Search

Acute promyelocytic leukemia (APL), a subtype of acute myeloid leukemia, is the prototype of a cancer that can be cured by differentiation therapy using combined retinoic acid (RA) and chemotherapy. Acute promyelocytic leukemia is caused by chromosomal translocations, which in the large majority of cases generate the prototypic promyelocytic leukemia-retinoic-acid receptor alpha (PML-RARalpha) an oncogenic fusion protein formed from the retinoic-acid receptor alpha and the so-called PML protein. The fusion protein leads to the deregulation of wild type PML and RARalpha function, thus inducing the differentiation block and an altered survival capacity of promyelocytes of affected patients. A plethora of studies have revealed molecular details that account for the oncogenic properties of acute promyelocytic leukemia fusion proteins and the events that contribute to the therapy-induced differentiation and apoptosis of patients' blasts. Illustrating the beneficial mechanisms of action of retinoids for acute promyelocytic leukemia patients this review goes on to discuss a plethora of recently recognized molecular paradigms by which retinoids and rexinoids, alone or in combination with other compounds, regulate growth, differentiation and apoptosis also in non-acute promyelocytic leukemia cells, highlighting their potential as drugs for cancer therapy and prevention.

Multiple studies performed in in vitro and in vivo settings have confirmed the cancer therapeutic and cancer preventive capacity of retinoids and rexinoids. These compounds mediate their actions through the retinoid and rexinoid receptors, respectively, which exist in multiple isoforms and form a plethora of distinct heterodimers. Despite their apparent anticancer potential, with one exception the molecular basis of this activity has remained largely elusive. The exception concerns acute promyelocytic leukemia (APL), the prototype of retinoic acid-dependent differentiation therapy, for which both the molecular nature of the disease and the mechanism of action of retinoids are well understood. However, retinoids and rexinoids are active beyond the borderlines of the well-defined chromosomal translocation that gives rise to curable APL. In this context, particularly interesting is that retinoic acid induces a member of the tumor necrosis factor family, tumor necrosis factor-related apoptosis inducing ligand (TRAIL) or Apo2L. This ligand is exceptional in that it is capable of inducing apoptosis in cancer cells but not in normal cells. It is possible that this connection to the TRAIL signaling pathway contributes to the anti-tumor activity of retinoids and rexinoids. This review focuses on what is presently known about the regulation of cell life and death by the retinoid/rexinoid and TRAIL signaling pathways.

As a promiscuous dimerization partner the retinoid X receptor (RXR) can contribute to signaling by multiple nuclear receptors. However, the impact of RXR cosignaling and the possible existence of an RXR homodimer signaling pathway are largely unexplored. We report here on the separation of RXR homo- and heterodimerization as an essential step towards the elucidation of the roles of RXR homo- and heterodimers in retinoid-rexinoid signaling. RXR homodimerization was specifically disrupted by single mutations in the RXR dimerization interface. In contrast, even multiple mutations did not fully impair RXR heterodimerization with retinoic acid receptor (RAR). Importantly, the mutation of mouse RXRalpha (mRXRalpha) Tyr402 substantially weakened RAR heterodimerization while concomitantly increasing homodimerization. Not only did this lead to cooperatively enhanced RXR homodimer binding to DR1 or DR5 elements, but unexpectedly, the mutant acquired significant binding efficiency for noncognate DR3 or DR4 elements as well. The increased stability of RXR homodimers on DR1 correlated with increased transcriptional activity of mRXRalpha(Y402A) on DR1-based reporter genes. Weak, if any, heterodimerization was observed with thyroid, vitamin D(3), or peroxisome proliferator-activating receptors. A model accounting for the structural impact of the Tyr402 mutation on dimerization is discussed. These results provide the basis for a genetic replacement of wild-type RXRs by mutants like mRXRalpha(Y402A) to elucidate the physiological impact of RXR homo- and heterodimerization.

Nuclear receptors recruit coregulator complexes through both their AF-1 and AF-2 activation domains. Here we demonstrate that TIF2, a p160 coactivator, is able to bridge the two activation domains of the retinoic acid (RA) receptor isotype RARalpha1, resulting in synergistic activation of transcription. Bridging requires the presence of motifs in region A of RARalpha1 and in the activation domain AD1 of TIF2. Notably, only RARalpha1 exerted this interaction, which requires additional unknown factors. This is the first observation of a RAR isotype-selective coactivator interaction. Because another p160 coactivator, SRC-1, has no effect, this is also the first demonstration of a difference between the members of this coactivator family.

Large-scale gene expression profiling with DNA microarrays opens new dimensions to molecular biology but still lacks the overall precision of traditional low-scale techniques. We developed a novel strategy of data processing linking search stringency to quality indicators for efficient detection of low-level, regulated genes. Using retinoid-induced differentiation of NB-4 promyelocytic cells, the variation of expression profiles between biological duplicates was studied and compared with the changes induced by all-trans retinoic acid (atRA) treatment. An analysis of 4320 genes showed that retinoic acid has mainly geneactivating function in NB-4 cells. Treatment with atRA for 18 hours induced metabolic genes that may be associated with cell differentiation and signaling factors triggering later events leading to apoptosis; cytokine genes were among the highest stimulated by atRA. Notably, we identified a regulatory loop inhibiting MYC action: as MYC was downregulated, a cognate repressor of MYC was upregulated.

PIAS3, a member of the protein inhibitor of activated STAT family, was found to interact in vivo and in vitro with TIF2, a previously described coactivator for nuclear receptors. The interaction is mediated by two distinct non-contiguous regions of TIF2. We found that TIF2-PIAS3 interaction occurs through a unique domain of PIAS3, very rich in acidic residues and conserved throughout the PIAS family. PIAS3 modulates the ability of TIF2 to mediate ligand-enhanced transcription activation positively or negatively, for different steroid receptors. Taken together, our results indicate a potential role of PIAS3 as transcriptional modulator of TIF2-mediated signalling.

Human TIF2 (hTIF2) is a member of the p160 family of nuclear receptor coactivators, which includes SRC-1 and p/CIP. Although the functions of hTIF2 and of its mouse homolog (GRIP1 or mTIF2) have been clearly established in vitro, their physiological role remains elusive. Here, we have generated mice lacking mTIF2/GRIP1 and examined their phenotype with a particular emphasis on reproductive functions. TIF2(-/-) mice are viable, but the fertility of both sexes is impaired. Male hypofertility is due to defects in both spermiogenesis (teratozoospermia) and age-dependent testicular degeneration, and TIF2 expression appears to be essential for adhesion of Sertoli cells to germ cells. Female hypofertility is due to a placental hypoplasia that most probably reflects a requirement for maternal TIF2 in decidua stromal cells that face the developing placenta. We conclude that TIF2 plays a critical role in mouse reproductive functions, whereas previous reports have not revealed serious fertility impairment in SRC-1(-/-) or p/CIP(-/-) mutants. Thus, even though the three p160 coactivators exhibit strong sequence homology and similar activity in assays in vitro, they play distinct physiological roles in vivo, as their genetic eliminations result in distinct pathologies.

Amphioxus, the closest living invertebrate relative of the vertebrates, has a notochord, segmental axial musculature, pharyngeal gill slits and dorsal hollow nerve cord, but lacks neural crest. In amphioxus, as in vertebrates, exogenous retinoic acid (RA) posteriorizes the embryo. The mouth and gill slits never form, AmphiPax1, which is normally downregulated where gill slits form, remains upregulated and AmphiHox1 expression shifts anteriorly in the nerve cord. To dissect the role of RA signaling in patterning chordate embryos, we have cloned the single retinoic acid receptor (AmphiRAR), retinoid X receptor (AmphiRXR) and an orphan receptor (AmphiTR2/4) from amphioxus. AmphiTR2/4 inhibits AmphiRAR-AmphiRXR-mediated transactivation in the presence of RA by competing for DR5 or IR7 retinoic acid response elements (RAREs). The 5' untranslated region of AmphiTR2/4 contains an IR7 element, suggesting possible auto- and RA-regulation. The patterns of AmphiTR2/4 and AmphiRAR expression during embryogenesis are largely complementary: AmphiTR2/4 is strongly expressed in the cerebral vesicle (homologous to the diencephalon plus anterior midbrain), while AmphiRAR expression is high in the equivalent of the hindbrain and spinal cord. Similarly, while AmphiTR2/4 is expressed most strongly in the anterior and posterior thirds of the endoderm, the highest AmphiRAR expression is in the middle third. Expression of AmphiRAR is upregulated by exogenous RA and completely downregulated by the RA antagonist BMS009. Moreover, BMS009 expands the pharynx posteriorly; the first three gill slit primordia are elongated and shifted posteriorly, but do not penetrate, and additional, non-penetrating gill slit primordia are induced. Thus, in an organism without neural crest, initiation and penetration of gill slits appear to be separate events mediated by distinct levels of RA signaling in the pharyngeal endoderm. Although these compounds have little effect on levels of AmphiTR2/4 expression, RA shifts pharyngeal expression of AmphiTR2/4 anteriorly, while BMS009 extends it posteriorly. Collectively, our results suggest a model for anteroposterior patterning of the amphioxus nerve cord and pharynx, which is probably applicable to vertebrates as well, in which a low anterior level of AmphiRAR (caused, at least in part, by competitive inhibition by AmphiTR2/4) is necessary for patterning the forebrain and formation of gill slits, the posterior extent of both being set by a sharp increase in the level of AmphiRAR. Supplemental data available on-line

The recent elucidation of several molecular paradigms by which retinoids regulate growth, differentiation, and apoptosis highlights their promise as drugs for cancer therapy and prevention. Several novel signaling pathways by which retinoids induce cell death have been identified recently. They comprise (a) the induction by RARalpha-selective retinoids of the tumor-selective death ligand TRAIL that kills acute promyelocytic leukemia (APL) cells in a paracrine mode of action, which is the cause of retinoic acid-induced apoptosis after maturation: (b) a novel RARalpha-independent rexinoid-PKA cross-talk that induces maturation of both ATRA-sensitive and ATRA-resistant APL cells and does not invoke ligand-induced alteration of PML-RARalpha signaling, stability, or compartmentalization; and (c) a novel rexinoid signaling pathway that triggers apoptosis of immature APL cells and may correspond to a default death pathway that is operative in the absence of "survival" factors. This rexinoid apoptosis is inhibited by RXR but not RAR antagonists and is distinct from that triggered by RAR agonists, which control cell maturation and postmaturation apoptosis. Here we discuss the promise of retinoids for cancer treatment and prevention with an emphasis on the recently identified mechanisms by which they control (cancer) cell proliferation.

Crystal structure and co-regulator interaction studies have led to a general mechanistic view of the initial steps of nuclear receptor (NR) action. Agonist-induced transconformation of the ligand-binding domain (holo-LBD) leads to the formation of co-activator complexes, and destabilizes the co-repressor complexes bound to the ligand-free (apo) LBD. However, the molecular basis of retinoid-X receptor (RXR) 'subordination' in heterodimers, an essential mechanism to avoid signalling pathway promiscuity, has remained elusive. RXR, in contrast to its heterodimer partner, cannot autonomously induce transcription on binding of cognate agonists. Here we show that RXR can bind ligand and recruit co-activators as a heterodimer with apo-retinoic-acid receptor (apo-RAR). However, in the usual cellular environment co-repressors do not dissociate and they prohibit co-activator access because co-regulator binding is mutually exclusive. Accordingly, RXR subordination can be overcome in heterodimers that bind co-repressor weakly or in cells with a high co-activator content. We identify two types of RAR antagonists that differentially modulate co-regulator interaction, and we demonstrate that synergy between RAR ligands and RXR agonists results from increased interaction efficiency of a single p160 with the heterodimer, requiring two intact receptor-binding surfaces on the co-activator.

Retinoids have a reputation for being both detrimental and beneficial: they are teratogens, but they also have tumour-suppressive capacity. Cell biology and genetics have significantly improved our understanding of the mechanisms that underlie the anti-proliferative action of retinoids. Recent elucidation of the pathways that are activated by retinoids will help us to exploit the beneficial aspects of this powerful class of compounds for cancer therapy and prevention.

The balance between cell proliferation and programmed cell death (apoptosis) determines body patterns during animal development and controls compartment sizes, tissue architecture and remodeling. The removal of primordial structures by apoptosis allows the organism to develop sex specifically and to adapt for novel functions at later stages; apoptosis also limits the size of evolving structures. It is a ubiquitous function that is essential for all cells. Although inappropriate regulation or execution of apoptosis leads to disease, such as cancer, there is now evidence for its great therapeutic potential. This would be particularly true if apoptosis could be targeted at defined cell compartments, rather than acting ubiquitously like chemotherapy. Here, we discuss the potential of nuclear receptor ligands, many of which act through their cognate receptors in defined body compartments as modulators of cell life and death, with special emphasis on the molecular pathways by which these receptors affect cell-cycle progression, survival and apoptosis.

The therapeutic and preventive activities of retinoids in cancer are due to their ability to modulate the growth, differentiation, and survival or apoptosis of cancer cells. Here we show that in NB4 acute promyelocytic leukemia cells, retinoids selective for retinoic-acid receptor-alpha induced an autoregulatory circuitry of survival programs followed by expression of the membrane-bound tumor-selective death ligand, TRAIL (tumor necrosis factor-related apoptosis-inducing ligand, also called Apo-2L). In a paracrine mode of action, TRAIL killed NB4 as well as heterologous and retinoic-acid-resistant cells. In the leukemic blasts of freshly diagnosed acute promyelocytic leukemia patients, retinoic-acid-induced expression of TRAIL most likely caused blast apoptosis. Thus, induction of TRAIL-mediated death signaling appears to contribute to the therapeutic value of retinoids.

In the past few years our understanding of nuclear receptor (NR) action has been dramatically improved. This is due to to advancements in three fields, (i) 3D structure determination, (ii) analysis of the complexes formed between nuclear receptors and co-regulatory molecules, and (iii) the genetic analysis of nuclear receptor signalling by gene "knock out" and "knock in" technologies. The elucidation of the crystal structure of apo-, holo (agonist)- and antagonist-NR ligand-binding domain (LBD) complexes is of outstanding importance for our understanding of the structural principles, in particular of the ligand-induced allosteric alterations, that are at the basis of receptor action. The concomitant identification and functional analysis of co-regulators (TIFs, coactivators and co-repressors) previously predicted from squelching studies have provided the possibility to understand the propagation of the original signal from ligand binding through intramolecular allosteric effects to intermolecular interactions. Recent crystal data of receptor LBD heterodimers and LBD-agonist complexes with nuclear receptor interacting peptides of co-activators have provided molecular insights into receptor dimerization and receptor-coactivator interaction. Finally, analysis of the signalling compexes established over nuclear receptors, assembling enzymatic activities that can alter the acetylation status of chromatin at the promoter regions of target genes and (de)acetylate other transcription regulatory factors paves the way to a comprehension of receptor action at the chromatin level. But much remains to be learnt and the recent studies have pointed towards an enormous complexity of this signalling system. Insights into the mechanistic basis of promyelocytic leukemia and the role of retinoic acid in differentiation therapy have been obtained as a consequence of the above studies, justified the efforts and led to an increasing awareness of the nuclear receptor signalling systems in basic and applied research. Here we will review recent data with the focus on what we have learnt about the interplay between NR structure and function to provide a view of the early steps of nuclear receptor action.

Nuclear receptors are members of a large family of ligand-inducible transcription factors that regulate gene programs underlying a plethora of (patho)physiological phenomena. The recent determination of the crystal structures of nuclear receptor ligand-binding domains has provided an extremely detailed insight into the intra- and intermolecular mechanisms that constitute the initial events of receptor activation and signal transduction. Here, a comprehensive mechanistic view of agonist and antagonist action will be presented. Furthermore, the novel class of partial agonists-antagonists will be described and the multiple challenges and novel perspectives for nuclear-receptor-based drug design will be discussed.

The activation function AF2 in the ligand-binding domain of estrogen receptors ER alpha and ER beta signals through the recruitment of nuclear receptor coactivators. Recent evidence indicates that coactivators, such as the transcription intermediary factor TIF2, also bind to and transactivate the N-terminal AF1 function of the two ERs. We have generated TIF2 mutant proteins that are deficient in either AF1 or AF2 interaction and use these mutants to investigate the relative contribution of both AFs to TIF2 recruitment and transactivation. We observe that TIF2 is capable of interacting simultaneously with both the isolated N- and C-terminus of ER alpha in transfected mammalian cells and in vitro, indicating that TIF2 can bridge both receptor domains. The concomitant interaction of TIF2 with both AFs results in synergistic activation of transcription. Thus, synergy between ER alpha AF1 and AF2 is a result of the cooperative recruitment of TIF2 and/or other members of the p160 coactivator family.

Both the human retinoic acid receptor alpha (hRARalpha) and a constitutively active mutant (F318A) of the mouse retinoid X receptor alpha (mRXR alpha F318A) ligand-binding domains were separately overexpressed in Escherichia coli, copurified as a heterodimer in a two-step procedure, and cocrystallized with an RAR alpha-specific antagonist by using polyethylene glycol 10,000 as precipitant. The crystals grew in the hexagonal space group P6(1)22 displaying the unit cell parameters a = b = 116.6 A and c = 207.8 A. They diffracted X-ray to a limit of 2.2-A resolution. The asymmetric unit comprises one heterodimer and the crystal contains 60% solvent. The structure was determined by molecular replacement and is currently being refined., (Copyright 2000 Academic Press.)

The crystal structure of a heterodimer between the ligand-binding domains (LBDs) of the human RARalpha bound to a selective antagonist and the constitutively active mouse RXRalphaF318A mutant shows that, pushed by a bulky extension of the ligand, RARalpha helix H12 adopts an antagonist position. The unexpected presence of a fatty acid in the ligand-binding pocket of RXRalpha(F318A is likely to account for its apparent "constitutivity." Specific conformational changes suggest the structural basis of pure and partial antagonism. The RAR-RXR heterodimer interface is similar to that observed in most nuclear receptor (NR) homodimers. A correlative analysis of 3D structures and sequences provides a novel view on dimerization among members of the nuclear receptor superfamily.

Although retinoic acid receptor alpha (RARalpha) agonists induce the maturation of t(15;17) acute promyelocytic leukemia (APL) cells, drug treatment also selects leukemic blasts expressing PML-RARalpha fusion proteins with mutated ligand-binding domains that no longer respond to all-trans retinoic acid (ATRA). Here we report a novel RARalpha-independent signaling pathway that induces maturation of both ATRA-sensitive and ATRA-resistant APL NB4 cells, and does not invoke the ligand-induced alteration of PML-RARalpha signaling, stability or compartmentalization. This response involves a cross-talk between RXR agonists and protein kinase A signaling. Our results indicate the existence of a separate RXR-dependent maturation pathway that can be activated in the absence of known ligands for RXR heterodimerization partners.

Several dibenzodiazepine derivatives were identified as novel retinoid X receptor (RXR) antagonists on the basis of inhibitory activity on retinoid-induced cell differentiation of human promyelocytic leukemia cells HL-60 and transactivation assay using retinoic acid receptors (RARs) and RXRs in COS-1 cells. 4-(5H-2,3-(2,5-Dimethyl-2,5-hexano)-5-n- propyldibenzo[b,e][1,4]diazepin-11-yl)benzoic acid (HX603, 6c) is an N-n-propyl derivative of an RXR pan-agonist HX600 (6a), and exhibited RXR-selective antagonistic activity. Similar RXR-antagonistic activities were observed with 4-(5H-2,3-(2,5-dimethyl-2,5-hexano)-5-methyl- 8-nitrodibenzo[b,e][1,4]diazepin-11-yl)benzoic acid (HX531, 7a) and 4-(5H-10,11-dihydro-5,10-dimethyl-2,3-(2,5-dimethyl- 2,5-hexano)-dibenzo[b,e][1,4]diazepin-11-yl)benzoic acid (HX711, 8b), which also inhibited transactivation of RARs induced by an RAR agonist, Am80. These compounds inhibited HL-60 cell differentiation induced by the combination of a low concentration of the retinoid agonist Am80 with an RXR agonist (a retinoid synergist, HX600). These results indicated that HX603 (6c), and the related RXR antagonists inhibit the activation of RAR-RXR heterodimers as well as RXR homodimers, which is a distinct characteristic different from that of the known RXR antagonist, LG100754 (9).

Background: Many synthetic retinoids have been generated that exhibit a distinct pattern of agonist/antagonist activities with the three retinoic acid receptors (RARalpha, RARbeta and RARgamma). Because these retinoids are selective tools with which to dissect the pleiotropic functions of the natural pan-agonist, retinoic acid, and might constitute new therapeutic drugs, we have determined the structural basis of their receptor specificity and compared their activities in animal and yeast cells., Results: There are only three divergent amino acid residues in the ligand binding pockets (LBPs) of RARalpha, RARbeta and RARgamma. We demonstrate here that the ability of monospecific (class I) retinoid agonists and antagonists to bind to and induce or inhibit transactivation by a given isotype is directly linked to the nature of these residues. The agonist/antagonist potential of class II retinoids, which bind to all three RARs but depending on the RAR isotype have the potential to act as agonists or antagonists, was also largely determined by the three divergent LBP residues. These mutational studies were complemented by modelling, on the basis of the three-dimensional structures of the RAR ligand-binding domains, and a comparison of the retinoid agonist/antagonist activities in animal and yeast cells., Conclusions: Our results reveal the rational basis of RAR isotype selectivity, explain the existence of class I and II retinoids, and provide a structural concept of ligand-mediated antagonism. Interestingly, the agonist/antagonist characteristics of retinoids are not conserved in yeast cells, suggesting that yeast co-regulators interact with RARs in a different way than the animal cell homologues do.

This was the second in a bi-annual series of workshops held in Europe, supported by the European Molecular Biology Organization (EMBO), dealing with nuclear receptor (NR) structure and function. Exciting, novel information was presented on both the structure and functionality of NRs, whose members are increasingly recognized as families of ligand-regulable transcription factors with a major impact on a variety of (patho)physiological phenomena, ranging from the regulation of cell growth and death to cholesterol homeostasis and drug metabolism. There is no doubt that NRs are of interest to researchers who are trying to understand the molecular events involved in the corresponding signaling pathways and drug companies involved in drug design.

RAR-RXR heterodimeric complexes (RARalphaDeltaAB-RXRalphaDeltaAB) bound to cognate DR5 DNA response elements were purified to apparent structural and functional homogeneity using a novel versatile immobilized metal affinity chromatography (IMAC) copurification procedure. Dynamic light scattering studies indicated that the complexes were more than 85% monodisperse. By electron microscopy the negatively stained RAR-RXR-DNA complexes appeared homogeneous and corresponded to a dimeric arrangement of the molecules. Using heterodimers purified according to this procedure we demonstrate ligand binding of RXR in the context of the RAR-RXR heterodimer-DNA complex. The present copurification procedure is rapid and has yielded high quality heterodimer-DNA complexes suitable for both structural and biochemical studies., (Copyright 1999 Academic Press.)

This conference was the first of a series of biannual meetings covering important aspects of receptor behavior and was held as a satellite to the IUPHAR Congress in Munich. Among its aims were the definition and mechanisms of efficacy at structurally distinct receptors, receptor dynamics (receptor and agonist trafficking), agonist-induced states and models of receptor activation. Agonist influence on receptor signaling, receptor promiscuity in signal transduction, properties of agonists and the concept of inverse agonism were also considered. In addition, mechanisms of receptor desensitization and sensitization, receptor internalization, determination of agonist-binding sites, point mutation studies and transduction in cell-cell interactions were examined. Though this meeting was targeted mainly at scientists dealing with membrane receptors, Patrick Humphrey (Glaxo Institute of Applied Pharmacology, Cambridge, UK) who chaired the first session, described four receptor classes: ion channels; G protein-coupled receptors; enzyme-associated receptors; and transcriptional regulatory receptors. One of the presentations also dealt with the structure-function relationship of nuclear receptors.

Nuclear receptors belong to a superfamily of ligand-inducible transcription factors that, in addition to directly regulating their cognate gene programs, can also mutually interfere with other signaling pathways. The recent identification of selective agonists/antagonists of the glucocorticoid, retinoid and estrogen receptors suggests that it might be possible to selectively elicit only a subset of the nuclear receptor functions that are induced by the natural ligand, with the aim of increasing the functional and, perhaps, tissue selectivity of nuclear receptor ligands and reducing unwanted side effects.

In the past few years our understanding of nuclear receptor action has dramatically improved as a result of the elucidation of the crystal structures of the empty (apo) ligand-binding domains of the nuclear receptor and of complexes formed by the nuclear receptor's ligand-binding domain bound to agonists and antagonists. Furthermore, the concomitant identification and functional analysis of co-regulators (transcriptional intermediary factors [TIFs], comprising co-activators and co-repressors) previously predicted from squelching studies, have deepened this understanding. Recent data have provided the structural basis for the specific recognition of ligands and the molecular mechanisms of agonism and antagonism, enabling us to gain a comprehensive view of the early steps of nuclear receptor action.

The nuclear receptor (NR) coactivator TIF2 possesses a single NR interaction domain (NID) and two autonomous activation domains, AD1 and AD2. The TIF2 NID is composed of three NR-interacting modules each containing the NR box motif LxxLL. Mutation of boxes I, II and III abrogates TIF2-NR interaction and stimulation, in transfected cells, of the ligand-induced activation function-2 (AF-2) present in the ligand-binding domains (LBDs) of several NRs. The presence of an intact NR interaction module II in the NID is sufficient for both efficient interaction with NR holo-LBDs and stimulation of AF-2 activity. Modules I and III are poorly efficient on their own, but synergistically can promote interaction with NR holo-LBDs and AF-2 stimulation. TIF2 AD1 activity appears to be mediated through CBP, as AD1 could not be separated mutationally from the CBP interaction domain. In contrast, TIF2 AD2 activity apparently does not involve interaction with CBP. TIF2 exhibited the characteristics expected for a bona fide NR coactivator, in both mammalian and yeast cells. Moreover, in mammalian cells, a peptide encompassing the TIF2 NID inhibited the ligand-induced AF-2 activity of several NRs, indicating that NR AF-2 activity is either mediated by endogenous TIF2 or by coactivators recognizing a similar surface on NR holo-LBDs.

Unliganded bacterially expressed RXR alpha lacking the N-terminal region AB (apo-RXR alpha delta AB) was found in solution as an apparent mixture of 165 kDa tetramers and 42 kDa monomers which could be quantitatively separated by gel filtration and non-denaturing gel electrophoresis. Under identical conditions both liganded (holo-) and apo-RAR alpha delta AB were present as single monomeric species. apo-RXR alpha delta AB tetramers, as well as dimers of the apo-RXR ligand binding domain (apo-LBD), dissociated readily into monomers when exposed to their cognate ligand 9-cis retinoic acid (9c-RA). The apo-RXR alpha delta AB tetramer bound only transiently to a cognate DR1 response element, and was converted into DR1-apo-RXR alpha delta AB homodimer complexes indistinguishable from those generated by cooperative DNA binding of apo-RXR alpha delta AB monomers. In the absence of DNA, the addition of 9c-RA greatly accelerated the formation of heterodimers with the apo-RAR alpha delta AB heterodimerization partner. No RXR alpha delta AB or RAR alpha delta AB homodimers could be observed in solution, but upon mixing of the two receptor monomers stable heterodimers could be isolated which bound to DR5 response elements in a highly cooperative manner. In these heterodimers, RXR alpha delta AB interacted with its cognate ligand as efficiently as in RXR alpha delta AB homodimers. The presence of ligand did not alter the stability of RXR alpha delta AB homodimer or RXR alpha delta AB-RAR alpha delta AB heterodimer complexes on DR1 and DR5 response elements, respectively. These in vitro data support a model in which RXR tetramers could serve as an inactive pool with the dual function of: (i) rapidly supplying large amounts of RXR heterodimerization partners upon 9c-RA generation; and (ii) allowing RXR homodimer formation on "accessible" cognate response elements in the absence of 9c-RA. These events may represent a ligand-dependent regulatory mechanism controlling the availability of the promiscuous RXR dimerization partner that is engaged in multiple nuclear receptor signalling pathways.

Surface-enhanced Raman scattering and fluorescence were used to investigate the interactions of all-trans retinoic acid with the gamma-type retinoic acid receptor. Raman data revealed a significant attenuation in intensity of the bands originating from the retinoic acid polyenic chain upon receptor binding, with the spectrum being dominantly that of the beta-ionone ring. Fluorescence measurements supported the hydrophobic character of the ligand binding. These novel spectroscopic results are fully consistent with the published X-ray crystallographic data and suggest that these techniques may be valuable additional tools to characterize the interactions of agonists and antagonists with residues in the ligand-binding pockets of retinoid receptor homo- and heterodimers.

In human HL-60 promyelocytic leukemia cells, diazepinylbenzoic acid derivatives can exhibit either antagonistic or synergistic effects on the differentiation-inducing activities of natural or synthetic retinoids, the activity depending largely on the nature of the substituents on the diazepine ring. Thus, a benzolog of the retinoid antagonist LE135 (6), 4-(13H-10,11,12,13-tetrahydro-10, 10,13,13,15-pentamethyldinaphtho[2,3-b][1,2-e]diazepin-7-yl) benzoic acid (LE540, 17), exhibits a 1 order of magnitude higher antagonistic potential than the parental LE135 (6). In contrast, 4-[5H-2,3-(2,5-dimethyl-2,5-hexano)-5-methyldibenzo[b,e] [1,4]diazepin-11-yl]-benzoic acid (HX600, 7), a structural isomer of the antagonistic LE135 (6), enhanced HL-60 cell differentiation induced by RAR agonists, such as Am80 (2). This synergistic effect was further increased for a thiazepine, HX630 (29), and an azepine derivative, HX640 (30); both synergized with Am80 (2) more potently than HX600 (7). Notably, the negative and positive effects of the azepine derivatives on retinoidal actions can be related to their RAR-antagonistic and RXR-agonistic properties, respectively, in the context of the RAR-RXR heterodimer.

Mutations of a single residue in the retinoid X receptor alpha (RXRalpha) ligand-binding pocket (LBP) generate constitutive, ligand-binding-competent mutants with structural and functional characteristics similar to those of agonist-bound wild-type RXR. Modelling of the mouse RXRalphaF318A LBP suggests that, like agonist binding, the mutation disrupts a cluster of van der Waals interactions that maintains helix H11 in the apo-receptor location, thereby shifting the thermodynamic equilibrium to the holo form. Heterodimerization with some apo-receptors (retinoic acid, thyroid hormone and vitamin D3 receptors) results in 'silencing' of RXRalphaF318A constitutive activity, which, on the other hand, efficiently contributes to synergistic transactivation within NGFI-B-RXR heterodimers. RAR mutants disabled for corepressor binding and/or lacking a functional AF-2 activation domain, do not relieve RXR 'silencing'. Not only RAR agonists, but also the RAR antagonist BMS614 induce conformational changes allowing RXR to exert constitutive (RXRalphaF318A) or agonist-induced (wild-type RXR) activity in heterodimers. Interestingly, the RXRalphaF318A constitutive activity generated within heterodimers in the presence of BMS614 requires the integrity of both RXR and RAR AF-2 domains. These observations suggest that, within RXR-RAR heterodimers, RAR can adopt a structure distinct from that of the active holo-RAR, thus allowing RXR to become transcriptionally responsive to agonists.

Some of the most potent antiinflammatory and immunosuppressive agents are synthetic glucocorticoids. However, major side effects severely limit their therapeutic use. The development of improved glucocorticoid-based drugs will require the separation of beneficial from deleterious effects. One possibility toward this goal is to try to dissociate two main activities of glucocorticoids, i.e. transactivation and transrepression. Screening of a library of compounds using transactivation and AP-1 transrepression models in transiently transfected cells identified dissociated glucocorticoids, which exert strong AP-1 inhibition but little or no transactivation. Importantly, despite high ligand binding affinity, the prototypic dissociated compound, RU24858, acted as a weak agonist and did not efficiently antagonize dexamethasone-induced transcription in transfected cells. Similar results were obtained in hepatic HTC cells for the transactivation of the endogenous tyrosine amino transferase gene (TAT), which encodes one of the enzymes involved in the glucocorticoid-dependent stimulation of neoglucogenesis. To investigate whether dissociated glucocorticoids retained the antiinflammatory and immunosuppressive potential of classic glucocorticoids, several in vitro and in vivo models were used. Indeed, secretion of the proinflammatory lymphokine interleukin-1beta was severely inhibited by dissociated glucocorticoids in human monocytic THP 1 cells. Moreover, in two in vivo models, these compounds exerted an antiinflammatory and immunosuppressive activity as potent as that of the classic glucocorticoid prednisolone. These results may lead to an improvement of antiinflammatory and immunosuppressive therapies and provide a novel concept for drug discovery.

The retinoid X receptor (RXR) regulates target gene transcription via its association with cognate DNA response elements either as a homodimer or as a heterodimer with a number of other nuclear receptors. We previously demonstrated that, in solution, RXR forms tetramers with a high affinity and that ligand binding leads to dissociation of receptor tetramers to smaller species. Here it is shown that RXR tetramers form stable complexes with direct repeats (DR-1 or DR-5) or palindromic (TREpal) response elements. Binding of RXR tetramers to cognate DNA occurs with a significantly higher affinity as compared with dimers. Ligand binding by DNA-bound RXR tetramers results in their dissociation to DNA-bound dimers, a process that is completely reversed upon removal of the ligand. Formation of stable tetramer-DNA complexes requires binding of two oligonucleotides/tetramer. It is proposed that ligand-dependent modulation of the oligomeric state of RXR is a regulatory feature of this nuclear receptor.

4-[5H-2,3-(2,5-Dimethyl-2,5-hexano)-5-methyldibenzo[b,e][1,4 ]diazepin-11-yl]benzoic acid (HX600), as well as its oxa- (HX620) and thia- (HX630) analogs, enhanced the activity of retinoic acid and a receptor alpha (RAR alpha)-selective agonist Am80 in HL-60 cell differentiation assays. HX600 synergizes with Am80 by binding to, and transactivating through, the RXR subunit of the RXR-RAR heterodimer. HX600 exhibited RXR pan-agonist activity in transient transfections with a DR1-based reporter gene and synergized with RA-bound RAR alpha and RAR beta in inducing transcription from a DR5-based reporter. In addition, all three compounds at high concentrations acted as RAR pan-antagonists in stably transfected RAR "reporter cells." These efficient synergists bind only weakly with RXRs in vitro, suggesting that they are RXR-RAR heterodimer-selective activators. These HX retinoids exhibited dual functionality, since they affected signalling through both retinoid receptor families (RARs and RXRs).

The natural ligands of the progesterone (PR) and androgen (AR) receptors, progesterone and testosterone, differ only by their 17 beta-substitution. To identify within the AR and PR ligand-binding domains (LBDs) the sequences responsible for the differential recognition of these ligands, chimeric LBDs assembled from five homologous AR/PR 'cassettes' linked to the GAL4-DNA binding domain were constructed, and their ligand binding and transactivation characteristics were determined. Replacing the central cassette 3 of PR by that of AR generated a progesterone- and testosterone-responsive PR LBD with the AR residues 788-RHLS-791 being specifically involved in testosterone recognition, while the introduction of the C-terminal PR cassette 5 into AR conferred progestin responsiveness onto the AR LBD. These results suggest that residues within AR 788-RHLS-791 interact with the testosterone 17 beta-OH, while PR cassette 5 apparently contains the amino acid(s) specifically involved in the recognition of the progesterone 17 beta-acetyl group. However, ligand binding and transactivation by these chimeras were significantly decreased compared with those of the parental LBDs, indicating that residues located outside of these cassettes contribute to the proper positioning of the steroids in the AR and PR ligand-binding pockets (LBPs). Indeed, certain AR/PR chimeras acquired efficient ligand binding, but were unable to transactivate, indicating that the ligand was improperly bound in the chimeric. LBP and could not induce the conformational changes leading to a transcriptionally competent activation function (AF-2) within the LBD. The properties of the various LBD chimeras are discussed in view of the recently solved three-dimensional structures of the retinoid X receptor alpha apo- and retinoic acid receptor gamma holo-LBDs.

A 28-kDa fragment (residues 178-423) of the human retinoic acid receptor gamma, hRARgamma D3E, encompassing the ligand-binding domain (LBD) was overproduced in Escherichia coli and purified as a monomer to more than 95% purity and homogeneity. The Kd for all-trans retinoic acid binding was 0.6 +/- 0.1 nM. Crystals of the LBD complexed with all-trans retinoic acid were grown at pH 7 from sodium acetate in the presence of detergents using the vapor diffusion method. They diffract to 2.0 A using a synchrotron radiation (lambda=0.91 A) and belong to the tetragonal space group P4(1)2(1)2 with unit cell parameters a=b=60.6 A and c=155.3 A, one monomer per asymmetric unit, a solvent content of ca. 33%, and a Vm value of approximately 2 A3/dalton.

Mutation of isoleucine 747 to threonine in the C-terminal part of the ligand-binding domain (LBD) of the human glucocorticoid receptor (GR) alters the ligand specificity for transactivation. Natural glucocorticoids such as cortisol or corticosterone were completely inactive with the mutant 1747T, whereas synthetic steroids like dexamethasone efficiently stimulated GR 1747T-mediated transactivation. However, the corresponding ligand dose-response curve for dexamethasone-induced transactivation was shifted to higher concentrations when compared with that obtained with the wild type GR. Neither this shift nor the inability of cortisol to activate the 1747T mutant was due to an altered in vitro ligand-binding affinity. In the canonical three-dimensional structure of nuclear receptor LBDs, isoleucine 747 is in the direct vicinity of residues that contribute to the ligand-binding pocket. Moreover, it is located in the C-terminal LBD region, which harbors the conserved core of the activation function AF-2 and undergoes a ligand-induce transconformation, required to generate the surface interacting with putative transcriptional intermediary factors/coactivators of AF-2. The phenotype of 1747T mutant is discussed in view of the possible consequences of the mutation on the various events which, according to the model, lead to a transcriptionally competent AF-2.

Vertebrate Hox genes are required for the proper organization of structures along the rostrocaudal axis. Hoxd-11 is expressed in the posterior part of the embryo, up to the level of prevertebra 27, and its expression boundary is reproduced by a Hoxd-11/lacZ transgene. Expression of this transgene anterior to prevertebra 27 is prevented by the silencing activity of a cis-acting element, region IX. Using transgenic mice, we show that Hoxd-11 repression by region IX is necessary to position the sacrum properly. This silencing activity depends on phylogenetically conserved sequences able to bind in vitro retinoic acid receptors and COUP-TFs. ES cells were used to generate mice carrying a subtle mutation that abolishes binding of nuclear receptors to region IX. Mutant mice display an anterior shift of their lumbosacral transition inherited as a codominant trait. In mutant embryos, expression of both Hoxd-11 and Hoxd-10 mRNAs in the prevertebral column is anteriorized. These results illustrate the sharing, in cis, of a single regulatory element in order to establish the expression boundaries of two neighboring Hoxd genes.

Signalling by all-trans retinoic acid is mediated through RXR-RAR retinoid receptor heterodimers, in which RXR has been considered to act as a transcriptionally silent partner. However, we show here that in cultured NB4 (ref. 6) human acute promyelocytic leukaemia cells treated with either an RAR-alpha-selective agonist alone, or certain RAR-alpha antagonists in combination with an RXR agonist, receptor-DNA binding is induced in vivo, resulting in expression of the target genes of retinoic acid as well as acute promyelocytic leukaemia protein (PML) relocation to nuclear bodies and differentiation before apoptosis. These results indicate that RAR-alpha ligands can induce two separate events: one enables RXR-RAR-alpha heterodimers to bind to DNA in vivo and allows RXR agonists to act; the other induces transcriptional activity of RAR-alpha. The availability of receptor-specific synthetic retinoids that can induce distinct receptor functions has potential in extending the therapeutic repertoire of retinoids.

Nuclear receptors (NRs) act as ligand-inducible transcription factors which regulate the expression of target genes upon binding to cognate response elements. The ligand-dependent activity of the NR activation function AF-2 is believed to be mediated to the transcription machinery through transcriptional mediators/intermediary factors (TIFs). We report here the cloning of the 160 kDa human nuclear protein TIF2, which exhibits all properties expected for a mediator of AF-2: (i) it interacts in vivo with NRs in an agonist-dependent manner; (ii) it binds directly to the ligand-binding domains (LBDs) of NRs in an agonist- and AF-2-integrity-dependent manner in vitro; (iii) it harbours an autonomous transcriptional activation function; (iv) it relieves nuclear receptor autosquelching; and (v) it enhances the activity of some nuclear receptor AF-2s when overexpressed in mammalian cells. TIF2 exhibits partial sequence homology with the recently isolated steroid receptor coactivator SRC-1, indicating the existence of a novel gene family of nuclear receptor transcriptional mediators.

The activity of the ligand-inducible activation function 2 (AF-2) contained in the ligand binding domain (LBD) of nuclear receptors (NRs) is thought to be mediated by transcriptional intermediary factors (TIFs). We have recently reported the isolation and characterization of two novel mouse proteins, designated TIF1 and mSUG1, that interact in a ligand-dependent fashion with the LBD (region E) of several NRs in vivo as well as in vitro. Remarkably, these interactions require the conserved core motif of the AF-2 activating domain (AF-2 AD) and can be blocked by AF-2 antagonists. TIF1 and mSUG1 might therefore represent TIFs/mediators for the ligand-dependent AF-2 of NRs. By comparing the interaction properties of these two putative TIFs with different NRs including the oestrogen (ER), thyroid hormone (TR), vitamin D3 (VDR), retinoic acid (RAR alpha) and retinoid X (RXR) receptors, we demonstrate that: (i) RXR alpha efficiently interacts with TIF1, but not with mSUG1, whereas TR alpha interacts much more efficiently with mSUG1 than with TIF1, and RAR alpha, VDR and ER efficiently interact with both TIF1 and mSUG1; (ii) the amphipathic alpha helix core of AF-2 AD is differentially involved in the interactions of RAR alpha with TIF1 and mSUG1; and (iii) the AF-2 AD cores of RAR alpha and ER are similarly involved in their interaction with TIF1, but not with mSUG1. Thus the interaction interfaces between the various NRs and either TIF1 or mSUG1 may vary depending on the nature of both the receptor and the putative mediator of its AF-2 function. We discuss the possible roles of TIF1 and mSUG1 as mediators of the transcriptional activity of the AF-2 of NRs.

Using a yeast two-hybrid system we report the isolation of a novel mouse protein, mSUG1, that interacts with retinoic acid receptor alpha (RAR alpha) both in yeast cells and in vitro in a ligand- and AF-2 activating domain (AF-2 AD)-dependent manner and show that it is a structural and functional homologue of the essential yeast protein SUG1. mSUG1 also efficiently interacts with other nuclear receptors, including oestrogen (ER), thyroid hormone (TR), Vitamin D3 (VDR) and retinoid X (RXR) receptors. By comparing the interaction properties of these receptors with mSUG1 and TIF1, we demonstrate that: (i) RXR alpha efficiently interacts with TIF1, but not with mSUG1, whereas TR alpha interacts much more efficiently with mSUG1 than with TIF1, and RAR alpha, VDR and ER efficiently interact with mSUG1 and TIF1; (ii) the amphipathic alpha-helix core of the AF-2 AD is differentially involved in interactions of RAR alpha with mSUG1 and TIF1; (iii) the AF-2 AD cores of RAR alpha and ER are similarly involved in their interaction with TIF1, but not with mSUG1. Thus, the interaction interfaces between the different receptors and either mSUG1 or TIF1 may vary depending on the nature of the receptor and the putative mediator of its AF-2 function. We discuss the possibility that mSUG1 and TIF1 may mediate the transcriptional activity of the AF-2 of nuclear receptors through different mechanisms.

The 2.0-A crystal structure of the ligand-binding domain (LBD) of the human retinoic acid receptor (RAR)-gamma bound to all-trans retinoic acid reveals the ligand-binding interactions and suggests an electrostatic guidance mechanism. The overall fold is similar to that of the human RXR-alpha apo-LBD, except for the carboxy-terminal part which folds back towards the LBD core, contributing to the hydrophobic ligand pocket and 'sealing' its entry site. We propose a 'mouse trap' mechanism whereby a ligand-induced conformational transition repositions the amphipathic alpha-helix of the AF-2 activating domain and forms a transcriptionally active receptor.

We describe here a novel class of cis-acting response elements for retinoid, vitamin D, and estrogen receptors which are widely spaced (10 to 200 bp) direct repeats (DRs) of the canonical 5'-AGGTCA half-site recognition motif (DR10 to DR200). In contrast to the specificity previously observed with shortly spaced DRs (DR1 to DR5), the different receptors bind promiscuously to these novel elements to activate transcription in the presence of retinoic acid (RA), vitamin D, or estrogen. The greatest RA-dependent transactivation, seen with DR15, was similar to that observed with the canonical DR5. Both RA receptors and retinoid X receptors contribute to transactivation through widely spaced DR elements. With the estrogen receptor, DR15 was one-third as efficient as the classical palindromic response element. A further increase of spacer lengths progressively decreased the efficiency of transactivation. No transactivation was seen with widely spaced DRs when the thyroid and retinoid X receptors were coexpressed in the presence of their ligands. The progesterone receptor was also unable to transactivate through a DR10 element composed of its cognate binding motifs. These results considerably extend the response element repertoire of nuclear receptors and suggest the existence of promiscuous transcriptional regulation through common response elements, as well as the possibility of receptor "cross-talk."

Many of the effects of retinoids on cells are mediated by the transcription factors known as retinoid nuclear receptors, but the mechanisms by which retinoids regulate the activity of the receptors are not known. It was previously shown that the retinoid X receptor (RXR) forms tetramers with a high affinity. In the present work it is demonstrated that binding of 9-cis-retinoic acid to RXR leads to rapid dissociation of receptor tetramers. In addition, fluorescence anisotropy studies indicate that ligand-binding results in a significant conformational change such that holo-RXR is more compactly folded as compared to the apo-protein. These findings suggest that the initial event in signaling by 9-cis-retinoic acid is a change in the oligomeric state of RXR. The data also imply that tetramer formation is a regulatory feature of the pathway by which RXR mediates the effects of retinoids on gene transcription.

The ligand binding domain (LBD) of the human retinoid X receptor alpha (hRXR alpha) was overproduced in Escherichia coli and purified to more than 95% purity and functional homogeneity. Circular dichroism spectra of the purified RXR alpha LBD indicated that the protein was composed predominantly of alpha-helical structures and coils. Crystals were grown from ammonium citrate using the vapor diffusion method against a reservoir containing 100 mM Pipes (pH 7.0) and 1.5 M ammonium citrate. They belong to the hexagonal space group P6(3)22 with unit cell parameters a = b = 110.8 A and c = 109.9 A, alpha = beta = 90 degrees, gamma = 120 degrees, and they diffract X rays to a resolution limit of 2.5 A using synchrotron radiation. The asymmetric unit of the crystals contains one molecule with a solvent content of approximately 55% and a Vm value of 3.6 A3/dalton.

Protein-protein interactions allow the retinoid X receptor (RXR) to bind to cognate DNA as a homo- or a heterodimer and to participate in mediating the effects of a variety of hormones on gene transcription. Here we report a systematic study of the oligomeric state of RXR in the absence of a DNA template. We have used electrophoresis under nondenaturing conditions and chemical crosslinking to show that in solution, RXR alpha forms homodimers as well as homotetramers. The dissociation constants governing dimer and tetramer formation were estimated by fluorescence anisotropy studies. The results indicate that RXR tetramers are formed with a high affinity and that at protein concentrations higher than about 70 nM, tetramers will constitute the predominant species. Tetramer formation may provide an additional level of the regulation of gene transcription mediated by RXRs.

The crystal structure of the human retinoid-X receptor RXR-alpha ligand-binding domain reveals a previously undiscovered fold of an antiparallel alpha-helical sandwich, packed as dimeric units. Two helices and one loop form the homodimerization surface, and hydrophobic heptad repeats participate in stabilizing the fold. The existence of a ligand-binding pocket is proposed that would allow 9-cis retinoic acid to interact with different functional modules, including the AF-2 activating domain. Several lines of evidence indicate that the overall structure is a prototype fold of ligand-binding domains of nuclear receptors.