| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Lady Davis Institute for Medical Research, Sir Mortimer B. Davis Jewish General Hospital (S.A.G., B.G., R.L., L.K.B., Y.E., L.P., M.A.T.); and Departments of Human Genetics (S.A.G., L.P., M.A.T.), Oncology (J.W.), Biology (L.P.), Medicine (L.P., M.A.T.), and Pediatrics (L.P.), McGill University, Montréal, Québec, Canada H3T 1E2
Address all correspondence and requests for reprints to: B. Gottlieb, Ph.D., Lady Davis Institute for Medical Research, Sir Mortimer B. Davis Jewish General Hospital, 3755 Cote Ste. Catherine Road, Montreal, Quebec, Canada H3T 1E2. E-mail: bruce.gottlieb{at}mcgill.ca.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Using mammalian two-hybrid assays, hAR amino/carboxyl (N/C)-terminal interactions of the mutant receptors were analyzed in the presence and absence of the hAR coactivator transcription intermediary factor 2 (TIF2). The mutations conferred decreased hAR N/C-terminal interaction, i.e. mild (
1.5-fold), partial (2-fold), and complete (10-fold), that mirrored the degree of AIS. All mutant LBDs showed a 2- to 3-fold increase in N/C-terminal interactions when TIF2 was cotransfected, although of a magnitude still less than that of wild-type LBD with TIF2.
The ligand-selective properties of the Ser814Asn mutant were also clearly reflected by the N/C-terminal interactions. Thus, measurement of N/C-terminal interactions may assist in the molecular analysis of mutant hARs associated with AIS.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The steroid receptors belong to the larger family of nuclear receptors that have a well characterized and conserved modular structure (2, 3). The hAR is comprised of a polymorphic N-terminal domain (NTD), a central DNA-binding domain (DBD), and a well conserved C-terminal ligand-binding domain (LBD). The hAR NTD harbors hormone-independent transcription activation function 1 (AF1) (4). Conserved FXXLF and WXXLF motifs also reside within the hAR NTD (5), forming in part the interface for the interaction of the NTD with hormone-dependent AF2 in the LBD (6).
Evidence supporting direct interaction between amino- and carboxyl-terminal regions of the hAR comes from mammalian two-hybrid and glutathione-S-transferase fusion protein studies indicating that the amino-terminal sequence FXXFL motif [amino acids (aa) 23–27] mediates interaction with the AF2 region of the LBD in an androgen-dependent manner (5). In contrast, several lines of evidence point to an indirect N/C interaction, mediated or at least partially influenced by the family of p160 coactivators. There is evidence to demonstrate the interaction between p160 coactivators and the NTDs of the estrogen receptor (7), progesterone receptor (8), and AR (9) in addition to a second interaction with their respective LBDs. Moreover, the AF-1 activity of each of these nuclear receptors was enhanced by this p160 interaction, indicating that AF-1, similar to AF-2, works at least partially through p160 recruitment. Furthermore, Ma and colleagues (9) observed positive functional interactions between the AR NTD and a C-terminal fragment of glucocorticoid receptor interacting protein 1 (GRIP1), suggesting a mechanism for AF1/AF2 coordination whereby p160 coactivators bridge communications between the NTD and LBD through direct contacts. This hypothesis is further supported by experiments demonstrating p160-mediated enhancement of the ligand-dependent, cooperative trans-activation of target genes (10).
The LBDs of steroid receptors also stabilize homodimerization and through their activation domains recruit coactivator molecules that act as bridging factors to preinitiation complexes and RNA polymerase II, thereby directing the transcriptional activation of ligand-responsive genes. Mutations that selectively interfere with the cooperation between the NTD and the LBD dramatically impair receptor activity (11, 12). Mutational studies in the estrogen, progesterone, retinoic acid, and thyroid hormone receptors demonstrate that the AF2 region is integral to the function of a family of closely related p160 coactivator proteins including N-acetyl coenzyme A1/steroid receptor coactivator 1 (SRC1), GRIP1/transcription intermediary factor 2 (TIF2)/N-acetyl coenzyme A2/SRC2, and CIP/RAC3/SRC3/AIB1/ACTR/TRAM1 (13). These proteins interact with the LBD through distinct, conserved LXXLL motifs and with the NTD through regions outside of these motifs (6). Through their histone acetyl-transferase activity, the p160 coactivators increase the transcriptional activation of target genes (14, 15).
The rat p160 coactivator TIF2 and its mouse homolog GRIP1 have been shown to interact through three nuclear receptor boxes (LXXLL motifs) with the coactivator-interacting core signature sequence in the AR LBD (12, 15, 16) and, independently of these motifs, with the AF1 region of the AR (6, 9). However, unlike other steroid receptors, the AR shows clear preference for certain LXXLL motifs (17). Mutations of the nuclear receptor boxes completely eliminated the coactivator function of TIF2 on many nuclear receptors, including the AR (18). These motifs thus orchestrate AR N/C-terminal interactions that are crucial to the functional integrity of the receptor.
Recent crystallization of the AR LBD (19) has confirmed the previously proposed LBD structure and offers great potential for furthering our understanding of AR dynamics. We have elucidated the molecular consequences of five distinct AR LBD mutations found in various patients with AIS. These aa substitutions are either in helix 5 or 9 or the region between helixes 7 and 8.
In this study we report that defects in N/C-terminal interactions offer an explanation for the pathogenicity of these mutations despite normal or near-normal androgen binding kinetics. We also show a strong correlation between the strength of N/C-terminal interactions and the corresponding clinical phenotype, stressing the importance of the structural conformation of the receptor.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Subject 4007, a Caucasian boy, was first seen at 18 yr of age with bilateral gynecomastia and impaired virilization (little facial hair, no chest hair, and high pitched voice) (20). Testicular volume was 15 ml bilaterally, and pubic hair was Tanner stage IV. His total plasma testosterone was 49.4 nM (normal, 12–35 nM), and his LH was 37.5 IU/liter (normal, 4–23 IU/liter). Sperm analysis on a semen sample revealed a density of 35 million/ml, but 60% were dead, 80% were immotile at 2.5 h, 13% were tapered, and 11% were amorphous. When examined 1 yr later, the total plasma testosterone level was 51 nM, and LH was 13.9 IU/liter, with a similar degree of impaired virilization. A repeat sperm analysis revealed a density of only 5 million/ml; 90% were dead, none was motile at 25 h, and 17% were amorphous. No family history was reported. The patient was diagnosed with MAIS (21).
Subject 2: 41560; S814N
Subject 41560 first presented at age 22 yr with a small penis (5 x 2.5 cm), infertility, and increased T and LH levels (22). His puberty (onset, 12 yr) generated little facial or body hair, and he developed gynecomastia, but normal axillary hair. Testicular volume was 10 ml, and pubic hair was Tanner stage III. His plasma testosterone level varied from 55–75 nM, and his basal plasma LH level ranged from 30–33 IU/liter. Sperm count was 15.5 million/ml; 44% were motile, and 79% had abnormal morphology. He had a positive family history, being the maternal first cousin to two pairs of similarly affected brothers (23). The patient was diagnosed with partial androgen insensitivity (PAIS) (21).
Subject 3: 3287; E772A
Subject 3287, an Inuit Eskimo boy, presented at 2 yr and 3 months of age because of severe perineal hypospadias with chordee (20). The scrotum was fused but empty, and the testes were palpable at the external inguinal rings. Pelvic ultrasonography did not reveal any female internal organs. At 3 yr and 5 months of age, his total serum testosterone level was less than 1 nM, and his LH and FSH values were 4.8 and 7.9 IU/liter, respectively. No family history was reported. The patient was diagnosed with PAIS (21).
Subject 4: KIL; V866M
Subject KIL, of Scottish origin, presented at 3.5 months, when she was found to have testes in bilateral hernial sacs and normal female external genitalia (24, 25, 26). At 17 months, examination showed a vagina 1–2 cm long, but no cervix or uterus. No family history was reported. The patient was diagnosed with CAIS (21).
Subject 5: LEL; R774C
Subject LEL, whose mother is French-Canadian, was 10 yr old when she presented with inguinal testes. They were normally differentiated but immature, with segments of vas deferens attached. Her external female genitalia were normal (27). She had a positive family history, with an older sister who had primary amenorrhea and presented for inguinal herniorrhaphy at age 15 yr and a maternal aunt with menarche at 18 yr of age. The patient was diagnosed with CAIS (21).
All five patients had normal 46,XY male karyotypes.
Cell culture
The genital skin fibroblast (GSF) strains were developed in our laboratory from small pieces of labium majus, scrotal, or preputial skin taken from subjects and normal controls with informed consent according to protocols approved by the hospital ethics committee. Normal and mutant GSFs as well as transfected COS-1 and CV-1 monkey kidney cells were maintained in Eagle’s MEM supplemented with 5–10% fetal bovine serum, 1 mM pyruvate, 10 mg/liter garamycin, 60 mg/liter penicillin G, and 60 mg/liter streptomycin.
Kinetic properties of AR mutant strains
Specific androgen-binding activity, maximum androgen-binding capacity, apparent equilibrium dissociation constants (Kd), and rate constant of dissociation (k) of various androgen-AR complexes were determined on monolayer cultures as described previously (20, 28). Finasteride (10-7 M) was added to the cultures to inhibit endogenous 5
-reductase activity that might degrade androgens.
Expression and reporter constructs
The p virus transcriptional regulator protein 16 (pVP16)pVP16-AR (TAD20Q) plasmid, coding for the N-terminal trans-activation domain (NTD) of wild-type AR (aa 1–565) fused in-frame with the VP16 activation domain (VP16AD), the pM-AR(LBD) plasmid coding for hAR aa 659–919 fused in-frame with the galactosidase 4 (GAL4)-DBD, and the pGAL4-luciferase (Luc) reporter vector containing GAL4-DNA-binding sites and a Luc reporter gene were provided by Dr. E. L. Yong (National University of Singapore) (29). The pVPhAR plasmid coding for full-length hAR fused in-frame with the VP16AD was provided by Dr. E. Wilson (University of North Carolina, Chapel Hill, NC) (30).
Each LBD mutation was recreated in the pSVhAR.BHEXE expression vector (31, 32) by the overlap extension method (33) using appropriate primers. Each mutation was then transferred to the pcDNA3 expression vector by double-digesting each mutant pSVhAR.BHEXE vector with NheI/BamHI and religating the AR cDNA into similarly digested pcDNA3.
The mammalian two-hybrid pM-AR(LBD) mutant expression constructs were created by a double digest of the mutant pcDNA3-hAR constructs with either XbaI/TthIII or XbaI/BstBI and religating in-frame to the similarly digested wild-type pM-AR(LBD) expression vector. To show that the appropriate mutation was incorporated, the pM-AR(LBD) LEL (R774C) construct was digested with ScaI to verify the creation of a new restriction endonuclease site resulting from the C
T substitution mutation at codon 774. The remaining four mutant expression constructs were sequenced to confirm the fidelity of the cloning procedures.
The pVP16AD-TIF2 plasmid was constructed by performing a double digest of pSG5-TIF2 with HindIII/XbaI, excising a fragment of TIF2 encoding the three conserved LXXLL signature sequences (aa 632–1716), followed by in-frame ligation downstream of the pVP16 activator domain. The cytomegalovirus promoter (pCMV)-ßgal, mouse mammary tumor virus promoter (pMMTV)-GH, and MMTV-Luc reporter constructs and the expression pSG5-TIF2 vector have previously been described (29).
Mammalian cell culture and N/C-terminal interaction assays
CV-1 cells were cultured in OptiMEM (Life Technologies, Inc., Grand Island, NY) supplemented with 5% fetal bovine serum. hAR N/C-terminal interaction assays were performed in vivo in a mammalian cellular environment, using the mammalian two-hybrid system. In brief, 2 x 106 CV-1 cells were seeded in T-25 flasks (Nalgene, Nunc International Corp., Naperville, IL) 24 h before transfection. Just before transfection, the cells were rinsed twice with PBS and then transfected with the appropriate DNA constructs. Four micrograms of reporter plasmid, pGAL4-Luc, 0.8 µg pCMV-ßgal (34), and the indicated amounts of pVP16-AR(TAD20Q) and the various pM-AR(LBD) expression constructs were combined with 30 µl Lipofectamine (Life Technologies, Inc.) according to the manufacturer’s procedure. Twenty-four hours after transfection, the transfected cells were replated in four 35-mm tissue culture wells, two of which were treated with 3 nM ligand [mibolerone (MB) or methyltrenolene (MT; R1881), synthetic nonmetabolizable androgens, testosterone (T) or dihydrotestosterone (DHT)]. Seventy-two hours after transfection, the cells were harvested, and then lysed in 1x Reporter Lysis Buffer (Promega Corp., Madison, WI). The cell debris was centrifuged, and 20 µl of the supernatant were added to 50 µl of the Luc reagent to quantify Luc activity according to Promega Corp. protocols. The total protein content (Pierce Chemical Co., Rockford, IL) and ß-galactosidase activity (35) using BCA protein assay reagent were subsequently determined to control for transfection efficiency.
Immunoblot analyses of extracts from transfected cells
Transfected COS-1 cells were washed twice with PBS, and the pellets were resuspended in SDS-PAGE sample buffer and heated at 95 C for 5 min. The proteins were separated 8% SDS-polyacrylamide gels, then transferred onto a nitrocellulose membrane (Amersham Pharmacia Biotech, Arlington Heights, IL), and immunoblotted with the mouse monoclonal antibody, SC510 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) directed against GAL4-DBD fusion proteins or the rabbit polyclonal antibody, PG-21, which recognizes the first 21 N-terminal aa of the human AR. Horseradish peroxidase-conjugated sheep antimouse or goat antirabbit IgG secondary antibodies (Zymed Laboratories, Inc., South San Francisco, CA), were used to detect the immunocomplexes with enhanced chemiluminescence Western blot reagents (Amersham Pharmacia Biotech) according to the manufacturer’s instructions.
Trans-activation assays
To determine the trans-activational properties of each of the mutant hAR strains, 4 µg of the pMMTV-GH reporter vector were cotransfected with 1 µg of each of the mutant pcDNA AR expression vectors along with 1 µg pCMV-ßGAL to assess transfection efficiency. GH concentrations in the culture medium were measured using an hGH ELISA kit (MEDICORP, Inc., Montréal, Canada). GH values were normalized for cellular protein, determined by Lowry assays, and for transfection efficiency, determined by ßGAL assays.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
To investigate the molecular mechanisms underlying the androgen-insensitive phenotypes of five patients with AR LBD point mutations, these mutations were recreated in mammalian expression vectors. The locations of these LBD mutations are in the following helixes: E772A and R774C in helix 5, S814N in the interhelical region between helixes 7 and 8, and R866M and R871G in helix 9, as shown in Fig. 1
. The biochemical phenotype of these mutant AR receptors in GSFs and transfected COS-1 cells are listed in Table 1. Whole cell binding assays in COS-1 cells showed that AR mutants R871G, S814N, and E772A possess normal Kd values for MB and MT (data for MT not shown), whereas V866M displayed slightly elevated Kd values (3-fold increase compared with normal).
|
|
-DHT (data for MB and DHT not shown), but both E772A and V866M displayed abnormally elevated k values. R774C had negligible binding activity (i.e. receptor negative) at 37 C. S814N was unusual in that it displayed ligand selective k values; the values were elevated for MB and DHT, but not for MT (22).
The trans-activation ability of each of the mutant ARs was assessed using the pMMTV-GH reporter in COS-1 cells in the presence of the synthetic ligand MB (Fig. 2). All five mutant hARs trans-activated the GH reporter gene to a lesser extent than the normal AR: R871G, 60%; S814N, 80%; E772A, 40%; V866M, 60%; and R774C, 0%. Of the five mutants, S814N was most comparable to the wild-type receptor. To further characterize the mutants we examined the interaction of TIF2 with receptor from patients with two clearly different phenotypes, namely R871G that has MAIS, and E772A that has PAIS (Fig. 3). As might be expected, TIF2 did a much better job of rescuing R871G than E772A, going from 50% of wild-type trans-activation activity in the absence of TIF2 to 75%. On the other hand, with E772A, addition of TIF2 resulted in trans-activation activity that was only 25% the activity of the wild-type AR in the presence of TIF2.
|
|
To further study the molecular effects of LBD mutations, AR N/C-terminal interactions were measured using the mammalian two-hybrid system. All five mutants showed severely reduced N/C-terminal interactions compared with wild-type AR (Fig. 4, 
). R871G (MAIS), S814N (PAIS), and E772A (PAIS) LBDs showed approximately 25% of the wild-type N/C-terminal interaction (Fig. 4, iv, v, and vi), V866M (CAIS) showed only 10% (Fig. 4, vii), and R774C (CAIS) showed only 4% of wild-type activity (the value is too small to be shown in Fig. 4, viii).
|
, 
). In every case, the strength of the N/C-terminal interaction correlated closely with the phenotypic degree of androgen insensitivity, as we observed only 68% of wild-type N/C interaction for R871G (Fig. 4, iv), 48% for S814N (Fig. 4, v), 50% for E772A (Fig. 4, vi), 12% for V866M (Fig. 4, vii), and negligible activity for R774C (Fig. 4, viii). Full-length constructs were also used because there are several areas within the full-length AR that could participate in stabilizing N/C-terminal interactions, interactions that would not be measured using N-terminal constructs alone.
Coexpression of TIF2, a p160 coactivator, enhanced the N/C-terminal interaction of three of the receptor LBDs assayed by approximately 3- to 12-fold, but not, interestingly enough, E772A (Fig. 5, vi) and R774C, which showed negligible N/C-terminal interaction initially (Fig. 5, viii). In the presence of TIF2, R871G showed only 38% of the wild-type N/C-terminal interaction (Fig. 5, iv), S814N showed 25% (Fig. 5, v), E772A showed 5% (Fig. 5, vi), V866M showed 3% (Fig. 5, vii), and R774C had negligible activity (Fig. 5, viii). Cell extracts of COS-1 cells transfected with the GAL4-LBD constructs were immunoblotted with a GAL4 antibody. This revealed that the encoded proteins were expressed at similar levels, suggesting that impaired N/C-terminal interactions were not the result of impaired protein expression (Fig. 6).
|
|
), the exception being R774C, which showed negligible TIF2/LBD interaction.
|
Previous characterization of the S814N mutant as a ligand-selective receptor (22, 38) led us to further investigate the nature of this mutation by testing its N/C-terminal interaction in the presence of four different hormones: T, DHT, MB, and MT. In each case in the presence and absence of TIF2, the S814N mutant displayed between 40–60% of wild-type N/C-terminal interactions (Fig. 8). Upon the addition of TIF2, N/C-terminal interactions of the wild-type AR increased 3.8- to 5.4-fold in the presence of all hormones, while N/C interactions of the mutant receptor increased 3.4- to 6.5-fold for all hormones except T. In the presence of T, TIF2 failed to rescue the S814N N/C-terminal interactions, as the increase in activity was only 1.3-fold (Fig. 8). The ability of TIF2 to rescue the S814N mutant AR N/C-terminal interactions in the presence of DHT, MB, and MT, but not T, further illustrates the ligand-selective nature of this mutation.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
1.5-fold), partial (2-fold), and complete (10-fold), that mirrored the degree of AIS. This interaction was abolished completely for the R774C mutation, which confers complete AIS. These results stress the importance of N/C-terminal interactions in the functional integrity of the AR and support the idea that disrupted N/C interaction is a key to molecular defects in many AIS patients in whom LBD mutations have not resulted in severe impairment of androgen-binding ability. Interestingly, the use of full-length hAR with the mutant LBDs increased interaction, but in a way that mirrored the mutant phenotypes. This suggests that other areas of the AR besides the N- and C-terminal regions may be involved in stabilizing protein-protein interactions. These results are particularly provocative in light of the continuing discussion of the possible N/C-terminal interactions in AR dimers and their parallel or antiparallel nature (30). Overexpression of TIF2 was shown to partially rescue N/C interaction for three of the mutants, but not E772A and R774C. Additional TIF2, however, failed to return N/C-interactions to wild-type levels even for those mutants that had increased N/C-terminal interaction in the presence of TIF2. When direct TIF2/LBD interactions were tested, there appeared to be no direct relationship between the strength of the TIF2/LBD interaction and the degree of androgen insensitivity conferred by the specific AR gene mutations. These results suggest that the reduced transcriptional activity of the mutants is mainly due to compromised N/C interaction rather than impaired TIF2 interaction.
Earlier studies have shown the importance of helixes 3 and 4 of nuclear receptor LBDs combined with helix 12 in the formation of a helical loop that functions as a highly complementary recognition site for the LXXLL motifs of coactivators (12). Furthermore, studies of the F725L mutation, found in a patient with PAIS, show severely impaired N/C-terminal interaction and trans-activation (41). F725 resides in helix 3 and the corresponding residue F367 in estrogen receptor are reported to form part of the interface for the recognition by GRIP1 nuclear box II peptide (40). This suggests a functional importance, with the structural conformation of helixes 3, 4, and 12 forming an interface for p160 coactivator interaction. Our results suggest that helix 5, which harbors both E772 and R774C residues, might also be involved in this helical loop structure. TIF2 failed to rescue the N/C-terminal interactions of these two mutations, which suggests that the problem may lie in the lack of interaction of the mutant LBDs with the p160 coactivator. These two residues might be integral to the coordination of helix positioning, and changing the Gly or Arg to Ala or Cys at position 772 or 774, respectively, may upset the structure, disrupting the formation of the coactivator interface binding site. Interestingly, a mutation in aa N727K was recently shown to lead to a similar disruption of interaction between receptor domains and GRIP1/TIF2 (12). Our results suggest that impairment of the N/C-terminal interaction in all five mutant hARs could be a result of or could result in disruptions in the efficient recruitment of coactivators that normally bind to the NTD of the AR in a hormone-dependent manner.
The study of these mutations is especially useful because they can illuminate functional subdomains that reside in the LBD that do not mediate androgen binding, but are important for interactions with coactivators and subsequent coordination of intramolecular interactions that bring the AR into the necessary structural conformation for efficient transcriptional activity. In fact, an earlier study describes an additional mutation, M886V, that also preserved ligand binding while disrupting NTD-LBD and LBD-TIF2 interactions (29). This functional dichotomy, which separates regions for ligand binding from regions for coactivator binding, is not implausible considering that intermolecular interactions depend on surface properties, whereas the ligand binding pockets are buried within the hydrophobic cores of all steroid receptor LBDs crystallized to date (29). Mutations may therefore affect many AR processes, including NTD-LBD and LBD-LBD interactions, DNA binding, and trans-activation competence. This is consistent with the integrative role that coactivators such as TIF2 may have in these postligand-binding events.
An earlier study investigating the molecular consequences of seven substitution mutations in the LBD of the hAR demonstrated that helixes 3, 4, and 12 are critical in mediating N/C-terminal interactions (41). Our results show, however, that the regions of the LBD critical to N/C interactions are not necessarily rigidly defined and may be extended to AR helix 9 (V866M and R871G), helix 5 (E772A and R774C), and the interhelical region between helixes 7 and 8 (S814N). In fact, it is plausible that every helix in the LBD is important in providing the exact structural conformation necessary for the functional integrity of the AR and that the helixes directly responsible for N/C-terminal interaction should be more broadly defined.
When the S814N mutant was tested with different androgens, both N/C-terminal interactions, and the ability of TIF2 to rescue this interaction differed immensely depending on the androgen used. The N/C interaction was minimal in the presence of T, whereas the bulkier androgens, DHT, MB, and MT, markedly increased the N/C-terminal interaction. Similar results were observed with TIF2, where TIF2 failed to rescue the N/C-terminal interaction in the presence of T, but increased this interaction between 3- to 6-fold in the presence of the remaining three androgens. Thus, our data show that the S814 residue is one of the aa critical in distinguishing responses to different androgens. Consistent with these data, the ligand-specific nature of the M742V mutant, which differs in transcriptional ability depending on the ligand used, displayed maximal N/C-terminal interaction in the presence of DHT, MB, or MT and minimal activity in the presence of T (36). DHT and MB, but not T, appear to be able to elicit a conformational change in the mutant receptor.
Furthermore, the restorative effects of MB and MT in vitro raise the possibility of directed hormonal therapy. In fact, empirical treatment with the androgen analog mesterolone (1-methyl-DHT, 1-methyl-17ß-hydroxy-5-androstan-3-one) in another subject with partial AIS harboring the AR N727K mutation was associated with marked improvement in sperm parameters, impregnation, and delivery of a healthy child (12, 42). Withdrawal of mesterolone therapy was associated with reversion to poor sperm production, suggesting that mesterolone had a corrective effect on this mutant AR in vivo. Our data further support an emerging paradigm with respect to AR mutations in the LBD and male infertility; pathogenicity is mediated via reduced interdomain and coactivator interactions, and androgen analogs that are corrective in vitro may be indicated for potential hormonal therapy (12).
Finally, in assessing the possible clinical value of such an approach, it is important to note, as we have previously, that the prospects for therapy and management at least for PAIS and MAIS will be clearly dependent on a thorough understanding of the exact mechanism of androgen insensitivity in each subject (21). This has been reflected in the fact that long-term androgen therapy has only produced significant results when the specific effects of the mutant AR are well understood (43). In the present study it is interesting to note that in the three subjects who had either PAIS or MAIS, a substantial improvement in the trans-activational activity of their ARs was achieved by adding the p160 coactivator TIF2. Further, in the case of the ligand-selective mutant AR the addition of one synthetic hormone together with TIF2 also had a substantial effect on its trans-activational activity. Although further work will be required to firmly establish the observations of this study, it is hoped that the present approach may add another element to the therapeutic options available in the clinical treatment of AIS.
|
Footnotes |
|---|
Received August 19, 2002.
Accepted January 29, 2003.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
and ß. Mol Cell Biol 19:8226–8239-dihydrotestosterone-binding activity in labium majus skin fibroblasts. Am J Med Genet 4:401–441[CrossRef][Medline]
-helical segments of the DNA-binding domain in the human androgen receptor. Hum Mol Genet 3:21–27This article has been cited by other articles:
 |
|
 |
|
  S. Bai and E. M. Wilson Epidermal Growth Factor-Dependent Phosphorylation and Ubiquitinylation of MAGE-11 Regulates Its Interaction with the Androgen Receptor Mol. Cell. Biol., March 15, 2008; 28(6): 1947 - 1963. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. B. Askew, R. T. Gampe Jr., T. B. Stanley, J. L. Faggart, and E. M. Wilson Modulation of Androgen Receptor Activation Function 2 by Testosterone and Dihydrotestosterone J. Biol. Chem., August 31, 2007; 282(35): 25801 - 25816. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Kohler, A.-L. Delezoide, B. Boizet-Bonhoure, M. J McPhaul, C. Sultan, and S. Lumbroso Coexpression of Wilms' tumor suppressor 1 (WT1) and androgen receptor (AR) in the genital tract of human male embryos and regulation of AR promoter activity by WT1 J. Mol. Endocrinol., May 1, 2007; 38(5): 547 - 554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. M. Wilson Muscle-Bound? A Tissue-Selective Nonsteroidal Androgen Receptor Modulator Endocrinology, January 1, 2007; 148(1): 1 - 3. [Full Text] [PDF]
|
|
|
|
|
|
J Jaaskelainen, A Deeb, J W Schwabe, N P Mongan, H Martin, and I A Hughes Human androgen receptor gene ligand-binding-domain mutations leading to disrupted interaction between the N- and C-terminal domains. J. Mol. Endocrinol., April 1, 2006; 36(2): 361 - 368. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. A. Elhaji, I. Stoica, S. Dennis, E. O. Purisima, and M. A. Trifiro Impaired helix 12 dynamics due to proline 892 substitutions in the androgen receptor are associated with complete androgen insensitivity Hum. Mol. Genet., March 15, 2006; 15(6): 921 - 931. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Yoshikawa, K. Yamamoto, N. Shimizu, S. Yamada, C. Morimoto, and H. Tanaka The Distinct Agonistic Properties of the Phenylpyrazolosteroid Cortivazol Reveal Interdomain Communication within the Glucocorticoid Receptor Mol. Endocrinol., May 1, 2005; 19(5): 1110 - 1124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-L. Hsu, Y.-L. Chen, H.-J. Ting, W.-J. Lin, Z. Yang, Y. Zhang, L. Wang, C.-T. Wu, H.-C. Chang, S. Yeh, et al. Androgen Receptor (AR) NH2- and COOH-Terminal Interactions Result in the Differential Influences on the AR-Mediated Transactivation and Cell Growth Mol. Endocrinol., February 1, 2005; 19(2): 350 - 361. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Umar, C. A. Berrevoets, N. M. Van, M. van Leeuwen, M. Verbiest, W. J. Kleijer, D. Dooijes, J. A. Grootegoed, S. L. S. Drop, and A. O. Brinkmann Functional Analysis of a Novel Androgen Receptor Mutation, Q902K, in an Individual with Partial Androgen Insensitivity J. Clin. Endocrinol. Metab., January 1, 2005; 90(1): 507 - 515. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. A. Elhaji, J. Hui Wu, B. Gottlieb, L. K. Beitel, C. Alvarado, G. Batist, and M. A. Trifiro An Examination of How Different Mutations at Arginine 855 of the Androgen Receptor Result in Different Androgen Insensitivity Phenotypes Mol. Endocrinol., August 1, 2004; 18(8): 1876 - 1886. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||
or 
) LBDs (1.5 µg), pVP16-TAD (aa 1–565, an AR:VP16 AD fusion protein), pGAL-4-Luc (4 µg), and pCMV-ßgal (800 ng), with (+) or without (-) pSG5-TIF2 (1.5 µg). For TIF2 negative cells, 1.5 µg pcDNA3 were added. Transfected cells were incubated in the presence or absence of 3 nM T (i), DHT (ii), MB (iii), or MT (iv). Twenty-four hours after transfection, the cells were washed and received fresh medium with or without the appropriate androgen, and the culture was continued for an additional 48 h. Luc activity was corrected for transfection efficiency and protein levels. Normalized Luc values are expressed relative to that of wild-type LBD in the presence of MB but without TIF2 (WT LBD + NTD +MB = 100%). The data show the mean of four different sets of experiments performed in duplicate. The fold increase in N/C-terminal interaction elicited by full-length TIF2 is depicted above the relevant bar.