| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of Pharmacology and Medicine, New York University School of Medicine, New York, New York 10016
Correspondence: Address all correspondence and requests for reprints to: Herbert H. Samuels, M.D., or Muktar Mahajan, Ph.D., Departments of Pharmacology and Medicine, New York University School of Medicine, 550 First Avenue, New York, New York 10016. E-mail: herbert.samuels{at}med.nyu.edu, or muktar.mahajan{at}med.nyu.edu
 |
Abstract |
|---|
 Top Abstract I. IntroductionII. Cloning, Domain Structure,...III. NRC Interacts with...IV. NRC Is an...V. Conclusions and Future...References |
|---|
|
I. Introduction |
|---|
|
TopAbstractI. IntroductionII. Cloning, Domain Structure,...III. NRC Interacts with...IV. NRC Is an...V. Conclusions and Future...References |
|---|
Type I NRs exist as apoproteins either in the cytoplasm (e.g., GR) or the nucleus (e.g., ER
) bound to heat shock proteins, whereas many type II receptors are thought to bind to their target DNA sequences in the absence of their cognate ligands (e.g., TRs and RARs). The transformation of an inactive aporeceptor to a transcriptionally active form is a complex cellular process that involves: 1) binding with ligands such as endocrine hormones, fatty acids, cholesterol derivatives, and products of lipid metabolism; and 2) change in the conformation of the NR to an active form that allows for 3) interaction with other coregulator proteins that potentiate the ability of the NR to activate or repress transcription of target genes. To date, ligands for a large number of NRs known as orphan receptors remain unknown. However, the mechanisms leading to transcription activation by these receptors is likely to be regulated similar to NRs with known ligands. Extensive studies from a large number of laboratories have contributed to our present understanding of the transformation of an aporeceptor to an active form, the role of ligand and modular structure of NRs and the role of DNA binding, dimerization, nuclear localization, and activation domains of these receptors.
NRs share a common modular structure (Fig. 1
) comprising an N-terminal "A/B" domain, a central zinc finger-containing DNA-binding "C" domain [DNA-binding domain (DBD)], a "hinge region" referred to as the "D" region, and a C-terminal "EF" domain. The A/B domain is the most variable region of the NRs and, for some receptors, harbors an activation function referred to as activation function 1 (AF1). The A/B domain has been shown to mediate ligand-independent transcriptional enhancement when expressed as a fusion with a heterologous DBD (e.g., yeast Gal4). The C-terminal EF region (and for some receptors DEF) binds ligand and comprises the ligand-binding domain (LBD). The LBD harbors a ligand-dependent activation function referred to as AF2 (1, 2). The crystallographic structures of various NRs, without and with their cognate ligands or antagonists, have been solved and have provided important structural insights into the mechanism of transcriptional activation, repression, and coregulator interaction (3, 4, 5, 6, 7). In summary, the NR LBD is organized into 12 -helical regions that reorient upon ligand binding. Helix 12 reorients to a position on the LBD and contributes to the formation of a hydrophobic groove, which allows for the docking of cofactors to the LBD surface, which, in turn, leads to transcriptional activation by the receptor. Although helix 12 was initially considered to be an activation domain and has been referred to as AF2 in the literature, its role is to contribute to a conformational change in the receptor leading to AF2 activity (8).
|
Beyond the p160 family members, the search for novel coactivators for NRs used both two-hybrid screening and biochemical purification of protein complexes associated with ligand-bound NRs. Two similar, if not identical, multiprotein coactivator complexes referred to as DRIPs (vitamin D receptor-interacting proteins) (18) and TRAPs (thyroid hormone receptor-associated proteins) (19) were isolated from HeLa cells using in vivo (TRAPs) and in vitro (DRIPs) biochemical approaches. DRIP205/TRAP220 is the protein component of the DRIP/TRAP complex that directly associates with NRs. PPAR-binding protein (PBP) is identical to DRIP205/TRAP220 and had been identified earlier in a yeast two-hybrid screen as a putative coactivator for PPAR (20). The DRIP/TRAP complex is one of several related complexes that share components and are thought to modulate the activities of a variety of transcription factors including the NRs, simian virus 40 virus promoter-specific transcription factor (Sp1), nuclear factor-
B (NF-B) (p65), herpes simplex virus protein 16, adenovirus early expressed protein 1 (E1A), and p53 (18, 21, 22, 23). In addition to the p160 family of coactivators and the DRIP/TRAP complex, cAMP response element-binding protein (CREB)-binding protein (CBP)/p300 plays an important role in the function of NRs and other transcription factors (10, 24, 25). Although the N-terminal region of CBP/p300 has been shown to interact with ligand-bound NRs in vitro, its interaction with NRs in vivo has not been established (26). CBP/p300 contains an intrinsic histone acetyl transferase activity, and numerous studies suggest that CBP/p300 appears to act as a transcriptional integrator that is recruited to the NRs and other transcription factors through their association with other coactivators (e.g., the SRCs and NRC) (13, 14, 27, 28, 29).
The discovery of RIP140 (30), a ligand-dependent interacting protein for ER, and subsequent studies with RIP140 led to the identification of a signature LxxLL interaction motif found in virtually all coactivators (26, 31). Studies from a number of laboratories documented that NRs associate with the LxxLLs of coactivator proteins (28, 31, 32, 33) via a ligand-dependent conformational change in the receptor LBD that repositions helix 12 as described above. This conformational change leads to the formation of a hydrophobic groove on the NR surface that functions as a docking site for the coactivator LxxLL motif interaction (32, 34, 35).
Coactivators are thought to act by modifying local chromatin structure and/or through direct interaction with components of the transcription factor IID basal transcription apparatus (36, 37). Thus, liganded NRs bind members of the p160 family, that recruits CBP/p300 (and CBP-bound p/CAF) to a target gene promoter. This recruitment locally modifies chromatin structure through the CBP/p300 and p/CAF histone acetyltransferase (HAT) activities. In addition, members of the p160 family associate with CARM1 (coactivator-associated arginine methyltransferase 1) and PRMT1 (protein arginine methyltransferase 1), which potentiate receptor activation. The effect of CARM1 and PRMT1 on receptor activity is thought to result from chromatin changes through histone methylation (38, 39). Thus, PRMT1 has been shown to methylate Arg-3 of histone H4, which enhances the extent of acetylation of H4 tails by CBP/p300 (38). The role of the DRIP/TRAP complex is less well defined although recent studies indicate that the TRAP complex interacts with the TAFII components of the TFIID complex and affects both basal activity and stimulation of transcription by activators (36). Chromatin immunoprecipitation assays (ChiPs) indicate that p160 coactivators and CBP/p300 are recruited to NR target genes rapidly (30 min to 1 h) after ligand binding (40). DRIP/TRAP is subsequently recruited to the receptor-bound promoter, and ChiP assays show periodic cycling of p160 coactivators and DRIP205/TRAP220. These findings suggest that p160 coactivators and CBP/p300 modify chromatin and allow for the subsequent recruitment of the DRIP/TRAP complex, which may target the RNA polymerase II apparatus (40).
In addition to the p160/SRC family and the DRIP/TRAP complex, nearly 50 other putative coactivators have been identified thus far (http://nursa.org). Coactivators that have been extensively studied include PPAR coactivator-1 (PGC-1), nuclear receptor interacting factor 3 (NRIF3), and NRC. PGC-1 was isolated as a coactivator for PPAR
involved in thermogenesis in brown adipose tissue (41). NRIF3 acts as a receptor-selective coactivator for the TRs and the RXRs, and interaction with these NRs occurs though a novel LxxIL motif (42, 43). NRC (nuclear receptor coregulator) (28), also referred to as activating signal cointegrator-2 (ASC-2) (44), 250-kDa receptor associated protein (RAP250) (45), thyroid hormone receptor-binding protein (TRBP) (70), and PPAR-interacting protein (PRIP) (46), was cloned and characterized as a coactivator by a number of laboratories. AIB3 is an N-terminal variant of NRC (NCBI accession no. 208277), which was first identified as one of the genes on chromosome 20 that was overexpressed in BT-474 breast cancer cells (47). The National Center for Biotechnology Information (NCBI) has annotated human NRC and the other identical factors as NCOA6 (nuclear receptor coactivator 6) (NCBI accession no. NM_014071). NRC has emerged as an important coactivator not only for NRs but also for a number of other well-known transcription factors such as c-Fos (cellular homolog of the vfos oncogene), c-Jun (cellular homolog of an oncogene in the avian sarcoma virus 17), CREB, NF-B, activating transcription factor-2 (ATF-2), heat shock factors, E2F-1, serum response factor (SRF), and retinoblastoma gene product (Rb) (28, 44, 48, 49, 50, 51, 52, 53, 54). The biological importance of NRC as an essential, nonredundant, coactivator is reflected by recent biochemical and genetic studies involving NRC knockout mice (55, 56, 57, 58). The objective of this review is to summarize studies on NRC from a number of laboratories and to highlight the recent advances in our understanding of how NRC functions as a transcriptional coregulator.
|
II. Cloning, Domain Structure, Functional Organization, and Expression of the Nuclear Hormone Receptor Coregulator/Coactivator (NRC) |
|---|
|
TopAbstractI. IntroductionII. Cloning, Domain Structure,...III. NRC Interacts with...IV. NRC Is an...V. Conclusions and Future...References |
|---|
as a bait (45), and PRIP from a mouse liver cDNA library using PPAR as a bait (46), whereas TRBP was cloned from a rat pituitary GC cell library using TR as a bait (49). Our laboratory cloned both rat and human NRC cDNAs (28). Rat NRC was cloned from a rat pituitary GH4C1 cell cDNA library. GH4C1 and related somatotroph cell lines have been extensively used to study transcriptional activation of the GH and prolactin genes by NRs. These cells express a wide variety of NRs at levels comparable to those found in various tissues in vivo (
5,000–10,000 receptors per cell). Because these cells are highly responsive to NR ligands, even at low levels of receptors, we considered that they may contain novel coregulators not previously identified in less differentiated cells (e.g., HeLa). Thus, we constructed a yeast two-hybrid cDNA library from GH4C1 cells in the pJG4–5, which allows for conditional expression of cDNAs in yeast that are expressed as fusions with the B42 activation domain. Using LexA-TR as bait, we identified a T3-dependent interactor of approximately 300 amino acids rich in Glu and Pro containing a single LxxLL motif that we referred to as NRC (28). The human expressed sequence tag database contained an approximately 6-kb cDNA (KIAA0181) that appeared to be an ortholog of rat NRC. Based on the genomic sequence, this human NRC cDNA clone appeared to lack the first N- terminal 62 amino acids including the initiator ATG. A full-length human NRC cDNA was generated by 5'-rapid amplification of cDNA ends. The human NRC/ASC-2/RAP250/TRBP clone (28, 44, 45, 49) and the mouse PRIP clone (46) were characterized in detail by our laboratory and others. The following structural and functional evidence support the notion that it functions as a coactivator for NRs: 1) human NRC contains two potential LxxLL NR-interaction motifs (NR boxes); 2) NRC interacts with ligand-bound NRs and potentiates the activity of a variety of NRs in mammalian cells; 3) a short region derived from rat and human NRC containing the NR box acted as a dominant-negative inhibitor of the activity of endogenous NRs; and 4) human full-length NRC displayed some sequence similarity with CBP/p300.
B. Domain structure and functional organization of NRC
Human NRC is an approximately 250-kDa protein containing 2063 amino acids (Fig. 2A). Structure-function analysis of NRC in yeast and in mammalian cells (28) revealed a modular structure consisting of: 1) a potent activation domain (AD1); 2) a second Glu, Pro-rich activation domain (AD2); 3) LxxLL-1, which plays an essential role for ligand-dependent interactions with a wide variety of NRs and LxxLL-2, which is more limited in its NR interactions; 4) a dimerization domain near LxxLL-1; and 5) an inhibitory region at the C terminus rich in Ser, Thr, and Leu.
|
600 amino acids) contains a region we designated as "STL," which is rich in Ser, Thr, and Leu. The STL region is inhibitory in the context of full-length NRC, and its deletion enhances the intrinsic activation potential of AD1 and AD2 in NRC (28). Thus, the role of the STL domain may be to control the transcriptional output of NRC through inter- and intramolecular interactions. The STL region contains nine predicted protein kinase A phosphorylation sites in two clusters and 14 predicted casein kinase II phosphorylation sites in four clusters. Thus, the function of the STL region may be modulated by phosphorylation- dephosphorylation, which could alter the activity of NRC. Although rich in Leu, the C-terminal region of NRC does not contain any apparent leucine zipper-like motifs. However, the high content of Leu, Ile, and Val may facilitate certain hydrophobic-driven interactions of NRC with other proteins to regulate the activation potential of NRC in vivo.
For a broad spectrum of NRs, NRC functions primarily as a NR AF2-dependent coactivator. As with other coactivators, helix 12 of the NR AF2 domain is critical in the formation of an interface that contacts the NRC LxxLL-1 module. Point mutations in helix 12 of chicken TR (L398R), or deletion of helix 12, abolish the association of the NR with NRC in vitro and in vivo (28). Although an initial report with ASC-2(NRC) suggested that NRs interact through a region (amino acids 586–860) that did not contain LxxLL-1 (44), other studies indicated that LxxLL-1 was important for interaction with a wide variety of NRs including TRs, RXRs, RARs, GR, ERs, VDR, and the PPARs (28, 45, 49). Mutations of LVNLL to AVNAA or to AVNAL or to LVAAA abrogated the association of NRC with NRs in vitro (28, 45, 49) and in vivo (28). In contrast with NRC, members of the p160 family of coactivators contain multiple LxxLL modules, all of which associate with a variety of NRs, albeit with some differences in specificity (33). Through the use of combinatorial peptide libraries of LxxLL motifs and surrounding residues, LxxLL modules have been characterized as class 1, 2, or 3 depending on the nature of the amino acid residue at position –1 or –2 of LxxLL (LxxLL = +1+2+3+4+5, respectively) (59). The class 2 LxxLL module with a Pro at –2 has been shown to interact with various NRs. The LxxLL-1 region of NRC (LTSPLLVNLLQSDIS) resembles the class 2 LxxLL module and contains a Pro at the –2 position. Members of the p160 family of coactivators contain multiple class 1 LxxLL modules (SRLxxLL), all of which associate with a variety of NRs albeit with some difference in specificity. In contrast, NRC contains a single functional LxxLL-1 module that associates with a variety of NRs in vitro and in vivo with high affinity. Interestingly, the Pro at the –2 position NRC LxxLL-1 motif is also found in the LxxLL (PILTSLL) of TRAP220/DRIP205/PBP and RIP140 LxxLL (PILYYML). The NRC LxxLL module and surrounding amino acids are very hydrophobic and interact with virtually all liganded NRs with high affinity.
In yeast two-hybrid assays, full-length NRC interacts with ligand-bound NRs with similar or higher affinity than full-length SRC-1, and the NRC LxxLL-1 motif appears to display broad specificity yet high affinity toward a large spectrum of NRs (28). The Ser at the –3 position (Ser884) of LxxLL-1 in TRBP(NRC) has also been shown to be important for binding with the LBDs of NRs, particularly for ERß, TR, and RXR (60). Although phosphorylation of Ser 884 by MAPK in vitro reduced its interaction with NRs, it is not known whether such regulation by phosphorylation occurs in vivo (60). Studies with an approximately 300-amino acid region from rat NRC containing LxxLL-1 and AD2 indicated that LxxLL-1 enhances both ligand-dependent activity and the intrinsic basal activity of AD2 (28). Thus, the LxxLL-1 module of NRC embedded in AD2 influences not only the association of NRC with NRs but also the activation potential of AD2.
NRC contains a second C-terminal LxxLL motif designated as LxxLL-2 (EAPTSLSQLLDNSGA), which does not contain a conserved hydrophobic amino acid residue at the –1 position (28). LxxLL-2 does not interact with TR, RAR, RXR, or GR, although it interacts with ER with approximately 10-fold lower affinity compared with LxxLL-1 (28). It has also been reported that ASC-2(NRC) can associate with LXR through LxxLL-2 (61, 62). Interestingly, LxxLL-2 does not resemble any of the known LxxLL motifs identified thus far in various coregulators or from combinatorial phage peptide libraries.
C. NRC is expressed as various isoforms
NRC is expressed as an mRNA of approximately 8–9 kb in various human tissues, which include spleen, thymus, testis, ovary, peripheral blood leukocytes, and brain (28). NRC mRNAs of 6.8 kb, 4.5 kb, and 3.6 kb of varying abundance was also detected in these tissues (28). Interestingly, the 3.6-kb transcript was the predominant NRC mRNA species detected in heart and skeletal muscle, whereas the 4.5-kb transcript was the major form found in testes (28). Mouse PRIP (8- to 9-kb mRNAs) encodes a protein of 2068 amino acids (46). The identity among human, mouse, and rat NRCs is more than 90% at the amino acid level. Analysis of the gene organization of human RAP250(NRC) on chromosome 20q11 predicts that the gene spans approximately 111 kb and consists of 15 exons and 14 introns (63). Alternative splicing of the gene likely reflects the various sized mRNA identified in different tissues. The sequence of RAP250/NRC gene promoter indicates that the promoter is TATA-less and contains four predicted binding sites for Sp1, and sites for C/EBP and Myc/Max (63), suggesting that the gene might be modulated by growth-regulatory signals.
The distribution of NRC protein in mice, studied by immunochemistry using antibody against AIB3/TRBP and RAP250, indicates that relatively higher levels of NRC are present in cells of endocrine target tissues, including testicular Sertoli cells, follicular granulosa cells, and epithelial cells of the prostate, uterus, and mammary gland, as well as kidney tubules (64). NRC protein expression was also detected in thyroid and parathyroid cells and the pancreatic islets of Langerhans. Although medium to low expression was reported in a variety of tissues, the relatively higher levels of NRC observed in reproductive and many types of endocrine tissues suggests that NRC supports NR functions in reproduction and regulation of the endocrine system (64).
The 4.5-kb mRNA species found in human testis was identified as an alternatively spliced form of human RAP250 encoding 1070 amino acids (1–971 and 1965–2063) (45). This isoform lacks LxxLL-2 and a large component (amino acids 972-1964) of the C terminus. A partial cDNA encoding the LxxLL-1 and AD2 followed by a stop codon and a poly A tail was isolated from a rat pituitary GH4C1 cell cDNA library (28). This rat NRC (rNRC.1) reflects an mRNA that lacks LxxLL-2 and the entire C terminus containing the STL region and likely represents one of the shorter NRC isoforms expressed in the pituitary. Because the C-terminal STL region appears to harbor an inhibitory domain, shorter isoforms lacking this region may have elevated intrinsic activity in vivo. The rat genome database has recently annotated a sequence for rat NRC mRNA that encodes a predicted protein of 1844 amino acids. The assembled rat NRC protein lacking parts of AD2 and the C-terminal STL region is presently hypothetical.
D. NRC/AIB3 is amplified and overexpressed in various cancers
A second isoform of human NRC/TRBP/ASC-2/RAP250 has been identified and referred to as AIB3. AIB3 is a 2001-amino acid protein that differs from NRC in its N terminus where the first 88 amino acids of NRC are replaced by 26 different amino acids in AIB3 (NCBI accession no. AF208277). Human NRC is localized on chromosome 20 (20q11). The AIB3 gene was initially mapped to the 20q11 segment of chromosome 20 during a search for amplified and overexpressed genes mapping to chromosome 20q in breast cancer (47). Using fluorescence in situ hybridization analysis, the AIB3/ASC-2 copy number was increased to a moderate level (four to six copies) in 14 of 335 (4.2%) cases of breast cancer and to a high level (more than six copies) in 15 of 335 (4.5%) cases of breast cancer (44). In addition, AIB3 mRNA was also detected in 11 different breast cancer cell lines with the highest expression in BT-474 cells (44). The AIB3 gene was also found to be amplified in lung and colon cancers. The high level of AIB3 expression in breast cancer cell lines does not correlate with the level of ER expression. It should be noted that the probes used to assess gene amplification or mRNA expression do not differentiate between AIB3 and NRC(ASC-2) mRNAs given the fact that the two mRNAs are nearly identical in length and virtually identical in sequence. Therefore, it is unclear whether it is AIB3 or NRC (or both) that are overexpressed in these tumors. Amplification and overexpression of NRC/AIB3 in various cancers underscores its functional importance (discussed later) as a prosurvival gene necessary for normal growth and development.
E. NRC exists as homodimers
Type II NRs, such as TR, RAR, RXR, PPAR, and VDR, primarily activate transcription from target DNA as heterodimers formed with RXR, whereas type I NRs, such as ER, PR, and GR, act primarily as homodimers (1). p160 Coactivators with multiple LxxLL motifs are directly involved in interactions with various NR dimers (35). Similarly, TRAP220 contain two LxxLL motifs, one of which interacts primarily with RXR, whereas the other interacts preferentially with VDR or TR (22). Thus, a single molecule of TRAP220 with two LxxLL motifs has the potential to bind NR/RXR heterodimers. In contrast, NRC contains a single essential LxxLL-1 motif that is necessary for binding with both type I and type II NRs. An interesting question relates to how NRC with a single LxxLL-1 motif binds to and activates NR dimers. This possibility appears to occur through the formation of NRC homodimers, which has been shown to occur using both yeast two-hybrid and in vitro binding assays (28). The homodimerization domain of NRC has been mapped to 146 amino acids (amino acids 849–995 of human NRC) and the corresponding homologous sequence in rat NRC (M. A. Mahajan and H. H. Samuels, in preparation). Thus, a homodimer of NRC with two LxxLL-1 motifs has the potential of binding receptor homo- and heterodimers with high affinity.
F. Ligand-bound nuclear NRs lead to a conformational change in NRC
An interesting paradigm for regulation of the activity of coactivators or other proteins is through conformational transitions that render the molecules active, inactive, stable, unstable, or competent for complex formation or signal transduction. It is well known that the TR LBD undergoes a conformational change upon ligand binding that is necessary for binding of the LBD with LxxLL modules of coactivators (1). An interesting question is whether ligand-bound NR leads to a conformational transition of the coactivator from a relatively inactive to active form. Thus, the activation properties of a coactivator could be modulated through conformational alteration if the activation domain of the coactivator molecule is buried or unexposed when not engaged with ligand-bound NR. Yeast and mammalian two-hybrid assays were used to provide evidence for a conformational change in NRC after association with ligand-bound NR (28). Although LexA-NRC is minimally active when expressed alone in yeast, the activity of a LexA-reporter gene was markedly enhanced by 9-cis-RA when LexA-NRC was expressed with the RXR-LBD lacking a heterologous activation domain (28). This result suggested that association with ligand-bound NR leads to a conformational change in NRC that exposes its AD1 or AD2 activation domains. Similar evidence for a conformational change in NRC by liganded NR was also provided by experiments in mammalian cells using a vector expressing the Gal4-DBD linked to a region of NRC that contained only the LxxLL-1 region and AD2. Although Gal4-DBD-LxxLL-AD2 is moderately active in mammalian cells when expressed alone, its activity is markedly enhanced when coexpressed with the liganded TR-LBD (28). Because the LBDs of liganded NRs are thought to bind only a single LxxLL motif of a coactivator through the hydrophobic groove of the liganded-LBD (35), the enhanced activation of Gal4-DBD-LxxLL-AD2 by liganded NR suggests a conformational change in NRC leading to enhanced AD2 activity. Similarly, PGC-1 has been shown to undergo a conformational change upon NR binding (65). Thus, like NRs that undergo a conformational change when they bind their cognate ligands, the activity of certain coactivators appears to be masked until they associate with ligand-bound NR.
G. Dominant-negative forms of NRC
Expression of a region of a coactivator containing an NR-interaction domain but lacking activation regions might be expected to act as a dominant-negative inhibitor of activation by liganded NRs. Because a broad spectrum of both type I and type II NRs bind LxxLL-1 of NRC, ectopic expression of a region of NRC containing LxxLL-1 might be expected to block ligand-dependent activation. Interestingly, a similar study using the NR-interacting domain of GRIP-1(SRC-2) blocked the effect of GR associated with activator protein-1 (AP-1) but not DNA-bound GR, suggesting that DNA binding of receptors may influence the ability of the receptor to interact with receptor-interacting domains of coactivators (66). Recently, an 80-amino acid region (DN1) of ASC-2(NRC) containing LxxLL-1 expressed in transgenic mice was found to block activation by NRs (67). Transgenic expression of another region (DN2) (amino acids 1431–1511) containing the LxxLL-2 region of ASC-2(NRC) was found to block the activity of LXRs in vivo (68). The livers from the DN2 transgenic mice showed similar changes as seen in livers from LXR–/– mice. These results suggest that NRC plays a role in cholesterol and lipid metabolism in the liver, presumably through regulation of LXR (68). An unexpected result of the transgenic expression of DN1 or DN2 is that these factors were reported to selectively block the activity of endogenous ASC-2 but not other coactivators. Because DN1 containing LxxLL-1 and DN2 containing LxxLL-2 would be expected to compete with the binding of other coactivators for liganded NRs, the mechanism(s) by which these LxxLL regions specifically block the activity of ASC-2, but not other coactivators, is unclear and requires further study.
|
III. NRC Interacts with and Forms Complexes with a Variety of Transcription Factors |
|---|
|
TopAbstractI. IntroductionII. Cloning, Domain Structure,...III. NRC Interacts with...IV. NRC Is an...V. Conclusions and Future...References |
|---|
). Because NRC is a large protein with multiple domains, it is likely that distinct regions of NRC might associate with factors involved in transcription such as transcriptional activators, components of the basal transcription apparatus, the DRIP/TRAP complex, or chromatin-remodeling factors. Thus, like CBP, NRC may act as a transcriptional cointegrator through its interaction with a variety of transcription factors including the NRs. Consistent with this model, NRC plays a role in activating not only NRs but also other factors such as c-Fos, c-Jun, NF-B, ATF-2, CREB, heat shock factors, SRF, and Rb (28, 44, 48, 49, 50, 52, 53, 54). The remainder of this section reviews the factors that have been identified to interact with and possibly play a role in the action of NRC.
A. CBP/p300
NRC has no apparent HAT structure or HAT activity, although this has not been rigorously examined. However, NRC could modulate chromatin structure by recruiting factors with HAT activity to the promoter (e.g., CBP/p300). Consistent with this notion is the finding that NRC forms a high-affinity complex with CBP in vivo (28). Expression of NRC in 293 cells followed by extraction, affinity adsorption, and Western blotting indicated that a significant amount of CBP in the cell is associated with NRC. Association of NRC with CBP involves the C-terminal region of NRC, which includes AD2. p300 was difficult to detect in the same Western blot, suggesting that NRC might preferentially associate with CBP in vivo. This preferential association may reflect, in part, the nonredundant physiological functions of CBP and p300. Although NRC associates with CBP in mammalian cells, no direct association of full-length NRC with full-length CBP was identified in yeast (M. A. Mahajan and H. H. Samuels, unpublished). This suggests that the association of NRC with CBP in mammalian cells may be mediated through another factor. In contrast with these findings, another study reported that ASC-2(NRC) associated with various regions of CBP in yeast and by in vitro binding, although full-length CBP was not examined (61). Other in vitro binding studies indicated that the C-terminal region of p300 (amino acids 1661–2414) can associate with the C-terminal region of TRBP(NRC) (49). Whether association of NRC with CBP/p300 is essential for ligand-dependent activation by NRs or other transcription factors by NRC has not been directly addressed. However, ligand-dependent activation of NRs by NRC was completely blocked by expression of E1A, which is known to inactivate CBP/p300 (28). Taken together, these results support the notion that NRC associates with a CBP/p300 and that NRC may act as a coactivator by recruiting a complex containing CBP/p300 and associated factors to ligand-bound NRs on gene promoters.
B. DNA-dependent protein kinase catalytic subunit (DNA-PKc), DRIP130, and ASCOM
The TRBP(NRC) C-terminal region (amino acids 1237–2063) has been shown to associate in vitro with some components of a DNA-dependent protein kinase complex (49). The components included the catalytic subunit (DNA-PKc), the regulatory subunits (Ku70 and Ku80), poly(ADP-ribose)-polymerase, and DNA topoisomerase I. DNA-PK in vitro phosphorylates TRBP(NRC)-containing amino acids 714–999 in a DNA-dependent manner and amino acids 1237–2063 in a DNA-independent manner (49). In addition, TRBP activates DNA-PK in vitro, the phosphorylation pattern of which differs from that of DNA-mediated activation. Although the role of DNA-PK in recombination and DNA repair has been very well characterized (69), the involvement of DNA-PK in transcription is less clear. However, studies of TRBP activity in cells deficient in DNA-PK indicate a marked reduction in TRBP-mediated enhancement of GR activity, suggesting a role for DNA-PK in transcriptional activation (70). DRIP130, a component of the DRIP/TRAP mediator complex, was reported to associate with the C-terminal region of TRBP(NRC) in vitro (amino acids 1237–2063) (49). Whether DRIP130 and NRC synergistically activate NRs and/or other transcription factors has not been explored.
A steady-state approximately 2-MDa complex referred to as ASCOM (ASC-2 complex) consisting of about eight proteins was isolated from HeLa nuclei using a monoclonal antibody against ASC-2(NRC) (71). In addition to ASC-2(NRC), the complex includes the Trithorax-related proteins ALL-1/MLL, ALR-1(MLL2), ALR-2, HALR (MLL-3), ASH2, a 66-kDa retinoblastoma binding protein, RBQ3, and 48-kDa - and ß-tubulin proteins. Given the composition and the nature of proteins associated with ASC-2(NRC), the ASCOM complex is clearly distinct from other known coactivator complexes identified for NRs and other transcription factors. How ASCOM acts to regulate transcription is not known, although recombinant ALR-1 and HALR exhibit weak histone H3 lysine 4 methylation activities in vitro. Using ChiP assays, ASH2, ALR, and ASC-2 were reported to be recruited to the RARß and the p21WAF1 gene promoters, which are activated by liganded RAR, and this recruitment was inhibited when the DN1 dominant-negative peptide containing LxxLL-1 was expressed (71). DN1 did not affect the recruitment of TRAP220 or SRC-1 to the p21WAF1 promoter. This is surprising because NRC(ASC-2), TRAP220, and all the p160 coactivators are thought to use the same interaction surface of liganded RAR. Although the mechanism of such specific inhibition is unclear (71), a possible explanation is that the DN1 LxxLL-1 region blocks dimerization of NRC, thus preventing its binding to NR dimers (28). Although ASC-2(NRC) was shown to be a component of the ASCOM complex, it is unclear whether ASC-2(NRC) acts only through this complex to activate NRs and other transcription factors such as c-Fos, c-Jun, CREB, and NF-B. In addition to CBP/p300 and the ASCOM complex, a number of other NRC-interacting factors, referred to as NIF-1 (50), coactivator activator (CoAA) (72), coactivator protein for AP-1 and ER receptor (CAPER) (73) and PRIP-interacting methyltransferase (PIMT) (74), which may play important roles in the actions of NRC, have been identified.
C. NRC-interacting factor-1 (NIF-1) and NIF-2
NIF-1 was isolated with a yeast two-hybrid screen using LexA-NRC (amino acids 849-2063) as bait, which contains the C-terminal half of NRC and includes LxxLL-1, AD2, LxxLL-2, and the STL region (50). Both rat and human NIF-1 proteins were cloned and characterized. NIF-1 is a zinc finger protein that is highly conserved across avian, mouse, rat, and human species. Human NIF-1 is a 1342-amino acid nuclear protein that contains a number of highly conserved domains including six zinc fingers, an N-terminal acid-rich region, an LxxLL motif, and a C-terminal leucine zipper-like motif. The LxxLL motif of NIF-1 does not directly interact with type I or type II NRs (50). Human NIF-2 was identified as an alternatively spliced isoform that is identical to NIF-1 but lacks the first three zinc fingers (50). NRC associates with NIF-1 through a region containing 146 amino acids (amino acids 849–995) of NRC, referred to as the NIF-interaction domain (NIF-ID). NIF-1 interacts with the NIF-ID of NRC through a region containing zinc finger 6. Although the NIF-ID of NRC contains LxxLL-1, this motif is not involved in interaction with NIF-1 (50). It is interesting that although the NIF-ID and the NR interaction region of NRC are in close physical proximity, associations of NIF-1 and NRs with NRC are not mutually exclusive. Zinc fingers 1, 2, and 3 in NIF-1 form a unique structure referred to as BED fingers (75), a motif conserved in Drosophila BEAF and DREF (76) and several other eukaryotic transcription factors such as Sp1 (77). BED finger proteins are thought to function as either activators or repressors by modifying local chromatin structure upon binding to GC-rich sequences (75). Thus, NIF-1 may modulate chromatin structure as a part of a coactivator complex and facilitate the activation potential of NRs and other factors. Another possibility is that NIF-1 may also bind insulator DNA sequences directly and facilitate the process of transcription. Interestingly, NIF-2 lacks the BED fingers but retains the ability to bind NRC through zinc finger 6. This raises the possibility that NIF-2 may be a naturally occurring dominant-negative form of NIF-1, although this has not been verified experimentally. Both NIF-1 and NIF-2 are ubiquitously expressed in a number of human tissues, although NIF-1 mRNA levels are expressed at much higher levels than NIF-2 (50). NIF-1 contains a number of potential phosphorylation sites for various kinases that may play a role in modulating the function of NIF-1. Although NIF-1 does not directly interact with ligand-bound NRs, it markedly enhances ligand-dependent activation (50). In addition to NRs, NIF-1 enhances the transcriptional activity of c-Fos and c-Jun (50). Thus, NIF-1 displays a similar activation profile as NRC and interacts with NRC in vivo. Because NIF-1 does not directly bind to NRs, its action is likely through its association with a primary coactivator such as NRC. Factors that do not directly bind transcriptional activators but modulate the activity of coactivators, which directly bind NRs and other factors, have been referred to as cotransducers (50). Cotransducer-like molecules have also been referred to as secondary coactivators.
D. RNA-binding proteins, CAPER, CoAA, and PIMT
In addition to NIF-1, three other cotransducer-like proteins, referred to as CAPER, CoAA, and PIMT, were identified in yeast two-hybrid screens to interact with NRC/ASC-2/PRIP/TRBP. In contrast with NIF-1, CAPER, CoAA, and PIMT contain RNA recognition motifs (RRMs). Although the extent of activation of NRs by PIMT, CAPER, and CoAA is only moderate (1.8- to 3-fold), this does not exclude important roles for these factors because many established coactivators only moderately enhance activation in transfection assays. CAPER (coactivator protein for AP-1 and ER receptor) was isolated from a mouse liver cDNA library using the C-terminal region (amino acids 1172–1729) of ASC-2(NRC) as bait (73). CAPER is identical to the nuclear autoantigens HCC1.3 and HCC1.4 reported in hepatocarcinoma (78). HCC1.3 and HCC1.4 are identical except for six additional amino acids in HCC1.4. Mouse CAPER (HCC1.3) was reported to enhance activation of ER and c-Jun. CAPER contains three RRMs and associates with c-Jun and the ligand-bound ER-LBD region via RRM3, whereas the C terminus of CAPER was shown to bind ASC-2(NRC) (73). CAPER did not appear to enhance activation of other NRs such as RARs, RXRs, TR, and GR. The basis for ER specificity was not clarified, but the specificity implies that CAPER may not enhance ER function through association with ASC-2(NRC), which interacts with and enhances activation of NRs including RARs, RXRs, TR, and GR.
Another RRM-containing factor referred to as coactivator activator (CoAA) is a 669-amino acid protein isolated from a GC cell cDNA library in a yeast two-hybrid screen using the C-terminal region (amino acids 1641–2063) of TRBP(NRC) as bait (72). CoAA contains two RRMs near its N terminus whereas its C-terminal region interacts with NRC. Whether CoAA interacts directly with various NRs has not been examined. Expression of CoAA enhances activation by a number of transcription factors including NF-B, CREB, AP-1, TR, ER, and GR. CoAA was also shown to associate with DNA-PK-regulatory subunit Ku86 and poly(ADP-ribose)-polymerase from GH3 cells, suggesting that CoAA may exist as part of a DNA-PK/TRBP complex in vivo.
PIMT was identified in a yeast two-hybrid screen using a human liver cDNA library with the C-terminal region (amino acids 773-2067) of PRIP(NRC) (74). PIMT is a ubiquitously expressed putative RNA methyltransferase, found in K-homology motifs of many RNA-binding proteins, that contains an invariant GXXGXXI motif near its N terminus. Transient expression of PIMT enhances the activity of PPAR and RXR, which is further enhanced by expression of PRIP(NRC). The putative methyltransferase activity of PIMT does not appear to be involved in its role as an activator because deletion of the methyltransferase domain does not affect its activating function (74). Like NRC, PIMT homodimerizes in vitro and may also form homooligomers (74). PIMT does not appear to methylate histones but binds CBP/p300 and PBP(DRIP205/TRAP220) (79). CAPER, CoAA, PIMT, and NIF-1 were not identified as components of the ASCOM complex (71). Thus, these factors may associate transiently with the ASCOM complex and are lost during purification or are components of other novel NRC protein complexes.
|
IV. NRC Is an Essential Coactivator and a Pleiotropic Modulator Affecting Growth, Development, Reproduction, Apoptosis, and Wound Healing |
|---|
|
TopAbstractI. IntroductionII. Cloning, Domain Structure,...III. NRC Interacts with...IV. NRC Is an...V. Conclusions and Future...References |
|---|
NRC–/– MEFs have also been useful in examining the effect of NRC on the activity of various NRs, and these cells display a reduction in the activities of RXR, RAR, TR, and PPAR (55, 56, 57, 89). Although different laboratories have reported differences in the extent of activation by NRs in NRC–/– MEFs, the overall conclusion is that NRC contributes to the activities of RXR, RAR, TR, and PPAR.
B. Growth and reproductive phenotypes of NRC+/–129S6 newborn and adult mice
Although NRC+/– mice in the C57/129 mixed genetic background appear outwardly normal and grow and reproduce similar to NRC+/+ mice, the penetrance of a specific phenotype is frequently dependent on the genetic background of the mouse strain. Because the embryonic stem cells used for homologous recombination are derived from the 129S6 strain, the effect of NRC heterozygosity was examined in 129S6 mice generated in our laboratory (58). These NRC+/–129S6 mice exhibit a number of interesting phenotypes primarily involving growth, reproduction, and wound healing (58).
1. Growth retardation phenotype.
NRC+/–129S6 mice exhibit a neonatal growth phenotype in which the newborn pups are approximately 10–15% smaller than their wild-type littermates (58). However, after weaning the NRC+/– mice exhibit "catch-up" growth and within 2–3 months appear similar in size to their NRC+/+ littermates. Although similar growth retardation has also been noted among NRC+/– neonates in the C57/129 background, it is much less profound and is found in less than 5% of NRC+/–C57/129 neonates. Interestingly, approximately 3% of the NRC+/–129S6 newborn pups are extremely growth stunted, weighing 70% less than their wild-type littermates (Fig. 3). These mice often die before weaning. However, a small number of these mice survive and exhibit "catch-up" growth and after 3 months appear similar in size to wild-type mice (58). Interestingly, the growth phenotype found in NRC+/–129S6 mice resembles that found with TIF-2–/– (SRC-2) mice, which also exhibit transient postnatal growth retardation (83). pCIP–/– (SRC-3) newborn mice are also uniformly smaller in size, but the weight deficit persists through adulthood. This is thought to reflect a role for pCIP–/– (SRC-3) in mediating effects through the IGF-I pathway (84, 90). Thus, the growth phenotype of pCIP–/– mice differs from that of NRC+/– mice because pCIP–/– mice remain small throughout their adult life. The molecular mechanisms underlying growth retardation in NRC+/–129S6 mice have not yet been defined but likely reflect an effect of NRC on one or more transcription factors that play an important role in growth (e.g., AP-1).
|
/ß–/– male mice become completely sterile within 6 months (91). Because NRC acts as a coactivator for LXR (68), the progressive loss of fertility in NRC+/– male mice may reflect this effect of NRC on the LXRs in the testes.
C. NRC+/– mice exhibit a wound-healing phenotype resulting from a defect in keratinocyte migration
Although NRC+/–C57/129 mice have been reported to appear normal (55, 56, 57), we noted that as they grow older they exhibit a wound-healing phenotype (58). The mice spontaneously develop skin lesions or ulcers around the neck, ears, snout, and facial area (Fig. 4). These regions correspond to the areas where mice groom and likely scratch themselves. This occurs in approximately 25% of both male and female NRC+/–C57/129 as well as NRC+/– 129S6 mice. These spontaneous lesions develop as early as 4 months of age in the NRC+/–C57/129 mice and somewhat earlier in the 129S6 background (58).
|
). This lack of keratinocyte migration was reproduced using ex vivo skin explant cultures from 2-d-old C57/129 NRC+/– mice (58). Whereas NRC+/+ keratinocytes showed robust migration when incubated with epidermal growth factor (EGF), keratinocytes from the NRC+/– explants showed little or no response to EGF (Fig. 5).
|
6 and ß1) (92). These integrins are thought to play an important role in keratinocyte migration and in self-renewal of epidermal stem cells leading to their gradual depletion over time. Other microarray studies (93) indicate that the majority of down-regulated genes identified in the keratinocytes overexpressing c-Myc are involved in cell adhesion and the cytoskeleton, including expression of 4ß6 integrin, which may be considered as a marker of epidermal stem cells.
The similar phenotypes found in the mice heterozygous for expression of K14/c-Myc and in the NRC+/– mice suggest that NRC may play a role in regulating the genes targeted by overexpressed c-Myc. For example, NRC is a potent coregulator of c-Jun activity (48, 49, 50), which plays an important role in regulating genes involved in wound healing (integrins, laminins, and extracellular matrix genes in the epithelium) (94). In addition, c-Jun has recently been shown to stimulate expression of heparin-binding EGF in the basal layer of wounded epithelium which, in turn, stimulates signaling by EGFR to regulate keratinocyte migration through regulation of focal adhesion kinase (95). Because NRC is known to play an important role in c-Jun signaling (48, 49, 50), the impaired wound-healing phenotype of NRC+/– mice can be explained, at least in part, by a reduction in c-Jun activity. In addition to c-Jun, other transcription factors and signaling molecules (e.g., NF-B and the PPARs) involved in wound healing are known to be regulated by NRC (28, 44, 46, 49, 50, 57). Both PPAR and PPARß have been shown to be important in wound healing (96, 97). Although both PPAR and PPARß are not expressed in adult epithelium (except for hair follicle keratinocytes), both are rapidly expressed after the development of a wound, and PPARß remains expressed throughout the entire wound-healing process (97). PPAR is thought to play an important role in the initial inflammatory process, whereas PPARß is involved in keratinocyte proliferation (97). PPARß+/– mice exhibit a significant delay in wound closure, and in vitro studies with PPARß+/– keratinocytes indicate a decrease in keratinocyte adhesion and mobility (97). PPARß also acts to block apoptosis of keratinocytes by regulation of the acute transforming retrovirus thymoma protein kinase 1 pathway through stimulation of expression of integrin-linked kinase (ILK) and 3-phosphoinositide-dependent kinase 1 (PDK1) and repression of phosphatase and tensin homolog deleted from chromosome 10 (PTEN) (98). The precise mechanism(s) by which a reduction of NRC in the skin of NRC+/– mice leads to chronic skin wounds needs to be further defined. However, its role as a coactivator for c-Jun and the PPARs and its role as a prosurvival antiapoptotic factor suggests that NRC acts through regulation of one or more of the above factors and functional pathways involved in the wound-healing process.
|
V. Conclusions and Future Perspectives |
|---|
|
TopAbstractI. IntroductionII. Cloning, Domain Structure,...III. NRC Interacts with...IV. NRC Is an...V. Conclusions and Future...References |
|---|
B, ATF-2, heat shock factors, E2F-1, SRF, and Rb (28, 44, 48, 49, 50, 51, 52, 53, 54). Thus, like CBP/p300, NRC may function as transcriptional cointegrator of NRs and other important transcription factors involved in growth, proliferation, cytokine signaling, metabolism, the immune response, and apoptosis. Given the wide spectrum of factors that are regulated by NRC, an interplay of signaling pathways is expected to be affected as a result of NRC expression. It would be of significant interest to determine whether these factors associate directly with NRC or whether they interact with other components of an NRC protein complex. In this regard, many of the factors described above have been shown to interact with CBP/p300, suggesting that NRC may function to enhance the activity of such factors though its interaction with CBP/p300, which occurs in vivo. In addition to conformational alterations in NRC mediated by liganded NRs, and possibly other transcription factors, posttranslational modifications that may regulate NRC activity (e.g., phosphorylation of the C-terminal STL-inhibitory region) could play an important role in modulating the intrinsic activity of NRC. Such phosphorylation of NRC by protein kinase A and other kinases may serve to communicate and integrate regulatory events mediated by NRs or other transcription factors and the signaling cascades generated by cell surface receptors. NRC has been shown to interact with a number of factors in vitro and in vivo (CBP/p300, DNA-PKc, DRIP130, NIF-1, CAPER, CoAA, PIMT) and to be part of the ASCOM complex). This raises the possibility that NRC may be a component of distinct transcriptional regulatory complexes as has been found for the various "mediator" (DRIP/TRAP) complexes, which have somewhat different protein compositions. In addition, a systematic analysis of protein complexes of the yeast proteome indicates that many proteins are shared by distinct protein complexes that otherwise have very different compositions. Given the complexity of the intermolecular network of NRC and its multiple functions, it is likely that NRC is a component of a variety of regulatory complexes in the cell.
Unlike the p160/SRC family of coactivators, NRC is a single-copy gene in the human genome. Because NRC appears to modulate the activity of many transcription factors, it is not surprising that deletion of both NRC alleles is embryonic lethal. Thus, like CBP, p300, or TRAP220, NRC appears to be an essential gene involved in the regulation of growth, development, and cell survival. Although full-length NRC has been studied in detail, various sized mRNAs are expressed in different tissues that likely reflect alternative splicing of the NRC gene. Most of these isoforms, some of which are tissue specific, have not been well characterized. Depending on the composition of the protein, such isoforms may mediate specific effects or selectively interact with specific protein complexes in the cell. Thus, some of the described isoforms lack the C-terminal STL region that interacts with CBP/p300, DRIP130, and DNA-PKc while retaining the interaction domain for NIF-1, PIMT, and CoAA. In addition, such an isoform would lack LxxLL-2, which exhibits preference for the LXRs. Furthermore, a NRC isoform lacking the inhibitory C-terminal STL region, but which retains LxxLL-1 and AD2, might be expected to be a more active isoform of NRC. Thus, although NRC is a single-copy gene, tissue-specific alternative splicing of the gene may generate isoforms that may exhibit differences in intrinsic activity and/or lead to specific or restricted target tissue response(s) compared with full-length NRC.
In summary, NRC/ASC-2/RAP250/TRBP/PRIP is an essential coregulator of NRs and other transcription factors and exhibits pleiotropic effects on a wide variety of processes. Recent studies of NRC+/– mice indicate that haploid expression of NRC leads to a variety of phenotypes indicating that NRC+/– mice may prove useful in deciphering the role(s) of NRC in various processes in vivo. These studies, and studies on the various NRC isoforms and NRC complexes, should lead to a more comprehensive understanding of the contribution(s) of NRC in mediating effects of the NRs and other factors involved in controlling growth, development, apoptosis, wound healing, reproduction, and metabolic regulation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
First Published Online December 7, 2004
Abbreviations: AD, Activation domain; AF, activation function; AIB3, amplified in breast cancer 3; AP-1, activator protein 1; AR, androgen receptor; ASC-2, activating signal cointegrator-2; ASCOM, ASC-2 complex; ATF-2, activating transcription factor-2; CAPER, coactivator protein for AP-1 and ER receptor; CBP, cAMP response element binding protein (CREB)-binding protein; c-Fos, cellular homolog of the vfos oncogene; ChiP, chromatin immunoprecipitation; c-Jun, cellular homologue of an oncogene in the avian sarcoma virus 17; CoAA, coactivator activator; CREB, cAMP response element-binding protein; DBD, DNA-binding domain; DNA-PKc, DNA dependent protein kinase catalytic subunit; dpc, days post conception; DRIP, vitamin D receptor-interacting protein; E1A, adenovirus early expressed protein 1; EGF, epidermal growth factor; ER, estrogen receptor; GR, glucocorticoid receptor; GRIP, GR-interacting protein; HAT, histone acetyltransferase; LBD, ligand-binding domain; LXR, liver X receptor; MEF, mouse embryo fibroblast; MR, mineralocorticoid receptor; NCoA, nuclear receptor coactivator; NF-B, nuclear factor B; NIF, NRC-interacting factor; NR, nuclear hormone receptor; NRC, nuclear receptor coactivator/coregulator; PBP, PPAR binding protein; p160, 160 kDa proteins of the SRC family; pCIP, p300/CBP/cointegrator-associated protein; PGC-1, PPAR coactivator-1; PIMT, PRIP-interacting methyltransferase; PPAR, peroxisome proliferator-activated receptor; PR, progesterone receptor; PRIP, PPAR interacting protein; RAP250, 250-kDa receptor associated protein; RAR, retinoic acid receptor; Rb, retinoblastoma gene product; RIP, receptor-interacting protein; RRM, RNA recognition motif; RXR, retinoid X receptor; Sp1, SV40 virus promoter-specific transcription factor; SRC, steroid receptor coactivator; SRF, serum response factor; STL, serine, threonine, leucine; TIF, transcriptional intermediary factor; TR, thyroid hormone receptor; TRAP, TR-associated protein; VDR, vitamin D receptor.
|
References |
|---|
|
TopAbstractI. IntroductionII. Cloning, Domain Structure,...III. NRC Interacts with...IV. NRC Is an...V. Conclusions and Future...References |
|---|
