Abstract
Defining the mechanisms that confer metastatic ability on cancer cells is an important goal towards prevention of metastasis. A gene array screen between a non-metastatic prostate cancer cell and its metastatic derivative line revealed decreased expression of Raf kinase inhibitor protein (RKIP) in the metastatic cell line. This finding is consistent with the possibility that loss of RKIP is associated with metastasis. RKIP is expressed in many tissues including brain, lung, and liver. RKIP blocks Raf-induced phosphorylation of MEK. In addition to its modulation of Raf signaling, RKIP modulates both G-protein signaling and NF-kappaB activity. The impact that RKIP has on multiple signaling pathways grants it the ability to play a role in several cellular functions including membrane biosynthesis, spermatogenesis, and neural signaling. Novel cellular functions for RKIP continue to be identified, several of which contribute to cancer biology. For example, RKIP promotes apoptosis of cancer cells, which suggests that loss of RKIP in cancer will protect cancer cells against cell death. Additionally, restoration of RKIP expression in a metastatic prostate cancer cell line does not effect primary tumor growth, but it does inhibit prostate cancer metastasis. These parameters identify RKIP as a metastasis suppressor gene, which suggest that it or proteins it interacts are putative molecular targets to control metastasis. These findings are supported by the observation that RKIP expression is decreased in metastases of prostate cancer patients, compared to normal prostate or the primary prostate tumor. In this review, RKIP biology and its role in cancer will be described.

Keywords: prostate carcinoma, metastasis, signal transduction, apoptosis, PEBP

1. Introduction
Raf kinase inhibitor protein (RKIP), a member of the phosphatidyl ethanolamine binding protein (PEBP) family, is a small, cytosolic protein originally purified from bovine brain [1-3]. The RKIP family of proteins is highly conserved and does not share significant homology with any other protein family [4]. RKIP was initially identified as human PEBP where it was shown to have a role in lipid metabolism and phospholipid membrane biogenesis. Recent investigations have identified that RKIP inhibits Raf-mediated activation of MEK, which accounts for it current name. Aberrant RKIP activity is associated with an increasing number of diseases via its association with signal transduction pathways [1, 3-9]. In this mini-review, the basic biology of RKIP and its role in cancer biology is discussed.

2. Tissue Localization
RKIP is synthesized locally in many tissues where it has been shown to be present in the cytoplasm and at the plasma membrane as determined by immunohistochemical staining [1]. While RKIP homologues can be found in testicular and epididymal luminal secretions, it is not found in blood, saliva, milk, uterine fluid, parotid fluid, prostate secretions, or seminal vesicle secretions [10].

Rat RKIP expression has been found in oligodendrocytes and Schwann cells of the neuronal tissue; spermatids, Leydig cells, and epididymal epithelium of the testis; steroidogenic cells of the adrenal gland zona fasciculata; proximal kidney tubule epithelium; enterocytes, goblet cells, and plasma cells of the small intestine; plasma cells of the lymph node; plasma cells and megakaryocytes of the spleen; heart; liver; and epididymis [11]. Some expression has also been found in bronchioles of the lung, mesenteric lymph node, oviduct, ovary, lactating mammary glands, uterus, and thyroid. RKIP expression has also been found in normal tissue and nonmetastatic prostate cancer cells, but is expressed weakly in metastatic prostate cancer cells [12].

3. Functions of RKIP
RKIP appears to have a variety of functions depending on the tissue in which it is localized. Several lines of evidence suggest that it is involved with mammalian spermatogenesis and male fertility. For example, rat epididymal secretions and sperm plasma membranes contain proteins with sequences similar to bovine brain RKIP [13]. Furthermore, RKIP released from spermatozoa may be involved with membrane biogenesis and maintenance of antigen segregation in spermatozoa [14]. Studies in the rat testis show that RKIP proteins may be involved in organization of the seminiferous epithelium or the transfer of phosphatidylethanolamine to other germ cells [15]. Due to its presence in Leydig cells, Frayne, et al. suggested a role for RKIP as a lipid carrier or binding protein within the rat testis that contributes to membrane organization during spermatogenesis [16].

Although RKIP is expressed in multiple tissues of the rat, higher expression levels can be found in the testis, brain oligodendricytes, Schwann cells, Purkinje cells of the cerebellum and within cortical and hippocampal layers of the brain [11, 16-20]. In rat medial septal nuclei, RKIP was found to enhance in vitro acetylcholine synthesis by upregulating choline acetyltransferase and possibly stimulating cholinergic neuronal pathways [7, 20-22].

RKIP interacts with small GTP-binding proteins, yet not GTP itself [23] and can be purified along with µ opioid receptors via morphine affinity chromatography using tissue derived from rat brain [24]. Grandy, et al., speculated that RKIP was a membrane-associated protein which may alter opioid binding via an enzymatic- or structural-induced reaction [24]. Using hydrophobic cluster analysis and molecular modeling, Schoentgen et al. showed that the bovine RKIP may [25] possess a potential nucleotide binding site and suggested that it may belong to the kinase family and promote the transfer of hydrophobic ligands to the plasma membrane [26]. Co-expression of human RKIP with human opioid or somatastatin receptors (G-protein-coupled receptors) in Xenopus laevis oocytes provided in vivo evidence that RKIP could modulate G-protein-coupled signaling [27]. These studies, along with its widespread distribution in tissues and multiple species, provide evidence that RKIP is involved with cell regulatory and cell signaling mechanisms.

The role of RKIP in cell signaling was identified in a yeast two-hybrid assay for screening clones from a human T-cell library that bound to Raf-1 kinase binding domains [3]. RKIP was shown to bind Raf-1, MEK-1 and weakly bind to ERK-2, interfering with MEK phosphorylation and activation by Raf-1. However, RKIP was not a substrate for Raf-1 or MEK. RKIP did not bind to Ras, nor possess kinase activity. It appears that RKIP acts to set the threshold for Raf-1 activation and subsequent activation of the MEK/ERK pathway. Raf-1 dissociates from its complex with MEK in the presence of RKIP (Summarized in Figure 1). As a result, downstream mitogen-activated protein kinase (MAPK) signaling is interrupted and diminished. As stated earlier, RKIP can bind to Raf-1 or MEK, yet not at the same time, and binding to either one is enough to cause downstream inhibition [28]. In addition, it was postulated that RKIP may be involved in growth, transformation, and differentiation [3] as these pathways are often deregulated in cancer.

Figure 1. Role of RKIP in Raf signaling.
(A) In the inactive state, RKIP is unphosphorylated and binds to activated Raf, inhibiting its ability to activate MEK. Also, G-protein-coupled receptor kinase-2 (GRK-2) binds to G-protein-coupled receptor (GPCR) resulting in inhibition of its activity.
(B) Cell membrane receptors are bound by ligand and protein kinase C (PKC) is activated leading to phosphorylation of RKIP. The phosphorylated RKIP releases Raf, which can then phosphorylate MEK, which in turns phosphorylates ERK. The phosphorylated RKIP can then bind to GRK-2, causing it to release, which allows GPCR to phosphorylate its downstream targets, including Raf. The overall effect is a positive reinforcement of Raf activity.

Protein kinase C (PKC), which phosphorylates target proteins that control growth, differentiation and transcription, can inactivate RKIP through phosphorylation of RKIP on serine 153 and alleviate its inhibition of Raf-1 [5]. PKC is normally recruited to the plasma membrane and activated by diacylglycerol. Its location near the plasma membrane may place it in close proximity to RKIP, which also binds to phospholipids [29]. As a result, PKC along with RKIP, function as unique selective regulators of the Raf-1/MEK/ERK growth factor signaling cascade. When RKIP is phosphorylated, it releases from Raf-1 and can bind onto G-protein-coupled receptor kinase-2 (GRK-2) preventing GRK-2's ability to inhibit G-protein-coupled receptor activity [30].

In summary, RKIP impacts multiple signaling pathways, and RKIP activity itself is regulated by PKC. Due to its involvement in several signaling pathways, RKIP modulates cellular functions that are dependent or altered by these signaling pathways.

4. Apoptosis
Apoptosis is a physiological cell self-destruction program that has been implicated in multiple biological and pathological processes including cancer. Beside the inappropriate expression of tumor promoting genes and the silencing of tumor suppressor genes the acquisition of resistance toward apoptosis is perhaps the next important landmark in malignant tumor development. Therefore, understanding the underlying mechanisms of apoptosis is of obvious importance in determining the efficacy of cancer treatments. In drug-curable malignancies, apoptosis is a prominent mechanism associated with the induction of tumor remission. Further, the expression of apoptosis modulators within a tumor appears to correlate with its sensitivity to traditional therapies. In mammals, the signaling pathways leading to apoptosis converge on activation of either caspase-8 or -9 [31]. Caspase-8 is mainly activated by CD95 and related TNF (tumor necrosis factor) receptor. The activation of caspase-9 is usually initiated by stress-induced signals inside the cell requiring pro-apoptotic molecules released from the mitochondria and Apaf-1 [32, 33]. In addition, apoptosis can also be induced by caspase-independent mechanisms. The molecular mechanism of caspase-independent apoptosis is unclear and is likely to involve mitochondria [34]. Caspase-dependent apoptosis pathways involving CD95 and mitochondria have been partially delineated and include both shared and unique components. The core component is comprised of caspase-3 and its downstream substrates. It appears that the regulation of apoptosis occurs mainly at levels upstream of caspase-3 and involves the activation cleavage of caspase-8 and -9 [35].

Chemotherapeutic agents kill cancer cells by multiple mechanisms. Although the primary intracellular targets of drug action are rather distinct, it has become evident that chemotherapeutic drugs cause cellular stress and induce various signaling pathways leading to apoptosis. There are two main mediators of stress-induced apoptosis in cancer cells after treated with chemotherapeutic drugs. They are the p53 family proteins and the stress-activated protein kinase (SAPK, also known as also Jun-N-terminal kinase or JNK) [36-38]. Rapid induction of p53 and other family members p73 and p63 function is achieved in response to DNA damage through post-translational mechanisms [39, 40]. The transcriptional activity of p53 family of proteins is important for its pro-apoptotic function. p53 and other members of the family can induce the expression of proteins involved in the mitochondrial pathway and in the death receptor pathway. JNKs can regulate the activity of AP-1 transcription factors. Known pro-apoptotic target genes for AP-1 are CD95L and TNF-alpha.

Cancer cells can acquire resistance to apoptosis by various mechanisms. It can involve regulators of various apoptosis signaling pathways. For example, cancer cells can acquire resistance by up-regulating the anti-apoptotic molecules BCL2 or down-regulating the expression of pro-apoptotic molecule BAX [41]. It can also involve pro-survival signaling pathways that impinge on apoptosis signaling. Pro-survival signaling pathways that are parts of cell failsafe mechanism against transformation include the PI3/AKT, Raf and NF-kappaB. We have evidence suggesting down-regulating the expression of RKIP is another mechanism cancer cell employed to evade apoptosis.

RKIP is a much more abundant protein than Raf-1, and raises the possibility that it may have additional intracellular targets. Indeed, we have shown that RKIP also antagonizes the signal transduction pathways that mediate the activation of the transcription factor NF-kappaB in response to TNF-alpha? and interleukin 1 beta (IL-1beta) stimulation [42]. Modulation of RKIP expression levels affected NF-kappaB signaling independently of the MAPK pathway. Genetic epistasis analysis involving the ectopic expression of kinases acting in the NF-kappaB pathway indicated that RKIP acts upstream of the kinase complex that mediates the phosphorylation and inactivation of the inhibitor of NF-kappaB (IkappaB). In vitro kinase assays showed that RKIP antagonizes the activation of IkappaB kinase (IKK) activity elicited by TNF-alpha. RKIP physically interacted with four kinases of the NF-kappaB activation pathway, NF-kappaB inducing kinase (NIK), transforming growth factor beta (TGF-beta activated kinase (TAK1), IKKalpha and IKKbeta. This mode of action bears striking similarities to the interaction of RKIP with Raf-1 and MEK1 in the MAPK pathway.

Although the molecular mechanism of how RKIP inhibits Raf and NF-kappaB signaling pathways has been partially delineated, little is known about the biological relevance of the inhibition of both pathways by RKIP. Consistent with the inhibitory effects of RKIP on NF-kappaB and Raf signaling, we have recently found that RKIP can induce apoptosis in certain prostate and breast carcinoma cell lines (unpublished data). The effect of RKIP on apoptosis signaling pathway appears to be specific to cancer cells as we have not observed any pro-apoptotic effect of RKIP on primary or immortalized non-transformed cells. We observed that RKIP induces apoptosis by modulating the activities of Raf and NF-kappaB. Taken together, these data strongly implicate that RKIP can modulate cancer cell resistance to chemotherapy. Loss or RKIP may have an overall anti-apoptotic effect.

5. RKIP suppresses prostate cancer metastasis.
Metastasis is defined as the formation of progressively growing secondary tumor foci at sites discontinuous from the primary lesion [43]. The metastatic process is a multi-step mechanim in which a metastatic cancer cell escapes from the primary tumor, enters the circulation, invades a distant tissue site and grows into a macroscopic tumor at the target site. Since many steps are required for metastasis to occur, it may be possible to block metastasis by inhibiting a single gene that is required for the completion of any one of these steps [43]. Consistent with this possibility, several studies have shown that the loss of function of specific genes called metastasis suppressor genes (MSG) is an important event in the progression to malignancy [44-47]. Due to their ability to regulate the metastatic process, MSG are potential diagnostic and therapeutic targets. Accordingly, identification of MSG may lead to advances in prostate cancer therapy.

To begin to identify prostate cancer MSG, we examined the difference in gene expression between a non-metastatic prostate cancer cell line and a metastatic prostate cancer cell line. We found that RNA expression of several genes was altered between these two lines [48]. RKIP was one a few genes found to be expressed at a lower level in the metastatic compared to the non-metastatic cell line. This suggested the possibility that loss of RKIP was associated with the development of metastasis.

To explore the relation of RKIP expression to clinical prostate cancer, we examined RKIP protein expression in non-neoplastic prostate tissue, primary prostate cancer and prostate cancer metastases. We found that RKIP expression level was highest for benign tissue, lower for cancerous tissue (declining with increasing Gleason score), and absent in metastases (P<.001, Mantel-Haneszel chi-square test). These results provided strong evidence that loss of RKIP is associated with the development of prostate cancer metastases. However, these data do not demonstrate that RKIP functionally contributes to the metastatic process.

To examine the function of RKIP during prostate cancer progression, we modulated RKIP expression in prostate cancer cells to determine the effect of different RKIP levels on the prostate cancer cells metastatic ability. Modulating RKIP expression had no effect on the ability of the cells to grow in vitro or on their ability to form colonies in soft agar. These results suggest that modulation of RKIP expression has no effect on these two primary tumorigenic properties of human prostate cancer cells. However, to examine whether changes of RKIP expression are associated with cancer cell invasiveness, we measured the effect of modulating RKIP expression on the in vitro invasive ability of the prostate cancer cells. Increasing RKIP expression in metastatic cancer cells decreased in vitro invasive ability. Conversely, decreasing RKIP expression, using antisense, in non-metastatic prostate cancer cells increased in vitro invasive ability. These results suggest that RKIP expression is inversely associated with the invasiveness of prostate cancer cells in vitro.

To determine if the in vitro results had relevance to in vivo metastasis, we determined if increasing RKIP expression decreased metastasis in a murine model. Accordingly, we implanted into mice prostates either (1) metastatic prostate cancer cells transfected with empty vector so they expressed low basal levels of RKIP or (2) the same prostate cancer cells that were engineered to express increased levels of RKIP. Tumor growth at the injection site in the prostates was identical between both groups. In contrast, increasing RKIP expression in the tumor cells resulted in decreasing the number of mice that developed lung metastases by 70%. Furthermore, in the mice that had received cells expressing increased RKIP and that developed metastases, the number of metastases was far fewer than in the mice that had received the cells expressing low levels of RKIP. Additionally, there was less vascular formation and less vascular invasion in the primary tumors derived from the mice that received the cells engineered to express RKIP. Taken together, these results suggest that RKIP functions as a suppressor of metastasis through decreasing angiogenesis and vascular invasion.

The specific signaling mechanisms through which RKIP modulates metastasis and invasion are not known. However, due to its effects on MEK/ERK, G-protein-couple receptors and NF-kappaB pathway, it is likely that one or more of these pathways play a role.

6. Conclusions
Prior to being named RKIP, this protein had been recognized as PEBP for many years. PEBP had contributed to many different physiologic activities including reproduction and neurophysiology. Recent research activity has identified that RKIP regulates an important signaling cascade, i.e., the Raf-MEK-ERK kinase cascade. In addition to this role, RKIP also modulates G-protein and NF-kappaB signaling. RKIP has been shown to contribute to several anti-cancer activities including induction of apoptosis and inhibition of metastasis. As prostate cancer progresses, it loses RKIP expression, which in turn, promotes metastasis. These observations suggest that restoring RKIP expression or inhibiting effectors downstream of Raf that are normally blocked by RKIP will have potent anti-cancer effects.

Acknowledgments
This work was supported by the National Cancer Institute Grant CA098513.

References
[1] P. C. Simister, M. J. Banfield, R. L. Brady, The crystal structure of PEBP-2, a homologue of the PEBP/RKIP family, Acta Crystallogr D Biol Crystallogr 58 (2002) 1077-1080.

[2] B. S. Vallee, P. Tauc, J. C. Brochon, R. Maget-Dana, D. Lelievre, M. H. Metz-Boutigue, N. Bureaud, F. Schoentgen, Behaviour of bovine phosphatidylethanolamine-binding protein with model membranes. Evidence of affinity for negatively charged membranes, Eur J Biochem 268 (2001) 5831-5841.

[3] K. Yeung, T. Seitz, S. Li, P. Janosch, B. McFerran, C. Kaiser, F. Fee, K. D. Katsanakis, D. W. Rose, H. Mischak, J. M. Sedivy, W. Kolch, Suppression of Raf-1 kinase activity and MAP kinase signalling by RKIP, Nature 401 (1999) 173-177.

[4] M. J. Banfield, J. J. Barker, A. C. Perry, R. L. Brady, Function from structure? The crystal structure of human phosphatidylethanolamine-binding protein suggests a role in membrane signal transduction, Structure 6 (1998) 1245-1254.

[5] K. C. Corbit, N. Trakul, E. M. Eves, B. Diaz, M. Marshall, M. R. Rosner, Activation of Raf-1 signaling by protein kinase C through a mechanism involving Raf kinase inhibitory protein, J. Biol. Chem. (2003) M210015200.

[6] M. Maki, N. Matsukawa, H. Yuasa, Y. Otsuka, T. Yamamoto, H. Akatsu, T. Okamoto, R. Ueda, K. Ojika, Decreased expression of hippocampal cholinergic neurostimulating peptide precursor protein mRNA in the hippocampus in Alzheimer disease, J Neuropathol Exp Neurol 61 (2002) 176-185.

[7] S. Mitake, K. Ojika, E. Katada, Y. Otsuka, N. Matsukawa, O. Fujimori, Accumulation of hippocampal cholinergic neurostimulating peptide (HCNP)-related components in Hirano bodies, Neuropathol Appl Neurobiol 21 (1995) 35-40.

[8] S. Mitake, E. Katada, Y. Otsuka, N. Matsukawa, T. Iwase, T. Tsugu, O. Fujimori, K. Ojika, Possible implication of hippocampal cholinergic neurostimulating peptide (HCNP)-related components in Hirano body formation, Neuropathol Appl Neurobiol 22 (1996) 440-445.

[9] Y. Tsugu, K. Ojika, N. Matsukawa, T. Iwase, Y. Otsuka, E. Katada, S. Mitake, High levels of hippocampal cholinergic neurostimulating peptide (HCNP) in the CSF of some patients with Alzheimer's disease, Eur J Neurol 5 (1998) 561-569.

[10] F. Schoentgen, P. Jolles, From structure to function: possible biological roles of a new widespread protein family binding hydrophobic ligands and displaying a nucleotide binding site, FEBS Lett 369 (1995) 22-26.

[11] J. Frayne, A. McMillen, S. Love, L. Hall, Expression of phosphatidylethanolamine-binding protein in the male reproductive tract: immunolocalisation and expression in prepubertal and adult rat testes and epididymides, Mol Reprod Dev 49 (1998) 454-460.

[12] Z. Fu, P. C. Smith, L. Zhang, M. A. Rubin, R. L. Dunn, Z. Yao, E. T. Keller, Effects of raf kinase inhibitor protein expression on suppression of prostate cancer metastasis, J Natl Cancer Inst 95 (2003) 878-889.

[13] R. Jones, L. Hall, A 23 kDa protein from rat sperm plasma membranes shows sequence similarity and phospholipid binding properties to a bovine brain cytosolic protein, Biochim Biophys Acta 1080 (1991) 78-82.

[14] A. C. Perry, L. Hall, A. E. Bell, R. Jones, Sequence analysis of a mammalian phospholipid-binding protein from testis and epididymis and its distribution between spermatozoa and extracellular secretions, Biochem J 301 ( Pt 1) (1994) 235-242.

[15] P. T. Saunders, C. McKinnell, M. R. Millar, J. Gaughan, K. J. Turner, B. Jegou, V. Syed, R. M. Sharpe, Phosphatidylethanolamine binding protein is an abundant secretory product of haploid testicular germ cells in the rat, Mol Cell Endocrinol 107 (1995) 221-230.

[16] J. Frayne, C. Ingram, S. Love, L. Hall, Localisation of phosphatidylethanolamine-binding protein in the brain and other tissues of the rat, Cell Tissue Res 298 (1999) 415-423.

[17] E. Katada, S. Mitake, N. Matsukawa, Y. Otsuka, Y. Tsugu, O. Fujimori, K. Ojika, Distribution of hippocampal cholinergic neurostimulating peptide (HCNP)-like immunoreactivity in organs and tissues of young Wistar rats, Histochem Cell Biol 105 (1996) 43-51.

[18] E. Katada, K. Ojika, S. Mitake, R. Ueda, Neuronal distribution and subcellular localization of HCNP-like immunoreactivity in rat small intestine, J Neurocytol 29 (2000) 199-207.

[19] C. Moore, A. C. Perry, S. Love, L. Hall, Sequence analysis and immunolocalisation of phosphatidylethanolamine binding protein (PBP) in human brain tissue, Brain Res Mol Brain Res 37 (1996) 74-78.

[20] M. Taiji, N. Tohdoh, K. Ojika, Neuronal expression of hippocampal cholinergic neurostimulating peptide (HCNP)-precursor mRNA in rat brain, J Neurosci Res 45 (1996) 202-215.

[21] N. Tohdoh, S. Tojo, M. Kimura, T. Ishii, K. Ojika, Mechanism of expression of the rat HCNP precursor protein gene, Brain Res Mol Brain Res 45 (1997) 24-32.

[22] K. Ojika, E. Katada, N. Tohdoh, S. Mitake, Y. Otsuka, N. Matsukawa, Y. Tsugu, Demonstration of deacetylated hippocampal cholinergic neurostimulating peptide and its precursor protein in rat tissues, Brain Res 701 (1995) 19-27.

[23] S. Bucquoy, P. Jolles, F. Schoentgen, Relationships between molecular interactions (nucleotides, lipids and proteins) and structural features of the bovine brain 21-kDa protein, Eur J Biochem 225 (1994) 1203-1210.

[24] D. K. Grandy, E. Hanneman, J. Bunzow, M. Shih, C. A. Machida, J. M. Bidlack, O. Civelli, Purification, cloning, and tissue distribution of a 23-kDa rat protein isolated by morphine affinity chromatography, Mol Endocrinol 4 (1990) 1370-1376.

[25] L. C. Robinson, K. Tatchell, TFS1: a suppressor of cdc25 mutations in Saccharomyces cerevisiae, Mol Gen Genet 230 (1991) 241-250.

[26] F. Schoentgen, N. Seddiqi, S. Bucquoy, P. Jolles, L. Lemesle-Varloot, K. Provost, J. P. Mornon, Main structural and functional features of the basic cytosolic bovine 21 kDa protein delineated through hydrophobic cluster analysis and molecular modelling, Protein Eng 5 (1992) 295-303.

[27] T. Kroslak, T. Koch, E. Kahl, V. Hollt, Human phosphatidylethanolamine-binding protein facilitates heterotrimeric G protein-dependent signaling, J Biol Chem 276 (2001) 39772-39778.

[28] K. Yeung, P. Janosch, B. McFerran, D. W. Rose, H. Mischak, J. M. Sedivy, W. Kolch, Mechanism of suppression of the Raf/MEK/extracellular signal-regulated kinase pathway by the raf kinase inhibitor protein, Mol Cell Biol 20 (2000) 3079-3085.

[29] M. D. Bazzi, M. A. Youakim, G. L. Nelsestuen, Importance of phosphatidylethanolamine for association of protein kinase C and other cytoplasmic proteins with membranes, Biochemistry 31 (1992) 1125-1134.

[30] K. Lorenz, M. J. Lohse, U. Quitterer, Protein kinase C switches the Raf kinase inhibitor from Raf-1 to GRK-2, Nature 426 (2003) 574-579.

[31] S. H. Kaufmann, M. O. Hengartner, Programmed cell death: alive and well in the new millennium, Trends Cell Biol 11 (2001) 526-534.

[32] I. Budihardjo, H. Oliver, M. Lutter, X. Luo, X. Wang, Biochemical pathways of caspase activation during apoptosis, Annu Rev Cell Dev Biol 15 (1999) 269-290.

[33] K. Newton, A. Strasser, Cell death control in lymphocytes, Adv Immunol 76 (2000) 179-226.

[34] D. Green, G. Kroemer, The central executioners of apoptosis: caspases or mitochondria?, Trends Cell Biol 8 (1998) 267-271.

[35] N. A. Thornberry, Y. Lazebnik, Caspases: enemies within, Science 281 (1998) 1312-1316.

[36] J. Kleeff, M. Kornmann, H. Sawhney, M. Korc, Actinomycin D induces apoptosis and inhibits growth of pancreatic cancer cells, Int J Cancer 86 (2000) 399-407.

[37] M. Kuwabara, K. Takahashi, O. Inanami, Induction of Apoptosis through the Activation of SAPK/JNK Followed by the Expression of Death Receptor Fas in X-irradiated Cells, J Radiat Res (Tokyo) 44 (2003) 203-209.

[38] O. Sordet, Q. A. Khan, K. W. Kohn, Y. Pommier, Apoptosis induced by topoisomerase inhibitors, Curr Med Chem Anti-Canc Agents 3 (2003) 271-290.

[39] T. Truong, G. Sun, M. Doorly, J. Y. Wang, M. A. Schwartz, Modulation of DNA damage-induced apoptosis by cell adhesion is independently mediated by p53 and c-Abl, Proc Natl Acad Sci U S A 100 (2003) 10281-10286.

[40] M. Marabese, F. Vikhanskaya, C. Rainelli, T. Sakai, M. Broggini, DNA damage induces transcriptional activation of p73 by removing C-EBPalpha repression on E2F1, Nucleic Acids Res 31 (2003) 6624-6632.

[41] S. Cory, D. C. Huang, J. M. Adams, The Bcl-2 family: roles in cell survival and oncogenesis, Oncogene 22 (2003) 8590-8607.

[42] K. C. Yeung, D. W. Rose, A. S. Dhillon, D. Yaros, M. Gustafsson, D. Chatterjee, B. McFerran, J. Wyche, W. Kolch, J. M. Sedivy, Raf kinase inhibitor protein interacts with NF-kappaB-inducing kinase and TAK1 and inhibits NF-kappaB activation, Mol Cell Biol 21 (2001) 7207-7217.

[43] D. R. Welch, C. W. Rinker-Schaeffer, What defines a useful marker of metastasis in human cancer?, J Natl Cancer Inst 91 (1999) 1351-1353.

[44] B. A. Yoshida, M. M. Sokoloff, D. R. Welch, C. W. Rinker-Schaeffer, Metastasis-suppressor genes: a review and perspective on an emerging field, J Natl Cancer Inst 92 (2000) 1717-1730.

[45] J. T. Dong, H. Suzuki, S. S. Pin, G. S. Bova, J. A. Schalken, W. B. Isaacs, J. C. Barrett, J. T. Isaacs, Down-regulation of the KAI1 metastasis suppressor gene during the progression of human prostatic cancer infrequently involves gene mutation or allelic loss, Cancer Res 56 (1996) 4387-4390.

[46] X. Yang, D. R. Welch, K. K. Phillips, B. E. Weissman, L. L. Wei, KAI1, a putative marker for metastatic potential in human breast cancer, Cancer Lett 119 (1997) 149-155.

[47] P. S. Steeg, G. Bevilacqua, L. Kopper, U. P. Thorgeirsson, J. E. Talmadge, L. A. Liotta, M. E. Sobel, Evidence for a novel gene associated with low tumor metastatic potential, J Natl Cancer Inst 80 (1988) 200-204.

[48] Z. Fu, I. Dozmorov, E. Keller, Osteoblasts produce soluble factors that induce a gene expression pattern in non-metastatic prostate cancer cells, similar to that found in bone metastatic prostate cancer cells, Prostate 51 (2002) 10-20.

Back . . .