Screening natural products for inhibitors of quinone reductase-2 using ultrafiltration LC-MS

Abstract

Inhibitors of quinone reductase-2 (NQO2; QR-2) can have antimalarial activity and antitumor activities or can function as chemoprevention agents by preventing the metabolic activation of toxic quinones such as menadione. To expedite the search for new natural product inhibitors of QR-2, we developed a screening assay based on ultrafiltration liquid chromatography-mass spectrometry that is compatible with complex samples such as bacterial or botanical extracts. Human QR-2 was prepared recombinantly, and the known QR-2 inhibitor, resveratrol, was used as a positive control and as a competitive ligand to eliminate false positives. Ultrafiltration LC-MS screening of extracts of marine sediment bacteria resulted in the discovery of tetrangulol methyl ether as an inhibitor of QR-2. When applied to the screening of hop extracts from the botanical, Humulus lupulus L., xanthohumol and xanthohumol D were identified as ligands of QR-2. Inhibition of QR-2 by these ligands was confirmed using a functional enzyme assay. Furthermore, binding of xanthohumol and xanthohumol D to the active site of QR-2 was confirmed using X-ray crystallography. Ultrafiltration LC-MS was shown to be a useful assay for the discovery of inhibitors of QR-2 in complex matrixes such as extracts of bacteria and botanicals.

Figure 1

Ultrafiltration LC-MS screening of resveratrol (2.5 μg/mL) for binding to QR-2 (80 μg/mL). LC-MS chromatograms of ultrafiltrates containing resveratrol released from active QR-2 (solid line) or denatured QR-2 (dashed line) as a control. Enhancement of the resveratrol peak (detected at m/z 227 as the deprotonated molecule during negative ion electrospray) compared to the control indicates specific binding to QR-2.

Figure 2

Ultrafiltration LC-MS detection of tetrangulol methyl ester as a ligand of QR-2 (80 μg/mL) in an extract of an Actinomyces sp. (2 μg/mL) from marine sediment. Computer-reconstructed negative ion electrospray mass chromatograms of the signals of m/z 317 using negative ion electrospray are shown for the ultrafiltrates obtained with active QR-2 (solid line) or denatured QR-2 as a control (dashed line).

Figure 3

Screening of an ethanolic hop extract (2 μg/mL) for ligands to QR-2 (80 μg/mL) by using ultrafiltration LC-MS. Compared to the control, which contained denatured enzyme (dashed line), LC-MS with negative ion electrospray LC-MS showed several peaks that were enhanced due to specific binding to QR-2 (solid line). A) The deprotonated molecule of xanthohumol at m/z 353 was detected at a retention time of 27.6 min; and B) isomeric compounds of m/z 369 containing one more oxygen than xanthohumol were detected including xanthohumol D eluting at 22.8 min. Identification of xanthohumol and xanthohumol D was based on comparison with authentic standards.

Figure 4

Ultrafiltration LC-MS analysis of a hop extract as shown in Figure 3 (solid line), except that 25 μM of resveratrol was included in a competition experiment (dashed line). Resveratrol, a high affinity ligand added at high concentration, displaced xanthohumol and xanthohumol D from QR-2 indicating that all three compounds compete for the same binding site.

Figure 5

Negative ion electrospray tandem mass spectra of the deprotonated molecules of A) xanthohumol at m/z 353; and B) xanthohumol D at m/z 369. Product ion mass spectra were obtained using collision-induced dissociation.

Figure 6

Crystal structure of QR-2 in complex with xanthohumol D. Xanthohumol D is colored in blue according to atom, and a hydrogen bond is shown as a grey dashed line. The 2Fo-Fc map is shown in grey mesh and contoured to 1.0σ, and the Fo-Fc map is shown in red mesh and contoured to 3.0σ.