Quantitative profiling of bile acids in rat bile using ultrahigh-performance liquid chromatography–orbitrap mass spectrometry: Alteration of the bile acid composition with aging

Journal of Chromatography B, 1031 (2016) 37–49

Contents lists available at ScienceDirect

Journal of Chromatography B journal homepage: www.elsevier.com/locate/chromb

Quantitative profiling of bile acids in rat bile using ultrahigh-performance liquid chromatography–orbitrap mass spectrometry: Alteration of the bile acid composition with aging Gakyung Lee a,b , Hyunbeom Lee a , Jongki Hong b , Soo Hyun Lee d , Byung Hwa Jung a,c,∗ a

Molecular Recognition Research Center, Korea Institute of Science and Technology, Seoul 02792, Republic of Korea Department of Basic Pharmaceutical Science, Kyung Hee University, Seoul 02447, Republic of Korea c Department of Biological Chemistry, Korea University of Science and Technology, Daejeon 34113, Republic of Korea d Department of Medical Records and Health Information Management, College of Nursing and Health, Kongju National University, Kongju 32588, Republic of Korea b

a r t i c l e

i n f o

Article history: Received 31 May 2016 Received in revised form 7 July 2016 Accepted 7 July 2016 Available online 9 July 2016 Keywords: Aging Bile acids (BAs) Bile UPLC–Orbitrap-MS

a b s t r a c t Bile acids (BAs) play important roles in physiological functions, including the homeostasis of cholesterol and lipids and as ligands for G protein-coupled receptors (GPCRs). With the increasing importance of BAs, analytical methods for their quantification and screening have been developed. However, due to the diverse range and variety of BAs with different activation potency, a simple, effective, and sensitive method is required to screen BAs for accurate quantification and identification. This paper presents an application of ultrahigh-performance liquid chromatography-orbitrap mass spectrometry (UHPLCLTQ–Orbitrap MS) for profiling BAs in bile. Using this method, along with the accurate quantification of 19 targeted BAs, 22 unknown BAs were detected and characterized by their fragmentation patterns. The method is beneficial for screening most of the BAs (quantitatively and qualitatively) in rat bile with simple preparation in a single run. The sample dilution ranges of each BA were optimized depending on the concentration of BAs in the bile to obtain good peak separation and accurate data. The method validation was performed successfully using charcoal-treated bile and the intra and inter-day coefficients of variation were less than 20% for all BAs while the recovery were above 88.5% except for the lithocholic acid. The method was applied to profile the age-dependent changes in the contents of rat BAs. Through statistical analysis, we found that as the rats aged, unconjugated BAs and glycine-conjugated BAs decreased or were unaffected, while taurine-conjugated BAs were increased in general. Among the unknown BAs, 5 of the taurine-conjugated BAs increased, while a glycine-conjugated BA decreased, in agreement with the trends of the targeted BAs. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Bile acids (BAs), the major constituents of bile [1], are well known for their role in the regulation of cholesterol homeostasis and lipid absorption. They also have functions in the ligands of nuclear receptors, such as the farnesoid X receptor (FXR) [2–4], Gprotein-coupled bile acid receptor (GPBAR1, also known as TGR5) [5], vitamin D receptor (VDR) [6], and pregnane X receptor (PXR) [7], which are related to liver disease. BAs are also known to be

∗ Corresponding author at: Molecular Recognition Research Center, Korea Institute of Science and Technology, 5, Hwarang-ro 14-gil, Seongbuk-gu, Seoul 02792, Republic of Korea. E-mail address: [email protected] (B.H. Jung). http://dx.doi.org/10.1016/j.jchromb.2016.07.017 1570-0232/© 2016 Elsevier B.V. All rights reserved.

cytotoxic [8] and cancer promoters [9]. With the increasing importance of BAs, the development of analytical methods for screening BAs in various biological fluids and tissues has been investigated by many researchers [10–16]. However, due to the BA structural similarity (Fig. 1), screening requires an analytical method with high sensitivity and specificity for efficient analysis. Because BAs exist in various amounts in biofluids and tissues, it is difficult to select a specific BA for quantification. In addition, a wide range of unknown BAs has been detected in various biofluids and tissues [17]. For the detection of bile acids, gas chromatography–mass spectrometry (GC–MS) has been used conventionally, providing high sensitivity and specificity [10]. Nevertheless because of the complicated and time-consuming preparation steps associated with GC–MS, including hydrolysis or derivatization techniques, liquid

38

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

Fig. 1. Chemical structures of bile acids.

chromatography (LC)-based methods were developed in the last decade. Although several detector-based (UV, fluorescence) assays coupled with LC have been developed, they still have disadvantages, such as limited sensitivity and specificity, and limited coverage of detectable complex biological matrices [11–13]. Recently, high-performance liquid chromatography–tandem mass spectrometry (HPLC–MS/MS) [18–21] and ultrahigh-performance liquid chromatography–tandem mass spectrometry (UHPLC–MS/MS) methods [14,16] were applied for BAs analysis. However, these methods have limitations in the separation of isomers, such as ␣-, ␤-, ␻-muricholic acid [14]. In particular, when using multiple reaction monitoring (MRM) mode, a wide range of unknown BAs in bile cannot be characterized, and small amounts of BAs cannot be detected due to the low sensitivity. For this purpose, high-resolution mass spectrometers (HRMS), such as Orbitrap or time-of-flight (ToF), have been increasingly used for quantification, identification, and characterization of various compounds [22]. These types of MS offer high resolution (>60,000 fwhm), accurate mass measurement (<5 ppm), excellent sensitivity and complete MS/MS fragment information. Nicholson et al. reported a comprehensive study on BA profiling and quantification method using UPLC coupled with Q-ToF mass spectrometer [16]. In addition, very few studies have been performed for the identification of unknown BAs and the quantification of major BAs at the same time [16,17]. Generally the quantitative and qualitative analysis of BAs were performed using two different instruments, comparing the two data sets became difficult. A simultaneous, simple, and rapid method that could analyze BAs in one run in a single instrument would be helpful. In this study, a method that can perform quantification and identification of BAs in bile simultaneously was developed using UPLC-LTQ–Orbitrap MS. BAs were extracted by solid phase extraction (SPE), and the concentration range for the quantification was optimized according to those of the individual BAs. The MS scan mode was used to quantify the targeted BAs and to detect unknown

BAs, while MS/MS mode was used to acquire the information of unknown BAs. We applied this method to study the age-dependent alteration of the bile composition in rats. Aging is highly correlated with the incidence of various diseases related to the liver and gastrointestinal tract [23]. BAs are correlated with the progression of liver gastrointestinal disease [24–26]. Therefore, investigation of the composition of BAs in bile is a useful way to study the metabolism related to many physiological conditions and to predict the potential therapeutic markers for longevity [27]. Furthermore, detected unknown BAs showing significant changes during aging are attractive targets for further studies to elucidate their biological functions in aging. 2. Materials and methods 2.1. Materials Compounds, ␣-muricholic acid (␣MCA), ␤-muricholic acid (␤MCA), ␻-muricholic acid (␻MCA), tauro-␣-muricholic acid (T␣MCA), and tauro-␤-muricholic acid (T␤MCA) were purchased from Steraloids, Inc. (Newport, RI, USA). Cholic acid (CA), chenodeoxycholic acid (CDCA), deoxycholic acid (DCA), lithocholic acid (LCA), ursodeoxycholic acid (UDCA), tauro-cholic acid (TCA), tauro-chenodeoxycholic acid (TCDCA), tauro-deoxycholic acid (TDCA), tauro-lithocholic acid (TLCA), tauro-ursodeoxycholic acid (TUDCA), glyco-cholic acid (GCA), glyco-chenodeoxycholic acid (GCDCA), glyco-deoxycholic acid (GDCA), glyco-lithocholic acid (GLCA), glycol-ursodeoxycholic acid (GUDCA), d4 -cholic acid (d4 CA), formic acid, and activated charcoal were purchased from Sigma-Aldrich (St. Louis, MO, USA). d4 -Chenodeoxycholic acid (d4 -CDCA) and d4 -lithocholic acid (d4 -LCA) were purchased from Toronto Research Chemicals Inc. (North York, Ontario, Canada). HPLC-grade acetonitrile and MS-grade methanol were obtained from SK chemicals (Ulsan, Republic of Korea). Ultrapure water (18.2 M cm) was obtained from a Milli-Q apparatus from Milli-

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

39

Table 1 Calibration curves information of each BAs. Compound name Internal standard Dilution factor (fold)

Adduct ion [M−H]−

Range (ng/mL)

R2

QC concentration LLOQ, L, M, H (ng/mL)

Weighting factor

CA ␣MCA ␤MCA ␻MCA UDCA CDCA DCA LCA

d4 -CA d4 -CA d4 -CA d4 -CA d4 -CDCA d4 -CDCA d4 -CDCA d4 -LCA

500 25 25 25 25 25 25 25

407.2803 407.2803 407.2803 407.2803 391.2854 391.2854 391.2854 375.2905

0.5–2500 5–8000 5–8000 1–5000 1–100 1–100 1–100 2–200

0.9992 0.9993 0.9997 0.9991 0.9916 0.9994 0.9980 0.9999

0.5, 75, 750, 1500 5, 250, 2500, 5000 5, 250, 2500, 5000 1, 150, 1500, 3000 1, 4.5, 45, 90 1, 4.5, 45, 90 1, 4.5, 45, 90 2, 7.5, 75. 150

1/x2 1/x2 1/x2 1/x2 1/x2 1/x 1/x2 1/x

Glycine-conjugated Bile acids GCA GUDCA GCDCA GDCA GLCA

d4 -CA d4 -CDCA d4 -CDCA d4 -CDCA d4 -LCA

500 25 25 25 25

464.3018 448.3068 448.3068 448.3068 432.3119

100–5000 10–1000 100–10000 50–5000 1–100

0.9997 0.9980 0.9983 0.9987 0.9973

100, 150, 1500, 3000 10, 45, 450, 900 100, 450, 4500, 9000 50, 150, 1500, 3000 1, 4.5, 45, 90

1/x 1/x 1/x 1/x 1/x2

Taurine-conjugated Bile acids TCA T-MCA (␣+␤) TUDCA TCDCA TDCA TLCA

d4 -CA d4 -CA d4 -CDCA d4 -CDCA d4 -CDCA d4 -LCA

500 500 500 500 500 25

514.2844 514.2844 498.2895 498.2895 498.2895 482.2946

100–5000 400–20000 100–5000 50–2500 50–2500 20–1000

0.9958 0.9977 0.9999 0.9988 0.9992 0.9998

100, 150, 1500, 3000 400, 500, 5000, 10000 100, 150, 1500, 3000 50, 75, 750, 1500 50, 75, 750, 1500 20, 45, 450, 900

1/x2 1/x2 1/x 1/x 1/x 1/x

Unconjugated Bile acids

pore (Milford, MA, USA). Oasis HLB SPE cartridges were purchased from Waters (Milford, MA, USA).

2.2. Instruments An Ultimate 3000 UHPLC system consisting of an autosampler and a column oven coupled to an LTQ Orbitrap Velos ProTM system mass spectrometer (Thermo Scientific, San Jose, CA, USA) with heated electrospray ionization source (HESI) was used. The software packages Xcalibur 2.2, Tune Plus 2.7, and Chromeleon MS Link 6.80 (all Thermo Scientific) were used to control the entire system and to perform the data analysis.

2.3. Liquid chromatography and mass spectrometric conditions All chromatographic separations were performed with an ACQUITY UPLC® BEH C18 column (2.1 × 100 mm, 1.7 ␮m, Waters, Milford, MA, USA). The mobile phase was composed of 0.1% formic acid in either 1% acetonitrile (v/v, mobile phase A) or 99% acetonitrile (v/v, mobile phase B). At a flow rate of 0.4 mL/min, the elution gradient was as follows: 100% to 75% mobile phase A from 0 to 5 min; linear increase to 40% mobile phase B from 5 to 13 min; linear increase of mobile phase B from 40% to 75% over 12 min; reequilibration with 100% mobile phase A from 26.5 to 30 min. The column was maintained at 50 ◦ C, and the injection volume was 5 ␮L for each sample. The samples were kept at 4 ◦ C in an autosampler during the analysis. An LTQ Orbitrap Velos ProTM mass spectrometer equipped with HESI in negative ionization mode was used for the detection of BAs. The HESI parameters were as follows: sheath gas flow rate, 40 arb; auxiliary gas flow rate, 10 arb; sweep gas flow rate, 1 arb; source voltage, 4.80 kV; capillary temperature, 350 ◦ C; S–lens RF level, 67.9%; heater temperature, 40 ◦ C. The MS mode was optimized as follows: for the MS full scan mode, resolution, 60,000; AGC target, 1E06; scan range, 110–2000 m/z. For MS/MS mode, resolution, 17,500; AGC target, 2E04; scan range, relative to parent mass.

2.4. Preparation of stock solution, calibration standards and quality control samples Stock solutions (1 mg/mL) of BAs and internal standards (ISTDs) containing d4 -CA, d4 -CDCA, and d4 -LCA were prepared in methanol. For the blank matrix, bile samples diluted in deionized water (25-fold) were incubated with 100 mg/mL of activated charcoal for 1 h to strip the matrix of endogenous bile acids. Then, the mixtures were centrifuged at 14,000 rpm for 10 min, and the supernatants were incubated once more with activated charcoal to repeat the stripping process. After the second centrifugation, the supernatants were filtered, and the filtrate was used for calibration and validation. The working solutions of the ISTDs were 1 ␮g/mL in methanol and were prepared from the stock solutions (Table 1). All stock solutions and working solutions were stored at −20 ◦ C before use. Calibration solutions were prepared by serial dilution of the working solutions with methanol and were subsequently spiked into rat bile. The final calibration ranges for each analyte are summarized in Table 1. Based on the calibration ranges in Table 1, the concentration ranges for the quality control (QC) samples (low, medium, and high concentrations) are summarized in Table 1.

2.5. Sample preparation procedure Solid-phase extraction (SPE) using Oasis-HLB SPE cartridges was used for sample extraction. Bile samples were diluted 25-fold and 500-fold with deionized water, 100 ␮L of diluted bile was spiked with 10 ␮L of ISTDs (1 ␮g/mL), vortexed, and loaded onto the cartridges. SPE cartridges were pre-conditioned with 1 mL of methanol and 1 mL of water serially. Loaded cartridges were washed with 2 mL of water and eluted with 2 mL of methanol. The eluate was evaporated under a gentle stream of nitrogen gas using a TurboVap LV (Caliper Life Sciences Inc., Hopkinton, MA, USA). The residue was reconstituted with 100 ␮L of 50% methanol and vortexed for 30 s. Prepared samples were transferred into a vial, and 5 ␮L of the reconstituted solution was injected into the UPLC-LTQ–Orbitrap MS.

40

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

Table 2 Extraction recoveries and matrix effects of BAs at each QC concentration in rat bile. Data are presented using the average of each QC concentrations (n = 5). Recovery (%)

Matrix effect (%)

LLOQ

Low

Medium

High

Low

Medium

High

Unconjugated CA ␣-MCA ␤-MCA ␻-MCA UDCA CDCA DCA LCA

95.68 95.77 99.23 97.92 88.49 99.39 90.98 71.66

98.41 99.84 100.5 98.51 99.76 92.53 89.59 64.3

98.65 99.6 99.77 99.87 98.59 94.13 92.48 82.8

95.77 96.15 95.44 95.4 95.24 91.53 90.53 79.24

106.6 104.9 106.4 107 105.1 108.9 102.8 111.4

104.2 105.7 105.8 102.1 109 107.9 105.5 114.7

99.39 99.6 100.9 98.4 94.91 94.46 96.46 96.01

Glycine-conjugates GCA GUDCA GCDCA GDCA GLCA

93.46 97.9 90.06 89.16 74.18

96.35 99.29 91.81 89.26 68.62

99.26 99.16 95.63 94.96 79.56

95.17 96.57 93.96 93.19 83

104 104.9 104.6 105.5 108.7

106.9 103.5 104.2 105.2 109.9

101.5 100.9 101.4 97.44 95.06

Taurine-conjugates T-MCA(␣+␤) TCA TUDCA TCDCA TDCA TLCA

95.62 99.89 97.83 88.55 89.19 75.71

98.87 99.53 99.47 91.18 89.67 71.72

98.85 98.8 99.6 95.12 95.03 83.52

96.59 94.68 95.26 94.91 92.43 81.58

103.5 106.4 107.2 108.8 106.4 112.5

105.5 104.8 104.1 104.6 102.6 101.6

100.2 101.3 100.5 94.44 102 97.74

2.6. Validation of the analytical method

2.6.2. Precision and accuracy The precisions and accuracies of the intra- and inter-day analyses were evaluated using solutions with relative LLOQ values and QC samples. The intra- and inter-day variations were determined using 5 replicates of spiked samples on the same day and on 5 different days. The precision was determined using the coefficient of variation (CV, %) calculated from the ratio of the relative standard deviation to the mean of the measured analyte concentration, and the accuracy was calculated from the% bias [(measured − theoretical)/measured concentrations, %]. The acceptable range of precision and accuracy for the intra- and inter-

2.6.1. Calibration curves A 100 ␮L volume of the blank matrix was spiked with calibration standard containing 19 targeted BAs and 10 ␮L of ISTDs to construct a calibration curve with at least 6 calibration points based on the appropriate range of each BA. Calibrations were analyzed by the best weighting factors between 1/x and 1/x2 (x is the concentration of each analyte) using linear or quadratic regression of the peak area ratio of the analyte to the IS versus nominal concentrations.

Table 3 The inter- and intra-day accuracy and precision for BAs in rat bile. Intra-day (n = 5) LLOQ

Inter-day (n = 5) Low

Medium

High

LLOQ

Low

Medium

High

Accuracy Precision Accuracy Precision Accuracy Precision Accuracy Precision Accuracy Precision Accuracy Precision Accuracy Precision Accuracy Precision (%) (CV%) (%) (CV%) (%) (CV%) (%) (CV%) (%) (CV%) (%) (CV%) (%) (CV%) (%) (CV%) Unconjugated 101.65 CA 100.77 ␣-MCA 99.98 ␤-MCA 99.48 ␻-MCA 98.93 UDCA 106.76 CDCA 101.22 DCA 97.22 LCA

6.65 1.72 2.12 2.18 2.95 1.56 4.21 3.88

92.20 98.67 99.88 99.59 99.5 100.10 102.42 99.32

2.90 3.09 1.46 0.68 2.02 4.28 1.89 3.43

100.00 101.19 96.79 100.87 104.61 99.82 99.31 100.33

2.69 1.44 2.24 3.14 2.17 1.95 3.56 2.96

100.05 97.04 94.75 99.58 102.26 99.58 100.21 102.73

0.89 0.97 1.53 1.51 1.71 1.06 0.97 0.89

100.54 100.07 100.02 99.92 96.01 102.67 100.63 101.14

1.18 0.55 1.56 5.8 4.3 2.03 3.02 3.61

90.82 100.02 100.11 100.91 100.05 100.64 97.22 99.68

3.42 2.77 3.39 0.93 2.11 3.01 5.81 1.89

100.45 100.48 100.21 100.62 108.29 103.50 100.73 100.38

1.61 2.83 2.39 1.58 4.52 3.81 2.37 2.20

100.00 99.43 95.15 100.48 100.37 100.01 100.04 101.96

1.34 3.63 1.72 3.07 0.71 1.24 3.47 2.98

Glycine-conjugates 93.29 GCA 95.82 GUDCA 91.27 GCDCA 92.51 GDCA 100.34 GLCA

1.39 3.57 2.41 0.97 3.16

97.11 103.38 102.27 101.30 96.46

3.20 2.78 4.35 7.22 6.74

100.55 99.02 95.14 98.25 102.39

1.16 1.08 1.91 2.92 3.44

103.40 100.15 99.8 98.22 100.79

0.66 2.63 4.44 2.40 4.03

100.61 97.47 90.8 90.26 98.91

5.14 2.20 5.20 2.56 3.91

98.20 101.45 105.48 106.28 97.78

3.86 2.67 5.82 3.93 8.16

106.55 98.59 96.45 97.81 101.90

5.04 3.63 5.83 5.72 4.80

104.90 99.43 100.20 98.56 101.45

1.70 1.20 5.14 2.57 5.32

Taurine-conjugates 99.30 TMCA 100.98 TCA 90.24 TUDCA 91.60 TCDCA 99.67 TDCA 98.69 TLCA

0.40 1.12 4.69 2.13 2.19 3.78

101.22 99.70 99.35 98.79 99.54 99.82

2.72 1.92 1.51 2.91 0.56 1.49

99.70 99.85 98.72 100.59 97.69 102.37

1.69 0.86 4.05 3.79 2.62 7.56

103.55 110.43 97.23 99.08 98.60 97.45

0.90 1.94 2.29 1.89 2.73 5.52

100.12 100.54 95.84 92.26 93.10 96.67

0.85 3.41 1.95 5.59 5.55 6.45

101.73 100.29 100.53 100.41 100.08 98.48

2.55 4.11 3.01 3.42 2.11 1.24

101.94 101.28 100.69 101.69 100.01 101.44

3.88 5.47 6.99 4.54 1.77 9.02

103.28 108.14 98.52 99.74 99.37 100.47

3.80 1.94 1.08 1.73 1.97 3.29

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

day variation is within ±15% (LLOQ: ±20%) according to the FDA guidelines [28]. 2.6.3. Recovery and matrix effect Low, medium, and high concentrations of QC samples were used to determine the recovery and matrix effects from five replicates. The recovery was evaluated by comparing the peak areas of the analytes obtained from the samples spiked into charcoal-treated bile, pre- and post-extraction. The matrix effect was determined by comparing the post-extraction spiked bile and the standard solution of the same concentration. 2.6.4. Sample stability Sample stability was evaluated under 5 different conditions: short- and long-term stability, autosampler stability, freeze-thaw stability and reconstitution stability. All stability tests were performed by analyzing five replicates of low, medium and high concentrations of the QC samples. The short-term stability and the long-term stability tests were carried out by keeping the samples at room temperature for 3 h and −80 ◦ C for 4 weeks. The autosampler stability test was evaluated comparing the prepared samples before and after keeping them in the LC autosampler (4 ◦ C) for 24 h. The freeze-thaw stability test was carried out by comparing the samples before and after the freeze-thaw cycle (frozen at −80 ◦ C and then thawed completely at room temperature 3 times). For the reconstitution stability test, the stability was compared between samples that were reconstituted immediately and samples that were stored at room temperature for 24 h before reconstitution. The stability was calculated as the ratio of the measured analyte concentration to the initial analyte concentration. The samples were considered stable when the test values were within the acceptable limit of accuracy of ±15%. 2.7. Animal study Male Sprague-Dawley rats were purchased from Orient Bio Inc. (Seongnam, Korea). They were separated into 3 groups (n = 5) and housed separately for 6 weeks, 6 months, and 15 months. All animals were housed in a light-, temperature-, and humiditycontrolled environment. All animal experiments were approved by the Institutional Animal Ethics committee of the Korea Institute of Science and Technology, Korea. After housing for the respective period, the bile duct was cannulated with PE-10 tubing and bile was collected through the cannula for 2 h [29]. The samples were stored at −80 ◦ C until use.

41

2.8. Data processing and statistical study Xcalibur software was used for quantification of the 19 targeted BAs. Quantification was performed using the MS scan data and extracting the ion chromatogram of each m/z value. For the identification of unknown BAs, peak extraction and alignment were performed using SIEVETM software (Thermo Scientific, San Jose, CA, USA). Statistical significance of the differences in BAs was evaluated by the Kruskal-Wallis test for three-way comparison and the Wilcoxon rank-sum test for pairwise comparison using MetaboAnalyst (www.metaboanalyst.ca), a web-based program. For the comparison of the change in unknown BAs with age, peak intensity was used instead of concentration. P-values less than 0.05 were considered statistically significant for group differences. 3. Results and discussion 3.1. Method development 3.1.1. Sample preparation Each BA has a large variation and diverse concentration range in bile. Therefore, the preparation of bile for simultaneous analysis of BAs is difficult to optimize with respect to the calibration range and the dilution factor (Table 1). In the method presented here, dilution factors were optimized into two groups according to each concentration of bile acid to obtain accurate quantification. A 500-fold dilution was used to quantify CA, TMCA, TCA, TUDCA, TCDCA, TDCA, and GCA, and a 25fold dilution was used for the others. The 19 targeted bile acids were quantified simultaneously using the above factors. For the sample preparation, protein-precipitation and solid-phase extraction (SPE) were investigated to increase the extraction recovery. Among those tests, good recovery was obtained using SPE with Oasis-HLB SPE cartridge (Table 2), it was used for the following sample preparations. To make a stripped matrix for the calibration curve and validation of the assay method, charcoal treatment was performed with different incubation times and different numbers of incubations. Finally, blank bile was prepared after treatment with 100 mg/mL activated charcoal for 1 h, twice. 3.1.2. UPLC-LTQ–Orbitrap MS method The columns (ACQUITY UPLC® T3 and ACQUITY UPLC® BEH C18 ), mobile phase (methanol- and acetonitrile-based) and various gradient conditions were tested to obtain better peak separation. A C18

Fig. 2. Chromatograms of 19 BAs and IS (1 ␮g/mL). 8 unconjugated BAs (shown in black), 5 glycine-conjugated BAs (shown in green), 6 taurine-conjugated BAs (shown in blue), and 3 deuterium-labeled standard BAs (shown in red). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).

42

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

Fig. 3. Representative chromatograms of BAs in rat bile. (A) TMCA, TUDCA, TCA, GCA, TCDCA, TDCA, CA and 13 unknown BAs peaks in 500-fold diluted bile sample; (B) ␣MCA, ␤MCA, ␻MCA, UDCA, CDCA, DCA, GUDCA, GCDCA, GDCA, GLCA, TLCA and 9 unknown BAs peaks in 25-fold diluted bile sample (LCA was not detected in rat bile samples).

column and a mobile phase system of 0.1% formic acid in methanol with a gradient were used as the optimized method, which generated well-separated peaks with a sharp peak shape (Fig. 2) for all of the unconjugated, glycine-conjugated, and taurine-conjugated BAs and even the untargeted BAs from rat bile (Fig. 3). Importantly, the baseline separation of the 3 isomers of MCA (␣, ␤, ␻) was successfully achieved in this method, which was difficult in previous reports [18,21] that used similar run times (22–25 min). Because the 3 isomers of MCA have individual functions and different toxicities, it is important to quantify them accurately. The MS parameters, including sheath gas, auxiliary gas, and sweep gas flows, spray voltage, and capillary temperature were

optimized by direct infusion of the BAs mixture. The adduct ions of each BA used for quantification are summarized in Table 1. When the total ion chromatograms for the bile samples were extracted with the adduct ions in Table 1, several unknown peaks were detected in small amounts (Fig. 3). 3.2. Method validation 3.2.1. Calibration curve In this study, the calibration ranges were investigated in detail because there are various concentrations of BA present in the bile tested. The calibration range was determined based on the sample intensity so that it is able to fit within the range (Table 1). For

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

43

Fig. 4. Boxplot of targeted 19 major BAs in rat bile (6W; 6weeks, 6 M; 6months, 15 M; 15months). The bile acids are divided into unconjugated BAs (A), glycine-conjugated BAs (B), and taurine-conjugated BAs (C). †: p < 0.05 for three way comparison (Kruskal-Wallis test) *: p < 0.05 for pair-wise comparison (Wilcoxon rank-sum test)

44

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

Table 4 The stability (%) test of BAs. 4weeks (−80 ◦ C)

3 h (room temp.) Low

Freeze-thaw (3 cylcles)

Autosampler (4 ◦ C)

Reconstitute

Medium

High

Low

Medium

High

Low

Medium

High

Low

Medium

High

Low

Medium

High

Unconjugated 99.36 CA ␣-MCA 99.47 ␤-MCA 99.74 ␻-MCA 96.97 UDCA 100.12 100.3 CDCA 95.55 DCA 100.95 LCA

95.93 95.71 97.16 95.95 94.68 96.14 94.72 100.26

98.05 97.8 96.02 94.71 100.18 98.06 100.08 97.77

100.55 100.13 100.7 100.53 100.03 100.24 100.11 99.99

98.09 97.74 97.46 97.98 98.28 99.43 100.13 100.94

96.81 96.5 97.33 98 95.58 97.06 96.36 97.6

91.44 97.16 98.1 97.52 100.58 97.02 92.96 93.72

98.09 95.69 93.53 92.14 95.96 95.94 98.02 98.89

95.7 95.69 95.01 94.72 93.03 95.01 96.35 92.35

100.44 102.03 100.64 99.24 100.93 102.53 98.43 101.94

100.53 100.02 98.72 98.42 100.1 100.3 100.53 102.91

99.61 98.25 97.79 96.8 100.68 100.4 100.14 98.9

95.69 98.02 97.54 98.51 100.1 94.46 95.38 95.69

96.49 95.43 94.61 94.24 97.57 97.03 98.03 99.77

98.12 95.86 96.6 95.53 96.41 97.86 96.49 100.55

Glycine-conjugates 99.56 GCA GUDCA 98.32 GCDCA 97.94 GDCA 99.8 GLCA 93.58

95.75 96.67 97.74 97.85 94.53

95.17 100.98 100.45 100.84 99.33

100.73 100.57 100.66 100.58 97.75

98.52 98.09 98.83 99.46 96.08

94.1 99 98.22 98.99 99.6

96.46 105.33 103.66 101.17 102.97

95.63 97.74 98.83 96.59 100.75

93.36 97.31 95.33 97.5 96.42

100.9 100.15 100.63 100.38 98.44

98.43 100.74 100.27 100.14 100.05

97.05 100.51 100.32 100.56 99.09

97.63 100.67 100.33 100.45 103.73

94.32 99.29 100.49 100.03 100.9

92.85 97.42 96.63 98.27 101.08

Taurine-conjugates 98.69 TMCA 97.86 TCA 98.45 TUDCA TCDCA 98.44 98.41 TDCA 94.24 TLCA

94.14 94.95 96.7 97.45 97.81 99.61

95.06 95.19 99.47 100.67 101.58 102.14

98.33 97.68 97.44 100.97 100.37 100.61

98.99 99.12 99.55 99.21 99.37 94.16

97.39 97.2 99.68 100.26 98.74 100.88

99.08 100.56 108.41 103.63 104.98 116.84

95.31 93.89 100.09 99.14 99.88 98.19

93.98 96.33 98.39 98.9 98.74 92.71

101.25 100.56 100.01 100.41 100.62 100.22

98.93 98.53 100.52 100.84 100.23 101.67

97.75 98.84 99.18 100.38 101.12 101.46

99.55 97.76 100.53 100.35 101.1 106.08

93.98 95.78 99.33 100.65 100.83 99.41

94.21 95.1 96.56 97.46 97.3 96.5

Table 5 Age-related concentration (median) changes and p-values of 19 targeted BAs in rat bile. Median (␮g/mL)

p-value

trend

6weeks

6months

15months

3group†

6 W vs. 6M*

6 W vs. 15M*

6 M vs. 15M*

Unconjugated BAs

␣MCA ␤MCA ␻MCA CA UDCA CDCA DCA Total

12.02 27.41 7.37 215.34 0.21 0.24 0.23 261.00

1.82 5.75 0.77 70.37 0.18 0.15 0.09 78.72

0.11 0.24 0.07 1.74 0.02 0.05 0.03 2.25

0.013 0.011 0.006 0.013 0.015 0.021 0.015 0.013

0.151 0.222 0.032 0.310 0.222 0.222 0.421 0.310

0.016 0.016 0.016 0.016 0.016 0.016 0.016 0.016

0.032 0.016 0.032 0.016 0.032 0.063 0.016 0.016

↓ ↓ ↓ ↓ ↓ ↓ ↓ ↓

Glycine-conjugated BAs

GCA GUDCA GCDCA GDCA GLCA Total

934.87 10.27 65.14 60.89 0.32 1069.76

871.28 6.77 51.55 17.93 0.03 1001.01

616.29 4.31 22.85 3.23 0.03 646.31

0.139 0.295 0.118 0.011 0.034 0.105

0.548 0.310 0.310 0.095 0.032 0.310

0.063 0.286 0.032 0.016 0.032 0.063

0.286 0.413 0.730 0.032 0.905 0.286

– – – ↓ ↓ –

Taurine-conjugated BAs

TMCA TCA TUDCA TCDCA TDCA TLCA Total

746.27 4721.68 719.61 490.02 399.47 3.78 6234.73 7951.96

1965.38 7686.34 729.93 1108.80 576.86 1.58 13204.93 14248.16

2108.53 7280.33 1821.22 1423.36 387.93 3.10 15439.75 16273.94

0.058 0.010 0.221 0.075 0.131 0.150 0.014 0.020

0.095 0.008 0.310 0.095 0.222 0.222 0.016 0.020

0.032 0.016 0.111 0.063 0.730 1 0.016 0.016

0.730 0.556 0.286 0.556 0.063 0.063 0.556 0.556

↑ ↑ – – – – ↑ ↑

total BAs

†p values for three way comparison (Kruskal-wallis test); *p values for pair-wise comparison (Wilcoxon rank-sum test). Bold p values indicate significance p < 0.05.

an instance, the concentration of CA in a sample ranges from 1 to 600 ␮g/mL, but UDCA ranges from 20 to 500 ng/mL. The best weighting factor (1/x2 or 1/x, where x is the concentration of the analytes) was selected to obtain a well-fitted regression model for the construction of the calibration curve for each BA’s accurate quantification. The correlation coefficients (r2 ) for all BAs were higher than 0.99 (Table 1). 3.2.2. Extraction recovery and matrix effect Table 2 shows the extraction recovery and matrix effect of the 19 bile acids. The range of extraction recovery of all analytes was 88.5–100.5%, except LCA and its conjugates. Particularly, LCA had low recovery (64.3–82.8%) but did not seriously affect the accuracy

or precision of the analysis. There was no significant matrix effect in the detection of bile acids in rat bile (94.4–114.7%).

3.2.3. Accuracy and precision The accuracy and precision of the intra- and inter-day measurements are listed in Table 3. The intra-day and inter-day variations in the accuracy were 96.3–111.6% and 93.4–105.4%, respectively, and the variations in the precision were 1.5–10.6% and 4.5–13.0%, respectively. All results were acceptable according to the FDA guidelines [28].

Table 6 Characterization of unconjugated and glycine-conjugated unknown BAs by MS/MS fragmentation. RT (min) Molecular formula

unconjugated BA 1 BA 2 BA 3 BA 4 BA 5 BA 6 BA 7 Glycineconjugates

G-BA 1 G-BA 2 G-BA 3 G-BA 4 G-BA 5 G-BA 6

Parent ion

MS/MS fragment m/z (relative intensity, %)

predicted [M−H]− m/z

measured [M−H]− m/z

[M−H]−

[M–H–H2 O]− [M–H–H2 O–H2 ]−

[M–H–2H2 O]−

[M–CO2 ]− [M–H2 CO2 ]−

[M–H–CO2 –H2 O]− [M–H–H2 CO2 –H2 O]−

[M–H–CO2 –2H2 O]− Other [M–H–H2 CO2 –2H2 O]−

10.59 11.71 12.22 12.60 14.11 16.10 16.99

C24 H40 O5 C24 H40 O5 C24 H40 O5 C24 H40 O5 C24 H40 O5 C24 H40 O4 C24 H40 O4

407.280 407.280 407.280 407.280 407.280 391.285 391.285

407.281 407.277 407.278 407.279 407.279 391.283 391.283

– – – – – 391 (4) –

389(75) 387 (49) 389 (5) 389 (44) 387 (11) 389 (32) 387 (41) 389 (100) 373 (22) –

371 (41) 371 (20) 371 (17) 371 (9) – 355 (35) 355 (15)

361 (10) 363 (14) 363 (9) – – 347 (59) 345 (100) 345 (100)

345 (100) 343 (20) 345 (100) 343 (49) 345 (100) 343 (7) – 345 (15) 343 (15) 329 (33) 327 (26) –

327 (12) 327 (44) 325 (27) 327 (8) – – – –

9.88 10.51 10.76 13.02 13.15 14.18

C26 H43 NO6 C26 H43 NO6 C26 H43 NO6 C26 H43 NO6 C26 H43 NO5 C26 H43 NO5

464.302 464.302 464.302 464.302 448.307 448.307

464.300 464.300 464.301 464.300 448.306 448.306

– 464 (13) – –

446 (20) 446 (5) 446 (11) 446 (12) 430 (16) 430 (4)

– – – 420 (100) 404 (100) 404 (100)

420 (100) 418 (11) 418 (100) 418 (100) 418 (11) 402 (51)

402 (21) 400 (5) 400 (14) 400 (20) –

384 (5) 384 (4) 384 (14) –

353 (12) 405 (12) 347 (100) 329 (11) – 365 (8) – – – –

386 (89)

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

Table 7 Characterization of taurine-conjugated unknown BAs by MS/MS fragmentation. RT (min)

Molecular formula

Parent ion

MS/MS fragments m/z(relative intensity, %)

predicted [M−H]− m/z

measured [M−H]− m/z

[M−H]−

T-BA 1 T-BA 2 T-BA 3 T-BA 4

9.93 10.05 10.56 12.56

C26 H45 NO7 S C26 H45 NO7 S C26 H45 NO7 S C26 H45 NO7 S

514.2844 514.2844 514.2844 514.2844

514.2823 514.2819 514.2822 514.2820

514 (100) 514 (100) 514 (100) 514 (100)

[M–H–O]− [M–H–H2 O]− 496 (4.15) 496 (6) 496 (3) 496 (7.7)

[M–H–2O2 ]− [M–H–H2 O–O2 ]−

[M–H–SO3 –H2 ]− [M–H–SO3 –2H2 ]− 430 (1.4) 430 (0.8) 432 (0.6) 430 (1.3) 432 (1.0) 432 (2.1)

[M–H–SO3 –H2 –H2 O]− [M–H–SO3 –H2 H2 O– [M–H–SO3 –2H2 –H2 O]− C2 H5 N]− 414 (1.0) 412 (1.7) 371 (0.8) 414 (0.9) 412 (0.8) – 414 (0.7) 412 (0.5) 371 (0.3) – –

[M–H–SO3 –H2 –2H2 O– C2 H5 N]− – 353 (0.4) – 353 (2.2)

448 (1.9) – 448 (0.4) –

T-BA 5 T-BA 6

14.20 10.16

C26 H45 NO7 S C26 H45 NO6 S

514.2844 498.2895

514.2820 498.2875

514 (100) 498 (100)

498 (54.0) 496 (9.3) 480 (1.9)

446 (8.2) 434 (1.9) 432(4.0)

– 416 (1.2) 414 (1.0)

– –

– –

– –

T-BA 7

12.04

C26 H45 NO6 S

498.2895

498.2889

498 (100)

–

434 (17.3)

416 (10.3)

–

–

–

T-BA 8

12.55

C26 H45 NO6 S

498.2895

498.2878

498 (100)

–

434 (5.8) 432 (4.0)

416 (2.9) 414 (2.3)

T-BA 9

18.60

C26 H45 NO6 S

498.2895

498.2878

498 (82)

–

434 (30.2)

–

–

–

–

Others 342 (15.8) 342 (4.6) – 390 (26.7) 372 (23.0) – 372 (1.5) 386 (1.4) 370 (16.0) 388 (32.5) 406 (89.2) 462 (7.2) 406 (27) 388 (5.4) 370 (3.2) 406 100) 370 (20.2)

45

46

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

Fig. 5. The change of bile acid composition in young (A), and aged (B) rat bile. When compared to the young rats, the taurine-conjugated BAs were increased and unconjugated BAs were decreased among the primary bile acids of the aged rats. For the secondary bile acids, glycine-conjugated BAs and unconjugated BAs were both decreased in the aged rats.

3.2.4. Stability Stability tests were evaluated by short-term stability, long-term stability, autosampler stability, freeze-thaw stability, and reconstitution stability. All results were within ±15%, indicating that all BAs were stable at room temperature for 3 h and for 4 weeks at −80 ◦ C. In the freeze-thaw cycle test, all analytes were stable after three cycles, except TLCA (116.84%) at low concentration (45 ng/mL). However, because the concentration of TLCA in the tested bile was greater than 700 ng/mL, the result of the freeze-thaw stability test for TLCA at lower concentration is not significant in this study. Other QC samples of different concentrations were stable, within ±15%. The autosampler stability and the stability of the reconstitution were also in an acceptable range (Table 4). 3.3. Analysis of BAs in rat bile 3.3.1. Changes in the concentrations of 19 targeted BAs during aging The new method was used to analyze the BAs in rat bile that had been collected from 6-week-, 6-month-, and 15-month-old rats. The concentrations of age-related rat bile are shown using a boxplot

in Fig. 4. Taurine-conjugated BAs were the major forms in the rat bile, composing more than 80% of the total BAs pool. Although the total BA concentrations increased with age, different patterns of concentration changes in the three classes of BAs were observed. Unconjugated BAs were present in a very small amount, except CA (LCA was not detected), and significantly decreased with age. In a statistical analysis, ␻MCA showed a statistically significant decrease in a three-way comparison and pairwise comparison of the 3 different age groups, whereas all other unconjugated BAs showed only a statistically significant decrease in the three-way comparison (Fig. 4a). A significant difference between the primary (CA, CDCA, ␣MCA, ␤MCA) and secondary (DCA, UDCA, ␻MCA, LCA) BAs was not found. For conjugated BAs, except for GCA, the glycine-conjugated BAs were found in a small amount in rat bile. Secondary BAs conjugated to glycine (GDCA and GLCA) showed a clear decreasing trend with age, which was statistically significant. Other glycine-conjugated BAs showed negligible change (Fig. 4b). Unlike the glycine-conjugated BAs, the taurine-conjugated primary BAs showed an increasing trend in general. TCA was the most signif-

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

47

Fig. 6. MS/MS spectrum and prediction of fragmentation of Unconjugated BA (A), Glycine-conjugated BA (B), and Taurine-conjugated BA (C).

icantly increased primary BA. The secondary BAs conjugated to taurine showed no significant changes (Fig. 4c). The median of each

BA in the 3 groups and the p-values of the three-way and pair-wise comparison are summarized in Table 5.

48

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

Fig. 7. Boxplot of 6 unknown BAs in aging rat bile that were significantly altered (6W; 6weeks, 6 M; 6months, 15 M; 15months). †: p < 0.05 for three way comparison (Kruskal-Wallis test) *: p < 0.05 for pair-wise comparison (Wilcoxon rank-sum test)

A notable alteration of the BA compositions in the bile of aging rats is that as the rats age, the overall level of unconjugated BAs and glycine-conjugated secondary BAs decreased, while the taurineconjugated primary BAs increased (Fig. 5). The gut microbiome is known to regulate BAs through functions including deconjugation, oxidation, and dihydroxylation of BAs [30,31]. The microbiome regulates BAs through a positive feedback antagonism of FXR in the liver and intestine, resulting in maintaining the hydrophilic BAs and agonism of FXR, resulting in unconjugated hydrophobic BA [30]. The microbiome has functions including deconjugation, oxidation, and dihydroxylation of BAs [31]. These functions are important to maintain the balance of hydrophilicity of the BA pool, which may be associated with disease states. For example, it was demonstrated that inflammatory bowel disease leads to decreased BA deconjugation and desulfation activities, which lead to modification in the BA composition and may contribute to chronic inflammation [32]. Furthermore, alteration of the gut microbiota is known to have an effect on health via BA composition change [33]. Therefore, we assumed that dysbiosis of the gut microbiota and an altered BA pool were mutually related. A decrease in the number and species diversity of many beneficial or protective anaerobes, such as bacteroides and bifidobacteria, decreases the functionality of microflora in the elderly [34]. Particularly, bacteroides and bifidobacteria are representative bacterial genera of the gut microbiota that have a function in the deconjugation reaction for BAs [31]. Several strains of bacteroides revealed much higher activity against taurine than glycine conjugates [35]. Taurine-conjugated BAs were predominant in rat bile (over 80%); therefore, taurine conjugates were significantly affected by the changes in intestinal microflora. During aging, due to the lack of bacteroides activity, the primary taurine-conjugated BAs increased in concentration, while the glycine-conjugated primary BAs were unaffected. Although the secondary conjugated BAs should have similar trends as the primary conjugated BAs, the results were unexpected. The secondary taurine-conjugated BAs were not changed, while the secondary glycine-conjugated BAs were significantly decreased in concentration (Fig. 5). This may be due to the effect of the intestinal microflora change in the aged rats. We believe that the decrease in the total

number of species of microflora affected the dehydroxylation and epimerization of the primary BAs into secondary BAs [36]. 3.3.2. Untargeted detection and characterization of potential unknown BAs In addition to 19 targeted bile acids, 22 unknown BAs were detected in the MS scan data (Fig. 3) and were characterized using the MS/MS fragmentation data. The fragmentation patterns were specific to those of BAs, and using the exact mass values, we identified the conjugates and designated the BAs accordingly (Fig. 6). The fragmentation patterns of the [M−H]− ions for the BAs are shown in Fig. 5. Glycine-conjugated and unconjugated BAs showed typical fragmentation of a sequential loss of H2 O, CO2 , H2 CO2 or H2 from the parent structure (Table 6), similar to what has been reported by others [17]. Taurineconjugated BAs also had typical fragmentation patterns, including [M−H]− , [M–H–H2 O]− , [M–H–H2 O–O2 ]− , [M–H–SO3 –H2 ]− , [M–H–SO3 –H2 –H2 O]− , [M–H–SO3 –H2 –H2 O–C2 H5 N]− , and [M–H–SO3 –H2 –2H2 O–C2 H5 N]− . Fig. 5 illustrates the fragmentation pattern of taurocholic acid, a taurine-conjugated BA. The newly designated taurine-conjugated BAs are summarized in Table 7. Among the newly designated BAs, six showed significant changes, including a conjugate with glycine and 5 conjugates of taurine (Fig. 7). The glycine-conjugated BA (G-BA 4) decreased, while the 5 taurine-conjugated BAs (T-BA 1, 3, 4, 5, 9) increased significantly in the bile of aged rats. 4. Conclusions In this study, we developed a UPLC-LTQ–Orbitrap-MS method that simultaneously performs quantitative and qualitative analysis of 41 BAs. Through this method, we successfully quantified 19 major BAs and identified 22 unknown BAs. Furthermore, this method was successfully applied to an age-related rat bile sample. We identified clinically important BA biomarkers of aging through quantitative analysis of 19 targeted BAs. Through the identification of 22 unknown BAs, their change in the rat bile supported the result of targeted BAs. We believe that the identified unknown BAs in rat bile would have important functions and they are worth considering as the subjects of the further study. Through this simple,

G. Lee et al. / J. Chromatogr. B 1031 (2016) 37–49

sensitive, accurate method, investigators can profile most BAs in biological samples simultaneously. Furthermore, clinically significant BAs related to various diseases, including gastrointestinal and liver disease, may be identified through this method.

Acknowledgements This study was supported by the Creative Fusion Research Program through the Creative Allied Project funded by the Korea Research Council of Fundamental Science and Technology (CAP-12-1-KIST) and the Bio-Synergy Research Project (NRF2013M3A9C4078145) of the Ministry of Science, ICT and Future planning through the National Research Foundation.

References [1] C. Thomas, R. Pellicciari, M. Pruzanski, J. Auwerx, K. Schoonjans, Targeting bile-acid signalling for metabolic diseases, Nat. Rev. 7 (2008) 678–693. [2] H. Wang, J. Chen, K. Hollister, L.C. Sowers, B.M. Forman, Endogenous bile acids are ligands for the nuclear receptor FXR/BAR, Mol. Cell. 3 (1999) 543–553. [3] M. Makishima, A.Y. Okamoto, J.J. Repa, H. Tu, R.M. Learned, A. Luk, M.V. Hull, K.D. Lustig, D.J. Mangelsdorf, B. Shan, Identification of a nuclear receptor for bile acids, Science 284 (1999) 1362–1365. [4] D.J. Parks, S.G. Blanchard, R.K. Bledsoe, G. Chandra, T.G. Consler, S.A. Kliewer, J.B. Stimmel, T.M. Willson, A.M. Zavacki, D.D. Moore, J.M. Lehmann, Bile acids: natural ligands for an orphan nuclear receptor, Science 284 (1999) 1365–1368. [5] Y. Kawamata, R. Fujii, M. Hosoya, M. Harada, H. Yoshida, M. Miwa, S. Fukusumi, Y. Habata, T. Itoh, Y. Shintani, S. Hinuma, Y. Fujisawa, M. Fujino, A G protein-coupled receptor responsive to bile acids, J. Biol. Chem. 278 (2003) 9435–9440. [6] M. Makishima, T.T. Lu, W. Xie, G.K. Whitfield, H. Domoto, R.M. Evans, M.R. Haussler, D.J. Mangelsdorf, Vitamin D receptor as an intestinal bile acid sensor, Science 296 (2002) 1313–1316. [7] J. He, S. Nishida, M. Xu, M. Makishima, W. Xie, PXR prevents cholesterol gallstone disease by regulating biosynthesis and transport of bile salts, Gastroenterology 140 (2011) 2095–2106. [8] A.F. Hofmann, Detoxification of lithocholic acid, a toxic bile acid: relevance to drug hepatotoxicity, Drug Metab. Rev. 36 (2004) 703–722. [9] B.S. Reddy, K. Watanabe, J.H. Weisburger, E.L. Wynder, Promoting effect of bile acids in colon carcinogenesis in germ-free and conventional F344 rats, Cancer Res. 37 (1977) 3238–3242. [10] B. Alme, A. Bremmelgaard, J. Sjovall, P. Thomassen, Analysis of metabolic profiles of bile acids in urine using a lipophilic anion exchanger and computerized gas-liquid chromatorgaphy–mass spectrometry, J. Lipid Res. 18 (1977) 339–362. [11] A. Roda, C. Cerre, P. Simoni, C. Polimeni, C. Vaccari, A. Pistillo, Determination of free and amidated bile acids by high-performance liquid chromatography with evaporative light-scattering mass detection, J. Lipid Res. 33 (1992) 1393–1402. [12] S.S. Rossi, J.L. Converse, A.F. Hofmann, High pressure liquid chromatographic analysis of conjugated bile acids in human bile: simultaneous resolution of sulfated and unsulfated lithocholyl amidates and the common conjugated bile acids, J. Lipid Res. 28 (1987) 589–595. [13] J. Goto, M. Saito, T. Chikai, N. Goto, T. Nambara, Studies on steroids. CLXXXVII. Determination of serum bile acids by high-performance liquid chromatography with fluorescence labeling, J. Chromatogr. 276 (1983) 289–300. [14] J. Huang, S.P. Bathena, I.L. Csanaky, Y. Alnouti, Simultaneous characterization of bile acids and their sulfate metabolites in mouse liver plasma, bile, and urine using LC–MS/MS, J. Pharm. Biomed. 55 (2011) 1111–1119. [15] Y. Alnouti, I.L. Csanaky, C.D. Klaassen, Quantitative-profiling of bile acids and their conjugates in mouse liver bile, plasma, and urine using LC–MS/MS, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 873 (2008) 209–217.

49

[16] M.H. Sarafian, M.R. Lewis, A. Pechlivanis, S. Ralphs, M.J. McPhail, V.C. Patel, M.E. Dumas, E. Holmes, J.K. Nicholson, Bile acid profiling and quantification in biofluids using ultra-performance liquid chromatography tandem mass spectrometry, Anal. Chem. 87 (2015) 9662–9670. [17] S.E. Jantti, M. Kivilompolo, L. Ohrnberg, K.H. Pietilainen, H. Nygren, M. Oresic, T. Hyotylainen, Quantitative profiling of bile acids in blood, adipose tissue, intestine, and gall bladder samples using ultra high performance liquid chromatography–tandem mass spectrometry, Anal. Bioanal. Chem. 406 (2014) 7799–7815. [18] M. Scherer, C. Gnewuch, G. Schmitz, G. Liebisch, Rapid quantification of bile acids and their conjugates in serum by liquid chromatography–tandem mass spectrometry, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 877 (2009) 3920–3925. [19] C. John, P. Werner, A. Worthmann, K. Wegner, K. Todter, L. Scheja, S. Rohn, J. Heeren, M. Fischer, A liquid chromatography–tandem mass spectrometry-based method for the simultaneous determination of hydroxy sterols and bile acids, J. Chromatogr. A 1371 (2014) 184–195. [20] L. Humbert, M.A. Maubert, C. Wolf, H. Duboc, M. Mahe, D. Farabos, P. Seksik, J.M. Mallet, G. Trugnan, J. Masliah, D. Rainteau, Bile acid profiling in human biological samples: comparison of extraction procedures and application to normal and cholestatic patients, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 899 (2012) 135–145. [21] D. Tagliacozzi, A.F. Mozzi, B. Casetta, P. Bertucci, S. Bernardini, C. Di Ilio, A. Urbani, G. Federici, Quantitative analysis of bile acids in human plasma by liquid chromatography–electrospray tandem mass spectrometry: a simple and rapid one-step method, Clin. Chem. Lab. Med. 41 (2003) 1633–1641. [22] J. Wang, W. Chow, J. Chang, J.W. Wong, Ultrahigh-performance liquid chromatography electrospray ionization Q-Orbitrap mass spectrometry for the analysis of 451 pesticide residues in fruits and vegetables: method development and validation, J. Agr. Food Chem. 62 (2014) 10375–10391. [23] I.H. Kim, T. Kisseleva, D.A. Brenner, Aging and liver disease, Curr. Opin. Gastroenterol. 31 (2015) 184–191. [24] S.P. Bathena, R. Thakare, N. Gautam, S. Mukherjee, M. Olivera, J. Meza, Y. Alnouti, Urinary bile acids as biomarkers for liver diseases I. Stability of the baseline profile in healthy subjects, Toxicol. Sci. 143 (2015) 296–307. [25] S.P. Bathena, R. Thakare, N. Gautam, S. Mukherjee, M. Olivera, J. Meza, Y. Alnouti, Urinary bile acids as biomarkers for liver diseases II. Signature profiles in patients, Toxicol. Sci. 143 (2015) 308–318. [26] T. Melgarejo, D.A. Williams, N.C. O’Connell, K.D. Setchell, Serum unconjugated bile acids as a test for intestinal bacterial overgrowth in dogs, Dig. Dis. Sci. 45 (2000) 407–414. [27] D. Gems, Long-lived dwarf mice: are bile acids a longevity signal? Aging Cell 6 (2007) 421–423. [28] Department of Health Human Services, Guidance for industry bioanalytical method validation, Drug Eval. Res. (2001) (for the http://www.fda.gov/cder/ guidance/index.htm). [29] J.E. Manautou, C. Chen, Collection of bile and urine samples for determining the urinary and hepatobiliary disposition of xenobiotics in mice, Curr. Protoc. Toxicol. (2005) 7 (Chapter 5, Unit5). [30] J.M. Ridlon, D.J. Kang, P.B. Hylemon, J.S. Bajaj, Bile acids and the gut microbiome, Curr. Opin. Gastroenterol. 30 (2014) 332–338. [31] P. Gerard, Metabolism of cholesterol and bile acids by the gut microbiota, Pathogens 3 (2013) 14–24. [32] H. Duboc, S. Rajca, D. Rainteau, D. Benarous, M.A. Maubert, E. Quervain, G. Thomas, V. Barbu, L. Humbert, G. Despras, C. Bridonneau, F. Dumetz, J.P. Grill, J. Masliah, L. Beaugerie, J. Cosnes, O. Chazouilleres, R. Poupon, C. Wolf, J.M. Mallet, P. Langella, G. Trugnan, H. Sokol, P. Seksik, Connecting dysbiosis, bile-acid dysmetabolism and gut inflammation in inflammatory bowel diseases, Gut 62 (2013) 531–539. [33] M. Trauner, P. Fickert, H. Tilg, Bile acids as modulators of gut microbiota linking dietary habits and inflammatory bowel disease: a potentially dangerous liaison, Gastroenterology 144 (2013) 844–846. [34] E.J. Woodmansey, Intestinal bacteria and ageing, J. Appl. Microbiol. 102 (2007) 1178–1186. [35] N. Masuda, Deconjugation of bile salts by bacteroids and clostridium, Microbiol. Immunol. 25 (1981) 1–11. [36] F.P. Martin, M.E. Dumas, Y. Wang, C. Legido-Quigley, I.K. Yap, H. Tang, S. Zirah, G.M. Murphy, O. Cloarec, J.C. Lindon, N. Sprenger, L.B. Fay, S. Kochhar, P. van Bladeren, E. Holmes, J.K. Nicholson, A top-down systems biology view of microbiome-mammalian metabolic interactions in a mouse model, Mol. Syst. Biol. 3 (2007) 112.