Development and validation of a hydrophilic interaction liquid chromatography–tandem mass spectrometry method for the quantification of lipid-related extracellular metabolites in Saccharomyces cerevisiae
Highlights
► A new analytical method to determine extracellular levels of lipid-related metabolites. ► Five metabolites are separated in 20 min with lower limit of quantification at 0.5 nM. ► The developed method is successfully applied to monitor the extracellular levels of metabolites in S. cerevisiae.
Introduction
The metabolome represents the full collection of low-molecular weight chemical species within a cell or biological system, and is considered the endpoint of “omics analysis” [1], [2], [3]. Current research on metabolomic investigation consists of four complementary approaches: target analysis, metabolic profiling, metabolic fingerprinting, and metabolic footprinting [4], [5], [6]. All of these approaches are usually applied to investigate differences in metabolite concentrations after alterations in the biological environment or upon genetic modification. The measurement of extracellular metabolites secreted from the intracellular volume into the growth medium is termed “metabolic footprinting”. Footprinting analysis offers important technical advantages over the analysis of intracellular compounds, referred to as “metabolic fingerprinting” [7], [8], [9]. First of all, the extracellualar metabolome is generally quite stable owing to the relative lack of enzymes that can convert the metabolites into other products. Therefore, the time-consuming quenching steps associated with the analysis of intracellular metabolites are not required when extracellular metabolites are analyzed. In addition, the extracellular metabolome is generally simplified in terms of the number of metabolites present and their concentration ranges, as compared to the intracellular metabolome [7]. Metabolic footprinting has been used in the classification of microbial mutants for functional genomics studies by employing mass spectrometry (MS) [3], [10], [11], [12] or nuclear magnetic resonance (NMR) analysis [13]. For yeast and mammalian cells, metabolic footprinting has been performed under a limited number of conditions, and has primarily focused on central carbon metabolites [14], [15], [16], [17], [18]. To our knowledge, no metabolic footprinting study has focused on lipid metabolites.
The simple eukaryote, Saccharomyces cerevisiae, has been used as a model to study many aspects of cell biology, including lipid metabolism [19], [20], [21], [22], [23]. We have chosen to analyze five major phospholipid metabolites in the media of S. cerevisiae as a means of monitoring aspects of phospholipid synthesis and turnover in the organism: inositol, choline, glycerol 3-phosphate, glycerophosphoinositol, and glycerophosphocholine. The lipid precursors inositol and choline regulate phospholipid biosynthesis at the transcriptional level in S. cerevisiae [24]. Thus, they are frequently added to and removed from medium to study and manipulate the biosynthetic pathways. Inositol and choline can also be produced by the cell via lipid turnover events. GroP is a metabolite that can be produced through phospholipid turnover and is also a precursor involved in phospholipid biosynthesis. Finally, extracellular GroPIns and GroPCho are produced through the hydrolytic cleavage of both acylester bonds of plasma membrane-associated glycerophospholipids by phospholipase B type enzymes [20], [21], [25]. Thus, extracellular GroPIns and GroPCho can be monitored as an indicator of phospholipase B activity encoded by the PLB1, PLB2, and PLB3 genes. Overall, the concentration of each of these metabolites in the medium is the result of two processes: their production via enzymatic activity and their uptake via plasma membrane transport. By knowing the extracellular levels of these metabolites, we will gain insight into the status of the phospholipid biosynthetic and catabolic pathways in S. cerevisiae.
This method was developed in order to create a simplified and sensitive method for the absolute and simultaneous quantitation of the levels of important phospholipid metabolites in the media of S. cerevisiae. Previous studies in which one or more of the compounds has been monitored usually involve the use of radioactivity, as the compounds have no UV or fluorescence detectable groups. For example, 3H or 14C labeled inositol or choline, and 32P-H3PO4 are typically utilized [26], [27], [28]. However, measuring all of these compounds simultaneously with multiple radioactive compounds is technically challenging, even if a good separation system is at hand. In addition, the use of radioactive molecules can also present a quantitation problem. For example, since S. cerevisiae can both synthesize and import inositol and choline, but radioactive labeling only allows for the discrimination of inositol and choline transported into the cell, not that derived from de novo synthesis. In contrast, MS detects all chemical species, regardless of their mode of synthesis. Finally, the method described here is an advance in terms of the LC separation of the compounds. While published procedure has described the chromatographic separation of choline-containing metabolites, and inositol-containing metabolites, we know of no published procedure for the LC separation of all five of these compounds simultaneously [29], [30], [31].
Two published studies are pertinent to the method described here. A LC–MS/MS method using a β-cyclodextrin-bonded column was described for the quantitative analysis of only internal GroPIns in rat cell lines, but other metabolites were not included in this method [32]. In contrast, our method analyzes five lipid-related metabolites, including GroPIns, in the extracellular medium of S. cerevisiae. Another study used normal phase LC–MS to quantitate several water-soluble metabolites, including GroPIns and GroPCho, from rat brain tissue [33]. Our method differs from that of Kopp et al. [33] and Dragani et al. [32] in that we perform extensive method validation, we analyze an overlapping but different set of metabolites, and we use HILIC chromatography, which results in decreased retention times as compared to normal phase.
Here, we present a hydrophilic interaction liquid chromatography–tandem mass spectrometry (HILIC–MS/MS) method for the quantification of five lipid-related extracellular metabolites in S. cerevisiae cells. A liquid–liquid extraction procedure has been combined with an extensive optimization of HILIC–MS/MS methodology to provide effective and reliable chromatographic separation of the analytes. The method is highly sensitive and has been thoroughly validated according to the bioanalytical method validation guidelines for industry as specified by the US Food and Drug Administration (FDA) (www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm070107.pdf).
Section snippets
Materials and reagents
Standard compounds including glycerophosphoinositol (l-α-glycerophospho-d-myo-inositol), glycerophosphocholine (l-α-glycerophosphorylcholine), glycerol 3-phosphate (l-α-glycero-phosphate), myo-inositol, and choline chloride were purchased from Sigma–Aldrich (St. Louis, MO, USA). The deuterated internal standard (IS), choline chloride-d9, was obtained from Cambridge Isotopes (Andover, MA, USA). Ammonium acetate was purchased from Fisher Scientific (Pittsburgh, PA, USA). Optima grade methanol and
Optimization of MS/MS conditions
The standards of the analytes (Fig. 1) were first characterized by total ion scan and product ion scan through direct infusion to ascertain their precursor ions and to select product ions for use in MRM mode. The MS/MS parameters, specifically fragmentor voltage and collision energy, were evaluated for the best response of the parent ion and daughter ion, respectively, using the automatic optimization process and then transferred to the MRM method. GroPCho, GroP, and choline were found to give
Conclusions
We have successfully developed and validated an efficient analytical method based on HILIC–MS/MS for the quantification of lipid-related extracellular metabolites in S. cerevisiae. In comparison with traditional radiolabeling methods and other LC–MS/MS methods, our approach is simple, robust, and well suited to metabolomic studies.
Acknowledgement
Mass spectrometry instrumentation was purchased in part from National Science Foundation (Grant MRI DBI-0821401).
References (46)
- et al.
Trends Biotechnol.
(1998) - et al.
Trends Plant Sci.
(2004) - et al.
Anal. Biochem.
(2006) - et al.
TrAC Trends Anal. Chem.
(2008) TrAC Trends Anal. Chem.
(2008)- et al.
Trends Biotechnol.
(2004) J. Lipid Res.
(2008)- et al.
J. Biol. Chem.
(1994) - et al.
J. Biol. Chem.
(1999) - et al.
J. Biol. Chem.
(2007)
J. Biol. Chem.
J. Biol. Chem.
J. Biol. Chem.
J. Chromatogr. B
Chem. Biol.
J. Chromatogr. A
J. Chromatogr. B
Anal. Biochem.
Mol. Cell
Nat. Rev. Microbiol.
Biotechnol. Lett.
Mass Spectrom. Rev.
Anal. Bioanal. Chem.
Cited by (15)
High-throughput LC-MS quantitation of cell culture metabolites
2019, BiologicalsCitation Excerpt :Full metabolite profiling remains a challenge due to the highly diverse physiochemical properties and wide concentration ranges (up to five orders of magnitude) of metabolites in the cell culture [19,20,24]. Various analytical methods have been utilized to identify and quantify different groups of metabolites, including gas or liquid chromatography (GC or LC) with fluorescence or UV detection [17,20,26,27], mass spectrometry (MS) [23,27–33], and nuclear magnetic resonance spectroscopy (NMR) [34,35]. In this study, we developed a LC-MS method to simultaneously analyze key metabolites in the core metabolic pathways that are essential for energy homeostasis and macromolecular synthesis [3–8,36,37], including amino acids [8,12,38–40], vitamins [8,36,41], glucose [39,42–44], and nucleic acid related compounds [30,45–47].
Loss of ypk1, the yeast homolog to the human serum- And glucocorticoid-induced protein kinase, accelerates phospholipase b1-mediated phosphatidylcholine deacylation
2014, Journal of Biological ChemistryCitation Excerpt :The cells were pelleted by centrifugation at 1,600 × g for 3 min, and the supernatant containing the extracellular metabolites was filtered through 0.2-µm cellulose acetate filters. The method of analysis for these extracellular metabolites by mass spectrometry was described previously (45, 46). A 250-µl aliquot of supernatant was added to an internal standard (choline-d9), and water-soluble lipid metabolites were extracted by the addition of 2.25 ml of chloroform/methanol (2:1, v/v).
Quantitative metabolic profiling of grape, apple and raspberry volatile compounds (VOCs) using a GC/MS/MS method
2014, Journal of Chromatography B: Analytical Technologies in the Biomedical and Life SciencesCitation Excerpt :The intraday precision of the retention time was 0.06% for apples, 0.03% for raspberries and 0.02% for grapes, while interday precision was 0.1%, for apples, 0.05% for raspberries and 0.1% for grapes. All precision parameters were within the acceptance criteria, which is ±15%, as recommended by FDA bioanalitical method validation guidelines [31]. In order to show the feasibility of using the method for the profiling of different fruits, the analysis was performed on 5 different varieties of apples (Crismon Crisp, Golden Delicious, Fuji, Gala and Pink Lady), raspberries (Himbo Top, Rubyfall, Glen Ample, Autumn Treasure and Sugana) and grapes (Sangiovese, Grüner Veltliner, Sauvignon Blanc, Chardonnay and Gewürztraminer), chosen for the study in order to evaluate the differences in the profile of volatile compounds within a given fruit and between different fruits (Tables 1–3).
Method validation strategies involved in non-targeted metabolomics
2014, Journal of Chromatography ACitation Excerpt :Though this validated method is not enough for validating a final biomarker for clinical use, which requires a target method with a proper, classical method validation, but results are more reliable compared to non-validated methods at least for further data processing. There are very few papers that used these criteria not only to check method reproducibility but to successfully validated and applied on bio-fluids or tissues [36,72–76]. In disease screening, non-targeted metabolomics results are ended-up with a prediction of class belonging, using multivariate statistical algorithm, usually first applying non-supervised principal component analysis (PCA).
Fast separation and determination of free myo-inositol by hydrophilic liquid chromatography
2014, Carbohydrate ResearchCitation Excerpt :HPLC methods with ELSD have been used for determination of d-chiro-inositol15 (LOD was 0.1 μg) or several carbohydrates16 (LOD was 8.3 mg/L). Sample derivatization has been applied for inositol determination by HPLC17 in blood (LOD was 0.2 mg/L for plasma) with UV-detection or by GC–MS.18,19 MS detection has been employed in a UHPLC method20 (LOD was 0.2 mg/kg) or HILIC21,22 (LOD were 27 ng or 5 μg/L). Clearly, above mentioned methods hyphenated with MS exhibit lower limits of detection.
A rough guide to metabolite identification using high resolution liquid chromatography mass spectrometry in metabolomic profiling in metazoans
2013, Computational and Structural Biotechnology JournalCitation Excerpt :If the goal of a metabolomics study is to monitor well characterised metabolites for instance in disease monitoring, then low resolution mass spectrometry might be more appropriate since the target compounds are known and tandem mass spectrometry is a method which is highly standardised with regard to being able to obtain quantitative accuracy. Some research groups have set up more specific biomarker screens based on low resolution tandem mass spectrometry where a large number of standard compounds have been used to standardise the methods [8–13]. However, for many research groups biomarker discovery remains a more general screening operation.
- 1
This author is deceased.