Functions of Normal and Malignant Prostatic Stem/Progenitor Cells in Tissue Regeneration and Cancer Progression and Novel Targeting Therapies

Abstract

This review summarizes the recent advancements that have improved our understanding of the functions of prostatic stem/progenitor cells in maintaining homeostasis of the prostate gland. We also describe the oncogenic events that may contribute to their malignant transformation into prostatic cancer stem/progenitor cells during cancer initiation and progression to metastatic disease stages. The molecular mechanisms that may contribute to the intrinsic or the acquisition of a resistant phenotype by the prostatic cancer stem/progenitor cells and their differentiated progenies with a luminal phenotype to the current therapies and disease relapse are also reviewed. The emphasis is on the critical functions of distinct tumorigenic signaling cascades induced through the epidermal growth factor system, hedgehog, Wnt/β-catenin, and/or stromal cell-derived factor-1/CXC chemokine receptor-4 pathways as well as the deregulated apoptotic signaling elements and ATP-binding cassette multidrug transporter. Of particular therapeutic interest, we also discuss the potential beneficial effects associated with the targeting of these signaling elements to overcome the resistance to current treatments and prostate cancer recurrence. The combined targeted strategies toward distinct oncogenic signaling cascades in prostatic cancer stem/progenitor cells and their progenies as well as their local microenvironment, which could improve the efficacy of current clinical chemotherapeutic treatments against incurable, androgen-independent, and metastatic prostate cancers, are also described.

• I. Introduction

• II. Functions of Normal and Malignant Adult Prostatic Stem/Progenitor Cells

• III. Functions of Normal Adult Prostatic Stem/Progenitor Cells in the Homeostatic Maintenance of the Prostate Gland

  • A. In vivo and ex vitro characterization of adult prostatic stem/progenitor cell properties
  • B. Paracrine and autocrine regulation of prostatic stem/progenitor cell proliferation and differentiation by the interplay of androgens, growth factors, and integrins
• IV. Functions of Prostatic Cancer Stem/Progenitor Cells in Prostate Cancer Initiation and Progression

  • A. Heterogeneity of prostate cancers derived from distinct prostatic cancer stem/progenitor cells
  • B. In vitro and in vivo prostatic cancer stem/progenitor cell models

• V. Molecular Mechanisms Involved in the Therapeutic Resistance of Prostatic Cancer Stem/Progenitor Cells and Their Further Differentiated Progenies

• VI. Novel Therapeutic Strategies against the Locally Advanced and Metastatic Hormone-Refractory Prostate Cancers

  • A. Targeting of PC stem/progenitor cells and their further differentiated progenies
  • B. Targeting of the local tumor microenvironment of PC stem/progenitor cells and their further differentiated progenies

• VII. Conclusions and Perspectives

I. Introduction

THE DEVELOPMENT OF earlier diagnostic tests in the past few years has led to a more effective therapeutic intervention for patients diagnosed with prostate cancer (PC) (1,2,3,4,5,6). Among the current clinical treatments, radical prostatectomy, hormonal therapies, radiotherapy, and/or adjuvant chemotherapy generally show beneficial effects and a significant curative rate for the patients diagnosed with localized PCs in the early stages (3,5,6,7). Unfortunately, the accumulation of genetic and/or epigenic alterations leading to an enhanced expression of numerous growth factor and cytokine signaling cascades during PC progression may confer an androgen-independent (AI) phenotype to PC cells and thereby contribute to the resistance to androgen deprivation therapies and disease relapse (3,8,9,10,11,12,13). For patients at high risk of progression to more advanced PCs or those diagnosed in the late stages with metastatic and hormone-refractory PCs (HRPCs), the nonhormonal systemic chemotherapeutic regimens may thus represent another clinical therapeutic option (3,7,9,11,12,13,14,15). The current chemotherapeutic treatments against the metastatic HRPCs include docetaxel plus prednisone and/or estramustine phosphate, or mitoxantrone plus prednisone (3,11,12,13,14,15,16,17,18,19,20,21). The choice of using the docetaxel- or mitoxantrone-based therapeutic regimens in first- or second-line treatments against metastatic HRPCs is based on the observations indicating that these chemotherapeutic treatments may improve the quality of life of the patients. However, these chemotherapies, which principally offer palliative care of metastatic bone pain, generally show no significant, or in case of docetaxel-based regimens, modest beneficial effects for improving the overall survival rate and outcome of patients in the clinics, and they ultimately result in the death of patients (16,17,18,21,22). At present, few therapeutic options exist to treat patients with metastatic HRPCs after a lack of response to docetaxel- or mitoxantrone-based therapies. Therefore, the identification of novel drug targets and cytotoxic agents for improving the efficacy of current chemotherapeutic treatments against the locally advanced and metastatic HRPCs is essential to counteract PC progression and thereby to prevent the formation of metastases at distant sites, including bone marrow, and disease relapse.

Numerous recent lines of evidence indicated that the persistence of androgen receptor negative (AR⁻) PC stem/progenitor cells, also designated as PC-initiating cells, in primary and secondary neoplasms, which might be resistant to current antihormonal therapy, radiotherapy, and/or chemotherapeutic treatments, may contribute to PC recurrence (10,23,24,25,26,27,28,29,30,31). More specifically, a very small subpopulation of CD133⁺/CD44⁺/α₂β₁^high-integrin/AR⁻ PC stem/progenitor cells, comprising about 0.1–3.0% of total PC cells, has been identified and isolated from primary and metastatic PCs (23,31,32). These isolated PC stem/progenitor cells gave rise to differentiated tumor cells in vitro and in vivo possessing a secretory luminal phenotype like the original tumor cells, such as the expression of AR (23,31). Moreover, an intermediate PC cell subpopulation expressing the basal and luminal cell-specific markers, such as cytokeratins, CK5 and CK18, respectively, and which represent only a minor fraction in the total PC cell population, has been identified in primary and metastatic PCs (33,34,35,36). Importantly, the sustained activation of numerous developmental signaling cascades in PC cells, including PC stem/progenitor cells, may cooperate in inducing a more complete epithelial-mesenchymal transition (EMT) program and thereby provide to them a more malignant behavior (Figs. 1 and 2). Among them, there are epidermal growth factor (EGF)/EGF receptor (EGFR), sonic hedgehog (SHH/PTCH/GLI), Wnt/β-catenin, stromal cell-derived factor-1 (SDF-1) also designated as CXCL12/CXC chemokine receptor-4 (CXCR4) and/or TGF-β cascades (3,10,37,38,39,40,41,42,43,44,45,46,47,48). These oncogenic pathways may contribute to the sustained growth, migration, invasion, and metastasis of PC cells as well as the resistance to current clinical therapies and disease relapse. Additionally, the changes in the local microenvironment of PC cells, including the release of diverse soluble factors by activated myofibroblasts and immune cells in tumor-reactive stroma, may also influence their malignant transformation and thereby promote the PC transition to a more aggressive disease state (3,28,40,44,45,49,50,51,52). In this review, we provide a particular view on the potential implications of normal and malignant prostatic stem/progenitor cells in tissue homeostasis and disease development with respect to new stem cell concepts and the accumulating body of experimental evidence in this research field. Specifically, we describe recent advancements related to the establishment of physiological functions of prostatic stem/progenitor cells in maintaining homeostasis of the prostate gland during the adult lifespan. Moreover, we discuss the possible pathophysiological implication of their malignant counterpart, the PC stem/progenitor cells, in PC initiation and progression to aggressive and recurrent disease states. We also review the recent works that have led to the identification of novel drug targets for the development of more effective combination therapies against metastatic HRPCs. The emphasis is on new combined targeted therapies aimed at eliminating the total mass of PC cells, which consist of PC stem/progenitor cells with a basal or intermediate phenotype and their further differentiated progenies with a luminal phenotype.

Figure 1.

Proposed model of the molecular events associated with the PC initiation and progression through the malignant transformation of prostatic stem/progenitor cells. A, The anatomic localization of undifferentiated prostatic adult stem/progenitor cells expressing different stem cell-like markers including CD133/CD44/α₂β₁^high-integrin in basal cell compartment in normal prostate epithelium and their possible symmetric division in homeostatic conditions. B, The possible malignant transformation of prostate stem/progenitor cells induced through genetic and/or epigenic alterations resulting via their asymmetric division to the generation of a heterogenous population of poorly, moderately, and well-differentiated PC stem/progenitor cells and PC initiation. C, The possible accumulation of genetic and/or epigenetic alterations in tumorigenic PC stem progenitor cells leading to the sustained activation of numerous growth factor signaling cascades and their acquisition of a migratory phenotype during EMT program and PC progression to the invasive and metastatic disease stages.

Figure 2.

Scheme showing the possible oncogenic signaling pathways involved in the stimulation of sustained growth, survival, migration, invasion, and drug resistance of PC cells. The intracellular signal transducing elements and expression of target gene products induced through the activation of EGFR family members, sonic hedgehog (SHH/PTCH/GLI), Wnt/β-catenin, and SCF-1/CXCR4 pathways and possible cross-talks between these cascades are shown. Moreover, the chemotherapeutic drug efflux mediated through the ABC-multidrug transporters such as ABCG2 pump is also illustrated. BCRP-1/ABCG2, Breast cancer resistance protein-1; CDK, cyclin-dependent kinase; LRP, low-density lipoprotein receptor-related protein; MEK, extracellular signal-related kinase kinase; MMPs, matrix metalloproteinases; PLC-γ, phospholipase C-γ; PTCH, hedgehog-patched receptor; SHH, sonic hedgehog ligand; uPA, urokinase plasminogen activator; Wnt, wingless ligand.

II. Functions of Normal and Malignant Adult Prostatic Stem/Progenitor Cells

The prostatic epithelium is derived during embryogenesis from the differentiation of a stem/progenitor cell population localized in the embryonic endodermal urogenital sinus (UGS) epithelium from which prostatic epithelial buds develop under the influence of UGS mesenchyme (51,53,54,55,56,57). The normal prostate gland is constituted essentially of three cell types, including basal and secretory luminal epithelial cells and neuroendocrine cells within the prostatic stratified epithelium. Accumulating lines of evidence suggest that the prostatic stem/progenitor cells may also persist in the mature prostatic epithelium in adults within the specialized microenvironments, or “niches” localized in the basal compartment (Fig. 1) (53,55,56,58,59). The genetic and/or epigenic alterations occurring in prostatic stem/progenitor cells combined with the changes in their local microenvironment consisting of stromal cells and extracellular matrix components, and more particularly during aging, may however lead to their acquisition of aberrant functions and malignant transformation (60). These prostatic stem/progenitor cells with a deregulated behavior can contribute to certain hyperproliferative disorders affecting prostate gland epithelium, such as basal cell hyperplasia, benign prostatic hypertrophy or hyperplasia (BPH also designated as BHP) and PCs, that commonly afflict the middle-aged and elderly men (3,23,24,53,61,62,63,64,65,66,67,68). Based on the experimental observations indicating the critical importance of prostatic stem/progenitor cells in the homeostatic maintenance of prostate gland as well as their implication in PC development, several recent works have been carried out to enrich, isolate, and characterize these cells and their malignant counterpart, PC stem/progenitor cells with stem cell-like properties. Among the methodologies frequently used, there are the immunohistochemical analyses and diverse enrichment and isolation strategies, including magnetic-activated cell sorting and fluorescence-activated cell sorting (FACS) with antibodies directed against certain stem cell-like surface markers such as CD133, CD44, α₂β₁-integrin, and/or CXCR4 (23,29,49,64,65,69,70,71,72,73,74,75). Moreover, the DNA label-retaining cell analyses have also been performed to detect slow-cycling prostatic stem cells and the Hoechst side population (SP) method to characterize the prostatic epithelial cells with the stem cell-like properties including their high dye efflux ability associated with the high expression levels of ATP-binding cassette (ABC) multidrug efflux pumps (27,29,32,73,74,76,77). Hence, it has been observed that a very small subpopulation of nonmalignant or malignant prostatic stem/progenitor cells may be detected and isolated from normal and malignant prostatic tissue specimens from healthy individuals and patients with PCs and well-established nonmalignant or malignant human prostatic cell lines. In regard to this, we are reporting the recent works that have led to the identification and in vivo and in vitro characterization of specific functional properties of human and rodent normal prostatic stem/progenitor cells and the molecular events that may contribute to their malignant transformation during PC initiation and progression.

III. Functions of Normal Adult Prostatic Stem/Progenitor Cells in the Homeostatic Maintenance of the Prostate Gland

A. In vivo and ex vitro characterization of adult prostatic stem/progenitor cell properties

Several recent investigations revealed that a very small subpopulation of multipotent and undifferentiated prostate stem/progenitor cells, comprising about 0.1–3.0% of the total prostatic epithelial cell population, principally reside within specialized areas or “niches” localized in the basal cell layer of acinar and ductal regions of the human prostate gland (Fig. 1) (24,30,32,49,55,56,59,64,70,74,78,79,80,81,82,83). This small fraction of human adult prostatic stem/progenitor cells, which is enriched near the basement membrane, expresses several specific markers, including CD133 (also designated as prominin-1 or AC133), CD44, α₂β₁^high-integrin, breast cancer resistance protein-1/ABCG2, telomerase reverse transcriptase (TERT), and CK5/14 (Fig. 1) (24,32,49,55,56,58,59,64,70,74,78,80,81,82,83). Adult prostatic stem/progenitor cells, however, did not express detectable levels of luminal markers such as AR and prostate-specific antigen (PSA). Therefore, these cells did not directly depend on androgens for their survival and proliferation. Moreover, human CD133/α₂β₁^high-integrin prostatic stem/progenitor cells possess a high in vitro proliferative potential and are able to reconstitute fully differentiated prostatic structure-like acini in male nude mice in vivo (70). In rodents, the adult prostatic stem/progenitor cells, which express CD133, stem cell antigen-1 (Sca-1), TERT, α₆-integrin, Notch 1, ABCG2, and antiapoptotic factor Bcl-2 but not AR, are enriched in the proximal region in prostatic ducts and the urethra (3,30,69,84,85,86,87,88,89). The mature prostatic tissue normally undergoes limited regeneration under physiological conditions, and thus the prostatic stem/progenitor cells may be maintained in a quiescent state due to their local microenvironment through a complex network of soluble growth factors, such as TGF-β secreted by the nearboring stromal mesenchymal cells, that tightly regulate their functions in a paracrine manner (3,51,90). The prostatic stem/progenitor cells may display a high self-renewal capacity and give rise, by asymmetric division, to distinct highly proliferative transit-amplifying (TA)/intermediate cell subpopulations in pathological conditions, such as intense injuries (3,30,59,64,78,80,85). The TA/intermediate cells, which are organized as a hierarchical population, may express different levels of CD44 and/or α₂β₁-integrin, which are stem cell-like markers, and an intermediate phenotype [p63, CK5/14, CK8/18, prostate stem cell antigen (PSCA) and AR^−/low] between the basal and secretory cells of prostatic epithelium (3,36,59,64,78,80,83,85,91,92,93). In fact, a decline in the expression of stem/progenitor cell-like markers on TA/intermediate cells upon acquisition of a more differentiated phenotype during prostatic epithelium regeneration appears to occur gradually. The TA/intermediate cell subpopulations may ultimately regenerate by subsequent divisions all of the mature cell lineages in stratified epithelium consisting of basal, neuroendocrine, and luminal cells in vitro and in vivo, and thereby they can participate in restoring the prostate gland after injuries (59,78,80). More specifically, the majority of secretory luminal terminally differentiated epithelial cells localized at the apical region of the prostate epithelium only express luminal markers, such as high levels of AR, PSA, and CK8/18, and are dependent on androgens for their survival and growth (Fig. 1) (30,58,59,80,83,92). In contrast, the basal and neuroendocrine cells, which are devoid of nuclear AR, are AI for their survival and proliferation (30,83,92,94). The mature basal epithelial cell subpopulation found within the basal compartment adjacent to basement membrane also expresses several specific markers, including p63, CK5/14, CK19, and glutathione S-transferase π (GST π), whereas neuroendocrine cells express chromogranin A and synaptophysin (30,55,56,58,83,92). Additionally, the intermediate prostatic stem/progenitor cells expressing basal-luminal markers CK5/18 and/or p63 have also been detected in normal and malignant prostate epithelium, suggesting that these cell types may play an important role in prostate regeneration and pathogenesis (33,34,35,36,67,95,96,97,98). Along these lines of thought, it is noteworthy that a positive staining of a very small subpopulation (clusters) of PC cells has been observed in certain primary and metastatic prostate adenocarcinoma specimens using basal cell-specific 34βE12 antibody directed against CK1/5/10/14 (35,99,100,101). In fact, 34βE12 antibody is generally used as a clinical diagnostic tool to distinguish between the PC lesions in which the basal cells are generally destroyed and benign prostatic lesions that retain the cytokeratin-expressing basal cells. Therefore, this observation underlines the importance of considering the possible presence of a small population of PC cells expressing the basal cell-like markers such as cytokeratins during the clinical diagnosis of patients, and more particularly, when the basal cell-reactive 34βE12 antibody is used.

B. Paracrine and autocrine regulation of prostatic stem/progenitor cell proliferation and differentiation by the interplay of androgens, growth factors, and integrins

Numerous investigations indicated that an interplay of signaling cascades initiated by the androgens, integrins, and growth factors, such as keratinocyte growth factor (KGF), also designated as fibroblast growth factor (FGF)-7, EGF, TGF-α/EGFR, IGF-I, sonic hedgehog SHH/PTCH, and Notch, may influence the prostatic epithelial cell proliferation and/or differentiation in paracrine and/or autocrine manners (3,8,39,49,53,59,65,70,80,81,85,102,103,104,105,106,107,108,109,110,111). These factors can participate in prostate organogenesis during embryogenesis or prostate gland development and regeneration in fetal, postnatal, and adult life (8,49,53,65,70,80,81,85,105,107,110). More specifically, androgens may induce a differentiation of prostate epithelial cells by acting on the adjacent stromal cells, causing them to secrete the soluble growth factors that stimulate the differentiation of prostatic stem/progenitor cells. In support of this, the tissue recombination experiments with AR-negative epithelium and normal UGS mesenchyme expressing wild-type AR showed that androgens may mediate their effects on AR⁺ mesenchymal cells by stimulating the release of diffusible growth factors such as KGF, which in turn may induce the growth and differentiation of prostatic epithelium (51,59,112). Further experimental evidence supporting the paracrine role of androgens on prostatic epithelium differentiation is based on the observation that the adult rodent prostate undergoes involution after the surgical castration, and the exogenous androgen treatment may reverse this degenerative effect (113,114,115,116,117). In this regard, the castration triggers apoptotic death of the luminal prostatic epithelial cells that are dependent on androgens for their survival. In contrast, the cells within the basal compartment near basement membrane, including AR⁻ prostatic stem/progenitor cells may survive and subsequently be restimulated by exogenous androgen application to regenerate the functional prostatic epithelium (113,116,117). The results from a recent study have also revealed that the castration may be accompanied by the up-regulated expression of EGFR in the prostatic epithelial cells, suggesting that the EGF and TGF-α/EGFR system may contribute to the regeneration of prostatic epithelium (107).

In addition, the androgens can also mediate their differentiating effect through a direct action on the prostatic epithelial cells. For instance, the results from recent studies have shown a direct effect of androgens on prostatic TA/intermediate cells isolated from human prostatic tissue obtained after transurethral resection of the prostates of BPH patients or cystoprostatectomy for bladder cancer (64,65). More specifically, the treatment of CD133⁻/α₂β₁^high-integrin/AR^−/low prostatic TA/intermediate cell subpopulations, which have been enriched from the total basal cells of the prostate gland with an androgen analog, R1881, down-regulated the α₂β₁-integrin expression level and induced their differentiation into epithelial cells with a terminally differentiated epithelial cell-like phenotype including the expression of AR, PSA, prostatic acid phosphatase, and CK18 (65). The treatment of CD133⁻/α₂β₁^high-integrin/AR^−/low cells, which express the FGF receptor tyrosine kinase (FGFR2) possessing a high affinity for KGF, with an anti-β₁-integrin antibody or KGF also suppressed α₂β₁^high-integrin expression and induced their differentiation into cells expressing the luminal markers including AR, prostatic acid phosphatase, and CK18 (64). In rodents, it has also been observed that α₆-integrin positive-murine prostate stem cells, when grown in Matrigel media containing its ligand, extracellular protein laminin, can form the clonogenic spheroid structures termed as prostate spheres that spontaneously differentiate into basal cells or PSCA-expressing TA cells (88). The addition of 5α-dihydrotestosterone in cell culture media, which stabilized AR and induced its nuclear translocation, also triggered the differentiation of cells into progenies possessing a luminal phenotype (88). A fraction of sphere cells endowed with the self-renewal capacity and multilineage differentiating potential can form prostate gland-like structures containing the basal and secretory luminal cell layers in vivo (88).

In light of these experimental observations, it appears that a very small subpopulation of undifferentiated and multipotent AR⁻ prostatic stem/progenitor cells can participate along the lifespan to generate moderately differentiated TA/intermediate cell subpopulations. The TA cells, in turn, can replenish the bulk mass of further differentiated AR⁺ epithelial cells with the secretory functions and thereby maintain the prostate homeostasis or regenerate the prostatic epithelium after tissue injuries. Unfortunately, the changes in prostatic epithelium-mesenchyme interactions may be accompanied by the malignant transformation of prostatic stem/progenitor cells into PC-initiating cells (3,30,36,40,50,67,83,118).

IV. Functions of Prostatic Cancer Stem/Progenitor Cells in Prostate Cancer Initiation and Progression

The accumulation of genetic and/or epigenic alterations occurring in prostatic stem/progenitor cells during the lifespan combined with the changes in their local microenvironment, and more particularly after intense prostatic epithelium injuries such as chronic proliferative inflammatory atrophies of the prostate gland, may result in their malignant transformation into PC stem/progenitor cells showing aberrant growth and differentiation potential (Fig. 1) (3,30,31,36,50,67,83,118). In general, PC development is accompanied by a continued transition from low- to high-grade prostatic intraepithelial neoplasias that may culminate in well-established localized PCs. The primary prostatic neoplasms may subsequently progress into locally invasive forms whose event may be accompanied by the spread of PC cells at the surrounding tissues, including the lymph nodes, and/or metastasize to distant tissues/organs such as the bone marrow, brain, lungs, and/or liver (Fig. 1) (30,40). This notion is well supported by the identification of a rare subpopulation of human PC stem/progenitor cells comprising about 0.1–3.0% of total PC cells by immunohistochemical staining in malignant prostatic adenocarcinomas and metastatic neoplasms (23,24,29,32,82). The PC stem/progenitor cells isolated from primary neoplasms express several prostatic stem cell-like markers such as CD133, CD44, α₂β₁^+/high-integrin, CK5/14, CK18, and/or CXCR4 but lack AR and PSA luminal marker expression (23,24,29,31,32,82). In regard to this, it has been reported that the CD133/α₂β₁^+/high-integrin PC stem cells isolated from primary PC (P4E6), when grafted orthotopically in a matrigel plug containing human prostatic stroma, were able to form multiple intraprostatic tumors in nude mice showing a histology like the original Gleason 4 grade PC (31). Importantly, certain PC stem/progenitor cells, unlike their normal counterpart, prostatic stem/progenitor cells with a basal phenotype (CK5/14), appear to possess an intermediate phenotype based on the coexpression of basal (CK5/14) and luminal (CK8/18) cytokeratin markers (33,34,35,36,95,96,98). These findings suggest that these PC cells may represent early progenies with an intermediate phenotype, which have been preserved even after the destruction of the basal epithelial cell layer during PC progression. This line of thought is well supported by the observation that the hedgehog signaling elements, receptor Patched 1 (PTCH1) and GLI transcription factor, were frequently found to colocalize with p63 in CD44/CK8/14-expressing prostatic hyperplasia basal cells and in the PC cells but were rarely detected in normal basal cells (67). On the basis of these observations, it has been proposed that the hedgehog signaling activation may induce a transitory differentiation of prostatic stem/progenitor cells into CD44⁺/p63^−/+ hyperplasia basal cells with an intermediate phenotype (CK8/14), and this early transforming event may culminate in tumorigenesis by giving rise to CD44-, PTCH1-, and GLI-expressing PC cells (67).

Hence, it seems that certain CD133⁺/AR⁻ PC stem/progenitor cells may successfully complete the multistep selective process and, more particularly, the acquisition of a migratory phenotype during the EMT program at primary neoplasms that may allow them to invade and migrate at distant metastatic sites such as bone marrow where they can form the established metastases (Fig. 1). Further studies are necessary to establish more precisely the specific gene expression pattern that characterizes CD133⁺/AR⁻ PC stem/progenitor cells and to distinguish them from their further differentiated progenies detected in different poorly to moderately differentiated PC subtypes during the disease progression.

A. Heterogeneity of prostate cancers derived from distinct prostate cancer stem/progenitor cells

Although the PC stem/progenitor cells may arise from the malignant transformation of normal prostatic stem cells that acquire an aberrant growth and differentiation potential, it is likely that the mutations in a wide variety of oncogenes and tumor suppressor genes may lead to different deregulated molecular pathways during PC initiation and progression (3,28,30,40,44,109,119,120,121,122,123,124,125). Therefore, the occurrence of distinct genetic and/or epigenic alterations in PC stem/progenitor cells concomitant with the changes in their local microenvironment may activate different oncogenic cascades during prostate carcinogenesis and thereby contribute to the intratumoral heterogeneity and/or development of diverse PC subtypes. For instance, the results from a study consisting of the deletion of the tumor suppressor gene, phosphatase and tensin homolog deleted on chromosome 10 (PTEN) in murine prostatic epithelium, showed that the loss of the PTEN gene product, lipid phosphatase, is accompanied by an expansion of Sca-1/Bcl-2/CK5/p63⁺-expressing prostatic stem/progenitor cells and their differentiation into a Sca-1/Bcl-2/CK5/p63^−/low cell subpopulation (124). These early progenies, which are localized near the basement membrane of the prostatic epithelial compartment, can give rise to a further differentiated cell subpopulation that dissociates from the basement membrane (124). The PTEN inactivation in a murine model in vivo, which activates the phosphatidylinositol 3′-kinase (PI₃K)/Akt signaling pathway, may result in the prostatic intraepithelial neoplasia lesions that progress into invasive and metastatic PCs recapitulating the molecular events seen during human PC progression (120,123,124). Furthermore, the combined loss of PTEN and other tumor suppressor genes, such as p53, or a lack of cyclin-dependent kinase inhibitor, p27^KiP1 expression may promote tumorigenesis and rapidly lead to the aggressive, invasive, and lethal PCs in murine models in vivo (120,122). Similarly, the deletion of both p53 and retinoblastoma (Rb) tumor suppressor genes in mouse prostatic epithelium may also induce a malignant transformation of AI Sca-1⁺ prostatic stem/progenitor cells in the proximal region of the ducts (121). These malignant Sca-1⁺ prostatic stem/progenitor cells, in turn, can give rise to a heterogeneous population of PC cells that express the neuroendocrine (chromogranin A and synaptophysin) and luminal epithelial (AR and CK8) markers but lack expression of basal cell marker CK5 in a microenvironment-dependent manner (121). In addition, the recombination of human prostatic carcinoma-associated fibroblasts (CAFs), which secrete the high levels of SDF-1 and TGF-β1, with the nonmalignant human prostatic epithelial cell line BHP-1 also induced their malignant transformation in vivo (45,50). This oncogenic event was mediated at least in part through the activation of TGF-β1 that stimulated the Akt pathway via the up-regulation of CXCR4 expression on PC cells (45,50). The p63/CK14/CK8/18-expressing BHP-1 cell sublines established from the tumors generated by transformed BHP-1 cells, gave the colonies in soft agar assays in vitro whereas parental BHP-1 cells did not. Moreover, p63/CK14/CK8/18 BHP-1 cells formed the poorly to moderately differentiated tumors resembling the original primary tumors when grafted beneath the renal capsule of athymic mice (50). Several lines of evidence have indicated that the PC stem/progenitor cells, like their normal counterpart, prostatic stem/progenitor cells, may reexpress the markers specific to mature and differentiated basal (CK5 and p63), luminal (AR), and/or neuroendocrine cells (nestin) depending upon the particular microenvironment prevalent at the primary and secondary neoplasms. For instance, the human TERT-overexpressing PC epithelial cell lines termed HPET or clonally derived HPET cells with stem cell-like characteristics (CD44⁺, CD133⁺, nestin⁺, Oct-4⁺, Nanog⁺, Sox2⁺ and AR/PSA⁻) have recently been established from prostatic adenocarcinoma specimens. These TERT⁺ HPET cells gave rise to the differentiated PC cells that expressed the markers specific to mature basal, neuroendocrine, and AR⁺ luminal cells in vivo (101). The parental and clonal HPET cells endowed with a self-renewal potential were also able to reconstitute the human prostatic tumors with a histopathological architecture and Gleason grade 8 to 9 similar to the original patients’ adenocarcinomas from which they were derived after serial transplantation in vivo (101). Similarly, human TERT-immortalized primary malignant RC-92a/hTERT showing a CD133⁺/CD44⁺/CK1/5/10/14⁺/CK18⁺/CXCR4⁺ and AR/PSA⁻ phenotype and endowed with a self-renewal potential were also able to differentiate into AR-expressing cells (29).

In addition, an enhanced expression and activation of TERT and/or diverse growth factor and chemokine signaling pathways such as EGF-EGFR, hedgehog, Wnt/β-catenin, estrogen/estrogen receptor, TGF-β, and/or SDF-1/CXCR4 in PC cells may also contribute to PC initiation and/or its progression to a metastatic state (Figs. 1 and 2) (3,28,29,37,44,67,87,108,109,126). The stimulation of these oncogenic cascades may enhance the expression of diverse signaling elements such as cyclin D1, MYC, Bcl-2, survivin, matrix metalloproteinases, urokinase-plasminogen activator, cyclooxygenase-2 (COX-2), ABC transporters, and transcriptional repressors of E-cadherin, snail, slug, and twist in PC stem/progenitor cells and/or their differentiated progenies. Hence, the activation of these tumorigenic cascades may contribute to the sustained growth, survival, migration, invasion, and/or metastasis of PC cells to distant sites as well as their resistance to current therapeutic treatments and disease relapse. Furthermore, the results of numerous recent studies have also revealed that the chromosomal rearrangements implicating the oncogenic ETS family transcription factors, and which fuse the 3′-end ERG, ETV1, ETV4, or ETV5 gene to the 5′-region of TMPRSS2 as well as ETV1 and untranslated region from SLC45A3, HERV-K 22q11.3, C15ORF21, or HNRPA2B1 gene, may be detected in certain PC subtypes (127,128,129,130,131,132,133,134,135,136,137,138,139,140). More particularly, a gene fusion consisting of androgen-regulated TMPRSS2 (21q22.2) and ERG (21q22.3) appears frequently to occur during PC development (127,128,129,130,131,132,133,134,136,137,138,139). Importantly, although the expression of TMPRSS2-ERG gene fusion may be up-regulated in response to the androgen treatment in AR⁺ PC cells, it has also been observed that an AR⁻ NCI-H660 prostatic cell line derived from a lymph node metastasis taken from a patient before therapy harbored TMPRSS2-ERG gene fusion and expressed ERG oncoprotein in an AI manner (132,141). Together these observations suggest that these fusion oncoptoteins like the wild-type ETS oncoproteins could mediate the transactivation of ETS target genes that may be involved in the malignant transformation of PC cells. Further studies are however essential to establish whether certain of these chromosomal rearrangements including TMPRSS2-ERG gene fusion may occur in PC stem/progenitor cells and/or their early progenies, and thereby contribute to their acquisition of more aggressive behavior during PC progression.

On the other hand, the malignant transformation of PC cells, and more particularly during the EMT process at the primary neoplasm, as well as their abilities to form the metastases at distant sites may also be influenced by their local microenvironment (3,28,30,40,44,45,49,50,51,52). The reciprocal PC cell-microenvironment interactions implicate the interplay of a complex network of diverse diffusible soluble growth factors and cytokines released by the tumor and host cells in reactive stroma and changes in the extracellular matrix components. Among host stromal cells, there are tumor microvasculature-associated endothelial cells, activated myofibroblasts, and immune cells such macrophages that may be recruited at the primary and secondary neoplastic sites (Fig. 1) (40). All of these molecular events may contribute to the progression of localized PCs to metastatic disease stages in promoting the tumor growth, angiogenesis and PC cell migration, invasion, and adhesion at distant tissues/organs where they can form the well-established metastases.

Hence, the specific oncogenic alterations occurring in PC stem/progenitor cells concomitant with the changes in their local microenvironment during PC progression may influence the phenotype of their progenies at the primary neoplasm and metastatic sites and thereby alter the differentiation state of the tumor subtype and the aggressiveness of the disease.

B. In vitro and in vivo prostate cancer stem/progenitor cell models

Recent experimental lines of evidence have revealed that several well-established nonmalignant and malignant prostatic cell lines may contain a prostatic stem/progenitor cell subpopulation that may be enriched or isolated by SP Hoechst technique and FACS with the specific antibodies directed against prostatic stem cell-like surface markers (27,58,71,72,73,75,83,86,142). These well-characterized nonmalignant and malignant prostatic stem/progenitor cells may then constitute the in vitro and in vivo cell models to define the molecular events involved in maintaining the prostatic epithelium differentiation and inducing the malignant transformation of prostatic stem/progenitor cells during PC progression. In this respect, a recent FACS study has revealed the presence of a small prostatic stem/progenitor cell subpopulation that expressed CD133/ABCG2/α₂β₁^+/high-integrin but lacked AR and PSA protein expression in nonmalignant and nonimmortalized human PrEC and immortalized parental RWPE-1 and PZ-HPV-7 epithelial cells corresponding to about 1.3, 0.24, and 1.75% of the total cell population, respectively (143). It has also been observed that the prostatic stem/progenitor cells from these well-established normal prostatic epithelial cell lines differentiated in vitro in low Ca²⁺ medium but with a limited potential into the TA/intermediate cells expressing different markers such as PSCA protein, a late intermediate marker (83). Importantly, the poorly differentiated and tumorigenic AR⁻ PC3 cell line with intermediate phenotype (CK5/18), which contains a small subpopulation of PC progenitor cells expressing stem cell-like markers (CD133⁺ and CD44^high), can also form poorly differentiated tumors in vivo that resemble the original patient’s tumor (72,144,145). Similarly, it has been reported that the PC cells constituting xenograft tumors established from LAPC9 or DU145 cells may be organized as a functional hierarchy where CD44⁺/α₂β₁^+/high or α₂β₁^−/low-integrin-expressing cells display a higher tumorigenic potential than the CD44⁻ cell fraction in nude mice (71,72,73). Additionally, the results from a recent study have revealed that the very small subpopulation of CD133⁺/CD44⁺/α₂β₁⁺-integrin DU145 isolated from the parental AR⁻ DU145 cell mass, which expresses a high level of β-catenin and MYC but a low level of apoptotic factor Bax, showed a self-renewal, extensive differentiating, and proliferative abilities and a higher tumorigenic potential relative to the CD133⁻/CD44⁺/α₂β₁^−/low-integrin DU145 cell fraction in vivo (75).

Hence, the use of PC stem/progenitor cells with stem cell-like properties from well-established cell lines should constitute useful preclinical in vivo cell models to establish the molecular mechanisms involved in the treatment resistance as well as to estimate the efficacy of new antitumoral drugs to eliminate these PC-initiating cells that can contribute to tumor formation, metastasis, and disease relapse. In this matter, we describe the molecular mechanisms that may be responsible, at least in part, for the drug resistance of PC stem/progenitor cells and their further differentiated progenies and disease relapse as well as the novel targeting therapies for overcoming treatment resistance and improving the current clinical therapies for patients diagnosed with locally advanced and metastatic HRPCs.

V. Molecular Mechanisms Involved in the Therapeutic Resistance of Prostatic Cancer Stem/Progenitor Cells and Their Further Differentiated Progenies

An understanding of the molecular mechanisms involved in the resistance of highly tumorigenic and metastatic PC stem/progenitor cells and their further differentiated progenies to current therapeutic regimens is of immense clinical interest. This information will enable us to establish novel strategies for a more effective treatment of locally advanced and/or metastatic HRPCs. Indeed, the intrinsic or acquired therapeutic resistance of PC stem/progenitor cells to current clinical antiandrogen, radiation, and chemotherapeutic treatments may be causally linked and contribute to treatment failure and disease relapse (Fig. 3). The cancer stem/progenitor cell concept suggests that the bulk mass of differentiated PC cells with a luminal phenotype does not possess the ability to form new tumors. Based on this, it is likely that AR⁻ PC stem/progenitor cells can contribute to PC recurrence (27,28,30,31,44). This concept is consistent with the observations that the patients with PCs initially respond to androgen deprivation therapy, which induces a massive rate of apoptotic death of bulk mass of androgen-dependent AR⁺ PC cells in the luminal compartment, and thereby clinically leads to tumor regression and a partial remission (Fig. 3). However, the emergence of HRPCs, which may be due to the persistence of AI and chemoresistant PC cells, including AR⁻ PC stem/progenitor cells that are able to initiate and drive tumor growth in primary and secondary neoplasms, may cause the disease relapse even after several years (Fig. 3) (8,27,28,30,44,86).

Figure 3.

Scheme showing the possible molecular events associated with the resistance of PC cells to current clinical treatments and new targeting therapies against locally advanced and/or metastatic HRPCs. Based on the PC stem/progenitor concept, it is likely that the androgen-independent and highly tumorigenic and migrating PC cells including AR⁻ PC stem/progenitor cells may persist after the current therapeutic treatments including androgen-deprivation therapies. Hence, they can contribute to the tumor regrowth and disease relapse. In contrast, novel combination therapies targeting the total mass of PC cells including AR⁻ PC stem/progenitor cells and their differentiated AR⁺ progenies as well as their local microenvironment including the host stromal cells such as activated myofibroblasts and endothelial cells could lead to an eradication of the global PC cell population and inhibition of the neoangiogenesis process. These new targeting therapies could induce a total tumor regression and thereby result in a complete remission of patients diagnosed with aggressive and incurable PCs.

In addition, the acquisition of drug resistance by PC cells may also result from the genetic and/or epigenic changes occurring in tumor cells during disease progression after continued exposure to therapeutic agents for a long time (3,9,27,28,30,44,77,146,147,148,149,150,151,152,153,154,155,156). The development of simultaneous resistance to several drugs and/or radiation therapy may also occur during the transition from localized PC into invasive and metastatic stages. Among the molecular events associated with the occurrence of multidrug resistance (MDR) phenomenon, there are the enhanced expression and/or activity of ABC superfamily transporter genes that encode the transmembrane proteins acting as the drug efflux pumps. The MDR phenomenon is frequently involved in the resistance to diverse unstructurally related chemical compounds, such as distinct cytotoxic chemotherapeutic agents including docetaxel and mitoxantrone (Fig. 2) (27,77,86,146,147,148,149,150,151,155,156,157,158,159). In this regard, the immunohistochemical analyses have revealed that the expression of a MDR transporter, P-glycoprotein encoded by the MDR1 gene, was elevated in 96 patients with PC, and this enhanced expression correlated with tumor grade, stage, and PSA levels (160). Moreover, several studies revealed that the metastatic PC cell lines, DU145 and PC3 cells, express high levels of multidrug efflux pumps including ABCG2 and MDR-associated protein-1 that may contribute to the SP fraction detected by the Hoechst technique (147,148,149,150,151,156,158,159,161). This suggests then the possible implication of these ABC multidrug transporters in the drug resistance of PC cells, including PC stem/progenitor cells, and in disease recurrence in certain patients. Additionally, the drug resistance may also be caused by the changes in the expression and/or activity of drug targets including the β-tubulin and/or topoisomerase II-α and -β, modification in GST π, DNA repair enzymes, and/or antiapoptotic vs. apoptotic signaling cascade elements in PC cells during PC progression (9,77,146,147,149,162,163). As a matter of fact, it has been observed that the expression levels of MDR-associated protein-1, topoisomerase IIα, GST π, p53, and Bcl-2 detected in PC tissue specimens at the time of surgical intervention enhanced with the Gleason grade (146). The up-regulation of glucosylceramide synthase, cholesterol, caveolin-1 and antiapoptotic factors such as Bcl-2, survivin, nuclear factor-κB (NF-κB), and PI₃K/Akt concomitant with the down-regulation of cellular ceramide levels may also occur during PC progression (3,8,9,77,146,156,162,163,164,165,166,167). These molecular events may increase the resistance of PC cells including PC stem/progenitor cells to current therapeutic treatments. In addition, the enhanced expression and/or sustained activation of numerous growth factor, chemokine, and cytokine signaling pathways, including EGF-EGFR, hedgehog SHH-PTCH, Wnt/β-catenin, estrogen/estrogen receptor, TGF-β, and SDF-1/CXCR4 axes in PC cells may also contribute to their sustained growth, survival, migration, invasion, and/or metastasis during the initiation and progression from localized PCs into advanced and metastatic disease states (Fig. 2) (3,8,37,38,39,42,43,45,47,48,68,168,169,170,171,172,173,174,175,176,177,178). In particular, the enhanced expression of EGFR and its ligands, EGF, TGF-α, and amphiregulin, hedgehog, and β-catenin signaling elements occurring in PC cells during the PC progression has been associated with the development of AI disease, a high rate of disease relapse, and decreased survival of patients after therapy (38,109,169,179,180,181,182,183,184,185,186). Importantly, several recent lines of evidence have revealed that the activation of these oncogenic signaling pathways in AI PC stem/progenitor cells may contribute to their malignant transformation into migrating, invasive, and metastatic PC cells and disease recurrence (29,67,102,156,187). In support of this, it has been reported that an accumulation of EGFR-positive tumor cells expressing intermediate markers CK5/18 occurred in a subset of patients with HRPCs (33,96). This suggests that the number of CK5/18- and EGFR-positive PC progenitor cells with an intermediate phenotype may rise during androgen deprivation and disease relapse, supporting the benefit of using the EGFR inhibitor to target this neoplastic cell subpopulation. Similarly, the CD44⁺ marker and hedgehog signaling element, PTCH1, were expressed and colocalized in PC cells with an intermediate phenotype CK14/8 (156). Moreover, the expression levels of β-catenin and hedgehog signaling element, smoothened (SMO), were higher in highly tumorigenic CD44⁺ DU145 cell subpopulation relative to the weakly tumorigenic CD44⁻ DU145 cell fraction (72). These observations suggest that the EGFR, hedgehog, and Wnt/β-catenin signaling cascades may be involved in the malignant transformation of PC stem/progenitor with an intermediate phenotype and disease relapse. Therefore, the development of a novel therapy by the combined targeting of these tumorigenic signaling elements, which are activated in AI and metastatic PC stem/progenitor cells as well as their supporting local microenvironment, could represent a more promising strategy for improving the current therapeutic treatments and the overall survival of patients diagnosed with metastatic HRPCs.

VI. Novel Therapeutic Strategies against the Locally Advanced and Metastatic Hormone-Refractory Prostate Cancers

Although surgical tumor resection, hormonal therapies, radiotherapy, and adjuvant chemotherapy, alone or in combination, show beneficial effects and a significant curative rate in treating patients with localized PCs in the early stages, the development of locally advanced and/or metastatic HRPCs eventually results in disease recurrence. Because most patients who undergo potentially curative resection for advanced and/or metastatic HRPCs subsequently relapse due to the persistence of foci and micrometastases, systemic chemotherapy may represent another option to eradicate the PC cells, including the highly tumorigenic PC stem/ progenitor cells that can drive tumor growth at primary neoplasms and distant metastatic sites. In this regard, we are reporting the recent advances that have led to the identification of novel therapeutic strategies targeting PC stem/progenitor cells and their further differentiated progenies as well as their local microenvironment, which would allow us to improve the efficacy of current antiandrogen and chemotherapeutic treatments against locally advanced and/or metastatic HRPCs.

A. Targeting of PC stem/progenitor cells and their further differentiated progenies

Several investigations revealed that EGFR, hedgehog, Wnt/β-catenin, and/or SDF-1/CXCR4 signaling cascades as well as ABC multidrug efflux pumps could represent important targets for overcoming the intrinsic or acquired resistance of PC stem/progenitor cells and/or their further differentiated progenies to current clinical therapies (Fig. 4) (3,9,37,45,46,47,108,109,126,155,172,184,185,186,188,189,190,191,192,193). Particularly, recent studies have revealed that the blockade of these tumorigenic signaling cascades could be beneficial as adjuvant therapy in the early phases of PC for decreasing the risk of relapse as well as at the late stages for improving the efficacy of current androgen deprivation therapy, radiotherapy, and/or systemic chemotherapy and patient survival rates. In this regard, it has been observed that the blockade of the EGFR pathway by anti-EGFR antibody (cetuximab, also designated erbitux, mAb-C225, and IMC-C225) or EGFR tyrosine kinase inhibitor (gefinib, erlotinib, and EKB-569) caused an arrest in the G₁ phase of the cell cycle, inhibited invasion, and/or induced apoptosis in metastatic PC cells in vitro and in vivo (3,169,171,190,191,192,193,194,195,196,197,198,199,200,201,202). More specifically, the preclinical and clinical trials with a low dose of specific inhibitor of EGFR tyrosine kinase activity, gefitinib, have indicated a potential benefit of using this type of agent, which generally shows a good bioavailability and few side effects in combination with other chemotherapeutic drugs used for numerous cancer types overexpressing EGFR, including PCs (203). Although the use of gefitinib as a monotherapy is generally not very effective, it has been observed that this agent may increase the cytotoxic effects induced by diverse chemotherapeutic agents, such as docetaxel, paclitaxel, and platinum compounds, cisplatin and carboplatinum, on metastatic and AI PC3 cells in vitro and in vivo (170,204). Moreover, gefitinib might also inhibit the AR activity and act in cooperation with the antiandrogenic agents such as bicalutamide (205,206). Similarly, the inhibition of the hedgehog cascade, either by using SMO signaling element inhibitor, cyclopamine, or the anti-SHH antibody, has been observed to result in an inhibition both of the growth via a blockade in the G₁ phase of the cell cycle and the invasion of metastatic PC cells in vitro and in vivo, whereas the normal prostate epithelial cells were insensitive to the cytotoxic effects of these agents (3,38,39,42,43,169,188,207). Of particular therapeutic interest, the continuous cyclopamine treatment of poorly differentiated and highly tumorigenic PC3 cell xenografts established in nude mice in vivo also resulted in apoptotic cell death and tumor regression without a sign of the adverse effects on normal cells and tumor recurrence after 72 h of treatment cessation, suggesting that PC-initiating cells have been eradicated (39). The expression levels of ABCG2 and MDR1 (ABCB1) multidrug efflux pumps in PC3 cells were also decreased after the cyclopamine treatment, supporting the potential benefit of using the cyclopamine to counteract the MDR mediated by these transporters in PC cells (156). In this regard, the results from our recent studies have also indicated that the combined use of lower doses of gefinitib and cyclopamine with docetaxel or mitoxantrone induced a growth arrest and a higher rate of apoptotic death on diverse metastatic and androgen-sensitive AR⁺ LNCaP-C33 cells and AI AR⁺ LNCaP-C81 and AR⁻ DU145 and PC3 cells than the individual drugs or two-drug combination treatment (170,208). Additionally, the combined use of the conventional treatments including antiandrogenic therapies with the differentiating agents or epigenic modulators of gene transcription that are able to promote the differentiation of PC cells including PC stem/progenitor cells and/or their early TA progenies into their further differentiated progenies and induce a growth arrest and/or apoptotic death of PC cells also may represent a potential therapeutic strategy against locally advanced and aggressive PC forms (209,210,211,212,213,214). In support of this, the result from numerous studies indicated that the use of differentiating agents such as retinoic acids, including natural active metabolite of vitamin A, all-trans-retinoic acid, and other retinoids, retinoic acid metabolism-blocking agents, and/or histone deacetylase inhibitors, alone in combination therapies, induced the differentiation, growth inhibition, and/or apoptotic death of PC cells in vitro and in vivo (210,215,216,217,218,219,220,221,222,223,224,225,226,227,228). Additional investigations are however necessary to establish more precisely the molecular mechanism(s) of anticarcinogenic effects mediated by these agent types as well as to ascertain whether they are able to eradicate the total mass of PC cells including immature PC stem/progenitor cells in vitro and in vivo without major systemic toxicity.

Figure 4.

Scheme showing the novel therapeutic strategies against locally advanced and/or metastatic HRPCs by targeting different oncogenic signaling cascade elements in PC cells. The pharmacological agents acting as the potent inhibitors of the tumorigenic signaling cascades including the selective inhibitors of the EGF-EGFR system (gefitinib, erlotinib, and CI1033), SMO hedgehog signaling element (cyclopamine), CXCR4 antagonist, and monoclonal antibody (mAb) directed against CXCR4 or Wnt ligand are indicated. Moreover, the potential anticarcinogenic effects induced in PC cells through the blockade of these tumorigenic cascades including the inhibition of cell proliferation, survival, migration, invasion, and metastasis are also indicated. Particularly, the possible stimulation of the increase of mitochondrial membrane permeability (MMP), activation of caspase cascades, production of reactive oxygen species (ROS), and/or DNA damage induced by the chemotherapeutic drugs are illustrated. These molecular events may contribute to triggering the activation of apoptotic signaling cascades leading to PC cell death. In addition, the potent inhibitory effect mediated by a specific inhibitor on ABCG2-multidrug efflux pump and whose event may lead to the intracellular chemotherapeutic drug accumulation is shown.

In addition, the altered expression and/or activity of TERT and apoptosis-regulatory signaling elements including the up-regulation of antiapoptotic factors (Bcl-2 family proteins, MYC, survivin, clusterin, PI₃K/Akt, and NF-kB), and/or the down-regulation of tumor suppressor gene products (p53 and PTEN), as well as the enzymes involved in the metabolism of proapoptotic ceramide may also occur in PC cells during the progression to AI disease and thus contribute to treatment resistance (3,9,46,77,123,124,146,164,165,168,229,230,231,232,233). Therefore, the targeting of one or several of these deregulated signaling elements may constitute promising approaches for improving the current androgen deprivation, radiation, and docetaxel- or mitoxantrone-based therapies. Particularly, the inhibition of TERT activity, which is up-regulated in proliferating PC cells including the immature PC progenitor cells during the disease progression, has great therapeutic promise (214,234,235). For instance, it has been observed that a combination of the active form of vitamin D3, 1α,25-dihydroxyvitamin D3, and 9-cis-retinoic acid inhibited TERT activity and growth of PC3 cell xenografts established in nude mice in vivo (235). Moreover, the inhibition of the ceramide degradation by using a specific inhibitor of sphingosine kinase-1 significantly improved the growth inhibitory and antimetastatic effects induced by docetaxel on the orthotopically implanted PC3 cells in nude mice in vivo (232). The combination of docetaxel plus bcl-2 and c-myc antisense oligodeoxynucleotides also produced a significant tumor growth inhibition of PC3 cell xenografts that resulted in a significant increase in mice survival (233). Similarly, the combined use of mitoxantrone plus adenoviral-mediated p53 gene transfer, Ad5CMV-p53 and antisense oligodeoxynucleotide targeting clusterin gene, also completely inhibited sc implanted AI PC3 cell-derived tumor growth as well as lymph node metastases from orthotopic PC3 cell-derived tumors in 60 and 100% of mice, respectively (230). Several recent works have revealed the substantial benefit to targeting the supporting tumor microenvironment of PC cells for improving the efficacy of treatments against aggressive PCs.

B. Targeting of the local tumor microenvironment of PC stem/progenitor cells and their further differentiated progenies

The targeting of the local microenvironment “niche” and stromal components that may influence the malignant transformation of PC stem/progenitor cells and their further differentiated progenies during PC progression and the angiogenic process also represent other promising therapeutic strategies (Figs. 3 and 4) (40,45,46,52,119,236). The combined use of antiangiogenic agents such as a specific antibody or an inhibitor of vascular epidermal growth factor (VEGF), VEGF receptor VEGFR), and/or COX-2 with other cytotoxic drugs acting on PC cells may notably counteract the tumor growth and invasion, and thereby prevent disease relapse (3,46,194,237,238). For instance, it has been observed that the combination of docetaxel, or anti-EGFR antibody, cetuximab, with sunitinib malate (SU11248), an oral multitarget tyrosine kinase inhibitor targeting VEGFR-1, -2, and -3, platelet-derived growth factor receptors α and β, stem cell factor receptor (KIT), and FMS-like tyrosine kinase-3 receptor (FLT3), induced a supra-additive inhibitory effect on tumor growth of PC3 xenografts established in mice in vivo (239). New therapeutic strategies targeting the prostatic CAFs and/or bone marrow-derived or vascular wall-resident endothelial progenitor cells or mesenchymal stem cells, which can contribute to the tumorigenesis by giving rise to new tumor endothelial cells or activated myofibroblasts, respectively, also merit further investigation (240,241,242). In this regard, the results from a recent study have revealed that the combined use of low doses of a specific inhibitor of fibroblast-myofribroblast transition, halofuginone and docetaxel, synergistically inhibited the PC cell xenograft-derived tumor development in nude mice (243). Additionally, because it has been reported that the secretion of the high levels of SDF-1 and TGF-β1 by human CAFs in tumor stroma may cooperate for inducing the malignant transformation of CXCR4-expressing BPH-1 cells, the blockade of both the SDF-1-CXCR4 and TGF-β1 pathways by using the specific antagonists or antibodies could also represent a potential strategy to counteract PC development (Fig. 4) (45,50).

Hence, it appears that the simultaneous targeting of PC stem/progenitor cells and their further differentiated progenies as well as their local microenvironment could be more effective to counteract the PC transition to invasive and metastatic stages. Future works with the isolated and well-characterized PC stem/progenitor cells from patients’ malignant tissues or established cell lines should allow us to estimate the efficacy of novel therapeutic agents to eliminate the highly tumorigenic PC stem/progenitor cells and their differentiated progenies, and thereby to improve the current therapeutic treatments against metastatic HRPCs.

VII. Conclusions and Perspectives

Altogether, these recent investigations have revealed that normal and malignant CD133⁺/CD44⁺/α₂β₁^high-integrin/AR⁻ prostatic stem/progenitor cells can give rise to different TA/intermediate cell types with distinct proliferative and differentiating capabilities between the prostatic stem/progenitor cells and terminally differentiated secretory epithelial cells during the multiple stages associated with the prostatic tissue regeneration process and PC development, respectively. Further studies are necessary to establish and define more precisely the specific markers and functional properties of normal CD133⁺/CD44⁺/α₂β₁^high-integrin/AR⁻ prostatic stem/progenitor cells vs. their more committed progenies and malignant counterpart, PC stem/progenitor cells. The identification of the intrinsic and extrinsic factors that may regulate the self-renewal ability and/or differentiating potential of normal and malignant prostatic stem/progenitor in vitro and in vivo, and in particular, the combined effects of the activation of diverse signaling cascades on their behavior, should provide insight into the molecular events that may be involved in prostate tissue homeostasis and pathologies. It will be of interest to establish the differences between the gene expression patterns observed for normal and malignant CD133⁺ prostatic stem/progenitor cells compared with the CD133⁻ subpopulation to elucidate their unique functions in prostate physiology and pathophysiology.

In addition, the determination of the implications of AR⁻ PC stem/progenitor cells in tumor growth, metastases, dormancy phenomenon, and PC recurrence after the clinical therapies should also provide important information for overcoming treatment resistance and disease relapse. Further in vitro and in vivo characterization of functional properties of human PC stem/progenitor cells isolated from primary or secondary tumors from patients and well-established PC cell lines should lead to the establishment of new in vitro and in vivo PC cell models in which the genetic and phenotypic changes observed could be comparable to those seen during human PC initiation, progression, and metastasis in the clinics. These novel PC stem/progenitor cell models should help in identifying the deregulated gene products associated with the molecular etiology and progression of benign vs. malignant prostatic diseases as well as the novel drug targets for the development of new therapeutic options to prevent or treat the BPH and the aggressive and recurrent PCs. The in vitro and in vivo evaluation of anticarcinogenic effects induced by the current androgen deprivation, radiation, and/or chemotherapeutic treatments, alone or in combination with novel therapeutic agents targeting the AR⁻ PC stem/progenitor cells and their local microenvironment, also merits future investigation. Particularly, it will be of interest to estimate the specific implication of the tumorigenic signaling cascades induced through EGFR, hedgehog, and/or SDF-1/CXCR4 pathways, which provide a critical role in PC progression and disease recurrence, in the malignant transformation of PC stem/progenitor cells and their progenies. The establishment of the PC stem/progenitor and their progenies made resistant to specific or multiple chemotherapeutic drugs should also provide important information regarding the diverse molecular mechanisms that may be responsible for the acquisition of drug resistance after therapeutic intervention consisting of a long exposition of PC cells to drugs currently used in the clinics. These additional investigations should allow us to identify new signaling elements in PC stem/progenitor cells that are involved in treatment resistance and which could be targeted for improving the current clinical antiandrogen deprivation therapies as well as docetaxel- or mitoxantrone-based chemotherapies against metastatic HRPCs.

Hence, these further investigations should lead to the development of more effective and safe clinical treatments for patients diagnosed with locally advanced and/or metastatic HRPCs by blocking the tumor angiogenic process and eradicating the total mass of PC cells consisting of PC stem/progenitor cells and their further differentiated progenies at primary and secondary neoplasms. These novel treatments could prevent disease relapse, and thereby improve the overall survival of patients with metastatic HRPCs that remain incurable with the current therapeutic regimen options.