| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular and Medical Pharmacology (D.J.M.), University of California Los Angeles School of Medicine, Los Angeles, California 90095; The Prostate Centre at Vancouver General Hospital (D.J.M., S.D., C.C.N.), Vancouver, British Columbia, Canada V6H 3Z6; and University of Southern California/Norris Cancer Center, Keck School of Medicine (G.A.C.), Los Angeles, California 90033
Correspondence: Address all correspondence and requests for reprints to: David J. Mulholland, Department of Molecular and Medical Pharmacology, 650 Charles E. Young Drive, Center for Health Sciences 23-234, University of California Los Angeles School of Medicine, Los Angeles, California 90095. E-mail: dmulholland{at}mednet.ucla.edu
 |
Abstract |
|---|
 Top Abstract I. IntroductionII. The Wnt/ß...III. Wnt/ß-Catenin/Tcf...IV. Nuclear Receptor Modulation...V. Summary and Therapeutic...References |
|---|
|
I. Introduction |
|---|
|
TopAbstractI. IntroductionII. The Wnt/ß...III. Wnt/ß-Catenin/Tcf...IV. Nuclear Receptor Modulation...V. Summary and Therapeutic...References |
|---|
Novel ligands of NRs are continuously being developed and include synthetic steroid agonists and antagonists, ligands that alter fatty acid synthesis, analogs of vitamins A and D, ligands with antidiabetic qualities, and ligands with anticancer attributes (7, 8). Therapeutic intervention is well exemplified by the use of RA (9) and tamoxifen (10), both of which function as chemopreventative agents. Thus, NRs and their cognate ligands serve as potent regulators of development, cell differentiation, and normal physiology but may also have important implications for pathologies such as cancer (11). Given the fundamental role that NRs serve in maintaining a normal cellular milieu, it is not surprising that NRs and their cognate ligands can functionally interact with potent oncogenic systems, such as Wnt, to elicit changes in cellular adhesion and oncogenesis.
It is generally appreciated that a loss of cell adhesion is concomitant with metastatic cancer of endocrine tissues (12, 13). As such, a large number of reports have appeared during the previous 5–10 yr pertaining to immunohistochemical distribution of cadherins and catenins in endocrine-related cancers, which have been subject to several comprehensive reviews (14, 15). However, the usefulness of immunohistochemistry to evaluate changing expression of cell adhesion molecules as a predictor of clinical outcome is frequently confusing and unreliable, suggesting that alternative approaches for analysis of endocrine/adhesion interactions are necessary. In the previous 4–5 yr, there has been a tremendous expansion in the number of reports pertaining to the functional interactions between NRs and the canonical, Wnt signaling pathway cascade. Increased laboratory technology, enhanced generation of recombinant Wnts, and the development of pertinent transgenic animals have made it clear that these interactions are of greater functional importance than previously appreciated. Evaluation of NR interactions with canonical Wnt/ß-catenin signaling will likely aid in supplementing knowledge obtained by immunohistochemical analysis of pathology specimens and systemic endocrinology.
The Wnt signaling pathway is pivotal to gene expression (16), cell adhesion (17), and tissue development (18). Wnt ligands and ß-catenin/T cell factor (Tcf) signaling are also potent initiators of human oncogenesis (19) such that mutations in regulatory molecules, including the adenomatous polyposis coli (APC) tumor suppressor and ß-catenin, have been shown to be key predictors of cancer progression (20). With the detection of ß-catenin mutations in primary prostate cancers (PrCas) (21) followed by the exciting identification of a functional AR/ß-catenin interaction (22), a flurry of studies have evaluated the nature and diversity of ligand-sensitive interactions between AR and the ß-catenin/Tcf signaling axis. Recently, exciting evidence for an oncogenic role for AR in PrCa was described by way of a conditional, transgenic mouse model harboring a single AR mutation (E231G), ultimately leading to metastatic PrCa (23). Increased AR expression has been consistently associated with development of cells that become supersensitive to androgen, facilitating the development of an "androgen-independent" phenotype. Also, increased AR expression is necessary and sufficient to convert PrCa to an ablation-resistant state (24). Therefore, anything affecting AR signaling, such as ß-catenin, may be expected to have profound effects on PrCa progression. Furthermore, it is not surprising that other members of the NR family have been shown to respond in an equally dynamic manner. Thus, it is with interest that we assess how members of the NR superfamily can be modulated by components of the highly oncogenic Wnt signaling pathway.
The emergence of studies demonstrating cross-regulation of Wnt/ß-catenin/Tcf signaling with that of NRs provides an enticing platform on which to evaluate alterations in cell adhesion and transcription potentially occurring during endocrine cancer progression.
Wnt ligands, ß-catenin, Tcfs, cyclin D1, and glycogen synthase kinase ß (GSK3ß) have all been shown to alter NR function by events including transcriptional activation, repression, and phosphorylation. Conversely, NRs and their ligands confer dynamic effects upon Wnt function, as demonstrated by their abilities to promote dramatic alterations in E-cadherin, ß-catenin gene targets, and Wnt-regulated physiology. Furthermore, GSK3ß, has emerged as a more promiscuous kinase than previously appreciated with the capacity to regulate NRs both directly, by way of posttranslational modification, and indirectly by directing IGF-I-dependent accumulation of ß-catenin.
Thus, in this review we summarize and interpret recent reports implicating functional interactions between NRs, including AR, RAR, RXR, glucocorticoid (GC) receptor (GR), TR, VDR, ER, and PPAR, and Wnt signaling with respect to alterations in gene expression, pathology, translational relevance, and potential therapeutic opportunities.
|
II. The Wnt/ß-Catenin/Tcf Signaling Pathway |
|---|
|
TopAbstractI. IntroductionII. The Wnt/ß...III. Wnt/ß-Catenin/Tcf...IV. Nuclear Receptor Modulation...V. Summary and Therapeutic...References |
|---|
-catenin; 2) the cytoplasm; and 3) the nucleus in association with other transcription factors (17). These pools of ß-catenin associate with partners such as E-cadherin or Tcf, mostly by way of the 12 Armadillo repeats, each containing three -helices (32). Soluble and nonsoluble cytosolic forms of ß-catenin are strictly regulated by the proteosome/ubiquitination system that consists of GSK3ß, axin, and APC (33, 34, 35, 36, 37).
Upon increased cellular levels and nuclear accumulation, ß-catenin binds the amino terminus of the high-mobility group binding protein, Tcf, and promotes its interaction with target DNA sequences (A/T A/T CAAAG) (38), thereby promoting displacement of the Tcf repressors, Groucho and CtBP (39). This, in turn, leads to a concomitant recruitment of coactivators such as CREB binding protein (CBP)/p300 (40), Brgl, and CARM1 (41) and an overall de-repression of Tcf transcription (42, 43). Activation of the Wnt/ß-catenin/Tcf signaling pathway (Fig. 1A
), either by disengagement of the APC/axin/GSK3 complexing or by Wnt activation promotes induction of downstream gene targets such as cyclin D1 (44), c-myc (45), PPAR
(46), Tcf-1 (38, 47), matrilysin (48) and CD44 (49). Induction of these genes has dramatic effects on cell and tissue development and oncogenesis (50, 51, 52, 53, 54). Negative Wnt signaling (Fig. 1B) occurs by CK-1 phospho-priming of ß-catenin; GSK3ß phosphorylation of free ß-catenin at Ser 33, 37, 45 and Thr 41 (55); and complexing with APC. This allows recognition by the F-box protein, ßTrCP, and ultimately degradation by the ubiquitin/proteosome pathway (56). ß-Catenin levels are also regulated by Siah, a p53-inducible component that functions independently of GSK3ß to recruit APC to target ß-catenin for proteosome degradation (57, 58).
|
|
III. Wnt/ß-Catenin/Tcf Modulation of Nuclear Receptor Function |
|---|
|
TopAbstractI. IntroductionII. The Wnt/ß...III. Wnt/ß-Catenin/Tcf...IV. Nuclear Receptor Modulation...V. Summary and Therapeutic...References |
|---|
). For example, Wnt-4 expression is required, in a progesterone-dependent manner, for mammary gland development, as exemplified by the fact that Wnt-4 –/– mice fail to form ducts upon glandular transplantation to Wnt 4 +/+ mice (62). Wnts are also potentiators of endocrine morphogenesis because mammary epithelium stimulation with ectopic Wnt-1 can overcome the genetic loss of PR. Wnt-1 overexpression can promote breast cancer, as demonstrated in mouse mammary tumor virus (MMTV)-Wnt-1 transgenic mice that, when crossed with mice null for ER, produce offspring that develop hyperplasia and tumorigenesis by age 36–48 wk (63, 64). Nevertheless, it is also apparent that Wnt-1 cannot substitute for the role that ER serves in ductal morphogenesis (63, 64). That Wnt-1 can promote tumorigenesis in the absence of estrogens or ER further demonstrates that estrogen-independent mammary oncogenesis may be compensated for by growth factor pathways such as Wnt. These data also imply that therapeutic inhibition of Wnt-driven ß-catenin/ER interactions may merely serve to reroute Wnt proliferative signals through other effectors of proliferation.
|
The partnership between Wnts and PPAR not only demonstrates a compelling example of how a Wnt-NR partnership can be essential for tissue morphogenesis, but also preempts another theme put forth in this review, that NRs may have potent repressive effects upon the Wnt/ß-catenin/Tcf axis (Fig. 3D). In the presence of cognate ligand, PPAR, including isoforms , 
, and , regulates growth, differentiation, apoptosis (70), and insulin sensitivity (71). Activation of PPAR is regulated by association with lipophilic ligands, including endogenous prostaglandins or synthetic thiazolidinediones (TZDs). In the presence of adipogenic stimuli, PPAR/RXR heterodimers function to promote differentiation of preadipocytes to adipocytes (72). Curiously, Wnt signaling functions as a promoter of preadipocyte growth and proliferation, it also functions as a potent inhibitor of adipogenesis. For preadipocyte differentiation to occur, critical modulation of Wnt signaling must take place (73, 74, 75). Mechanistically, a reciprocal interplay of inhibitory signals occurs between PPAR and Wnt/ß-catenin signaling, such that in the absence of adipogenic stimuli, Wnts 1 and 10 promote growth and cell proliferation of preadipocytes by activating the Wnt targets, cyclin D1 and c-myc, while simultaneously inhibiting PPAR. Cyclin D1 and c-myc facilitate these effects by binding PPAR and the CCAAT/enhancer-binding protein (C/EBP) transcription factor, respectively (44, 76, 77). The presence of adipogenic stimuli, such as troglitazone, promotes induction of PPAR and C/EBPß to facilitate a mandatory reduction in Wnts 1 and 10 (Fig. 3D). Expression of C/EBPß and C/EBP also coincides with phosphorylation of the ß-catenin regulatory domain and its proteosome degradation (73, 74, 78). Inhibition of adipogenesis relies upon CK1/GSK3ß phosphorylation of ß-catenin and not the Siah/p53 axis of degradation (GSK3ß independent) (203), although some have conjectured that PPAR may also promote degradation of ß-catenin by a pathway that is strictly proteosome dependent and APC/GSK3ß/p53 independent (79). PPAR activation by TZD is also capable of directly influencing cyclin D1 in a cAMP-response element binding protein (CREB)-dependent, ß-catenin-independent manner resulting in hepatocyte arrest (80). PPAR possesses tumor suppressor functions that have been proposed to be elicited by activation of phosphatase and tensin homolog deleted on chromosome 10 (PTEN), a lipid phosphatase and negative regulator of Akt/protein kinase B (81, 82). These data imply that PTEN-mediated antagonism of PI3K/Akt signaling and reduced inhibition of GSK3ß may serve as a means to enhance adipogenic stimulation. This mechanism is congruent with other reports that have implicated PTEN as a negative regulator of ß-catenin by way of the Akt/GSK3ß axis (83, 84). The liver X receptor (LXR) may also participate in regulation of adipogenesis. However, unlike PPAR the LXR appears as a potential negative regulator of adipogenesis, possibly by collaboration with ß-catenin (75). However, these observations remain preliminary with more investigation being required before a functional partnership between LXR and ß-catenin in lipogenesis can be assigned. In addition to being highly expressed in adipose tissue, both PPAR and PPAR are thought to serve important roles in colon carcinogenesis, specifically as immediate targets of Wnt/ß-catenin/Tcf activation. ß-Catenin and Tcf may serve both to bind and coactivate PPAR-driven response elements, thereby activating PPAR target genes including fatty-acid binding protein-2, adipophilin-2, and keratin 20 (85). Increased expression of PPAR is also evident in both human colonic tumors (86) and the APCMin mouse model (85, 87, 88). Importantly, synthetic agonists of PPAR have also been shown to induce increased colonic tumor mass in APCMin mice, suggesting a link between a high-fat diet and carcinogenesis of the colon (88). Thus, although the role of PPAR as a contributor to colon carcinogenesis is considered somewhat contentious (89), convincing in vitro and in vivo evidence suggests PPAR isoforms are activated and overexpressed in colon cancer systems.
|
AR and ß-catenin interact by direct binding and complexing, as ascertained by yeast two-hybrid analysis (95, 96), GST pull downs (95, 97), coimmunoprecipitations (22, 95, 98), and transcriptional reporter assays (22, 95, 96, 98, 99). Other NRs that have been demonstrated to interact with ß-catenin include the RAR (100, 101, 102, 103), VDR (104), RXR (105), PPAR (78, 80, 85), and most recently ER (107). To date, however, neither the PR, GR, or TR have been shown to directly interact with ß-catenin (97, 99). Despite this, NRs failing to directly associate with ß-catenin, in many instances, interact with other members of the Wnt signaling pathway including Wnts, GSK3ß, cyclin D1, Tcf4, and Tcf1.
AR/ß-catenin interactions are ligand sensitive, whereby complexing occurs in the presence of dihydrotestosterone (DHT), or R1881, less in the absence of ligand (22, 98), and not in the presence of the pure AR antagonist, bicalutamide (96, 98). Transient transfections of deletion mutant expression plasmids and yeast two-hybrid studies suggest that the ligand binding domain of AR (ARLBD) is necessary and sufficient for AR/ß-catenin interactions (95, 96). Reduced AR/ß-catenin binding, in the presence of pure AR antagonists, may be explained by formation of unfavorable stoichiometry by helix 12 of ARLBD, with respect to its binding pocket, thereby preventing efficient coactivator binding (108, 109). Despite the conservation in structure between NRs, it is clear that the ARLBD has unique structural aspects that facilitate binding to ß-catenin. LXXLL binding motifs (L = leu, X = any amino acid) contained within the ARLBD serve a different function than most other NRs. In AR, LXXLL binding regions of the LBD serve to mediate NTD (containing FXXLF) and LBD interactions (110, 111), whereas in most other NRs, LXXLL binding motifs serve primarily to recruit transcriptional coactivators (112). Mutation of ARLBD helices 3 and 12 results in disruption of AR/ß-catenin binding, with alteration of helix 3 affecting binding of all three, ß-catenin, the ARNTD, and the transcriptional regulator, TIF2 (transcriptional initiation factor) (96). If ß-catenin/ARLBD binding depends on functional LXXLL binding motifs found within the AR, it is conceivable that ß-catenin is required for changes in structural conformation that are coincident with ligand binding and dissociation of heat shock proteins. Regions of ß-catenin necessary for interaction with AR have also been well defined. ß-Catenin Arm repeats 1–6 are required because mutation of repeats 5 or 6 abolishes binding, coactivation, and nuclear cotranslocation interactions with AR (95). Importantly, Arm repeats 5 or 6 also bind Tcf4 and E-cadherin (32, 113), suggesting the therapeutic possibility of simultaneously disrupting the coactivating effects that ß-catenin may confer upon either AR or Tcf4. That overexpression of Tcf4 or E-cadherin blocks ß-catenin interactions with the AF2 region of AR indicates that ß-catenin binds these molecules in close proximity of the Arm repeats (96, 98). ß-Catenin also contains LXXLL motifs, found on the second -helix of Arm repeats 1, 7, 10, and 12 (32, 114). However, deletion mutants of repeats 7, 10, and 12 suggest that these sites may not be necessary for AR/ß-catenin binding (95, 96), as has also been suggested to be the case for interactions between the RAR and ß-catenin (100). Structurally, this may be explained by the fact that leucine residues of the Arm repeats are buried within hydrophobic cores, possibly rendering them inaccessible to NR binding (114). ER associates with ß-catenin (107) and both Tcf4 and Tcf1 isoforms, although with varying physiological outcomes (115). Whereas Tcf4 antagonizes ER, Tcf1 promotes transactivation of ER on estrogen response elements contained in an osteopontin promoter. This exemplifies differential regulation between a NR and Tcf isoforms and suggests that the relative abundance of Tcf isoforms may dictate the physiology of consequence of Wnt stimulation (115). Genetic interactions between ER and ß-catenin have been identified in Drosophila and can promote an estrogen-dependent, hyperplastic phenotype in the eye of Drosophila. Although these data imply that ER/ß-catenin complexing may promote growth and tumorigenesis, the significance of ER/ß-catenin interactions remains to be considered in other systems (107).
Correlated with the binding capacity of coregulators is their ability to alter transcription. ß-Catenin is a potent transcriptional coactivator of AR (22, 41, 95, 96, 98, 99), RAR (100, 117), VDR (104), and ER (107). In accordance with the lack of binding observed between PRß and GR (95, 96, 97, 100), ß-catenin promotes little coactivation of these NRs (95). ß-Catenin does, however, transactivate AR on minimal transcriptional reporters (95, 96, 97, 118) as well as endogenous targets such as prostate-specific antigen (PSA) (118) at a magnitude similar to CBP (95) and steroid receptor coactivator 1 (22, 95), thereby demonstrating the potency of ß-catenin as an AR regulator. Interestingly, ß-catenin is more effective as an AR coactivator in cell lines harboring endogenous AR (119, 120), suggesting that the ability of ß-catenin to enhance AR coactivation is sensitive to the endogenous cellular milieu of coregulators.
The affinity of ß-catenin/AR interactions is likely attributable to the unique structural identity of the ARLBD but is likely also accounted for by differences in the supporting network of coregulators between cell lines. Interestingly, although cells expressing high levels of ß-catenin do not show increased VDR coactivation upon overexpression, of exogenous cells that express ß-catenin, little nuclear ß-catenin elicit a potent 1,25(OH)2D3-dependent transactivation of VDR response elements (104, 121). This effect could be attributable to the presence of high levels of nuclear ß-catenin sequestering the available transcriptional machinery. In MCF7 breast cancer cells, activated forms of ß-catenin (Ser37Ala mutation) have been shown to coactivate RAR, but not RXR, on RAR response elements or MMTV-Tet-response element luciferase reporter vectors. These data affirm that ß-catenin activates only select NRs and that these interactions are likely dependent upon structural regions unique to each NR. The ability of ß-catenin to serve as an AR activator likely also depends on variations in PI3K/Akt signaling status between cell lines. Constitutively activated PI3K signaling, found frequently in many PTEN –/– breast and PrCa cell lines, promotes decreased GSK3ß function and, consequently, high levels of ß-catenin (83, 84, 122). Therefore, AR coactivation may be less apparent in cells containing high levels of total ß-catenin with relatively little opposition by the Wnt degradative system. System differences in ß-catenin coactivation effects could also reflect the availability of AR pools to bind and import ß-catenin to the nucleus (95, 97, 99).
Genetic silencing of PTEN is a frequent genetic event in advanced PrCa and has been clearly associated with accumulation of nuclear ß-catenin (83, 84) and cyclin D1 (76). Despite the loss of PTEN and accumulation of ß-catenin, activation of ß-catenin/Tcf gene targets is low in most advanced PrCa systems suggesting that the pro-proliferative effects of ß-catenin are mediated via AR, as opposed to ß-catenin’s cognate receptor, Tcf.
Advanced PrCa is often treated by total androgen ablation therapy; however, the ultimate phenotype is one of androgen independence (AI) and death (123). Transcriptional coregulators are hypothesized to serve a critical role in promoting a more aggressive AR during androgen-independent PrCa. Altered ligand responsiveness of AR has been postulated as a major mechanism by which PrCa continues to proliferate in low androgen environments (24). As such, several lines of evidence indicate that ß-catenin may promote the oncogenicity of AR. In light of this, it is significant that ß-catenin increases AR-mediated gene activation not only in the presence of DHT, but also in the presence of the weaker adrenal androgen, androstenedione (22), a steroid remaining present in chemically castrated patients. Importantly, in addition to the AR-specific coactivators ARA70 and ARA50, ß-catenin is one of the few coactivators to enhance transcription in LNCaP cells upon treatment with 17ß-estradiol (22). ß-Catenin also coactivates mutant forms of AR that are clinically relevant, including AR-W741C and AR-T877A, mutations found in PrCa cell lines isolated from hormone refractory PrCa patients that have been treated with bicalutamide (AR W741C) (119, 124) and in lymph node metastatic lesions, respectively. These findings indicate that ß-catenin acts as a coactivator of both wild-type (WT) and mutant AR. Therefore, ß-catenin is not only altering the specificity of AR toward certain ligands but is also acting as a pure coactivator. The status of Wnt signaling in many PrCa cell lines (PC3, LNCaP) and bladder cancer (TSU) cell lines is low, likely due to relatively low levels of endogenous Tcf (98, 119, 120). Despite this, several lines of evidence indicate that ß-catenin is important for progression of PrCa. Cre-mediated excision of the ß-catenin (exon3) regulatory domain develops hyperplasia and transdifferentiation in mice at 18 wk of age but without metastatic behavior (125). In a similar model, stabilized ß-catenin appears to be important for the initiation of prostatic neoplastic lesions (126), a phenotype comparable to intestinal polyps, the precursor of invasive carcinoma and colon cancer. Also, gain-of-function, truncated forms of ß-catenin occurring in metastatic prostate and breast specimens have been shown to preferentially locate to the nucleus, possibly serving as an additional "pool" of ß-catenin to promote cell proliferation during the androgen-independent phenotype (127).
AR/ß-catenin interactions have distinct clinical relevance and therapeutic options. For example, ß-catenin recruitment to AR is readily detected in a PrCa subline containing an AR-W741C mutation (isolated from a hormone refractory PrCa specimen) (118), suggesting that ß-catenin/AR complexing may be increased in hormone refractory PrCa. Furthermore AR/ß-catenin interactions are reduced or abolished in the presence of AR antagonists, suggesting that loss of ß-catenin function may be associated with the beneficial effects of antiandrogen therapy for PrCa patients (118). ß-Catenin/ARLBD interactions have been demonstrated to be dependent upon a single AR lysine (K720) for binding (96), a site also necessary for proper ARNTD and TIF2 interactions (128). Thus, small molecule targeting of K720 may serve as an attractive target, assuming that potential disruption of AR/TIF2 interactions can be tolerated. Certainly advances in peptidometics make this an appealing option, although the obvious challenge would be delivery without any severe systemic outcomes. Given that a moderate degree of homology exists between NRs that are known to bind ß-catenin, including AR, RAR, RXR, VDR, and ER, the obvious therapeutic challenge will be the tissue-specific targeting of NR/ß-catenin interactions. However, yeast two-hybrid studies indicate that unlike AR, RAR does not bind the Armadillo repeats of ß-catenin (101), emphasizing that differences exist in the manner by which NRs bind ß-catenin. Furthermore, although AR, VDR, and RAR have been shown to efficiently bind ß-catenin, interactions between RXR and ß-catenin are likely weak or transient because binding either has been difficult to detect (105) or has not detected at all (100). Binding of NRs with ß-catenin is likely not reliant on NRDBD regions, because several NRs, including GR and PR, have failed to demonstrate an interaction with ß-catenin, despite the fact that most NRs contain nearly identical DBDs. These notions suggest the possibility of targeting complexes dependent on the ARLBD (helices 3, 4, 5) and ß-catenin Arm repeats (1, 2, 3, 4, 5, 6) without detrimental effects on interactions that are mediated through other AR domains (95). Thus, small molecule inhibitory approaches targeting NR/ß-catenin interactions are an enticing viable therapeutic approach for disruption of Wnt stimulatory effects on NRs.
C. Cyclin D1
Cyclin D1, a regulator of G0/G1 cell cycle progression, is a primary target of Wnt signaling and closely correlates with Tcf gene activation in many cell types (44, 50, 76). Cyclin D1 expression serves as a clinical predictor of poor prognosis in breast cancer (129, 130), ovarian cancer (131, 132), and thyroid cancer (133). However, most studies concur that up-regulation of cyclin D1 in prostate adenocarcinomas is a rare cell cycle event and that cyclin D1 expression does not correlate with tumor grade or progression (134, 135, 136). Cyclin D1 is also poorly correlated with ß-catenin signaling activity in PrCa cells (98). Despite this, an interesting relationship exists between AR and cyclin D1. Cyclin D1 can bind, ligand independently, to the ARNTD resulting in repressed AR activity (137, 138, 139). It also promotes both cyclin-dependent kinase-dependent mitogenesis and antimitogenic events dictated by the AF-1 domain of AR (138, 139). Mutational disruption of cyclin D1/cyclin-dependent kinase 4 interactions does not, however, interfere with the repressive effects that cyclin D1 has on AR, suggesting cell cycle-independent interactions (140). That AR expression is increased during advanced PrCa suggests that cyclin D1 may be progressively inhibited. Analysis of in vivo mouse models with stabilized expression of ß-catenin recapitulates these observations and demonstrates that stabilized ß-catenin induces little change in cyclin D1 levels but promotes significant increases in c-myc (126, 139).
Because both AR and cyclin D1 are promoters of cell proliferation, further evidence will be required to decipher the significance of the apparent negative feedback loop that cyclin D1 confers upon AR. Regardless, cyclin D1 can be considered one of the few bona fide AR-specific, corepressors, as demonstrated both in androgen-dependent and -independent environments (139). Interestingly, cyclin D1 also inhibits epidermal growth factor receptor and IGF, both of which are prosurvival factors leading to stimulation of androgen-independent growth (139). Because AR can repress ß-catenin/Tcf signaling and its target, cyclin D1 (98, 120, 141), it is possible that endogenous cyclin D1 levels never achieve steady-state levels necessary to repress AR during progressive PrCa (Fig. 2). Thus, it is tempting to speculate that cells with hyperactive Wnt signaling could incur negative selection against AR as suggested by the inverse correlation between the presence of AR and colon cancer (142). In contrast to AR and TR, cyclin D1 assumes a role as an activator of ER gene regulation (143, 144, 145) and prognostic factor for breast cancer (130, 146). Although ER/cyclin D1 binding requires an LXXLL motif in cyclin D1, this motif is not required for AR/cyclin D1 binding (144). By forming a trimeric complex with ER and steroid receptor coactivator-1, cyclin D1 can enhance estrogen transcription, estrogen dependently (143, 144). If cyclin D1/ER interactions are clinically important, then their disruption by antiestogens could prove relevant clinically (143).
|
). For example, GSK3ß binds and phosphorylates both AR (150, 151) and GR (152) at sites within the hinge and ligand binding regions resulting in decreased gene activation and cell proliferation. In accordance, overexpression of the GSK3ß inhibitor, Akt, or treatment with LiCl results in increased GR (152) and AR activity (150). This implies that endocrine pathologies, with mutational silencing of PTEN and Akt-inactivation of GSK3ß may allow for more promiscuous NR function. Short interfering RNA-mediated silencing of GSK3ß results in suppression of androgen-stimulated gene expression in PSA-secreting cells (153). This suggests that although a minimal level of GSK3ß is required for proper AR expression, higher levels, achieved in lesser PI3/Akt-sensitive cells, decrease AR activity. Interestingly, introduction of Akt-resistant forms of GSK3ß (deleted amino acids 1–9) has been reported to promote nuclear colocalization of GSK3ß and AR (150); however, the physiological significance of this remains to be determined.
The dynamics of GSK3ß/NR interactions cannot be properly interpreted without consideration of Akt, an upstream regulator of GSK3ß and key dictator in the determination of how GSK3ß influences AR function. Loss of PTEN expression and the resulting constitutive activation of PI3K/Akt signaling (154, 155, 156) results in repressed GSK3ß function as frequently observed in PrCa and breast cancer (157, 158, 159) (Fig. 2B). In these systems, GSK3ß direct regulation of NRs is likely limited; however, in systems with reduced PI3K/Akt signaling, AR likely undergoes simultaneous inactivation by Akt/protein kinase B and GSK3ß-dependent phosphorylation, with the net outcome likely dependent upon the activity of PI3K signaling and the local concentration of androgens. GSK3ß has also been demonstrated to mediate the coactivating effects that ß-catenin confers upon AR, thereby reflecting the functional integration of Wnt and PI3K signaling in PrCa cells (83, 84, 160). Both Wnt3a (65, 66) and IGF-I (161) enhance AR coactivation and production of endogenous PSA. Therefore, an attractive hypothesis is that Wnt and PI3K/Akt growth factors phosphoinactivate GSK3ß promoting stabilization, nuclear localization of ß-catenin (98, 99), enhanced ß-catenin/AR interactions (122), and enhanced proliferation (65, 66). Ultimately, however, it appears that the status of PTEN dictates these events and that, upon its loss, creates an environment that is highly amenable for NR gene activation.
|
IV. Nuclear Receptor Modulation of the Wnt/ß-Catenin/Tcf Signaling Axis |
|---|
|
TopAbstractI. IntroductionII. The Wnt/ß...III. Wnt/ß-Catenin/Tcf...IV. Nuclear Receptor Modulation...V. Summary and Therapeutic...References |
|---|
). This trend is observed not only by in vitro assays but also in vivo, thus suggesting that NRs and endocrine pathologies harboring activated Wnt/ß-catenin signaling may be subject to therapeutic manipulation. Importantly, an increasing number of agonistic and antagonistic synthetic ligands are being identified that may influence the inhibitory actions that NRs appear to confer upon Wnt signaling. The modes are diverse and, in many cases, cell line dependent; however, it is clear that NRs have the potential for potent repression of Wnt/ß-catenin/Tcf signaling (Table 2).
|
AR repression of Wnt signaling is not without clinical relevance because pathological expansion of the AR polyglutamine tract (increases of 20 to 51 glutamine repeats) results in diminished, inhibitory effects on TOPFLASH reporter activity. These observations provide a potential connection between Wnt signaling and Kennedy’s disease, a disease characterized by expanded glutamine repeats (141). They also implicate the expression of NRs with reduced colon carcinogenesis, a disease characterized by heightened ß-catenin/Tcf activity. Although surmising a causal, in vivo relationship between the presence of AR and reduced oncogenic ß-catenin/Tcf signaling is based on correlative data, it is interesting that noncancerous, colon patient samples were scored positive for total AR expression at a higher frequency than those samples obtained from patients with colon cancer (142, 164). Interestingly, these studies did not observe any significant variations in colonic AR expression with respect to either sex and age of patients (164). Considering the reports documenting the repressive effects of NRs on ß-catenin/Tcf signaling, it is tempting to conclude that loss of AR, and possibly other NRs, contributes to heightened Wnt signaling activity. Although speculative, the ability of androgens to repress Wnt activity suggests the possibility that DHT analogs, targeting colon cancers, could reduce oncogenic Wnt/ß-catenin/Tcf activity.
Retinoids are natural and synthetic derivatives of vitamin A that regulate gene activation through the RAR/RXR NR family (165). Importantly, retinoids have potent anticancer functions and have been shown to be effective in treatment of cancers of the lung (166), colon (167, 168), and prostate (169). Retinoid-activated RAR is a potent repressor of ß-catenin/Tcf signaling in retinoid-sensitive cells as exemplified in Caco-2 and HT29 colon cancer cells (100, 101, 102) and bronchial epithelium (170, 171) (Fig. 3, B and C). RA induction of E-cadherin expression, differentiation, and reduction of cyclin D1 may occur by diminishing Tcf sites of ß-catenin (102, 103). However, in SKBR3 breast cancer cells, the AP-1 pathway appears to be predominantly inhibited, rather than the Tcf axis (117), illustrating the broad manner by which retinoids exert their anticancer effects. Activation of the VDR with its metabolite ligand, 1,25(OH)2 vitamin D3, can repress Wnt/ß-catenin/Tcf signaling and in some instances can promote dramatic alteration in the integrity of the adherens junction, increase differentiation (104, 117), and decrease oncogenic cell signaling (100). Using SW480 colon cancer cells as a template, reduced ß-catenin expression and gene activation is achieved by a means independent of APC and based on a mechanism of ß-catenin reciprocity. This is supported by observations indicating that overexpression of WT Tcf-4, but not mutant ß-catenin non-binding Tcf, can repress VDR response element-mediated transcription (104). VDR-mediated repression of Wnt target genes, PPAR, Tcf-1, matrilysin, cyclin D1, and CD44 occurs before nuclear export of ß-catenin, suggesting that a transcriptional event such as depletion of Tcf sites, preempts nuclear exit of ß-catenin (104). Clearly, binding of ß-catenin to VDR, Tcf, and E-cadherin is competitive and 1,25(OH)2D3-sensitive. 1,25(OH)2 vitamin D3 induces increased VDR/ß-catenin complexing in addition to enhancing E-cadherin expression and sequestering of Tcf-bound ß-catenin (104). Although VDR/ß-catenin complexing and transrepression of ß-catenin/Tcf gene activation are achieved in the presence of low concentrations (10–11 to 10–7 M) of 1,25(OH)2D3, repression of Tcf gene activation is VDR dependent because SW480 sublines (SW480-R and SW620 cells) not containing endogenous VDR show little 1,25(OH)2D3-induced adhesion. The effects that RAR and VDR confer upon E-cadherin transcription appear to be independent of regions regulated by the cadherin regulatory molecule, Snail (104, 172), suggesting that VDR ligands could induce E-cadherin gene activation, possibly by modulation promoter methylation status.
Given the potent transcriptional and morphological effects that RA and 1,25(OH)2D3 can elicit in cells with hyperactive Wnt/ß-catenin/Tcf signaling, deciphering the mechanism by which E-cadherin is altered would appear to be of therapeutic importance. The combined use of demethylating agents with VDR ligands could be dually effective in enhancing differentiation and preventing invasive properties. 1,25(OH)2D3 analogs, including the deltanoids EB1089 and KH1060, demonstrate increased efficacy compared with 1,25(OH)2D3 in reducing ß-catenin/Tcf reporter activity and, as a result, are being used to treat neoplasia (173). Interestingly, clinical data support an inverse correlation between 1,25(OH)2 vitamin D3, dietary intake, and sunlight exposure with the incidence of colon cancer (174, 175). Thus, it is tempting to speculate that this physiology could, perhaps, be explained by a dramatic redistribution of ß-catenin from the nucleus to cell membrane, an effect exemplified in colon cancer cell lines in response to vitamin D3 (104).
Reciprocal transrepression of the Wnt/ß-catenin/Tcf axis by NRs is a mode of transcriptional repression that can be clearly applied to endocrine systems regulated by RAR, VDR, and AR as supported by the following. 1) RAR repression of ß-catenin/Tcf is concomitant with sequestering of shared cofactors with histone acetyl transferase activity. 2) NR deletion constructs, including AR 
A/B (NTD deletion) or VDR AF-2, dramatically reduce the ability of NRs to inhibit Wnt/ß-catenin/Tcf gene activation (101). Similar mutations in RAR (A/B, AF-2, and 408) do not inhibit TOPFLASH (101). That these altered regions are necessary for the coactivating effects that ß-catenin confers on NRs further implicates its role in a mechanism for NR transrepression. In addition to a reciprocal balance of ß-catenin, CBP/p300, a coactivator for ß-catenin (40) and many NRs (176, 177), has also been implicated as a major interacting protein between both Tcf and NRs (101) (Fig. 3B).
B. Glucocorticoid receptor
The GR and its ligand, dexamethasone, also modulate the integrity of E-cadherin/ß-catenin complex; however, the mechanisms by which this occurs appear not to be by direct modulation of ß-catenin/Tcf. Fascin is a negative regulator of epithelial adherens junctions in mammary epithelial tumor cells (178). Upon exposure to dexamethasone, Fascin facilitates assembly of E-cadherin and ß-catenin complexes (179, 180) resulting in enhancement of tight and adherens junctions (181, 182, 183). Thus, dexamethasone activation GR recruits ß-catenin to the cell membrane to promote differentiation of mammary epithelium. Ligand-activated GR also modulates the Wnt target, c-myc, through alteration of PI3K-dependent pools of GSK3ß. Inhibition of PI3K, using the inhibitors LY294002 and wortmannin, abrogates the ability of dexamethasone to modulate GSK3 (184), again affirming that PI3K-dependent growth factors may cross-regulate with NRs by convergence at GSK3ß. GC and Wnt/ß-catenin/Tcf interactions have also been observed with changes in bone mineralization. In osteoblast cultures, GCs activate GSK3ß by way of a reduction in phosphorylated AktSer473, thus promoting ß-catenin degradation (185, 186). If Wnts contribute to osteoblast differentiation, then GCs appear to repress this morphological change, results that are recapitulated by the fact that mutations in the Wnt binding receptor LRP5 also appear to dampen Wnt promotion of osteoblast differentiation. That dexamethasone inhibits phosphorylated Akt, activates GSK3 activation, and promotes ß-catenin degradation further underscores that PI3K/Akt and Wnt pools of GSK3ß can functionally interact (185, 186). GC modification of the adherens junctions also appears to be dependent on changes in Ras (184), whereby alterations in Rho A may lead to an altered actin cytoskeleton including connecting partners, - and ß-catenin (187). With the ability to stabilize adherens junctions, GCs may be beneficial in restoring polarization, morphology, or adhesiveness of poorly differentiated cancer cells.
C. Thyroid/retinoid X receptor
Wnt silencing is observed in response to T3 treatment. T3 can induce a striking, time-dependent inhibition of ß-catenin of GC pituitary cells that is concomitant with up-regulation of axin levels (188). T3 may also mitigate its inhibitory effects by down-regulation of IGF, a known activator of the Wnt pathway (189). Activation of TRß by T3 promotes repressed ß-catenin/Tcf and cyclin D1 gene activation in 293T cells and SW480 cells (133). Reduced levels of TRß have been detected in colon cancers compared with nonmalignant samples, suggesting a selection against T3/TR signaling in Wnt-mediated oncogenesis. Undoubtedly, future studies will consider the effects that T3 has on cell junctional morphology, growth, and tumorigenesis. RXR also inhibits Wnt/ß-catenin signaling, but appears to follow a different paradigm than RAR or that of other NRs. The activating ligand, AGN194204, promotes RXR binding of ß-catenin, resulting in degradation and reduced ß-catenin/Tcf complexing (105) by a mechanism that is independent of APC but dependent upon steric hindrance of ß-catenin (163). Thus, the mode by which RXR inhibits ß-catenin appears to be distinct from the mode by which RAR, AR, and VDR reduce activity of the ß-catenin/Tcf axis. Whereas RAR appears to decrease signaling by competition of nuclear cofactors, RXR operates by facilitating the degradation of ß-catenin (105). The reasons for differential effects of RA treatment on Tcf transcription are unclear, although cell line variation in basal ß-catenin/Tcf activity, relative levels of cadherins, and ratios of RAR/RXR or RXR/RXR dimers may be important. The effects of retinoids on Wnt signaling are not restricted to alteration of epithelial adhesion but can be extended to aspects of mammalian development (190, 191). Mammalian teratocarcinoma cells require increased Tcf gene activation during RA-dependent morphogenesis and formation of the primitive endoderm (192). Retinoid-mediated repression of several Wnt genes has also been implicated as a required step in the differentiation of neuronal (NT2) cells (193). Thus, retinoid RAs (all-trans RA or 9-cis-RA) appear to provide a two-fold anticancer effect by reducing ß-catenin/Tcf signaling and also up regulating E-cadherin levels. Stabilization of adherens junctions, in general, promotes sequestering of ß-catenin to the plasma membrane and decreased Tcf gene activation (102, 117). These mechanisms, perhaps, contribute to the effectiveness of RA acting as a chemopreventive agent in cancers with hyperactive Wnt signaling (194, 195).
Although NRs and their ligands are, in many instances, potent negative regulators of the Wnt/ß-catenin/Tcf axis, it is apparent that several mechanisms can be put forth by which NRs inhibit Wnt/ß-catenin/Tcf signaling including: 1) competition for a limited pool of ß-catenin, and/or common cofactors, (between Tcf sites and NRs); 2) degradation of ß-catenin and possibly recruitment of transcriptional repressors; and 3) a direct interaction between NRs and Tcfs, promoting chromatin remodeling. Whereas cell line-dependent differences in Tcf and NRs are likely functionally important, we favor the first scenario for VDR, AR, and RAR for the following reasons: 1) NR constructs with mutant ß-catenin binding sites do not reduce ß-catenin/Tcf signaling; 2) NR-agonist-mediated repression of ß-catenin/Tcf signaling is relieved by overexpression of Tcf4; 3) antagonized NRs do not impinge on ß-catenin/Tcf signaling; and 4) changes in NR/ß-catenin and Tcf/ß-catenin complexing are observed as a function of ligand exposure. RXR, TR, and PPAR appear to modulate Wnt/ß-catenin/Tcf signaling by a different mechanism, which is predominantly by way of GSK3-dependent phosphor-priming for proteosome-dependent degradation (Fig. 3.)
|
V. Summary and Therapeutic Implications |
|---|
|
TopAbstractI. IntroductionII. The Wnt/ß...III. Wnt/ß-Catenin/Tcf...IV. Nuclear Receptor Modulation...V. Summary and Therapeutic...References |
|---|
-helices 3, 4, and 5 of the ARLBD or first six Arm repeats of ß-catenin could prove effective. Alternatively, inhibition of extracellular growth factors, including Wnt3 and IGF-I, that are capable of ß-catenin mediated activation of AR may serve as a viable option. Those endocrine pathologies harboring activated Tcf transcription could make use of viral delivery of vectors coding death genes (e.g. fadd) under the control of Tcf response elements (196). Mounting evidence suggests that mutational silencing of PTEN may promote ß-catenin and AR gain-of-function. However, because PTEN loss appears to be a late genetic event in PrCa, small-molecule inhibition of ß-catenin may be most effective during late stage or androgen-independent cancers, as opposed to early prevention. The ability of synthetic ligands to promote dramatic enhancement of junctional proteins and, thus, cellular differentiation suggests the potential for targeted use of NR ligands for treatment of Wnt-activated cancers. Recognized NR/Wnt interactions assume developmental roles and have implications for endocrine oncogenesis. A large body of evidence implicates ß-catenin involvement in endocrine cancers; however, it also remains to be determined whether Wnt and ß-catenin/Tcf complexing are initiators of oncogenesis or act in concert with other signaling pathways to promote cell proliferation. Important animal model studies, including the Wnt-1/ER knockout mouse and prostate-specific expression of stabilized ß-catenin, do indicate that Wnt signaling can carry out functions necessary for endocrine oncogenesis. Because activating ß-catenin mutations provide gain-of-function, the development of pharmacological inhibitors could be applicable for inhibiting Wnt signaling in cells of endocrine tumors harboring such mutations. However, a therapeutic challenge is specificity because inhibitor studies targeting mutant forms of ß-catenin have also shown to nonspecifically inhibit WT forms (197). Development of potent analogs, based on the properties of AR, RAR, VDR, GR, or PPAR ligands, may serve useful in opposing endocrine cancers by reducing nuclear ß-catenin while simultaneously increasing adhesion and differentiation. However, given that a variety of NR ligands can inhibit Wnt signaling, it is tempting to speculate whether PrCa hormone withdrawal therapy could, in fact, enhance Wnt/ ß-catenin signaling and cell proliferation, thus promoting androgen-independent PrCa. This notion is consistent with data describing increased ß-catenin levels in advanced PrCa tissue cores, most of which were obtained from patients who have undergone neoadjuvant hormone therapy (198).
The function of Wnt signaling is context specific because components of other major prosurvival pathways can clearly have a major impact on normal and pathogenic Wnt signaling activities. Although PTEN expression is lost only in some endocrine cancers, its capacity to alter NR transcription, promote cellular redistribution of ß-catenin, and induce expression of E-cadherin suggests that pharmacological mimicking of its tumor suppressor qualities may be of value in retarding NR transcription and anchorage independence. Abrogation of upstream PI3K/Akt signaling can be achieved by other regulators including SH2-containing inositol phosphatase (199) and C-terminal modulator protein (CTMP) (200), regulators of phospholipids [phosphatidylinositol-3,4,5-triphosphate (PIP3)] and total Akt phosphorylation, respectively. Thus, it has become apparent recently that cross-regulation of GSK3ß pools by PI3K/Akt or Wnt stimulation can promote altered AR function. Parallel convergence of PI3K/Akt and Wnt signaling upon GSK3ß to promote ß-catenin and AR gain-of-function is of potential interest. In these instances, activation of GSK3ß may prove to have a beneficial effect reducing the growth-promoting effects of cells harboring activated Wnt and PI3K signaling. Cells that undergo simultaneous Wnt and PI3K activation would experience a tremendous selection against pools of GSK3ß. That GSK3ß is repressed in cells with loss of PTEN suggests that any increases in its expression could influence growth, cell cycle, and oncogenic effects of PI3K and Wnt signaling. The therapeutic possibility is, therefore, for GSK3ß activators or molecules capable of mimicking the tumor suppressor qualities of GSK3ß pools to degrade ß-catenin and reduce NR transcription. Nevertheless, therapeutic challenges will be to 1) identify upstream regulators of GSK3ß, and 2) evaluate their potential for tissue-specific activation.
The intersection of NRs and Wnt signaling is a developing area but also one of rapid expansion. Clearly, the means by which NRs interact with the Wnt/ß-catenin/Tcf axis are diverse and, in many cases, remain to be delineated. Although the majority of reports are generated by in vitro systems, increasing implications to in vivo systems are being put forth. The ability of NR ligands to alter Wnt/ß-catenin/Tcf signaling in a manner that promotes dramatic changes in cell morphology and physiology provides motivation to provide answers for these outstanding questions. Determination of how NR signaling intersects with the Wnt/ß-catenin/Tcf axis to regulate cell cycle events that are both normal and pathological in nature will likely evolve into an area of tremendous promise for development of targeted small-molecule inhibitors and intervention.
|
Footnotes |
|---|
First Published Online August 26, 2005
Abbreviations: AF, Activation function; AI, androgen independence; APC, adenomatous polyposis coli; AR, androgen receptor; CBP, CREB binding protein; C/EBP, CCAAT/enhancer-binding protein; CREB, cAMP-response element binding protein; DBD, DNA binding domain; DHT, dihydrotestosterone; ER, estrogen receptor; GC, glucocorticoid; GR, GC receptor; GSK3ß, glycogen synthase kinase 3 ß; LBD, ligand binding domain; LXR, liver X receptor; MMTV, mouse mammary tumor virus; NR, nuclear receptor; NTD, amino-terminal domain; PI3K, phos-phatidyl inositol 3-kinase; PIP3, phosphatidylinositol-3,4,5-triphosphate; PPAR, peroxisome proliferator-activated receptor; PR, progesterone receptor; PrCa, prostate cancer; PSA, prostate-specific antigen; PTEN, phosphatase and tensin homolog deleted on chromosome 10; RA, retinoic acid; RAR, RA receptor; RXR, retinoid X receptor; Tcf, T cell factor; TIF, transcriptional initiation factor; TR, thyroid receptor; TZD, thiazolidinediones; VDR, vitamin D receptor; WT, wild-type.
|
References |
|---|
|
TopAbstractI. IntroductionII. The Wnt/ß...III. Wnt/ß-Catenin/Tcf...IV. Nuclear Receptor Modulation...V. Summary and Therapeutic...References |
|---|
