Use of mitogenic cascade blockers for treatment of C-Raf induced lung adenoma in vivo: CI-1040 strongly reduces growth and improves lung structure

Oncogene-based therapeutics is a novel approach to inhibit proteins, which are essential for the initiation and maintenance of malignancies [1]. Agents such as Herceptin for the treatment of advanced breast cancer and Gleevec for chronic myelogenous leukemia have proven that cancer therapies targeting specific molecular alterations in signaling pathways are successful [2]. The Ras-MAP kinase pathway has a central role in regulating tumor cell growth and survival, differentiation and angiogenesis and has been targeted for therapeutic intervention in the past [3, 4]. C-Raf kinase and MEK are downstream effectors of the Ras signaling cascade. Both kinases are essential for cellular homeostasis and induce both proliferation and survival by suppression of apoptosis [5]. Raf and Ras mutations found in human malignancies convey constitutive activity to these signaling molecules thereby converting them into an oncogenic state [6].

In this study the Raf inhibitor BAY 43-9006 [7] and the MEK inhibitor CI-1040 [8] were tested as potential drugs in a transgenic mouse lung cancer model [9]. We have previously established this mouse model by expressing mutated, constitutively active C-Raf kinase (C-Raf BxB) under the control of the human surfactant protein C (SP-C) promoter [10]. C-Raf BxB lacks the regulatory NH₂-terminal sequences including the Ras interaction domain. Lung targeted expression of constitutively active C-Raf induced lung adenomas within 4 months of life [10]. Although these adenomas are stable for more than one year, deficiencies in other genes such as Bcl-2 or p53 were found to modulate adenoma growth or even switch the phenotype of tumor cells, respectively [11, 12]. The effects of BAY 43-9006 and CI-1040 were assessed in this lung adenoma model by daily intraperitoneal injections of these drugs at concentrations of 100 mg per kg body weight over a period of three weeks to four months old C-Raf-BxB mice.

We were able to show inhibition of ERK phosphorylation in vitro by both, the MEK inhibitor CI-1040 and the Raf inhibitor BAY 43-9600. In addition, both drugs inhibited ERK phosphorylation to a comparable level when tested on cell lines in tissue culture (data not shown). Although both inhibitors reached comparable serum concentrations after intraperitoneal administration (data not shown), the MEK inhibitor was active in mice whereas the Raf inhibitor was not. Why BAY 43-9006 is not active in our Raf dependent lung tumor mouse model and whether Raf and MEK might fulfill different functional roles with respect to lung adenoma formation needs to be further addressed in detail in future.

One possible explanation for the observed effects may relate to differences in the in vivo accessibility of these inhibitors to sub-cellular sites where C-Raf is localized. Alternatively, inhibition of Raf kinases might be more difficult to achieve due to multiple feedback mechanisms [15], whereas the regulation of MEK is less complex and inhibition of MEK is easier achieved in the living organism. Along this line it was already discussed previously that inhibitors targeting the kinase activity of Raf might not be useful as anticancer drugs because inhibition of Raf is always counterbalanced by reactivation [16]. Finally, it was recently suggested that C-Raf has signaling properties that are independent of its protein kinase activity [17]. Therefore, it has also to be considered that BAY 43-9006-bound kinase-inactive C-Raf has still transforming activity in pneumocyte type II cells. But this possibility appears unlikely in the light of earlier results demonstrating the loss of transforming activity and concomitant gain of dominant-negative function of c-Raf genes with mutations that inactivate the protein kinase activity of C-Raf [18‐20]. Moreover, there is a direct correlation between reversal of a transformed phenotype by BAY43-9006 and Phospho-ERK suppression (Sirrenberg and Rapp, unpublished data).

Previous work on CI-1040 has already established that this drug has a cytostatic effect on subcutaneously implanted colon tumors of human and mouse origin harboring RAS mutations [21]. In addition, in a mouse model of metastatic melanoma that was dependent on a B-Raf mutation CI-1040 was preventing the formation of new pulmonary metastases and caused rapid regression of already established pulmonary metastases [22]. Both publications as well as our report indicate that CI-1040 has a broad spectrum of antitumor activity and will be useful for the treatment of cancers that are dependent on mutations or overexpression of members of the mitogenic cascade upstream of ERK.

We also determined the effects of both inhibitors on expression of Bmi-1. The number of Bmi-1 positive cells in lung sections was reduced more than twofold by CI-1040, whereas BAY 43-9006 had no effect. The observed effects of CI-1040 may be limited to proliferation as neither apoptosis nor alveolar differentiation was directly affected. Bmi-1 was originally cloned as c-Myc cooperating oncogene in murine lymphomas [23, 24]. A role for Bmi-1 in human tumorigenesis was recently suggested by the discovery of Bmi-1 amplification in mantle cell lymphomas [25] and of Bmi-1 overexpression in a variety of tumors including non-small cell lung cancer [26]. Since CI-1040 might be more effective in inhibiting the proliferation of Bmi-1 positive adenoma cells, we want to address in future whether Bmi-1 is a self-renewal factor of lung stem cells or lung adenoma stem cells.

These experiments demonstrate an essential role of MEK/ERK signaling for the maintenance of Raf induced transformation in vivo. Effects on the ERK pathway in vivo may be predictive for effective tumor drug design and therapy of Raf driven tumors. CI-1040 is currently evaluated in phase II trials and may be a promising anti-proliferative drug for Ras and/or Raf-dependent human malignancies.