Hexavalent chromium induces centrosome amplification through ROS-ATF6-PLK4 pathway in colon cancer cells.

Centrosome amplification (CA) refers to a numerical increase in centrosomes resulting in cells with more than two centrosomes. CA has been shown to initiate tumorigenesis and increase the invasive potential of cancer cells in genetically modified experimental models. Hexavalent chromium is a recognized carcinogen that causes CA and tumorigenesis as well as promotes cancer metastasis. Thus, CA appears to be a biological link between chromium and cancer. In the present study, we investigated how chromium triggers CA. Our results showed that a subtoxic concentration of chromium-induced CA in HCT116 colon cancer cells, resulted in the production of reactive oxygen species (ROS), activated ATF6 without causing endoplasmic reticulum stress, and upregulated the protein level of PLK4. Inhibition of ROS production, ATF6 activation, or PLK4 upregulation attenuated CA. Inhibition of ROS using N-acetyl-l-cysteine (NAC) inhibited chromium-induced activation of ATF6 and upregulation of PLK4. ATF6-specific siRNA knocked down the protein level and activation of ATF6, and upregulated PLK4, with no effect on ROS production. Knockdown of PLK4 protein had no effect on chromium-induced ROS production or activation of ATF6. In conclusion, our results suggest that hexavalent chromium induces CA via the ROS-ATF6-PLK4 pathway and provides molecular targets for inhibiting chromium-mediated CA, which may be useful for the assessment of CA in chromium-promoted tumorigenesis and cancer cell metastasis.

The binding between NPM and H2B proteins signals for the diabetes-associated centrosome amplification.

Diabetes not only increases the risk for cancer but also promotes cancer metastasis. Centrosome amplification (CA) is sufficient to initiate tumorigenesis and can enhance the invasion potential of cancer cells. We have reported that diabetes can induce CA, with diabetic pathophysiological factors as the triggers, which involves the signaling of nucleophosmin (NPM). Thus, CA can serve as a candidate biological link between diabetes and cancer. In the present study, we attempted to identify the NPM binding partners and investigated whether the binding between NPM and its partner mediated the CA. We confirmed that high glucose, insulin, and palmitic acid cancer could elicit CA in the HCT16 colon cancer cells and found that the experimental treatment increased the binding between NPM and H2B, but not between p-NPM and H2B. The molecular docking analysis supported the fact that NPM and H2B could bind to each other through various amino acid residues. The treatment also increased the colocalization of NPM and H2B in the cytosol. Importantly, disruption of the NPM1-H2B complex by individual knockdown of the protein level of NPM or H2B led to the inhibition of the treatment-evoked CA. In conclusion, our results suggest that the binding between NPM and H2B proteins signals for the CA by high glucose, insulin, and palmitic acid.