment appeared more susceptible to proteolysis for H874Y (Fig. 5). In the presence of hormone, both wildtype and mutant proteins showed essentially the same pattern, with the dominant fragment of 29 kDa. Additional structural analysis involved steady-state fluorescence emission spectroscopy. The AR-DBD-LBD polypeptide contains four tryptophan residues and 13 tyrosines, which were excited at 278 nm. The steadystate emission spectra for the wild-type and mutant

receptors were very similar, with a peak emission (␭max) for tryptophan of 306–308 nm, which is consistent with each protein being folded and the tryptophans being buried and shielded from solvent (data not shown). However, interestingly, a strong peak was observed at 278 nm for the mutants H874Y and D879G compared with the wild-type and T877A mutant, presumably due to absorption by steroid, even

Table 1. Binding Affinity of DHT

a

Receptor Protein

Kd (nM)

Bs (fmol/ml cell extract)

Wild type H874Y T877Aa D879G

0.77 0.77 0.38–0.58 0.59

101 21 ND 22

Zhao et al. (45) (CV1 and LNCaP cells) and Mizokami et al. (46) (LNCaP cells). ND, Not determined.

Fig. 4. Expression of AR-LBD in Bacteria A, Schematic representation of the human AR together with the construct encoding the DBD-LBD. B, Coomassie Blue-stained gel of partially purified wild-type and mutant AR polypeptides in the absence (⫺) or presence (⫹) of 50 ␮M DHT. The receptor polypeptide and the bacterial chaperone protein DnaK are indicated by the arrowheads.



Fig. 5. Partial Trypsinization of AR-DBD-LBD Polypeptides The Western blots on the left show the fragments generated after trypsin digestion in the absence of hormone for the wild-type (WT), H874Y, T877A, and D879G AR polypeptides. The Western blots on the right show the patterns of digestion in the presence of 50 ␮M DHT for wild-type and mutant receptor polypeptides. Receptor polypeptides were detected using antibody C19 (Santa Cruz Biotechnology, Inc.) recognizing epitopes in the AR-LBD. Open arrowhead, Intact proteins; filled arrowheads, 35- and 29-kDa polypeptides, respectively. Note the reduced digestion of the full-length wild-type and T877A mutant recombinant proteins in the absence of ligand (middle left panel), most likely reflecting variability in the partial digestion reaction.

though the spectra had been corrected for free steroid. This might suggest subtle differences in hormone binding or dissociation kinetics for H874Y and D879G. Taken together, the structural analysis supports the conclusion that mutating histidine 874 or aspartic acid 879 does not lead to any gross alterations in structure, as was also demonstrated by the crystal structure of T877A bound to DHT (31). One of the best-characterized protein-protein interactions involving the AR is between AR-NTD and ARLBD. This involves an FxxFL motif within the first 21 amino acids of the AR-NTD and an area overlapping with that bound by the p160 coactivators in the LBD (35, 36). We therefore investigated the possible effect of the H874Y and T877A mutations on interactions with the AR-NTD. The AR-NTD was synthesized and radiolabeled in a rabbit reticulocyte lysate system and

incubated with wild-type or mutant AR-DBD-LBD proteins immobilized on the surface of a Scintiplate well. After extensive washing, the bound radioactivity was measured directly and plotted relative to a BSA control. Figure 6 shows that the wild-type and mutant proteins showed similar levels of ligand-dependent binding of AR-NTD. The transactivation activities of both wild-type and mutant receptors were also compared by studying the protein-protein interaction profiles with coregulatory proteins. Figure 7 shows liganddependent interactions with members of the p160 family of coactivators, steroid receptor coactivator-1a (SRC-1a), transcriptional intermediary factor-2 (TIF-2), and activator of thyroid hormone and retinoid receptor/amplified in breast cancer 1 (ACTR/AIB1), for the wild-type and H874Y and T877A mutant AR polypeptides. The binding of p160 coactivators to the T877A



Fig. 6. Interaction of AR-NTD with Wild-Type (WT) and Mutant AT-LBDs Radiolabeled AR-NTD polypeptide was incubated with either wild-type or mutant AR-DBD-LBD polypeptides in the absence (⫺ DHT) or presence (⫹ DHT) of hormone, and the bound radioactivity was measured after incubation and extensive washing. Binding was measured in triplicate, and the results from at least two independent experiments are presented as the mean ⫾ SD relative to a BSA control (set at 1).

polypeptide was comparable with that to the wild-type receptor, but binding to the H874Y mutant receptor polypeptide was increased by 40–50%. The binding of p160 coactivators to the LBD has been shown to involve LxxLL motifs in the coactivator proteins and a hydrophobic cleft and charge clamp on the surface of the LBD. A lysine residue, equivalent to position 720 in the AR-LBD, has been shown to be critical residue for coactivator binding, making up one end of the charge clamp (37, 38). The mutation K720E has been reported in prostate cancer (39) and would be predicted to disrupt the binding of p160 proteins and thus act as a negative control. We expressed and purified the ARDBD-LBD polypeptide with the K720E mutation (Fig. 4B) and tested the binding of coactivator proteins. Figure 7 shows that K720E has greatly reduced or no binding to each of the p160 family members. Taken together, the protein-protein interaction studies suggest that mutating H874 selectively alters interactions with members of the p160 family of coactivator proteins. To study the possible consequences of increased p160 binding on receptor function, we cotransfected the wild-type and mutant AR with SRC-1a in COS cells. Figure 8 shows that cotransfection of SRC-1a resulted in a significant enhancement of reporter gene activity with the H874Y mutant receptor in the presence of DHT, without significantly altering receptor levels. Note that under the conditions of this experiment, the wild-type and mutant receptors all showed approximately a 2.5-fold ligand activation of reporter gene activity in the absence of exogenous SRC-1a. The activity of the K720E receptor showed a modest decrease in activity, compared with wild-type, in the absence of cotransfected SRC-1a (Fig. 8). All three receptors were expressed at comparable levels (Fig. 8, inset). Taken together, these studies reveal an increased binding of p160 coactivators and increased transactivation activity for the H874Y mutant receptor.

Fig. 7. Binding of p160 Coactivator Proteins to Wild-Type (WT) and Mutant AR LBD Polypeptides The p160 family members SRC-1a, TIF-2, and ACTR/AIB1 were radiolabeled and incubated with immobilized AR-DBDLBD in the absence (⫺ DHT) or presence (⫹ DHT) of hormone as described in Materials and Methods. After incubation and extensive washing, the bound radioactivity was measured. Binding was measured in triplicate, and the results from at least two independent experiments are presented as the mean ⫾ SD relative to a BSA control (set at 1).

DISCUSSION In this study we have undertaken a detailed analysis of residues at the C terminus of helix 10/11 in the AR-LBD that are mutated in prostate cancer; in particular, histidine 874. Mutation of histidine 874 was originally identified in a patient with hormone refractory prostate cancer (21) and was later found to be the mutation present in the CWR22 human prostate cancer xenograft model (40). This point mutation results in a receptor protein that is


Fig. 8. Coactivation of Wild-Type (WT) and H874Y AR Transcription Activities by SRC-1a COS-7 cells were cotransfected with expression plasmids for wild-type or mutant AR together with (GRE)2TATA-luciferase reporter gene and either an empty vector or a plasmid expressing SRC-1a. Cells were incubated with vehicle control (ethanol) or 100 mM DHT for 24 h. The cells were harvested, and luciferase activity and protein concentration were determined in cell extracts. Experiments were carried out in duplicate, and the results are presented (mean ⫾ SD) for at least two independent experiments. The absence and presence of receptor for DHT and SRC-1a are indicated by the ⫺ and ⫹ signs, respectively. Inset, Western blot for the AR protein in duplicate. Cell extracts were resolved by SDS-PAGE, and proteins were transferred to nitrocellulose. The blot was then probed with an antibody against the LBD. The positions of the full-length receptor proteins are indicated. RLU, Relative light units.

now responsive to adrenal androgens, nonandrogenic steroids, and antiandrogens (Refs. 21 and 40–42 and this study). A similar reporter gene activation profile is observed for the well-characterized T877A mutation, originally identified in the LNCaP cell line (29) and subsequently in patients after androgen withdrawal treatment (21). Both H874Y and T877A were activated to similar degrees by DHEA, but showed differences with estradiol and progesterone. H874Y was more readily activated by progesterone, showing only modest, if any, activation with estradiol. This contrasts with the previous findings of Taplin et al. (21) and Tan et al. (40), who reported higher activity for the H874Y mutant receptor with estradiol than for T877A. Similarly, Taplin et al. (21) found that T877A was activated to a greater extent with progesterone than H874Y. In our reporter gene system, H874Y was more active with 100 nM progesterone than both the wild-type and T877A receptors, in agreement with the study by Tan et al. (40). We also observed differences in the responses of mutant receptors to nonsteroidal antiandrogens, with T877A being activated by flutamide, nilutamide, and bicalutamide, whereas H874Y was only activated by nilutamide. The activation profile of a given receptor mutation could have important consequences when choosing second-line antiandrogen therapies. The activation of wild-type and T877A mutation by bicalutamide was initially surprising, because this nonsteroidal antiandrogen has previously been reported only as a


pure antagonist of both wild-type and mutant receptors (43). However, recently Sawyers and co-workers (44) demonstrated that in a prostate cancer xenograft model, increasing the AR mRNA/protein level was sufficient to produce a hormone-refractory state and convert antagonists to agonists. Thus, overexpression of the receptor may have contributed to this response. However, the H874Y mutation was expressed at similar levels and showed no agonist activity with bilcalutamide. The levels of coactivator proteins have also been shown to influence the response of the AR to antagonists (11). In contrast to both H874Y and T877A, and, indeed, the wildtype receptor, mutating D879G resulted in a reduction in reporter gene activity with androgenic steroids and no concomitant increase with the other ligands tested. This was somewhat surprising given the assumption that this would be a gain of function mutation. This residue has also been found mutated to a tyrosine in partial AIS (45). This study also reported a loss of ligand binding for mutation H874R in complete AIS. Consistent with reports for other AR-LBD mutations found in prostate cancer, there was no significant difference in the affinity of DHT binding to H874Y or D879G compared with that to the wild-type receptor, with Kd values of 0.77, 0.59, and 0.77 nM, respectively, which is also in good agreement with the published affinity for T877A of 0.38–0.58 nM (46, 47). However, it was striking that the levels of receptor competently binding DHT were only 20% those of the wild-type protein. Thus, the specific activity for the H874Y and D879G mutants is significantly higher than that for the wild type, and the increased efficiency in gene activation may be a contributing factor for prostate cancer cells harboring these mutant AR proteins. This reduction in ligand-bound receptor in COS-7 cells may reflect subtle structural alteration in the ligand-binding pocket, which would be supported by the apparent increased dissociation of DHT observed in the fluorescence experiments in vitro. However, we cannot rule out that a population of mutant receptors does not bind steroid at all, perhaps as a result of limiting chaperone activity in COS-7 cells. The above reporter gene studies reveal that H874Y and T877A both result in receptors that are promiscuous in their ligand-binding/activation profile, and this has been argued to represent a selection advantage for the tumor cells in a low androgen environment. Recently, the crystal structure for T877A bound to DHT has been determined. This very clearly illustrates that removing the side chain of threonine and replacing it with alanine creates a larger binding pocket, which can then accommodate bulkier substitutions on the D ring. Furthermore, it removes a direct hydrogen bond from the threonine to the 17␤-hydroxyl group (30, 31). Using partial proteolysis of purified recombinant AR-DBD-LBD polypeptides, we showed that there was no obvious structural alteration for any of the three mutations in the absence or presence of hormone. Histidine 874 is not thought to be directly involved in contacting the hormone molecule from the structures of the AR-LBD available with the agonist bound and appears to face away from the ligand-binding


pocket (30, 31). However, using the DeepView/SwissPdbViwer program (www.expasy.org/spdbv) (48), it is possible to model the consequences of changing this residue for a tyrosine. Figure 9 shows sculpted views of the ligand-binding pocket with DHT bound (1T7T.pdb) for the wild-type histidine residue and the mutated tyrosine and indicates no gross conformational changes in LBD. Changing the histidine to a tyrosine results in nine possible rotamer positions for the mutated residue. Figure 9B shows the one with the least steric hindrance as well as an additional H bond to tyrosine 739; Fig. 9, C and D, shows alternative positions for the tyrosine side chain, pointing away from tyrosine 739, that have steric clashes with neighboring residues, arginine 871, glutamine 875, and the helix 12 residues, valine 903 and leucine 907. Energy minimization calculations suggest that the conformation shown in Fig. 9B is very similar to that of the wild-type residue, whereas the other two positions illustrated are energetically unfavorable. However, it is possible that the steric clashes predicted for these positions could lead to a local disruption in the structure, creating a larger ligand-binding pocket that can accommodate nonandrogenic steroids. Furthermore, such rearrangement of these residues may alter the activation function AF2 transactivation surface and allow enhanced binding of p160 coactivators. Previous modeling of H874Y and T877A (49) also predicted no dramatic conformational changes, which would be in agreement with our trypsinization data. Unlike other members of the nuclear receptor superfamily, the AR exhibits little or no AF2 activity (50, 51) and shows preference for the motif FxxLF, present in the


AR-NTD and some AR coactivator proteins, over the LxxLL motifs present in the p160 coactivator family (52, 53). Recently, the structural basis for this selectively was revealed (54–56). X-Ray crystallography studies showed that the AR-NTD FxxLF motif forms a charge clamp with glutamic acid 897 in helix 12 and lysine 720 at the end of helix 3, and the hydrophobic residues fits better into the surface pocket on the LBD (55, 56). In contrast, an LxxLL motif peptide fails to make hydrogen bond contacts with the glutamic acid residue in helix 12 and makes fewer hydrophobic contacts with the surface of the LBD (55, 56). With H874Y, we observed increased binding of p160 family members and increased coactivation of gene expression. This suggests that mutating the histidine 874 improves the ability of glutamic acid 897 to form a charge clamp with the LxxLL motifs present in the p160 proteins and/or binding of the leucine residue to the hydrophobic pocket on the surface of the AR-LBD, without altering the interaction with the AR-NTD. The K720E mutation showed little or no ligand-dependent binding to p160 family members, but did show a response, albeit reduced, in vivo. This most likely reflects the fact that the p160 proteins bind and coactivate AF1 in the AR-NTD (reviewed in Refs. 14 and references therein). Thus, although the wild-type AR shows little AF2 activity and correspondingly weaker interactions between p160 coactivators and the LBD compared with other members of the nuclear receptor superfamily, the present results demonstrate that a mutation in the LBD can significantly enhance LBD-coactivator interactions and lead to increased transactivation activity. It remains to be determined whether the enhanced interaction of p160 coac-

Fig. 9. Modeling of the H874Y Change in AR-LBD A, Structure of the wild-type receptor with histidine at position 874. Only the histidine (green) and surrounding residues are shown for simplicity. Threonine 877 and the ligand DHT are highlighted in blue and yellow, respectively. Hydrogen bonds are shown as green dotted lines. B–D, Different rotomers of changing histidine 874 to tyrosine. See text for detailed discussion.



tivators with the H874Y mutant AR-LBD results from an increase in binding affinity and/or a decrease in the off rate. The above-described cotransfection studies were carried out in a monkey kidney cell line, COS-7. Because the cellular environment (i.e. coactivator repertoire and levels) may have a profound effect on receptor function, it will be important to assess wild-type and H874Y mutant ARs in a prostate cell background. Although no studies are available that permit a direct comparison, the response profiles of the H874Y mutant receptor to different ligands in a reporter gene assay in the prostate cell line DU145, in yeast (42) and CV1 cells (21, 40, 41), and in the present study are all in generally good agreement. Thus, in the context of the mutations studied in this report, the intrinsic change from the receptor protein, rather than the cellular environment, may be the critical factor in determining the AR response. Increased responses of the AR in prostate cancer cells may also result from alterations in coactivator binding or activity. The mutations H874Y and T877A did not lead to significant changes in the ligand-dependent interaction of the AR-LBD with the NTD. However, strikingly, the H874Y mutation showed enhanced interaction with three members of the p160 coactivator family, and in cotransfection studies, SRC-1a enhanced the activity of the full-length H874Y mutant AR compared with the wild-type receptor or a point mutation that disrupts p160 binding. Overexpression of p160 coactivators, in particular, TIF-2, has been observed in recurrent prostate cancer patients (57), and increased levels of SRC-1 and the coactivator ARA55 were correlated with a response of patients to complete androgen ablation therapy (12). Chang and co-workers (11) showed that even the wildtype AR would respond to nonandrogen ligands such as estradiol by overexpression of coregulator proteins, ARA70, ARA55, and, to a lesser degree, SRC-1. In conclusion, we have shown that the change of histidine 874 to tyrosine can be accommodated without dramatic alterations in the structure of the LBD, but the local changes in the region of the ligand-binding pocket/ AF2 domain may be sufficient to permit a wider range of steroids and nonsteroid ligands to act as agonists. In addition, we found that this mutation led to enhanced

coactivator interactions and transactivation activity. Either or both of these phenotypes would result in a more active AR and could contribute to the progression of tumors to an androgen-independent, ARdependent state. The findings also emphasize the importance of a highly conserved motif in helix 10/11 for selective hormone binding and coactivator-dependent transactivation.

MATERIALS AND METHODS Plasmid Construction and Site-Directed Mutagenesis The sequence for the human AR DBD-LBD fragment (amino acids 528–910) was amplified using the Expand system (Roche, St. Louis, MO), and the resulting PCR product was cloned into pET-19b, containing an N-terminal histidine tag, to give pET-AR-DBD-LBD. Stratagene’s QuikChange sitedirected mutagenesis kit (La Jolla, CA) was used to introduce point mutations into plasmids pSVARo (58) and pET-ARDBD-LBD, which were then used for mammalian expression of the full-length AR and bacterial expression of the AR-DBDLBD, respectively. Primers were designed so as to introduce the desired point mutation (underlined nucleotides) and included or removed a restriction enzyme cleavage site that could be used in screening for positive clones (Table 2). Protein Expression and Purification Wild-type and mutant AR-DBD-LBD plasmids were transformed into BLR cells [recA⫺ derivatives of BL21 (DE3) cells] and grown in 2⫻ TY medium [1.6% (wt/vol) Bactotryptone, 1.0% yeast extract, 0.5% sodium chloride] supplemented with 0.5% (wt/vol) glucose and ampicillin. The cultures were allowed to grow until they had reached an OD of approximately 0.6–0.8 at 600 nm and were induced with 0.5 mM isopropyl-␤-D-thiogalactoside, and transferred to a 23 C shaking incubator for 2 h. Proteins were induced in both the presence and absence of 50 ␮M DHT dissolved in ethanol. Cells were lysed as described previously, and the soluble recombinant proteins were purified by Ni2⫹-nitriloacetate agarose affinity chromatography. Proteins were analyzed by SDS-PAGE and dialyzed against 100 mM HEPES-KOH (pH 7.9), 150 mM NaCl, 10% glycerol, 0.5 mM EDTA, 0.1% Nonidet P-40, 2 mM dithiothreitol, and either 50 mM DHT or 0.1% ethanol vehicle only. The protein concentration was determined by the method of Bradford (59).

Table 2. Primer Sequences Mutation

Primer Sequence (5⬘–3⬘)

K720E

Forward Reverse

GTGGTCAAGTGGGCCGAAGCCTTGCCTGGCTTC GAAGCCAGGCAAGGCTTCGGCCCACTTGACCAC

H874Y

Forward Reverse

CTATTGCGAGAGAGCTCTATCAGTTCACTTTTGACCTGC GCAGGTCAAAAGTGAACTGATAGAGCTCTCTCGCAATAG

T877A

Forward Reverse

CGAGAGAGCTGCATCAGTTCGCGTTTGACCTGCTAATC GATTAGCAGGTCAAACGCGAACTGATGCAGCTCTCTCG

D879G

Forward Reverse

GCGAGAGAGCTCCATCAGTTCACTTTTGGCCTGCTAATC GATTAGCAGGCCAAAAGTGAACTGATGGAGCTCTCTCGC

Underlining represents nucleotide changes introduced into wild-type sequence.


Structural Analysis of Proteins: Partial Proteolysis with Trypsin Partial proteolytic cleavage of histamine-tagged wild-type and mutant AR-DBD-LBD proteins was used as a means of analyzing the structural stability of each protein. Varying concentrations of trypsin were used to digest 25 pmol of each purified protein in reactions made up to a final volume of 10 ␮l with protease buffer [20 mM HEPES-KOH (pH 7.9), 10% (vol/vol) glycerol, 0.2 mM EDTA, 5 mM MgCl2, 20 mM CaCl2, 60 mM KCl, and 100 nM dihydrotestosterone or 0.1% (vol/vol) ethanol]. The reactions were left at room temperature for 2–10 min and were stopped with the addition of 2⫻ sodium dodecyl sulfate sample buffer and heating at 75 C for 5 min. The digest products were analyzed by Western blotting of proteins resolved by12.5% SDS-PAGE using the AR C19 antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA). Protein-Protein Interactions Target proteins were radiolabeled with [35S]methionine (Amersham Biosciences, Arlington Heights, IL) using the TNTcoupled transcription/translation reticulocyte lysate system (Promega Corp., Madison, WI). Two hundred microliters of 100 nM histamine-tagged AR-DBD-LBD protein diluted in binding buffer [20 mM HEPES-KOH (pH 7.9), 100 mM KCl, 10% (vol/vol) glycerol, 5 mM MgCl2, 0.2 mM EDTA, 0.2 mM phenylmethylsulfonylfluoride, 0.5 mM ␤-mercaptoethanol, and 100 nM DHT or 0.1% ethanol as required] was added to wells of a 96-well Scintiplate (Wallac, Gaithersburg, MD) in triplicate and incubated for 3 d at 4 C to allow the proteins to absorb to the plate. The wells were then blocked with BSA before incubation with labeled target proteins for 90 min. The wells were extensively washed, and the bound radioactivity was measured in a micro-␤-counter (Wallac). Mammalian Cell Reporter Gene Assays COS-7, a monkey kidney fibroblast cell line, cells were cultured in 90% (vol/vol) DMEM with 4 mM L-glutamine adjusted to contain 1.5 g/liter sodium bicarbonate and 4.5 g/liter glucose (Invitrogen Life Technologies, Inc., Carlsbad, CA) supplemented with 10% (vol/vol) fetal bovine serum (Biowest, Nuaille´, France). Cells were transiently transfected with 1.5 ␮g AR expression plasmid and 1.5 ␮g 2GRE-TATA-luciferase reporter plasmid (60, 61) using Lipofectamine 2000 (Promega Corp.) in serum-free DMEM. In cotransfection studies with p160 coactivator proteins, 1 ␮g receptor expression plasmid, 2 ␮g reporter gene, and 4 ␮g p160 coactivator-expressing vector DNA or empty plasmid control were used. Cells were maintained under serum-free conditions for 24 h before treatment with ethanol vehicle or selected hormones as detailed in Figs. 2, 3, and 8.


activity for each sample was then normalized for protein levels by dividing the raw relative light units by the sample protein concentration, which was determined by performing a Bradford assay. The data were plotted as a bar graph using Excel (Microsoft Corp., Redmond, WA). The assay was performed in duplicate, and measurements for each sample were taken in duplicate so the SE could be calculated for each set of samples. Five microliters of each sample was also analyzed by Western blotting to determine AR expression levels and ensure comparability of transfection efficiency between different samples using the AR441 antibody (Santa Cruz Biotechnology, Inc.), which detects an epitope in the N-terminal portion of the AR. Ligand Binding Assay: Scatchard Analysis The binding affinity of DHT to the full-length AR in COS-7 cells transfected with wild-type and mutant AR was determined using tritiated DHT (5␣-dihydrol-[1,2,4,5,6,7-3H]testosterone; Amersham Biosciences) in the absence or presence of a 200-fold excess of unlabeled ligand. The radioactivity bound in samples after the washing steps was then measured in a scintillation counter and used to plot a graph displaying the specific, total, and nonspecific binding of radiolabeled DHT. The data obtained from these measurements were used to plot a Scatchard graph from which the binding affinities (Kd) of DHT to wild-type and mutant AR were calculated. The total binding was converted to femtomoles and divided by unbound ligand, also in femtomoles. Bound steroid/free was plotted against bound steroid to give a straight line from which the Kd (binding affinity) of the receptor could then be calculated from the slope (Kd ⫽ 1/slope).

Acknowledgments We gratefully acknowledge the following for providing plasmid DNA reagents: Drs A. O. Brinkmann (Erasmus University, Rotterdam, The Netherlands), P. Chambon (Institut de Ge´ne´tique et de Biologie Mole´culaire et Cellulaire, Strasbourg, Germany), R. Evans (The Salk Institute, San Diego, CA), and B. W. O’Malley (Baylor College of Medicine, Houston, TX).

Received June 13, 2005. Accepted July 27, 2005. Address all correspondence and requests for reprints to: Dr. Iain J. McEwan, School of Medical Sciences, Institute of Medical Sciences Building, University of Aberdeen, Foresterhill, Aberdeen, Scotland, United Kingdom AB25 2ZD. E-mail: [email protected]. This work was supported by Medical Research Council Strategic Postgraduate Studentship G78/7180 (to J.D.).

Luciferase Assay Forty-eight hours after transfection, the cells were harvested, and luciferase activity was measured. The luciferase activity of each sample was measured in a Lumat LB9501 luminometer (Berthold Technologies, Oakridge, TN). Five microliters of each sample was added to 350 ␮l luciferase assay buffer (15 mM MgSO47H2O, 30 mM GlyGly (pH 7.8), and 2 mM Na2ATP) and placed in a 5-ml, 75 ⫻ 12-mm luminometer tube (Sarstedt, Leicester, UK). The tube was inserted inside the measuring chamber of the luminometer, then 100 ␮l 0.5 mM luciferin (Molecular Probes, Inc., Eugene, OR) dissolved in 30 mM GlyGly (pH 7.8) was injected into the sample, and the light emission produced by the reaction between the reagent and sample was measured over 10 sec in the luminometer photodetector. The values obtained for each sample in the luminometer were recorded as relative light units. The luciferase

REFERENCES 1. Toms J 2004 CancerStats Monogr 2004:80 2. Ruijter E, van de Kaa C, Miller G, Ruiter D, Debruyne F, Schalken J 1999 Molecular genetics and epidemiology of prostate carcinoma. Endocr Rev 20:22–45 3. Taplin ME, Ho SM 2001 Clinical review 134: the endocrinology of prostate cancer. J Clin Endocrinol Metab 86: 3467–3477 4. Gelmann EP 2002 Molecular biology of the androgen receptor. J Clin Oncol 20:3001–3015 5. Visakorpi T, Hyytinen E, Koivisto P, Tanner M, Keinanen R, Palmberg C, Palotie A, Tammela T, Isola J, Kallioniemi OP 1995 In vivo amplification of the androgen receptor


6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

gene and progression of human prostate cancer. Nat Genet 9:401–406 Bubendorf L, Kononen J, Koivisto P, Schraml P, Moch H, Gasser TC, Willi N, Mihatsch MJ, Sauter G, Kallioniemi OP 1999 Survey of gene amplifications during prostate cancer progression by high-throughout fluorescence in situ hybridization on tissue microarrays. Cancer Res 59: 803–806 Palmberg C, Koivisto P, Kakkola L, Tammela TL, Kallioniemi OP, Visakorpi T 2000 Androgen receptor gene amplification at primary progression predicts response to combined androgen blockade as second line therapy for advanced prostate cancer. J Urol 164:1992–1995 Linja MJ, Savinainen KJ, Saramaki OR, Tammela TL, Vessella RL, Visakorpi T 2001 Amplification and overexpression of androgen receptor gene in hormone-refractory prostate cancer. Cancer Res 61:3550–3555 Gnanapragasam VJ, Robson CN, Leung HY, Neal DE 2000 Androgen receptor signalling in the prostate. Br J Urol Int 86:1001–1013 Culig Z, Klocker H, Bartsch G, Hobisch A 2002 Androgen receptors in prostate cancer. Endocr Related Cancer 9:155–170 Yeh S, Kang HY, Miyamoto H, Nishimura K, Chang HC, Ting HJ, Rahman M, Lin HK, Fujimoto N, Hu YC, Mizokami A, Huang KE, Chang C 1999 Differential induction of androgen receptor transactivation by different androgen receptor coactivators in human prostate cancer DU145 cells. Endocrine 11:195–202 Fujimoto N, Mizokami A, Harada S, Matsumoto T 2001 Different expression of androgen receptor coactivators in human prostate. Urology 58:289–294 Olapade-Olaopa EO, MacKay EH, Taub NA, Sandhu DP, Terry TR, Habib FK 1999 Malignant transformation of human prostatic epithelium is associated with the loss of androgen receptor immunoreactivity in the surrounding stroma. Clin Cancer Res 5:569–576 McEwan IJ 2004 Molecular mechanisms of androgen receptor-mediated gene regulation: structure-function analysis of the AF-1 domain. Endocr Relat Cancer 11: 281–293 Newmark JR, Hardy DO, Tonb DC, Carter BS, Epstein JI, Isaacs WB, Brown TR, Barrack ER 1992 Androgen receptor gene mutations in human prostate cancer. Proc Natl Acad Sci USA 89:6319–6323 Klocker H, Culig Z, Hobisch A, Cato AC, Bartsch G 1994 Androgen receptor alterations in prostatic carcinoma. Prostate 25:266–273 Evans BA, Harper ME, Daniells CE, Watts CE, Matenhelia S, Green J, Griffiths K 1996 Low incidence of androgen receptor gene mutations in human prostatic tumors using single strand conformation polymorphism analysis. Prostate 28:162–171 Marcelli M, Ittmann M, Mariani S, Sutherland R, Nigam R, Murthy L, Zhao Y, DiConcini D, Puxeddu E, Esen A, Eastham J, Weigel NL, Lamb DJ 2000 Androgen receptor mutations in prostate cancer. Cancer Res 60:944–949 Lamb DJ, Puxeddu E, Malik N, Stenoien DL, Nigam R, Saleh GY, Mancini M, Weigel NL, Marcelli M 2003 Molecular analysis of the androgen receptor in ten prostate cancer specimens obtained before and after androgen ablation. J Androl 24:215–225 Suzuki H, Sato N, Watabe Y, Masai M, Seino S, Shimazaki J 1993 Androgen receptor gene mutations in human prostate cancer. J Steroid Biochem Mol Biol 46: 759–765 Taplin ME, Bubley GJ, Shuster TD, Frantz ME, Spooner AE, Ogata GK, Keer HN, Balk SP 1995 Mutation of the androgen-receptor gene in metastatic androgen-independent prostate cancer. N Engl J Med 332:1393–1398 Taplin ME, Bubley GJ, Ko YJ, Small EJ, Upton M, Rajeshkumar B, Balk SP 1999 Selection for androgen


23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

receptor mutations in prostate cancers treated with androgen antagonist. Cancer Res 59:2511–2515 Wallen MJ, Linja M, Kaartinen K, Schleutker J, Visakorpi T 1999 Androgen receptor gene mutations in hormonerefractory prostate cancer. J Pathol 189:559–563 Haapala K, Hyytinen ER, Roiha M, Laurila M, Rantala I, Helin HJ, Koivisto PA 2001 Androgen receptor alterations in prostate cancer relapsed during a combined androgen blockade by orchiectomy and bicalutamide. Lab Invest 81:1647–1651 Hyytinen ER, Haapala K, Thompson J, Lappalainen I, Roiha M, Rantala I, Helin HJ, Janne OA, Vihinen M, Palvimo JJ, Koivisto PA 2002 Pattern of somatic androgen receptor gene mutations in patients with hormonerefractory prostate cancer. Lab Invest 82:1591–1598 Taplin ME, Rajeshkumar B, Halabi S, Werner CP, Woda BA, Picus J, Stadler W, Hayes DF, Kantoff PW, Vogelzang NJ, Small EJ, Cancer and Leukemia Group B Study 9663. 2003 Androgen receptor mutations in androgenindependent prostate cancer: Cancer and Leukemia Group B Study 9663. J Clin Oncol 21:2673–2678 Tilley WD, Buchanan G, Hickey TE, Bentel JM 1996 Mutations in the androgen receptor gene are associated with progression of human prostate cancer to androgen independence. Clin Cancer Res 2:277–285 Buchanan G, Greenberg NM, Scher HI, Harris JM, Marshall VR, Tilley WD 2001 Collocation of androgen receptor gene mutations in prostate cancer. Clin Cancer Res 7:1273–1281 Veldscholte J, Berrevoets CA, Ris-Stalpers C, Kuiper GG, Jenster G, Trapman J, Brinkmann AO, Mulder E 1992 The androgen receptor in LNCaP cells contains a mutation in the ligand binding domain which affects steroid binding characteristics and response to antiandrogens. J Steroid Biochem Mol Biol 41:665–669 Matias PM, Donner P, Coelho R, Thomaz M, Peixoto C, Macedo S, Otto N, Joschko S, Scholz P, Wegg A, Basler S, Schafer M, Egner U, Carrondo MA 2000 Structural evidence for ligand specificity in the binding domain of the human androgen receptor. Implications for pathogenic gene mutations. J Biol Chem 275:26164–26171 Sack JS, Kish KF, Wang C, Attar RM, Kiefer SE, An Y, Wu GY, Scheffler JE, Salvati ME, Krystek Jr SR, Weinmann R, Einspahr HM 2001 Crystallographic structures of the ligand-binding domains of the androgen receptor and its T877A mutant complexed with the natural agonist dihydrotestosterone. Proc Natl Acad Sci USA 98:4904–4909 Mayer MP, Rudiger S, Bukau B 2000 Molecular basis for interactions of the DnaK chaperone with substrates. J Biol Chem 381:877–885 Kallio PJ, Janne OA, Palvimo JJ 1994 Agonists, but not antagonists, alter the conformation of the hormone-binding domain of androgen receptor. Endocrinology 134: 998–1001 Kuil CW, Berrevoets CA, Mulder E 1995 Ligand-induced conformational alterations of the androgen receptor analyzed by limited trypsinization. Studies on the mechanism of antiandrogen action. J Biol Chem 270: 27569–27576 He B, Wilson EM 2002 The NH2-terminal and carboxylterminal interaction in the human androgen receptor. Mol Genet Metab 75:293–298 Steketee K, Berrevoets CA, Dubbink HJ, Doesburg P, Hersmus R, Brinkmann AO, Trapman J 2002 Amino acids 3–13 and amino acids in and flanking the 23FxxLF27 motif modulate the interaction between the N-terminal and ligand-binding domain of the androgen receptor. Eur J Biochem 269:5780–5791 Darimont BD, Wagner RL, Apriletti JW, Stallcup MR, Kushner PJ, Baxter JD, Fletterick RJ, Yamamoto KR 1998 Structure and specificity of nuclear receptor-coactivator interactions. Genes Dev 12:3343–3356


38. McInerney EM, Rose DW, Flynn SE, Westin S, Mullen TM, Krones A, Inostroza J, Torchia J, Nolte RT, AssaMunt N, Milburn MV, Glass CK, Rosenfeld MG 1998 Determinants of coactivator LXXLL motif specificity in nuclear receptor transcriptional activation. Genes Dev 12:3357–3368 39. Kleinerman DL, Troncosa P, Pisters LL, Navone NM, Hsieh JT, Logothetis CJ, von Eschenbach AC, Sleddens HFBM, van der Kwast TH, Brinkmann AO, Schroder FH, Trapman J 1996 J Urol 155:624 40. Tan J, Sharief Y, Hamil KG, Gregory CW, Zang DY, Sar M, Gumerlock PH, DeVere White RW, Pretlow TG, Harris SE, Wilson EM, Mohler JL, French FS 1997 Dehydroepiandrosterone activates mutant androgen receptors expressed in the androgen-dependent human prostate cancer xenograft CWR22 and LNCaP cells. Mol Endocrinol 11:450–459 41. Fenton MA, Shuster TD, Fertig AM, Taplin ME, Kolvenbag G, Bubley GJ, Balk SP 1997 Functional characterization of mutant androgen receptors from androgenindependent prostate cancer. Clin Cancer Res 3:1383–1388 42. Shi XB, Ma AH, Xia L, Kung HJ, de Vere White RW 2002 Functional analysis of 44 mutant androgen receptors from human prostate cancer. Cancer Res 62:1496–1502 43. Hara T, Miyazaki J, Araki H, Yamaoka M, Kanzaki N, Kusaka M, Miyamoto M 2003 Novel mutations of androgen receptor: a possible mechanism of bicalutamide withdrawal syndrome. Cancer Res 63:149–153 44. Chen CD, Welsbie DS, Tran C, Baek SH, Chen R, Vessella R, Rosenfeld MG, Sawyers CL 2004 Molecular determinants of resistance to antiandrogen therapy. Nat Med 10:33–39 45. Chavez B, Mendez JP, Ulloa-Aguirre A, Larrea F, Vilchis F 2001 Eight novel mutations of the androgen receptor gene in patients with androgen insensitivity syndrome. J Hum Genet 46:560–565 46. Zhao XY, Malloy PJ, Krishnan AV, Swami S, Navone NM, Peehl DM, Feldman D 2000 Glucocorticoids can promote androgen-independent growth of prostate cancer cells through a mutated androgen receptor. Nat Med 6:703–706 47. Mizokami A, Koh E, Fujita H, Maeda Y, Egawa M, Koshida K, Honma S, Keller ET, Namiki M 2004 The adrenal androgen androstenediol is present in prostate cancer tissue after androgen deprivation therapy and activates mutated androgen receptor. Cancer Res 64:765–771 48. Guex N, Peitsch MC 1997 SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 18:2714–2723 49. McDonald S, Brive L, Agus DB, Scher HI, Ely KR 2000 Ligand responsiveness in human prostate cancer: structural analysis of mutant androgen receptors from LNCaP and CWR22 tumors. Cancer Res 60:2317–2322 50. Simental JA, Sar M, Lane MV, French FS, Wilson EM 1991 Transcriptional activation and nuclear targeting sig-


51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

nals of the human androgen receptor. J Biol Chem 266: 510–518 Jenster G, van der Korput HA, Trapman J, Brinkmann AO 1995 Identification of two transcription activation units in the N-terminal domain of the human androgen receptor. J Biol Chem 270:7341–7346 He B, Lee LW, Minges JT, Wilson EM 2002 Dependence of selective gene activation on the androgen receptor NH2- and COOH-terminal interaction. J Biol Chem 277: 25631–25639 Hsu CL, Chen YL, Yeh S, Ting HJ, Hu YC, Lin H, Wang X, Chang C 2003 The use of phage display technique for the isolation of androgen receptor interacting peptides with (F/W)XXL(F/W) and FXXLY new signature motifs. J Biol Chem 278:23691–23698 Dubbink HJ, Hersmus R, Verma CS, van der Korput HA, Berrevoets CA, van Tol J, Ziel-van der Made AC, Brinkmann AO, Pike AC, Trapman J 2004 Distinct recognition modes of FXXLF and LXXLL motifs by the androgen receptor. Mol Endocrinol 18:2132–2150 He B, Gampe Jr RT, Kole AJ, Hnat AT, Stanley TB, An G, Stewart EL, Kalman RI, Minges JT, Wilson EM 2004 Structural basis for androgen receptor interdomain and coactivator interactions suggests a transition in nuclear receptor activation function dominance. Mol Cell 16: 425–438 Hur E, Pfaff SJ, Payne ES, Gron H, Buehrer BM, Fletterick RJ 2004 Recognition and accommodation at the androgen receptor coactivator binding interface. PLoS Biol 2:E274 Gregory CW, He B, Johnson RT, Ford OH, Mohler JL, French FS, Wilson EM 2001 A mechanism for androgen receptor-mediated prostate cancer recurrence after androgen deprivation therapy. Cancer Res 61:4315–4319 Brinkmann AO, Faber PW, van Rooij HC, Kuiper GG, Ris C, Klaassen P, van der Korput JA, Voorhorst MM, van Laar JH, Mulder E 1989 The human androgen receptor: domain structure, genomic organization and regulation of expression. J Steroid Biochem 34:307–310 Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248–254 Blok LJ, Themmen AP, Peters AH, Trapman J, Baarends WM, Hoogerbrugge JW, Grootegoed JA 1992 Transcriptional regulation of androgen receptor gene expression in Sertoli cells and other cell types. Mol Cell Endocrinol 88:153–164 Berrevoets CA, Doesburg P, Steketee K, Trapman J, Brinkmann AO 1998 Functional interactions of the AF-2 activation domain core region of the human androgen receptor with the amino-terminal domain and with the transcriptional coactivator TIF2 (transcriptional intermediary factor 2). Mol Endocrinol 12:1172–1183

Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.