Polar metabolomics using trichloroacetic acid extraction and porous graphitic carbon stationary phase.

doi:10.21203/rs.3.rs-3784240/v1

Download PDF

Research Article

Polar metabolomics using trichloroacetic acid extraction and porous graphitic carbon stationary phase.

https://doi.org/10.21203/rs.3.rs-3784240/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 15 Jul, 2024

Read the published version in Metabolomics →

You are reading this latest preprint version

Accurately identifying and quantifying polar metabolites using untargeted metabolomics has proven challenging in comparison to mid to non-polar metabolites. Hydrophilic interaction chromatography and gas chromatography-mass spectrometry are predominantly used to target polar metabolites. However, the scope of these methods are limited and the extraction and post-analyses can be laborious. Using serum from 6 healthy individuals we demonstrate a simple one step MilliQ + 10% trichloroacetic acid (TCA) extraction combined with porous graphitic carbon liquid chromatography – mass spectrometry (LC-MS) that can reliably retain (Coefficient of variation [CV] <0.30) a range of polar metabolites with logp as low as -9.1 (Range:-9.1-5.6). The use of QreSS^TM (Quantification, Retention, and System Suitability) internal standards determined this method consistently (CV<0.16), and effectively (70-130% recovery) recovers polar to mid-polar metabolites within biologically relevant concentration ranges. Quality control dilution series demonstrated ~80% of annotated metabolites can be accurately (Pearsons’s correlation coefficient >0.80) quantified within their concentration range. Repeatability of this method was demonstrated through clustering of repeated extractions from a single sample. This LC-MS method is better suited to covering the polar segment of the metabolome than current methods.

polar metabolites

trichloroacetic acid

porous graphitic carbon

LC-MS

HILIC

untargeted metabolomics

We present a simple, fast and novel LC-MS based method.
TCA efficiently extracts a range of biologically relevant polar to mid-polar metabolites and is compatible with RPLC.
PGC retains and resolves more polar metabolites than conventional RP stationary phases.
Polar metabolites are accurately, reproducibly and precisely, qualitatively and quantitively measured within their dynamic range.

Polar metabolites have important biological functions in energy metabolism, biosynthesis and signalling. Dysregulation of polar metabolite concentrations can manifest as disease phenotypes. For example, increased concentrations of branched-chain amino acids are positively correlated with obesity (Bagheri et al. 2019; Szczerbinski et al. 2022) aspartate, glutamate, hypoxanthine, lactate, proline, and pyroglutamate concentrations are elevated in women with cervical cancer. By default Khan et al. (2019), isoleucine, phenylalanine and tyrosine concentrations cumulatively correlate with increased risk of diabetes (T. J. Wang et al. 2011). The ability to easily and accurately quantify a large range of these compounds in readily accessible samples (e.g. blood) will enable identification of polar metabolites causatively or correlatively associated with diseases, providing evidence of potential mechanisms contributing to disease phenotypes.

Untargeted metabolomics seeks to identify and quantify all metabolites in a sample (Roca et al. 2020). However, this is not currently possible due to their diverse properties and abundances. The range of metabolite solubilities, the sensitivity and specificity of separation techniques available and ionisation abilities necessitate multiple methods to target non-polar metabolites (lipids), mid-polarity & polar metabolites, and those that only ionize in one or another polarity. Metabolite polarity mainly determines which metabolomic method is most suitable to accurately identify and quantify it (Roca et al. 2020). To capture the entire metabolome, it is necessary to use multiple methods, although this is difficult due to cost, and restrictions in time, instrument access and practical experience (Haggarty and Burgess 2017).

Reverse phase liquid chromatography [RPLC] on in silica particulate columns with octadecylsilane stationary phase is the most reproducible and widely applicable chromatographic technique for metabolite identification and quantification (Zeki et al. 2020). Conventional RPLC retains and resolves metabolites with partition coefficient values higher than zero (logp > 0) i.e. mid- to non-polar metabolites (Dudzik and García 2021; Roca et al. 2020). Chromatographic resolution of more polar metabolites (logp < 0) typically necessitates either hydrophilic interaction chromatography [HILIC] or derivatisation for analysis by gas-chromatography [GC] (Rakusanova et al. 2023; Roca et al. 2020; Zeki et al. 2020). In this paper we evaluate chromatographic resolution of polar metabolites using porous graphitic carbon [PGC] as a stationary phase. PGC exhibits increased retention of polar metabolites over C18 phases with RPLC, and shorter between-run stabilisation periods (Ramautar and De Jong 2014). The extraction for PGC chromatography is simpler than the extraction methods required for HILIC (Ramautar & De Jong, 2014).

The high organic solvent starting conditions that are needed in HILIC require that samples are dissolved in a matched matrix. Whilst methanol and acetonitrile are commonly used to deproteinate metabolites for LC-MS-based analysis, it has long been known that this yields poor recovery of amino acids (Sedgwick et al. 1991). Aqueous solutions of perchloric or trichloroacetic acid [TCA] can be used as an alternative deproteinisation strategy to organic solvents. Such aqueous extracts are incompatible with HILIC, but they are compatible with RPLC and positive electrospray ionisation mass spectrometry. However, most RPLC stationary phases have limited or no retention of metabolites with logp < 0. Ortmayr et al. (2015) demonstrated that PGC columns can separate 30 polar metabolites that were not retained by RPLC, such as amino acids and sugar phosphates. Here we describe our work to explore the retention and separation power of this relatively underappreciated stationary phase for the resolution of the polar metabolome. We demonstrate the ability of a novel LC-MS method that is sensitive, repeatable and accurately identifies and quantifies a range of polar metabolites, offering a new solution for untargeted polar metabolomics.

Objective

We set out to validate the performance of a novel LC-MS method with a TCA extraction and porous graphitic carbon stationary phase.

2.1 Ethics

Serum collection was approved by the Central Health and Disability Ethics Committee of the University of Auckland (20/CEN/69). The study was prospectively registered with the Australian New Zealand Clinical Trials Registry at www.anzctr.org.au (ACTRN12620000629932p).

2.2 Solvents and Chemicals

LC-MS grade solvents and modifiers were obtained (methanol, Merck; acetonitrile, Optima LC-MS grade; formic acid and ammonium formate, Thermo Fisher Scientific (Waltham, MA, USA)). Type 1 water was generated using a Millipore unit (Merck Millipore, Auckland, New Zealand). Trichloroacetic acid was obtained from Merck. QReSS™ (Quantification, Retention, and System Suitability) internal standard kit (cat no. MSK-QReSS-KIT) was obtained from Cambridge Isotope Labs Inc. (Tewksbury, MA, USA).

2.3 Human serum samples

Serum was collected from six healthy individuals and stored at -80°C prior to analysis. Samples were de-identified prior to use. Three technical repeats were extracted from a single serum sample from each individual. An additional QC sample was created by combining equal parts of all six individuals’ serum samples.

2.4 Sample preparation

Pre-extraction, 10 µL of QReSS internal standard mixture (1.25µL of QReSS internal standard1, 1.25 µL of QReSS internal standard2, 7.5µL of water) is added to a 20 µL aliquot of serum in a 1.5 mL Eppendorf tube and vortexed (3s). To extract metabolites from the serum and internal standard mix, 50 µL of 90:10 mix of milliQ water [MQ] and TCA is added to the Eppendorf tube and vortexed (3s). The sample is then incubated on ice (5 min), while the metabolites separate from serum. Samples are centrifuged (12,000 Relative Centrifugal Force [RCF], 5 min, 4^oC). Retain 50 µL of the supernatant and dilute it four-fold with MQ. Vortex (3s) prior to LC-MS analysis.

Linearity of the analysis is assessed via serial extractions using varying volumes (20 µL, 30 µL, 40 µL, 50 µL) of the supernatant.

2.5 Liquid chromatography – Mass spectrometry (LC-MS) conditions

LC-MS was performed using an Accela 1250 pump (Thermo Fisher Scientific, Waltham, MA, USA), a Thermo TriPlus autosampler maintained at 8°C (Thermo Fisher Scientific, Hampton, VA, USA), a HotDog5090 column oven and a porous graphitic Supel™ Carbon LC 2.7 μm HPLC Column (150 mm × 2.1 mm × 2.7 μm) maintained at 30 °C. Flow was 0.3 mL/min and the injection volume was 20 µL. The mobile phase is (A) MQ H₂O with 10 mM ammonium formate, 0.1% formic acid, (B) methanol with 10 mM ammonium formate, 0.1% formic acid, (C) acetonitrile with 0.1% formic acid.

The mass spectrometer was a Q-Exactive (Thermo Scientific, Dreieich, Germany) with a heated electrospray ionisation [ESI] source. Source voltage was 3.5 kV, S-lens RF level 50, heated capillary temperature 263 °C, AGC target 1^e6 and resolution 17,500. Mass spectra were acquired in positive polarity and centroid mode. The scan range over the entire run was m/z 50–750 with stepped normalised collision energy (NCE) of 20, 30, 40 eV.

Data-dependent acquisition mode [DDA] was used to acquire data for qualitative analyses. The mobile phase gradient starts at 0% B and C for 1 min, then ramped down to 0% B and 10% C over 2 mins, then ramped down to 0% B and 30% C over 2 mins, then ramped down to 0% B and 97% C over 3 mins. The gradient was subsequently held at 10% B and 0% C for 1 min before being re-equilibrated to 0% B and 0% C from 9 to 15 min.

2.6 Data Processing

Chromatographically resolved features with MS/MS data were processed using MS-Dial (v4.9) for peak picking, alignment, integration and spectral matching. Features were annotated by reference to an in-house MS/MS library based on the 2020 NIST library (National Institute of Standards & Technology, Gaithersburg, MD, USA) (Tsugawa et al. 2015). Peak areas were blank-subtracted and normalised to recovery of the internal standard with the closest retention time. Ontological classifications for the annotations were scraped from the Human Metabolome Database v4.0 [HMDB] (Wishart et al. 2018), where present.

We observed 11 of the 16 compounds in the QReSS internal standard that ionise in positive mode (Supplementary figure 1). Tryptophan was not included in recovery calculations as its signal is lost rapidly in successive runs. Of the remaining 10 observed metabolites, six (i.e. creatinine, hypoxanthine, phenylalanine, thymine, tyrosine and vitamin B3) are recovered within the range of 70-130%, which is considered acceptable by targeted method validation criteria (figure 1e)(Alseekh et al. 2021; APVMA 2014). However, Alseekh et al. (2021) suggest that internal standards recovered up to 50%, if linear and reproducible, may be considered acceptable. QReSS alanine and ethanolamine recoveries were ~36%. Leucine was just below the threshold with 65% recovery. All detected internal standards had a mean Coefficient of Variation [CV] between 0.03-0.16. Labelled alanine and ethanolamine (<70% recoveries) had mean CVs of 0.07 and 0.06, respectively.

We detected 14,685 features and collected 1,907 MS/MS spectra that allowed us to annotate 274 features with a dot product (match) factor above 0.75. The percent of annotated features which met CV criteria 0.15, 0.20 and 0.30, were 66.8%, 83.7% and 90.9%, respectively (Figure 2a).

We conducted a dilution series of post-extraction human serum supernatant to assess the linearity of the recovery response (i.e. coefficient of determination [R²] vs. Pearson correlation coefficient [PCC]) for each metabolite). The R² value is a measure of the proportion of variance in the detected abundance of a metabolite that is explained by the actual change in concentration of that metabolite due to the dilution. The relationship between the R² and PCC is quadratic, and both have specified threshold limits of > 0.80 for untargeted metabolomics (Fda and Cder 2019). The PCC and R² threshold was met by 79.2% and 68.0% of our annotated metabolites, respectively (figure 2c). Three metabolites had negative PCC values, in which absolute levels |R²| did not pass the > 0.80 limit (Supplementary figure 2). For example, theophylline had the largest negative PCC (-0.64) from actual concentration increase to abundance detected. However, peak visualisation for theophylline identified multiple irregularly shaped peaks likely due to partial elution. The PCC for internal standards ethanolamine, 1,4-butanediamine, alanine, leucine, phenylalanine, vitamin B3 and thymine had high PCCs with values of 0.97, 0.94, 0.97, 0.95, 0.94 and 0.86. Tyrosine was lower, at 0.52. All internal standards had R² thresholds above 0.87, except thymine (0.73) and tyrosine (0.27).

Clustering the samples according to metabolite abundance identified greater between individual variation than was observed for repeated extractions from the same serum sample (figure 2b).

Twenty-one metabolite classes were annotated using this method (figure 3). Most of these identified metabolites were carboxylic acids and their derivatives (55.4%). Sub-classifications within this class include amino acids, peptides, and analogues. By comparison, the next most abundant classes of compounds, fatty acyls occur at 8.46% of the annotated and class-identified metabolites, 7.48% are organooxygen compounds, 3.85% are organonitrogen compounds and 3.85% are benzene and substituted derivatives.

Polar metabolites represent an important fraction of the metabolome. Our novel RPLC-MS/MS method uses a PGC column to quantify 14,685 features in human serum and collect MS/MS spectra to annotate 1,907 features. Fifty five percent of annotated features were carboxylic acids and derivatives. Triplicate analyses of six individual’s serum demonstrated that 90% of the annotated features had CVs lower than the widely used untargeted metabolomics threshold of 0.30 (Brunius et al. 2016). Appropriate CV thresholds for metabolite analyses are determined according to the aim of the study. For example, less strict CV thresholds are typically applied to exploratory untargeted metabolomics analyses to increase metabolome coverage (Brunius et al. 2016). Thresholds for acceptable CVs for features in untargeted metabolomics vary but CVs between 0.20 (Dudzik et al., 2018) to 0.30 (Brunius et al. 2016) are generally accepted. Six internal standards (i.e. creatinine, hypoxanthine, phenylalanine, thymine, tyrosine and vitamin B3) had recoveries within 70–130% of the spiked blanks, five of these with CVs < 0.10, an additional four internal standards (i.e. leucine, alanine, ethanolamine and 1,4-butanediamine) met the acceptable 0.15 > CV for targeted metabolomics, specified by the United States Food and Drug Agency [FDA] (Fda and Cder 2018). Thus, we claim the method we developed accurately and precisely quantifies polar metabolites in human serum.

Consensus criteria for the evaluation of the performance of untargeted metabolomics methods are still to be universally adopted. However, we recognise the comments of Dudzik et al., Kirwan et al., Rakusanova et al. (Dudzik et al. 2018; Kirwan et al. 2022; Rakusanova et al. 2023) which include: the need to clearly communicate thought-out methodology; QCs and blanks to assess method consistency, spiked internal standards relevant to the targeted metabolites; and to keep the metabolite co-efficient of variation as low as possible. We incorporated these suggestions into our demonstration of the accuracy, reproducibility and utility of a novel LC-MS/MS method that identifies and quantifies polar metabolites. Our results compare well with results of HILIC or derivatisation-GC-MS methods (Rakusanova et al., 2023; Zeki et al., 2020). For example, Zhang et al. (2023) detected 755 positive ions in serum samples from 131 hypertensive children. HILIC methods (e.g. Zhang et al. (2023)) often require repetitive drying and reconstitution steps, using a combination of methanol and acetonitrile extractions. These methods are time consuming and modify metabolites. Our method requires a simple single extraction and centrifugation step, that can be completed within 15 minutes. Furthermore, each sample takes only 15 minutes of LC-MS time to analyse, including re-equilibration of the column. Thus, our method is quick, easy and able to accurately and precisely identify a wide range of polar metabolites.

Aqueous TCA extraction and porous graphite LC-MS/MS identifies polar metabolites with logp values ranging from − 9.1 to 5.6 (Supplementary Fig. 3). Metabolites with logp < 0 are typically described as polar. Approximately 70% of annotated metabolites captured by our method are polar. Oxidised glutathione is the annotated metabolite with the lowest partition coefficients (-9.1).

Derivatisation-GC-MS is useful for polar metabolomics. However, derivatisation-GC-MS tends to detect fewer features than LC-MS and requires additional pre-processing steps that often impact sample integrity. For example, a recent GC-MS study detected 384 features (Eylem et al. 2022). Similarly, Rey-Stolle et al. detected only 241 features and of these 153 had a CV < 30 and 123 had a CV < 20 (Rey-Stolle et al. 2022). Furthermore, when used in conjunction with LC-MS to extend coverage to the non-polar fraction of the metabolome, GC-MS requires extra training and is costly (Haggarty and Burgess 2017).

Previously, TCA as a metabolite extraction solvent in combination with NMR-based metabolomics was claimed to be a poor deproteinisation reagent that introduces artifacts (Daykin et al. 2002). We have demonstrated aqueous TCA to be an effective deproteinisation and extraction solvent for a wide range of polar metabolites and did not observe similar artifacts. Ninety percent of annotated metabolites, predominantly ‘carboxylic acids and derivatives’, were consistently recovered with a CV < 0.30. The disadvantage of using TCA is that it is a relatively strong acid (pKa = 0.66) that can hydrolyse esterified compounds. Moreover, the high concentration of halogenated acid is inappropriate for ESI- modes, as it is anionic and quenches ionisation. Although ~ 80% of metabolites ionise to some degree in ESI+. Additionally, the water component of the TCA solution is more polar than methanol or acetonitrile and thus the process is less effective at extracting mid- to non-polar metabolites. However, this is advantageous when the mid- to non-polar metabolites are not the identified targets of the method. This is particularly true with PGC which irreversibly adsorbs molecules with planar saturated hydrocarbon moieties such as tryptophan.

Polar internal standards (i.e. creatinine, tyrosine, hypoxanthine, thymine and vitamin B3) were highly recovered by the paired TCA extraction and PGC chromatography method. By contrast, polar internal standards alanine and ethanolamine were not, possibly due to their loss during extraction, or ion suppression (Alseekh et al. 2021). Quantitation of creatinine, tyrosine, hypoxanthine, thymine and vitamin B3 were highly linear with abundance PCC values of 0.97, when compared to dilution, suggesting that they are not entering the machine early enough to be affected by void enhancement or suppression. Tryptophan appears to be lost throughout the run possibly because it has an aromatic moiety that is strongly retained by the PGC, and only partially elutes. A stronger solvent mixture might resolve this, adding extra value to the method we have developed.

The use of internal standards is a major strength of our study, particularly as the concentrations are at the lower end of their adult physiological quantities acquired from HMDB, demonstrating that the assay yields robust and biologically relevant information (Wishart et al. 2018). This is in part because metabolite identification and quantitation is restricted by the mass spectrometer dynamic range. Dynamic range is determined by metabolite abundance, methodology and instrument selection and can be tested by determining metabolite linearity by serial dilutions. Accurate quantification of metabolites is important to enable deciphering differences in metabolite concentrations between groups or individuals over time. Non-linearity can distort these differences to make them less or more significant than they are. Therefore, parametric tests, which use numerical value differences to determine significance, cannot be used on these metabolites. However, for those metabolites that are positively correlated (99.8%) non-parametric statistical tests that use ranks from observed data instead of the size of relative abundance difference, can be used to determine observable differences between samples.

Linearity is particularly important for internal standards, as these are used to normalise other metabolites and to reduce batch effects. Thus, non-linearity in internal standards may lead to distortions in the relative abundances of metabolites that are being normalised. QreSS internal standards, 1,4-butanediamine, alanine, leucine, phenylalanine, vitamin B3, and thymine had high observable r values to actual concentration therefore are reliable internal standards to use. However, tyrosine was relatively poorly correlated and is therefore less reliable as an internal standard for sample preparation and detectability. Interestingly tyrosine sensitivity seems to decline throughout the run, which would describe why its CVs are high in comparison to the other internal standards. This is likely due to tyrosine like tryptophan also being aromatic.

Unsurprisingly as metabolite concentration is more highly correlated to detected abundance, the R² of metabolite concentration is increased. The variance explained by the fitted regression model for 68.0% of annotated metabolites is higher than the acceptable 0.80 threshold, suggesting that the actual change in concentration for over two thirds of annotated metabolites is driving the differences we observed. Most internal standards sit above the 0.8 threshold and their detected abundance is relative to their actual concentration. Thus, the internal standards are reliable for normalisation across the tested concentration range.

‘Carboxylic acids and derivatives’, and ‘fatty acyls’ are the top classifications of our annotated metabolites that were identified in HMDB. These metabolites can be annotated to level 2 of the MSI guidelines (Sumner et al. 2007). Carboxylic acids are polar metabolites, while fatty acyls can be polar or non-polar dependent on the attached hydrocarbon chain. These classes have previously been identified in more complex LC-MS and derivatisation-GC-MS methods (Bian et al. 2020; Yu et al. 2021; Zhu et al. 2016). However, many metabolites were unable to be successfully annotated, and fewer than 50% (130/274) of them could be classified using the HMDB 2020 library. Increasing the number of annotated and classified metabolites would likely result in even greater diversity in our classifications.

Future developments of this method might incorporate stronger solvents to enable quantification of metabolites that strongly bind to the PGC. Further validation of this method using predefined metabolomic mixes (i.e. credentialing kit) may be used to show the MS/MS acquired features are not contaminants (L. Wang et al. 2019).

Metabolite extraction using aqueous TCA, combined with PGC chromatography provides accurate and precise, qualitative and quantitative data on a range of biologically relevant polar metabolites in particular carboxylic and amino acids. We collated widely accepted thresholds and values to rigorously evaluate our method. Using consensus untargeted metabolomics criteria, more than 80% of the annotated features detected using this method can be quantified accurately and within their dynamic range. Importantly to our aim, most of these metabolites were polar. Further repeatability of this method was demonstrated through clustering of repeated extractions from a single sample. The use of spiked QReSS internal standards demonstrated the precision of recovery of most polar and some mid-polar internal standards. Those that were not well recovered may require a stronger mobile phase solvent to elute from the PGC. Even without these additional improvements, this LC-MS method is comparable, if not better suited to covering the polar segment of the metabolome than current methods. Researchers that want a simple and quick method to accurately and precisely, qualitatively and quantitatively detect the polar segment of the metabolome will find this method useful.

Acknowledgements

Availability of data and materials

The raw data files generated in this study are available on request but cannot be shared outside the country of origin for reasons of data sovereignty.

Funding

The serum samples used in this study were collected as part of the project Tūhauora: A functional beverage containing the taonga species kawakawa (Piper excelsum) for South Asian markets, funded by High-Value Nutrition Ko Ngā Kai Whai Painga National Science Challenge as grant number HVN1917 with co-funding from Wakatū Incorporation, Te Tauihu, Aotearoa New Zealand. Frankie Day’s PhD was funded by a UoA Scholarship. Chris Pook's Fellowship was funded by a donation from Shundi Group.

Authors and Affiliations

Liggins Institute, Waipapa Taumata Rau—The University of Auckland, 85 Park Road, Private Bag 92019, Auckland 1142, New Zealand

Frankie Day, Farha Ramzan Justin O’Sullivan & Chris Pook

The Maurice Wilkins Centre, The University of Auckland, Auckland, New Zealand

Justin O’Sullivan

MRC Lifecourse Epidemiology Unit, University of Southampton, United Kingdom

Justin O’Sullivan

Australian Parkinson’s Mission, Garvan Institute of Medical Research, Sydney, New South Wales, 384 Victoria Street, Darlinghurst NSW 2010, Australia

Justin O’Sullivan

A*STAR Singapore Institute for Clinical Sciences, Singapore, Singapore

Justin O’Sullivan

Authors contributions

Conceptualization, FD and CP; investigation, FD, FR and CP; writing—original draft preparation, FD and CP; writing—review and editing, CP and JOS; visualization, FD; supervision, CP and J.O.S.

Corresponding authors

Correspondence to Chris Pook

Ethics declarations

Ethics approval

This study was conducted in accordance with the Declaration of Helsinki, and was approved by the Southern Health and Disability Ethics Committee (20/STH/236). The studies are prospectively registered with the Australian New Zealand Clinical Trials Registry at www.anzctr.org.au (accessed on 14 March 2022) TOAST: ACTRN12621000311853.

Consent to participate

Written informed consent was obtained from all subjects involved in this study.

Consent for publication

All authors approve the manuscript for publication.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Alseekh, S., Aharoni, A., Brotman, Y., Contrepois, K., D’Auria, J., Ewald, J., et al. (2021). Mass spectrometry-based metabolomics: a guide for annotation, quantification and best reporting practices. Nature Methods 2021 18:7, 18(7), 747–756. https://doi.org/10.1038/s41592-021-01197-1
Bagheri, M., Djazayery, A., Farzadfar, F., Qi, L., Yekaninejad, M. S., Aslibekyan, S., et al. (2019). Plasma metabolomic profiling of amino acids and polar lipids in Iranian obese adults. Lipids in Health and Disease, 18(1), 1–9. https://doi.org/10.1186/S12944-019-1037-0/FIGURES/3
Bian, X., Qian, Y., Tan, B., Li, K., Hong, X., Wong, C. C., et al. (2020). In-depth mapping carboxylic acid metabolome reveals the potential biomarkers in colorectal cancer through characteristic fragment ions and metabolic flux. Analytica Chimica Acta, 1128, 62–71. https://doi.org/10.1016/J.ACA.2020.06.064
Brunius, C., Shi, L., & Landberg, R. (2016). Large-scale untargeted LC-MS metabolomics data correction using between-batch feature alignment and cluster-based within-batch signal intensity drift correction. Metabolomics : Official journal of the Metabolomic Society, 12(11). https://doi.org/10.1007/S11306-016-1124-4
Daykin, C. A., Foxall, P. J. D., Connor, S. C., Lindon, J. C., & Nicholson, J. K. (2002). The Comparison of Plasma Deproteinization Methods for the Detection of Low-Molecular-Weight Metabolites by 1H Nuclear Magnetic Resonance Spectroscopy. Analytical Biochemistry, 304(2), 220–230. https://doi.org/10.1006/ABIO.2002.5637
Dudzik, D., Barbas-Bernardos, C., García, A., & Barbas, C. (2018). Quality assurance procedures for mass spectrometry untargeted metabolomics. a review. Journal of Pharmaceutical and Biomedical Analysis, 147, 149–173. https://doi.org/10.1016/J.JPBA.2017.07.044
Dudzik, D., & García, A. (2021). Untargeted metabolomics methods to analyze blood-derived samples. Neuromethods, 159, 173–187. https://doi.org/10.1007/978-1-0716-0864-7_14/FIGURES/2
Eylem, C. C., Nemutlu, E., Dogan, A., Acik, V., Matyar, S., Gezercan, Y., et al. (2022). High-Throughput Single-Step plasma sample extraction optimization strategies with experimental design for LC-MS and GC–MS integrated metabolomics and lipidomics analysis. Microchemical Journal, 179, 107525. https://doi.org/10.1016/J.MICROC.2022.107525
Fda, & Cder. (2018). Bioanalytical Method Validation Guidance for Industry Biopharmaceutics Bioanalytical Method Validation Guidance for Industry Biopharmaceutics Contains Nonbinding Recommendations. http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/
default.htmand/orhttp://www.fda.gov/AnimalVeterinary/GuidanceComplianceEnforcement/GuidanceforIndustry/default.htm. Accessed 17 April 2023
Fda, & Cder. (2019). M10 BIOANALYTICAL METHOD VALIDATION. Accessed 2 August 2023
Haggarty, J., & Burgess, K. E. (2017). Recent advances in liquid and gas chromatography methodology for extending coverage of the metabolome. Current Opinion in Biotechnology, 43, 77–85. https://doi.org/10.1016/J.COPBIO.2016.09.006
Khan, I., Nam, M., Kwon, M., Seo, S. S., Jung, S., Han, J. S., et al. (2019). LC/MS-Based Polar Metabolite Profiling Identified Unique Biomarker Signatures for Cervical Cancer and Cervical Intraepithelial Neoplasia Using Global and Targeted Metabolomics. Cancers 2019, Vol. 11, Page 511, 11(4), 511. https://doi.org/10.3390/CANCERS11040511
Kirwan, J. A., Gika, H., Beger, R. D., Bearden, D., Dunn, W. B., Goodacre, R., et al. (2022). Quality assurance and quality control reporting in untargeted metabolic phenotyping: mQACC recommendations for analytical quality management. Metabolomics, 18(9), 70. https://doi.org/10.1007/S11306-022-01926-3
Ortmayr, K., Hann, S., & Koellensperger, G. (2015). Complementing reversed-phase selectivity with porous graphitized carbon to increase the metabolome coverage in an on-line two-dimensional LC-MS setup for metabolomics. Analyst, 140(10), 3465–3473. https://doi.org/10.1039/C5AN00206K
Rakusanova, S., Fiehn, O., & Cajka, T. (2023). Toward building mass spectrometry-based metabolomics and lipidomics atlases for biological and clinical research. TrAC Trends in Analytical Chemistry, 158, 116825. https://doi.org/10.1016/J.TRAC.2022.116825
Ramautar, R., & De Jong, G. J. (2014). Recent developments in liquid-phase separation techniques for metabolomics. Bioanalysis, 6(7), 1011–1026. https://doi.org/10.4155/BIO.14.51
Rey-Stolle, F., Dudzik, D., Gonzalez-Riano, C., Fernández-García, M., Alonso-Herranz, V., Rojo, D., et al. (2022). Low and high resolution gas chromatography-mass spectrometry for untargeted metabolomics: A tutorial. Analytica Chimica Acta, 1210, 339043. https://doi.org/10.1016/J.ACA.2021.339043
Roca, M., Alcoriza, M. I., Carlos Garcia-Ca, J., Naveras, ~, & Lahoz, A. (2020). Reviewing the metabolome coverage provided by LC-MS: Focus on sample preparation and chromatography-A tutorial. https://doi.org/10.1016/j.aca.2020.12.025
Sedgwick, G. W., Fenton, T. W., & Thompson, J. R. (1991). Effect of protein precipitating agents on the recovery of plasma free amino acids. https://doi.org/10.4141/cjas91-116, 71(3), 953–957. https://doi.org/10.4141/CJAS91-116
Sumner, L. W., Amberg, A., Barrett, D., Beale, M. H., Beger, R., Daykin, C. A., et al. (2007). Proposed minimum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI). Metabolomics : Official journal of the Metabolomic Society, 3(3), 211–221. https://doi.org/10.1007/S11306-007-0082-2
Szczerbinski, L., Wojciechowska, G., Olichwier, A., Taylor, M. A., Puchta, U., Konopka, P., et al. (2022). Untargeted Metabolomics Analysis of the Serum Metabolic Signature of Childhood Obesity. Nutrients, 14(1), 214. https://doi.org/10.3390/NU14010214/S1
Tsugawa, H., Cajka, T., Kind, T., Ma, Y., Higgins, B., Ikeda, K., et al. (2015). MS-DIAL: data-independent MS/MS deconvolution for comprehensive metabolome analysis. Nature methods, 12(6), 523–526. https://doi.org/10.1038/NMETH.3393
Validation of analytical methods for active constituents and agricultural products | Australian Pesticides and Veterinary Medicines Authority. (2014, July 1). https://apvma.gov.au/node/1048. Accessed 13 September 2023
Wang, L., Naser, F. J., Spalding, J. L., & Patti, G. J. (2019). A Protocol to Compare Methods for Untargeted Metabolomics. Methods in molecular biology (Clifton, N.J.), 1862, 1. https://doi.org/10.1007/978-1-4939-8769-6_1
Wang, T. J., Larson, M. G., Vasan, R. S., Cheng, S., Rhee, E. P., McCabe, E., et al. (2011). Metabolite profiles and the risk of developing diabetes. Nature Medicine 2011 17:4, 17(4), 448–453. https://doi.org/10.1038/nm.2307
Wishart, D. S., Feunang, Y. D., Marcu, A., Guo, A. C., Liang, K., Vázquez-Fresno, R., et al. (2018). HMDB 4.0: the human metabolome database for 2018. Nucleic acids research, 46(D1), D608–D617. https://doi.org/10.1093/NAR/GKX1089
Yu, Y., Tan, P., Zhuang, Z., Wang, Z., Zhu, L., Qiu, R., & Xu, H. (2021). Untargeted metabolomic approach to study the serum metabolites in women with polycystic ovary syndrome. BMC Medical Genomics, 14(1), 1–15. https://doi.org/10.1186/S12920-021-01058-Y/FIGURES/5
Zeki, Ö. C., Eylem, C. C., Reçber, T., Kır, S., & Nemutlu, E. (2020). Integration of GC–MS and LC–MS for untargeted metabolomics profiling. Journal of Pharmaceutical and Biomedical Analysis, 190, 113509. https://doi.org/10.1016/J.JPBA.2020.113509
Zhang, K., Liu, Y., Liu, L., Bai, B., Shi, L., & Zhang, Q. (2023). Untargeted Metabolomics Analysis Using UHPLC-Q-TOF/MS Reveals Metabolic Changes Associated with Hypertension in Children. Nutrients, 15(4), 836. https://doi.org/10.3390/NU15040836/S1
Zhu, Q. F., Zhang, Z., Liu, P., Zheng, S. J., Peng, K., Deng, Q. Y., et al. (2016). Analysis of liposoluble carboxylic acids metabolome in human serum by stable isotope labeling coupled with liquid chromatography–mass spectrometry. Journal of Chromatography A, 1460, 100–109. https://doi.org/10.1016/J.CHROMA.2016.07.017

No competing interests reported.

SupplementaryMaterials.docx

Download PDF

Journal Publication

published 15 Jul, 2024

Read the published version in Metabolomics →

Editorial decision: Revision requested
16 May, 2024
Reviews received at journal
16 May, 2024
Reviewers agreed at journal
22 Apr, 2024
Reviews received at journal
13 Feb, 2024
Reviewers agreed at journal
29 Jan, 2024
Reviewers agreed at journal
29 Jan, 2024
Reviewers invited by journal
21 Dec, 2023
Submission checks completed at journal
21 Dec, 2023
Editor assigned by journal
21 Dec, 2023
First submitted to journal
20 Dec, 2023

You are reading this latest preprint version

Polar metabolomics using trichloroacetic acid extraction and porous graphitic carbon stationary phase.

Status:

Journal Publication

Version 1

Abstract

Figures

Highlights

1. Introduction

2. Materials and methods

2.1 Ethics

2.2 Solvents and Chemicals

2.3 Human serum samples

2.4 Sample preparation

2.5 Liquid chromatography – Mass spectrometry (LC-MS) conditions

2.6 Data Processing

3. Results

4. Discussion

5. Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1