Cell Transformation, Invasion, and Angiogenesis: A Regulatory Role for Ornithine Decarboxylase and Polyamines?

In this issue of the Journal, Kubota et al. (1) report that overexpression of ornithine decarboxylase (ODC) triggers mitogen-activated protein (MAP) kinase activity, which in turn implies a proportional increase in invasiveness. They demonstrate that transfection of mouse 10T12 fibroblasts with rat ODC complementary DNA (cDNA) elicited morphologic transformation; the ODC transformants formed colonies in monolayer tissue culture and proliferated in semisolid soft agar. Kubota et al. suggest that the ODC-induced cell transformation mechanism can be explained, at least in part, by activation of the MAP kinase pathway, since the ODC transformants that they studied displayed increased protein kinase activity against myelin basic protein used as an MAP kinase substrate in in-gel assays. In addition, they postulate that deregulated ODC activity may influence invasion of a cancer cell. This was evidenced by migration of ODC transformants through a reconstituted basement membrane-coated filter in Boyden chambers in vitro. Furthermore, Kubota et al. show increased secretion of matrix metalloproteinase (MMP)-2, a 72-kd progelatinase considered to be one of the key players in extracellular matrix degradation. The results obtained by these investigators provide strong evidence that ODC activity and polyamines may affect a number of processes that have an impact on tumor development, including abnormal cell proliferation and invasion. Polyamines-putrescine, spermidine, and spermine-are small cationic organic molecules that are indispensable for cell proliferation and differentiation. All eukaryotic cells contain one or more of the polyamines. Their concentrations vary during the cell cycle, and one of the first events in proliferating cells is induction of polyamine synthesis, preceding both nucleic acid and protein syntheses. Beyond that, however, it is hard to elucidate the specific biologic functions of polyamines. The key enzyme in the polyamine biosynthetic pathway is ODC. In its active form, ODC is a dimer of two identical subunits of 51 kd. It is one of the most strictly regulated enzymes known. ODC has a turnover rate ranging from a few minutes to 1 hour. In contrast, the average life of mammalian enzymes is counted in days. This rapid turnover is essential for fast and dramatic changes in ODC levels in response to many kinds of growth stimuli that affect the rate of enzyme synthesis, making its induction pattern more “ pulse-like.” The activity of ODC is regulated by many putative mechanisms. These mechanisms include transcriptional control of gene expression, suppression of translation, post-translational modifications, and proteolytic destruction. It is now well established that deregulated ODC activity and polyamine metabolism are commonly found in a variety of animal and human cancers. It is interesting that, during recent years, one of the key questions in polyamine biology (i.e., does elevated ODC activity contribute, or is it merely one of the consequences of neoplastic cell growth?) has been extensively explored with the aid of advanced molecular biology techniques. The supporting evidence that cell transformation is dependent on increased ODC activity and polyamine levels rests in part on the following findings: (a) When full-length human ODC cDNA was expressed from different mammalian vectors with promoters of different strengths, the transfected, immortalized NIH 3T3 and Rat-1 fibroblasts showed morphologic transformation. Importantly, the degree of transformation was related to the levels of ODC expression. There appears to be a minimum threshold level of ODC activity beyond which ODC overexpression induces cell transformation. The strongest ODC expression was elicited by the Moloney leukemia virus promoter, and these transformed cells showed many of the well-characterized properties of cancerous cells. These properties include alterations in cell shape, cell attachment, intercellular communication, and growth factor requirements (2). (b) Luk and co-workers (3) provided further evidence of a causative role for deregulated ODC activity. Their NIH 3T3based human ODC transformants induced tumors in nude mice, implicating increased tumorigenicity in vivo. (c) Shantz and Pegg (4) elegantly used a different model- 4E-P2 cells derived from NIH 3T3 cells that overexpress a translation initiation factor eIF-4E. 4E-P2 cells exhibit a transformed phenotype, most likely due to translational relief of messenger RNAs (mRNAs) encoding proteins that regulate cell growth. One of these mRNAs is ODC. The proliferation of 4E-P2 cells in soft agar was inhibited by alpha-difluoromethylornithine (DFMO) treatment, an irreversible inhibitor of ODC activity, providing more evidence for deregulated ODC activity in neoplasia. Recently, Shantz et al. (5) reported that expression of a C-terminal truncated, dominant-negative ODC mutant reversed the transformation of 4E-P2 cells. In addition, they showed that high ODC levels are critical for the transformed phenotype; when the ODC activity of 4E-P2 cells was reduced by expression of the above mutant by an average of 60%, the tumors formed were smaller and showed delayed progression in nude mice. (d) Gilmour and co-workers (6) have provided the first evidence that elevated ODC activity and polyamine levels contribute to transformation of primary cultures of normal, diploid keratinocytes and fibroblasts. In their experiments, they overexpressed C-terminal truncated mouse ODC cDNA from a replication-defective retroviral vector. Their results indicate that ODC overexpression by itself is not sufficient to induce solid tumors in nude mice, but cooperation of ODC with activated c-rasHa gene induced tumorigenicity of subcutaneously injected epidermal 308 and SP-1 cells. It is interesting that all tumors arising from the ODC-infected 308 or SP-1 cells were vascularized, implicating polyamine-mediated angiogenesis. (e) One of the critical tests for a requirement of ODC in cell transformation is to overproduce it in transgenic mice. Significantly, Megosh et al. (7) have targeted C-terminal truncated mouse ODC cDNA expression to a specific subpopulation of epidermal keratinocytes via bovine keratin IV promoter. In this in vivo model, constitutive ODC overexpression induced spontaneous tumor formation at high frequency, indicating that an increase in ODC levels is required for cell transformation and is not a secondary, pleiotypic response to oncogene activation. In contrast, overexpression of full-length human ODC cDNA from its own promoter did not result in any tumor formation in transgenic mice (8). The discrepancy between the two reports appears to be a question of ODC activity rising above a critical point that brings on a transformed state in appropriate tissue. Although the concept of ODC as a novel proto-oncogene is becoming more widely accepted, relatively little is known so far about the ODC-induced cell transformation mechanisms. Therefore, the study by Kubota et al. (1) on the involvement of MAP kinases in deregulated cell proliferation of ODC transformants is important. The current evidence implies that enhanced tyrosine phosphorylation is a frequent hallmark of cancer cells and is associated with deregulated phosphotransferase activity of protein tyrosine kinases, decreased phosphatase activity, or both, resulting in inappropriate phosphorylation and activation of cellular substrates (9). On the other hand, cell transformation is consistent with a reduced concentration of F-actin microfilaments. It may well be that the disassembly of actin cytoskeleton defines, at least partly, the onset of transformed cell morphology and the loss of anchorage to the extracellular matrix (10). This concept is further supported by the discovery that increased ODC activity and elevated polyamine levels fine-tune alterations in the tyrosine phosphorylation of proteins, one or more of which may be F-actin-binding phosphotyrosine-containing proteins, in the assembly of cortical cytoskeleton upon cell transformation (2,11). That is, the cellular actin cytoskeleton may be remodeled by polyamine-dependent phosphorylations. These findings suggest that one of the most intriguing roles of polyamines could be their impact on the net level of total tyrosine phosphorylation, perhaps by increasing protein tyrosine kinase activity of some (key) signaling proteins, acting as phosphatase inhibitors, or both. However, to show this phenomenon unambiguously is a tedious challenge, since so far only 300 of the predicted 2000 protein tyrosine kinase and 1000 phosphatase genes have been identified (9). In addition, presumably many of these gene products are associated with regulation of the organization of the actin cytoskeleton. To find out which signaling protein tyrosine kinases and phosphatases are required for DNA synthesis in response to neoplastic cell growth, the ODC transformants provide an interesting model with which to investigate this daunting task. Immunoblotting of total cellular proteins with antiphosphotyrosine revealed one band of about 130 kd that had undergone increased tyrosine phosphorylation in the ODC transformants (2). It seems that there is a cluster of proteins of this size, and one of them is characterized as Src-associated p120 (12), an armadillo-like protein reported to be a central regulator of cell adhesion, cytoskeletal interactions, and tumor progression (13). Another is characterized as Src-associated p130, later designated p130CAS(14), which has been shown to be a common regulator of cell transformation (12). p130CAS is an adaptor protein, which has an NH2-terminal SH3 domain followed by 15 potential tyrosine phosphorylation sites and many proline-rich regions (14). Based on its relatedness to signaling molecules, p130CAS recruits and tethers together many SH2-and SH3-containing signaling molecules. As a result, it stabilizes the assembly of aberrant focal adhesion plaques by which mobile, transformed cells adhere transiently into the extracellular matrix. The function of the hyperphosphorylated p130CAS is only recently beginning to emerge, as it has been found to form complexes with a number of signaling transducers. These transducers include PI-3 kinase, PLC-γ, and SH2-/SH3-adaptor proteins Grb2, Nck, and Crk, all of which couple p130CAS to the Ras pathway (15). The road via Ras to MAP kinases provides the first level of understanding of how cellular activities are regulated all the way from the plasma membrane to MAP kinases. Indeed, it offers a nearly complete trace of a molecular chain in which the most efficient growthpromoting signal amplification can be achieved by arranging protein kinases from plasma membrane to the nucleus into series. Moreover, it allows feedback, cross-talk, and branching, by which short-lived tyrosine phosphorylation is converted to much longer lasting serinephosphorylation. It may be that, upon activation of the Ras pathway, p130CAS provides a nucleus for formation of a specific protein complex (15,16) that normally serves as a signaling unit but inappropriately signals the growth and the metastatic phenotype of a cancer cell. Apart from p130CAS , a novel cytoplasmic 70-kd-associated protein tyrosine kinase, first identified and characterized from ODC transformants, may also be a common mediator of cell transformation (12,17). Its phosphorylation on tyrosine and serine residues is sensitive to treatment with herbimycin A, a protein tyrosine kinase inhibitor that reverses the transformed morphology, including reappearance of normal distinct F-actin stress fibers. Of note is the function of this kinase downstream of ODC. This finding has two implications for polyamine biology: 1) Deregulation of a 70-kd-associated protein tyrosine kinase activity as a consequence of overexpression of ODC provides proof that increased ODC activity and elevated polyamine levels play a role in tyrosine phosphorylation. 2) The results raise the possibility that, in a variety of cancers, aberrant ODC expression may transduce its mitogenic potential by modulating signal transduction as an intermediate. Whether this novel protein tyrosine kinase is one of those kinases needed for the assembly of focal adhesion plaques remains to be seen. Another independent line of evidence also suggests the involvement of protein tyrosine kinases in ODC-induced transformation of NIH 3T3 cells. Moshier et al. (18) have reported that alterations in the signaling of epidermal growth factor receptor (EGF-R) may partly mediate transformation as well. The EGF-R is a prototype protein tyrosine kinase, in which the ligandbinding domain and the protein tyrosine kinase activity are separated by the plasma membrane. In their ODC overexpression model, the EGF-R activity is constitutively elevated and the cells are markedly more responsive to transforming growth factor-? ligand than normal cells. Whether this result is due to overexpression of the appropriate receptor is not reported. In addition, Moshier et al. transfected their ODC transformants with a dominant-negative EGF-R mutant and treated the cells with tyrphostin AG-25, a potent selective inhibitor of EGF-R. By these methods, the ODC transformants regained contact inhibition and anchorage-dependent growth. An aggressive tumor cell dissociates from the primary tumor and invades the basement barrier in a three-step process: 1) attachment to basement membranes, 2) dissolution of the extracellular matrix, and 3) migration. In this editorial, the activation of destructive enzymes is of particular interest. The overexpression and proteolytic activation of serine protease urokinase-type plasminogen activator and MMPs, including type IV collagenase, 92-kd gelatinase, 72-kd gelatinase, and stromelysins, are common features of an invasive cancer cell. For instance, 72-kd gelatinase (MMP-2) is secreted as a zymogen, and proteolytic removal of its NH2-terminus is followed by activation in a temporal and spatial fashion with respect to the cell attachment and migration (19). The presence of MMP-2 as the 62-kd activated form is a feature of an invasive phenotype. Kubota et al. (1) show that their ODC transformants secrete more 72-kd MMP-2 into the conditioned medium than do the control counterparts in gelatin-substrate zymogram analysis. In addition to the possibility that polyamines may turn on the proteolytic cascade by activating latent extracellular matrix proteases, previous indirect investigations (20,21) have indicated a pivotal role for polyamines not only in the growth of primary tumors but also in the formation of new blood vessels. Whether ODC transformants alone can evoke angiogenesis, an obligate requirement for an asymptomatic lesion to become malignant, remains an interesting issue to be clarified. When subjected to environmental insults, cells have diverse mechanisms by which they induce neoplastic growth. Current knowledge implies that ODC is a central convergence point of growth-promoting signals. The model depicted in suggests that ODC and polyamines are regulators of cell proliferation and transformation. In addition, it attempts to link a series of events arising from the loss of anchorage to extracellular matrix, including invasion and angiogenesis. In the proposed model, ODC becomes constitutively activated during cell transformation induced by v-src (2,11), c-HarasVal12(22), and neu (23) oncogenes. Recently, ODC has been shown to be a transcriptional target of the action of c-Mycheterodimers (24,25). In addition, ODC has proved to be a novel proto-oncogene whose overexpression by itself is sufficient to induce transformation in fibroblasts (1-7). Increased ODC activity is followed by activation of a 70-kd-associated protein tyrosine kinase (12). Presumably, it participates in phosphorylation of intracellular substrates, such as Src-associated p120 and p130CAS , that are needed for regulation of cytoskeletal architecture and cell-to-cell interactions. p130CAS serves as an intracellular glue trap (15,16), which in turn may function additively in tumorigenicity, providing an amplification loop of a primary aberrant growth-promoting signal by activating the Grb2complex and thereby the second round of the Ras pathway involving the MAP kinases (1). Such a transformed fibroblast attaches transiently to the extracellular matrix, releasing proteolytic proenzymes such as MMP-2 that, upon activation, degrade matrix components (1,19); the fibroblast then migrates through the breached matrix. Finally, an invasive fibroblast may express and secrete some angiogenic activators other than polyamines (20,21). In parallel, expression of some angiogenesis-inhibiting factors may be decreased, thereby abolishing normal constraints on angiogenesis. In the proposed model, polyamines are suggested to play a pivotal role because their concentrations define the level of protein tyrosine phosphorylation and the organization of the actin cytoskeleton (2,11,12). In addition, polyamines may affect invasion by activating latent extracellular matrix proteases, and they may mediate additively tumor angiogenesis (20,21). Fig. 1. Schematic tumorigenic fibroblast in which ornithine decarboxylase (ODC) is positioned in response to the constitutively activated Ras-signal transduction pathway. Ras protein functions as a critical intermediate between epidermal growth factor receptor (EGF-R) protein tyrosine kinase and downstream serinekinases (cRaf-1, mitogen-activated protein [MAP] kinase kinase [MAPKK], and MAP kinases [MAPK]). The activated Grb2complex links the constitutively activated EGF-R with Ras. The cascade issued by deregulated EGF-R, Neu, RasVal12 , or v-Src results in transcriptional activation of the ODC gene by the action of c-Mycheterodimers. Altered ODC transcription subsequently results in increased ODC activity and elevated polyamine levels, which in turn activates a 70-kilodalton (kDa)-associated protein tyrosine kinase (PTK), leading to further phosphorylation of proximal cellular substrates such as Src-associated p120 and p130CAS . Hyperphosphorylated p130CAS may then pass the once amplified growth signal again to the Ras pathway, thus functioning additively in ODC-triggered aberrant cell proliferation. In addition, invadopodia of an invasive fibroblast like this may release proteolytic matrix metalloproteinases (MMPs) locally, such as MMP-2, which are involved in pathologic destruction of the surrounding extracellular matrix. Finally, the onset of tumorigenicity and invasion may coincide with increased secretion of angiogenesis-stimulating factors and decreased release of angiogenic inhibitors. In this hypothetical model, polyamines are proposed to serve as critical mediators in all stages of tumor progression, including deregulated Ras signaling, invasion, and angiogenesis. The model of ODC and polyamines as regulators of cell transformation, invasion, and angiogenesis has certain limitations. First, to date no transforming versions of Grb2 and MAP kinases have been identified in any human tumors. Second, the model does not apply to the incipient growth phase of a cancer cell embedded within an asymptomatic lesion. In this avascular phase, the cell population is limited. As indicated above, in many studies (2,5–7), there was a minimum threshold level of ODC activity that had to be exceeded before cell transformation occurred. The missing link is how endogenous ODC is constitutively activated to such high levels during progression of a human tumor to full malignancy. To date, the only reports of genetic changes involving the ODC gene deal with its rearrangement with immunoglobulin γ1 in mouse myeloma cells (26) and its amplification with N-myc in human neuroblastoma (27). As yet, there have been no reports on point mutations. Because current knowledge is lacking with regard to ODC signaling, the proposed model () implies an ODC-triggered feedback amplification loop(s) of a mitogenic MAP kinasesignaling cascade that, after many rounds, would result in a persistently high ODC activity. It remains to be seen whether the EGF-R-mediated (18) MAP kinase pathway (1) is required for the cell transformation mechanism(s) of fully malignant ODC transformants, displaying highly elevated ODC activities. The other caveat is that one of the most critical tests for the requirement for ODC in cell proliferation is to inhibit its production by disrupting the ODC gene. Without these data, it still remains unproven whether ODC plays such a critical role in the growth of normal and neoplastic cells as has been thought. Despite these concerns, reduction of ODC activity by different means should not be overlooked in therapeutic interventions (28,29). Financially supported by Emil Aaltonen Foundation, Magnus Ehrnrooth Foundation, Finnish Cultural Foundation, and Wihuri Foundation. I thank Erkki Hölttä and Vijay Kumar for reading the manuscript and for their helpful suggestions.