Increasing evidences including results from our laboratory demonstrated that aberrant expression and activation of G6PD promoted cell proliferation and adaptation in a serial of cancers and might serve as a potential anticancer target [16–18]. In these studies, the oncogenic functions and underlying molecular mechanisms of G6PD in cancer progression have partially been clarified. However, the question why G6PD is over-activated in different types of cancers is largely unknown.
G6PD activation could be regulated at both transcriptional and post-translational levels. It has been reported that Nrf2 could contribute to elevated G6PD expression in hepatocellular carcinoma through the transcriptional regulatory effect [41]. Ma et al. reported that Plk1 interacted with G6PD, promoted the formation of G6PD active homodimer, and therefore promoted cancer cell growth [42]. Meanwhile, Jiang et al. suggested that p53 formed a complex with G6PD, but inhibited its activities by restraining G6PD homodimer formation [43]. On the contrary, TAp73, a structural homolog of p53, could activate G6PD transcription, and facilitate oncogenic cell proliferation [44]. Although G6PD could be promoted by the positive feedback regulatory action of pSTAT3 through transcriptional regulation [18], the reason for ectopic G6PD gene overexpression in ccRCC still needs to be unraveled.
Previous studies from our laboratory demonstrated that G6PD overexpression lead to the increment of G6PD-NADPH-NOX4-dependent ROS accumulation and then pSTAT3 and MAKP signaling over-activation in ccRCC [18, 19]. However, several problems, such as alterations in ROS-triggered signaling pathways or other relevant regulators responsible for G6PD high expression, remained to be clarified. This study unraveled the interactive signaling pathways involved in G6PD overexpression, including ROS, NF-κB, and pSTAT3. Each of these pathways has been well studied in the tumorigenesis and development of a series of human cancers [21, 28, 32, 45]. ROS production and clearance are closely related to intracellular redox homeostasis and can be regulated by many factors. Compared with normal cells, tumor cells, especially derived from kidney cancers, are in a strong state of oxidative stress [46]. It is noteworthy that G6PD plays an important role in maintaining ROS homeostasis by balancing NADP/NADPH, GSH/GSSG, and other redox systems [16, 17]. Preliminary study showed that not only G6PD regulated ROS production and pSTAT3 activation, but ROS accumulation also impacted G6PD expression [18]. Therefore, it was hypothesized that ROS-involved signaling pathways might participate in G6PD dysregulation, perform cross-talk, and form a feedback loop to promote ccRCC tumorigenesis. The present study demonstrated that significantly reduced or increased activities of the NF-κB signaling pathway was found in ccRCC cells following treatment with ROS scavenger or stimulator, respectively (Fig. 1), which was consistent with the changes in pSTAT3 signaling activity and G6PD expression. The aforementioned evidences suggested that pSTAT3 and NF-κB signaling were transcriptional regulators and might play synergistic effects on G6PD overexpression in ccRCC. After comprehensive investigation, it was concluded that ROS induced NF-κB and pSTAT3 signaling over-activation. These two signaling pathways not only activated each other, but also formed a p65/pSTAT3 transcriptional complex and exhibited a reciprocal regulatory effect on promoting G6PD transcription (Fig. 7).
As the first and rate-limiting enzyme of PPP, G6PD not only mediates glucose catabolism and maintains cell redox homeostasis, but also generates sufficient precursors and plays crucial roles in the biosynthesis of lipids and nucleic acids to meet the requirement of cancer cell for rapid proliferation and progression [17, 43]. The present study suggested novel proliferation strategies of ccRCC cancer cells via dual oncogenic transcriptional factors, NF-κB and STAT3, which are over-activated in ccRCC and cooperatively facilitate ccRCC proliferation through inducing G6PD overexpression and then cell cycle regulators such as CyclinD1 and CDK4. Additionally, previous results described that G6PD high expression was positively correlated to lymph node metastasis, Fuhrman grade, and TNM stage of ccRCC, indicating that G6PD might be involved in promoting ccRCC metastasis. Although the mechanism of G6PD in ccRCC was not fully revealed, it was faithful that in addition to the proliferation-promoting effect, the work model unraveled in present study—NF-κB and pSTAT3 synergistically drove G6PD overexpression (Fig. 7)—might display more functions, especially the potential to mediate tumor metabolic reprogramming and facilitate metastasis of ccRCC patients. Therefore, more investigations are necessary to be carried out for clarifying these hypotheses.
In addition, when the endogenous expression of p65 was knocked down by RNAi lentivirus, not only p65 and G6PD expression levels were declined, a significant reduction of p105 and p50 was also found in ccRCC cell lines (Fig. 2F-G). These results indicated an interaction between p65 and p50—two functional activators of NF-κB signaling pathway. Additionally, p50 also responded to ROS stimulation (Fig. 1C-D) and could form a complex with pSTAT3 (data not shown). However, p50 recruited on neither the potential NF-κB-binding site (Fig. 2E) nor the pSTAT3-binding site (Fig. 3B) on the G6PD promoter, suggesting that p50 may be potential to interact with pSTAT3 and regulate G6PD expression, but the complex might occupy other undiscovered binding sites. Similar experiments to verify the impact of p50 on G6PD expression and the interaction between p65, p50, and pSTAT3 in ccRCC still need to be performed in future studies.
Moreover, G6PD has been pinpointed as a new biomarker in acute myeloid leukemia (AML), and its overexpression positively correlated with poor prognosis of AML patients [47]. Meanwhile, targeting G6PD induces apoptosis and enhances chemotherapeutic antitumor effects via ROS-mediated damage in certain cancer, including AML, lung cancer, breast cancer, and colorectal cancer [47, 48]. Previous study also indicated that G6PD was a potential prognostic biomarker and a promising therapeutic target for ccRCC treatment [18]. However, whether G6PD inhibition exert any antitumor effects in ccRCC is far from being clarified. The present study investigated whether the nicotinamide analog 6-AN could affect the proliferation of ccRCC both in vitro and in vivo. 6-AN, a known competitive inhibitor of G6PD, could modulate the cytotoxicity of antineoplastic treatments, and is undergoing preclinical investigation in certain cancers [40, 49]. As shown in Fig. 6A-B, a significantly reduced proliferation rate of 786-O cells was found in vitro. However, the inhibition of tumor growth was not as dramatic as that mediated by G6PD silencing in our previous study [18], which most probably due to low drug bioavailability. Nevertheless, the inhibition of tumor growth was associated with decreased levels of G6PD, pSTAT3, and p65 in isolated tumor samples (Fig. 6F-G), demonstrating that abnormally activated pSTAT3 and NF-κB signaling pathways were positively correlated to G6PD overexpression in vivo and G6PD inhibition exhibited tumor-suppressive activities in ccRCC. Although developing clinical application are still challenging, the present studies indicated that G6PD-based gene therapy might provide an adjunctive approach to ccRCC treatment.