Two studies find a critical role for retinoid X receptor (RXR) in the development of acute promyelocytic leukemia, highlighting RXR as a potential target for the development of anti-cancer therapies.
It is well established that the fusion of retinoic acid receptor (RAR) with another protein that promotes self association is implicated in the pathogenesis of acute promyelocytic leukaemia (APL). Treatment based on targeting RAR is effective for many patients; however, some patients do not respond or become resistant to this treatment. Two papers now show that retinoid X receptor (RXR) has an essential role in different models of APL, indicating new therapeutic avenues.
RAR fusion proteins form homodimers, and these can in turn form tetra- or oligomeric complexes. In addition, heteromeric complexes that include RXR are also formed, and targeting RXR can reduce the oncogenic effect of PML–RAR. However, the specific effects that these different complexes have on APL development remain unclear. To address this, Bernd Zeisig and colleagues focused on a rare RAR fusion protein, STAT5b–RAR (SR), that was initially cloned from a patient with APL. Gel-filtration analyses showed that deletion of the N-terminal (N) homotetramerization domain or loss of the coiled-coil (CC) homodimerization domain in the STAT5b half of SR resulted in a loss of homo-oligomeric forms. CC mutants were still able to form homotetramers, but N mutants only formed homodimers, and were unable to transform haematopoietic cells in vitro. Because the oncogenic effect of RAR fusion proteins relies on DNA binding, the authors examined whether this was compromised in the mutants. Both SR-N and surprisingly SR-CC showed reduced DNA binding, despite the fact that SR-CC can transform cells. Furthermore, homo-oligomerized RAR fusion proteins did not bind DNA efficiently, even though they also transform cells in vitro. These results indicate that intrinsic homo-oligomeric DNA binding is not crucial for transformation by RAR fusion proteins in vitro. Heterodimerization with RXR was essential for efficient DNA binding, and for interactions with chromatin modifiers that mediate RAR fusion protein suppression of RAR target gene expression. Moreover, disruption of RXR function suppresses RAR fusion protein transformation. The authors conclude that drugs that target RXR or that could reduce RAR fusion protein complexes to homodimers are potential future treatments for APL.
Jun Zhu and colleagues have also investigated the function of RXR in APL. They looked at the more common fusion between PML and RAR. RXR is known to be present in the PML–RAR complex, so these authors made a PML–RAR mutant that is unable to bind RXR, but can still form homo-oligomeric complexes. Interestingly, they found that this mutant can transform primary haematopoietic cells in vitro, but that it cannot cause APL in vivo. Instead, mice that express the mutant PML–RAR develop a myeloproliferative-like disease that has so far never progressed to APL. Both wild-type and mutant PML–RAR can bind DNA in vitro, indicating that RXR is not essential for DNA binding. However, despite the ability of the mutant to bind DNA in vitro, only wild-type PML–RAR was able to induce the suppression of RAR target genes in vivo. These authors further showed that the binding of RXR to PML–RAR results in post-translational modification of RXR, notably its sumoylation. Therefore, PML–RAR might enable the localization of active sumoylation machinery near PML–RAR binding sites, thus sumoylating RXR and chromatin-regulatory proteins, which might alter their function.
Both papers indicate that a greater understanding of the functional interactions of RAR fusion proteins, both in vitro and in vivo, is required to identify new therapeutic targets for this disease.
Nicola McCarthy
References
Zeisig, B. B. et al. Recruitment of RXR by heterotetrameric RAR fusion proteins is essential for transformation. Cancer Cell12, 36–51 (2007) | Article | PubMed |
Zhu, J. et al. RXR is an essential component of the oncogenic PML/RAR complex in vivo. Cancer Cell12, 23–35 (2007) | Article | PubMed |
(RXR
N) homotetramerization domain or loss of the coiled-coil (