The transcription corepressor CtBP is often recruited to the target promoter via interaction with a conserved PxDLS motif in the interacting repressor. In this study, we demonstrate that CtBP1 was SUMOylated and that its SUMOylation profoundly affected its subcellular localization. SUMOylation occurred at a single Lys residue, Lys428, of CtBP1. CtBP2, a close homolog of CtBP1, lacked the SUMOylation site and was not modified by SUMO-1. Mutation of Lys428 into Arg (K428R) shifted CtBP1 from the nucleus to the cytoplasm, while it had little effect on its interaction with the PxDLS motif. Consistent with a change in localization, the K428R mutation abolished the ability of CtBP1 to repress the E-cadherin promoter activity. Notably, SUMOylation of CtBP1 was inhibited by the PDZ domain of nNOS, correlating with the known inhibitory effect of nNOS on the nuclear accumulation of CtBP1. This study identifies SUMOylation as a regulatory mechanism underlying CtBP1-dependent transcriptional repression.

Transcriptional repression mediated through histone deacetylation is a critical component of eukaryotic gene regulation. Here we demonstrate that the class II histone deacetylase HDAC4 is covalently modified by the ubiquitin-related SUMO-1 modifier. A sumoylation-deficient point mutant (HDAC4-K559R) shows a slightly impaired ability to repress transcription as well as reduced histone deacetylase activity. The ability of HDAC4 to self-aggregate is a prerequisite for proper sumoylation in vivo. Calcium/calmodulin-dependent protein kinase (CaMK) signalling, which induces nuclear export, abrogates SUMO-1 modification of HDAC4. Moreover, the modification depends on the presence of an intact nuclear localization signal and is catalysed by the nuclear pore complex (NPC) RanBP2 protein, a factor newly identified as a SUMO E3 ligase. These findings suggest that sumoylation of HDAC4 takes place at the NPC and is coupled to its nuclear import. Finally, modification experiments indicate that the MEF2-interacting transcription repressor (MITR) as well as HDAC1 and -6 are similarly SUMO modified, indicating that sumoylation may be an important regulatory mechanism for the control of transcriptional repression mediated by both class I and II HDACs.

Posttranslational modification with SUMO1 regulates protein/protein interactions, localization, and stability. SUMOylation requires the E1 enzyme Aos1/Uba2 and the E2 enzyme Ubc9. A family of E3-like factors, PIAS proteins, was discovered recently. Here we show that the nucleoporin RanBP2/Nup358 also has SUMO1 E3-like activity. RanBP2 directly interacts with the E2 enzyme Ubc9 and strongly enhances SUMO1-transfer from Ubc9 to the SUMO1 target Sp100. The E3-like activity is contained within a 33 kDa domain of RanBP2 that lacks RING finger motifs and does not resemble PIAS family proteins. Our findings place SUMOylation at the cytoplasmic filaments of the NPC and suggest that, at least for some substrates, modification and nuclear import are linked events.

Androgen receptor is a ligand binding transcription factor that controls several physiological processes ranging from the development of the male sexual organs to the acquisition of secondary sex characteristics. It is composed of a carboxy-terminal ligand binding domain, a centrally located DNA binding domain and an amino terminal modulator region. Detailed study on the DNA and carboxy-terminal regions have been carried out, but only limited information is available on the activity of the N-terminus. With the use of truncated and chimeric receptor constructs we have demonstrated in transient transfection experiments that the N-terminus of the androgen receptor contributes to DNA binding, transactivation and transrepression functions of the receptor. We have shown that specific sequences at the N-terminus are needed for transactivation but we were unable to identify discrete sequences in this region for the DNA binding and transrepression functions. Sequences from the transcription factor NFI/X3 that bear no homology to the N-terminus of the androgen receptor nevertheless functionally replaced it in enhancing DNA binding, transrepression but not transactivation functions of the receptor. Thus, it appears that the structure rather than sequence specific elements determines the contribution of the N-terminus of the androgen receptor to DNA binding and transrepression functions.

Expansion of CAG trinucleotide repeats in androgen receptor gene is present in patients with a rare X-linked inherited form of motor neuron disorder termed Kennedy's disease or spinal and bulbar muscular atrophy (SBMA). This is a late onset progressive disease often associated with mild signs of androgen insensitivity. Defects in androgen receptor (AR) action have been linked to the expansion of the CAG trinucleotide repeats and postulated to be the cause of the disease. We have identified a trinucleotide repeat of 45 in the N-terminus of the AR in two brothers with SBMA and several members in their family (range in the general population is 11-35). Treatment of the patients with androgens failed to improve their clinical symptoms and provided no hint of an anomalous function of the AR. Consistently, functional analysis of the mutant receptor showed hormone binding, transactivation and transrepression potentials identical to that of the wild-type receptor. These results together argue against SBMA being a loss of function mutation of the AR.

The androgen receptor (AR) plays a central regulatory role in prostatic carcinoma and is a target of androgen ablation therapy. Recent detection of mutant receptors in tumor specimens suggest a contribution of AR alterations to progression towards androgen independence. In a specimen derived from metastatic prostate cancer we have reported a point mutation in the AR gene that leads to a single amino acid exchange in the ligand binding domain of the receptor. Another amino acid exchange resulting from a point mutation was also identified 15 amino acids away from our mutation. This mutation was detected in the AR gene isolated from an organ-confined prostatic tumor. Here we report the functional characterization of the two mutant receptors in the presence of adrenal androgens and testosterone metabolites, These studies were performed by cotransfecting androgen-responsive reporter genes and either the wild-type or mutant AR expression vectors into receptor negative DU-145 and CV-1 cells. The indicator genes used consisted of the promoter of the androgen-inducible prostate-specific antigen gene or the C' Delta 9 enhancer fragment from the promoter of the mouse sex-limited protein driving the expression of the bacterial chloramphenicol acetyl transferase gene. Cotransfection-transactivation assays revealed that the adrenal androgen androstenedione and two products of testosterone metabolism, androsterone and androstandiol, induced reporter gene activity more efficiently in the presence of the mutant receptors than in the presence of the wild-type receptor. No difference between wild-type and mutant receptors was observed in the presence of the metabolite androstandione. The interaction of receptor-hormone complexes with target DNA was studied in vitro by electrophoretic mobility shift assays (EMSA). Dihydrotestosterone and the synthetic androgen mibolerone induced a faster migrating complex with all receptors, whereas the androgen metabolite androstandione induced this complex only with the two mutant receptors. Androsterone and androstandiol were inactive in the EMSA. These aberrant properties of the mutant receptors in the presence of adrenal androgens and products of androgen metabolism may be of importance in the course of the prostate cancer, especially during androgen ablation therapy.

Steroid receptors activate and repress genes. An important class of genes that they repress is controlled by the transcription factor AP‐1. The activity of AP‐1 is inhibited by the receptor, a mechanism exploited for the therapy of various forms of pathological hyperproliferation in humans. We show here by point mutations in the DNA binding domain and by the choice of steroid ligands that repression of AP‐1 activity and transactivation functions of the glucocorticoid receptor (GR) are separable entities. While DNA binding and activation of glucocorticoid‐regulated promoters require GR dimerization, we present data that suggest that repression is a function of GR monomers.