In normal noncancerous tissue from radical prostatectomy specimens, immunohistochemistry localizes ERK to the cytoplasm of most cells of the prostate including the epithelial, basal, and stromal cells 12. Despite the abundance of ERK it does not appear to be active in the epithelial layer of normal prostatic tissue, but up to 80% of cells in the stroma and basal layers will stain positively for phosphorylated ERK (p-ERK) within the nucleus 3. Gioeli et al also described p-ERK staining in normal prostate tissue adjacent to areas of prostate cancer and found that ERK activation was directly related to poor histologic/prognostic features 4.

Nearly all studies involving this pathway have been examined using prostate cancer cell lines. There are 40 prostate cell lines available in the ATCC catalog of both normal and cancerous tissue. The most commonly used cell lines are the androgen sensitive LNCaP cells, isolated from a cancerous supraclavicular lymph node, and the androgen insensitive cell lines DU145 and PC3, derived from brain and bone metastasis respectively. Of note, DU145 cell lines have basal ERK activation from paracrine/autocrine factors that is not demonstrated in other cell lines. Karyotypes have been described for these lines and more recently for a variety of the other cells lines available 5. Studies of kinase activation in normal prostate tissue cell lines are remarkably absent from the literature.

The most well studied ligands/mitogens in prostatic cells are epidermal-derived growth factor (EGF), transforming growth factor (TGF)-α, and insulin-like growth factor (IGF). The mechanism of action of these proteins is well described in many reviews 678. Generally, the ligands interact with a membrane receptor and transmit a signal to a cytosolic tyrosine kinase. EGF and TGF-α share about 35% sequence homology and bind to the same receptor, the epidermal growth factor receptor (EGFR). The ERK module appears to be a primary signal relay as inhibition of this activation prevents cellular proliferation induced by EGF as well as numerous other mitogens, which operate by transactivating the EGF receptor 9. There are numerous effectors of the ERK pathway in prostate cancer cells and these are described in Table 1 and 2. These stresses and agents are of considerable interest in that they are potential manipulators of this cascade.

Androgenic manipulation has been a mainstay of prostate cancer treatment for over 60 years, but the clinically challenging cancers will grow aggressively in the absence of androgens. Of considerable interest in prostate physiology is the relationship between androgens, the androgen receptor, and the MAP kinase cascade. The effect of the potent androgen dihydrotestosterone (DHT) is still unclear as it activated ERK in one study but not in numerous others with LNCaP cells 91011. Regardless of the effect on ERK, the contribution of MAPK to cellular proliferation due to DHT appears to be small relative to other pathways. With the apparent poor evidence suggesting direct androgen stimulation of ERK, the focus has been redirected on Interleukin (IL) -6 and the communication links between this important cytokine and the androgen receptor.

The interaction of IL-6 with the MAPK pathways is of particular interest for this is suspected to be a major autocrine factor in the progression of hormone refractory prostate cancer. There is an excellent review of the intracellular activities initiated by IL-6 12. IL-6 appears to be able to transactivate the androgen receptor in the absence of androgens at the N-terminal domain as well as increase the mRNA for the androgen receptor. Activation of androgen receptor by IL-6 involves the ERK pathway among others 1314. ERK also appears to be involved in the phosphorylation of steroid receptor co-activator (SRC) -1, which binds to the androgen receptor 15. LNCaP cells are also sensitive to IL-6 as an interesting study demonstrated that IL-6 exposed tumors injected into nude mice demonstrated an abrogation of proteins involved in cell cycle control. Inhibition with PD98059 was able to retard the cancer growth of these IL-6 exposed cells 16. Our preliminary studies demonstrate that IL-6 expression by PC3, a line of hormone refractory prostate cancer cells is in part regulated by ERK signal transduction pathway (Koul et al in press]

EGF is extremely important in the study of cancer. Medications that interact with this receptor are regularly used in breast malignancies and they are being examined in many other tumor types. Interference with the EGFR either through prevention of ligand binding, disruption of surface expression, or prohibition of cytosolic protein interaction can inhibit ERK activation. Multiple studies have demonstrated interaction with EGFR by one of the aforementioned means can affect prostate cancer cell growth and invasion in PC3 and DU145 cells 17. Tyrphostin AG825 and ZM252868 are two promising molecules that act in this fashion 1819. G protein coupled receptors appeared to be very important in communicating with the EGF receptor in prostate physiology 20. G proteins can transactivate the EGFR through a variety of means, which include cytosolic protein activation and metallomatrix protein pro-ligand cleavage. Numerous physiologic molecules normally activate this pathway including lysophosphatidic acid, bombesin, adenosine triphosphate (ATP), and 5-HETE 212223242526. A number of other metabolites appear to operate specifically via the metallomatrix protein pathway and these pathways can be inhibited with methyl selenium molecules 27.

Numerous cytosolic proteins are involved in the intracellular events leading to ERK activation. Among these important molecules are ras, PTEN, ID-1, DOC-2/DAB2, Protein kinase C (PKC) epsilon and some of the integrin subtypes 2829303132. One of the intracellular proteins that transmit signals from the EGF receptor to the modular MAPK pathway is ras. Isoprenylation allows ras to approach the membrane, a requirement for interaction with the EGFR. This chemical reaction is facilitated by HMG-COA reductase and the statin-family of drugs and phenyl acetate can inhibit this process 3334.

The first part of the modular MAPK cascade resulting in ERK activation is the Raf molecule, which includes three isoforms, Raf-1, RafA, and RafB. These molecules have not been extensively studied with regards to prostate cancer. One interesting study used the LNCaP cell line transfected with a constitutively active form of Raf-1, which increased ERK activity and decreased plating and cloning efficiency 35. This observation is contrary to other studies by suggesting that ERK activation may have tumor suppressor effects or perhaps chronic high level activation may have different responses. Raf kinase inhibitor protein (RKIP) is a protein that inhibits activation of MEK/ERK and appears to be a metastasis suppressor gene. Fu et al investigated clinical samples of local and metastatic prostate tissue and demonstrated immunohistochemical presence of RKIP was inversely related to histological grade. Additionally, no RKIP antigen was found in metastatic deposits. These investigators also demonstrated no effect of this protein on tumor growth, but in vitro and in vivo evidence of decreased metastasis 36. These results support results from our lab that show inhibition of p42/p44 MAP kinase affects in vitro clonogenic potential in PC3 cells. Taken together these results suggest that inhibition of this pathway may be effective in preventing metastatic deposits, but not gross tumor growth.

Immediately downstream of ERK, numerous proteins are activated that generally contribute to invasive potential, cell proliferation/differentiation, or survival in adverse environments. The continuation of vascular endothelial growth factor (VEGF) secretion despite cell contact appears to be regulated by ERK in LNCaP cell lines with inactivated focal adhesion kinases (FAK) via a ras-independent pathway 37. Peroxisome proliferators-activated receptor (PPAR) Gamma appears to be an important element in regulation of proliferation and differentiation of PC3 cells. This protein is phosphorylated (inactivated) by ERK activation, thus unshackling cell growth from this control 38. Prostate specific antigen (PSA) promotion is an ERK sensitive phenomenon in androgen independent conditions and this may explain the continuing rise in PSA in hormone refractory prostate cancer 39. A study using PC3 cells showed that prostaglandin E2 was able to promote cell survival in oxygen poor environments by increasing hypoxia inducible factor-1α (HIF-1alpha) protein levels via a pathway inhibited by PD98059 40. Another study examined fibroblast-derived growth factor (FGF)-1 stimulation in LNCaP cells and this stimulation resulted in an increase in promatrilysin via ERK/STAT3 pathways 41. This enzyme is associated with increased prostate cell invasion. Radiation exposed DU145 cells had increased production of DNA repair proteins, XRCC1 and ERCC1 by ERK related mechanisms 42. Additionally, ERK may be involved in the expression of integrins as UO126 was able to inhibit binding to the promoter region of the alpha 6-integrin gene in only PC3 cells 43. Androgen-sensitive LNCaP cells grown in an androgen deprived environment can develop neuroendocrine phenotypes in certain clones. ERK is constitutively activated in these cells via an elevated level of receptor-type protein-tyrosine phosphatase alpha. Inhibition of ERK with PD98059 prevents neuroendocrine differentiation and the increased level of neuron-specific enolase 44.

The effect of phospho-ERK on prostate cancer cells can either be one of reducing apoptosis or more commonly one of induction of cellular proliferation. Certainly, the relative activation of the ERK isoforms can have variable cellular effects, but this has not been evaluated in detail in prostate cancer. A few studies have demonstrated ERK dependent apoptosis in prostate cancer cells. Resveratrol in DU145 cells and phenylethyl isothiocyanate in PC3 cells both produce rapid phosphorylation of ERK and cause apoptosis 4546. Other MAPK pathways do not appear necessary for this apoptotic event even though they may be activated. A few other studies view more indirect evidence that ERK activation can inhibit proliferation or other similar effects. In particular, vitamin D is able to activate ERK and also has inhibitory effects on cellular proliferation, but a causal relationship was not established 47. Also, bryostatin 1 was able to induce apoptosis and Raf1/ERK activation in LNCaP cells that over express PKCα 48.

There is a substantial body of evidence supporting ERK and its effects on survival and proliferation. Both 13-(S)-HODE and 5-HETE cause PC3 cells to grow most likely by the ERK pathway 2638. Also, inhibition of the ERK pathway has resulted in either decreased survival or increased apoptosis in cancer cell lines. In particular, 4-HPR in several studies documented apoptotic effects in multiple prostate cancer cell lines and these effects were increased with the inhibition of the ERK pathway using PD98059 4950. This suggests that under situations of cellular distress ERK attempts to rescue the cell from death. Numerous other micronutrients have shown cell growth inhibition primarily through negative regulation of the ERK pathway 51. MEK1/2 inhibitors potentiate apoptosis caused by Tyrphostin AG825 and a check point abrogator UCN-01 in breast and prostate cancer cells 1852. Also, ERK inhibition is able to increase radiation induced apoptosis in DU145 cells 535455.

There is a growing body of evidence that identify ERK as an enzyme responsible for increased invasive and metastatic potential in prostate cancer. Our studies in PC3 cells demonstrate that ERK signal transduction pathway plays only a minor role in growth and proliferation of these cells, but is essential for clonogenic activity, cell migration and invasion [Koul S et al in preparation]. Our studies demonstrate that ERK signal transduction does not appear to play a role in cell growth, but is essential for new colony formation. This raises interesting aspects in modulation for oncologic control. Clearly, invasion and metastasis are the elements of tumors that make them malignant and the cause of patient suffering. However, ERK inhibition may not affect already developed metastatic sites. This might lend to early use of inhibitors or modulators of this pathway while disease burden is still low and/or possibly as prevention of tumor metastasis.

With such a large body of evidence supporting a role for ERK MAP kinase signal transduction pathway in promoting tumorigenesis in prostate cancer, we recognize that p42 / p44 MAP kinase signal transduction pathway may serve as a novel target for the treatment of prostate cancer.

The presence of active JNK in normal tissues from prostatectomy specimens is somewhat controversial, even in the same research groups 12. Most studies show that JNK expression or activation is increased in neoplastic cells 3. Additionally, JNK activity appears to be inversely related to MKP-1 expression. An interesting sidebar to the use of transgenic adenocarcinoma of mouse prostate (TRAMP) mice by Uzgare et al showed that JNK expression can increase in poorly differentiated tumors without an apparent increase in activation. JNK is generally activated by cellular stress, however, numerous molecules are able to phosphorylate the enzyme [Table 3]. 2-methoxyestradiol (ME), an endogenous metabolite of estradiol, can synergize with taxol to increase JNK activation and enhance the apoptotic effect of this chemotherapeutic agent 56. N-(4-hydroxyphenyl)retinamide (4-HPR) is a synthetic analog of all-trans retinoic acid and has shown some promise in localized and preventative breast cancer treatment. Poorly differentiated androgen-insensitive PC3 cells seem somewhat resistant to 4-HPR induced JNK phosphorylation, while androgen-sensitive LNCaP cells appear to be quite sensitive correlating with clinical behavior in breast cancer studies 5758. Numerous other agents have been studied regarding JNK activation in prostate cancer cell lines and these include the following: Ghosh et al showed that inhibition of arachidonate 5-lipoxygenase (an enzyme that converts arachidonic acid to 5-HETE) caused rapid depletion of 5-HETE and activation of JNK, which triggered apoptosis in LNCaP and PC3 cells 59. This suggests 5-HETE can be a critical cell cycle regulator as it activates ERK when abundant. The unique ability to activate JNK might not only come from antioxidants. Reactive oxygen species, in of themselves, can increase JNK in a study with multi-cellular prostate cancer spheroids 60. The phorbol esters are used widely to experimentally promote tumor growth and are well known for activation of PKC. Thapsigargin is a potent inhibitor of the sarcoplamsic/endoplasmic reticulum calcium ATPase (SERCA) pump, which results in rapid increase in intracellular calcium ion.

While JNK appears to be involved in the cross talk between many pathways there is some research into receptors that may activate JNK. In particular, the TNF-related apoptosis inducing ligand (TRAIL) receptor can activate JNK, although this effect is not necessary for apoptosis caused by this receptor suggesting other pathways are activated 6162. Certain proteins in the cell can play a roll in activating the JNK modular pathway. Prostate apoptosis response (PAR) – 4 appears to supplement JNK activity as loss of this protein caused hyperactivation of NF-kappaB and impairment of JNK and p38 63. In DU145 cells normally deficient of the Retinoblastoma (Rb) gene, reactivation of this gene is required for gamma irradiation induced JNK phosphorylation. Additionally, this study showed that mutant jun blocked radiation induced apoptosis 64. Fas also plays a role in JNK activation, as well as STE20-like kinase and Janus kinase/signal transducer and activator of transcription (JAK/STAT) 126566. Immediately upstream is from JNK is MKK 4 and 7 which have not been studied in great detail with regard to prostate cancer. However, one study showed that MKK 4 acts as one of seven genes that are metastasis suppressers without suppressing tumor growth 67. A number of proteins negatively regulate JNK. As mentioned previously, arachidonate 5-lipoxygenase which converts arachidonic acid to 5-HETE provides a substrate that appears to prevent JNK activation 59. Additionally, NF-kappaB activation is able to prevent JNK phosphorylation caused by 4-HPR 49. MAPK phosphatases have an important role in the regulation of kinase activation. Phenylethyl isothiocyanate (PEITC) increases JNK activity through the inactivation of phosphatases (M3/6) in LNCaP cells 68. Another study showed that over expression of MKP-1 and DU145 cells blocked the activation of JNK 69. JNK is able to activate numerous functions in the prostate cancer cell, which include both transcription factors and functional proteins. The most well known substrate for JNK is c-jun and JNK activation is synonymous with c-jun phosphorylation. JNK is able to activate the transcription factor AP1 70. JNK when activated can phosphorylate serine 62 on Bcl-xL, which effectively prevents this protein's anti-apoptotic effects 56. JNK is also able to activate numerous different caspases including 3, 8, and 9, as well as prevent nucleosome formation 5971. The specific activity of the different isoforms is still unknown. As lab techniques improve, undoubtedly we will have better answers. An interesting study that examined the role of JNK2 used serial analysis of gene expression (SAGE) in PC3 cells. These authors found that genes involved in DNA repair, mRNA turnover, and drug resistance were down regulated by JNK2 inactivation 72. This suggests a role for JNK2 in cell saving. Also multiple MAPKs including JNK appear to play a role in the upregulation of the urokinase-like plasminogen activator (U-PA) promoter in PC 3 cells 73. Regarding gross cellular function in the cell, authors have shown that activation of JNK regulates the cytoskeleton; prevention of nucleosome formation and mitochondrial dysfunction appear to also be major events following JNK activation 596669. JNK appears to be overwhelmingly involved in apoptotic pathways shown by multiple studies using a variety of molecules, protein and hormones to activate JNK. JNK activation has been shown to be a crucial step in the apoptosis induced by nonsteroidal anti-inflammatory drugs in human colon cancer cells 74. Hypoxia appears to be another condition in which JNK is phosphorylated. This was shown in male rats that were castrated and subsequently had the environment of the ventral prostate gland examined 75.

JNK activation is not a necessity for apoptosis as demonstrated in multiple studies using inhibitors or dominant negative cell lines. Despite overwhelming studies suggesting JNK activation and apoptosis, several well-performed studies suggest that JNK inactivation is beneficial. Different roles for JNK1 and JNK2 are of interest and poorly understood. As mentioned previously, JNK2 inactivation prevented the up regulation of genes involved in DNA repair, mRNA turnover and drug resistance 72. Other studies using anti-sense forms of JNK have revealed some interesting findings. One study has suggested that JNK is more active in growth and proliferation 76. These authors exposed human prostate cancer lines to anti-sense JNK1 and JNK2 and found that anti-sense JNK1 inhibited growth and anti-sense JNK2 inhibited proliferation. This study suggests that JNK is a potential target for prostate cancer growth. Another study reviewed the methods for sensitizing prostate cancer cells to cisplatin, by expression of p53 and anti-sense JNK 77.