Metabolic Profiling of Bile Acids in Human and Mouse Blood by LC

Subscriber access provided by CMU Libraries - http://library.cmich.edu

Article

Metabolic profiling of bile acids in human and mouse blood by LC-MS/ MS in combination with phospholipid-depletion solid-phase extraction Jun Han, Yang Liu, Renxue Wang, Juncong Yang, Victor Ling, and Christoph H. Borchers Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/ac503816u • Publication Date (Web): 12 Dec 2014 Downloaded from http://pubs.acs.org on December 23, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Metabolic profiling of bile acids in human and mouse blood by LC-MS/MS in combination with phospholipid-depletion solid-phase extraction

Jun Han†, Yang Liu†, Renxue Wang‡, Juncong Yang†, Victor Ling‡, and Christoph H. Borchers*†§

†University of Victoria - Genome BC Proteomics Centre, University of Victoria, Vancouver Island Technology Park, 3101−4464 Markham Street, Victoria, British Columbia V8Z 7X8, Canada ‡BC Cancer Agency, 600 West 10th Avenue, Vancouver, Canada, V5Z 4E6 §Department of Biochemistry and Microbiology, University of Victoria, Petch Building Room 207, 3800 Finnerty Road, Victoria, British Columbia V8P 5C2, Canada

(*) Correspondence should be addressed to: Dr. Christoph H. Borchers, University of Victoria - Genome British Columbia Proteomics Centre 3101-4464 Markham St, Victoria, BC Canada V8Z 7X8 Phone (250)483-3221; Fax: (205)483-3238 Email: [email protected]

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 31

Abstract To obtain a more comprehensive profile of bile acids (BAs) in blood, we developed an ultrahigh performance liquid chromatography/multiple-reaction monitoring-mass spectrometry (UPLC/MRM-MS) method for the separation and detection of 50 known BAs. This method utilizes phospholipid-depletion solid-phase extraction as a new high-efficiency sample preparation procedure for BA assay. UPLC/scheduled MRM-MS with negative ion electrospray ionization enabled targeted quantitation of 43 and 44 BAs, respectively, in serum samples from seven individuals with and without fasting, as well as in plasma samples from six cholestatic gene knockout mice and six age- and gender-matched wild-type (FVB/NJ) animals. Many minor BAs were identified and quantitated in the blood for the first time. Method validation indicated good quantitation precision with intra-day and inter-day relative standard deviations of ≤ 9.3% and ≤10.8%, respectively. Using a pooled human serum sample and a pooled mouse plasma sample as the two representative test samples, the quantitation accuracy was measured to be 80% to 120% for most of the BAs, using two standard-substance spiking approaches. To profile other potential BAs not included in the 50 known targets from the knockout versus wild-type mouse plasma, class-specific precursor/fragment ion transitions were used to perform UPLC/MRM-MS for untargeted detection of the structural isomers of glycine- and taurine-conjugated BAs, and unconjugated tetra-hydroxy BAs. As a result, as many as 36 such compounds were detected. In summary, this UPLC/MRM-MS method has enabled the quantitation of the largest number of BAs in the blood thus far, and the results presented have revealed an unexpectedly complex BA profile in mouse plasma.

2


Page 3 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Bile acids (BAs) are steroidal C24 carboxylic acids derived from cholesterol metabolism in hepatocytes. BA synthesis involves two major pathways (i.e., the classic neutral pathway and the alternative acidic pathway), multiple steps and at least 16 enzymes.1,2 BA synthesis produces primary cholic acid (CA) and chenodeoxycholic acid (CDCA). Within hepatocytes, BAs are conjugated with taurine or glycine as well as metabolized by hydroxylation, sulfation, and glucuronidation. The BAs and their conjugates are secreted from the hepatocytes into the canalicular space and are stored in the gallbladder.3,4 Bile and BAs are released into the small intestine after eating to aid digestion. In the intestines, CA and CDCA are dehydroxylated by intestinal bacteria to form lipophilic secondary deoxycholic acid (DCA) and lithocholic acid (LCA). Most of these compounds are reabsorbed in the distal ileum and return to the liver via enterohepatic circulation. The BAs that escape the enterohepatic circulation pass into the colon where they undergo further bacterial metabolism such as deconjugation, oxidation-reduction, epimerization, 7-dehydroxylation and esterification before their excretion in feces.5 Because of these complex synthetic and metabolic processes, at least several dozen BAs have thus far been identified in humans, rodents and other animals.6,7 BAs facilitate the absorption of lipids, cholesterol, and fat-soluble vitamins in the intestinal tract. Studies have also demonstrated that BAs have many other important physiologic functions including regulation of their own synthesis, endocrine, paracrine functions, and energy expenditure.8 BAs act as the signaling molecules in controlling the metabolism of glucose and lipids in the enterohepatic system and energy homeostasis in the peripheral tissues.9-11 At high physiological concentrations, lipophilic BAs cause oxidative and nitrosative stress, DNA damage, apoptosis, and mutation, and can lead to genomic instability, development of apoptosis resistance, 3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 31

and cancer.12,13 BAs have been found to be etiologic agents of cancers of the esophagus,14 stomach,15 and small intestine.16,17 Because of the diverse functions and roles of BAs, a more comprehensive determination of BAs in different biological samples will be important for the diagnostic, therapeutic, and prognostic management of the associated diseases. BAs were first measured by gas chromatography (GC) with related detection techniques.18-22 Due to the technical limitations of GC for the analysis of polar compounds such as BA conjugates, liquid chromatography-tandem mass spectrometry (LC-MS/MS) now dominates the analysis of BAs in biological samples.23-28 Thus far up to 31 BAs in human and rodent blood and liver29 and up to 32 of BAs and their sulfates in mouse liver, plasma, bile, and urine30 have been quantitated by LC-MS/MS using multiple-reaction monitoring (MRM). With the accumulation of new evidence for the involvements of BAs in more and more human physiological and pathological conditions, there has been a renewed interest in improved measurement of various BAs in different biological samples. Thus, the purpose of this work was to develop an ultrahigh-performance (UP) LC/MRM-MS method for the more comprehensive analysis of BAs in human and mouse blood, in combination with phospholipid depletion as an improved and efficient sample preparation procedure. EXPERIMENTAL Materials and Methods. LC/MS grade acetonitrile, methanol, water, formic acid, and HybridSPE®-Phospholipid 96-well plates (50 mg/2 mL) were obtained from Sigma-Aldrich (St Louis, MO). Oasis HLB cartridges (30 mg/1 mL) were purchased from Waters (Billerica, MA). The authentic compounds of 27 unconjugated, 8 glycine conjugated, and 11 taurine conjugated BAs were ordered from either Steraloids (Newport, RI), Santa Cruz Biotechnologies (Santa Cruz, CA), or Toronto Research Chemicals (Toronto, Ontario, Canada). Fourteen deuterium (D4 or 4


Page 5 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


D6)-labeled BAs were purchased from Steraloids, Santa Cruz Biotechnologies, Toronto Research Chemicals, or C/D/N Isotopes (Pointe-Claire, Quebec, Canada), and were used as isotopelabeled internal standard (IS) for quantitation. The details on these authentic compounds are provided in Supporting Information Table S1. In addition, 4 tetra-hydroxy bile acids (THBAs), i.e., 3α,6α,7α,12α-THBAs, 3α,6β,7α,12α-THBAs, 2α,3α,7α,12α-THBAs, and tauro3α,6α,7α,12α-THBA, were custom-synthesized and provided by Victor Ling’s group at British Columbia Cancer Agency (BCCA), Vancouver, BC, Canada. Human Serum. Five human serum samples collected from five healthy donators were obtained from Bioreclamation LLC (Westbury, NY), pooled at an equal volume, and used for human blood assay development. Seven pairs of human fasting versus non-fasting serum samples provided by Cathie Garnis at the University of British Columbia (UBC) were drawn and prepared from seven healthy volunteers one hour after eating a fatty meal and again, three weeks later, after fasting overnight during a previous study.31 The sera were prepared and were stored at -80 oC until used. The use of the human samples in this work was approved by the Ethics Committee of the University of Victoria. Mouse Plasma.

Equal (2 mL) volumes of five mouse plasma samples, obtained from

Bioreclamation LLC, were pooled and used for mouse blood assay development.

Using

approved protocols of the UBC’s Committee on Animal Care according to the guidelines of the Canadian Council on Animal Care, twelve plasma samples of six eight-week old mdr2-/knockout (KO) female mice and six age- and gender-matched wild-type (WT) FVB/NJ mice were used for the detection and quantitation of the systemic circulation pool of BAs in the two strains.

5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 31

The mdr2 gene encodes the liver-specific multidrug resistance P-glycoprotein, Mdr2, responsible for transport of phospholipids from the hepatocyte across the canalicular membrane into the bile.32 Secretion of phospholipids into the bile is necessary to neutralize BA toxicity. The “naked” toxic BAs secreted in these mice cause inflammation and fibrosis in cholangiocytes and bile ducts, and eventually lead to hepatocellular carcinoma (HCC) at age of 16 months.33 Therefore, mdr2-/- mice are an inflammation-associated animal model that resembles human HCCs in many aspects.34

The six mdr2-/- mice in the current analysis displayed evident

cholangitis and cholestasis as demonstrated by their liver indicator profiles. Standard Solutions and Calibration Curves. Stock solutions (1 mg/mL each) of the 50 BAs were individually prepared from their authentic compounds in methanol for the unconjugated BAs and in 50% methanol for the glycine and taurine conjugates. These stock solutions were further diluted with 50% methanol to give final concentrations of 0.002 to 20 µg/mL. These standard solutions were used to determine the lower limits of detection (LLODs) and the lower limits of quantitation (LLOQs). A mixed-standard solution containing 100 ng/mL of each of the 14 D4- or D6-labeled BAs was prepared in 50% methanol and was used as the IS solution. For preparation of the calibration curves, each working standard solution was mixed with an equal volume of the IS solution. Sample Preparation. Four different sample preparation approaches were evaluated: phospholipid-depletion SPE (PD-SPE) using a Sigma-Aldrich HybridSPE®-Phospholipid 96well plate,35 protein precipitation (PPT), reversed-phase SPE, and reversed-phase SPE with high pH (hi-pH SPE) sample loading. The details of the PPT, reversed-phase SPE, and hi-pH SPE methods for sample preparation are described in Supporting Information.

6


Page 7 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Phospholipid-Depletion SPE (PD-SPE). Fifty-µL aliquots of the IS solution were placed into individual 1.5-mL Eppendorf tubes and then dried in a SPD-1010 vacuum concentrator (Thermo Scientific Inc., Waltham, MA). To each tube, a 50-µL aliquot of human serum or mouse plasma and 100 µL of 1% formic acid in water were added. The tubes were vortexed and 350 µL of acetonitrile were added. The tubes were sonicated for 30 s in a 0 oC water bath and then centrifuged at 16,000 x g and 4 oC for 10 min in a micro-centrifuge. The entire supernatant was loaded into a single well of a HybridSPE®-Phospholipid 96-well plate (each well had been activated with 2 mL of methanol and pre-conditioned with 2 mL of 70% aqueous acetonitrile containing 0.2% formic acid prior to use). After sample loading, the flow-through fractions were collected into 2-mL wells of a 96-well collection plate under a 4-psi positive pressure. Each well was then washed with 0.5 mL of 70% aqueous acetonitrile containing 0.2% formic acid. The flow-through fractions were collected, pooled, and dried in the vacuum concentrator. The residues were reconstituted in 100 µL of 50% methanol. Ten µL were injected for UPLC/MRMMS. To measure those BAs whose endogenous concentrations were higher than the upper LOQs, each solution were diluted 1:5 (v/v) with a mixture of the IS solution and 50% methanol (1:1, v/v) for the human serum samples or diluted 1:50 (v:v) with the same mixture for the mouse plasma samples before re-injection. For comparison of PD-SPE with the following three sample preparation approaches, no IS was spiked in.

UPLC/MRM-MS. An Ultimate 3000 RSLC system (Dionex Inc., Amsterdam, the Netherlands) coupled to a 4000 QTRAP mass spectrometer (AB Sciex, Concord, ON, Canada) via a Turbo Ionspray electrospray ionization (ESI) source was operated in the negative ion mode. A BEH C18 (2.1 Х150 mm, 1.7 µm) UPLC column (Waters Inc. Milford, MA) was used for the gradient 7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 31

elution, with 0.01% formic acid in water (solvent A) and 0.01% formic acid in acetonitrile (solvent B) as the mobile phase. The gradient was optimized at 25% to 40% B in 12 min and then 40% to 75% B in 14 min. The column was washed with 100% B for 2 min and equilibrated with 25% B for 4 min between injections. The flow rate was 0.35 mL/min and the column was maintained at 45 oC. Optimization of the MRM transitions of the 50 BAs and the scheduled MRM operation parameters are described in the Supporting Information and these transitions are listed in Supporting Information Table S2. To detect other potential BAs in the mouse WT versus mdr2-/plasma samples, UPLC/MS/MS with unscheduled MRM scans was used. To do this, three Q1 to Q3 ion transitions common to each class of unconjugated THBAs, glycine and taurineconjugated BAs were included. A 20-ms dwell time for each transition of all the conjugated BA groups and a 60-ms dwell time for each transition of the unconjugated THBAs and tauro-THBAs were used. The collision energy for each group-specific MRM transition was the median of the collision energies for the same transition for all the isomeric BAs in each group. Evaluation of phospholipid depletion efficiency was performed by UPLC/in-source collisioninduced dissociation (CID)-MS/MS on the same UPLC-4000 QTRAP MS system. The in-source CID conditions were optimized to be a declustering potential of 150 volts using an MRM transition of m/z 153 (Q1) to m/z 153 (Q3) -- a specific reporter ion for monitoring phospholipids except for sphingomyelin in the negative ion mode36 -- and a low collision energy of 10 volts. Complementary confirmation of the potentially unknown BAs detected by UPLC/MRM-MS was carried out by UPLC/Fourier transform mass spectrometry (FTMS) and UPLC/MS/MS using

8


Page 9 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


CID on a Thermo LTQ-Orbitrap Fusion mass spectrometer. The detailed procedure is described in the Supporting Information.

RESULTS AND DISCUSSION Optimization of UPLC and MRM. To more comprehensively analyze BAs in human serum and mouse plasma, 50 BAs -- the largest number involved in an MRM-based assay thus far -were selected as the targets for assay development. These included 3 unconjugated THBAs, 1 tauro-THBA, 12 oxo and keto BAs, and the unconjugated and conjugated allocholic acids, most of which were used in an LC/MRM-MS assay for the first time. The 50 BAs are listed in Figure 1 and their optimized MRM transitions are summarized in Supporting Information Table S2. For quantitation of the 30 unconjugated BAs, their quasi-molecular (M-H)- ions, instead of precursor ion to product ion transitions, were used for both the Q1 and Q3 ion isolations. This scan mode, known as pseudo-MRM, was used because the CID fragment ions of these unconjugated BAs showed low ion intensities which were insufficient for quantitation.29 For example, the (M-H)-/(M-H)- pairs for CA and DCA showed about 10 and 40 times of the ion intensities, respectively, as compared to the MRM transitions resulting from their (M-H2O-H)ions, while LCA, a lipophilic mono-OH BA, did not produce any CID fragment ion that could be detectable by UPLC/MRM-MS in this work. Of the 50 targets involved in the assay development, 42 BAs have structural isomers, and most of the structural isomers, e.g., tauro-α-muricholic acid (tauro-α-MCA), tauro-β-MCA, tauro-γMCA, tauro-ω-MCA, tauro-hyocholic acid (also known as tauro-λ- or γ- MCA), and tauro-CA, shared the same MRM transitions as listed in Supporting Information Table S2. To quantitate all

9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 31

the isomeric BAs, several 15-cm long reversed-phase UPLC columns were tested and compared, and the BEH C18 UPLC column gave the best separation of the different BA isomers when 0.01% formic acid in water (solvent A) and 0.01% formic acid in acetonitrile (solvent B) were used as the mobile phase solvents for binary gradient elution. Under the optimized liquid chromatographic and MRM conditions, as described in the “UPLC/MRM-MS” section, baseline separation of the different structural isomers was achieved, as shown in Figure 1, except tauroursodeoxycholic acid (compound 38) and taurohyodeoxycholic acid (compound 39) which were only partially resolved, with a chromatographic separation factor (Rs) of ca. 1.2. Figures 1A and 1B show the UPLC/MRM-MS chromatograms of the 30 unconjugated BAs and the 20 conjugated BAs, respectively, and the corresponding 14 D-labeled ISs. Sample Preparation. PD-SPE was tested as a new sample preparation approach in this work to provide quantitative analysis of the detectable BAs in human serum and mouse plasma. Blood samples have a high content of phospholipids.37 The removal of phospholipids by PD-SPE, based on a mechanism of the Lewis acid – base interactions,38 helps reduce matrix effects in LC/ESI-MS analysis.39 Using aqueous acetonitrile as the solvents, different concentrations of formic acid in the solvents were tested for sample loading and analyte elution. The measured peak areas of various BAs detected in human serum and mouse plasma showed that the use of 50% to 80% aqueous acetonitrile containing 0.1% to 0.5% formic acid as the solvents did not result in statistical differences in the chromatographic peak areas for all of the detected BAs. For consistency in sample preparation, 70% acetonitrile containing 0.2% formic acid was selected as the solvent for PD-SPE for all the subsequent experiments. Three common procedures including PPT,26,29 reversed-phase SPE,40 and reversed-phase SPE with high pH (hi-pH SPE) for sample loading,41 which have been used for the preparation of

10


Page 11 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


blood, tissue, and fecal samples for the LC/MRM-MS quantitation of BAs, were compared with PD-SPE. Supporting Information Figure S1 shows bar graphs of the measured peak areas plus the standard deviations (SDs) of nine representative BAs detected in human serum, three from each class of unconjugated, glycine-conjugated, and taurine-conjugated BAs, using 50-µL aliquots of the human serum test sample in triplicate for sample preparation. As shown, PD-SPE resulted in not only the highest peak areas but also the lowest SDs for seven of the nine BAs among these four approaches (p value ≤ 0.01, T-test), except for two conjugated BAs, i.e., glycodeoxycholic acid and taurohyodeoxycholic acid, which displayed similar peak areas and SDs with PD-SPE and reversed-phase SPE (p value > 0.01, T-test). This comparison also revealed that PD-SPE had a higher reproducibility than other three sample preparation approaches. To determine the extent of phospholipid removal by PD-SPE, UPLC/insource CID/MRM-MS was used to monitor the levels of the residual phospholipids in the samples after sample preparation with PD-SPE. Supporting Information Figure S2 shows the representative ion current chromatograms using the pseudo-MRM transition of m/z 153 (Q1) to m/z 153 (Q3) for monitoring phospholipids in the negative ion mode from two samples, one prepared by PD-SPE and the other by PPT. This figure indicates that the phospholipids in the human serum sample were completely depleted by PD-SPE, as compared to the phospholipid depletion-free PPT approach. The successful removal of phospholipids from serum samples by PD-SPE improved the detection of BAs, particularly for those lipophilic species with their retention times between 20 and 26 min. Sensitivity and Linearity. The signal-to-noise (S/N) ratios measured from a series of diluted standard solutions of the 50 BAs were used to estimate the lower limit of detection (LLOD) for each BA, defined as the lowest on-column concentration that would yield at least 3 times the 11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 31

S/N-ratio. The 14 D-labeled BAs were used as the ISs for their corresponding non-D-labeled forms. For BAs without D-labeled analogues, LCD-D4 was used as the common IS in this work to compensate for analytical variations due to sample preparation. The lower limit of quantitation (LLOQ) for each BA was defined as the lowest concentration, within the linear range (R2 ≥ 0.998) of their respective calibration curves, which generated at least five times the S/N-ratio and gave the relative standard deviations (RSDs) of ≤ 20% (n=5). Supporting Information

Table S3 lists the measured LLODs and LLOQs for all the 50 BAs, with observed analytical sensitivities in a range of sub-nanomolar concentrations and with a linear concentration range from 256 to 4096 fold for different BAs. Assay Precision and Accuracy. The precision of quantitation was determined as the intra-day and inter-day RSDs for assay of the BAs that were quantifiable in the pooled human serum and mouse test samples. Table 1 and Supporting Information Table S4 list the intra-day and interday % RSDs for 33 and 34 of the 50 targeted BAs with their determined concentrations greater than the LLOQs in the two test samples. The intra-day RSDs were determined to be 1.0% to 9.3% and 1.3% to 8.4%, respectively, for the analyses of six 50-µL aliquots of the same human test sample and six 50-µL aliquots of the same mouse test sample, which were prepared and analyzed every four hours within the same day. The inter-day RSDs were 2.2% to 10.8% and 1.4% to 8.6%, respectively, for the analyses of 50-µL aliquots of the human test sample and 50µL aliquots of the mouse test sample on different days for six continuous days. These measured RSDs indicated good quantitation precision for all of these quantifiable BAs in the two samples. To measure the accuracy of the UPLC/MRM-MS method for the quantitation of all the 50 BAs, standard substance spiking tests were performed and two spiking strategies were employed. For the 33 and 34 BAs that were quantifiable in the two test samples, two levels of the standard

12


Page 13 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


substances at 100% and 250% of the measured endogenous concentrations were spiked into different 50-µL aliquots of the human serum and mouse plasma test samples (six replicates per spiking level), followed by sample preparation with PD-SPE and quantitation with UPLC/MRMMS. The recovery was calculated as % of the measured amount accounting for the spike-in amount. Table 1 and Supporting Information Tables S4 list the measured recoveries for the 33 and 34 BAs, respectively, which were quantified in the two test samples. Twenty-nine of the 33 BAs quantified in the human test sample showed recoveries in a range of 80.7% to 113.9% at the 100% spiking level, expect for 4 BAs which showed recoveries of between 67.9% and 76.8%. At the 250% spiking level, 30 of the 33 BAs quantified in the human test sample showed recoveries in the range of 82.3% to 115.5%. The other 3 BAs showed the recoveries as 128.7% for hyocholic acid (also known as λ- or γ-MCA), 68.8% for murocholic acid and 73.1% for tauro-β-MCA. In Supporting Information Table S4, 33 of the 34 BAs quantified in the mouse test sample show measured recoveries at the 100% spiking level in a range of 86.5% to 114.8%, except for tauro-CA that had a recovery of 79.4%. At the 250% spiking level, 31 of the 34 BAs quantified in the same sample showed recoveries in the range of 86.8% to 116.3%. For the remaining three low-abundance BAs, the recoveries were 120.8% for 3α,6α,7α,12α-THBA, 132.6% for Tauro-3α,6α,7α,12α-THBA and 125.7% for taurohyocholic acid. For those BAs that were not detected or were detected at concentrations below the LLOQs, standard substance spiking recovery tests were performed using a new strategy, i.e., by spiking two levels of the standard substances, equivalent to 5 and 12.5 times the LLOQ for each of the 17 and 16 BAs, respectively, which were not quantifiable in the two test samples. The % recoveries were calculated as the measured amount divided by the 5- or 12.5-times LLOQ x 100 for each of these BAs. Table 1 and Supporting Information Table S4 also list the recoveries

13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 31

measured in this way. As shown, at the 5-times LLOQ spiking level, recoveries of between 85.9% and 111.8% were achieved for 14 of the 17 BAs in the human serum test sample. For the other BAs, the recoveries were 72.4% for dehydrocholic acid, 125.1% for tauro-3α,6α,7α,12αTHBA, and 67.4% for glycodehydrocholic acid. At the 12.5-times LLOQ spiking level, recoveries of between 80.6% and 116.2% were achieved for 15 of the 17 BAs (all except tauro3α,6α,7α,12α-THBA and glycodehydrocholic acid). The former had a recovery of 121.3% and the latter had a recovery of 67.4%. In contrast, 13 of the 16 BAs spiked into the mouse plasma at the 5-times the LLOQs showed their recoveries between 81.2% and 114.0%, with the recoveries of the other 3 BAs being between 68.9% and 71.0%. At the 12.5-times LLOQ spiking level, 9 of the 16 BAs showed their recoveries in a range of 82.3% to 102.9%, while the remaining 7 BAs showed their recoveries between 68.3% and 79.4%. For all the recovery tests, six technical replicates were performed. Based on the measured recoveries as listed in Table 1 and Supporting Information Table S4, most of the 50 BAs showed the quantitation accuracies of 80% to 120%, with RSDs of ≤ 10.7% (n=6), which indicated that the combined PD-SPE and UPLC/MRM-MS method offered both precise and accurate quantitation for most of the 50 BAs in both the human serum and mouse plasma samples. Some BAs showed recoveries of 120% when the standard substances were spiked into the two test samples, which indicated significant ionization suppression or enhancement in ESI (i.e., matrix effects),42,43 which most likely resulted from coeluting compounds in the samples. The use of 13C-labeled versions of these BAs (if available) would overcome this issue. Determination of BAs in Fasting versus Non-fasting Human Serum. The new method was used to analyze BAs in the human serum samples that had been collected from seven healthy

14


Page 15 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


individuals with and without fasting. Forty-three of the 50 targeted BAs, including 26 unconjugated, 7 glycine-conjugated, and 10 taurine-conjugated species were quantitated in these samples. To our knowledge, this is the largest number of BAs that have been detected and quantitated in human blood thus far. The quantitated BAs included many low-abundance species, some of which have never been reported in human blood before -- e.g., 3α, 6β, 7α, 12α-THBA, 3α, 6α, 7α, 12α-THBA, and most of the oxo or keto BAs as shown in Supporting Information Table S5. This table lists the measured concentrations of these 43 BAs, and Supporting Information Figure S3 shows the total concentrations of these BAs in the serum samples of the seven individuals under fasting and non-fasting conditions, as well as the distributions between the unconjugated and conjugated species. All seven individuals showed significant increases in the total BA concentration in the peripheral blood circulation under non-fasting conditions. Glycine-conjugated BAs were the dominant forms in the blood of all of the individuals, with the molar %s ranging from 46.4% to 69.3% of the total concentrations under the fasting condition, and from 57.7% to 89.0% under the non-fasting condition. Although the total BA concentrations in all the individuals increased under non-fasting conditions, different patterns of concentration changes in the three classes of BAs were observed. Five individuals showed blood concentration increases in unconjugated BAs while the other two showed concentration decreases. For the conjugated BAs, one of the seven individuals showed a slight concentration decrease of glycineconjugated BAs and one individual showed a slight concentration decrease in taurine-conjugated BAs, while the other individuals showed significant concentration increases of both glycine and taurine-conjugated BAs in the blood. In humans, bile acid:CoA synthase5 and bile acid:amino acid transferase44 are the two key enzymes involved in conjugation of BAs to glycine and taurine in the liver, though the latter is also involved in conjugation of fatty acids to glycine.45 These

15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 31

observations reflected the large biological variability in humans, which might come from the significantly different activities of these two enzymes among the seven individuals. In humans, the BA pool consists of CA, CDCA, and DCA in an approximate 40:40:20 molar ratio.11 Based on the concentrations of the 43 BAs measured in this study, the average molar ratio of CA, CDCA and DCA, including their unconjugated and conjugated forms, in the seven humans was about 20:52:28 under the fasting conditions, and this ratio was 15:55:30 in the non-fasting individuals, which were different from the previously reported ratio.11 Metabolic Profiling of BAs in mdr2-/- and WT Mouse Plasma. Quantitation of Known BAs. From the scheduled MRM experiments, 38 of the 50 BAs could be quantitated in the WT mouse plasma samples and 44 of the 50 targeted BAs were quantitated in the plasma of mdr2-/- mice. This represents the largest number of BAs that have been detected and quantitated in mouse blood thus far. The concentrations of these measured BAs are listed in Supporting Information Table S6. It was noticed that nordeoxycholic acid, which was not detectable or was at very low concentrations in biological samples -- and which was once widely used as internal standard in GC for the determination of BAs46,47 -- was detected in all of the plasma samples from both the WT and mdr2-/- mice. In fact, this BA was detected as a very abundant species in all the six WT animals, with an average of 37.3% of the total molar concentration of the 38 BAs quantitated in the WT strain. In addition to nordeoxycholic acid, other major BAs detected in the WT mouse plasma included β-MCA, ω-MCA, CA, DCA, and their taurine conjugates. These BAs together with nordeoxycholic acid accounted for an average of 96.9% of the total amount. As shown in Supporting Information Table S6, the total concentrations of the quantitated BAs in the mdr2-/- -mouse plasma were significantly higher than in the WT mouse plasma, by an average of 69.4 fold. The accumulation of BAs in the

16


Page 17 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


plasma indicated cholestasis, i.e. the stagnation of bile and the accumulation of biliary constituents in the body. In addition to the significantly elevated total BA concentrations, the compositions of BAs in the plasma of the mdr2-/- mice were also different from those of the WT mice. The major BAs in the WT plasma were taurine-conjugated and unconjugated species, with average molar percentages of 51.1% and 48.8%, respectively, of the total. The remaining 0.1% BAs were glycine conjugated. In contrast, the 44 BAs measured in the mdr2-/- plasma were dominated by taurineconjugated species, with molar percentages ranging from 92.2% to 99.1% of the total amounts in the six mdr2-/- mice. Also, three taurine conjugated BAs -- i.e., tauro-β-MCA, tauro-ω-MCA, and tauro-CA -- were the predominant species among the 44 BAs in the mdr2-/-plasma, with average molar percentages of 59.7%, 18.8%, and 17.9% of the total, respectively. As observed, hydrophilic BAs, and the less-hydrophobic species that come from liver detoxification of the hydrophobic BAs, dominated the BA compositions in the cholestatic mdr2-/plasma. In addition to the three very abundant taurine conjugates (i.e., tauro-β-MCA, tauro-ωMCA, and tauro-CA), the major less abundant BAs that were detected in the mdr2-/- mouse plasma included tauro-α-MCA (1.0% ), ω-MCA(0.53%), β-MCA (0.46%), tauro-CDCA (0.42%), and taurohyodeoxycholic acid (0.21%). Also, two THBA species -- 3α,6β,7α,12α-THBA (0.057%) and tauro-3α,6α,7α,12α-THBA (0.073%) -- were detected at relatively high concentrations in the mdr2-/- mouse plasma compared to the WT mouse plasma. 3α,6β,7α,12αTHBA or 3α,6α,7α,12α-THBA, and their taurine conjugates have been reported as the major species formed in the urine of cholic acid-fed Fxr-/- mouse and are likely to be formed by 6hydroxylation of CA and its taurine conjugate.48 The appearance of these two THBAs in the cholestatic mdr2-/-mice suggests increased hepatic detoxification and reduction of lipophilic BAs

17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 31

in the liver.4,49 In addition, oxo (keto) species such as 7-ketodeoxycholic acid, 3-oxocholic acid, and 12-ketolithocholic acid also showed significantly higher concentrations in mdr2-/- mice than in WT mice. This is also consistent with enhanced hepatic detoxification in the cholestatic mouse liver. Also, allocholic acid, typically a fetal bile acid that reappears during rat liver regeneration50 and human and rat carcinogenesis,51,52 was also observed with significant concentration elevation (ca. 100 times) in the mdr2-/- plasma. Untargeted Detection of Potentially Unknown BAs. Though 50 BAs were involved in the targeted quantitation described above, it was found that there were still many potential BAs that were detectable in the WT and mdr2-/- mouse plasma samples. A panel of MRM transitions for each group of glycine- and taurine-conjugated BAs that contained 0 to 4 OH groups in their structures, and for unconjugated THBAs, were compiled for untargeted detection of potential BAs by UPLC/MRM-MS. Supporting Information Table S7 lists these MRM transitions. For each group, three structurally specific pairs of the Q1 to Q3 ion transitions were selected from Supporting Information Table S2. The declustering potential and collision energy voltages used were the median values from the parameters for all the isomeric members in each group, as listed in Supporting Information Table S2. Because the unconjugated dehydro- (i.e., no OH), and mono-, di-, and tri-OH BAs only showed low-abundance fragments upon CID or no detectable fragments at all (e.g., for LCA), and because pseudo-MRM, i.e., the use of (M-H)-/(M-H)- for the ion monitoring, lacks the specificity and selectivity required for reliable detection, no MRM transitions were included for detection of potentially unknown isomers of these unconjugated BAs. UPLC-MS/MS runs with unscheduled MRM scans were performed on the WT and mdr2-/plasma samples and up to 36 such compounds were detected for the groups of glycine

18


Page 19 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


conjugated BAs which contained 2 or 3 OHs, taurine conjugated BAs which contained 2 to 4 OHs, and unconjugated THBAs. No unknown isomeric compounds were detected for the groups of glycine and taurine conjugated BAs which contained no or only 1 OH in their structures. The numbers of the potential BAs detected in the plasma samples are listed in Supporting Information Table S7. To provide complementary confirmation on the detection of these potential BAs, UPLC/Fourier transform (FT) MS and UPLC-MS/MS by CID were performed on the same sample set using an ultrahigh-resolution LTQ-Orbitrap Fusion mass spectrometer, as described in the Supporting Information. As a result, all of the 36 compounds detected by UPLC/MRM-MS were also detected by UPLC/FTMS within a mass error of ±2.5 ppm. Figure 2A shows a comparison of the extracted ion current chromatograms between UPLC/MRM-MS and UPLC/FTMS for detection of the potentially unknown isomers of tauro-3α,6α,7α,12αTHBA in both of the WT and the mdr2-/- mouse plasma samples. As shown, at least 16 tauroTHBA isomers were detected and only one of these compound, tauro-3α,6α,7α,12α-THBA, was known. Similarly, up to 12 isomers of unconjugated THBAs were detected in the same way, with only three compounds (2α, 6α, 7α, 12α-THBA, 3α, 6α, 7α, 12α-THBA, and α, 6β, 7α, 12αTHBA) being known. Because of the high speed of MS scans and the high efficiency of highenergy CID on the LTQ-Orbitrap instrument, the MS/MS spectra were acquired for all of these potential BA compounds. Interpretation of the acquired tandem mass spectra supported the assignments of these compounds to each of their belonged BA groups. For example, all of the putative taurine conjugates showed the characteristic fragments for taurine at m/z 124.01, 106.98 and 79.96, while all of the detected unknown glycine-conjugated BA isomers showed the characteristic fragment for glycine at m/z 74.024 in addition to their structure-related fragments (M-CO2-H)-, (M-CO2-H2O-H)- and/or (M-H2O-H)-. Figure 2B displays three representative CID

19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 31

mass spectra for one of the detected isomers in each group of tauro-THBAs (I), glyco-tri-OH BAs (II) and unconjugated THBAs (III). It should be mentioned that most of these potential BAs displayed significantly higher abundances in the mdr2-/- mice than in the WT mice, as did the known BAs. The detection of these potential isomeric BAs in the plasma samples, on one hand, indicated a more complex BA profile in the mouse plasma than we previously anticipated, although further structural elucidation of these compounds is needed for determination of their identities. On the other hand, these results also demonstrated the capability of the combined UPLC/MRMMS, UPLC/ultrahigh resolution FTMS, and high-efficiency UPLC-MS/MS technique used in this study to detect and quantify many potentially unknown yet authentic endogenous compounds in biological samples. We believe that there are many other classes of such metabolites which are currently not covered by any metabolome databases.

CONCLUSION An improved UPLC/MRM-MS method that utilizes high-throughput-compatible 96-well phospholipid-depletion solid-phase extraction as a new sample preparation approach has been developed for quantitative analysis of targeted BAs in human and mouse blood. This method enabled separation of the 50 known BAs in a single analysis with high sensitivity detection in the sub-nM concentration ranges. This made it possible to successfully determine the concentrations of more than 40 of the known BAs in the human serum and mouse plasma samples, with a few minor and low abundance BAs being measured in these biological fluids for the first time. Phospholipid depletion was shown to be a superior sample cleanup and analyte enrichment procedure than three other sample preparation procedures (PPT, reversed-phase SPE and high-

20


Page 21 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


pH SPE), and efficiently removed phospholipids from the blood samples. Method validation demonstrated precise quantitation of BAs in the blood samples, and accurate quantitation for most of them. In addition, an MRM-based untargeted UPLC-MS/MS strategy was applied and led to the detection of many other potential BAs in the mouse WT and KO plasma samples, though further structural elucidation of the detected compounds is needed. In summary, this study provides a useful UPLC-MS/MS method for a more comprehensive analysis of BAs in the blood samples. Supporting Information. Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

ACKNOWLEDGMENTS This work at the University of Victoria-Genome BC Proteomics Centre was supported by the Genome Canada, Genome Alberta, and Genome BC funded “The Metabolomics Innovation Centre (TMIC)”, and by funding from Genome Canada and Genome BC through the “Science and Technology Innovation Centre (S&TIC)”. The research work in Dr. Victor Ling’s group was supported by funding from the Terry Fox Research Institute (TFRI) and the Canadian Institutes of Health Research (CIHR). We thank Jonathan Sheps and Lin Liu for the preparation and collection of the mouse samples, and Dr. Carol Parker for review of this manuscript.

21



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 31

REFERENCES (1)

Russell, D. W. Annu Rev Biochem 2003, 72, 137-174.

(2)

Russell, D. W. J Lipid Res 2009, 50 Suppl, S120-125.

(3)

Gerloff, T.; Stieger, B.; Hagenbuch, B.; Madon, J.; Landmann, L.; Roth, J.;

Hofmann, A. F.; Meier, P. J. J Biol Chem 1998, 273, 10046-10050. (4)

Wang, R.; Salem, M.; Yousef, I. M.; Tuchweber, B.; Lam, P.; Childs, S. J.;

Helgason, C. D.; Ackerley, C.; Phillips, M. J.; Ling, V. Proc Natl Acad Sci U S A 2001, 98, 2011-2016. (5)

Franklund, C. V.; de Prada, P.; Hylemon, P. B. J Biol Chem 1990, 265, 9842-

(6)

Hofmann, A. F.; Hagey, L. R. Cell Mol Life Sci 2008, 65, 2461-2483.

(7)

Hagey, L. R.; Moller, P. R.; Hofmann, A. F.; Krasowski, M. D. Physiol Biochem

9849.

Zool 2010, 83, 308-321. (8)

Sharma, R.; Long, A.; Gilmer, J. F. Curr Med Chem 2011, 18, 4029-4052.

(9)

Lefebvre, P.; Cariou, B.; Lien, F.; Kuipers, F.; Staels, B. Physiol Rev 2009, 89,

147-191. (10)

Nguyen, A.; Bouscarel, B. Cell Signal 2008, 20, 2180-2197.

(11)

Li, T.; Chiang, J. Y. J Lipids 2012, 2012, 754067.

(12)

Payne, C. M.; Bernstein, C.; Dvorak, K.; Bernstein, H. Clin Exp Gastroenterol

2008, 1, 19-47. (13)

Yoshimoto, S.; Loo, T. M.; Atarashi, K.; Kanda, H.; Sato, S.; Oyadomari, S.;

Iwakura, Y.; Oshima, K.; Morita, H.; Hattori, M.; Honda, K.; Ishikawa, Y.; Hara, E.; Ohtani, N. Nature 2013, 499, 97-101.

22


Page 23 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(14)

Dvorak, K.; Payne, C. M.; Chavarria, M.; Ramsey, L.; Dvorakova, B.; Bernstein,

H.; Holubec, H.; Sampliner, R. E.; Guy, N.; Condon, A.; Bernstein, C.; Green, S. B.; Prasad, A.; Garewal, H. S. Gut 2007, 56, 763-771. (15)

Dixon, M. F.; Mapstone, N. P.; Neville, P. M.; Moayyedi, P.; Axon, A. T. Gut

2002, 51, 351-355. (16)

Ross, R. K.; Hartnett, N. M.; Bernstein, L.; Henderson, B. E. Br J Cancer 1991,

63, 143-145. (17)

Kaplan, M. M.; Gershwin, M. E. N Engl J Med 2005, 353, 1261-1273.

(18)

Grundy, S. M.; Ahrens, E. H., Jr.; Miettinen, T. A. J Lipid Res 1965, 6, 397-410.

(19)

Jones, D. C.; Melchior, G. W.; Reeves, M. J. J Lipid Res 1976, 17, 273-277.

(20)

Makita, M.; Wells, W. W. Anal Biochem 1963, 5, 523-530.

(21)

Van Berge Henegouwen, G. P.; Ruben, A.; Brandt, K. H. Clin Chim Acta 1974,

54, 249-261. (22)

Vreken, P.; van Rooij, A.; Denis, S.; van Grunsven, E. G.; Cuebas, D. A.;

Wanders, R. J. J Lipid Res 1998, 39, 2452-2458. (23)

Janzen, N.; Sander, S.; Terhardt, M.; Das, A. M.; Sass, J. O.; Kraetzner, R.;

Rosewich, H.; Peter, M.; Sander, J. J Lipid Res 2010, 51, 1591-1598. (24)

Scherer, M.; Gnewuch, C.; Schmitz, G.; Liebisch, G. J Chromatogr B Analyt

Technol Biomed Life Sci 2009, 877, 3920-3925. (25)

Cai, X.; Liu, Y.; Zhou, X.; Navaneethan, U.; Shen, B.; Guo, B. Biomed

Chromatogr 2012, 26, 101-108. (26)

Tagliacozzi, D.; Mozzi, A. F.; Casetta, B.; Bertucci, P.; Bernardini, S.; Di Ilio, C.;

Urbani, A.; Federici, G. Clin Chem Lab Med 2003, 41, 1633-1641.

23



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(27)

Page 24 of 31

Trottier, J.; Bialek, A.; Caron, P.; Straka, R. J.; Heathcote, J.; Milkiewicz, P.;

Barbier, O. Dig Liver Dis 2012, 44, 303-310. (28)

Humbert, L.; Maubert, M. A.; Wolf, C.; Duboc, H.; Mahe, M.; Farabos, D.;

Seksik, P.; Mallet, J. M.; Trugnan, G.; Masliah, J.; Rainteau, D. J Chromatogr B Analyt Technol Biomed Life Sci 2012, 899, 135-145. (29)

Garcia-Canaveras, J. C.; Donato, M. T.; Castell, J. V.; Lahoz, A. J Lipid Res

2012, 53, 2231-2241. (30)

Huang, J.; Bathena, S. P.; Csanaky, I. L.; Alnouti, Y. J Pharm Biomed Anal 2011,

55, 1111-1119. (31)

MacLellan, S. A.; MacAulay, C.; Lam, S.; Garnis, C. BMC Clin Pathol 2014, 14,

(32)

Smit, J. J.; Schinkel, A. H.; Oude Elferink, R. P.; Groen, A. K.; Wagenaar, E.; van

27.

Deemter, L.; Mol, C. A.; Ottenhoff, R.; van der Lugt, N. M.; van Roon, M. A.; et al. Cell 1993, 75, 451-462. (33)

Katzenellenbogen, M.; Mizrahi, L.; Pappo, O.; Klopstock, N.; Olam, D.; Jacob-

Hirsch, J.; Amariglio, N.; Rechavi, G.; Domany, E.; Galun, E.; Goldenberg, D. Mol Cancer Res 2007, 5, 1159-1170. (34)

Mauad, T. H.; van Nieuwkerk, C. M.; Dingemans, K. P.; Smit, J. J.; Schinkel, A.

H.; Notenboom, R. G.; van den Bergh Weerman, M. A.; Verkruisen, R. P.; Groen, A. K.; Oude Elferink, R. P.; et al. Am J Pathol 1994, 145, 1237-1245. (35)

https://www.sigmaaldrich.com/content/dam/sigma-

aldrich/docs/Supelco/Datasheet/1/T713202_HybridSPE-PLus.pdf. (36)

Xia, Y. Q.; Jemal, M. Rapid Commun Mass Spectrom 2009, 23, 2125-2138.

24


Page 25 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(37)

Psychogios, N.; Hau, D. D.; Peng, J.; Guo, A. C.; Mandal, R.; Bouatra, S.;

Sinelnikov, I.; Krishnamurthy, R.; Eisner, R.; Gautam, B.; Young, N.; Xia, J.; Knox, C.; Dong, E.; Huang, P.; Hollander, Z.; Pedersen, T. L.; Smith, S. R.; Bamforth, F.; Greiner, R.; McManus, B.; Newman, J. W.; Goodfriend, T.; Wishart, D. S. PLoS One 2011, 6, e16957. (38)

Kweon, H. K.; Hakansson, K. Anal Chem 2006, 78, 1743-1749.

(39)

Pucci, V.; Di Palma, S.; Alfieri, A.; Bonelli, F.; Monteagudo, E. J Pharm Biomed

Anal 2009, 50, 867-871. (40)

Burkard, I.; von Eckardstein, A.; Rentsch, K. M. J Chromatogr B Analyt Technol

Biomed Life Sci 2005, 826, 147-159. (41)

Sakakura, H.; Kimura, N.; Takeda, H.; Komatsu, H.; Ishizaki, K.; Nagata, S. J

Chromatogr B Biomed Sci Appl 1998, 718, 33-40. (42)

Vogeser, M.; Seger, C. Clin Chem 2010, 56, 1234-1244.

(43)

Mulvana, D. E. Bioanalysis 2010, 2, 1051-1072.

(44)

Johnson, M. R.; Barnes, S.; Kwakye, J. B.; Diasio, R. B. J Biol Chem 1991, 266,

10227-10233. (45)

O'Byrne, J.; Hunt, M. C.; Rai, D. K.; Saeki, M.; Alexson, S. E. J Biol Chem 2003,

278, 34237-34244. (46)

van Berge-Henegouwen, G. P.; Allan, R. N.; Hofmann, A. F.; Yu, P. Y. J Lipid

Res 1977, 18, 118-122. (47)

Roda, E.; Aldini, R.; Mazzella, G.; Roda, A.; Sama, C.; Festi, D.; Barbara, L. Gut

1978, 19, 640-649. (48)

Cho, J. Y.; Matsubara, T.; Kang, D. W.; Ahn, S. H.; Krausz, K. W.; Idle, J. R.;

Luecke, H.; Gonzalez, F. J. J Lipid Res 2010, 51, 1063-1074.

25



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(49)

Page 26 of 31

Perwaiz, S.; Forrest, D.; Mignault, D.; Tuchweber, B.; Phillip, M. J.; Wang, R.;

Ling, V.; Yousef, I. M. J Lipid Res 2003, 44, 494-502. (50)

Monte, M. J.; Martinez-Diez, M. C.; El-Mir, M. Y.; Mendoza, M. E.; Bravo, P.;

Bachs, O.; Marin, J. J. J Hepatol 2002, 36, 534-542. (51)

El-Mir, M. Y.; Badia, M. D.; Luengo, N.; Monte, M. J.; Marin, J. J. Clin Sci

(Lond) 2001, 100, 499-508. (52)

Mendoza, M. E.; Monte, M. J.; El-Mir, M. Y.; Badia, M. D.; Marin, J. J. Clin Sci

(Lond) 2002, 102, 143-150.

26


Page 27 of 31

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 1. Endogenous concentration of bile acids in the human serum test sample and the precision and accuracy of quantitation No

Compound

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50

3α,6β,7α,12α-THBA 3α,6α,7α,12α-THBA Ursocholic acid Dioxolithocholic acid Dehydrocholic acid ω-Muricholic acid α-Muricholic acid 7-Ketodeoxycholic acid β-Muricholic acid 2α,3α,7α,12α-THBA 12-Ketochenodeoxycholic acid λ-Muricholic acid Murocholic acid 3-Oxocholic acid Allocholic acid Cholic acid Ursodeoxycholic acid Hyodeoxycholic acid 7-Ketolithocholic acid 6,7-Diketolithocholic acid 12-Ketolithocholic acid Nordeoxycholic acid Apocholic acid Chenodeoxycholic acid Deoxycholic acid Alloisolithocholic acid Isolithocholic acid Isodeoxycholic acid Lithocholic acid Dehydrolithocholic acid Tauro-3α,6α,7α,12α-THBA Taurodehydrocholic acid Tauro-α-muricholic acid Glycodehydrocholic acid Tauro-β-muricholic acid Tauro-ω-muricholic acid Tauro-λ-muricholic acid Tauroursodeoxycholic acid Taurohyodeoxycholic acid Glyco-λ-muricholic acid Taurocholic acid Glycoursodeoxycholic acid Glycohyodeoxycholic acid Glycocholic acid Taurochenodexycholic acid Taurodeoxycholic acid Glycochenodeoxycholic acid Glycodeoxycholic acid Taurolithocholic acid Glycolithocholic acid

Human serum C (nM) -* 1.66 46.5 5.63 0.47 23.9 0.36 0.45 0.85 40.6 109.5 3.13 8.81 21.8 8.85 178.6 494.7 9.59 9.48 9.50 3.85 4.84 0.82 4.89 26.0 109.3 377.0 1.63 295.5 66.3 113.3 1389.8 445.3 9.99

RSD% 7.7 4.1 3.4 8.9 9.8 9.1 5.6 3.9 1.5 3.0 4.9 5.7 4.8 5.7 1.4 1.9 9.3 2.3 4.3 6.9 8.9 8.7 3.3 7.9 6.8 3.3 7.3 2.5 2.8 4.7 2.3 6.4 4.4

Precision, RSD% (n=6) Intra-day 4.7 6.5 5.6 4.7 3.7 5.9 6.7 7.8 2.9 3.3 4.8 7.5 5.2 7.1 4.1 3.8 6.3 3.4 6.3 6.5 6.3 5.8 4 4.4 4.1 4.4 8.4 2.0 4.6 2.4 1.0 4.8 3.7

Inter-day 4.9 6.7 5.6 6.3 5.3 4.4 3.6 8.3 2.3 6.5 7.1 5.7 5.4 6.1 2.2 4.1 4.7 3.2 7.4 4.1 6.2 6.1 4.6 4.6 4.5 4.6 10.5 5.6 4.0 3.4 4.5 5.3 3.8

Accuracy – recovery % (n=6) standard spiking of endogenous levels 100% spiking RSD% 250% spiking RSD% 91.7 6.1 86.2 3.9 93.5 2.4 84.3 5.0 76.3 5.2 84.9 2.6 85.2 9.0 93.5 3.2 113.9 6.1 128.7 7.3 69.2 9.1 68.8 7.3 105.7 1.3 109.6 6.6 104.6 0.8 105.6 5.7 102.2 3.3 105.0 3.1 95.5 1.2 94.6 3.0 88.5 6.5 91.9 2.6 99.0 2.5 97.8 1.5 103.7 10.2 105.1 3.1 110.8 1.1 112.9 4.8 100.5 8.4 106.2 4.2 99.0 2.0 95.7 1.9 89.4 3.0 83.4 4.0 89.9 1.9 88.1 0.7 96.0 10.2 99.7 4.7 85.3 8.3 100.8 4.1 80.7 6.1 73.1 5.3 76.8 4.5 93.7 2.8 87.4 7.0 89.6 3.9 93.6 4.6 98.8 2.9 100.5 4.6 105.5 3.4 110.5 3.0 115.5 2.2 77.8 6.9 82.3 8.0 100.7 1.9 102.4 1.6 101.1 2.6 98.5 3.9 107.6 1.1 113.9 3.4 101.5 0.7 102.9 1.0 109.9 1.5 103.0 3.0 98.6 4.1 96.3 2.8

Accuracy – recovery % (n=6) standard spiking based on LLOQs 5x spiking RSD% 12.5x spiking RSD% 87.4 6.5 89.2 4.6 92.5 3.7 88.5 6.1 72.4 1.6 80.6 4.5 99.2 1.5 101.1 4.6 94.8 6.6 99.1 3.8 98.3 2.9 103.3 6.1 93.8 7.5 84.9 3.1 111.8 0.8 111.5 5.0 85.9 3.2 86.0 1.4 109.4 3.9 109.2 2.0 125.1 5.9 121.3 3.8 95.2 8.1 103.1 2.7 67.4 5.1 71.4 8.8 100.2 1.7 102.3 6.0 110.5 1.8 116.2 5.7 111.1 0.7 109.6 4.3 -

* Denotes not detectable or not applicable.

27



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 31

Figure Legends Figure 1. UPLC/MRM-MS of 30 unconjugated (A), 8 glycine-conjugated and 12 taurineconjugated BAs (B), and 14 deuterium (D)-labeled BAs as IS (ion signals shown in red). Figure 2. Untargeted detection of potentially unknown tauro-THBAs in WT and mdr2-/- mouse plasma by UPLC/MRM-MS (A, I) and complementary detection by UPLC/Fourier transform MS with accurate mass measurements (A, II). (B) Three representative collision-induced dissociation mass spectra by UPLC-MS/MS of the unknown isomers of tauro-THBAs (I), glycine-tri-OH BAs (II) and unconjugated THBAs (III).

28


Page 29 of 31

Figure 1.

Intensity, cps

23

13

5.0e6

16 15

3

4.0e6

6

3.0e6

8 7

12 14

9

2.0e6 1

A

24 25

22 28

17 21

18

2

1.0e6

19

29 26 27

10 11

45

30

20

0.0 4

6

8

10

2.5e6

Intensity, cps

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


14

16

18

20

22

24

39 38

2.0e6

40 41

37

1.5e6 1.0e6 31

32

5.0e5 0.0

12

4

46

43 44 42

26

B

49 48 50

47 45

33 34 35 36 6

8

(1) 3α,6β,7α,12α-tetrahydrox bile acid (THBA) (2) 3α,6α,7α,12α-THBA (3) ursocholic acid (4) dioxolithocholic acid (5) dehydrocholic acid (6) ω-muricholic acid (7) α-muricholic acid (8) 7-ketodeoxycholic acid (9) β-muricholic acid (10) 2α,3α,7α,12α-THBA (11) 12-ketochenodeoxycholic acid (12) λ-muricholic acid (13) murocholic acid (14) 3-oxocholic acid (15) allocholic acid (16) cholic and cholic-D4 acid (17) ursodeoxycholic and ursodeoxycholic- D4 acid

10

12

14 Time, min

16

(18) hyodeoxycholic acid (19) 7-ketolithocholic acid (20) 6,7--diketolithocholic acid (21) 12-ketolithocholic acid (22) 23-nordeoxycholic acid (23) apocholic acid (24) chenodeoxycholic and chenodeoxycholic-D4 acid (25) deoxycholic and deoxycholic-D4 acid (26) alloisolithocholic acid (27) isolithocholic acid (28) isodeoxycholic acid (29) lithocholic and lithocholic-D4 acid (30) dehydrolithocholic acid (31) tauro-3α,6α,7α,12α-THBA (32) taurodehydrocholic acid (33) tauro-α-muricholic acid (34) glycodehydrocholic acid

18 (35) (36) (37) (38) (39) (40) (41) (42) (43) (44) (45) (46) (47) (48) (49) (50)

20

22

24

26

tauro-β-muricholic acid tauro-ω-muricholic acid taurohyocholic acid tauroursodeoxycholic and tauroursodeoxycholic-D4 acid taurohyodeoxycholic acid glycohyocholic acid taurocholic and taurocholic-D4 acid glycoursodeoxycholic and glycoursodeoxycholic-D4 acid glycohyodeoxycholic acid glycocholic and glycocholic-D4 acid taurochenodeoxycholic and taurochenodeoxycholic-D4 acid taurodeoxycholic and taurodeoxycholic-D6 acid glycochenodeoxycholic and glycochenodeoxycholic-D4 acid glycodeoxycholic and glycodeoxycholic-D4 acid taurolithocholic acid glycolithocholic and glycolithocholic-D4 acid.

29



Figure 2

(II) UPLC/FTMS, XIC of m/z 530.2793 ± 2.5 ppm

1.2e4 6e3

4 12 3 56

Intensity

0

4

6 Time (min)

8

10

8 (Tauro-3α,6α,7α,12α-THBA)

1.6e5

1.6e5

2

1 23

0

56 7

2

9

11 13 14

4

15

10

4 1

3 2

0

6 5 7

9

2

15

11 14 10 13

4

16

6 Time (min)

(taurine-NH3-H)-

100

150

200

250

300 m /z

350

400

450

9.0e4

N H

HO

74.024

1.2e5 Intensity

(M-H2O-H)512.268

5.0e 4

OH OH

HO

H S O 3-

79.957

1.0e 5

-106.980 -79.957

124.007

1.5e 5

106.980

Intensity

2.0e 5

(glycine-H)-

SO3 H

8

-74.024 -46.025

(M-CO2-H)-

(M-HCOOH-H)HO

OH

(M-H2O-CO2-H)-

6.0e4

100

160

220

12

C O H O

3.0e4

500

10

-44.009

(II)

-124.007 N H

HO

(taurine-H)-

12

6.0e5

12

530.278

B.

O

10

mdr2-/-

O

(I)

8

9.0e5

3.0e5 8

6 Time (min)


16

6 Time (min)

4

1.2e6

mdr2-/-

4

4.0e4

101112 13 14 15

34

0

12

1.2e5 8.0e4


2.4e5

8e4

1011121314 15

2

WT

3.2e5 Intensity

1.8e4

16

WT


280 m/z

340

400

464.3018

2.4e4

Intensity

Intensity

16

(M-H2O-H)446.292

(I) UPLC/MRM/MS, 530.3/79.9 and 530.3/124.0

402.288 418.295 420.311

A.

460

O

(III)

OH

HO

1.0e4

423.275

-46.0254 OH

(M-HCOOH-H2O-H)-

341.248

4.0e3 2.0e3 300

320

340

360 m/z

(M-HCOOH-H)(M-H2O-H)405.264

(M-HCOOH-2H2O-H)-

6.0e3

377.269

OH

359.258

8.0e3 HO Intensity

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 31

380

400

420

30


Page 31 of 31

TOC Graphic Phosholipid depletion - UPLC/MRM-MS Intensity, cps

5.0e6

16 15

3

4.0e6

23

13

Unconjugated bile acids

6

3.0e6

8 7

12 14

9

1

24 25

22 28

17 18

2

2.0e6 1.0e6

21 19

29 26 27

10 11

45

30

20

0.0 4

6

8

10

2.5e6 Conjugated bile acids

Intensity, cps

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


38

2.0e6

14

16

18

20

22

24

26

46

43 44 42

22

24

26

49 48 50

47 45

33 34

1.0e6 31

32

5.0e5 0.0

40 41

37

1.5e6

12

39

4

35 36 6

8

10

12

14 Time, min

16

18

20

31


Metabolic Profiling of Bile Acids in Human and Mouse Blood by LC

Recommend Documents