Unexpected oncosuppressive role for STAT3 in KRAS-induced lung tumorigenesis

ABSTRACT

Signal transducer and activator of transcription 3 (STAT3) plays a critical role in the pathogenesis of several diseases and is considered a therapeutic target in solid cancers, including lung cancer. However, we recently demonstrated a tumor suppressive function of STAT3 in kirsten rat sarcoma oncogene homolog (KRAS)-driven lung cancer. Here, we discuss these findings and their consequences.

Lung cancer is the prime cancer-related cause of death in developed countries.¹ Predominant driver mutations in lung adenocarcinoma (AC) are activating mutations in the kinase domain of the epidermal growth factor receptor (EGFR) or in the kirsten rat sarcoma oncogene homolog (KRAS). These driver mutations appear to be mutually exclusive, and KRAS mutations primarily arise in patients with a history of smoking. Therapeutic strategies to target oncogenic KRAS have been unsuccessful to date; therefore, targeting downstream effectors or cooperative pathways of KRAS has emerged as an alternative route for blocking KRAS-driven oncogenic pathways.² In this context, signal transducer and activator of transcription 3 (STAT3)-mediated signaling is believed to contribute to KRAS-induced tumorigenesis. Indeed, the pro-oncogenic characteristics of STAT3 during tumor development were acknowledged in various in vitro and in vivo studies, supporting efforts to develop STAT3 inhibitors.³ However, other models of solid cancers have yielded controversial findings suggesting an opposite role of STAT3 in repression of tumor growth.⁴ These results indicated that STAT3 has distinct features depending on tumor tissue type or driver mutations. Importantly, lung cancer-related studies established the pro-tumorigenic functions of STAT3, which was found to be frequently activated in lung tumor patients.⁵ In our recent study, we discovered that STAT3 is a haploid tumor suppressor in KRAS-mutant lung cancer, as evidenced in an inducible Kras^G12D-driven mouse model and xenografted human lung AC cells.⁶ Intrigued by these surprising findings we performed a thorough analysis of Stat3-deficient tumor tissue. In addition to pronounced tumor angiogenesis, we observed massive infiltration of myeloid cells in Stat3-deficient tumor tissue indicating qualitative and/or quantitative changes in chemokine expression. We discovered elevated levels of the potent proangiogenic chemokine chemokine (C-X-C motif) ligand 1 (Cxcl1) as a cause of aggravated tumor growth. Indeed, we validated Cxcl1 as a crucial chemokine that enhances tumor growth in Stat3-deficient tumors by showing that treatment with an antagonist against chemokine (C-X-C motif) receptor 2 (Cxcr2) receptor pathway blocked tumor growth in Stat3-deficient tumors. Cxcl1 and its human ortholog interleukin-8 (IL-8) are important nuclear factor kappa B (NFκB)-induced target genes and in vitro experiments revealed that STAT3 negatively regulates NFκB-induced expression of CXCL1/IL-8 in human and mouse lung epithelial cells. Microscopic examination of the interplay between NFκB and STAT3 offered hints regarding the mechanism of STAT3-dependent CXCL1 downregulation: STAT3 retains the NFκB p65 subunit in the cytoplasm, thereby inhibiting CXCL1/IL-8 gene transcription. Hence, we concluded that the tumor suppressive function of STAT3 in lung cancer cells is based on interference with NFκB activation (Fig. 1a). Vice versa, as depicted in Figure 1b, STAT3 deletion enhanced NFκB-mediated CXCL1 secretion, resulting in aberrant myeloid cell infiltration and tumor angiogenesis and further fueling tumor growth.

Figure 1.

Functions of STAT3 in KRAS-mutant cancer cells. (A) Upon activation of signal transducer and activator of transcription 3 (STAT3), nuclear factor kappa B (NFκB) is retained in the cytoplasm, therefore STAT3 inhibits the expression of chemokine (C-X-C motif) ligand 1 (CXCL1)/interleukin-8 (IL-8). (B) In the absence of STAT3, enhanced NFκB activity induces increased expression of chemokine CXCL1/IL-8 in KRAS-mutant tumors. CXCL1 attracts tumor supportive myeloid cells and promotes angiogenesis thereby enhancing tumor growth and progression.

Clinical evidence supporting our experimental findings was obtained from datasets of 4 different patient cohorts. We confirmed that STAT3 activation and expression levels were reduced in KRAS-mutant patient samples compared to EGFR-mutant and KRAS-wild type tumors. Moreover, patients with high-grade tumors showed a significant reduction in STAT3 expression compared to those with lower grade tumors, suggesting that STAT3 expression is eventually lost during progression. Patients with a smoking history (a prime cause of KRAS mutations) and low STAT3 signature had a worse survival outcome than patients with higher STAT3 expression levels. Interestingly, this observation was not made in lung AC patients without smoking history (who more frequently exhibit EGFR mutations) indicating that patients harboring driver mutations other than KRAS, such as EGFR, may benefit from STAT3 inhibition.

We recognize the controversy created by our studies regarding the role of STAT3 in lung tumorigenesis. However, we suspect that different driver mutations in lung AC may account for the reported oncogenic versus tumor suppressive functions of STAT3. Indeed, the clinical data presented in our manuscript strongly support this assumption. Given the tremendous possibilities generated by novel tools for genome editing we expect several upcoming studies addressing these issues.

Taking these findings together, we propose that stratification of patients according to their tumor driver mutations would be essential in order to predict whether patients would benefit from STAT3 inhibition. We claim that STAT3 targeted therapy not only provides no benefit for lung AC patients with mutated KRAS, but moreover puts these patients at a high risk for aggravation of cancer progression.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.