In-line formation and identification of toxic reductive metabolites of aristolochic acid using electrochemistry mass spectrometry coupling.

Small-molecule metabolism has been extensively studied in the past decades, notably driven by the development of new pharmaceutical ingredients. The understanding of metabolism is critical to the anticipation of reactive metabolite formation in vivo that is often associated with toxicity. Electrochemistry has been proposed to simulate the oxidoreductive metabolism reaction catalyzed by cytochrome P450, a family of microsomal enzymes strongly involved in xenobiotic metabolism. The implementation of an electrochemical cell online with mass spectrometry allows for the fast formation and identification of the reaction end products. This study discusses the ability of the synthetic electrochemical approach to simulate a complex lactamization reaction that involves the formation of reactive metabolites. Aristolochic acid I was used as a model molecule to evaluate the ability of electrochemical simulation to generate nitroso, hydroxylamine, N-hydroxylactam, lactam, and nitrenium ion metabolites.

Unexpected benzimidazole ring formation from a quinoneimide species in the presence of ammonium acetate as supporting electrolyte used in the coupling of electrochemistry with mass spectrometry.

RATIONALE: Electrochemistry (EC) coupled to mass spectrometry (MS) has been used to study different phase-I reactions. Despite of the versatility of EC/MS, the effect of the nature of the supporting electrolyte on the formation of oxidation products has seldom been discussed during EC/MS experiments. Here, we present a comparison of two different supporting electrolytes and their effect on the identification of unstable intermediate oxidation species is discussed. METHODS: The oxidation of acebutolol was performed with a coulometric cell in the presence of two supporting electrolytes namely ammonium acetate and lithium acetate. Ultra-performance liquid chromatography/quadrupole time-of-flight mass spectrometry (UPLC/QTOFMS) using a binary gradient (water/acetonitrile) with positive electrospray ionization was used to identify the oxidation products in the presence and absence of glutathione. Chemical structure elucidations of the oxidation products were performed by high-resolution mass spectrometry (HRMS) and were also supported by nuclear magnetic resonance (NMR) measurements. RESULTS: From the electrochemical study and HRMS measurements, we demonstrate that the quinoneimide species resulting from the oxidative hydrolyses of acebutolol gives a benzimidazole ring product in the presence of ammonium acetate. Through the example of the oxidation of acebutolol, a correlation between the supporting electrolyte nature and oxidation product formation was established. The obtained results were supported by quantum mechanical calculations. CONCLUSIONS: We present here evidence of the side reactions induced by the presence of ammonia as supporting electrolyte during EC/MS measurements. Acebutolol was used as a model to postulate an uncommon and unexpected side reaction leading to benzimidazole ring formation. The findings may help to understand the identification of the intermediate species in the oxidative degradation process.

In situ ultrafast 2D NMR spectroelectrochemistry for real-time monitoring of redox reactions.

The in situ implementation of an electrochemical cell (EC) inside a nuclear magnetic resonance (NMR) spectrometer is extremely powerful to study redox reactions in real time and identify unstable reaction intermediates. Unfortunately, the implementation of an electrochemical device near the sensitive volume of an NMR probe significantly affects the quality of the NMR signal, inducing significant line broadening resulting in peak overlap and partial loss of the multiplet structures. Two-dimensional (2D) NMR spectroscopy allows one to bypass signal overlapping by spreading the peaks along two orthogonal dimensions, while providing precious information in terms of structural elucidation. Nevertheless, the acquisition of 2D NMR data suffers from long acquisition durations which are incompatible with fast redox processes taking place in solution. Here, we present a new approach to deal with this issue, consisting of coupling EC-NMR with ultrafast 2D spectroscopy, capable of recording 2D spectra much faster than conventional 2D NMR. This approach is applied to the real-time monitoring of a model reaction. Fast correlation spectroscopy (COSY) spectra are recorded every 3 min in the course of the 80 min reaction, leading to the unambiguous identification of one reaction intermediate and two reaction products. The evolution of 2D NMR peak volumes in the course of time provides further insight into the mechanism of this reaction involving an unstable intermediate. This study demonstrates the feasibility and the relevance of coupling in situ spectroelectrochemistry with ultrafast 2D spectroscopy to monitor real-time electrochemical reactions in the NMR tube.

Fast spatially encoded 3D NMR strategies for (13)C-based metabolic flux analysis.

The measurement of site-specific (13)C enrichments in complex mixtures of (13)C-labeled metabolites is a powerful tool for metabolic flux analysis. One of the main methods to measure such enrichments is homonuclear (1)H 2D NMR. However, the major limitation of this technique is the acquisition time, which can amount to a few hours. This drawback was recently overcome by the design of fast COSY experiments for measuring specific (13)C-enrichments, based on single-scan 2D NMR. However, these experiments are still limited by overlaps because of(1)H-(13)C splittings, thus limiting the metabolic information accessible for complex biological mixtures. To circumvent this limitation, we propose to tilt the (1)H-(13)C coupling into a third dimension via fast-hybrid 3D NMR methods combining the speed of ultrafast 2D NMR with the high resolution of conventional methods. Two strategies are described that allow the acquisition of a complete 3D J-resolved-COSY spectrum in 12 min (for concentrations as low as 10 mM). The analytical potentialities of both methods are evaluated on a series of (13)C-enriched glucose samples and on a biomass hydrolyzate obtained from Escherichia coli cells. Once optimized, the two complementary experiments lead to a trueness and a precision of a few percent and an excellent linearity. The advantages and drawbacks of these approaches are discussed and their potentialities are highlighted.

Fast quantitative 1H-13C two-dimensional NMR with very high precision.

Quantitative analysis by nuclear magnetic resonance (NMR) requires highly precise measurements to achieve reliable quantification. It is particularly true in (13)C site-specific natural isotope fractionation studied by nuclear magnetic resonance, where the range of values of (13)C isotopic deviations at natural abundance is highly restricted. Consequently, an NMR method capable of measuring δ(13)C ‰ values with a very high precision (a few per mil) is indispensable. This high degree of precision has already been achieved by one-dimensional (13)C acquisitions; however, this approach is limited by peak overlaps which reduce the precision of the isotope content determination, even for certain small molecules. It is therefore necessary to extend this promising methodology to a higher dimensionality. In this context, this paper aims at determining conditions that allow the achievement of two-dimensional (2D) (1)H-(13)C heteronuclear experiments with a precision of a few per mil in a reasonable time. Our results demonstrate that a high precision (repeatability of 2 per mil) can be reached with the (1)H-(13)C HSQC (Heteronuclear Single Quantum Correlation) experiment, thus satisfying the conditions needed to perform (13)C isotope analysis by 2D NMR. We also consider the impact of several approaches which have been proposed to reduce the duration of heteronuclear 2D experiments. Two of these common time-saving strategies, spectral aliasing and linear prediction, are fully compatible with the high-precision requirements of isotopic NMR, while a third one, nonuniform sampling, leads to dramatic precision losses. In conclusion, this study demonstrates the feasibility of very precise 2D NMR measurements and opens a number of application perspectives.