RXR: no longer a stepchild

The nuclear receptor superfamily contains members with known ligands, orphans, and RXR playing a subservient role as heterodimer partner. Taschner and colleagues demonstrate that RXRα may be more than a stepchild.

RARα and its heterodimer partner retinoid X receptor (RXR) are members of the nuclear receptor superfamily.¹  Nuclear receptor superfamily members have been documented to form heterodimers with RXR. These heterodimers act as transcription factors, binding to DNA response elements usually located in the 5′ untranslated regions of genes. Presence or absence of the ligand for the RXR heterodimer partner controls recruitment of coactivator or corepressor proteins to regulate gene transcription. However, 2 fundamental questions remain largely unresolved: (1) does RXR and its ligand 9-cis retinoic acid (9cRA) contribute to transcriptional regulation mediated by RXR heterodimers; and (2) do RXR/RXR homodimers play important roles in regulating gene transcription independently of RXR heterodimers?

In this issue of Blood, Taschner and colleagues present strong evidence that RXRα plays an instrumental role in controlling myeloid developmental cell fate. In a series of elegant experiments, the investigators demonstrated that down-regulation of RXRα protein is required for neutrophil development, while maintenance of RXRα supports development of monocytes. Taschner and colleagues first demonstrated this using primary CD34⁺ progenitor cell cultures containing a mixture of growth factors plus granulocyte colony-stimulating factor (G-CSF) or macrophage colony-stimulating factor (M-CSF) to drive neutrophil or monocyte development, respectively. They further verified that neutrophils and monocytes isolated from peripheral blood also showed this pattern of RXRα protein expression.

Next, Taschner and colleagues used expression vectors containing either functional full-length or truncated dominant-negative RXRαΔ in combination with RXRα agonist to determine whether the RXRα protein itself could alter myeloid development. In addition, they also tested the effects of either vitamin A (as RAR ligand all-trans retinoic acid [ATRA] or 9cRA) or vitamin D (as 1,25-dihydroxyvitamin-D3 [VD3]) in the presence and absence of these expression vectors, since vitamin A is known to promote neutrophil development, while vitamin D promotes monocyte development. Addition of RXRα agonist to cells transfected with either an empty vector or full-length RXRα switched cell fate choice from neutrophils to monocytes despite culture in neutrophil-promoting medium containing G-CSF. Further, addition of VD3, but not ATRA or 9cRA, promoted monocyte development in cultures containing G-CSF, and this effect was significantly inhibited by dominant-negative RXRα. In all cases tested, dominant-negative RXRα inhibited monocyte development and promoted neutrophil development. Further data presented by the authors suggest that RXRα inhibits neutrophil development at least partially by interfering with neutrophil proliferation.

Although Taschner and colleagues provide convincing evidence of a prominent role for RXRα in the later stages of myeloid development, their studies do not provide convincing evidence as to whether RXRα is functioning as a homodimer or heterodimer in this system. Recently, Vivat-Hannah and colleagues²  have described specific mutations of RXR that strongly influence the ability of this receptor to form heterodimers versus homodimers. Future studies using these mutated RXR constructs may be able to provide insight into this as yet unresolved question.