Article pubs.acs.org/ac
Intact Endogenous Metabolite Analysis of Mice Liver by Probe Electrospray Ionization/Triple Quadrupole Tandem Mass Spectrometry and Its Preliminary Application to in Vivo Real-Time Analysis Kei Zaitsu,*,†,‡,# Yumi Hayashi,†,§,# Tasuku Murata,∥ Tomomi Ohara,§ Kenta Nakagiri,§ Maiko Kusano,‡ Hiroki Nakajima,∥ Tamie Nakajima,⊥ Tetsuya Ishikawa,§ Hitoshi Tsuchihashi,‡ and Akira Ishii‡ †
In Vivo Real-Time Omics Laboratory, Institute for Advanced Research, Nagoya University, Furo-cho, Chikusa-ku, Nagoya, 464-8601, Japan ‡ Department of Legal Medicine & Bioethics, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya, 466-8550, Japan § Department of Radiological and Medical Laboratory Sciences, Nagoya University Graduate School of Medicine, 1-1-20 Daiko-Minami, Higashi-ku, Nagoya, 461-8673, Japan ∥ Shimadzu Corporation, 1 Nishinokyo-Kuwabaracho, Nakagyo-ku, Kyoto, 604-8511, Japan ⊥ College of Life and Health Sciences, Chubu University, 1200 Matsumoto-cho, Kasugai, Aichi 487-8501, Japan S Supporting Information *
ABSTRACT: Probe electrospray ionization (PESI) is a recently developed ionization technique that enables the direct detection of endogenous compounds like metabolites without sample preparation. In this study, we have demonstrated the first combination use of PESI with triple quadrupole tandem mass spectrometry (MS/MS), which was then applied to intact endogenous metabolite analysis of mice liver, achieving detection of 26 metabolites including amino acids, organic acids, and sugars. To investigate its practicality, metabolic profiles of control and CCl4-induced acute hepatic injury mouse model were measured by the developed method. Results showed clear separation of the two groups in score plots of principal component analysis and identified taurine as the primary contributor to group separation. The results were further validated by the established gas chromatography/MS/MS method, demonstrating the present method’s usefulness. In addition, we preliminarily applied the method to real-time analysis of an intact liver of a living mouse. We successfully achieved monitoring of the real-time changes of two tricarboxylic acid cycle intermediates, α-ketoglutaric acid and fumaric acid, in the liver immediately after pyruvic acid injection via a cannulated tube to the portal vein. The present method achieved an intact analysis of metabolites in liver without sample preparation, and it also demonstrates future possibility to establish in vivo real-time metabolome analysis of living animals by PESI/MS/MS.
M
acquisition and exclude bias from sample pretreatment.7,8 Furthermore, development of intact sample analysis method without sample preparation can become an epoch-making method for real-time analysis of living animals.9−11 Flow injection analysis (FIA) has recently been applied to metabolomics for its rapidity and high repeatability,12,13 though some sample pretreatment like protein precipitation is still necessary for FIA. To apply FIA/MS to real-time analysis of endogenous molecules, online desalting device is essential to directly connect the flow containing biological matrixes to the mass spectrometer.14−17 However, there has been a difficulty to
etabolomics has been widely extended to various scientific fields.1−3 For metabolomics analysis, sample preparation like extraction, concentration with evaporation or freeze-drying, and derivatization steps are mandatory to execute gas chromatography/mass spectrometry (GC/MS), liquid chromatography/tandem mass spectrometry (LC/MS/MS), or capillary electrophoresis/MS/MS,4−6 but these steps may potentially cause bias, which can lead to misinterpretation of the final results. To overcome this issue, the internal standard (IS) correction method is generally applied, though it is difficult to exclude bias from the sample preparation steps completely because of the large number of target compounds and the complexity of compounds’ properties treated in metabolome analysis. Therefore, analysis of intact biological tissues is one of the ultimate goals for metabolomics to obtain solid information on metabolome alteration, because it can achieve rapid data © XXXX American Chemical Society
Received: October 26, 2015 Accepted: February 23, 2016
A
DOI: 10.1021/acs.analchem.5b04046 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 1. Photograph of the probe electrospray ionization ion source combined with a triple quadrupole tandem mass spectrometer. The diameter of probe needle tip is ∼700 nm.
Table 1. Identified Metabolites, and SRM Transition and Collision Energy (CE) Used for PESI/MS/MS
a
name
polarity
SRM transition (m/z)
CE
name
polarity
3-hydroxybutyrate citric acid/isocitric acid D-glucose G6P glutaric acid glycine L-asparagine L-aspartic acid L-glutamic acid L-lactic acid L-malic acid L-serine pyruvic acid
nega neg neg neg neg neg neg neg neg neg neg neg neg
103.1 → 59.0 191.0 → 111.1 179.1 → 59.2 259.1 → 96.9 131.0 → 87.3 74.2 → 74.2 131.0 → 113.3 131.9 → 88.1 146.0 → 102.1 89.0 → 43.2 133.0 → 114.9 103.9 → 74.2 87.1 → 43.1
35 20 20 20 20 20 20 20 20 20 20 20 20
succinic acid taurine 2-aminobutyricacid L-glutamine L-histidine L-leucine/L-isoleucine L-methionine L-ornithine L-phenylalanine L-proline L-threonine L-tryptophan L-tyrosine
neg neg posb pos pos pos pos pos pos pos pos pos pos
SRM transition (m/z)
CE
→ → → → → → → → → → → → →
20 20 20 20 20 20 20 20 20 20 20 20 20
117.1 124.0 104.1 147.1 156.1 132.1 150.3 132.9 166.2 116.2 120.1 205.2 182.1
73.0 80.0 58.1 84.2 110.3 86.2 104.1 70.0 120.2 70.0 74.0 146.1 136.1
neg: ESI-negative mode. bpos: ESI-positive mode.
report on the combination use of PESI and triple quadrupole tandem mass spectrometry applied to metabolomics. In this study, we combined PESI with triple quadrupole tandem mass spectrometry (PESI/MS/MS) in order to develop an intact endogenous metabolite analysis method of mice liver samples and estimated its efficacy using carbon tetrachloride (CCl4)-induced acute hepatic injury in mice. In addition, we preliminarily tried to apply PESI/MS/MS to real-time analysis of liver in a living mouse.
develop such a device with low loss rate of target molecules like amino acids. Probe electrospray ionization (PESI) is a recent ionization technique developed by Hiraoka et al.18 This technique enables direct analysis of endogenous compounds in biological tissues like liver and heart without tedious sample preparation.11,19−22 In fact, Yoshimura et al. reported some application of PESI in combination with a time-of-flight mass spectrometer, resulting in the direct detection of some endogenous compounds like lipids.11,22 However, single mass spectrometry with electrospray ionization (ESI) has a limitation: low selectivity to compound identification due to single mass separation. In particular, it is quite difficult to identify endogenous compounds by ESI-mass spectrometry under the existence of a complicated matrix derived from biological specimens because specific product ions are generally not obtained by ESI-single mass spectrometry. Thus, combination use with tandem mass spectrometry, especially triple quadrupole tandem mass spectrometry, is crucial to improve selectivity for identifying endogenous compounds under the coexistence of matrix, especially with the wide dynamic range of concentrations of target compounds like metabolites. Moreover, PESI is a nonchromatographic technique and strongly requires specific determination of metabolites by tandem mass spectrometry, but there is no
■
EXPERIMENTAL SECTION
Reagents. CCl4, ethanol, and olive oil were purchased from Wako Pure Chemical Industries (Osaka, Japan). The stable isotope of L-glutamic acid (13C5, 15N1) was purchased from Taiyo Nippon Sanso Corporation (Tokyo, Japan). All other chemicals and reagents were purchased from Wako Pure Chemical Industries (Japan). All reagents used were of analytical grade or better quality. Intact Analysis of Liver Samples. An LCMS-8040 triple quadrupole tandem mass spectrometer with probe electrospray ion source (Shimadzu Corporation, Kyoto, Japan) was used. Mass resolution of the mass spectrometers was set to unit resolution, and the product ion scan mode m/z range was set from m/z 30 to [precursor ion’s m/z + 15]. B
DOI: 10.1021/acs.analchem.5b04046 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 2. (a) Schematics of an in vivo real-time analytical system using probe electrospray ionization. (b) Detail of the small cup (8 mm i.d.) attached to the surface of a mouse liver. The cup is a double-layered structure and a polyvinylidene chloride film (11 μm thickness) is used for the layers.
TICA Co. Ltd., Osaka, Japan) followed by attaching with aftermentioned plastic films, was mounted directly on the surface of the liver and fixed using the adhesive except at the center of the cup. The handmade cup has a double-layered structure using a polyvinylidene chloride film (11 μm thickness, AsahiKasei, Tokyo, Japan). To the cup, 150 μL of 50% aqueous ethanol solution with heparin (0.5 mg/mL) was poured just before analysis. Invasion depth of the needle tip to the liver was mechanically controlled: the length of the needle tip is 5 mm, and the height of the attached cup is 6 mm. We fixed the upperend of the cup at 6.5 mm distance from the lowest position of the needle (theoretical invasion depth is 0.5 mm). Under isoflurane anesthesia, a real-time analysis of liver by PESI/MS/ MS was executed. Pyruvic acid (10 mg) saline solution was injected via the cannulation tube to the mouse’s liver 4.5 min after the analysis started. Loop time of mass spectrometer and CPM were set at 250 ms and 0.1 Hz, respectively. Needle acceleration speed was set at 300 mm/s. Heat block and DL temperatures of the mass spectrometer were set at 30 and 250 °C, respectively. The heat block was covered with a heatresistant sheet (DEX paper, 0.7 mm thickness, Hirose Paper Mfg Co., Ltd., Japan) to prevent elevation of the mouse’s body temperature. SRM transitions and collision energies (CE) were set at m/z 115.0 → 71.0, CE 15 V for fumaric acid, and m/z 145.0 → 101.0, CE 10 V for α-ketoglutaric acid, respectively. These animal experiments were approved by the animal experiment committee of Nagoya University Graduate School of Medicine (No. 27457). Administration of CCl4 to mice, and the use of chloroform for sample pretreatment for GC/MS/MS were executed inside of a draft chamber to ensure that there was no exposure of CCl4 and chloroform to experimenters.
Details of the PESI ion source is shown in Figure 1. Both ESI positive and negative modes were used for analysis. Selected reaction monitoring (SRM) mode was used for relative comparison of metabolite levels. Polarities, SRM transitions, and collision energies were optimized using authentic standards as listed in Table 1. The PESI needle (tip diameter, 700 nm) was manufactured by Shimadzu Corporation (Kyoto, Japan). The loop times of data acquisition of the mass spectrometer were set at 102 ms for ESI positive and 220 ms for ESI negative. The cycle time of probe movement (CPM) was set at 4.55 Hz. Additionally, the number of maximum SRM transitions per run was set at 6 for ESI positive and 10 for ESI negative, and the data acquisition time was set at 0.25 min in order to obtain sufficient data points. Heat block and desolvation line (DL) temperatures of the mass spectrometer were set at 50 and 300 °C, respectively. Metabolite data were obtained by averaging the results from triplicate analyses for each sample. The probe needle was changed between sample analyses to prevent contamination. The ∼3 mm-square frozen mouse liver sample was set on a sample plate (Shimadzu Corporation), and 30 μL of 50% ethanol solution with or without the stable isotope L-glutamic acid serving as IS was added. The details of the animal experiments are described in the Supporting Information. Multivariate analysis was executed using SIMCA-P+ software (version 13, Umetrics, Umeå, Sweden) and Welch’s t-test was applied to the metabolites that contributed to group separation on score plots of principal component analysis (PCA) (significance level, 0.05). A GCMS-TQ8040 gas chromatograph-triple quadrupole tandem mass spectrometer (Shimadzu Corporation) was used for validation of the results obtained from the developed method.23,24 Analytical conditions and sample preparation details are provided in the Supporting Information. Pearson correlation coefficient was calculated using the R software package.25 Real-Time Analysis of Liver in a Living Mouse. The schematic of real-time analysis is shown in Figure 2a. A male 10-week 129/sv mouse (bred in our laboratory) was anesthetized by isoflurane for 5 min and laparotomy was performed.26 A 25-gauge infusion needle (Terumo, Tokyo, Japan) was cannulated into the hepatic portal vein and fixed using a medical adhesive for biotissues (Aron Alpha, Daiichi Sankyo Company, Tokyo, Japan). After exposure of the liver surface, a handmade small plastic cup (8 mm i.d., Figure 2b), which was made by cutting a polypropylene tube (INAOP-
■
RESULTS AND DISCUSSION
Analysis of Intact Liver by PESI/MS/MS. As shown in Table 1, 26 metabolites including amino acids, organic acids, and sugars were detected in mice liver directly by PESI/MS/ MS. As described above, we mainly targeted toward low-mass molecules in this study. We did not target high-mass molecules like phosphatidylcholines (PCs) and triacylglycerides (TAG), as they have many isomers and it was thought to be difficult to differentiate them by a nonchromatographic technique like PESI/MS/MS. As explained above, PESI requires up-and-down movement of the probe. Thus, the sum of the loop time of each transition and cycle time of the probe movement needed to be synchronized to avoid missing of data acquisition. The optimized values are described in the Experimental Section. C
DOI: 10.1021/acs.analchem.5b04046 Anal. Chem. XXXX, XXX, XXX−XXX
Article
Analytical Chemistry
Figure 3. Results of principal component analysis (PCA) applying the metabolite data obtained by PESI/MS/MS. (a) Score plots, (b) loading plots, and (c) box-whisker plot of the hepatic levels of taurine (***p < 0.001). The ellipse in the PCA score plots shows Hotelling’s T2 (
