Modulating nuclear receptor function: may the phos be with you.

Abstract

Gene expression is tightly regulated in response to both extracellular and intracellular signals, which control virtually every biological process. Intracellular signaling molecules such as steroid hormone, thyroid hormone, retinoids, and vitamin D exert their function through direct binding to their cognate nuclear receptors (1). Nuclear receptors are transcription factors that recognize specific sequences in target genes via a centrally located DNA-binding domain (DBD). Some receptors bind DNA as monomers, some as homodimers, and some as heterodimers with a common partner, the retinoid X receptor (RXR). Both the DBD and the COOH-terminal ligand-binding domain (LBD) contribute to dimerization. Many nuclear receptors repress transcription in the absence of ligand, due to a repression function in the LBD. Interaction of the LBD with ligand abolishes repression and activates transcription via a COOH-terminal activation domain (activation function 2, or AF2), which in some receptors works in tandem with an additional activation domain (AF1) in the NH2-terminus, or A/B region. There are over 150 mammalian members of the nuclear receptor superfamily. Many of these are orphan receptors, whose ligands are yet to be discovered. In contrast to the small lipophilic molecules that signal via nuclear receptors, extracellular signaling molecules such as peptide hormones and cytokines communicate with their intracellular targets through surface receptors, which activate signal transduction pathways that ultimately lead to regulation of gene expression mediated by transcription factors such as fos, jun, cAMP-response element binding protein (CREB), and others. Most often the mechanism involves direct phosphorylation of the transcription factor by a kinase that is activated as a result of the ligand-receptor interaction at the cell surface. Nuclear receptors are also targets of kinases involved in signal transduction. In this issue of the JCI, Kremer and colleagues report that activation of mitogen-activated protein kinase (MAPK) attenuates the ligand-dependent transactivation function of the vitamin D receptor (VDR) and its heterodimer partner RXR (2). This effect appears to be mediated by phosphorylation of human RXRα on a serine residue at amino acid 260 in the protein. This report not only provides a potential mechanism for ras-induced transformation of human keratinocytes, but also has broader implications because RXR heterodimerizes with several other nuclear receptors including thyroid hormone receptor, retinoic acid receptor, and peroxisome proliferator–activated receptor. Thus, it will be interesting to determine in future studies whether phosphorylation of RXR alters the biological activity of these receptors. Phosphorylation of nuclear receptors provides an important mechanism for crosstalk between signaling pathways. Phosphorylation has been shown to modulate the activity of many nuclear receptors (Table ​(Table1).1). Multiple kinase pathways have been implicated, including cAMP-dependent protein kinase, casein kinase, glycogen synthase kinase (GSK), jun kinase, cyclin-dependent kinases (Cdks), and MAPKs. All aspects of receptor function can be regulated, including DNA binding and dimerization, transcriptional activity, interaction with cofactors, and ligand-binding affinity. Table 1 Phosphorylation of nuclear receptors The mechanism by which phosphorylation of RXR inhibits vitamin D signaling is not clear. Phosphorylated RXR was not recognized by an antibody that recognized the nonphosphorylated protein, suggesting that phosphorylation resulted in a conformational change in the protein. Such a change in tertiary structure might alter affinity of the VDR/RXR heterodimer for transcriptional coregulators. This is an attractive hypothesis because ligand activation of nuclear receptors is clearly related to a conformational change that favors coactivator interaction (3). An alternative possibility is that the phosphate group has a more direct role in a critical function such as coregulator interaction, as has been observed for the interaction between phosphorylated CREB and its main coactivator CBP (4). Although the structures of the DBD and LBD have been solved for numerous receptors, no full-length receptor has been crystallized, nor has the structure of any phosphorylated nuclear receptor been determined. Thus, it remains to be determined which effects of phosphorylation listed in Table ​Table11 are due to conformational changes and which are due to more localized effects of the phosphate group on receptor function. Moreover, the combination of phosphorylation and ligand binding may in some cases lead to an entirely novel structure. The concept that nuclear hormone receptor function is determined by rapid, reversible, and combinatorial structural changes governing interactions with other cellular proteins provides a framework for understanding how the receptor can integrate a variety of signals in a physiological context. This level of understanding puts us light years ahead of where we were just a short time ago.