Abstract
Metabolomics aims at studying the whole metabolite set contained in biological media. The main objectives are to identify and quantify all the small molecules present in complex matrices in a comprehensive manner. Due to its ability to detect hundreds of metabolites in standard matrices, high resolution electrospray ionization mass spectrometry (HRMS) coupled to liquid chromatography is one of the most widely used analytical techniques in metabolomics.
Despite the constant improvements of mass spectrometers in terms of sensitivity, mass accuracy, resolution, and acquisition speed, identification of unknown metabolites remains challenging. Accurate metabolite identification is crucial to translate relevant analytical data into meaningful biological information. The first level of investigations when using HRMS data relies on the robust and accurate determination of an elemental composition (EC) for a given MS peak. Altogether, mass accuracy, mass resolution, and isotopic pattern accuracy directly influence the ability of HRMS instruments to perform accurate EC determination. Recent Orbitrap instruments can routinely achieve mass accuracies below 3 ppm (with external calibration), whereas Fournier transform ion cyclotron resonance (FTICR) mass spectrometers are characterized by an even better mass accuracy below 0.2 ppm. However, high mass accuracy alone is not sufficient to assign a unique and unambiguous EC to a measured mass. For instance, Kind and Fiehn showed that a mass spectrometer capable of 3 ppm mass accuracy and 2% error for isotopic abundance patterns should outperform mass spectrometers providing a mass accuracy of 0.1 ppm but results in poor performances in terms of relative natural isotope abundance (RIA) measurement. RIA has already been studied on different mass spectrometers, but these studies deal essentially with natural 13C1 isotope. The RIA of this isotope appears to be rather commonly underestimated when measured on an Orbitrap as well as on FTICR instruments. As an example, a mean RIA error standing between +5% and −50% is commonly observed, depending on ion intensity. Increased resolutions and low ion abundances also negatively impact the quality of the RIA measurement.
Examining the fine isotopic distribution, in particular that of the M + 2 isotopes, can also bring valuable information for refining EC list. However, although not a real issue for low mass metabolites, it is estimated that a full width at half-maximum (fwhm) resolution over 300,000 is required to discriminate between 15N and 13C1 but also between 34S, 18O, and 13C2 species for metabolites with a mass of about 500 Da. Until recently, only FTICR instruments could reach such required resolution, but the newly introduced high-field Orbitrap instruments, such as the Orbitrap Fusion, are now capable of achieving mass resolutions higher than 500,000 (at m/z 200, fwhm), for instance rendering 18O and 13C2 isotopic peaks distinguishable. Due to their natural abundance, 34S isotopes are quite convenient to study as recently done by Blake et al. on an FTICR instrument. However, and to the best of our knowledge, no work has been published regarding specific 18O RIA measurement for EC determination of unknown metabolites on FTICR or on an Orbitrap instrument.
The aim of this article is therefore to evaluate the effectiveness of the newly introduced high resolution Orbitrap Fusion for accurately measuring metabolite masses and relative isotopic abundances to expedite EC determination. A set of 50 standard molecules dissolved in a water–acetonitrile solution was first used in order to optimize analytical conditions and then after being spiked into a complex plasma metabolite extract to evaluate the potential impact of matrix interferences. In addition to being the first description of the Orbitrap Fusion for metabolomics, this report also provides optimized operating conditions for getting the best of this instrument in this context.
Read the French version.