Pattern recognition analysis of endogenous cell metabolites for high throughput mode of action identification: removing the postscreening dilemma associated with whole-organism high throughput screening.

Although whole-organism HTS can give clear indications of in vivo activity, typically few clues are given as to the mechanism of action (MOA), and determining the MOA for large numbers of active compounds can be costly and complex-an alternative approach is required. This report demonstrates that it is possible to conduct relatively high throughput MOA characterization of HTS hits utilizing a single sample preparation and analytical method. By monitoring a wide range of endogenous cellular metabolites via (1)H nuclear magnetic resonance spectroscopy, the MOA of herbicides can be predicted using computational methods to compare the metabolite perturbation patterns. Herbicides that induce a characteristic pattern of metabolic perturbation in maize include inhibitors of acetolactate synthase, acetyl co-enzyme A carboxylase, protoporphyrinogen oxidase, 5-enolpyruvylshikimate-3-phosphate synthase, and phytoene desaturase. In soya, photosystem II inhibitors can also be detected, further demonstrating that this method is not limited to inhibitors of enzymes that directly act upon endogenous metabolites, or a single species. The methods, including data analysis, can be readily automated, enabling relatively high throughput MOA elucidation of whole-organism screen hits. Additionally, for compounds with a novel MOA, this approach may lead to MOA identification faster than traditional methods. It is envisaged that application of these data analysis methods to other data types-for example, transcription (mRNA) or translation (protein) profiles-is likely to permit higher throughput with smaller sample requirements, along with ability to discriminate MOAs that are not adequately discriminated based upon endogenous metabolite profiles.

Mass spectrometrically detected directly coupled high performance liquid chromatography/nuclear magnetic resonance spectroscopy/mass spectrometry for the identification of xenobiotic metabolites in maize plants.

Reconstructed ion chromatograms have been used to identify relevant high performance liquid chromatography (HPLC) peaks in a directly coupled high performance liquid chromatography/nuclear magnetic resonance spectroscopy/mass spectrometry (HPLC/NMR/MS) experiment. This has been applied to a study of the metabolism of a model compound, 5-nitropyridone (2-hydroxy-5-nitropyridine), in maize plants grown hydroponically. By monitoring the on-flow reconstructed ion chromatogram corresponding to the 5-nitropyridone fragment at m/z 143, and additional molecular ions corresponding to metabolites identified as products from similar compounds, relevant peaks were identified rapidly for subsequent stopped-flow 1H NMR spectroscopic analysis. The combination of coupled HPLC/NMR/MS enabled the direct identification of three metabolites, namely the N-glucoside, N-malonylglucoside, and O-malonylglucoside. This work demonstrates the power of HPLC/NMR/MS for the structural elucidation of xenobiotic metabolites in complex biological matrices (such as plant material) with minimal sample preparation. In particular, using mass spectrometry for the initial identification of relevant HPLC peaks allows the analysis of complex samples without the necessity for other spectroscopic markers, such as 19F NMR signal for fluorinated compounds or UV spectroscopy for molecules with strong UV chromophores.

Application of directly coupled HPLC-NMR-MS/MS to the identification of metabolites of 5-trifluoromethylpyridone (2-hydroxy-5-trifluoromethylpyridine) in hydroponically grown plants.

Directly coupled HPLC-NMR-MS was used to characterize two major metabolites of 5-trifluoromethylpyridone (2-hydroxy-5-trifluoromethylpyridine), a model compound for herbicides, after it had been dosed into hydroponically grown maize plants. The combination of NMR and MS data allowed the identification of both of these metabolites, namely, the N-glucoside and O-malonylglucoside conjugates of the parent pyridone. This work demonstrates the efficiency and the potential application of HPLC-NMR-MS to the investigation of the metabolism of agrochemicals. The work also indicates that combination of the use of hydroponically grown plants and directly coupled HPLC-NMR-MS allows rapid identification of metabolites with little sample preparation.

The effect of methyl substituents on the in vitro metabolism of cyclopenta[a]phenanthren-17-ones: implications for biological activity.

The in vitro metabolism of 15,16-dihydrocyclopenta[a]phenanthren-17-one and its 11- and 12-methyl derivatives has been compared. All three compounds form trans-3,4-dihydrodiols having quasi-diequatorial conformations and 3R,4R configurations. The trans-3,4-dihydrodiol of the mutagenic, but non-tumorigenic 15,16-dihydrocyclopenta[a]phenanthren-17-one appears to undergo stereospecific epoxidation to a syn-diol epoxide. By contrast the 3,4-dihydrodiol of the nontumorigenic 15,16-dihydro-12-methylcyclopenta[a]phenanthren-17-one appears to undergo stereospecific epoxidation to an anti diol-epoxide equivalent to that generated from 15,16-dihydro-11-methylcyclopenta[a]phenanthren-17-one which is a strong carcinogen. These results are discussed with reference to the biological activities of the parent compounds.

The biological activity and activation of 15,16-dihydro-1,11-methanocyclopenta[a]phenanthren-17-one, a carcinogen with an obstructed bay region.

The proposal that an unobstructed bay region is a prerequisite for tumorigenic activity in cyclopenta[a]phenanthrene-17-ones is not supported by the observation of the tumorigenicity of 15,16-dihydro-1,11-methanocyclopenta[a]phenanthrene-17-one towards the skin of T.O. mice. The title compound is oxidised in vitro by a mixed function oxidase to produce, inter alia, a trans-3,4-dihydrodiol, postulated as the proximate tumorigen. Unequivocal identification of a second metabolite as a trans-1,2-dihydrodiol derivative demonstrates the potential for enzymatic oxidation within the obstructed bay region and supports the proposal that the ultimate tumorigen is a trans-3,4-dihydrodiol-anti-1,2-oxide. This is further substantiated by the chromatographic behaviour of the major hydrocarbon-nucleoside adduct derived from mouse skin treated with the parent compound in vivo. The structures of certain others of the metabolites produced in vitro are also considered.

Carcinogenicity and mutagenicity of some alkoxy cyclopenta[a]phenanthren-17-ones; effect of obstructing the bay region.

Six 11-O-alkoxy derivatives of the phenol 15,16-dihydro-11-hydroxycyclopenta[a]phenanthren-17-one were prepared and tested for their ability to initiate skin tumours on mouse skin after topical application, followed by promotion with croton oil. The 11-methoxy derivative was the most active, but was less so than the strong carcinogen 15,16-dihydro-11-methyl-cyclopenta[a]phenanthren-17-one (Ib). The 11-ethoxy derivative was somewhat less active as a tumour initiator than the 11-methoxy compound, and the 11-n-propoxy and 11-n-pentoxy derivatives were inactive. The 11-iso-propoxy and, surprisingly, the 11-n-butoxy compound possessed weak initiating activity. The phenol, which is too insoluble inorganic solvents to be tested in this way, proved to be about half as active as a skin tumour initiator compared with the 11-methyl derivative (Ib) when these compounds were injected s.c. in oil, and this was followed by topical promotion of dorsal skin remote from the site of injection. 15,16-Dihydro-15-methoxy-cyclopenta[a]phenanthren-17-one essentially lacked tumour initiating activity on mouse skin, but its 11-methyl homologue was moderately active. The carcinogenicity of the 11-alkoxy compounds was in general paralleled by their mutagenicity to Salmonella typhimurium TA100 in the Ames' test, with the exception of the 11-phenol which was not a mutagen under these conditions (plate assay).