Is the Achilles’ Heel for Prostate Cancer Therapy a Gain of Function in Androgen Receptor Signaling?

Multiple genetic changes are required to produce prostatic cancer cells that can metastasize and kill the patient (1, 2). The most fundamental consequence of these molecular abnormalities is that the rate of malignant prostate cell proliferation exceeds its rate of cell death (3). It is this disruption of cellular homeostasis that results in the continuous net growth producing the lethality of this devastating disease. Successful therapy for the 30,000 U.S. males dying of prostate cancer annually will require approaches that shift the cell kinetic balance such that the rate of prostatic cancer cell death exceeds cell proliferation without producing unacceptable host toxicity. To have a realistic chance of developing such successful therapy, identification of the molecular changes driving the imbalance in proliferation vs. death of malignant prostate cells is critical. Androgens are the major growth factors for the normal prostate, and its cognate receptor is fundamental for androgen signaling within the gland (4). Prostate cancers retain androgen receptor (AR) signaling pathways and thus are nearly universally responsive initially to androgen ablation therapy. Unfortunately, however, essentially all ablated patients eventually relapse. Because of this relapse, androgen ablation therapy is not curative, no matter how complete the ablation (5). A growing body of data has documented that this is due to the accumulation of molecular changes inducing gain of function in the AR signaling pathways during the progression of prostatic cancer. These gain of function changes result in prostate cancer cells that are resistant to androgen ablation because of their acquired ability to activate novel AR signaling pathways for their proliferation and survival without requiring physiological androgen ligand binding. These changes produce malignancy unique signaling pathways that, while androgen ligand-independent, are still dependent upon AR, and this provides a therapeutic Achilles’ heel for control of this devastating disease. The rationale for this statement is based on the following facts. First, the AR gene is located on the X chromosome, and thus males have only a single copy of this gene. Second, germline truncation mutations early in the first exon of the AR gene result in complete androgen insensitivity syndrome because no expression of AR protein occurs in these patients (6). Although such complete androgen insensitivity syndrome mutations prevent masculinization, they are not life threatening (6). This means that in prostate cancer patients with germline wild-type AR, systemic therapy that either selectively prevents AR expression or neutralizes its signaling ability should not be lethal to normal host tissues, except the male accessory sex tissues. These accessory sex tissues undergo regression by standard androgen ablation without affecting host survival. Therefore, such systemic AR-targeted therapy would have a unique AR-dependent therapeutic index because blocking AR signaling while eliminating the metastatic prostate cancer cells remaining after androgen ablation would not be life threatening. Thus, prostate cancers should provide a paradigm for successful rational drug development based on this unique therapeutic index. For such rational drug development to take advantage of the Achilles’ heel of prostate cancer, identification of the novel malignancy-acquired constitutive AR signaling pathways is critical. As a background for such research, this review will focus on what is presently understood concerning the mechanism for androgen regulation of normal cellular homeostasis in the prostate and the genetic changes responsible for gain of function in the AR-signaling pathways acquired during prostatic carcinogenesis and progression. The normal human prostate is a tubular-alveolar gland composed of a well developed stromal compartment containing nerves, fibroblasts, infiltrating lymphocytes and macrophages, endothelial cell capillaries, and smooth muscle cells surrounding glandular acini composed of a two-layered (i.e. basal and secretory luminal) epithelium (Fig. 1). Scattered throughout this epithelial compartment are occasional neuroendocrine (NE) cells that are characterized by expression of chromogranin A, serotonin, and neuron-specific enolase, but not AR (7). Functionally, the epithelium is composed of multiple stem cell units (8–14) (Fig. 2). In an individual stem cell unit, the stem cell that has the capacity for unlimited self-renewal characteristically expresses α2β1-integrins (15) but only rarely proliferates to provide progeny that differentiate to become either transit-amplifying or NE cells (10, 11). The stem and the majority of the transit-amplifying cells are believed to be located in the basal epithelial layer (Figs. 1 and 22). A subset of basal cells represent stem cells, and the remainder represent transit-amplifying cells. Cellular heterogeneity within the normal prostate. Histological architecture of the prostate is comprised of blood vessels that provide nutrients including androgens to the fibrous stromal layer, which primarily consists of fibroblasts and smooth muscle cells, and to the epithelial layer. Epithelium can be subdivided into a basal epithelium, which contains AR-negative proliferating cells, and secretory luminal epithelium, which consists of fully differentiated AR and p27 Kip1 positive, nonproliferating cells. Stem cell model of prostatic epithelial cell compartmentalization. The prostate gland consists of a number of stem cell units that arise from one stem cell. Such a stem cell is located in the basal epithelial layer of the prostate and, upon division, gives rise to a population of transit-amplifying cells. The latter divide in the basal layer, and a fraction of them differentiate and move into the secretory luminal epithelial layer. As transit-amplifying cells differentiate and move into a secretory luminal layer from the basal layer, they acquire expression of a number of genetic markers, as indicated. **, Low-level retention of expression by a subset of transit-amplifying (i.e. intermediate) cells; +, expression of marker; −, lack of detectable expression of marker. Basal cells express prostate stem cell antigen (PSCA), the p53-related p63 protein, c-Met, the plasma membrane receptor for hepatocyte growth factor (HGF), and the prosurvival protein bcl-2 (16–19). A subset of these cells shows proliferative activity as evidenced by positive staining for Ki-67, and these presumed transit-amplifying cells, which are also referred to as intermediate cells in the prostate, either do not express AR or express it at very low levels (20, 21). During the hierarchical expansion of prostatic epithelium cells, a minority of stem cell progeny differentiate into NE cells that secrete neuroendocrine peptides like bombesin, calcitonin, and PTH-related peptide (22). The majority of the stem cell progeny become transit-amplifying cells that eventually terminally differentiate into mature secretory luminal cells that are nonproliferative and positive for AR and p27Kip1cyclin-dependent kinase inhibitor (20, 21, 23) (Fig. 2). Because of this hierarchical expansion, these nonproliferating AR/p27Kip1-positive secretory luminal cells are quantitatively the major subtype of epithelial cells present in the normal prostate. These AR/p27Kip1-positive secretory luminal cells also express the prostate-specific differentiation markers, prostatic specific acid phosphatase, prostate-specific antigen (PSA), NKX 3.1, human glandular kallikrein-2 (hK2), prostate-specific membrane antigen (PSMA), and PSCA, as well as vascular endothelial growth factor (VEGF) (25–29). The transcriptional expression of these prostate-specific differentiation marker genes is enhanced by occupancy of the AR with physiological androgen [dihydrotestosterone (DHT)] and the subsequent binding of the occupied AR to androgen response elements in the promoter and enhancer sequences of these genes within the nuclei of these secretory luminal cells (27, 29–32). In contrast to the regulation of transcription of these prostate differentiation marker proteins, AR in the nuclei of these secretory luminal cells does not directly regulate the survival of these luminal cells, nor does it positively regulate the proliferation and survival of the prostatic epithelial stem and transit-amplifying cells. Instead, the survival of the secretory luminal cells and the proliferation of the transit-amplifying cells requires the androgen-dependent production of peptide growth factors by the prostatic stromal cells (33–35). These processes are initiated by testosterone diffusing from the capillary bed in the stromal compartment of the normal prostate across the basement membrane (BM) to enter the basal epithelial cells. These basal cells express type I and II/5 α-reductase proteins that enzymatically convert testosterone to 5 α-DHT (36). Once formed, DHT diffuses both into the secretory luminal cells in the epithelial compartment and back across the BM to the smooth muscle cells and fibroblasts in the stromal compartment. Secretory luminal cells also express type I/5 α reductase activity (36), thus further increasing their cellular level of DHT above that provided by the basal cells. Within these secretory luminal epithelial cell nuclei, this enhanced level of DHT binds to the AR and directly transcriptionally up-regulates the expression of the prostate-specific differentiation markers (prostatic specific acid phosphatase, PSA, hKh2, PSCA, NKX3.1, and PSMA) (16, 25, 27, 29–32) and also VEGF indirectly (28). These secretory luminal cells also express TGFβ1 (37). These growth factors diffuse across the BM to affect stromal cells. Specifically, VEGF affects the survival of the stromal endothelial cells (28), and TGFβ1 inhibits stromal cell proliferation and induces smooth muscle differentiation and neuronal trophism (38, 39). Binding of DHT to the AR within the nuclei of these stromal smooth muscle cells inhibits their expression of certain cytokines like TGFβ1 (40, 41) while enhancing their secretion of andromedins (i.e. androgen-induced stromal peptide growth factors) (42, 43). These andromedins diffuse back across the BM into the epithelial compartment where they interact with their specific cognate plasma membrane receptors of the secretory luminal cells generating intracellular signaling (e.g. down-regulation of TGβ receptors) needed to repress the apoptotic death pathway in the secretory luminal cells (44). Binding of the andromedins to the plasma membrane receptors of the transit-amplifying epithelial cells can recruit them into the cell cycle. If a sufficient systemic androgen level is not chronically maintained (e.g. after androgen ablation), then the level of DHT-occupied AR within prostatic stromal cells decreases to a level unable to maintain adequate expression of the stromally derived andromedins and unable to repress expression of TGFβ1 (40, 41). Without adequate andromedins, prostatic transit-amplifying epithelial cells remain proliferatively quiescent in Go and do not enter the cell cycle, whereas in the prostatic secretory luminal epithelial cells, lack of sufficient andromedins results in the up-regulation of expression of type I and TGFβ1 receptors The enhanced levels of TGFβ1 receptors in these secretory luminal cells are by the enhanced levels of TGFβ1 ligand by stromal cells after androgen ablation (40, 41). This enhanced TGFβ1 receptor signaling the apoptotic within the secretory luminal cells inducing their death (40, This apoptotic changes in the intracellular and and of the secretory luminal cells into apoptotic Because secretory luminal cells are the of VEGF production in the prostate, their death results in a of the prostatic VEGF levels (28). This of the VEGF levels results in the of the apoptotic death of a subset of stromal endothelial cells, blood Although secretory luminal cells undergo after androgen the basal stem and transit-amplifying cells do not A for this is that prostatic stromal cells express expression by these stromal cells is not by androgen occupancy of the AR in these stromal cells Basal stem and transit-amplifying cells express the plasma membrane cognate receptor for whereas secretory luminal cells do not Such signaling is for basal cell and proliferation Thus, after androgen the prostatic stromal cells to adequate levels of to to and signaling by the receptors of the basal cells, thus both blocking of and proliferation of these basal cells normal prostatic is to in only the transit-amplifying cells (i.e. intermediate cell to during the subsequent of these transit-amplifying cells (i.e. intermediate cell have a rate of proliferation (i.e. proliferation in in in such low do not express AR and thus are not by androgen to the These cells are however, on a of peptide growth factor andromedins in the for their survival and a rate of proliferation In contrast to the proliferation rate in in only of the epithelial cells are proliferating in normal prostatic in these cells are to physiological levels of andromedins present in (3). These the of how the in proliferation of the transit-amplifying cells to only and not net continuous prostatic epithelial growth the level of the stromally andromedins in the of physiological androgen is that AR signaling in the nuclei of the prostatic secretory luminal cells and the subset of transit-amplifying cells inhibits proliferation of these cells in the of continuous This mechanism has documented both human and prostate epithelial cells. These latter have that AR-negative intermediate prostatic epithelial cells are to express AR and then to physiological levels of their in proliferation is in the of andromedins with no upon cell survival These results that for prostatic epithelial cells, the AR as a growth its ability to as a growth such AR signaling also induces differentiation of these transit-amplifying cells from an intermediate to a secretory luminal cell This of proliferation to be to up-regulation of the kinase inhibitor The mechanism for this up-regulation in normal prostatic epithelial cells enhanced of the protein to transcriptional of expression of the in Although it is that prostate cancer from the epithelial the identification of the specific epithelial cell subtype in which the has only the focus of the for most prostatic is to be prostatic It is believed that from which in is to arise within normal prostate epithelium (Fig. The cell type of for however, is still A of carcinogenesis is that the arise in tissues in which cells acquire genetic in growth In normal human prostate epithelium, most cell in the basal cell compartment in which the stem and the transit-amplifying cells (20, 21). The majority of secretory luminal cells do not and are the terminally differentiated cells that the differentiated of the prostate, such as production and prostate cancer and cells and of secretory luminal cells (i.e. and PSA, and AR they also of the basal transit-amplifying cell compartment such as and self-renewal (Fig. 2). Thus, in these and transit-amplifying have from the basal into the secretory luminal compartment It has that the cell of for prostate cancer is an prostatic epithelial derived from the basal transit-amplifying population that the malignant molecular changes that gene expression and of both basal and secretory luminal cells model for prostatic A number of molecular and changes take as normal prostatic epithelium to prostatic Such changes are and be by and genetic factors are also an in cancer and progression. The of these intermediate initiated cells not to be within the prostate. Instead, they are in of glandular in which the luminal epithelial cells are proliferative and by within the Therefore, these have proliferative the of the following of these are to be an intermediate to early prostatic as with epithelium, is contains proliferating cells in the luminal layer, which is to of the luminal cells in have expression of the kinase inhibitor they express contains cells with of intermediate cells, which have as the cells for carcinogenesis in the contains very cells with cells in the luminal layer shows expression in the and α in of the cells, with a response to an and shows to and occurs to cancers the of these a model of prostate carcinogenesis has and in with and prostate epithelial cells for and (Fig. proliferation occurs as a response to epithelial cells. The proliferation occurs in cells with a transit-amplifying or intermediate and and and In this expression is in of the cells in as a Although in of the cells in expression is eventually in cells as the result of of the of the gene promoter such of the promoter is one of the molecular abnormalities of prostate cancer cells This these cells at for the accumulation of genetic with of the of these genetic changes by cells. This to their genetic driving further genetic producing cancers During the of prostate are changes in the AR signaling the proliferating transit-amplifying cells in the basal epithelial layer do not express the AR or express only low levels of As during their these cells eventually express levels of Once a AR level is the occupancy of AR by its ligand inhibits proliferation of these cells and induces their differentiation into secretory luminal cells. In the intermediate type of proliferating cells in expresses levels of AR, and such AR expression is further enhanced in proliferating cells in This that changes in the AR-signaling pathways at this early of cancer development because cells are proliferating and are not growth This gain of function ability the AR to the molecular signaling pathways the proliferation and survival of these initiated prostatic cells the in the normal prostate in which such growth regulation is initiated by AR binding to sequences in the nuclei of stromal cells, during prostatic AR binding within the cells this growth Because of these is a from to AR-signaling pathways in prostate cancer These gain of function changes to to and expression of survival and proliferation genes that are not by these in either normal transit-amplifying or secretory luminal cells In such gain of function AR signaling no but Such enhanced expression results in down-regulation of protein, enhancing proliferation of these cancer cells with these of these (i.e. pathways can still be dependent on the binding of androgen to its receptor in the nuclei of these cells (i.e. androgen and AR or they can be constitutive (i.e. of the binding of physiological androgens to the but still requiring AR in the nuclei of these malignant cells to the transcription of both secretory markers and also growth (i.e. androgen but still AR To the therapeutic of these an of the cellular heterogeneity and of prostate cancer cellular is ablation therapy, by or induces the of only androgen-dependent prostate cancer cells because these cells require a level of physiological androgen for their continuous proliferation and survival Unfortunately, androgen ablation is not prostate cancers are composed of of androgen-dependent cancer cells and also malignant that are androgen These latter cells are androgen because androgen occupancy of their AR is not required for their survival are of such prostate cancer cells. subtype a to androgen occupancy of its AR to its rate of cell such occupancy is not required for its survival. Thus, these cells are because their rate of growth is but not by androgen The subtype is androgen because androgen ablation decreases their rate of proliferation nor survival These of malignant are not by standard androgen and thus these are the malignant cells that eventually kill the patient It that, after androgen such androgen prostate cancer cells no express AR and that in such androgen cells, the AR no function in survival. This based on that the majority of and human (i.e. in cell from androgen ablation not express In contrast to this of prostatic cancers directly from patients androgen ablation AR In of such prostatic cancer, this is with genetic and in with AR mutations These results of AR in the of proliferation of in (i.e. prostate cancer cells resistant to androgen This is by a growing body of prostate cancer model that have documented that which with AR proliferation and of (i.e. androgen ablation prostatic cancer cells Thus, of these AR-signaling pathways should provide rational drug development for androgen prostatic To such however, a level of of the molecular for how such AR signaling is are binding proteins binding of transcriptional at the elements of The AR gene is located on the of the X (i.e. and a protein with an in and binding with transcriptional or a binding with binding and and a ligand binding which is also in and binding (Fig. This latter is also where protein (i.e. to the AR protein during its subsequent to its with results in the of This the of the proteins and thus the binding and of the occupied to elements present in the promoter and enhancer in genes (27, 29–32). of the AR and AR is by a gene located on the of the X (i.e. and, is present as a single gene copy cell. The gene is by as by in the transcriptional the is which is into a A number of are in AR binding and and a the a number of are believed to be for AR including located in the and in the ligand binding and binding located in the This AR binding further binding to specific of the AR by proteins (i.e. transcriptional proteins like and transcription factors like and to produce transcriptional that can activate or repress specific gene expression For of an transcriptional is in and that gene expression (Fig. is a of the transcriptional gene that and and and in transcription and in have that the AR-dependent of genes the mechanism for such binding of proteins to the AR (Fig. This the to its with activity such as or also to the This results in and binding of transcription factors such as and with the AR (Fig. of transcriptional by AR at the of the of the prostate-specific antigen gene. AR a transcription on the of genes like AR a in the and a number of transcription factors that recruit transcription and and transcription of the gene. the of this of normal AR the AR-signaling in prostate cancer cells can become (i.e. of genetic and to genetic gain of function AR mutations in low (i.e. in prostate cancer In the cells in and prostate cancer after androgen ablation have a level (i.e. of AR mutations molecular both human and prostate cancer has that mutations in the are in AR prostate cancer from androgen ablation whereas mutations in the ligand binding are androgen ablation mutations AR binding with specific transcriptional without the for occupancy of the ligand binding of the AR mutations have documented to physiological (e.g. and to to the of of transcriptional also can such of constitutive AR transcriptional without androgen ligand binding. For it has documented that expression of AR at low levels results in whereas levels of AR to transcriptional Because of androgen ablation prostate cancers AR this provide a mechanism for their In have that is the AR and the signaling pathway by of pathways protein kinase A certain peptide growth factors These have that AR is not signaling by or certain peptide growth factors (e.g. is unable to the transcription of prostate-specific marker genes like PSA, by prostate cancer cells. In AR is and can the expression of these marker genes in an For this latter of AR signaling by of AR with is Although such does not require of or ligand transcriptional signaling by the does require of by in the signaling initiated by binding to its plasma membrane receptor on prostate cancer cells the such binding to its receptor the and of transcription pathway This results in proteins in the Once and then is into the can AR, the to be into the where it can transcription Such to androgen ablation by prostate cancer cells both in and in These genetic and provide a for rational drug development for androgen ablation prostate cancer cells. The most to prevent the gain of function AR-dependent signaling is to prevent the expression of the AR protein this can be by a of molecular (e.g. and approaches (e.g. Although the approaches are specific for AR expression they are presently to The approaches have that like can to the enhancing the of the AR protein by Because a of proteins AR (e.g. such from the of their therapeutic index because they will AR in normal as well as cancer cells. is not to AR directly but to the ability of AR to interact and with the required for the of transcriptional Because these are malignancy-acquired gain of function these should be unique to androgen prostatic cancer cells, and thus these should not produce host toxicity. To such in cell model are because they for both for therapeutic and to for drug It has that the of the specific with AR at these prostate-specific differentiation genes (i.e. PSA, PSMA) in these model would also provide an of the AR regulate the genes for proliferation and survival of malignant prostate cells. In a number of such model cell however, is a regulation of malignant growth vs. regulation of the expression of prostate-specific markers and These results that growth and the expression of the marker genes are molecular they a for androgen AR can in prostate cancer cells for regulation of expression for the and genes Thus, the of prostate-specific differentiation marker gene regulation as model for AR of proliferation and survival of prostate cancer is to the for androgen regulation of the genes that directly affect the proliferation and survival of androgen ablation resistant prostate cancer cells. how AR genes within prostate cancer cells without the for binding its physiological ligand DHT is This should be because androgen response elements are also present in growth For have documented in the of the kinase inhibitor and the growth factor gene and are growth for prostate cancer cells resistant to androgen ablation whereas in contrast is not by normal prostate epithelial cells and is not an by normal prostate stromal cells These latter results that the AR-signaling pathways become in androgen ablation resistant prostate cancer cells. Although such a malignant these cells to become it also an Achilles’ heel for their This the that these latter genes can be as genetic model for of gain of function AR of androgen prostate cancer cells. such a rational drug development should be to convert to therapy for this devastating disease. the of and for in of the This by of androgen response basement growth hepatocyte growth human glandular ligand binding proliferative prostatic protein kinase prostate-specific prostate stem cell prostate-specific membrane and of vascular endothelial growth