"Getting from here to there"--mechanisms and limitations to the activation of the androgen receptor in castration-resistant prostate cancer

Abstract

Despite the clinical regression that typifies the initial response of advanced prostate cancer to gonadal testosterone depletion, tumors eventually progress. However, evidence supports the concept that signaling via the androgen receptor (AR) is important in progression to castration-resistant prostate cancer (CRPC).Steroid hormones are synthesized from cholesterol in a series of tightly regulated steps involving the cleavage of carbon-carbon bonds, the introduction of functional groups derived from activated molecular oxygen, and the oxidation and reduction of carbon-carbon and carbon-oxygen bonds. In the adrenal cortex and gonads, steroidogenesis is tightly regulated, very efficient, and highly directional. In contrast, steroid metabolism in peripheral tissues is characterized by competing enzymes and pathways, low efficiency, and great variability. Many steps are mechanistically and functionally irreversible, but some are not, and the repertoire of specific enzymes, intracellular redox state, and access to hormone precursors all contribute to steroid flux and accumulation.The investigation of steroid metabolizing enzymes in CRPC often assumes that the pathways and the patterns of metabolism mirror those defined in the adrenals and the gonads and validated by human deficiency syndromes. Unfortunately, several potential pathways using different enzymes might contribute substantially to androgen synthesis in CRPC. Finally, a number of mechanisms have been reported by which the AR is activated independent of ligand. Recent observations have suggested that AR forms with constitutive activity occur in CRPC, stimulating transcription without a requirement for ligand. This overview outlines a broad view of how the mechanisms by which the AR may be activated, whether by alternate pathways of androgen synthesis or the production of alternate forms of the AR, with an emphasis on what aspects must be accounted for when using model systems to explore the biology of human prostate cancer.

Figure 1

Steroidogenic pathways to testosterone and dihydrotestosterone. Adrenal DHEA may be converted to testosterone and dihydrotestosterone via conversion through androstenedione or androstenediol as the initial step. Alternative, de novo steroidogenesis from cholesterol through the “backdoor pathway” may occur through progesterone → 17α-hydroxyprogesterone → 5α-pregnan-17α-ol-3,20-dione → 5α-pregnane-3α,17α-diol-20-one → androsterone → 5α-androstane-3α,17β-diol → dihydrotestosterone.

Figure 2

Normal and abnormal adrenal steroidogenesis in the zona fasciculata. The normal pathway is shown on the left. Loss of function mutations in CYP21A2 (steroid 21-hydroxylase) lead to congenital adrenal hyperplasia (CAH; right), resulting in a massive increase in 17-hydroxyprogesterone. This 17-hydroxyprogesterone and upstream precursors are metabolized to androgens, probably via several distinct pathways.

Figure 3

Steroid flux from DHEA versus progesterone. LNCaP cells were treated with [³H]-DHEA (100 nM), [³H]-progesterone (100 nM), and [³H]-progesterone (100 nM) pretreated with 4-MA (10 μM), for 48 hours. Steroids were extracted from the medium, treated with glucuronidase and analyzed by thin-layer chromatography (not shown) or high-performance liquid chromatography (HPLC). DHEA is readily metabolized to androstenedione (AD), testosterone (T), dihydrotestosterone (DHT) and androsterone (AST). In contrast, for cells treated with progesterone (Prog), there was no detectable 17OH-progesterone (17OHP), T, DHT, or AST. No 17OHP is detected with pretreatment with the SRD5A inhibitor 4-MA, and little Prog metabolism occurs. These results suggest that the major metabolic pathway of Prog in LNCaP cells begins with 5α-reduction, rather than 17-hydroxylation.