The roles and significance of STAT3 in cancer biology have been extensively studied for more than a decade. Mounting evidence has shown that constitutive activation of STAT3 is a frequent biochemical aberrancy in cancer cells, and this abnormality directly contributes to tumorigenesis and shapes many malignant phenotypes in cancer cells. Nevertheless, results from more recent experimental and clinicopathologic studies have suggested that STAT3 also can exert tumor suppressor effects under specific conditions. Importantly, some of these studies have demonstrated that STAT3 can function either as an oncoprotein or a tumor suppressor in the same cell type, depending on the specific genetic background or presence/absence of specific coexisting biochemical defects. Thus, in the context of cancer biology, STAT3 can be a friend or foe. In the first half of this review, we will highlight the “evil” features of STAT3 by summarizing its oncogenic functions and mechanisms. The differences between the canonical and non-canonical pathway will be highlighted. In the second half, we will summarize the evidence supporting that STAT3 can function as a tumor suppressor. To explain how STAT3 may mediate its tumor suppressor effects, we will discuss several possible mechanisms, one of which is linked to the role of STAT3β, one of the two STAT3 splicing isoforms. Taken together, it is clear that the roles of STAT3 in cancer are multi-faceted and far more complicated than one appreciated previously. The new knowledge has provided us with new approaches and strategies when we evaluate STAT3 as a prognostic biomarker or therapeutic target.

We have previously described the existence of two phenotypically distinct cell subsets in ALK-positive anaplastic large cell lymphoma (ALK + ALCL) based on their differential responsiveness to a Sox2 reporter (SRR2), with reporter-responsive (RR) cells being more tumorigenic and chemoresistant than reporter-unresponsive (RU) cells. However, the regulator(s) of RU/RR dichotomy are not identified. In this study, we aim to delineate the key regulator(s) of RU/RR dichotomy. JASPER motif match analysis was used to identify the putative factors binding to SRR2 sequence. SRR2 probe pull-down assay and quantitate real-time PCR were performed to analyze the regulation of Sox2 transcriptional activity by MYC. Methylcellulose colony formation assay, chemoresistance to doxorubicin and mouse xenograft study were performed to investigate the biological functions of MYC. PCR array and western blotting were executed to study related signaling pathways that regulate MYC expression. Immunofluorescence and immunohistochemistry assay were initiated to evaluate the expression of MYC and its correlation with its regulator by chi-square test analysis in human primary tumor cells. We identified MYC as a potential regulator of RU/RR dichotomy. In support of its role, MYC was highly expressed in RR cells compared to RU cells, and inhibition of MYC substantially decreased the Sox2/SRR2 binding, Sox2 transcriptional activity, chemoresistance, and methylcellulose colony formation. In contrast, enforced expression of MYC in RU cells conferred the RR phenotype. The Wnt/β-catenin pathway, a positive regulator of MYC, was highly active in RR but not RU cells. While inhibition of this pathway in RR cells substantially decreased MYC expression and SRR2 reporter activity, experimental activation of this pathway led to the opposite effects in RU cells. Collectively, our results support a model in which a positive feedback loop involving Wnt/β-catenin/MYC and Sox2 contributes to the RR phenotype. In a mouse xenograft model, RU cells stably transfected with MYC showed upregulation of the Wnt/β-catenin/MYC/Sox2 axis and increased tumorigenecity. Correlating with these findings, there was a significant correlation between the expression of active β-catenin and MYC in ALK + ALCL primary tumor cells. A positive feedback loop involving the Wnt/β-catenin/MYC/Sox2 axis defines a highly tumorigenic cell subset in ALK + ALCL.

The phenomenon that malignant cells can acquire stemness under specific stimuli, encompassed under the concept of cancer cell plasticity, has been well-described in epithelial malignancies. To our knowledge, cancer cell plasticity has not yet been described in hematopoietic cancers. To illustrate and study cancer cell plasticity in hematopoietic cancers, we employed an in-vitro experimental model of ALK-positive anaplastic large-cell lymphoma (ALK+ALCL) that is based on the phenotypic and functional dichotomy of these cells, with cells responsive to a Sox2 reporter (i.e. RR cells) being significantly more stem-like than those unresponsive to the reporter (i.e. RU cells). H2O2 was employed to trigger oxidative stress. GFP expression and luciferase activity, readouts of the Sox2 reporter activity, were quantified by using flow cytometry and luciferase activity assay, respectively. Doxorubicin-resistance and clonogenicity were assessed by using the MTS, methylcellulose colony formation and limiting dilution assays. Western blotting and quantitative PCR were used to assess the expression of various members of the Wnt/β-catenin pathway. Pull-down studies using a Sox2 binding consensus sequence were used to assess Sox2-DNA binding. Quercetin and 10074-G5 were used to inhibit β-catenin and MYC, respectively. siRNA was used to downregulate Sox2. Under H2O2-induced oxidative stress, a substantial fraction of RU cells was found to convert to RR cells, as evidenced by their acquisition of GFP expression and luciferase activity. Compared to the native RU cells, converted RR cells had significantly higher levels of doxorubicin-resistance, clonogenicity and sphere formation. Converted RR cells were characterized by an upregulation of the Wnt/β-catenin/MYC/Sox2 signaling axis, previously found to be the key regulator of the RU/RR dichotomy in ALK+ALCL. Furthermore, Sox2 was found to bind to DNA efficiently in converted RR cells but not RU cells, and this finding correlated with significant elevations of several Sox2 downstream targets such as WNT2B and BCL9. Lastly, inhibition of β-catenin, MYC or Sox2 in RU cells significantly abrogated the H2O2-induced RU/RR conversion. We have demonstrated that cancer cell plasticity exists in ALK+ALCL, a type of hematopoietic cancer. In this cancer type, the Wnt/β-catenin/MYC/Sox2 axis is an important regulator of cancer cell plasticity.