I. Introduction

More than 10 % of American men older than 60 are affected by prostate cancer (PCa). Similarly PCa is highly prevalent in western countries, being the second most diagnosed malignancy after skin cancer and more importantly, the second most common cause of cancer-related male mortality (Greenlee et al, 2000). In addition, PCa incidence its directly associated with aging, meaning that with increased life expectancies, the number of PCa cases is predicted to raise over the coming years (Nelson et al, 2003). Prostate cancer is a relatively slowly progressing and indolent disease and when it remains localised it can be controlled first with surgery and later by anti-androgen therapies. However tumors reappearing after anti-androgen treatment are usually irresponsive to the existing treatments, more aggressive and prone to metastasize. Thus, androgen-independent prostate cancer (AIPC) constitutes a real life threat that accounts for the gross part of PCa mortality. Therefore, understanding the genetic basis for the progression to AIPC could contribute to the establishment of rationale alternative therapies for the later stages of this disease (Feldman BJ and Feldman D, 2001).

In adult men, the prostate gland is a small acorn-shaped tissue with ductal-acinar histology that lacks lobular organization. It is located below the bladder, surrounding the urethra and secretes proteins that are incorporated to the seminal fluid. However, it lacks a discernible role in fertility and its loss has no impact on viability. Therefore, the practical interest in studying the biology and development of the prostate is linked entirely to its pathological relevance.

Morphologically the human prostate can be divided in 3 distinct regions termed peripheral, transitional and central zones. This morphological distinction is important from the pathological perspective as prostate cancer (PCa) originates mainly from the peripheral zone, while the transition zone give rises to benign prostate hyperplasias (BPH) that are not evolutionary related to prostate cancer.

At the cellular level, the human prostate is composed of ducts and acini embedded in a stromal matrix of fibroblastic and myofibroblastic cells. In the glandular epithelium there are two predominant epithelial cell types, namely secretory luminal cells and basal epithelial cells (Bonkhoff and Remberger, 1996). Basal cells are set in the exterior part of the prostate ducts and they secrete components of the basement membrane. The luminal cells are located in the inner layer of the ducts and are involved in secreting components that are incorporated to the prostatic fluid. Interestingly, luminal cells express androgen receptor (AR), the prostate specific antigen (PSA) and are dependent on androgen for their proliferation, while basal cells are androgen-independent and do not express these markers. In addition, there is also a small proportion of a third cell type of non-epithelial origin in the prostate, neuro-endocrine cells, which precise role is not completely understood. Interestingly it is believed that a subset of the basal cells could present a stem cell-like potential giving rise to and being able to replenish the luminal cells. These human prostatic epithelial stem cells have been identified as a₂b₁^hi/CD133⁺, have a high proliferative potential in vitro and can reconstitute prostatic-like acini in nude mice (Richardson et al, 2004). The stem cell properties are underscored by the observation that hormone replacement in castrated animals results in a quick regeneration of the whole prostate gland (Sugimura et al, 1986). In addition, luminal cells are post-mitotic while basal cells are not. As in any other organ, there exists a significant interaction and crosstalk between the different cell types (epithelial and stromal) and this interaction impacts on the overall growth and development of the prostate (Liu et al, 1997).

The different cell types of the prostate can be identified by the expression of diverse cellular markers. For example, luminal cells are characterised mainly by expressing cytokeratin 8 and 18. On the other hand, basal cells express cytokeratin 5 and 14 in conjunction with other relevant markers such as CD44 (Liu et al, 1997). A more detailed analysis of cytokeratin expression in the prostate have identified different cell subpopulations expressing mixed combinations of these markers, what has been interpreted as suggestive of intermediate precursors originated during the differentiation of prostate cells (Xue et al, 1998). Further work in this direction will probably be interesting for understanding both normal prostate development and the formation and evolution of PCa lesions

III. Prostate cancer, progression and treatment

There are several abnormal pathologies of the prostate each one presenting different origins, progression and consequences. Hyperplasia on the transition zone of the prostate and a subsequent obstruction of the urethra is extremely frequent in aging men (Isaacs and Coffey, 1989). This condition is known as benign prostatic hyperplasia (BPH) and itself constitutes the second most common reason for surgery in men over 65 (Oesterling, 1995). However as we noted before, this fairly common malignancy does not constitute a precursor phase of PCa, which arises from a different region of the prostate.

Histological analysis of prostate lesions have identified a type of pre-malignant neoplastic lesions, referred as prostatic intraepithelial neoplasia (PIN) as the precursor of human prostate tumors. PIN lesions typically appear in the prostate peripheral zone and are multifocal in nature similarly to that observed with PCa (Abate-Shen and Shen, 2000). PIN has been classified in 4 grades according to their severity. Importantly, prostate ducts affected by PIN present disruption of the basal layer while still retaining an intact basement membrane, therefore impeding the stromal invasion by neoplastic cells. In contrast, invasive prostate carcinomas can invade the stromal tissue, as the basal lamina is lost from the prostate ducts. A reliable indicator of PCa development is the presence of high levels of PSA in serum. Elevated levels of PSA in serum are observed in invasive carcinomas but not in PIN (Mazzucchelli et al, 2000). The available clinical tests for measuring the levels of PSA in serum have allowed for an early detection of the disease. Therefore, intervention at earlier stages of PCa is being implemented and in the long term may contribute to improved survival. Several systems are used to classify the severity of PCa. The procedure most commonly used is known as Gleason score and was devised by Dr. Donald F. Gleason more than 30 years ago. It is based on the histological classification of haematoxylin/eosin stained prostate sections between five grades. They vary from 2 to 10 as the grade of the two more abundant patterns observed in every case is added up to calculate the Gleason score. Different studies have validated the accuracy of the Gleason score system by correlating it with patient survival or progression to metastatic state (Gleason, 1992).

A. Treatments for prostate cancer

Primary prostate cancer which remains localised to the prostate glands can be controlled by either prostatectomy or radiation therapy. The later usually removes or destroys local tumors. However, different strategies have to be applied for dealing with PCa after it invades other organs or for treating PCa tumors that reappear after the initial intervention. In these circumstances deprivation of androgens is the therapy of choice (Eisenberger et al, 1998). However it has to be considered that the effectiveness of androgen ablation in the management of advanced prostate cancer is limited. After certain time (the average being 2 years), PCa evolves to an androgen-independent disease with a poor prognosis and unfortunately with no curative therapies available (Feldman BJ and Feldman D, 2001). During the final stages, tumors progressively invade seminal vesicles and finally metastasize in other organs, primarily to the bone, resulting in osteoblastic tumors (Logothetis and Lin, 2005). Usually a lost of androgen dependency is observed at these final steps of the disease. This is partially caused by the selective pressure exerted by androgen ablation therapy.

IV. Androgen receptor signalling

Growth factors enhance proliferation and promote survival being involved in maintaining the normal homeostasis of tissues and organs. This is particularly true under physiological and pathological circumstances for the prostate as it is dependent on androgens for its normal growth and development. Also prostate cancer remains highly dependent on androgens during its initial stages. The main circulating androgen on males is testosterone that is produced in the testis and the adrenal glands (Wilson et al, 1983). Once inside prostate cells, the enzyme 5a-reductase converts it to the metabolically more active dihydrotestosterone (DHT) (Bruchovsky and Wilson, 1999). DHT act as the main physiological ligand of androgen receptor (AR) and it is much more active than testosterone, having higher affinity for the AR. The AR is a classical nuclear receptor, consisting of a ligand-binding domain, an amino terminal activating domain and two Zn-fingers that constitute the DNA-binding domain (Gao et al, 2005). Similar to other nuclear receptors, upon binding of ligand, the AR suffers a conformational change that facilitate its binding to different transcriptional co-activators that link it with the general transcriptional machinery (Hart, 2002). Many transcriptional targets of the AR have been identified, amongst them PSA being one prototypic target with a critical relevance for prostate cancer diagnosis (Kim and Coetzee, 2004). After its activation by androgens, AR can bind to DNA target sequences located on or near androgen-dependent genes, termed androgen responsive elements (AREs). Such elements are present in known androgen responsive genes such as the ones encoding for PSA or AR (Berger and Watson, 1989). Using different genomic approaches a broad set of genes that are responsive to androgens have been identified (Nelson et al, 2002).

Overall, through the modulation of multiple targets, androgens and AR mediate key processes involved in the normal development and function of the prostate (Bonkhoff and Remberger, 1996; Isaacs et al, 1992). During development, the pattern of expression of the AR dictates the androgen-responsiveness of the prostate and impact on its differentiation. In the mature prostate androgen does not only regulate the division and proliferation of prostate cells but also governs a cell death programme. In general, androgen also regulates several aspects of prostate cellular metabolism.

V. Anti-androgen therapy

The observation that castration induces regression of PCa, in a similar way as it induces involution of the prostate gland in animal models was the first suggestion that PCa relies on androgens for its maintenance and progression. Therefore, it became obvious that targeting this growth-addiction offered a therapeutic opportunity to tackle PCa progression (Huggins, 1967). Anti-androgen intervention can be implemented in different ways; one is orchiectomy (surgical removal of the testicles), but the most usual approaches involve pharmacological control.

The first of these chemical approaches consist in the use of super-agonists of the Gonadotropin releasing hormone (GnRH) (Labrie et al, 1986). They downregulate the GnRH receptors, leading to a suppressor of luteinizing hormone (LH) release, that finally causes an inhibition of testosterone secretion in the testis. Total androgen ablation or maximal androgen blockade is a more thorough approach that combines an anti-androgen together with super-agonists of GnRH (Labrie et al, 1993). In this way, not only the androgen production in the testis is blocked, but also anti-androgens can avoid the activation of the androgen receptor by the residual levels of androgen that continues to be produced in the adrenal gland.

It is important to remark that any of these strategies do not achieve a complete depletion of androgen, but result in androgen deprivation. Because of that, Scher and Sawyers proposed that the term ‘castration-resistant’ is more adequate to define these tumors (Scher and Sawyers, 2005). Overall, the fallout of these therapies is that they cannot be considered a cure of prostate cancer, but they just temporally avoid its progression for a limited time before the appearance of resistant tumors resulting in the inevitable progression of this disease. The current standard therapy for patients who have progressed on androgen deprivation is docetaxel and prednisone (Debes and Tindall, 2004). Although effective in prolonging life, this therapy is not curative and new approaches to treat AIPC are under development and consequently others should be investigated.

VI. Conversion to androgen independency

The transition of the prostate cancer cell to an androgen independent phenotype is a complex process that involves selection and outgrowth of pre-existing clones of androgen-independent cells as well as selection for genetic events that help the cancer cells survive and grow in androgen-independent conditions. These two mechanisms share a common initial point: prostate cancers are heterogeneous tumours comprised of different subpopulations of cells that therefore would respond differently to anti-androgens. This tumour heterogeneity may thus reflect a combination of their multifocal origin, ability to adapt to the environment and the genetic instability of the tumor (Abate-Shen and Shen, 2000).

Similarly to what happens with our current understanding of the genetic mechanism involved in the transition from cancer to a metastatic stage (Bernards and Weinberg, 2002), there are different theories for explaining the conversion of androgen-dependent PCa to androgen-independent prostate cancer (AIPC) (Feldman BJ and Feldman D, 2001).

The initial PCa tumors already contain a heterogeneous mix of cells with different genetic backgrounds and the selective pressure exerted by anti-androgen therapy favours the outgrowth of a population over other. One possibility is that cells resistant to androgen ablation emerge amongst androgen-resistant prostate stem cells. Although this remains a hypothesis, the recent identification of prospective prostate cancer stem cells suggests that such a scenario could be plausible (Collins et al, 2005). In any case, the anti-androgen treatment adds up to the global pressure of the tumor to outgrowth and survive and can contribute to select for cells able to proliferate independently of androgens. Thus, either these mutations happen as an early event that is selected during the evolution process, or perhaps the selection is triggered at a later stage because the mutations already conferred some advantage for tumor growth, even under androgen-dependent conditions (Abate-Shen and Shen, 2000). Therefore, one suggestion is that intermittent anti-androgen therapy could be successful in suppressing tumor growth without exerting such a strong pressure for the appearance of AIPC that would delay the progression to the more malignant stages of PCa (Abate-Shen and Shen, 2000). The fact that anti-androgen therapy exerts such a pressure for the selection of additional lesions during PCa is underscored for the diversity of mutations that can be observed upon different anti-androgen treatments, as we will describe more extensively below.

VII. Molecular pathways involved in AIPC

An overview of the signal transduction pathways triggered by androgen signaling suggests that alterations occurring at different levels could potentially result in the maintenance of androgen-independent growth. The different genetic events that are involved in AIPC progression (summarized in Table 1) can be classified taking into consideration their different ways of action (Feldman BJ and Feldman D, 2001).

A. Mechanisms dependent on AR and androgens.

1. Amplification of AR

There are several mechanisms for explaining how proliferation is sustained in a way dependent on the androgen receptor (AR) upon anti-androgen treatment or castration. Amplifications of the AR gene are observed in more than 30 % of the tumors that later will become androgen independent (Taplin and Balk, 2004). A comparison with the primary tumors from where these AIPC tumors originated shows that the AR amplification happened after treatment, what suggest that the pressure exerted by anti-androgen therapy probably allows for a selection of pre-existing clones in which the AR gene had been amplified (Visakorpi et al, 1995a). An interesting observation is that tumors with the AR gene amplified present a better outcome that the ones without AR amplifications (Koivisto et al, 1997).

Although we termed the tumors that present amplification of the AR gene as androgen independent, probably is more accurate name them as hypersensitive to low concentrations of androgen, as they still depend of the minimum amounts of androgen present in the organism to proliferate. This is consistent with the proposed nomenclature of ‘castration-resistant’ tumors proposed by Scher and Sawyers (Scher and Sawyers, 2005). Therefore, it can be suggested that total androgen ablation may be an advantage respect to monotherapy as it will reduce even further the effective levels of circulating androgens in blood (Palmberg et al, 2000).

Interestingly, an upregulation of the AR mRNA levels was the only consistent change observed in a set of isogenic prostate cancer xenograft models after progression to androgen independency (Chen et al, 2004). The conclusion inferred from this study is that even a subtle upregulation of AR mRNA levels causes a state of increased sensitivity to low levels of circulating androgens. Similar results could be predicted if increased expression of AR co-activators is achieved either because of their gene amplification or because their transcriptional upregulation.

2. Production of increased levels of androgen.

There are some hints that under anti-androgen therapy despite a sharp decrease of overall androgen in blood, the availability of DHT does not decrease to the similar extent (Labrie et al, 1986). One of the possible explanations is a compensatory regulatory loop that accounts for an increased conversion of testosterone to DHT. Interestingly polymorphisms in the 5a-reductase gene had been reported and some of these polymorphisms have been linked to a hyperactive enzyme being produced. These polymorphisms are observed more frequently in men of African origin that present an increased risk of PCa (Makridakis et al, 1997). No association of these mutations with AIPC conversion has been described, but the predisposition of individuals carrying these polymorphisms cannot be ignored.

B. Mechanisms dependent on AR but independent on androgens

1. AR sensitive to activation by alternative ligands

As we have noted, AR expression is conserved in many of the AIPC cases (Buchanan et al, 2001). In a significant subset of these cases (the exact percentage varying from report to report), gain of function mutations in the ligand-binding domain of the AR are observed. However, it seems clear that these mutations are selected as a consequence of anti-androgen treatment.

These AR gain of function mutations were first described in LNCaP cells (Veldscholte et al, 1994). LNCaP cells were derived from a metastasis to the lymph node of a patient with prostate cancer that had been treated with the anti-androgen flutamide (Horoszewicz et al, 1980). Their AR presents a mutation resulting in the substitution of threonine for alanine at position 877 (T877A).

As a result the AR mutant present in LNCaP cells becomes susceptible to be activated by ligands different of DHT. Notably flutamide, that it is used as an anti-androgen or androgen antagonist during therapy, behaves as an activator of the T877A AR (Horoszewicz et al, 1980). In this way, flutamide fuels the progression of PCa once this mutation on the AR is selected. Interestingly, the same AR mutation is observed in a number of different cases of AIPC. Also there is a subset of patients in clinic that present a decrease of the PSA levels after flutamide withdrawal, what would be consistent with flutamide upregulating the expression of PSA, an AR-dependent gene (Horoszewicz et al, 1980). Interestingly however, LNCaP cells which growth is stimulated by flutamide by virtue of the T877A mutations, still remain sensitive to casodex, other anti-androgen (Horoszewicz et al, 1980). This has been explained in structural studies examining the binding site of both drugs in the AR molecule (Horoszewicz et al, 1980). The T877A mutation also makes the AR sensitive to other ligands such as estrogens and progesterone. In addition to the T877A mutation a whole set of other AR mutations have been identified and catalogued (http://www.androgendb.mcgill.ca).

Some of these mutations in the AR gene have also been found to make the AR responsive to different ligands. For example a double mutation in the AR gene (T877A together with L701H) results in the AR being susceptible to activation by glucocorticoids (Zhao et al, 2000). Finally, AR mutations being selected during anti-androgen treatment can also be mimicked in animal models of AIPC conversion, further underscoring their importance.

2. Activation of AR independently of ligand

It has been suggested that alterations in proteins that act in conjunction with AR in transcriptional control could be involved in the progression to AIPC (Adachi et al, 2000). However, the existing evidence comes only from cell culture data. Perhaps one of the best indications that this could be meaningful comes from the fact that mutations on the AR affecting to residues involved in their interaction with co-activators are also involved in AIPC.

Based on work performed in other tumor types (specially in lymphoma) it has been proposed that the complex tumor suppressor and oncogene networks altered during cancer progression also influence the responses to cancer therapies (Schmitt et al, 2002). Thus, it can be envisioned that mutations that happen in tumors and result positively selected because they can favour tumor progression may impact in AIPC.

In this sense the AR pathway is inter-connected with other signaling transduction pathways and therefore alterations in those pathways would result in a modification of AR function that could provoke AIPC. For example, other growth survival pathways driven by extracellular receptors have been found to crosstalk and thus sustain AR-signaling. In that sense different evidence has been provided for the ability to cause AIPC by growth factors such as epidermal growth factor (EGF), insulin-like growth factor 1 (IGF-1) and keratinocyte growth factor (KGF) among others (Culig et al, 1994). Probably IGF-1 is the best studied example, but interestingly it is known that all of them, IGF-1, KGF and EGF act upstream of AR signaling, as the androgen independent effects they exert can be blocked by anti-androgens such as casodex. Conversely, this would downplay the effect that they can have during AIPC in conditions of complete anti-androgen therapy.

A clearer connection has been identified between the pathways triggered by receptor tyrosine kinase and AR signaling. Similar results have also been obtained for studies performed in other endocrine-dependent tumors (such as breast or ovarian cancers). The prototypic case of a growth factor receptor involved in hormone-independency is Her2/neu (Craft et al, 1999). Her2/neu is over-expressed or amplified in a subset of both breasts and ovarian tumours. Those studies have provided the foundation for developing agents interfering with Her2/neu signaling and their use in cancer control (Allen, 2002). Notably, these drugs have already been used successfully against non-small-cell lung cancer and certain metastatic breast tumours and are being applied now in the clinic.

Interestingly, studies performed on AIPC cell lines derived from xenografts from castrated mice did consistently show an upregulation of Her2/neu expression, what suggest that in AIPC operates a similar mechanism to that operating in breast and ovarian cancer (Craft et al, 1999). Forced over-expression of Her2/neu recapitulates those observations, thus suggesting that is sufficient for conversion to AIPC. Her2/neu growth promotion under AIPC conditions is also dependent on AR signaling, but differently to what it has been shown for IGF-1 and EGF, casodex cannot inhibit this process, therefore suggesting that this pathway is independent on the AR-ligand binding domain (Craft et al, 1999; Yeh et al, 1999). These results hint that, similarly to what already is in practise in the clinics for breast tumors, inhibition of Her2/neu signaling (through the use of antagonist antibodies, such as herceptin) could help to treat PCa. Also, perhaps this therapy should be used in combination with chemotherapeutic agents such as it is used in metastatic breast cancer (Slamon et al, 2001). Preliminary work performed in prostate cancer xenografts and cells suggested that indeed this could be the case (Agus et al, 1999). However, there is no current clinical data to support the use of drugs targeting Her2/neu for prostate cancer (Gross et al, 2004).

The crosstalk between the Her2/neu pathway and AR signaling ultimately results in the phosphorylation of AR what impacts into its inappropriate activation. As a result, it can be hypothesized that the components lying in the interphase between both pathways could drive AIPC conversion when its regulation is lost. The pathway has begun to be unveiled and involves at least MAPK signaling (Yeh et al, 1999). Additionally the phosphatidylinositol-3-OH kinase (PI3K) pathway is also activated by Her2/neu (Yakes et al, 2002) and it is also implicated in cross-talking with the AR (Lin et al, 2001). In fact, activation of the PI3K/AKT pathway can drive AIPC in a manner that is independent on both AR and androgens, as we will discuss below.

C. Mechanisms independent on AR and androgens

1. Activation of parallel survival pathways

An alternative mechanism for progression to AIPC is based on the induction of a positive growth signal independent on the AR that can overcome the growth inhibition imposed by anti-androgen therapies, thus establishing what has been defined as a bypass pathway (Feldman BJ and Feldman D, 2001).

A role for the PI3K/AKT pathway in PCa had been already suggested through the analysis of mutations happening in cancer. Particularly, the tumor suppressor PTEN, was first identified because it was found altered in prostate amongst other types of tumors (Li et al, 1997). The PI3K pathway integrates receptor tyrosine kinase signaling with the apoptotic network. One key mediator of PI3K signaling is the protein kinase AKT, that phosphorylates multiple downstream effectors causing deep changes in cellular physiology (Cantley, 2002). How AKT promotes survival is not completely understood, but it may involve direct phosphorylation of apoptotic regulators, increased cell cycle progression, decreased transcription of pro-apoptotic genes through inhibition of forkhead transcription factors, altered metabolism, or changes in mRNA translation that ultimately impact on cell death. Mutations that activate this signaling route, include amplifications of components of the PI3K pathway and the inactivation of the negative regulator lipid phosphatase PTEN, are common in human malignancies and amongst them PCa (Vivanco and Sawyers, 2002).

The PI3K pathway is engaged and activated by multiple extracellular receptors. PI3K is a lipid kinase that phosphorylates phosphatidyl inositols. This phosphorylation process is antagonized by PTEN that is a lipid phosphatase that removes the 3-phosphate from 3-phosphorylated inositol lipids. The phosphatidyl inositol tri-phosphate lipids act as second messengers and activate AKT. AKT is a protein kinase that once activated controls multiple targets involved in cell cycle progression or apoptosis survival amongst others. Among the substrates that AKT phosphorylates are the CDK inhibitor p27^KIP1, the apoptotic proteins BAD, AR, FOXO and many others (Datta et al, 1999; Reed, 2002).

Experiments analysing androgen-independent xenografts have shown increased levels of AKT activity (Graff et al, 2000). Interestingly, the ability of AKT to phosphorylate AR could be linked to this androgen independency (Lin et al, 2001). In addition, as the PI3K/AKT pathway is a critical regulator of apoptosis, it may be thought that part of the effect exerted by this gene in AIPC could be through the regulation of apoptosis by modulating independent parallel pathways.

Androgen, as we mentioned before, drives prostate cells into active cell cycle progression. Conversely androgen withdrawal results in a combination of a tight cell cycle arrest and apoptosis. Consequently, it can be thought that if alternative survival pathways are activated, the anti-proliferative effects caused by androgen withdrawal can be suppressed either partially or totally. Particularly, overexpression of apoptosis modulators may be an obvious target to improve cell survival. One of the first genes suggested to play such a role was Bcl2, that under normal circumstances has its expression restricted to basal cells that do not dependent on androgens. Conversely Bcl2 is not expressed in luminal cells that are androgen-dependent (McDonnell et al, 1992). Overexpression of Bcl2 is observed in the initial stages previous to PCa such as in PIN. Interestingly, aberrant expression of Bcl2 has been linked to hormone-independent prostate carcinomas (Gleave et al, 1999) and specifically Bcl2 is expressed during AIPC conversion while the parental tumors were showed not to express initially Bcl2. Presumably other modulators of the apoptotic network could have similar effects in driving AIPC-progression.

2. Activation of downstream pathways

Previous work has suggested the involvement of c-myc in prostate cancer and particularly in the progression to AIPC (Nupponen et al, 1998). In this case, the mechanism by which c-myc drives the progression of AIPC relies on the activation of pathways located downstream of the AR. Different studies had narrowed down a common amplicon detected during the conversion to AIPC to a short region spanning the chromosome 8q and containing the c-myc gene (Visakorpi et al, 1995b; Nupponen et al, 1998).

Fluorescence in situ hybridization have shown amplification of the c-myc gene in more than 70% of AIPCs (Nupponen et al, 1998) and a significant increase of c-myc amplification is observed after anti-androgen treatment (Kaltz-Wittmer et al, 2000). However, the mechanism used by c-myc to drive AIPC progression was not clear.

Recently we examined the effect of c-myc in AIPC progression by using LNCaP cells treated with the anti-androgen casodex (Bernard et al, 2003). In that work, we showed that overexpression of c-myc was sufficient to induce androgen-independent growth in casodex treated cells. Our data suggested that c-myc did not act through an increase of AR activity because c-myc did not alter the levels of AR-dependent genes such as PSA or PSMA. Functional data further confirm these results as neuroendocrine differentiation induced by AR inhibition (Ahlgren et al, 2000; Ismail et al, 2002; Wright et al, 2003) was not overcome by c-myc expression. In addition, AR silencing in c-myc–expressing cells did not prevent cell growth. More interestingly, what our results suggest is that c-myc is indeed a downstream target of AR, as the c-myc protein level is regulated by AR activity and c-myc is required for androgen-dependent growth (a schema depicting c-myc mode of action is presented in Figure 1). The precise mechanism of c-myc regulation by AR is not known, but our data suggest that it can happen at a posttranscriptional level.

We have also demonstrated that c-myc expression could immortalize human primary prostate epithelial cells by overriding the p16^INK4a/Rb pathway and inducing hTERT expression (Gil et al, 2005). The fact that c-myc can be activated by AR activity (Quarmby et al, 1987) supports a role for c-myc in the early stages of prostate cancer. Accordingly, it was demonstrated that directed c-myc expression in the prostate induces development of a prostatic intraepithelial neoplasia in the mouse (Zhang et al, 2000). Therefore, during PCa c-myc act probably as a pleiotropic oncogene that becomes activated or overexpressed and gives growth advantage to early tumors, but later on the progress of the tumor is also able to confer androgen independency.

During AIPC, c-myc is needed for AR-induced proliferation because it controls the activity of major growth related proteins. Consequently, without AR activity, but with constitutive expression of c-myc, the cells grow regardless of AR signaling, even if AR expression is maintained. In addition c-myc is an essential gene which deletion impairs the growth of both androgen-dependent and -independent cell lines. Therefore, in addition to the classical hypotheses of increased AR signaling or establishment of a bypass pathway (Feldman BJ and Feldman D, 2001), c-myc may induce androgen-independent growth through the activation of a downstream pathway. Indeed, we have shown that c-myc is regulated by AR and is required for AR-dependent as well as -independent growth, suggesting that c-myc may be involved in the development of AIPC, including that resulting from an increase of AR signaling.

IX. Conclusions

The hypothesis that mutations which drive cancer progression could also have a profound impact on the responses to therapies have already been proved for other types of oncogenic lesions (Schmitt et al, 2000). In this context, PCa, together with other hormone dependent cancers such as breast cancer and ovarian cancer constitute unique types as their growth rely heavily on hormones and the hormone dependency is also targeted during the anticancer therapy. Because of that, a better knowledge of the molecular mechanisms that can permit growth under anti-androgen conditions could result in the adoption of novel therapeutic approaches, either targeting genes identified through this process, or using our improved knowledge of the molecular events involved in AIPC to implement combined therapies. Also the opposite road can be devised, if any of the novel therapeutic approaches (either conventional or targeted) that are being developed for other tumor types are successful in treating PCa, the molecular pathways underlying the mechanism of action of the drugs can be investigated in the context of PCa. Overall, a better molecular knowledge will contribute to better treatments for avoiding this mortal stage of PCa, which incidence is predicted to scale in the coming years due to increased aging.

Acknowledgements

Jesús Gil is supported by the Medical Research Council.

References

Abate-Shen C and Shen MM (2000) Molecular genetics of prostate cancer. Genes Dev 14, 2410-2434.

Adachi M, Takayanagi R, Tomura A, Imasaki K, Kato S, Goto K, Yanase T, Ikuyama S, Nawata H (2000) Androgen-insensitivity syndrome as a possible coactivator disease. N Engl J Med 343, 856-862.

Agus DB, Scher HI, Higgins B, Fox WD, Heller G, Fazzari M, Cordon-Cardo C, Golde DW (1999) Response of prostate cancer to anti-Her-2/neu antibody in androgen-dependent and -independent human xenograft models. Cancer Res 59, 4761-4764.

Ahlgren G, Pedersen K, Lundberg S, Aus G, Hugosson J and Abrahamsson PA (2000) Regressive changes and neuroendocrine differentiation in prostate cancer after neoadjuvant hormonal treatment. Prostate 42, 274-279.

Allen TM (2002) Ligand-targeted therapeutics in anticancer therapy. Nat Rev Cancer 2, 750-763.

Berger FG and Watson G (1989) Androgen-regulated gene expression. Annu Rev Physiol 51, 51-65.

Bernard D, Pourtier-Manzanedo A, Gil J and Beach DH (2003) Myc confers androgen-independent prostate cancer cell growth. J Clin Invest 112, 1724-1731.

Bernards R and Weinberg RA (2002) A progression puzzle. Nature 418, 823.

Bonkhoff H and Remberger K (1996) Differentiation pathways and histogenetic aspects of normal and abnormal prostatic growth, a stem cell model. Prostate 28, 98-106.

Bruchovsky N and Wilson JD (1999) Discovery of the role of dihydrotestosterone in androgen action. Steroids 64, 753-759.

Buchanan G, Greenberg NM, Scher HI, Harris JM, Marshall VR and Tilley WD (2001) Collocation of androgen receptor gene mutations in prostate cancer. Clin Cancer Res 7, 1273-1281.

Cantley LC (2002) The phosphoinositide 3-kinase pathway. Science 296, 1655-1657.

Chen CD, Welsbie DS, Tran C, Baek SH, Chen R, Vessella R, Rosenfeld MG, Sawyers CL (2004) Molecular determinants of resistance to antiandrogen therapy. Nat Med 10, 33-39.

Collins AT, Berry PA, Hyde C, Stower MJ and Maitland NJ (2005) Prospective identification of tumorigenic prostate cancer stem cells. Cancer Res 65, 10946-10951.

Craft N, Shostak Y, Carey M and Sawyers CL (1999) A mechanism for hormone-independent prostate cancer through modulation of androgen receptor signaling by the HER-2/neu tyrosine kinase. Nat Med 5, 280-285.

Culig Z, Hobisch A, Cronauer MV, Radmayr C, Trapman J, Hittmair A, Bartsch G, Klocker H (1994) Androgen receptor activation in prostatic tumor cell lines by insulin-like growth factor-I, keratinocyte growth factor and epidermal growth factor. Cancer Res 54, 5474-5478.

Datta SR, Brunet A and Greenberg ME (1999) Cellular survival, a play in three Akts. Genes Dev 13, 2905-2927.

Debes JD, Tindall DJ (2004) Mechanisms of androgen-refractory prostate cancer. N Engl J Med 351, 1488-1490.

Eisenberger MA, Blumenstein BA, Crawford ED, Miller G, McLeod DG, Loehrer PJ, Wilding G, Sears K, Culkin DJ, Thompson IM Jr, Bueschen AJ, Lowe BA (1998) Bilateral orchiectomy with or without flutamide for metastatic prostate cancer. N Engl J Med 339, 1036-1042.

Feldman BJ and Feldman D (2001) The development of androgen-independent prostate cancer. Nat Rev Cancer 1, 34-45.

Gao W, Bohl CE and Dalton JT (2005) Chemistry and structural biology of androgen receptor. Chem Rev 105, 3352-3370.

Gil J, Kerai P, Lleonart M, Bernard D, Cigudosa JC, Peters G, Carnero A, Beach D (2005) Immortalization of primary human prostate epithelial cells by c-Myc. Cancer Res 65, 2179-2185.

Gleason DF (1992) Histologic grading of prostate cancer, a perspective. Hum Pathol 23, 273-279.

Gleave M, Tolcher A, Miyake H, Nelson C, Brown B, Beraldi E, Goldie J (1999) Progression to androgen independence is delayed by adjuvant treatment with antisense Bcl-2 oligodeoxynucleotides after castration in the LNCaP prostate tumor model. Clin Cancer Res 5, 2891-2898.

Graff JR, Konicek BW, McNulty AM, Wang Z, Houck K, Allen S, Paul JD, Hbaiu A, Goode RG, Sandusky GE, Vessella RL, Neubauer BL (2000) Increased AKT activity contributes to prostate cancer progression by dramatically accelerating prostate tumor growth and diminishing p27Kip1 expression. J Biol Chem 275, 24500-24505.

Greenlee RT, Murray T, Bolden S and Wingo PA (2000) Cancer statistics. CA Cancer J Clin 50, 7-33.

Gross ME, Jo S and Agus DB (2004) Update on Her-kinase-directed therapy in prostate cancer. Clin Adv Hematol Oncol 2, 53-64.

Hart SM (2002) Modulation of nuclear receptor dependent transcription. Biol Res 35, 295-303.

Horoszewicz JS, Leong SS, Chu TM, Wajsman ZL, Friedman M, Papsidero L, Kim U, Chai LS, Kakati S, Arya SK, Sandberg AA (1980) The LNCaP cell line--a new model for studies on human prostatic carcinoma. Prog Clin Biol Res 37, 115-132.

Huggins C (1967) Endocrine-induced regression of cancers. Science 156, 1050-1054.

Isaacs JT and Coffey DS (1989) Etiology and disease process of benign prostatic hyperplasia. Prostate Suppl 2, 33-50.

Isaacs JT, Lundmo PI, Berges R, Martikainen P, Kyprianou N and English HF (1992) Androgen regulation of programmed death of normal and malignant prostatic cells. J Androl 13, 457-464.

Ismail AH, Landry F, Aprikian AG and Chevalier S (2002) Androgen ablation promotes neuroendocrine cell differentiation in dog and human prostate. Prostate 51, 117-125.

Kaltz-Wittmer C, Klenk U, Glaessgen A, Aust DE, Diebold J, Lohrs U, Baretton GB (2000) FISH analysis of gene aberrations (MYC, CCND1, ERBB2, RB and AR) in advanced prostatic carcinomas before and after androgen deprivation therapy. Lab Invest 80, 1455-1464.

Kim J and Coetzee GA (2004) Prostate specific antigen gene regulation by androgen receptor. J Cell Biochem 93, 233-241.

Koivisto P, Kononen J, Palmberg C, Tammela T, Hyytinen E, Isola J, Trapman J, Cleutjens K, Noordzij A, Visakorpi T, Kallioniemi OP (1997) Androgen receptor gene amplification, a possible molecular mechanism for androgen deprivation therapy failure in prostate cancer. Cancer Res 57, 314-319.

Labrie F, Belanger A, Dupont A, Luu-The V, Simard J and Labrie C (1993) Science behind total androgen blockade, from gene to combination therapy. Clin Invest Med 16, 475-492.

Labrie F, Dupont A, Belanger A, St-Arnaud R, Giguere M, Lacourciere Y, Emond J, Monfette G (1986) Treatment of prostate cancer with gonadotropin-releasing hormone agonists. Endocr Rev 7, 67-74.

Li J, Yen C, Liaw D, Podsypanina K, Bose S, Wang SI, Puc J, Miliaresis C, Rodgers L, McCombie R, Bigner SH, Giovanella BC, Ittmann M, Tycko B, Hibshoosh H, Wigler MH, Parsons R (1997) PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast and prostate cancer. Science 275, 1943-1947.

Lin HK, Yeh S, Kang HY and Chang C (2001) Akt suppresses androgen-induced apoptosis by phosphorylating and inhibiting androgen receptor. Proc Natl Acad Sci U S A 98, 7200-7205.

Liu AY, True LD, LaTray L, Nelson PS, Ellis WJ, Vessella RL, Lange PH, Hood L, van den Engh G (1997) Cell-cell interaction in prostate gene regulation and cytodifferentiation. Proc Natl Acad Sci U S A 94, 10705-10710.

Logothetis CJ and Lin SH (2005) Osteoblasts in prostate cancer metastasis to bone. Nat Rev Cancer 5, 21-28.

Makridakis N, Ross RK, Pike MC, Chang L, Stanczyk FZ, Kolonel LN, Shi CY, Yu MC, Henderson BE, Reichardt JK (1997) A prevalent missense substitution that modulates activity of prostatic steroid 5a-reductase. Cancer Res 57, 1020-1022.

Mazzucchelli R, Colanzi P, Pomante R, Muzzonigro G and Montironi R (2000) Prostate tissue and serum markers. Adv Clin Path 4, 111-120.

McDonnell TJ, Troncoso P, Brisbay SM, Logothetis C, Chung LW, Hsieh JT, Tu SM, Campbell ML (1992) Expression of the protooncogene bcl-2 in the prostate and its association with emergence of androgen-independent prostate cancer. Cancer Res 52, 6940-6944.

Nelson PS, Clegg N, Arnold H et al (2002) The program of androgen-responsive genes in neoplastic prostate epithelium. Proc Natl Acad Sci U S A 99, 11890-11895.

Nelson WG, De Marzo AM and Isaacs WB (2003) Prostate cancer. N Engl J Med 349, 366-381.

Nupponen NN, Kakkola L, Koivisto P and Visakorpi T (1998) Genetic alterations in hormone-refractory recurrent prostate carcinomas. Am J Pathol 153, 141-148.

Oesterling JE (1995) Age-specific reference ranges for serum prostate-specific antigen. Can J Urol 2, 23-29.

Palmberg C, Koivisto P, Kakkola L, Tammela TL, Kallioniemi OP and Visakorpi T (2000) Androgen receptor gene amplification at primary progression predicts response to combined androgen blockade as second line therapy for advanced prostate cancer. J Urol 164, 1992-1995.

Quarmby VE, Beckman WC Jr., Wilson EM and French FS (1987) Androgen regulation of c-myc messenger ribonucleic acid levels in rat ventral prostate. Mol Endocrinol 1, 865-874.

Reed SI (2002) Keeping p27(Kip1) in the cytoplasm, a second front in cancer's war on p27. Cell Cycle 1, 389-390.

Richardson GD, Robson CN, Lang SH, Neal DE, Maitland NJ and Collins AT (2004) CD133, a novel marker for human prostatic epithelial stem cells. J Cell Sci 117, 3539-3545.

Scher HI and Sawyers CL (2005) Biology of progressive, castration-resistant prostate cancer: directed therapies targeting the androgen-receptor signaling axis. J Clin Oncol 23, 8253-8261.

Schmitt CA, Fridman JS, Yang M, Lee S, Baranov E, Hoffman RM, Lowe SW (2002) A senescence program controlled by p53 and p16INK4a contributes to the outcome of cancer therapy. Cell 109, 335-346.

Schmitt CA, Wallace-Brodeur RR, Rosenthal CT, McCurrach ME and Lowe SW (2000) DNA damage responses and chemosensitivity in the E mu-myc mouse lymphoma model. Cold Spring Harb Symp Quant Biol 65, 499-510.

Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, Fleming T, Eiermann W, Wolter J, Pegram M, Baselga J, Norton L (2001) Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344, 783-792.

Sugimura Y, Cunha GR and Donjacour AA (1986) Morphological and histological study of castration-induced degeneration and androgen-induced regeneration in the mouse prostate. Biol Reprod 34, 973-983.

Taplin ME and Balk SP (2004) Androgen receptor, a key molecule in the progression of prostate cancer to hormone independence. J Cell Biochem 91, 483-490.

Veldscholte J, Berrevoets CA and Mulder E (1994) Studies on the human prostatic cancer cell line LNCaP. J Steroid Biochem Mol Biol 49, 341-346.

Visakorpi T, Hyytinen E, Koivisto P, Tanner M, Keinanen R, Palmberg C, Palotie A, Tammela T, Isola J, Kallioniemi OP (1995a) In vivo amplification of the androgen receptor gene and progression of human prostate cancer. Nat Genet 9, 401-406.

Visakorpi T, Kallioniemi AH, Syvanen AC, Hyytinen ER, Karhu R, Tammela T, Isola JJ, Kallioniemi OP (1995b) Genetic changes in primary and recurrent prostate cancer by comparative genomic hybridization. Cancer Res 55, 342-347.

Vivanco I and Sawyers CL (2002) The phosphatidylinositol 3-Kinase AKT pathway in human cancer. Nat Rev Cancer 2, 489-501.

Wilson JD, Griffin JE, George FW and Leshin M (1983) The endocrine control of male phenotypic development. Aust J Biol Sci 36, 101-128.

Wright ME, Tsai MJ and Aebersold R (2003) Androgen receptor represses the neuroendocrine transdifferentiation process in prostate cancer cells. Mol Endocrinol 17, 1726-1737.

Xue Y, Smedts F, Debruyne FM, de la Rosette JJ and Schalken JA (1998) Identification of intermediate cell types by keratin expression in the developing human prostate. Prostate 34, 292-301.

Yakes FM, Chinratanalab W, Ritter CA, King W, Seelig S and Arteaga CL (2002) Herceptin-induced inhibition of phosphatidylinositol-3 kinase and Akt Is required for antibody-mediated effects on p27, cyclin D1 and antitumor action. Cancer Res 62, 4132-4141.

Yeh S, Lin HK, Kang HY, Thin TH, Lin MF and Chang C (1999) From HER2/Neu signal cascade to androgen receptor and its coactivators, a novel pathway by induction of androgen target genes through MAP kinase in prostate cancer cells. Proc Natl Acad Sci U S A 96, 5458-5463.

Zhang X, Lee C, Ng PY, Rubin M, Shabsigh A and Buttyan R (2000) Prostatic neoplasia in transgenic mice with prostate-directed overexpression of the c-myc oncoprotein. Prostate 43, 278-285.

Zhao XY, Malloy PJ, Krishnan AV, Swami S, Navone NM, Peehl DM, Feldman D (2000) Glucocorticoids can promote androgen-independent growth of prostate cancer cells through a mutated androgen receptor. Nat Med 6, 703-706.

AppleMark

Jesús Gil

Review Article

Received: 11 January 2006; Revised: 22 February 2006

Accepted: 23 March 2006; electronically published: April 2006