Background: Hypoxia-inducible factor-1α (HIF-1α) induces the expression of glycolysis-related genes, which plays a direct and key role in Warburg effect. In a recent study, honokiol (HNK) was identified as one of the potential agents that inhibited the HIF-1α signaling pathway. Because the HIF- 1α pathway is closely associated with glycolysis, we investigated whether HNK inhibited HIF-1α-mediated glycolysis.Methods: The effects of HNK on HIF-1α-mediated glycolysis and other glycolysis-related genes’ expressions, cancer cells apoptosis and tumor growth were studied in various human breast cancer models in vitro and in vivo. We performed the following tests: extracellular acidification and oxygen consumption rate assays, glucose uptake, lactate, and ATP assays for testing glycolysis; WST-1 assay for investigating cell viability; colony formation assay for determining clonogenicity; flow cytometry for assessing cell apoptosis; qPCR and Western blot for determining the expression of HIF-1α, GLUT1, HK2 and PDK1. The mechanisms of which HNK functions as a direct inhibitor of HIF-1α were verified through the ubiquitination assay, the Co-IP assay, and the cycloheximide (CHX) pulse-chase assay.Results: HNK increased the oxygen consumption rate while decreased the extracellular acidification rate in breast cancer cells; it further reduced glucose uptake, lactic acid production and ATP production in cancer cells. The inhibitory effect of HNK on glycolysis is HIF-1α-dependent. HNK also downregulated the expression of HIF-1α and its downstream regulators, including GLUT1, HK2 and PDK1. A mechanistic study demonstrated that HNK enhanced the self-ubiquitination of HIF-1α by recruiting two E3 ubiquitin ligases (UFL1 and BRE1B). In vitro, HNK inhibited cell proliferation and clonogenicity, as well as induced apoptosis of cancer cells. These effects were also HIF1α-dependent. In vivo, HNK inhibited tumor growth and HIF-1α-mediated glycolysis.Conclusion: HNK has an inhibitory effect on HIF-1α-mediated glycolysis in human breast cancer. Our research revealed a new mechanism of HNK as an anti-cancer drug, thus representing a novel strategy to improve the prognosis of cancer.

Background S100A6 and murine double minute 2 (MDM2) are important cancer-related molecules. A previous study identified an interaction between S100A6 and MDM2 by size exclusion chromatography and surface plasmon resonance experiments. The present study investigated whether S100A6 could bind to MDM2 in vivo and further explored its functional implication. Methods Co-immunoprecipitation, glutathione-S-transferase pull-down assay, and immunofluorescence were performed to determine the in vivo interaction between S100A6 and MDM2. Cycloheximide pulse-chase assay and ubiquitination assay were performed to clarify the mechanism by which S100A6 downregulated MDM2. In addition, clonogenic assay, WST-1 assay, and flow cytometry of apoptosis and the cell cycle were performed and a xenograft model was established to evaluate the effects of the S100A6/MDM2 interaction on growth and paclitaxel-induced chemosensitivity of breast cancer. The expressions of S100A6 and MDM2 in patients with invasive breast cancer were analyzed by immunohistochemistry. In addition, the correlation between the expression of S100A6 and the response to neoadjuvant chemotherapy was statistically analyzed. Results S100A6 promoted the MDM2 translocation from nucleus to cytoplasm, in which the S100A6 bound to the binding site of the herpesvirus-associated ubiquitin-specific protease (HAUSP) in MDM2, disrupted the MDM2–HAUSP–DAXX interactions, and induced the MDM2 self-ubiquitination and degradation. Furthermore, the S100A6-mediated MDM2 degradation suppressed the growth of breast cancer and enhanced its sensitivity to paclitaxel both in vitro and in vivo. For patients with invasive breast cancer who received epirubicin and cyclophosphamide followed by docetaxel (EC-T), expressions of S100A6 and MDM2 were negatively correlated, and high expression of S100A6 suggested a higher rate of pathologic complete response (pCR). Univariate and multivariate analyses showed that the high expression of S100A6 was an independent predictor of pCR. Conclusion These results reveal a novel function for S100A6 in downregulating MDM2, which directly enhances sensitivity to chemotherapy.