On the role of SAP kinase pathways in cellular responses to cancer treatment

Abstract: Inherited or acquired resistance to therapy is a major clinical problem in the treatment of cancer. Defective apoptosis signalling is increasingly acknowledged as an important mechanism of resistance. Better understanding of apoptosis regulation may therefore help improve therapy. This thesis investigates the involvement of stress-activated protein kinase (SAPK) pathways in cellular responses to the DNA-damaging agents cisplatin and ionizing radiation (IR). Focus is on the apical SAPK MEKK1, which in response to many apoptotic stimuli is cleaved, generating a kinase-active fragment, deltaMEKK1. Although deltaMEKK1 contributes to apoptosis, the underlying mechanisms are not known. Because Ras is commonly mutated in human tumours, the question how oncogenic Ras influences the sensitivity of cells to cisplatin is clinically important. This was studied in FR3T3 fibroblasts and sublines transformed by 12V-H-ras. In response to cisplatin, FR3T3 cells arrested in G2, while the ras-transformants accumulated in S-phase. dnMEKK1, a kinase-inactive mutant fragment of MEKK1, blocked apoptosis but not the S-phase accumulation. Interestingly, the ras-transformants showed higher levels of apoptosis at 24 h compared to parental FR3T3 cells, whereas in a 7-day colony formation assay they were more resistant than FR3T3. This could in part be explained by increased cisplatin sensitivity of FR3T3 when seeded at the low densities required for the colony assay. In addition, ras-transformed cells which had survived treatment proliferated faster than FR3T3. These results illustrate the need to carefully consider the choice of assay and relevant timing in drug sensitivity studies. dnMEKK1 blocked cisplatin-induced apoptosis in melanoma cell lines, but did not block activation of JNK. It also blocked cisplatin-induced activation of the Bcl-2 family member Bak which regulates cytochrome c release during apoptosis. Conversely, the corresponding kinase-active mutant, dpMEKK1, could on its own induce Bak activation within a similar time frame and to a similar extent as cisplatin. However, unlike cisplatin, dpMEKK1 was unable to induce the subsequent rapid formation of Bak complexes required for cytochrome c release and apoptosis. Whereas the JNK inhibitor SP600125 did not block the initial Bak activation, it did block complex formation, suggesting that the missing signal is provided by JNK. IR-induced apoptosis in the small cell lung carcinoma cell line U1285 was found to involve dnMEKK1-sensitive activation of Bak, and to require the activities of the SAP kinases JNK and p3 8. In the IR-resistant non small cell lung carcinoma cell line U 1810, however, IR did not activate JNK or p38. Neither were Bak/Bax activated, suggesting that their resistance is due to a block early in the apoptotic process. Nevertheless, cisplatin treatment of U 1810 cells did induce activation of JNK, Bak and apoptosis, indicating that cisplatin and IR signal via different pathways. Since dpMEKK1 could per se induce apoptosis in these cells, the resistance block must lie elsewhere. A model is presented, showing cisplatin-induced activation of at least two kinase pathways which separately contribute to Bak activation and complex formation. IR is hypothesized to activate at least one of these pathways. This thesis illustrates that apoptosis induction by cisplatin and IR involve complex molecular pathways, and that these pathways differ between the two agents. The finding that deltaMEKK1 and JNK regulate two distinct steps in the activation of Bak is consistent with the notion that the apoptotic process is carefully regulated. Increased insight into the mechanisms of apoptosis induction by conventional anticancer agents will contribute to improved treatment modalities and an increased understanding of the problem of resistance to therapy.