Androgen receptor functions in androgen insensitivity syndrome and prostate cancer

The effect of androgens is mediated by the androgen receptor (AR), a ligand-inducible transcription factor belonging to the steroid receptor superfamily. The AR has an N-terminal domain (NTD), DNA-binding domain (DBD), and the ligand-binding domain (LBD). Also, in-between the DBD and LBD a functional linker (hinge region) can be found. When the ligand is not present the AR is inactive and resides in the cytoplasm in complex with heat shock proteins (HSP). Upon ligand binding the AR undergoes a conformational change and get translocated to the nucleus. Once in the nucleus, the AR can start a search for its response elements in the DNA. There it will recruit co-regulators, which enable to regulation of gene expression in negative or positive fashion.

In the first part of this study we functionally characterized the dimerization of the AR via the LBD. We were able to show that the AR-LBD homodimerization is ligand-inducible in vivo, and that antiandrogens are unable to induce this interaction. Furthermore, we were able to show that mutations found in androgen insensitivity syndrome impair AR functions through their effect on LBD dimerization. Our data further shows that AR-LBD dimerization is essential for transactivation. Altogether, the current findings demonstrate the LBD dimerization surface is critical for the transcriptional activity of the AR and for androgen physiology.

In the second part of the study, we shed light on the involvement of AR mutations in enzalutamide resistance. The double mutant AR (F877L/T878A) is able to convert enzalutamide to a partial agonist of the receptor. All of our data converge to the same conclusion: enzalutamide binding to the AR is enhanced by the F877L mutation, without a major impact on the activation potential of this antagonist. Only when F877L and T878A co-occur, enzalutamide becomes a strong partial agonist of the AR. Our results are corroborated by clinical data since the double mutant AR has been found in a patient who progressed on enzalutamide. We were able to show that structurally diverse antagonists of the AR can still inhibit F877L as well as the double mutant AR, which might open a possibility for use of other antiandrogens when patients progress under enzalutamide treatment due to this specific AR mutation.

The third part of this study implicated c-Myc and metabolic alterations in a preclinical model of enzalutamide resistance. We have generated enzalutamide resistant prostate cancer cell lines and described their phenotype using next generation sequencing techniques (for RNA and chromatin analysis) and various functional assays. We could not detect, mutations, overexpression or splice variants of the AR in our model. Surprisingly, the AR signaling was seen to be reduced in the resistant cells. Pathway analysis and FAIRE sequencing revealed changes in metabolism (glycolysis and hypoxia pathways) and c-Myc signaling. Functional investigation has shown that mitochondrial function and glutathione content were altered, and could be important for the aggressive phenotype of these cells. Transcription factor c-Myc was overexpressed and overactive in the resistant cell lines. Importantly, overexpression of c-Myc in the parental cell line was sufficient to reduce their sensitivity to enzalutamide. Our results are supported by the clinical finding that tumors with MYC gain have a shorter time to progression under enzalutamide treatment, but whether there is a direct link and what the mechanism behind it will be and how we can exploit this therapeutically remains to be investigated.