Prostate cancer progression is highly dependent on androgen receptor (AR) in the early stage of the disease, and therefore responds well to androgen deprivation therapy (ADT). Adaptation of tumor cells to this androgen-low environment, particularly with newest 2nd generation therapeutics, has given rise to AR-deficient castration-resistant prostate cancer (CRPC), an uniformly lethal disease. The molecular mechanisms underlying this aggressive phenotype are poorly understood but recent work from the Hsieh lab (Human Biology Division) published in Science Translational Medicine last month helped answer some of the questions.
One of the main actors in prostate cancer initiation and progression is the eukaryotic translation initiation factor 4E (eIF4E) when the tumor suppressor phosphatase and tensin homolog (PTEN) is lost. However, past studies mostly demonstrated that the aberrant cellular processes induced by AR signaling were linked to mRNA transcription. However, it is unknown if and how AR coordinates the process of mRNA translation. Dr. Hsieh and his lab were intrigued by this and sought to investigate the relationship between AR signaling and protein synthesis.
The authors used a genetically engineered mouse model where PTEN is deleted specifically in the prostate gland (PTENL/L). Prostate cancer can develop in either intact or castrated animals, with the latter modeling androgen depletion and CRPC. Tumors from castrated animals displayed higher rates of de novo protein synthesis, suggesting an increased activity of eIF4E. Interestingly, whereas eIF4E expression was unchanged in castrated versus intact animals, its binding partner and negative regulator, 4EBP1, was downregulated in the CRPC animals. As AR and 4EBP1 expression are correlated in both human and murine tissues, the researchers assessed whether AR signaling could directly regulate the transcription of 4ebp1. They found that 4ebp1 mRNA level decreased when AR was low in the PTENL/L mice, in a human cell line and in patient-derived xenografts (PDX). Furthermore, after chromatin immunoprecipitation sequencing, they identified an Androgen Responsive Element (ARE) in the first intron of 4EBP1 and confirmed that binding of AR to this region increased 4ebp1 expression. Therefore, loss of androgen signaling after castration results in decreased levels of 4EBP1. As 4EBP1 competes with other transcription initiation factors (such as eIF4G) for binding to eIF4E, the authors decided to assess the effect of decreased levels of 4EBP1 on the formation of other eIF4E protein complexes. They performed an in vivo proximity ligation assay, a technique where a signal is observed when the two targets bind each other. They observed increased formation of the eIF4E-eIF4G complex relative to the eIF4E-4EBP1 complex in castrate animals, a change that tips the balance towards boosting translation initiation. This translational burst increased cell proliferation, as assessed by Ki67 staining in the prostate gland.
The authors next wanted to characterize the pool of mRNA that was targeted by eIF4E in CRPC cells. Following ribosome profiling, they identified 697 mRNA upregulated in tumors from castrate PTENL/L mice compared to intact animals. The majority of these mRNA possessed a conserved GC-rich region in the 5’ untranslated region. Binding of the 4EBP1/eIF4E complex to this region results in translation inhibition; conversely loss of AR signaling and downregulation of 4EBP1 results in an increased translation. Gene set enrichment analysis of these upregulated genes demonstrated that AR signaling represses genes not only invoved in cell proliferation, but also in translation, transcription and signal transduction. The translational control that AR exerts on prostate cancer cells is much broader than what the researchers expected. Dr. Andrew Hsieh explains: “Historically, it has been thought that the androgen receptor mainly controls gene expression through transcription regulation. Our work shows for the first time that the AR also use translation control to shape the global proteome.”
For the researchers of the Hsieh group, this discovery presented a possible therapeutic opportunity that they could test in vivo. They used a small molecule that disrupts eIF4E-eIF4G complex (4E1RCat), a compound that previously demonstrated in vivo activity. In two different models or AR-deficient protate cancer cells, treatment with 4E1RCat both slowed tumor growth and prolonged survival, providing encouraging preclinical data. As Dr. Hsieh emphasizes: “We have found that AR-null prostate cancer is exquisitely sensitive to inhibition of translation initiation.” In humans, this could be exploited through the use of mTOR pathway inhibitors that failed in unselected populations of prostate cancer patients.
This work was supported by the AACR, Burroughs Wellcome Fund, V Foundation, the DoD and the National Institutes of Health.
Fred Hutch/UW Cancer Consortium members Drs. Hsieh, Nelson, Beronja and Plymate contributed to this research.
Liu Y, Horn JL, Banda K, Goodman AZ, Lim Y, Jana S, Arora S, Germanos AA, Wen L, Hardin WR, Yang YC, Coleman IM, Tharakan RG, Cai EY, Uo T, Pillai SPS, Corey E, Morrissey C, Chen Y, Carver BS, Plymate SR, Beronja S, Nelson PS, Hsieh AC. The androgen receptor regulates a druggable translational regulon in advanced prostate cancer. Science Translational Medicine. 11(503). doi: 10.1126/scitranslmed.aaw4993.