Role of Steroid Hormone Coregulators in Health and Disease

Steroid hormone action is mediated mainly through the intracellular nuclear receptors. These receptors are ligand-inducible transcription factors belonging to the nuclear receptor superfamily and mediate diverse actions of sex steroids. They share three common functional domains: (i) amino-terminal transactivation domain containing activation function (AF)-1, (ii) central or DNA-binding domain (DBD) containing two zinc fingers like motifs, and (iii) carboxy-terminal ligand-binding domain containing AF-2 [1]. Following the binding of hormone to its specific receptor, the binary complex interacts with the responsive elements of the target genes and recruits an array of coregulators and chromatin modifiers. The involvement of coregulators makes the regulation more complex (fig. 1). These coregulators areclassified as coactivators when they activate the expression of genesand as corepressors when they inhibit the transcription of genes [2]. Previously, it was thought that coactivators and corepressors are present in separate complexes, but recent evidence indicates that these molecules exist in the same complex [3]. This complex has an opposite function and allows for efficient transcriptional control mechanisms depending on posttranslational modification of coregulators. In the majority of cases, transcription is activated by histone acetylation and repressed by histone deacetylation.

This review discusses the current understanding of coregulators and their mechanism of action, defects in coregulators, resulting diseases and therapeutic approaches. The information included in this review has been collected from the literature available in NCBI, ISI Web of Knowledge, ScienceDirect search engine, HighWire press and Karger press journals. Understanding the role of coregulators in steroid hormone action is important for developing a therapeutic strategy for the treatment of steroid-related diseases.

A large number of coactivators have been identified and characterized over the past few years [4] – several are listed in table 1. They are characterized by their interaction with the transcription machinery and activation function domains of nuclear receptors [5]. Based on the mechanism of function, coactivators can be categorized into three distinct multiprotein complexes: (i) SWI/SNF complex, which is associated with ATP-dependent alteration in chromatin structure; (ii) p160/CBP complex, which is primarily related to histone acetylation, and (iii) mediator complex, which is involved in the activation of RNA polymerase II and initiation of transcription.

Members of the SWI/SNF family of coactivators are the first described positive regulators of hormone action. They were initially identified in the yeast on the basis of a genetic screening required for regulation of mating type switching [6]. They mediate chromatin remodeling of swi/snf genes and support glucocorticoid receptor (GR)-dependent transactivation. However, mutations in these genes fail to transactivate a reporter gene in the presence of transfected GR. It has also been shown that GR co-immunoprecipitates with the SWI/SNF complex [7, 8]. Ligand-dependent transcriptional activation of GR or estrogen receptor (ER) requires several SWI gene products such as SWI1, SWI2 and SWI3, which are members of the SWI/SNF complex. The human homologue of SWI2 is termed as SNF2α and SNF2β or Brahma [9, 10]. Other members of the SW1/SNF family include the yeast general control non-repressed protein (GCN) 5, p300 and CBP [11, 12]. They possess histone acetyltransferase (HAT) activity which is involved in chromatin remodeling.

SRC-1, first discovered in human as a ligand-dependent protein of progesterone receptor (PR), belongs to the p160/CBP complex [13]. It interacts with helix 12 within AF-1 and AF-2 regions of ER [14]. The binding of SRC-1 is more efficient when it makes a bridge between AF-1 and AF-2 and interacts directly to the basal transcription machinery through TFII or TBP [15].

DRIP, purified previously as a member of the p160/CBP coactivator complex, has no HAT activity [16]. VDR interacts directly with a human multi-protein WINAC through the William syndrome transcription factor which contains both SWI/SNF components and DNA replication-related factors and exhibits ATP-dependent chromatin remodeling activity [17]. Another member of the p160/CBP complex, p300/CBP, interacts with roughly 100 proteins [18]. RAC3, also known as SRC-3/ACTR/AIB1/pCIP/TRAM-1, acts through a bipartite nuclear localization signal and importin α3 protein [19]. AIB (amplified in breast cancer) 1, a member of the p160 family, is highly expressed in ER- and PR-responsive endometrial carcinoma [20].

Higher order chromatin structure is regulated in part by covalent modifications of N-terminal histone tails, which in turn serve as platforms for the recruitment of signaling molecules that include non-histone proteins such as heterochromatin protein (HP1), nucleosomal remodeling and histone deacetylation complex [21]. Translocated in liposarcoma (TLS), a 65-kDa protein has been identified as TFII/RNA polymerase II-associated protein because of its sequence similarity to hTAF68 [22]. Its high-affinity binding enhances the expression of thyroid receptor (TR) proteins. Furthermore, sumoylation is responsible for chromatin modification of AR interacting protein 4 (ARIP4), which belongs to the SNF2-like family and acts as a chromatin modifier in AR regulation. However, the ability of ARIP4 to activate AR function is abolished with mutation in six potential SUMO (small ubiquitin-related modifier) attachment sites [23]. Other chromatin modifications may take place by sumoylation or binding of cofactors. For example, the chromatin remodeling factor Mi-2α interacts with human c-Myb and activates the transcription of hematopoietic genes [24].

As compared to coactivators, relatively less is known about corepressors (table 2). A true corepressor contains autonomous repression domain and directly interacts with the receptor. It also interacts with components of the transcription machinery and represses the transcription of a specific gene. The repression of transcription may be due to competition among molecules which bind to DNA or may be due to direct silencing of the basal transcription machinery [25]. Thus, a particular gene may be repressed through the recruitment of corepressor molecules to the gene and protein-protein interaction. TR-mediated repression is reversed by cotransfection with either the unliganded retinoic acid receptor (RAR) or C-terminus of the oncoprotein (v-ErbA) [26].

The corepressors are associated with nuclear hormone receptors and recruit histone deacetylase (HDAC)3 and other isoforms that mediate histone deacetylation [27]. Another interacting protein, ZNF366, acts directly or through ERα corepressors, namely RIP140 and CtBP, and inhibits the expression of estrogen-responsive genes by HDAC activity [28]. These findings strongly support the hypothesis that gene transcription is repressed through nucleosomal condensation caused by deacetylation of histones [29].

Both corepressors NCoR and SMRT exist in large protein complexes with estimated sizes of 1.5–2 MDa in HeLa nuclear extracts. Using conventional and immunoaffinity chromatography, HDAC3, transducin (β)-like 1 and a WD-40 repeat-containing protein have been isolated as subunits of the purified SMRT complex. The HDAC3-containing SMRT and NCoR complexes can bind to unliganded thyroid hormone receptors (TRs) in vitro. The injection of antibodies against HDAC3 or SMRT/NCoR into Xenopus oocytes leads to partial relief of repression by unliganded TR/RXR, suggesting both NCoR and SMRT as partner of HDAC-containing complexes and their function in transcriptional repression [30].

Steroid receptors in combination with their coactivators and corepressors mediate precise gene regulation ranging from the ligand-induced transcriptional activation to repression. As studies continue to unfold the role of coregulators in transcriptional regulation by nuclear steroid receptors, it is becoming evident that they are important for understanding the etiology of a variety of diseases such as obesity, diabetes, atherosclerosis, osteoporosis, cancer, neurodegenerative disorders, etc. The coregulators may also be used as therapeutic targets in steroid-related diseases. For example, breast cancer patients show resistance to anticancer drugs, and the majority of them eventually do not respond to tamoxifen therapy [31].

Mutations in CBP coactivator are involved in embryogenesis and chromosomal translocation of p300 gene that causes Rubinstein-Taybi syndrome. The loss of one functional copy of the CBP gene leads to the developmental abnormalities in Rubinstein-Taybi syndrome and possibly the propensity for malignancy [32]. It also causes open neural tube and heart defects in humans, as well as embryonic lethality in mice [33, 34]. The activity of CBP/p300 is also inhibited in neurodegenerative diseases [35].

Coactivators like Ciz1 and FHL2 (four and half LIM domain) interact with ERα and are involved in ovarian and breast cancers [36, 37]. The peroxisome proliferator-activated receptor (PPAR)γ coactivator-1 of the PGC-1 gene familyregulates lipid and carbohydrate metabolism and inflammation [38]. Knockout studies reveal the effects of two members of the p160 family on energy homeostasis. TIF2–/– mice are protected against obesity and display enhanced adaptive thermogenesis, whereas SRC1–/– mice are prone to obesity due to reduced energy expenditure [39]. In addition, TRAP220 knockout reveals an early gestational stage with heart failure and exhibits impaired neuronal development with extensive apoptosis [40].

Mutations in receptor gene also inhibit the binding of coactivators and lead to diseases. For example, androgen-insensitive syndrome is developed due to mutation in the DNA-binding domain of AR which prevents interaction of coactivators to the AF-1 region and transmission of activation signal [41]. This syndrome shows various degrees of impairment in the genital virilization [42, 43]. The transcriptional response to AR coactivator (ART-27) is increased by the majority of naturally occurring AR mutations in prostate cancer, but decreased by P340L mutation and androgen-insensitive syndrome [44].

Mutation in the corepressor or change in its interaction with the basal transcription machinery may lead to diseases. For example, mutations in MeCP-2, RB and DAX-1 result in the loss of corepressor interaction leading to embryonic diseases during development [45, 46]. Mutation in MeCP-2, which affects either the methyl-binding domain or the transrepressor domain by deacetylation, causes an X-linked neurodegenerative developmental disorder, Rett syndrome [47]. These mutations result in the loss of recruitment of the SIN3 corepressor to target genes [48]. Mutations in MTA (metastasis-associated protein) 2 are responsible for estrogen-responsive cancer [49]. An ERβ-associated corepressor NM23-H2 has been identified in the arteries from young subjects (27 ± 6 years) with benign intimal hyperplasia. The expression of NM23-H2 decreases with the advancement of atherosclerosis and inflammation in human [50].

Mutations in receptors also affect their interaction with corepressors and lead to a number of diseases. For example, acute myelogenic leukemia (AML), which results from translocations involving fusion of DBD of AML protein to the transcriptional repressor ETO/MTG8, is associated with aberrant interaction of corepressor with transcription factors [51]. Similarly, B-cell acute lymphoblastic leukemia involves the chromosomal translocation, which results in the fusion of repression domain of the Ets leukemia protein TEL to the large form of AML. In both cases, corepression arises from the recruitment of NCoR and SIN3-HDAC complex by ETO and TEL [52].

The synthetic compounds, which target HAT/HDAC complex or the receptor itself, are used in the therapy of diseases caused by defects in the binding of coregulators.

The reduction in HDAC activity and HDAC2 expression causes chronic obstructive pulmonary disease which may account for the amplified inflammation and resistance to the actions of corticosteroids. The reduction in HDAC2 expression may be secondary to the oxidative and nitrative stress as a result of cigarette smoking and severe inflammation. It occurs to different degrees in severe asthma, smoking asthmatic patients and cystic fibrosis. Similar mechanisms may also account for the steroid resistance seen in latent adenovirus infections. The reduction in HDAC activity induced by oxidative stress can be restored by theophylline, which acts through specific kinases and reverses steroid resistance in chronic obstructive pulmonary disease and other inflammatory lung diseases [53]. Several histone deacetylase inhibitors have been shown to inhibit tumor growth and now used in clinical trials for the treatment of cancer [54, 55]. The inhibition of HDAC1 by phenylbutyrate also reverses the ETO-mediated transcriptional repression and induces partial differentiation of AML-ETO cell line [56].

LXRs (LXRα and LXRβ) have a sterol-responsive property and can bind directly with several oxysterol metabolites. They are involved in cholesterol transport, glucose metabolism and inflammation. Thus, synthetic LXR ligands have been designed to treat disorders such as atherosclerosis and diabetes [57]. Further, statins modulate the sterol-responsive element-binding protein, which targets genes that are involved in cholesterol and fatty acid metabolism, namely hydroxyl-methyl-glutaryl acetyl Coenzyme-A (HMG-CoA) reductase, HMG-CoA synthase and the low-density lipoprotein receptor [58]. Similarly, insulin-like drugs target the PPARγ, whereas several anti-inflammatory drugs inhibit activation of nuclear factor-ĸB [59].

The expanding number of coregulators has increased the challenge and added the complexity to understanding the transcriptional regulation of steroid hormone action. One of the future challenges is to determine the specificity of coregulators. For example, the central nervous system is regulated by an array of NR coactivators such as p160/SRC family members, CBP/p300, BRG1, TRAP220 and PGC-1α as well as corepressors like NCoR and SMRT. In recent years, the studyof corepressor biology has become an area of active research.This has led to substantial progress in the identification ofa number of novel corepressors with critical roles in determiningthe action of NR. Emerging data re-emphasize that the regulation of NRby corepressor is as significant as that by coactivator. Although much more workremains to be done, we have begun to gain a deeper insightinto the transcriptional regulation of steroid receptors, therole played by chromatin modifications, and the contributionof coactivators and corepressors in the development of pathological conditions. Further study of key coregulators in physiologically relevantanimal models and human subjects combined withcell culture models will help to elucidate the detailed mechanism of steroid hormone action and develop the therapeutic strategies for the hormone-related diseases.

We acknowledge the financial support from the Indian Council of Medical Research and Department of Biotechnology, Government of India.

ACTR = Activator of the thyroid and retinoic acid receptor; ARA = androgen receptor activator; ARIP = androgen receptor-interacting protein; BRG = Brahma-related gene; CBP = CREB-binding protein; CiZ1 = Cip-interacting zinc finger protein 1; CREB = cAMP-response element-binding protein; DAX-1 = DSS-AHC critical region on the chromosome gene 1; DRIP = vitamin D receptor-interacting protein; ERAP-140 = estrogen receptor (ER)-associated protein 140; ETO = eight-twenty-one translocation; GRIP = glucocorticoid receptor interacting protein; MeCP-2 = methyl-CpG-binding protein; NCoR = nuclear receptor corepressor; pCAF = p300/CBP-associated factor; pCIP = p300/CBP cointegrator-associated protein; PGC = PPARγ coactivator; RAC = receptor-associated coactivator; RIP = receptor-interacting protein; SMRT = silencing mediator of retinoic acid and thyroid hormone receptor; SRC = steroid receptor coactivator; TAF = TBP-associated factor; TBP = TATA-binding protein; TIF = transcriptional intermediary factor.