Transcriptional activity of androgen receptor is modulated by two RNA splicing factors, PSF and p54nrb

Abstract

Nuclear receptors regulate gene activation or repression through dynamic interactions with coregulators. The interactions between nuclear receptors and RNA splicing factors link gene transcription initiation with pre-mRNA splicing, providing a coordinated control of the products of gene transcription. Here we report that two RNA splicing factors, PTB-associated splicing factor (PSF) and p54nrb, synergistically form protein complexes with the androgen receptor (AR) in a ligand-independent manner and inhibit its transcriptional activity. PSF does not affect AR protein stability, as in the case of the progesterone receptor, but impedes the interaction of AR with the androgen response element. Both splicing factors interact directly with mSin3A and attract mSin3A to the AR complex in a synergistic manner. The suppression of AR transcriptional activity by PSF and p54nrb is reversed by the inhibition of histone deacetylase activity. These data demonstrated that PSF and p54nrb complex with AR and play a key role in modulating AR-mediated gene transcription.

FIG. 1.

AR forms a protein complex with PSF and p54nrb. (A, B, and C) 293T cells were transiently transfected with vectors encoding pcDNA3 AR, His-PSF, and Flag-p54nrb, as indicated. WCLs were immunoprecipitated with antibodies against AR, His, or Flag tag. The associated proteins were immunoblotted to detect existence of AR, His-PSF, and Flag-p54nrb in the precipitate. (D) Endogenous AR/PSF/p54nrb complex was immunoprecipitated with LNCaP cell lysate with AR antibody in the presence of vehicle or 10 nM DHT. Associations of PSF and p54nrb were detected with PSF or p54nrb antibody. (E) GST or GST-PSF fusion proteins were immobilized on glutathione beads and incubated with WCL from LNCaP cells. Association of AR and p54nrb was detected by AR and p54nrb antibody. IP, immunoprecipitation; WB, Western blotting.

FIG. 2.

AR and PSF recruitment on PSA gene. (A) Schematic diagram of the PSA gene regulatory regions. Boxes depict AREs. The numbers represent the positions upstream from the PSA transcription start site. Primers were used to amplify these corresponding DNA fragments. (B) LNCaP cells were treated with vehicle or 10 nM DHT for 2 h. Cross-linked chromatin was immunoprecipitated with antibodies as indicated. The precipitated DNA was amplified using primers to amplify ARE I, III, and X described in panel A. (C) Re-ChIP assay was performed using LNCaP cells treated with vehicle or 10 nM DHT for 2 h. Chromatin extracts were immunoprecipitated with AR antibody followed by reimmunoprecipitation with AR, PSF, or rabbit IgG antibody. The precipitated DNA was amplified using primers described in panel A. (D) Chromatin extracts from LNCaP cells treated with 10 nM DHT for 0 to 90 min were immunoprecipitated with AR or PSF antibody, followed by PCR using primers described in panel A. IP, immunoprecipitation.

FIG. 3.

PSF inhibits transcriptional activity of AR. (A and B) 293T, PC-3, and LNCaP cells were transiently transfected with AR, PSF, or PSF-F expression vector and 3×ARE-Luc or MMTV-Luc reporter gene for 24 h. Cells were treated with vehicle or 10 nM DHT for additional 24 h. 293T cells were transiently transfected with AR and increasing doses as indicated (C) or a constant dose (0.5 μg) (D) of PSF expression vector together with MMTV-Luc and treated with vehicle or DHT for 24 h. (E) 293T cells were transfected with siPSF or scrPSF for 48 h. PSF protein expression levels were detected by immunoblotting with PSF antibody, and α-tubulin served as a loading control. (F) 293T cells were transiently transfected with siPSF or scrPSF together with an AR expression vector and MMTV-Luc for 48 h and then treated with vehicle or 10 nM DHT for an additional 24 h. Luc activities were normalized to β-Gal. The means ± standard deviations from two independent experiments performed in triplicate are shown.

FIG. 4.

AR's transcriptional activity is inhibited by p54nrb. (A and B) 293T and LNCaP cells were transiently transfected with AR, p54nrb, and MMTV-Luc vectors as indicated for 24 h and then treated with vehicle or 10 nM DHT for additional 24 h. (C) Schematic diagram describing vectors encoding Gal4 DBD fusion proteins with p54nrb or deletion mutants of PSF. 293T cells were transfected with pM, pM-PSF or its deletion mutants, pM-p54nrb, and pM-PRB (D) or with increasing doses of pM vectors without or with inserts of PSF, p54nrb, or PRB cDNA (E) together with G5-Luc reporter vector for 24 h. Luc activities were measured and normalized to β-Gal activity. The means ± standard deviations from two independent experiments performed in triplicate are shown. (F and G) Expression of pM and its fusion proteins was confirmed by immunoblotting with Gal4 DBD antibody.

FIG. 5.

Mechanisms of PSF inhibition of AR transactivation. (A and B) 293T cells were transfected with expression vectors as indicated. AR, PSF, and p54nrb were detected by Western blotting with specific antibodies, and α-tubulin served as a loading control. (C) Flag-tagged AR, PSF, or p54nrb protein was purified using M2-Sepharose beads with the lysate from 293 cells stably transfected with Flag-tagged expression vectors. Proteins were separated on SDS gels and stained with Coomassie blue. (D) Five micrograms of purified Flag-tagged proteins was subjected to Western blotting with M2 antibody to confirm their identities. (E) A gel shift assay was performed by incubating purified Flag-AR with γ-³²P-labeled double-strand ARE oligonucleotide in the presence of 10 nM DHT, and the AR/ARE complex was separated by nondenatured PAGE for 2 h. Nonlabeled ARE (lanes 5 and 6) and increasing doses of M2 antibody (lanes 9 and 10) were used to confirm the identity of the components in the migration-shifted band. Purified PSF and p54nrb were added to the reaction mixture (lanes 13 to 16). (F) LNCaP cells were transiently transfected with mock (CMV-Flag, Sigma), Flag-PSF, or Flag-p54nrb expression vectors. WCLs were immunoblotted with antibodies against AR, PSF, p54nrb, and Flag tag to confirm overexpression of PSF and p54nrb. (G) LNCaP cells transfected as indicated in panel F were treated with vehicle or 10 nM DHT for 2 h. Chromatin extracts were used to perform ChIP assays with AR antibody and rabbit IgG as a control, followed by PCR using primers described in the legend of Fig. 2A. (H) 293T cells were transfected with AR and/or PSF expression vector and MMTV-Luc for 24 h before the addition of 0 to 5 mM NaBu and 10 nM DHT for an additional 24 h. (I) 293T cells were transfected with pM or pM-PSF vector together with G5-Luc for 24 h before the addition of 0 to 5 mM NaBu for an additional 24 h. Luc activities were measured and normalized by β-Gal activity. The means ± standard deviations from two independent experiments performed in triplicate are shown.

FIG. 6.

Mechanisms of p54nrb inhibition of AR transactivation. (A) A GST pull-down assay was performed by incubating GST, GST-p54nrb, or GST-PSF with [³⁵S]methionine-labeled mSin3A (top). GST or GST fusion proteins used in the assay were separated on SDS gels and stained with Coomassie blue (bottom). (B) 293T cells were transfected with pM or pM-p54nrb vector together with G5-Luc and treated with 0 to 5 mM NaBu for 24 h. (C) 293T cells were transfected with AR, MMTV-Luc, and p54nrb or mock expression vector for 24 h before the addition of 10 nM DHT and 0 to 5 mM NaBu for an additional 24 h. (D) 293T cells were transfected with expression vectors as shown, and the WCLs were immunoprecipitated (IP) with AR antibody, followed by immunoblotting using antibodies against Flag-PSF, Flag-p54nrb, and endogenous mSin3A. (E) Cells were transfected with AR, PSF, and/or p54nrb expression as indicated together with MMTV-Luc. Cells were treated with 10 nM DHT for 24 h. (G) 293T cells were transfected with expression vectors as indicated and G5-Luc for 24 h. Luc activities were measured and normalized to β-Gal activity. The means ± standard deviations from two independent experiments performed in triplicate are shown. (H) LNCaP cells were maintained in phenol red-free medium with 10% charcoal-stripped serum for 48 h and then transfected with expression vectors as indicated. Cells were then treated with vehicle or 10 nM DHT for 4 days. PSA protein levels from the culture medium were measured by ELISA.

FIG. 7.

PSF and p54nrb enhance the recruitment of mSin3A and HDAC1 to the PSA gene. LNCaP cells transfected as indicated in the legend of Fig. 5F were treated with vehicle or 10 nM DHT for 2 h. Chromatin extracts were used to perform ChIP assays with mSin3A, HDCA1 antibody, or rabbit IgG as a control, followed by PCR using the primers described in the legend of Fig. 2A. CMV, cytomegalovirus.