Role of nutrient availability on proliferation and cell cycle excution of immortalized and kras transformed mouse fibroblastic

Mammalian cells proliferate, differentiate or die in response to extracellular signals as growth factors and nutrients. Cancer is essentially a disease in which cells have lost responsiveness to many of these signals and to normal checks on cell proliferation. Therefore, it may not be surprising that tumor cells, in order to meet the increased requirements of proliferation, often display fundamental changes in pathways of energy metabolism and nutrient uptake (Garber, 2006). In particular, several studies shown that the process of tumorigenesis is often associated with altered metabolism of two major nutrients, glucose and glutamine (Mazurek et al, 1998; Chiaradonna et al., 2006; Gaglio et al., 2009). Moreover, this metabolic changes and cellular sensitivity to these nutrients can be induced or influenced by oncogenic transformation as i.e. Ras mutation (Chiaradonna et al., 2006; Gaglio et al., 2009) that has been found in 25% of human cancers. Glucose and glutamine are involved in multiple pathways required for cell proliferation and survival both in normal and transformed cells. The role of these pathways in the survival of transformed cells is mostly based either from the fact that the pathways in question can be regulated by oncogenes, or that cell death following nutrients shortage is associated with changes in the activation state of these pathways. In particular, in this work has been studied the response of K-Ras transformed cells to glucose and glutamine availability. Transformed cells have been shown to have a particular dependence on aerobic glycolysis compared to normal cells. Indeed, proliferation analysis of asynchronous K-Ras transformed fibroblasts as compared to normal cells, grown both in high and low glucose (25mM and 1mM), indicated that transformed cells showed an higher proliferation potential in 25mM glucose and lost completely this proliferative advantage in low glucose (1mM). Moreover, the strong reduction of glucose availability, observed at 72hrs in cells grown in 1mM glucose, induced an enhanced apoptosis only in transformed cells. Indeed, the apoptotic process activated in normal cells was glucose independent and probably correlated to prolonged contact inhibition. This effect has been validated by both Annexin V/PI staining and caspases-3 activation. Since transformed cells were characterized by strong reduction of proliferation as well as apoptosis at low glucose concentration, it has been analyzed the effects of Ras activation and glucose shortage on the cell cycle machinery, in particular during G1/S transition in synchronized normal and transformed cells. These results indicated that the timing of G1/S transition execution was glucose independent in both cell lines. Therefore, oncogenic Ras expression is able to induce a greater sensitivity to glucose shortage as compared to normal cells (decreased proliferation and enhanced apoptosis) only if the glucose shortage is persistent. Another metabolic adaptation of cancer cells, that has been studied, is their propensity to exhibit increased glutamine consumption. The results indicated that in asynchronous normal cells, contact inhibition, regardless of glutamine availability, brings to down-regulation of Akt that together with AMPK up-regulation, observed at low glutamine, concurs to TOR pathway inactivation. As a result, the expression of cyclin D, E and A is down regulated, pRb phosphorylation is strongly reduced, p27kip1 level is increased and its localization becomes preferentially nuclear, establishing therefore a condition that bring to a G1- cell cycle arrest. In synchronous normal cells, glutamine shortage slows the G2/M transition, indicating a possible role of glutamine in such cell cycle phase. In K-ras transformed cells, in which the level of activated Ras-GTP is very high (Nagase et al., 1995) and the contact inhibition is less efficient (Nagase et al., 1995), the deprivation of glutamine affects Akt and AMPK in a way opposite to that observed in normal cells, leaving the TOR pathway at least partially activated. This event allows sizable expression of cyclin D (at least until 72 hrs), E and A, sustained pRb phosphorylation, decreased p27kip1 and its preferential cytoplasmic localization, conditions that, taken together, promote entrance into S phase. Surprisingly, in condition of glutamine shortage, transformed asynchronous cells accumulate in S phase. In synchronous transformed cells, glutamine shortage slows both the G1 to S and the G2/M transitions. Since glutamine is an important intermediate in purine and pyrimidine biosynthesis, glutamine exhaustion could deplete intracellular nucleotide pools, bringing in turn to a failure in the execution of a normal cell cycle (Christofk et al., 2008; Martinez-Diez et al., 2006). This hypothesis has been confirmed by the fact that the proliferation defect of transformed cells is rescued by adding the four deoxyribonucleotides (precursors of DNA polymerization) to low glutamine medium. Moreover, experiments performed in synchronized transformed cells have been shown that low-glutamine medium causes a 2 hrs delay in entering into S phase after serum re-addition. This effect on cell cycle timing is worsened by complete absence of glutamine, in which 4 hrs delay in entering into S phase was observed. These data strongly indicated that the effect of glutamine limitation in transformed cells was first to slow down the S phase traverse, then, when a more severe limitation was established, to stuck a large fraction of the cells population in S-phase. Indeed, addition of a mix of 10µM deoxyribonucleotides reverted completely S phase reaching. Therefore in cells exhibiting high metabolic rates, such as rapidly dividing cancer cells grown in vitro, glutamine, being the most readily available amino acid used as energy source, may became the major source to sustain protein and nucleic acid synthesis (Ziegler et al., 2001), especially when glucose levels are low and energy demand is high. However, analysis of the levels of mRNA, proteins and above all of ATP in normal and transformed cells grown in high and low glutamine availability, did not show particular differences, suggesting an important role of glutamine for nucleotides synthesis in K-ras transformed cells. In conclusion, glutamine shortage in K-ras transformed cells limits proliferation by inducing abortive S phase entrance, while glucose shortage in the same system enhanced cell death (Lopez-Rios et al., 2007; Mankoff et al., 2007). The differential effects of glutamine and glucose on cell viability are not a property of the transformed phenotype per se, but rather depend on the specific pathway being activated in transformation. It has been previously shown that nutrient shortage influence cell proliferation and G1/S transition of K-Ras transformed fibroblasts. To understand how intracellular and extracellular signals are transmitted to the cell-cycle machinery and how the latter adjusts its frequencies accordingly is one of the major challenges in molecular biomedicine. To this aim, a computational model of the cell cycle based on experimental data has been developed. Indeed biochemical and genetic studies can be combined with bioinformatics and biosystems approaches in order to sketch a plan of the regulatory circuits governing cell cycle progression in normal cells, firstly, and then in transformed cells. Taking in consideration that the trespassing of the Restriction Point influences the timing of the cell cycle execution and that such timing is influenced both by growth factors and nutrients availability, it has been initially identified the restriction point in normal mouse fibroblasts, synchronized in G0 by serum starvation and stimulated to re-enter into S-phase by readdition of serum. During the time course of re-entering into cell cycle, from G0 to G1/S phase, and in agree with restriction point reaching, it has been observed a gradual increase of cyclin D and cyclin E, a constant expression of Cdk4 and Cdk2 and an abrupt decrease of p27Kip1. Moreover, in quiescent cells, has been observed a completely nuclear localization of p27Kip1 and more cytoplasmic localization of Cdk4 and Cdk2. These data agreed with other results since, in most cases, the concentration of the kinase subunit is relatively constant, whereas the concentration of the cyclin subunit oscillates. This detailed study of G1/S transition in normal fibroblasts allowed a novel mathematical model develop. Because tumor cells often display a reduced dependence on growth factors or an increased dependence on some nutrients, an understanding of the cell cycle and a dynamical computational model that include regulatory aspect might help explain the changes leading to tumor formation.