The PKM2 inhibitor shikonin enhances piceatannol-induced apoptosis of glyoxalase I-dependent cancer cells.

Glyoxalase I (GLO I), a major enzyme involved in the detoxification of the anaerobic glycolytic byproduct methylglyoxal, is highly expressed in various tumors, and is regarded as a promising target for cancer therapy. We recently reported that piceatannol potently inhibits human GLO I and induces the death of GLO I-dependent cancer cells. Pyruvate kinase M2 (PKM2) is also a potential therapeutic target for cancer treatment, so we evaluated the combined anticancer efficacy of piceatannol plus low-dose shikonin, a potent and specific plant-derived PKM2 inhibitor, in two GLO I-dependent cancer cell lines, HL-60 human myeloid leukemia cells and NCI-H522 human non-small-cell lung cancer cells. Combined treatment with piceatannol and low-dose shikonin for 48 h synergistically reduced cell viability, enhanced apoptosis rate, and increased extracellular methylglyoxal accumulation compared to single-agent treatment, but did not alter PKM1, PKM2, or GLO I protein expression. Taken together, these results indicate that concomitant use of low-dose shikonin potentiates piceatannol-induced apoptosis of GLO I-dependent cancer cells by augmenting methylglyoxal accumulation.

Triphenylbismuth dichloride inhibits human glyoxalase I and induces cytotoxicity in cultured cancer cell lines.

Organobismuth compounds, i.e., organic-inorganic hybrid molecules composed of an organic structure and bismuth metal, have been reported to induce cytotoxicity in cancer cells; however, the target proteins associated with this cytotoxicity have not been elucidated. Herein, we investigated the inhibitory effect of five organobismuth compounds on human glyoxalase I (hGLO I), a promising target candidate for cancer therapy. Among these compounds, triphenylbismuth dichloride (Bi-05) exerted a strong inhibitory effect on hGLO I. Indeed, Bi-05 inhibited hGLO I in a dose-dependent manner with an IC50 value of 0.18 µM. Bi-05 also induced cytotoxicity in human leukemia HL-60 cells and human lung cancer NCI-H522 cells, both of which exhibit high expression levels of GLO I. However, the hGLO I-inhibiting and cytotoxic effects of Bi-05 disappeared when the bismuth atom was replaced with an antimony or phosphorus atom. Bismuth(III) nitrate had little inhibitory effect on hGLO I activity and only slightly reduced the viability of cancer cells. In the culture medium of Bi-05-treated HL-60 cells, the concentration of the GLO I substrate methylglyoxal was markedly elevated. In addition, Bi-05 treatment more strongly inhibited human lung cancer NCI-H522 cell (exhibiting high GLO I expression) proliferation than human lung cancer NCI-H460 cell (exhibiting low GLO I expression) proliferation. Furthermore, the cytotoxicity of Bi-05 was significantly decreased by pre- and co-treatment with the methylglyoxal scavengers N-acetyl-L-cysteine and aminoguanidine. Overall, these results suggest that Bi-05 treatment leads to the accumulation of methylglyoxal via GLO I inhibition, resulting in cytotoxic effects in cancer cells.

Addition of hydrophobic side chains improve the apoptosis inducibility of the human glyoxalase I inhibitor, TLSC702.

Glyoxalase I (GLO I) is a known therapeutic target in cancer. Even though TLSC702, a GLO I inhibitor that we discovered, induces apoptosis in tumor cells, exceptionally higher doses are required compared with those needed to inhibit GLO I activity in vitro. In this work, structure-activity optimization studies were conducted on four sections of the TLSC702 molecule to determine the partial structural features necessary for the inhibition of GLO I. Herein, we found that the carboxy group in TLSC702 was critical for binding with the divalent zinc at the active site of GLO I. In contrast, the side chain substituents in the meta- and para- positions of the benzene ring had little influence on the in vitro inhibition of GLO I. The CLogP values of the TLSC702 derivatives showed a positive correlation with the antiproliferative effects on NCI-H522 cells. Thus, two derivatives of TLSC702, which displayed either high or low lipophilicity due to the types of substituents at the phenyl position, were selected. Even though both derivatives showed comparable inhibitory effects as that of their parent compound, the derivative with the high CLogP value was distinctly more antiproliferative than TLSC702. In contrast, the derivative with the low CLogP value did not decrease cell viability in NCI-H522 and HL-60 cells. These findings suggested that structural improvements, such as the addition of hydrophobic moieties to the phenyl group, enhanced the ability of TLSC702 to induce apoptosis by increasing cell membrane permeability.