Chapter 1 Introduction

Cancer remains a prominent killer in industrialized nations. Distinguishing and removing cancer cells from normal cells continue to be key in the experimental design for therapy and prevention. Modern approaches to treating cancer take advantage of critical biochemical differences between cancer cells and normal cells - from radiation therapy to chemotherapy to experimental gene therapy. The Warburg hypothesis was based on the metabolic differences between cancer cells and normal cells, and proposed that increased glycolysis by transformed cells conferred a bio-energetic advantage for survival over normal counterparts under anoxic conditions (Anghileri, 1983). This hypothesis laid the foundation for cancer research strategies to find critical differences between transformed cellular processes and normal cellular processes.

When cells proliferate in the absence of appropriate driving signals, cancer is the undesirable consequence. Therefore, understanding biochemical dynamics of cell cycle progression may lead to target-specific therapy with improved side effect profiles. Ultimately, this understanding will enhance the survival rate of cancer patients, and perhaps make prevention a reality.

The Mammalian Cell Cycle

Encouraging progress in understanding cell cycle regulation occurred over the past five years. New regulatory targets were discovered and novel functions of known proteins were uncovered. An obvious requirement for the stringent control of cell cycle progression is the prevention of deregulated proliferation - loss of control may result in tumors and cancers. Equally devastating is the loss of normal cellular functions due to non-differentiation, loss of growth in a favorable environment, or continued growth in a hostile environment.

The mammalian cell cycle is stringently regulated and orderly process by which a cell reproduces. Such regulation ensures faithful reproduction of DNA for subsequent distribution to daughter cells. Transitional checkpoints are therefore crucial for a successful completion of the cell cycle. The mammalian cell cycle typically completes in 24 hours, where dramatic changes occur in cellular metabolism and cytoskeletal physiology. Each stage of the cell cycle is profiled by distinct protein complexes and phosphorylation events. Dysregulation at any checkpoint program may lead to a transformed phenotype.

Extracellular signals direct the cell-cycle "engine." Mitogenic and anti-proliferative signals trigger specific sequences of cytosolic events through which the signal is eventually relayed to the nucleus. Stages of the cell cycle are G1 (Gap 1), S (Synthesis), G2, M (Mitosis), and G0 (quiescence), with mitosis as the shortest and most visually dramatic process of the cycle after which two daughter cells result, Figure 1-1.

The G1 phase is a period of intense metabolic activity and regulation. Integration of intracellular metabolic states with extracellular signals leads to proliferation, differentiation, apoptosis, or quiescence. The "Restriction Point" (R-Point) describes a cellular commitment to enter DNA synthesis under suitable conditions. The Restriction point is also known as a checkpoint, where "the cell is arrested at a particular phase of the cycle due to a lack of appropriate signals" (Hartwell and Weinert, 1989). Restriction points receive signals from central- and peripheral cellular networks to determine subsequent cell fate. Factors mandating cellular consequence include cell surface-receptors, cytoplasmic proteins, and nuclear proteins. These factors couple extracellular signals to intracellular processes to determine cell cycle progression. Dysregulation of processes driven by these factors is a severe liability to the cell and therefore represents potential areas for therapeutic intervention.

TGF-β-Mediated Signaling and Modulation of Peptide Growth Factors

Signaling specificity is conferred by receptors and mediated through associated-kinases. The mammalian cell is furnished with receptors linked to interactive series of cytoplasmic networks for controlling cellular processes. Growth factors that signal through tyrosine-kinase receptor families include the epidermal growth factor (EGF), platelet-derived growth factor (PDGF) and transforming-growth factor-α (TGF-α). Transforming growth factor-β1 (TGF-β) signals through a serine/threonine-kinase receptor pathway.

TGF-β1 is a 112 amino-acid homodimer with nine cysteine residues, Figure 1-2. Members of the TGF-β family exhibit 65% to 88% amino acid homology, with the amino- and carboxyl- termini conserved at 97% between species. TGF-β1 is synthesized in pre-pro form and must be enzymatically cleaved for release from the extracellular matrix. TGF-β is one of the only few extracellular molecules signaling through a serine/threonine kinase receptor, Figure 1-3. The heterotetrameric signaling complex confers signal specificity at the level of ligand binding (type II receptor-directed) and signaling consequence (type I receptor-directed). TGF-β type II receptor (TβR-II) binds TGF-β and recruits TβR-I, which alone has a low affinity for TGF-β. Formation of receptor-growth factor complex is followed by trans-phosphorylation of the GS-domain on TβR-I by TβR-II. Activation of TβR-I leads to phosphorylation of downstream effectors in the TGF-β signaling pathway.

The TGF-β family of growth factors plays divergent roles in cells, including development, wound healing, and tissue remodeling. TGF-β1 is also known to exert multiple effects on metabolism, including glycolysis and system A-dependent amino acid transport. However, the TGF-β signaling pathway remains unclear. TGF-β is enigmatic by virtue of its pleiotropic effects that depend on environmental and cellular context. In cells of epithelial origin, TGF-β potently inhibits proliferation. In cells of mesenchymal origin such as fibroblasts, TGF-β stimulates cell cycle progression. What remains unknown is the underlying biochemical processes responsible for the context-dependent TGF-β response. If TGF-β stimulates and inhibits the growth of various cells through a common mechanism, where does the switch occur such that one cell type may be inhibited for growth while another cell type is stimulated? If distinct pathways are employed, what are these pathways?

Tissue assignment of various cell types may contribute to the difference in growth factor response. Tissues containing epithelial cells serve as protective barriers for organs, and epithelial cells are typically polarized. Fibroblastic cells, on the other hand, connect tissues and have the potential to differentiate into other types of supportive cells such as adipocytes, smooth muscle cells, and bone- and cartilage-producing cells. Unique differentiation potentials of epithelial cells versus fibroblasts may determine the response to TGF-β. Supporting this possibility is the observation that transformed epithelial cells resistant to TGF-β inhibition display an elongated, fibroblastic-like morphology. Therefore, mechanisms that are present in epithelial cells but absent in fibroblasts may either be lost or inactivated during the transformation process. Elucidation of pathways governing context-specific responses will be important when considering TGF-β as a therapeutic agent.

TGF-β is also a potent modulator of growth factor signaling. TGF-β is observed to delay the effects of early-acting growth factors such as EGF (epidermal growth factor), TGF-α, FGF (fibroblast growth factor), PDGF, and INF-γ (interferon-gamma). During wound repair, TGF-β stimulates components of the extracellular matrix (ECM), concomitant with the release of other growth factors at the site of injury. The interactive signaling between TGF-β and released peptide growth factors becomes important in examining extracellular matrix deposition during wound repair. The modulation of PDGF (platelet-derived growth factor) by TGF-β led to an early proposition for TGF-β mechanism as a secondary and inductive effect, from stimulating PDGF production (Moses et al, 1987). However, antibodies against PDGF-BB failed to block TGF-β stimulation in C3H-10T½ fibroblasts (Kim et al, 1993). Signaling mechanisms independent of PDGF stimulation must therefore be present. Identification of TGF-β receptors also lend support for a direct signaling effect by TGF-β as opposed to an indirect response through modulation of other growth factor signaling.

A wealth of literature points to TGF-β modulation of receptor tyrosine-kinase signaling. In human lung fibroblasts, TGF-β potentiated the mitogenic responses to FGF in a time- and dose-dependent manner by up-regulating protein levels of the FGF receptor (Thannickal et al, 1998). The role of TGF-β in various fibrotic disorders may lie in its induction of CTGF (connective tissue growth factor) in human fibroblasts whereas EGF, PDGF, FGF, and IGF-1 (insulin-like growth factor-1) were without such inductive effects (Tamatani et al, 1998). During embryonic lung branching morphogenesis, TGF-β modulated the stimulatory effects of EGF and PDGF-AA in the epithelium. Abolishing TGF-β signaling with dominant-negative type II TGF-β receptors correlated with a potentiation of the EGF- and PDGF-AA-mediated mitogenic response (Zhao et al, 1998). The presence of the TGF-β signal therefore appeared to blunt the mitogenic effects of EGF and PDGF during branching morphogenesis.

Identification of downstream effectors of TβRs (TGF-β receptors) first came from studies of the signaling pathways of a member of the TGF-β superfamily, the bone morphogenic proteins (BMP). Bone morphogenic proteins regulate vertebrate tissue and organ development through proliferation, differentiation, and apoptosis. Studies of BMP regulatory mechanisms lead to the cloning of SMAD 1, a human homologue of the MAD (mothers against decapentaplegic protein from Drosophila) and SMA (related gene from C. elegans). The smad proteins were recognized as a class of transcription-mediating factors in the TGF-β signaling pathway (Liu et al, 1996). Subsequent smads were identified and cloned: Smad 2 was found to mimic activin effects in xenopus and in mammalian cells; "Deleted in Pancreatic Cancer locus 4" (DPC 4), later identified as Smad 4 was inactivated in ~50% of pancreatic adenocarcinomas (Zhang et al, 1997). Smad 5 was proposed to direct formation of ventral mesoderm and epidermis (Suzuki et al, 1997). Smad 8 was isolated from a rat brain cDNA library and was found to transcriptionally activate mesoderm genes downstream of the ALK-2 (activin-receptor like kinase) signaling pathways (Chen et al, 1997).

Smads 1-4 comprised the activating circuit of TGF-β signaling whereas smads 6 and 7 were the inhibiting factors in an interactive paradigm, Figure 1-4. Smads 6 and 7 were identified as related proteins that inhibited TGF-β signals mediated by the other smad members (Nakao et al, 1997). When TGF-β receptors are activated, Smads relay the TGF-β signal through a series of phosphorylation and heterodimerization events. Smad 4-complexes localize to the nucleus and regulate transcriptional responses including the regulation of p21cip1 (Moustakas and Kardassis, 1998) and ATF-2- (cAMP-response element-binding protein) mediated pathways (Sano et al, 1999).

Extracellular Signals in G1-S Phase Progression

Communication in society helps maintain function and order; likewise, signal transduction pathways enable a cell to respond appropriately to extracellular changes for maintaining cellular function and integrity. A simplified signal transduction paradigm consists of a signal, a messenger, and a receiver, Figure 1-9. The signal is itself a primary messenger, relaying information to a secondary messenger ("mediating receiver"), which then passes the information (thereby becoming a "mediating signaler") to a subsequent messenger until the target effector is reached and the appropriate cellular response executed. Extracellular signals include temperature (permissiveness), nutrient availability, waste accumulation, presence of toxic chemicals, proximity with neighboring cells, and the presence of hostile cells. In addition, intracellular states may also modulate surface-receptor profiles. Complexity in signal transduction arises when signaling pathways interconnect.

Integration of central and peripheral signaling pathways determines cellular phenotype. Mammalian cells have many signaling pathways converging and diverging to relay extracellular signals and affect nuclear and cytoplasmic events leading to G1/S progression. Major signaling pathways include the protein kinase-C (PKC) pathway, mitogen-activated protein kinase (MAPK) pathway, phosphatidylinositol 3-kinase (PI 3-kinase) pathway, FK506-rapamycin-associated protein (FRAP) pathway, and p70 S 6-kinase (p70 S6 K) pathway.

Convergence or divergence of major signaling pathways appears to be cell-specific. Integration and bifurcation of PI 3-kinase and p70 S6-kinase pathways illustrate context-specificity in determining signal. Although the specific targets of p70 S6 K remain largely unknown, this ribosomal kinase is intimately tied to cell cycle progression at the level of immediate early genes required for cell cycle entry. Several studies showed p70 S6 K to be an effector downstream PI 3-kinase, where PI 3-kinase inhibition prevented p70 S6 K activation and cell cycle progression. Coupling p70 S6 K activation with MAPK-mediated upstream signaling pathways appeared to also be cell-type specific. In H4IIE cells, the catabolic enzyme glucose-6-phosphatase is strongly repressed by insulin in a PI 3-kinase-sensitive manner (Dickens et al, 1998). However, inhibiting the FRAP/p70 S6-kinase and MAPK pathways did not abolish insulin repression, suggesting divergence of PI 3-kinase pathway from FRAP/p70 S6-kinase and MAPK pathways, Figure 1-10. Amino-acid withdrawal correlated with the de-activation of p70 S6-kinase and dephosphorylation of eIF-4E-BP 1: cells became unresponsive to all stimuli where increasing ambient amino acids reactivated p70 S6-kinase activity but did not significantly affect phosphotyrosine-associated PI 3-kinase activity or MAPK activity (Hara et al, 1998). Insulin-dependent anabolic responses converged at FRAP/p70 S6-kinase and PI 3-kinase-sensitive regulatory pathways but diverged from the MAPK pathway, Figure 1-10.

Rapamycin and wortmannin are two emerging microbial products useful in dissecting PI 3-kinase-mediated- and FRAP/mTOR-mediated-signaling pathways in mammalian cells. Isolated from Streptomyces hygroscopicus, rapamycin is an immunosuppressor that binds to its cytoplasmic immunophilin receptor, FKBP12, Figure 1-11. Rapamycin delays cell cycle entry by prolonging G1. In serum-deprived NIH 3T3 cells, rapamycin delayed cyclin D1 mRNA accumulation, affected transcript stability, and increased proteasome-mediated cyclin D1 degradation rate. Reducing cyclin D1 may decrease active cyclin D1-cdk 4 complexes to render cyclin E complexes susceptible to p27kip1 sequestration and inhibition (Hashemolhosseini et al, 1998). Wortmannin is a fungal metabolite and potent PI 3-kinase inhibitor isolated from Penicillium wortmanni, Figure 1-12. While both wortmannin and 2-(4-morpholinyl)-8-phenyl-1[4H]-benzopyran-4-one (LY294002) inhibit PI 3-kinase regulated pathways, mechanisms of these two agents may be distinct. Cell-type specific inhibition contribute to the effectiveness of rapamycin and wortmannin on downstream substrates including components of translation and upstream kinases.

Accelerating and Braking Components of the Mammalian Cell Cycle Machinery

Regulatory components in cell cycle progression are conserved between species. This conservation provides valuable insight to the mechanisms governing these processes: complementation studies between yeast, xenopus egg, and mammalian cells lead to the identification of regulators at the G1/S, G2/M, and G0/G1 transitions (Pines, 1992). Understanding the regulatory strategies of cell cycle components at the G1/S transition requires an investigation of component interaction dynamics. Periodic expression of regulatory components in the mammalian cell cycle is implicit in the name "cyclins". Cyclins associate with specific cyclin-dependent kinases (cdk) to form a stable kinase complex. Activation of cyclin-cdk complexes requires chemical modification by an activating kinase. The activated cyclin-cdk complex is then capable of phosphorylating target substrates (Poon and Hunter, 1995). Cyclin-cdk complexes comprise an important accelerating force through phase-boundaries, Figure 1-7.

Cyclin E and Cyclin D1 are important G1-cyclins. Cyclin E forms an active kinase complex with cdk 2 during G1/S transit (Koff et al, 1991). Constitutive expression of cyclin E in fibroblasts shortened G1, decreased cell size, and reduced serum-requirement for S-phase entry (Ohtsubo and Roberts, 1993). Transformed fibroblasts capable of anchorage-independent growth lacked cyclin E-cdk 2 complex activation in late G1 and suggested that suppression of cyclin E complexes by cdk inhibitors may contribute to contact inhibition (Fang et al, 1996). In unattached, transformed human fibroblasts, up-regulation of cdk-complex inhibiting proteins p21cip1 and p27kip1 were unable to overcome anchorage independence sustained by high levels of cyclin E-cdk 2 activity. On the other hand, substrate-attached, normal fibroblasts exhibited nuclear-localized p21cip1- and p27kip1-sequestered cyclin E-cdk 2 complexes (Orend et al, 1998). Therefore, differential compartmentalization of both inactive complexes and high-level cyclin E complexes collectively contributed to an anchorage-independent phenotype. While little is known about the downstream effectors of cyclin E-cdk 2 complexes, recently identified targets included components of the pre-mRNA processing machinery, such as the spliceosome-associated proteins (SAP-) 114, 145, 155, and the snRNP core proteins. In this study, p21cip1-sensitive in vitro SAP-phosphorylation by cyclin E-cdk 2 linked pre-mRNA splicing to cell cycle regulation (Seghezzi et al, 1998).

Cyclin D1 was originally identified as a candidate oncogene located on chromosome 11q13 that was clonally rearranged in benign parathyroid tumors ("PRAD1", Motokura et al, 1991). Independent and subsequent cloning approaches led to the identification of cyclins D2 and D3 mapping respectively to chromosomes 12p13 and 6p21 (Xiong et al, 1992). While structural similarity existed across the three D-cyclin members, distinct roles (as opposed to functional redundancy) between each cyclin D member was inherent in the evolutionary conservation of each D-cyclin across species (Inaba et al, 1992). Initial investigations for cyclin D1 function centered on its interaction with the product of the retinoblastoma gene, pRB (Dowdy et al, 1993). Cyclin D1-cdk 4 complexes were shown to phosphorylate pRB, thereby releasing sequestered E2F(factor mediating transactivation of the adenoviral E2 promoter)-1/DP (dimerization partner)-1 heterodimers Figure 1-8. This release step served as a rate-limiting factor in G1/S progression (Kato et al, 1993). Supporting the relationship between this kinase-substrate and cancer development were immunohistochemical studies of primary esophageal carcinomas. Overexpressing cyclin D1 inactivated pRB (Jiang et al, 1998). Proliferative potential conferred by cyclin D1 may also span tissue regeneration, since cyclin D1-dependent kinase activity was demonstrated to increase during G1 phase following partial hepatectomy to promote liver regeneration (Kato et al, 1998).

Cdk-inhibitors brake against cyclin-cdk-mediated acceleration into the cell cycle and include members of the CIP family ("Cyclin Inhibitory Protein" p21, p27, and p57) and the INK4 family ("cdk 4-Inhibitor Kinase gene" p15, p16, p18, and p19). P21Cip1/Waf1 is a member of the CIP family and is expressed in response to DNA damage. P21cip1 was originally identified by a yeast two-hybrid system from immunoprecipitates of cyclins A, E, D1, and cdk 2. P21cip1 was observed to be a potent inhibitor of cyclin-cdk complexes (Harper et al, 1993), notably inhibiting DNA replication through preventive interaction with PCNA (proliferating cell nuclear antigen). P21cip1 affected cdk activity in liver repair as observed by the dramatic increase in hepatocytic cell cycle progression in p21-/- mice (Albrecht et al, 1998). Selective dysregulation of p21cip1 may contribute to glomerular cell hypertrophy in diabetic nephropathy, where high glucose-induced mesangial cell hypertrophy in vitro was observed to associate with increased p21cip1 (Kuan, 1998).

P27Kip1 was discovered as a heat labile factor accumulated during cell cycle arrest and provided the putative link between TGF-β-mediated growth-arrest to cell cycle components (Polyak et al, 1994). P27kip1 binds and inhibits cyclin-cdk complexes through the N-terminus. Since p27kip1 displayed 44% amino acid identity with p21cip1, these two proteins were categorized in the same family of cyclin inhibitors. Although p27kip1 inhibits the activation of cyclin E-cdk 2 complexes, p27kip1 binds with higher affinity to cyclin D1-cdk 4. Therefore, p27kip1 -inhibition of cyclin E activity may be effectively titrated with accumulating cyclin D1 complexes. In addition, p27kip1 can also inhibit cyclin D3-cdk 4 activity (Dong et al, 1998). In various systems, p27kip1 was shown to accumulate under treatment with inhibitors such as rapamycin, wortmannin, retinoic acid (Matsuo and Thiele, 1998), and flavopiridol (Brusselbach et al, 1998) and appears to be a target of diverse anti-proliferative signaling pathways. P27kip1 is notably regulated post-transcriptionally, maintained at the protein level rather than at the mRNA level.

P57kip2 was identified to behave similarly to p27kip1 by potently inhibiting G1- and S-phases at the level of cyclin-cdk complexes (Lee et al, 1995). P57kip2 mutations were notably present in the cancer-predisposing Beckwith-Wiedemann syndrome and were observed to transmit through maternal imprinting of chromosome 11p15 (Matsuoka et al, 1996). P57kip2 was also implicated in other maternally imprinted tumors including adrenocortical carcinoma and rhabdomyosarcoma (Hatada et al, 1996). Like p21, p57Kip2 displays a C-terminal PCNA-binding domain to prevent DNA replication in vitro and S-phase entry in vivo. Functionality of both the cdk- and PCNA-inhibiting domains is essential in p57-suppression of c-myc/ras-mediated transformation in primary cells (Watanabe et al, 1998).

Cdk 4- and cdk 6-inhibiting proteins (p15Ink4b and p16Ink4a) were cloned in 1995 and mapped to human chromosome 9p21 (Quelle et al, 1995). The INK4 inhibitors were proposed to be tumor suppressors based on frequency of deletion in human neoplasia (Herman et al, 1995). Alterations or deletions at this locus have been correlated with non-small cell lung cancer (de Vos et al, 1995), glioblastoma (Izumoto et al, 1995), thyroid cancer (Jones et al, 1996), and acute pediatric lymphoblastic leukemia (Guidal-Giroux et al, 1996). With the discovery of subsequent members including p18 and p19, the INK4 family of inhibitory proteins may be classified according to involvement in human cancer due to genetic/epigenetic mutations (p15 and p16 deletions) and rare somatic mutations (p18 and p19 deletions, Gemma et al, 1996).

P16ink4a fine-tunes the cell cycle network by competing with cyclin D1 for cdk 4, Figure 1-6. P16ink4a therefore indirectly affects p27kip1-mediated inhibition of accumulating cyclin D1 complexes. In vitro binding assays showed that both p21cip1 and p27kip1 interacted with cyclins D1, D2, D3, E, A, and to a lesser extent, cyclin B. The cyclin subunit is necessary for such inhibitor-cdk interaction. In contrast, p15ink4b and p16ink4a can bind cdk 4 and cdk 6 without the presence of the cyclin subunit. P21cip1 and p27kip1 thus may act as a broad-spectrum inhibitor of cdk function by participating in ternary complexes while p15ink4b and p16ink4a specifically interfere with cyclin D-dependent kinase complexes (Hall et al, 1995). Interactive dynamics between the inhibition of cyclin E- and cyclin D-complexes by p27 and p16ink4a therefore shift the pertinent inhibitory fractions of cdk: increased cyclin E expression inactivated p16ink4a-mediated growth suppression whereas over-expression of p16ink4a shifted cyclin E-cdk 2 to the p27-bound fraction (Jiang et al, 1998). Therefore, even when p16ink4a specifically abrogated cdk 4 complexes, the full inhibitory potential of p16ink4a requires inhibitory sequestration of both cyclin D- and cyclin E-complexes.

Interaction between cdk inhibitors spans both titration and regulatory mechanisms. Overexpressing p16ink4a in osteogenic sarcoma cells sequestered cdk 4 and cdk 6 from cyclin D and cognate inhibitor p27kip1 or p21cip1 (titration) while increasing cellular levels of p21cip1 (regulation). Consequently, cyclin E-cdk 2 complexes became inhibited (Mitra et al, 1999). Since p21cip1 lies in the p53-mediated pathway, p16ink4a may relay DNA damage signals towards modulation of p21cip1. Further investigation in this cell line demonstrated that induction of p16ink4a also prevented phosphorylation of pRB by the cyclin D1-cdk 4 complex and inhibited cdk 2-associated activity (McConnell et al, 1999). In this case, the titration mechanism between cdk inhibiting components involving p16ink4a re-assorted p27kip1-sensitive complexes by binding cdk 4- and cdk 6-associated complexes. Within the cdk 4- and cdk 6-binding INK4 family members, p16ink4a warranted significant attention due to its ability to form a stable binary complex with cdk 4 while all other INK members transiently bind cdk 4 and showed greater affinity for cdk 6 (Parry et al, 1999). Therefore, titration dynamics between p16ink4a, p27kip1, and p21cip1 play a key role in determining cell cycle progression at the G0/G1/S borders.

Critical for S-phase entry is the inactivating hyperphosphorylation of pRB by cyclin D1-cdk 4 to release E2F-1, Figure 1-8. E2F heterodimerizes with proteins in the DP family and becomes a transcriptionally active unit. Target promoters containing E2F binding sites include dihydrofolate reductase and thymidine kinase. Phase-G0/G1 association of E2F with hypophosphorylated pRB controls G1/S progression. Interaction between E2F and pRB occurs through the hydrophobic pocket domain of pRB. This domain is frequently targeted by oncoproteins including E1A, SV40 large T antigen, and the human papilloma virus E7 protein. The presence of 3 pRB family members, 6 E2F members, and 3 DP members allows varying complexes between these partners. Since retinoblastoma protein concentration does not vary during the cell cycle, cyclin D1-cdk 4-mediated phosphorylation during mid-G1 serves as determinant for G1 exit. Hyperphosphorylation of the serine/threonine residues of pRB continues into late G1 phase via cyclin E-cdk 2, S phase by cyclin A-cdk 2, and G2/M transits by cyclin B-cdc 2 (Sherr and Roberts, 1995). Retinoblastoma protein-phosphorylation may not be necessary in G1/S entry if G1-cyclin complexes were overexpressed, further demonstrating that cell cycle regulation by pRB rests at the level of phosphorylation. Osteosarcoma SAOS-2 cells are RB-deficient and may be accelerated for G1/S entry by co-expressing cyclin E- or cyclin D-complexes, or by injecting purified G1-cyclin complexes during G1 (Leng et al, 1997).

Roles of Cell Cycle Components in Development, Tumorigenesis, and Waning of Proliferative Potential

A developing embryo illustrates the visual drama of transition - macroscopic manifestation of microscopic dynamics underlying cellular development. Each seemingly homogeneous cell is signal-directed to evolve and differentiate toward population heterogeneity. Improved biotechnological methods have enabled an important dimension in understanding development: linking cell cycle components to macroscopic changes in tissue formation and embryonic transition. Specific associations of cyclin-cdk complexes and programmed induction of cell cycle components regulate development. Presence or absence of particular components in various organs points to roles that these proteins may play during embryogenesis. For example, CIP-family inhibitors such as p21cip1 and p27kip1 were observed in the heart while INK-family inhibitors were not detected in this organ (Koh et al, 1998). P27kip1 was further implicated in the differentiation potential of cardiomyocytes and myoblasts (Zabludoff et al, 1998). Moreover, studies of induction kinetics of cell cycle components in specific cell types showed that cyclin D1 was found primarily in external granular cells while p27kip1 was found in purkinje cells in postnatal cerebellum development (Watanabe et al, 1998b).

Introduction of genomic instability often leads to tumorigenesis, with cell cycle components participating in these abnormalities. Co-transfection of cyclin E with Ha-ras into primary rat embryo fibroblasts resulted in focus formation and growth on soft agar (Haas et al, 1997) and may therefore act as an oncoprotein with a potential to induce genomic instability. Furthermore, G1 cyclin complexes including p21cip1, p27kip1, and p16ink4a may abolish cyclin E-induced oncogenesis (Haas et al, 1997). Immunohistochemical staining of malignant lymphomas for cyclin E showed that high expression correlated with poor prognosis (Erlanson et al, 1998). Reduced p27kip1 levels correlated with tumor aggressiveness in certain human cancers: aberrant p27kip1 expression may be associated with human esophageal squamous cell carcinoma (Anayama et al, 1998). Levels of nuclear p27kip1 immunoreactivity in primary tumors could predict recurrence and survival among patients with localized prostate cancer to determine prognosis (Cote et al, 1998, Tsihlias et al, 1998).

P27kip1 was observed to safeguard against excessive cell proliferation in certain pathophysiologies. Renal functions of p27kip1 knockout mice were reduced with associated increase in glomerular cell proliferation, apoptosis, and matrix protein accumulation. P27kip1 thus may have a protective role against inflammatory injury in cells and tissues (Ophascharoensuk et al, 1998). In human malignant glioma cells, treatment with antisense telomerase resulted in either apoptosis or differentiation, regulation of that is dependent upon induction of p27kip1 or interleukin-1b-converting enzyme (ICE). Introduction of antisense p27kip1 oligonucleotides into telomerase-inhibited surviving subpopulations of malignant glioma cells induced apoptosis, suggesting a role in p27kip1 protection against apoptosis in differentiating glioma cells (Kondo et al, 1998).

Waning of proliferative or developmental potential of a cell marks its entry into senescence. Cdk-inhibitors are distinguishing markers of this protective mechanism against the emergence of immortal clones. Induction of the senescent phenotype in young fibroblasts via ectopic expression of p21cip1 and p16ink4a reduced the proliferative capacity that included pRB-hypophosphorylation, increased PAI-I expression, and the appearance of senescence-associated-b-galactosidase activity (McConnell et al, 1998).

Whereas modulating protein expression in tissues is important in determining cellular outcome, the titration mechanisms from interacting cell cycle components also contribute significantly. Varying protein levels of one particular cell cycle component redistributes all participating factors in that circuit. For example, up-regulation of cyclin D1 changes the ratio of cyclin D1-cdk 4 to p16ink4a-cdk 4 by stoichiometric competition of cyclin D1 with p16ink4a for cdk 4. Increasing cyclin D1-cdk 4 complexes in turn titrates p27kip1 from cyclin E-cdk 2, which frees cyclin E-cdk 2 for phosphorylation. Since signal-transducing events drive the interactions between these cell cycle components, manipulating mitogenic and inhibitory signals allows dissection of pathways crucial for key events in G1/S progression. Phase G1/S progression may therefore be studied under controlled conditions using selected signals known to stimulate or inhibit key cell cycle components to elucidate the biochemical and molecular nature of cell cycle events.

The Cancer Challenge

Biotechnological revolution has created novel therapeutic intervention. Understanding components of cellular signaling pathways is key to the rational design of small molecules. Dissecting the convergence or divergence of various signaling strategies provides a starting point in determining targets for therapeutic intervention. With improving gene therapeutics and molecular therapeutics, regulatory dysfunctions in cellular abnormalities including tumor formation may now be treated at the genetic and molecular level.

The cancer challenge continues to be distinguishing between normal cells and transformed cells. Identifying unique cell markers enables engineering of therapeutic molecules specific to the difference for treatment with the least amount of toxicity to normal cells. The dilemma is that many components dysregulated in the transformed state also have important cellular functions in the normal state. Targeting dysregulated components often sacrifice normal cellular processes. An example is utilizing the Ras oncogene as a possible therapeutic target. The protein product of ras oncogene is a farnesyltransferase often altered in transformation. Ras, in turn, activates various signaling molecules, all of which represents potential disruption points leading to transformation. However, ras also functions in normal cellular processes, and disrupting ras may sacrifice other critical cellular functions dependent on its activity.

Many current strategies in cancer therapy are still limited by toxicity to normal tissues. Drugs intercalating DNA to take advantage of increased replication rate in transformed cells also affect rapidly dividing normal cells, such as those of hair follicles and gastrointestinal tracts. Selectivity of anti-cancer compounds depends on the particular cancers. Combination therapy has shown greater efficacy than sole administration of a compound. New therapeutic applications include gene therapy, antisense therapy, angiogenesis inhibitors, and cytokine or antibody therapies. Targeting cellular death processes have also come into focus as therapeutic strategies.

Recently, targets of molecular therapy shifted from upstream signals to downstream cell cycle components such as cyclins and cyclin-inhibitors. Targeting proteins involved in cell proliferation is one approach to halting abnormal growth. Another approach is to target apoptotic pathways. Apoptosis is, like cell cycle progression, comprised of a common underlying network of modulatory and effector activities. Different cyclin-cdk pairs were observed to correspond to different instances of cell death. Understanding regulatory pathways responsible for changes in cell cycle dynamics by selective "pushing" (stimulatory) or "pulling" (inhibitory) component dynamics helps the mining for new treatment targets.

Since its discovery, TGF-β1 has shown excellent potential as a tissue wound repair and bone repair medication, as evidenced by an initial patent awarded to Genentech in 1989 (US Patent ID#4,886,747). However, because signaling mechanisms remain unclear in terms of how TGF-β modulates other growth factors as well as determining stimulation versus inhibition, this theoretically promising growth factor currently could not meet development criteria without further understanding of its biochemical mechanisms of action. Wortmannin and rapamycin are also promising anti-tumor drugs, although the toxicity of wortmannin limits clinical usage. Constructing analog compounds may bring comparable therapeutic effects without the corresponding cytotoxicity from wortmannin. Rapamycin has been successfully used to prevent graft rejection during transplantation and is currently in active clinical trials (Phase I/II) as a treatment for graft-versus-host disease in patients with hematologic disorders receiving bone marrow transplants from mismatched-related donors. This thesis investigates the roles of signaling pathways regulating cell cycle progression by TGF-β1, rapamycin, and wortmannin.

Literature Cited

Albrecht JH, Poon RY, Ahonen CL, Rieland BM, Deng C, Crary GS. Oncogene 1998 Apr 23;16(16):2141-50. Involvement of p21 and p27 in the regulation of CDK activity and cell cycle progression in the regenerating liver.

Anayama T, Furihata M, Ishikawa T, Ohtsuki Y, Ogoshi S. Int J Cancer 1998 Aug 21;79(4):439-43. Positive correlation between p27Kip1 expression and progression of human esophageal squamous cell carcinoma.

Anghileri LJ. Arch Geschwulstforsch 1983;53(1):1-8. Warburg's cancer theory revisited: a fundamentally new approach.

Brusselbach S, Nettelbeck DM, Sedlacek HH, Muller R. Int J Cancer 1998 Jul 3;77(1):146-52. Cell cycle-independent induction of apoptosis by the anti-tumor drug Flavopiridol in endothelial cells.

Chen Y, Bhushan A, Vale W. Proc Natl Acad Sci U S A 1997 Nov 25;94(24):12938-43. Smad8 mediates the signaling of the ALK-2 [corrected] receptor serine kinase.

Cote RJ, Shi Y, Groshen S, Feng AC, Cordon-Cardo C, Skinner D, Lieskovosky G. J Natl Cancer Inst 1998 Jun 17;90(12):916-20. Association of p27Kip1 levels with recurrence and survival in patients with stage C prostate carcinoma.

de Vos S, Miller CW, Takeuchi S, Gombart AF, Cho SK, Koeffler HP. Genes Chromosomes Cancer 1995 Nov;14(3):164-70. Alterations of CDKN2 (p16) in non-small cell lung cancer.

Dickens M, Svitek CA, Culbert AA, O'Brien RM, Tavare JM. J Biol Chem 1998 Aug 7;273(32):20144-9. Central role for phosphatidylinositide 3-kinase in the repression of glucose-6-phosphatase gene transcription by insulin.

Dong F, Agrawal D, Bagui T, Pledger WJ. Mol Biol Cell 1998 Aug;9(8):2081-92. Cyclin D3-associated kinase activity is regulated by p27kip1 in BALB/c 3T3 cells.

Dowdy SF, Hinds PW, Louie K, Reed SI, Arnold A, Weinberg RA. Cell 1993 May 7;73(3):499-511. Physical interaction of the retinoblastoma protein with human D cyclins.

Erlanson M, Portin C, Linderholm B, Lindh J, Roos G, Landberg G. Blood 1998 Aug 1;92(3):770-7. Expression of cyclin E and the cyclin-dependent kinase inhibitor p27 in malignant lymphomas-prognostic implications.

Fang F, Orend G, Watanabe N, Hunter T, Ruoslahti E. Science 1996 Jan 26;271(5248):499-502. Dependence of cyclin E-CDK2 kinase activity on cell anchorage.

Gemma A, Takenoshita S, Hagiwara K, Okamoto A, Spillare EA, McMemamin MG, Hussain SP, Forrester K, Zariwala M, Xiong Y, Harris CC. Int J Cancer 1996 Nov 27;68(5):605-11. Molecular analysis of the cyclin-dependent kinase inhibitor genes p15INK4b/MTS2, p16INK4/MTS1, p18 and p19 in human cancer cell lines.

Guidal-Giroux C, Gerard B, Cave H, Duval M, Rohrlich P, Elion J, Vilmer E, Grandchamp B. Br J Haematol 1996 Feb;92(2):410-9. Deletion mapping indicates that MTS1 is the target of frequent deletions at chromosome 9p21 in paediatric acute lymphoblastic leukaemias.

Haas K, Johannes C, Geisen C, Schmidt T, Karsunky H, Blass-Kampmann S, Obe G, Moroy T. Oncogene 1997 Nov 20;15(21):2615-23. Malignant transformation by cyclin E and Ha-Ras correlates with lower sensitivity towards induction of cell death but requires functional Myc and CDK4.

Hall M, Bates S, Peters G. Oncogene 1995 Oct 19;11(8):1581-8. Evidence for different modes of action of cyclin-dependent kinase inhibitors: p15 and p16 bind to kinases, p21 and p27 bind to cyclins.

Hara K, Yonezawa K, Weng QP, Kozlowski MT, Belham C, Avruch J. J Biol Chem 1998 Jun 5;273(23):14484-94. Amino acid sufficiency and mTOR regulate p70 S6 kinase and eIF-4E BP1 through a common effector mechanism.

Harper JW, Adami GR, Wei N, Keyomarsi K, Elledge SJ. Cell 1993 Nov 19;75(4):805-16. The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases.

Hartwell LH and Weinert TA. Science 1989 Nov 3;246(4930):629-34. Checkpoints: controls that ensure the order of cell cycle events.

Hashemolhosseini S, Nagamine Y, Morley SJ, Desrivieres S, Mercep L, Ferrari S. J Biol Chem 1998 Jun 5;273(23):14424-9. Rapamycin inhibition of the G1 to S transition is mediated by effects on cyclin D1 mRNA and protein stability.

Hatada I, Ohashi H, Fukushima Y, Kaneko Y, Inoue M, Komoto Y, Okada A, Ohishi S, Nabetani A, Morisaki H, Nakayama M, Niikawa N, Mukai T. Nat Genet 1996 Oct;14(2):171-3. An imprinted gene p57KIP2 is mutated in Beckwith-Wiedemann syndrome.

Herman JG, Merlo A, Mao L, Lapidus RG, Issa JP, Davidson NE, Sidransky D, Baylin SB. Cancer Res 1995 Oct 15;55(20):4525-30. Inactivation of the CDKN2/p16/MTS1 gene is frequently associated with aberrant DNA methylation in all common human cancers.

Inaba T, Matsushime H, Valentine M, Roussel MF, Sherr CJ, Look AT. Genomics 1992 Jul;13(3):565-74. Genomic organization, chromosomal localization, and independent expression of human cyclin D genes.

Izumoto S, Arita N, Ohnishi T, Hiraga S, Taki T, Hayakawa T. Cancer Lett 1995 Nov 6;97(2):241-7. Homozygous deletions of p16INK4A/MTS1 and p15INK4B/MTS2 genes in glioma cells and primary glioma tissues.

Jiang H, Chou HS, Zhu L. Mol Cell Biol 1998 Sep;18(9):5284-90. Requirement of cyclin E-Cdk2 inhibition in p16(INK4a)-mediated growth suppression.

Jones CJ, Shaw JJ, Wyllie FS, Gaillard N, Schlumberger M, Wynford-Thomas D. Mol Cell Endocrinol 1996 Jan 15;116(1):115-9. High frequency deletion of the tumour suppressor gene P16INK4a (MTS1) in human thyroid cancer cell lines.

Kato A, Ota S, Bamba H, Wong RM, Ohmura E, Imai Y, Matsuzaki F. Biochem Biophys Res Commun 1998 Apr 7;245(1):70-4. Regulation of cyclin D-dependent kinase activity in rat liver regeneration.

Kato J, Matsushime H, Hiebert SW, Ewen ME, Sherr CJ. Genes Dev 1993 Mar;7(3):331-42. Direct binding of cyclin D to the retinoblastoma gene product (pRb) and pRb phosphorylation by the cyclin D-dependent kinase CDK4.

Kim TA, Cutry AF, Kinniburgh AJ, Wenner CE. Cancer Lett 1993 Jul 30;71(1-3):125-32. Transforming growth factor beta 1-induced delay of cell cycle progression and its association with growth-related gene expression in mouse fibroblasts.

Koff A, Cross F, Fisher A, Schumacher J, Leguellec K, Philippe M, Roberts JM. Cell 1991 Sep 20;66(6):1217-28. Human cyclin E, a new cyclin that interacts with two members of the CDC2 gene family.

Koh KN, Kang MJ, Frith-Terhune A, Park SK, Kim I, Lee CO, Koh GY. J Mol Cell Cardiol 1998 Mar;30(3):463-74. Persistent and heterogenous expression of the cyclin-dependent kinase inhibitor, p27KIP1, in rat hearts during development.

Kondo S, Tanaka Y, Kondo Y, Hitomi M, Barnett GH, Ishizaka Y, Liu J, Haqqi T, Nishiyama A, Villeponteau B, Cowell JK, Barna BP. FASEB J 1998 Jul;12(10):801-11. Antisense telomerase treatment: induction of two distinct pathways, apoptosis and differentiation.

Kuan CJ, al-Douahji M, Shankland SJ. J Am Soc Nephrol 1998 Jun;9(6):986-93. The cyclin kinase inhibitor p21WAF1, CIP1 is increased in experimental diabetic nephropathy: potential role in glomerular hypertrophy.

Lee MH, Reynisdottir I, Massague J. Genes Dev 1995 Mar 15;9(6):639-49. Cloning of p57KIP2, a cyclin-dependent kinase inhibitor with unique domain structure and tissue distribution.

Leng X, Connell-Crowley L, Goodrich D, Harper JW. Curr Biol 1997 Sep 1;7(9):709-12. S-Phase entry upon ectopic expression of G1 cyclin-dependent kinases in the absence of retinoblastoma protein phosphorylation.

Liu F, Hata A, Baker JC, Doody J, Carcamo J, Harland RM, Massague J. Nature 1996 Jun 13;381(6583):620-3. Human type II receptor for bone morphogenic proteins (BMPs): extension of the two-kinase receptor model to the BMPs.

Matsuo T, Thiele CJ. Oncogene 1998 Jun 25;16(25):3337-43. p27Kip1: a key mediator of retinoic acid induced growth arrest in the SMS-KCNR human neuroblastoma cell line.

Matsuoka S, Thompson JS, Edwards MC, Bartletta JM, Grundy P, Kalikin LM, Harper JW, Elledge SJ, Feinberg AP. Proc Natl Acad Sci U S A 1996 Apr 2;93(7):3026-30. Imprinting of the gene encoding a human cyclin-dependent kinase inhibitor, p57KIP2, on chromosome 11p15.

McConnell BB, Starborg M, Brookes S, Peters G. Curr Biol 1998 Mar 12;8(6):351-4. Inhibitors of cyclin-dependent kinases induce features of replicative senescence in early passage human diploid fibroblasts.

McConnell BB, Gregory FJ, Stott FJ, Hara E, Peters G. Mol Cell Biol 1999 Mar;19(3):1981-9. Induced expression of p16(INK4a) inhibits both CDK4- and CDK2-associated kinase activity by reassortment of cyclin-CDK-inhibitor complexes.

Mitra J, Dai CY, Somasundaram K, El-Deiry WS, Satyamoorthy K, Herlyn M, Enders GH. Mol Cell Biol 1999 May;19(5):3916-28. Induction of p21(WAF1/CIP1) and inhibition of Cdk2 mediated by the tumor suppressor p16(INK4a).

Moses HL, Coffey RJ Jr, Leof EB, Lyons RM, Keski-Oja J. J Cell Physiol Suppl 1987;Suppl 5:1-7. Transforming growth factor beta regulation of cell proliferation.

Motokura T, Bloom T, Kim HG, Juppner H, Ruderman JV, Kronenberg HM, Arnold A. Nature 1991 Apr 11;350(6318):512-5. A novel cyclin encoded by a bcl1-linked candidate oncogene.

Moustakas A, Kardassis D. Proc Natl Acad Sci U S A 1998 Jun 9;95(12):6733-8. Regulation of the human p21/WAF1/Cip1 promoter in hepatic cells by functional interactions between Sp1 and Smad family members.

Nakao A, Afrakhte M, Moren A, Nakayama T, Christian JL, Heuchel R, Itoh S, Kawabata M, Heldin NE, Heldin CH, ten Dijke P. Nature 1997 Oct 9;389(6651):631-5. Identification of Smad7, a TGFbeta-inducible antagonist of TGF-beta signalling.

Ohtsubo M, Roberts JM. Science 1993 Mar 26;259(5103):1908-12. Cyclin-dependent regulation of G1 in mammalian fibroblasts.

Ophascharoensuk V, Fero ML, Hughes J, Roberts JM, Shankland SJ. Nat Med 1998 May;4(5):575-80. The cyclin-dependent kinase inhibitor p27Kip1 safeguards against inflammatory injury.

Orend G, Hunter T, Ruoslahti E. Oncogene 1998 May;16(20):2575-83. Cytoplasmic displacement of cyclin E-cdk2 inhibitors p21Cip1 and p27Kip1 in anchorage-independent cells.

Parry D, Mahony D, Wills K, Lees E. Mol Cell Biol 1999 Mar;19(3):1775-83. Cyclin D-CDK subunit arrangement is dependent on the availability of competing INK4 and p21 class inhibitors.

Pines J. Curr Opin Cell Biol 1992 Apr;4(2):144-8. Cell proliferation and control.

Polyak K, Kato JY, Solomon MJ, Sherr CJ, Massague J, Roberts JM, Koff A. Genes Dev 1994 Jan;8(1):9-22. p27Kip1, a cyclin-Cdk inhibitor, links transforming growth factor-beta and contact inhibition to cell cycle arrest.

Poon RY, Hunter T. Science 1995 Oct 6;270(5233):90-3. Dephosphorylation of Cdk2 Thr160 by the cyclin-dependent kinase-interacting phosphatase KAP in the absence of cyclin.

Quelle DE, Zindy F, Ashmun RA, Sherr CJ. Cell 1995 Dec 15;83(6):993-1000. Alternative reading frames of the INK4a tumor suppressor gene encode two unrelated proteins capable of inducing cell cycle arrest.

Sano Y, Harada J, Tashiro S, Gotoh-Mandeville R, Maekawa T, Ishii S. J Biol Chem 1999 Mar 26;274(13):8949-57. ATF-2 is a common nuclear target of Smad and TAK1 pathways in transforming growth factor-beta signaling.

Seghezzi W, Chua K, Shanahan F, Gozani O, Reed R, Lees E. Mol Cell Biol 1998 Aug;18(8):4526-36. Cyclin E associates with components of the pre-mRNA splicing machinery in mammalian cells.

Sherr CJ, Roberts JM. Genes Dev 1995 May 15;9(10):1149-63. Inhibitors of mammalian G1 cyclin-dependent kinases.

Suzuki A, Chang C, Yingling JM, Wang XF, Hemmati-Brivanlou A. Dev Biol 1997 Apr 15;184(2):402-5. Smad5 induces ventral fates in Xenopus embryo.

Tamatani T, Kobayashi H, Tezuka K, Sakamoto S, Suzuki K, Nakanishi T, Takigawa M, Miyano T. Biochem Biophys Res Commun 1998 Oct 29;251(3):748-52. Establishment of the enzyme-linked immunosorbent assay for connective tissue growth factor (CTGF) and its detection in the sera of biliary atresia.

Thannickal VJ, Aldweib KD, Rajan T, Fanburg BL. Biochem Biophys Res Commun 1998 Oct 20;251(2):437-41. Upregulated expression of fibroblast growth factor (FGF) receptors by transforming growth factor-beta1 (TGF-beta1) mediates enhanced mitogenic responses to FGFs in cultured human lung fibroblasts.

Tsihlias J, Kapusta LR, DeBoer G, Morava-Protzner I, Zbieranowski I, Bhattacharya N, Catzavelos GC, Klotz LH, Slingerland JM. Cancer Res 1998 Feb 1;58(3):542-8. Loss of cyclin-dependent kinase inhibitor p27Kip1 is a novel prognostic factor in localized human prostate adenocarcinoma.

Watanabe H, Pan ZQ, Schreiber-Agus N, DePinho RA, Hurwitz J, Xiong Y. Proc Natl Acad Sci U S A 1998a Feb 17;95(4):1392-7. Suppression of cell transformation by the cyclin-dependent kinase inhibitor p57KIP2 requires binding to proliferating cell nuclear antigen.

Watanabe G, Pena P, Shambaugh GE 3rd, Haines GK 3rd, Pestell RG. Brain Res Dev Brain Res 1998b Jun 15;108(1-2):77-87. Regulation of cyclin dependent kinase inhibitor proteins during neonatal cerebella development.

Xiong Y, Menninger J, Beach D, Ward DC. Genomics 1992 Jul;13(3):575-84. Molecular cloning and chromosomal mapping of CCND genes encoding human D-type cyclins.

Zabludoff SD, Csete M, Wagner R, Yu X, Wold BJ. Cell Growth Differ 1998 Jan;9(1):1-11. p27Kip1 is expressed transiently in developing myotomes and enhances myogenesis.

Zhang Y, Musci T, Derynck R. Curr Biol 1997 Apr 1;7(4):270-6. The tumor suppressor Smad4/DPC 4 as a central mediator of Smad function.

Zhao J, Sime PJ, Bringas P Jr, Gauldie J, Warburton D. Mech Dev 1998 Mar;72(1-2):89-100. Epithelium-specific adenoviral transfer of a dominant-negative mutant TGF-beta type II receptor stimulates embryonic lung branching morphogenesis in culture and potentiates EGF and PDGF-AA.