Growth factor stimulation of a resting G0 cell to enter the early G1 phase of the cell cycle and transition across the G1 restriction point into the late G1 phase, followed by entrance into S phase and DNA synthesis, requires the coordinated efforts of multiple cyclin:Cdk complexes and an increased metabolism. Previous research has demonstrated the roles of tumor-suppressor genes in the regulation of cell cycle progression, specifically the G1 phase of the cell cycle. An important negative regulator of G1 cell cycle progression at the restriction point is pRB, the product of the retinoblastoma tumor-suppressor gene. pRB targets cellular transcription factors, such as members of the E2F family. E2F transcription factors are involved in driving the expression of genes involved in DNA synthesis after transition through the restriction point into late G1. Transition across the restriction point irrevocably commits a cell to continue through the rest of the cell cycle. This key growth regulatory checkpoint balances the appropriate requisite level of metabolism with growth factor stimulation.
Phosphorylation on 16 cyclin-dependent kinase (Cdk) consensus sites by G1 cyclin:Cdk complexes regulates pRB. In G0 cells, pRB is unphosphorylated and does not associate with E2F transcription factors, suggesting that this form is inactive. As cells progress into early G1, pRB becomes hypophosphorylated on Cdk sites and associates with E2Fs. At the restriction point, pRB becomes inactivated by hyperphosphorylation on Cdk sites and E2Fs are released. Due to continued Cdk activity, pRB remains hyperphosphorylated through S, G2, and M phases. Thus, in response to growth factor stimulation, pRB is differentially regulated by cyclin:Cdk complexes in early and late G1. Although the RB gene is genetically altered in a low percentage of human malignancies (<1 percent), the p16INK4a, and cyclin D1, Cdk4, and Cdk6 proto-oncogenes are mutated in most, if not all, human malignancies. These observations underscore the importance of understanding the exact physiological and biochemical role that these genes play during G1-phase cell cycle progression and oncogenesis.
We are interested in understanding the complex interactions involved in regulation of cell cycle progression at the G1 restriction point and the aberrations that occur during oncogenesis.
Differential Regulation of pRB by G1 Cyclin:Cdk Complexes
In cycling normal and tumor cells, cyclin D:Cdk4/6 complexes are constitutively expressed and active. In contrast, cyclin E:Cdk2 complexes are activated at or near the late-G1 restriction point and remain active into S phase. Physiologically, pRB exists as a hypophosphorylated protein bound to E2F transcription factors, while cyclin D:Cdk4/6 complexes are active in early G1. However, congruent with activation of cyclin E:Cdk2 complexes, inactive, hyperphosphorylated pRB first appears at or near the restriction point. Thus, pRB oscillates between an active, hypophosphorylated form and inactive, hyperphosphorylated forms in cycling cells. These observations suggest that activation of cyclin E:Cdk2 complexes is the critical rate-limiting step to pRB inactivation.
To address this problem while avoiding overexpression of the cyclin:Cdk proteins, we sought to use the protein transduction method developed in our lab to introduce into ~100 percent of primary or transformed cells specific negative regulators of cyclin D:Cdk4/6 and cyclin E:Cdk2 complexes. Recombinant fusion proteins were made that harbored an amino-terminal protein transduction domain from the HIV TAT protein. Purified TAT-fusion proteins produced in bacteria rapidly (<10 min) transduce into 100 percent of all cell types. Transduction of TAT-p16 protein, a specific negative regulator of Cdk4/6, into primary cells resulted in a G1 cell cycle arrest, loss of hypophosphorylated pRB, and importantly, loss of pRB:E2F complexes. Specific inactivation of cyclin E:Cdk2 complexes by transduction of TAT-Cdk2 dominant-negative (DN) protein resulted in a G1 cell cycle arrest, loss of cyclin E:Cdk2 kinase activity, and loss of hyperphosphorylated pRB. Importantly, pRB remained hypophosphorylated and associated with E2F transcription factors in the presence of active cyclin D:Cdk4/6 complexes in TAT-Cdk2 DN treated cells.
We conclude that cyclin D:Cdk4/6 complexes activate pRB by hypophosphorylation in early G1 and that cyclin E:Cdk2 complexes are solely responsible for the physiologic hyperphosphorylation and inactivation of pRB at the late G1 restriction point. These observations reveal an entirely new level of G1 Cdk regulation of pRB. However, and perhaps more importantly, the rate-limiting activator of cyclin E:Cdk2 complexes at the restriction point remains unknown.
Inactive cyclin E:Cdk2 complexes are present in the early-G1 phase of cycling cells and become activated at the restriction point by an unknown mechanism. To address this problem, we revised the current model of G1 cell cycle progression using a data-driven mathematical modeling systems biology approach that incorporates the experimental observations from above. We constructed computational models of G1 progression to determine whether the models could account for the observed data. To account for the experimental observations, we included an unknown modifier controlled by metabolic signals to regulate activation of cyclin E:Cdk2 complexes. Treatment of cells with cell cycle inhibitors does not slow metabolism, whereas inhibitors of cellular metabolism induce cell cycle arrest, suggesting that metabolism directly or indirectly regulates G1 cell cycle progression. This change in network topology resulted in a revised model that accurately predicts the dynamics of Cdk2 activity and pRb inactivation. Therefore, we are performing experiments to identify the nature and regulation of the metabolic modifier.
In pursuing metabolic regulators of cell cycle progression, we discovered that intracellular reactive oxygen species (ROS) levels increase as cells progress through G1 and into S phase. High ROS levels induce oxidative damage and serve as a pathogen defense mechanism. However, cells also utilize low physiologic ROS levels for cell-to-cell communication (nitric oxide) and in proliferation. Inhibition of ROS by antioxidants activates a novel late-G1-phase checkpoint that arrests cells containing active cyclin E:Cdk2 complexes, inactive hyperphosphorylated pRB, but no cyclin A protein:Cdk2 complexes. Further analysis determined that the anaphase-promoting complex APCCdh1, which regulates cyclin A protein by degradation, remains active in antioxidant-treated cells. Thus, the checkpoint occurs after transition across the growth factor—dependent late-G1 restriction point and regulates accumulation of cyclin A protein (and likely others) by continued APCCdh1 activity, preventing initiation of DNA synthesis. This late G1 ROS metabolic checkpoint is conserved in yeast, suggesting that this is an intrinsic evolutionary feedback mechanism to monitor the status of cellular metabolism prior to DNA synthesis commitment. We are investigating the potential sources of ROS in the cell—including mitochondria, peroxisomes, growth factor receptor signaling, and NADPH-oxidase—and also the candidate species of reactive oxygen that may be involved.
Transduction of Full-Length Proteins
Peptides and proteins have been evolutionarily selected to perform specific functions and interact with specific downstream targets. Due to their size and composition, peptides and proteins have no bioavailability to enter cells. However, addition of the small (9-mer) cationic peptide from the HIV-1 TAT protein, termed a protein transduction domain (PTD), results in the rapid entrance into cells in culture and into most, if not all, tissues in mouse models in vivo. The TAT PTD has been used to deliver a wide variety of biologically active, macromolecular cargo for the treatment of multiple preclinical disease models, including cancer and stroke.
Although TAT-mediated transduction was first reported in 1988, the mechanism of transduction remained unknown. Because of TAT PTD's strong cell surface binding, early assumptions regarding cellular uptake suggested a direct penetration mechanism across the lipid bilayer by a temperature- and energy-independent process. However, using a transducible TAT-Cre recombinase reporter assay on live cells, we found that after an initial ionic cell surface interaction, TAT-fusion proteins are rapidly internalized by fluid-phase lipid raft-dependent macropinocytosis, a specialized form of endocytosis. Moreover, transduction was independent of interleukin-2 receptor/raft-, caveolae-, clathrin-mediated endocytosis and phagocytosis. Using this information, we developed a transducible, pH-sensitive fusogenic TAT-HA2 peptide that dramatically enhanced TAT-Cre escape from macropinosomes. Together these observations provide a working model for the mechanism of transduction into cells and allow further development of novel, biologically active, transducible macromolecular therapeutics.
In 85 percent of patients, metastatic ovarian peritoneal carcinomatosis is resistant to current chemotherapy treatment and results in a devastating 20 percent 5-year survival rate. Therapeutics that restore genes inactivated during oncogenesis are predicted to be more potent and specific than current therapies. Experiments with viral vectors have demonstrated the theoretical utility of expressing the p53 tumor-suppressor gene in cancer cells. However, clinically useful alternative approaches for introducing p53 activity into cancer cells are clearly needed. It has been hypothesized that direct reactivation of endogenous p53 protein in cancer cells will be beneficial, but few tests of this hypothesis have been carried out in vivo. We designed a transducible D-isomer RI-TATp53C' peptide that activates the p53 protein in cancer cells, but not normal cells. RI-TATp53C' peptide treatment of preclinical terminal peritoneal carcinomatosis and peritoneal lymphoma models results in significant increases in life span (>6-fold) and the generation of disease-free animals. These observations show that specific activation of endogenous p53 activity by a macromolecular agent is therapeutically effective in preclinical models of terminal human malignancy. Our results suggest that TAT-mediated transduction may be useful for delivery of macromolecular therapeutics to treat malignant cells in vivo.
Lastly, transmissible spongiform encephalopathies, including variant Creutzfeldt-Jakob disease in humans and mad cow disease in cattle, represent an emerging class of fatal neurodegenerative disorders that are characterized by template-directed protein misfolding of the host wild-type cellular prion protein (PrPC). PrPC is a cell surface–associated protein whose normal physiologic function is unclear; however, host cell exposure to infectious, protease-resistant prion protein (PrPSc) results in the conformational conversion of endogenous PrPC to the pathological PrPSc isoform. Recently, the presence of small amounts of cytoplasmic PrP has been shown to induce neuropathologies. The mechanism that exogenous PrP uses to enter cells remains unknown, however. We found that, similar to the mechanism of TAT-mediated protein transduction, prion proteins contain an amino-terminal cationic transduction domain that stimulates lipid raft–dependent macropinocytosis. Moreover, PrP contains a second amino-terminal domain that enhances escape from macropinosomes into cells. Moreover, treatment with inhibitors of macropinocytosis both prevented prion protein infection of cells and conversion of cellular prion protein into the pathological scrapie form of the protein. These observations provide a molecular mechanism for infection of neuronal cells by exogenous prion protein following host exposure to contaminated material.
