Exploratory analysis using MRM profiling mass spectrometry of a candidate metabolomics sample for testing system suitability

https://doi.org/10.1016/j.ijms.2021.116663Get rights and content

Highlights

  • Small molecule screening guided by chemical functionality was performed in human liver standard candidate from NIST.

  • Flow injection sample delivery followed by neutral loss, precursor ion and multiple reaction scans were used.

  • MRMs for lipids (N = 191), metabolites (N = 104) and exogenous compounds (N = 17) and their tentative attributions are reported.

  • Relative standard deviation measurement for the MRMs reported were lower than 25%.

  • The analytical workflow described allows data acquisition rate of 50 compounds/min, using two MRMs for each compound.

Abstract

Multiple reaction monitoring (MRM) profiling is an exploratory mass spectrometry (MS) method applicable to the initial screening of complex samples for small molecules based on their chemical functionalities. We report on the applicability and quality of this method to screen for metabolites, lipids and exogenous compounds in a candidate reference material, the Metabolomics System Suitability Research Grade Material (RGM 10122) which is being developed by the National Institute of Standards and Technology (NIST). In an initial discovery experiment, we recorded data using eighty neutral loss (NL) and precursor ion (Prec) MS/MS scans, selected from literature data as likely of value in recognizing the presence of potential lipids and metabolites. Then the NIST sample was re-examined combining precursor-to-product ion transitions from the discovery experiment with a list of 1357 known literature-based metabolite MRMs. This MRM profiling experiment gave a small set (191) of high-quality lipid specific MRMs for the sample. Similar experiments gave 104 and 17 metabolite and exosome MRM's. These MRM experiments, with a few exceptions, showed relative standard deviations (RSD) under 25% for individual tentatively assigned compounds. At a data acquisition rate of 50 compounds/min, using two MRMs for each compound, this approach allows quick surveys for easily detectable compounds in complex samples.

Introduction

Continued progress in small molecule analysis by mass spectrometry (MS) has reinforced the need for analytical standards for biological matrices, such as biofluids and tissues. Such complex standards are necessary to support the development and validation of workflows for sample preparation, measurement, and data analysis, and are required for the evaluation of inter-laboratory comparability. One of the first such initiatives involving the use of a human biofluid analytical standard for metabolomics was taken in 2010 when the LIPID Metabolites and Pathways Strategy (LIPID MAPS) consortium published the analytical results of NIST Standard Reference Material (SRM) 1950 Metabolites in Frozen Human Plasma and quantified the abundant lipid species [1] Subsequent use of this NIST SRM contributed to a harmonization effort undertaken through an interlaboratory comparison exercise involving 31 different sites [2].

A wide range of chemical functional groups occurs in individual lipids and metabolites and they can be categorized as polar (metabolites, some lipids, and most drugs) or non-polar (some lipids) compounds. The structural variety extends even further if one considers exogenous compounds such as synthetic chemicals, drugs and their metabolites; the human metabolome can be used as a basis to study exposure to exogenous compounds and hence to provide a framework for study of exposure–response relationships [3].

Full mass scan exploratory analysis for the identification of small molecule metabolites, lipids, and exogenous compounds typically utilizes commercial platforms often including liquid-chromatography (LC) coupled to high-resolution mass spectrometers (HRMS), especially time-of-flight (ToF) and Orbitrap instruments, or hybrids like the quadrupole/ToF. Sample separation by LC is followed by product ion MS/MS scans. It is important to note that even though the product ion scan MS/MS is the most common MS/MS experiment, this is just one of the four types of MS/MS experiments that can be performed in triple quadrupoles and other types of mass spectrometers. Two other MS/MS scans, namely the precursor ion scan (Prec) and the neutral loss scan (NL), allow profiling for chemical functionalities since these frequently yield class-diagnostic product ions or neutral losses [4]. Therefore, while the product ion scan is ideal for obtaining structural information on a specific precursor ion and hence a specific compound, Prec and NL are powerful scan modes for exploratory analysis of small molecules at the level of the chemical functional group.

From these considerations, an alternative exploratory method has emerged in which Prec and NL scans are used in experiments (or alternatively taken from the literature) to discover precursor-to-product transitions which are distinctive for particular functionalities. Once these are recognized, then individual samples can be screened rapidly for these chemical signatures using a set of transitions assembled in a multiple reaction monitoring (MRM) set. This procedure, under the name MRM profiling, has been described [5] and reviewed [6]. Chromatographic separation and the use of internal standards for structural confirmation or absolute quantification are reserved for a final target validation step [7] when such a step is deemed necessary to solve the particular problem in hand. Examples of published applications of the MRM profiling method include identification of cases of drug resistance in bacteria [8], profiling of epidermal lipids in a mouse model of dermatitis [9], lipid extract stability [10] and characterization of peri-ovarian adipose tissue [11]. MRM profiling has also been integrated with proteomics in analysis of the gold nanoparticle biocorona in healthy and obese populations [12]. Structural confirmation and absolute quantification have been performed in some studies, but the main objective of the MRM profiling method is to survey the chemical composition of the sample at a higher level than individual compound identification in order to guide further exploratory and validation efforts. This approach finds particular value in the classification of samples into related groups. We envisage that the MRM profiling approach can support efforts in which MS is applied to point-of-care analysis, specially using simple instruments and low cost materials such as paper spray ionization [13,14].

In this research, the MRM profiling method was applied for the exploratory analysis of lipids, metabolites, and exogenous compounds in the NIST material currently being developed as a Metabolomics System Suitability Sample (NIST RGM 10122), which will soon become a NIST standard. This sample is a dried polar extract of human liver intended for evaluation of instrument suitability and performance in advance of a metabolomics study. In addition, the NIST liver extract sample also served to provide information on the reproducibility of the MRM methodology.

Section snippets

RGM 10122 production

Liver tissues were cryo-homogenized at the NIST laboratory in Charleston (SC, USA) using established protocols [20]. Homogenized liver powder was extracted with 70% volume fraction ethanol in water and stored at −80 °C to facilitate protein precipitation. The liver sample was then centrifuged at 4 °C, the supernatant was transferred, and the centrifugation step was repeated. The final supernatant was diluted twice (2×) with 70% volume fraction ethanol in water, dispensed into autosampler vials

Results and discussion

This research interrogated for selected small molecule chemical functionalities in the NIST RGM 10122 polar liver extract using an MRM profiling strategy. A clear understanding of the four MS/MS scan modes in mass spectrometry (Prec, NL, Product ion scan and MRM) as outlined by Schwartz et al. 1990 and recently reviewed [6] is necessary to draw parallels between MRM profiling and the typical exploratory methods for small molecule composition in biological samples.

The experiments reported here

Authors’ contributions

MEE and CAM contributed equally to the experimental work and preparation of data for publication; TBS contributed systems knowledge and advice on presentation; TJPS made key contributions to data processing; CRF and RGC are responsible for conceiving the MRM profiling.

Disclaimer

Certain commercial equipment, instruments, or materials are identified in this paper in order to specify the experimental procedure adequately. Such identification is not intended to imply recommendation or endorsement by the National Institute of Standards and Technology, nor is it intended to imply that the materials or equipment identified are necessarily the best available for the purpose.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We acknowledge the support of the National Science Foundation (NSF) CHE 1905087; Indiana Clinical and Translational Sciences Institute funded, in part, by Award Number UL1TR002529 from the National Institutes of Health, National Center for Advancing Translational Sciences, Clinical and Translational Sciences; Bindley Bioscience Center; and São Paulo Research Foundation (FAPESP2019/03385-7 and n° 2018/11700-7). We appreciate the creativity, knowledge and effort of Clay Davis, Debra Ellisor,

References (38)

Cited by (5)

View full text