Carcinogenesis is caused by multiple co-operating genetic lesions leading to a progressive deregulation of cellular signalling and cell cycle restriction point control. The mutations involved result in oncogene activation or loss of tumor-suppresser gene function. We are studying the mechanisms by which these mutant genes (e.g. activated ras/raf + myc or ras/raf + loss of p53 function) co-operate in malignant cell transformation. Recently we have made significant progress in identifying the molecular events involved. We have learnt that single oncogenes are insufficient to cause malignant transformation because they simultaneously induce signals stimulating and inhibiting cell growth. As a result cell proliferation remains restricted. In contrast, co-operating oncogenic lesions act in concert to disable such inhibitory signals while reinforcing the growth-promoting stimuli. This involves integration of multiple signals converging on the regulation of cell cycle-dependent kinase complexes.
------------------------------------------------------------------------
Tumor suppresser p53 and cell cycle inhibitor p21cip1 determine cellular response to ras and raf oncogenes:

Oncogenic lesions co-operating with Ras/Raf cause a dramatic switch in Ras/Raf signal specificity. For example, in Schwann cells and fibroblasts activated Ras/Raf signaling leads to a G1-specific cell cycle arrest. In contrast, in presence of dominant-negative (dn)-p53 or Myc, Ras/Raf activation causes rapid proliferation. Understanding the molecular architecture of this switch is thus likely to be important for the development of mechanistic models for multi-step carcinogenesis. We have analyzed this switching process using inducible Raf-steroid hormone receptor chimeras which can be activated within minutes and thus allow real time analysis. In these experiments, we have demonstrated genetically and biochemically that the Raf-induced G1 arrest requires strong activation of MAP kinase activity followed by the inhibition of cyclin-dependent kinases (cdk) via induction of the cyclin/cdk inhibitory protein p21cip1. In Schwann cells p21cip1 induction by Raf requires p53 activity and loss of p53 function abrogates the Raf-induced G1 arrest. Thus, p53 activity and the expression level of p21cip1 determine the cellular response to Ras and Raf (Fig. 1). Raf is also unable to inhibit cell cycle progression in fibroblasts lacking a functional p21cip1, although the induction of p21cip1 by Ras/Raf is independent of p53 in this cell type.

Fig. 1 Signaling by co-operating oncogenes converges at the regulation of cyclin/cdk complexes. In Schwann cells Ras or Raf oncogenes induce a G1-specific cell-cycle arrest via induction of p21Cip1 and the concomitant inhibition of cyclin/cdk activity. In the absence of functional p53 however, the p21Cip1 induction is suppressed and the growth arrest abolished. Instead the ability of Raf to activate cyclin/cdk activity and to induce cell cycle progression is revealed. Thus, Raf elicits either growth inhibitory or stimulatory signals depending on the presence of functional p53.
------------------------------------------------------------------------ Raf signal specificity can be determined by signal intensity:

The ability of Ras and Raf to cause such opposing cellular responses as the induction of DNA synthesis and the inhibition of proliferation is not only due to signal integration of convergent signals. In NIH 3T3 cells both of these responses to Ras and Raf have been observed in the absence of any co-operating genes. However, it remained unclear how this difference in signal specificity is controlled. We have addressed this question using a regulatable Raf-steroid hormone receptor fusion protein, which allows the titration of its kinase activity in response to the hormone added. We found that in NIH3T3 cells only a strong Raf signal can cause a G1-specific cell cycle arrest due to induction of p21Cip1. In contrast, moderate Raf activity induces DNA synthesis. At the same time cyclin D expression is maximally induced, while p21Cip1 expression is not affected. Thus, Raf signal specificity can be determined by modulation of signal strength presumably through the induction of distinct protein expression patterns (Fig. 2).

Fig. 2 Raf activation induces signals activating and inhibiting G1-S progression. The specificity of the cellular response is determined by modulation of signal intensity leading to distinct patterns of gene expression. Moderate activation of Raf is sufficient to induce cyclin D1, which ultimately triggers Rb phosphorylation and S-phase entry. In contrast, strong activation of Raf causes an arrest in G1 via induction p21Cip1 which inhibits the activity of G1-specific Cyclin/Cdk complexes.
-----------------------------------------------------------------------
Induction of cell cycle progession by Myc:

To gain insight into the molecular mechanisms by which Myc can induce cell cycle entry, we have investigated the events that lead to Myc-dependent activation of cyclin E/Cdk2 kinase. For this purpose we used Rat1 fibroblasts expressing a regulatable Myc-oestrogen receptor fusion protein (MycER). We found that MycER activation by 4-OH-tamoxifen (OHT) in the absence of other exogenous signals rapidly induces cyclin E transcription in a protein synthesis-independent manner. However, induction of cyclin E transcription in the same cells under control of a GalER-dependent promoter is insufficient for activation of cyclin E-dependent kinase activity and induction of DNA synthesis. Since activation of MycER on the other hand leads to an increase in cylin E/cdk2 kinase activity within three to four hours, Myc must affect additional function(s). We therefore investigated whether Myc might affect the interaction of the cell cycle inhibitor p27Kip1 (the major Cdk2 inhibitor in Rat 1 cells) with cyclin E/Cdk2 complexes. In these experiments we found that the preexisting pool of cyclin E/Cdk2 complexes (approximately 80-90%) remained bound to p27Kip1 and inactive during the first 8 hr of Myc activation. However, Myc rapidly alters the ability of free p27Kip1 to bind to newly synthesised cyclin E/Cdk2 complexes which consequently are phosphorylated by cyclin activating kinase (CAK) and become active. At the same time the levels of p27Kip1 remain. This suggests that Myc causes activation of cyclin E/Cdk2 via a highly dynamic process involving at least two steps: increased synthesis of cyclin E and decreased avidity of p27Kip1.
------------------------------------------------------------------------
Rescue of Raf-induced cell cycle arrest by Myc:

To explore the molecular basis of co-operation between the two oncoproteins Raf and Myc we have designed a system in which these proteins can be regulated independently by fusion to distinct steroid hormone receptors. Activation of an androgen receptor-Raf fusion (RafAR) in NIH3T3 cells is followed by a rapid and prolonged activation of the MAP kinase pathway and after 8-10 hours the cells develop a characteristic transformed morphology. Moreover, as mentioned above, Raf activation leads to a p21Cip1-dependent G1 arrest in these cells. Since Myc is able to suppress p27Kip1 activity we wondered whether Myc could release the Raf-induced p21Cip1-dependent cell cycle block. Indeed, activation of MycER in RafAR arrested cells leads to reentry of the cells into the cell cycle which is preceeded by a reactivation of cyclin E/cdk2 kinase activity. This involves inhibition of p21Cip1 through similar mechanism as previously decribed for p27Kip1. Myc prevents binding of p21Cip1 to cyclin E/Cdk2 without affecting the levels of p21Cip1 expression. Thus Myc co-operates with Raf by interfering with p21Cip1 function.
------------------------------------------------------------------------
Myc controls cell cycle inhibitor activity via modulation of cyclin D protein synthesis rates:

As we found no evidence for modification of cyclin E/Cdk2 complexes or p21Cip1 by Myc, the possibility that Myc might induce the expression of a protein which can sequester the cell cycle inhibitors p21Cip1 and p27Kip1 appeared to us as the most simple hypothesis. Indeed, in pulse-labelling experiments we could show a Myc-dependent increase in the binding of two proteins to p21Cip1 and p27Kip1 which could be observed in several cell types. These two proteins were identified as the cyclins D1 and D2, and we confirmed that the synthesis of both proteins is rapidly induced after activation of Myc. As a consequence p21Cip1 and p27Kip1 bind preferentially to cyclinD/Cdk4,6 complexes. This suggested that cyclinD/Cdk4,6 complexes can function as the predicted sequestration proteins buffering p21Cip1 and p27Kip1 activities. Consistent with this conclusion primary cells from cyclin D1-/- and cyclin D2-/- mouse embryos, unlike their wild type controls, fail to respond to Myc with increased proliferation, although they undergo accelerated cell death in the absence of serum. Myc sensitivity of cyclin D1-/- cells can be restored by exogenous expression of cyclins D1 and D2, as well as a cyclin D1 mutant forming kinase-defective, Cki-binding cyclin/cdk complexes. The sequestration function of D cyclins thus appears essential for Myc-induced cell cycle progression but dispensable for apoptosis. Our results also predict that induction of cyclin E/Cdk2 activity is coupled to cyclin D synthesis rates. We tested this possibility using the cells in which we can manipulat cyclin E and cyclin D synthesis rates. As mentioned above, cyclin E induction does not lead to activation of cyclin E/Cdk2 activity and/or cell cycle progression because p27Kip1 is in sufficient excess when these cells form a monolayer in the absence of serum. Increasing the synthesis rate of cyclin D in these cells, however, is sufficient for the cells to activate cyclin E/Cdk2 kinase activity. Importantly, this can also be achieved with the cyclin D mutant unable to form kinase-active Cdk complexes. Thus genetic and biochemical data demonstrate that the sequestration function of cyclin D/Cdk complexes is essential for Myc-dependent G1/S progression.

Fig. 3 Myc induced cyclin E/Cdk2 activation involves two steps, the inhibition of Cki (p27Kip1 and p21Cip1) binding to cyclin E/Cdk2 and the induction of cyclin E synthesis rates. Sequestration of the Ckis is mediated via an induction of cyclin D1 and cyclin D2 protein synthesis rates. This causes the preferential association of Ckis with cyclinD/Cdk complexes. Concomittantly increased cyclin E protein synthesis feeds the pool of newly formed Cki-free cyclin E/Cdk2 complexes which become CAK-phosphorylated and kinase-active.
------------------------------------------------------------------------
Oncogene co-operation in cell regulation:

Co-operating oncogenic lesions are expected to increase the proliferative advantage of tumour cells. However, it is remarkable that as part of this process they enhance each others oncogenic potentials. This may be necessary because the Ras/Raf pathway is involved in many cellular responses to peripheral signals and specifies multiple and in part mutually exclusive cellular programmes such as cell cycle progression and cell cycle arrest. The decision, whether to progress through the cell cycle or to arrest can be determined either by signal strength or multiple signals converging at the regulation of cyclin/cdk complexes. Although dn-p53 and Myc co-operate with Raf through distinct mechanisms, both dn-p53+Raf and Myc+Raf signalling converges on the regulation of cyclin/cdk activities via modulation of the sythesis rates of cyclins and cell cycle inhibitors. This supports our view that cyclin/cdk complexes function as important signal integrators in growth control. To understand multi-step carcinogenesis at the molecular level. further investigation of these signal integration processess is essential. At the same time this approach will also serve to explore how cellular signalling networks operate to determine cellular decisions.
 

Deleu L, Shellard S, Alevizopoulos K, Amati B, Land H (2001) Recruitment of TRRAP required for oncogenic transformation by E1A. Oncogene, 20:8270-5

Perez-Roger I, Kim SH, Griffiths B, Sewing A, Land H (1999) Cyclins D1 and D2 mediate myc-induced proliferation via sequestration of p27(Kip1) and p21(Cip1). Embo J, 18:5310-20

Sewing A, Wiseman B, Lloyd AC, Land H (1997) High-intensity Raf signal causes cell cycle arrest mediated by p21Cip1. Mol Cell Biol, 17:5588-97

Perez-Roger I, Solomon DL, Sewing A, Land H (1997) Myc activation of cyclin E/Cdk2 kinase involves induction of cyclin E gene transcription and inhibition of p27(Kip1) binding to newly formed complexes. Oncogene, 14:2373-81

Lloyd AC, Obermuller F, Staddon S, Barth CF, McMahon M, Land H (1997) Cooperating oncogenes converge to regulate cyclin/cdk complexes. Genes Dev, 11:663-77