Genetic Dissection of IL-6 signalin in prostate cancerogenesis

Prostate cancer (PC) is the second most common cause of death in men with approximately 70,000 deaths in Europe annually. Fatalities are mainly due to a highly aggressive form of PC, which has escaped more detailed molecular characterization for improved diagnosis or prognosis. The PI3K pathway is activated in a majority of PC and reasoned to be the major oncogenic driver of these tumors. In order to simulate human PC in mice, PTEN (phosphatase and tensin homolog deleted on chromosome 10) was ablated in prostate epithelial cells. Such mice develop PC and a detailed analysis confirms they can serve as model for the human disease. We will therefore exploit this model to investigate the functional contribution of IL-6 to PC progression. The cytokine Interleukin-6 (IL-6) is a signaling molecule with a multitude of specific responses in different cell types. IL-6 activates the Jak/Stat pathway (Jak - Janus kinase, Stat signal transducer and activator of transcription), and has been shown to be strongly deregulated in a majority of PC. Stat3 has been shown to contribute to many tumor types as an oncogene, but Stats have also been described as tumor suppressors in specific circumstances. IL-6 negatively regulates proliferation in prostate cells, but it has been also described to promote tumor progression in PC via autocrine stimulation and the IL-6R is upregulated in a subset of malignant PC. Based on these findings IL-6 has been selected as a therapeutic target, but accumulating clinical evidence indicates it fails to provide the expected benefit. In PC the situation is complicated by the fact that IL-6 might also affect vascular, inflammatory and stromal cells in PC. An initial analysis of patient samples shows that loss of Stat3 expression significantly correlates with bad prognosis and Stat3 therefore can act as a tumor suppressor at least in a potential subset of PC. We have therefore chosen to analyze PC to characterize this novel subtype of malignant PC, as this might provide crucial information for individualized therapy of patients. The complex interactions between PC and host tissue are best analyzed in animal disease models, which reflect their full complexity. To explore IL-6 signaling in PC we have therefore chosen to delete IL-6 or Stat3 in the background of the PTEN model. Further analysis of the transgenic mouse models will be undertaken to identify the biological features supporting initiation, invasion and progression of PC after loss of IL-6 or Stat3. We propose to compare human PC patient samples and mouse tissue specimens using tissue micro arrays and gene expression analysis to assess whether our results are relevant to the situation in humans and specific subsets of PC can be determined on the basis of the IL-6/Stat3 status.

Prostate cancer (PCa) is one of the most prevalent tumors in men. Hyperactive IL-6/Stat3 signaling is an important driving force in PCa. Targeting the IL-6/Stat3 axis in PCa patients failed to provide therapeutic benefit. In this project we showed that genetic inactivation of IL-6 or Stat3 overcame cellular senescence and promoted cancer progression and metastasis in Pten- deficient PCa mouse models. We furthermore showed a new mechanism by which IL-6/Stat3 signaling linked cellular senescence with dissemination in PCa. We demonstrated that p19ARF is a novel direct target of Stat3. Stat3 loss in PCa downregulates the cyclin dependent kinase inhibitor p19ARF, which results in an increase of MDM2 and a decrease of p53 and PML expression, which suppresses senescence. In human PCa, loss of Stat3 and/or p14ARF expression correlates with an increased risk of disease recurrence. Therefore, Stat3/p14ARF are novel markers to stratify high- from low-risk PCa patients. This could be exploited diagnostically and lead to new therapies. Since loss of PTEN and Stat3 results in a metastatic phenotype, we furthermore investigated metastatic PCa relapsing from antiandrogen treatment and chemotherapies, which can exhibit recurrent genomic alterations, variant histology and cellular metabolism. We identified mutually exclusive STAT3 and PTEN deletions in liquid biopsies of PCa patients and from the TCGA database (Pencik et al., unpublished). In addition to the metastatic progression observed in a Pten/Stat3-deficient PCa mouse model we demonstrated activated mTORC1-dependent metabolic reprogramming which also caused mitochondrial dysfunctions. In contrast, constitutive active Stat3 led to direct prolongation of the lifespan, inhibited tumor formation and mTOR activity and stabilized mitochondrial respiration to restore normal cellular metabolism. We therefore hypothesized that IL-6/STAT3 signaling would be dependent on mTOR signaling (Kenner L., unpublished observation).