"A Steroid Receptor Coactivator, SRA, Functions as an RNA and is Present in an SRC-1 Complex".
Summary:
Nuclear receptors play critical roles in the regulation of eukaryotic gene expression. We report the isolation and functional characterization of a novel transcriptional coactivator, termed steroid receptor RNA activator (SRA). SRA is selective for steroid hormone receptors and mediates transactivation via their amino-terminal activation function. We provide functional and mechanistic evidence that SRA acts as an RNA transcript; transfected SRA, unlike other steroid receptor coregulators, functions in the presence of cycloheximide, and SRA mutants containing multiple translational stop signals retain their ability to activate steroid receptor-dependent gene expression. Biochemical fractionation shows that SRA exists in distinct ribonucleoprotein complexes, one of which contains the nuclear receptor coactivator steroid receptor coactivator 1. We suggest that SRA may act to confer functional specificity upon multiprotein complexes recruited by liganded receptors during transcriptional activation.
In this work we describe the isolation and functional characterization of a novel transcriptional coactivator termed SRA. SRA is different from other known coregulators in that it functions as an endogenous RNA transcript. We have defined several different features of this RNA: SRA is (1) a bona fide transcriptional coactivator, (2) selective for the AF1 of steroid receptors, (3) expressed as multiple isoforms in a cell-specific manner, and (4) present in a steady-state coregulator complex with the AF2 coactivator SRC-1.
We have described the isolation of three SRA isoforms deduced from sequencing of different cDNAs and genomic clones from different species. When overexpressed in mammalian cells, recombinant SRA, regardless of isoform or origin, enhanced steroid receptor-mediated transactivation without significantly enhancing the level of basal transcription of minimal or natural promoters. In assays of endogenous PR-mediated transcactivation, a typical enhancement of receptor gene activity of ~10-fold was achieved by coexpression of SRA. Antisense deoxyoligonucleotides added to cells reduced steroid receptor-induced transcription by up to 70%. In addition, we have shown that SRA reverses steroid receptor squelching in a dose-dependent manner. Hence, SRA exhibits many characteristics expected of a bona fide coactivator.
Despite certain functional similarities, SRA differs in some important aspects from many other coactivators in that its coactivation is selective and that it is an RNA. We have presented several independent lines of evidence that indicate that SRA selectively enhances steroid receptor-mediated transactivation but does not influence transactivation by type II nuclear receptors or by other transcription factors. In addition, using in situ hybridization analysis we have obtained evidence for both a selective expression pattern of SRA and a general colocalization in brain tissue with members of the steroid receptor family (data not shown). In our coimmuno-precipitation assays it appears that, unless SRC-1 is coexpressed, the N-terminal domain of steroid receptor is required for binding of SRA. In our reporter gene assay, SRA per se fails to enhance transcriptional potency of the AF2 receptor domain in cultured cells. The exact nature of the interaction of SRA with the AF1 domain of steroid receptors is as yet unclear. SRA was originally isolated in the yeast two-hybrid system, an assay designed to identify protein-protein interactions. In a reconstructed yeast system, SRA associated with the N-terminal domain of PR but not with a control hybrid (not shown). The lack of sequence homology within the amino terminus of steroid receptors suggests that SRA may interact indirectly with the AF1 of the receptors as part of a ribonucleoprotein complex. It is unlikely that protein-protein interactions between the bait construct and the GAL activation domain played any role in the isolation of SRA. Rather, we envision an interaction of SRA with the PR N-terminal bait, thereby recruiting it to the reporter gene site. The SRA-PR N terminus interaction is likely to have been supported by yeast proteins, possibly through a mediator with functional similarity to SRC-1, and such interactions would have favored reporter gene activity and resulted in a positive hit. This somewhat fortuitous isolation of SRA appears less puzzling when it is considered that yeast proteins contribute functionally to transcriptional activation by steroid receptors (Yoshinaga et al, 1992), even though steroid receptors are not expressed in yeast.
Although we do not totally exclude the existence of a translation product of SRA contained in certain cells at specific developmental stages, we have provided evidence to indicate that SRA exists and functions as an RNA transcript. First, we were not succesful in our attempts to translate the SRA clones in vitro or in vivo. Second, an affinity column containing a mAb raised against a sequence at the carboxy-terminal end of the putative ORF1 transcript failed to detect endogenous SRA in various cell lines tested. In addition, extensive mutagenesis of SRA, introducing multiple translational stop codons in all reading frames, did not affect the ability of these mutants to enhance PR transactivation. A final functional test was provided by transfection experiments in the presence of cycloheximide, in which SRA retained its ability to coactivate a reporter gene, while other protein coregulators such as SRC-1 and CBP did not.
The ability of RNA molecules to perform many functions that were
commonly attributed to proteins has been well documented. RNA molecules
perform enzymatic reactions such as trans-esterification (Jaeger,
1997) or catalysis of peptide bond formation (Zhang and Cech, 1997)
and can regulate gene expression in trans by structure (Jones and
Peterlin, 1994), by antisense RNA-RNA interaction (Lee, et al, 1993;
Crespi
et al, 1994), or by association of two genomic-sense RNAs (Sit et al,
1998). To our knowledge, however, SRA is different from eukaryotic transcriptional
coactivators in its ability to function as an RNA transcript to selectively
regulate the activity of a family of transcriptional activators.
1. Frenster JH, "Nuclear RNA Species Activate DNA Transcription Within Chromatin", FASEB Journal, Vol. 13, No. 7, A1506 (April 23, 1999).
2. "Nuclear Polyanions as De-Repressors of Synthesis of Ribonucleic Acid".
3. "Mechanisms of Repression and De-Repression w ithin Interphase Chromatin".
4. "Mated Models of Gene Regulation in Eukaryotes".
5. "Selective Gene De-Repression by De-Repressor RNA".
6. "Oncogenes as Molecular Targets within Active Chromatin".