Focused metabolomics using liquid chromatography/electrospray ionization tandem mass spectrometry for analysis of urinary conjugated cholesterol metabolites from patients with Niemann–Pick disease type C and 3β-hydroxysteroid dehydrogenase deficiency

Abstract

Background

Various conjugated cholesterol metabolites are excreted in urine of the patients with metabolic abnormalities and hepatobiliary diseases. We aimed to examine the usefulness of precursor ion scan and neutral loss scan for the characterization of conjugated cholesterol metabolites in urine.

Methods

A mixture of authentic standards of conjugated cholesterol metabolites was used for investigating the performance of the present method. The urine of patients with Niemann–Pick diseases type C and 3β-hydroxysteroid dehydrogenase deficiency were analysed by precursor ion scan of m/z 97, 74, and 124.

Results

A precursor ion scan of m/z 97 was effective for identifying conjugates with ester sulphates on hydroxyl groups whereas ester sulphates on phenolic alcohols were signalled by a neutral loss scan of 80 Da. Monosaccharide-conjugated cholesterol metabolites were signalled by a precursor ion scan of m/z 113. Although precursor ion scan of m/z 74 and 124 was effective for finding glycine- and taurine-conjugated metabolites, high intensity of product ions (m/z 74 and 124) disturbed measurement of other multiply conjugated metabolites. The urine samples contained many conjugated cholesterol metabolites, and there were several disease-specific intense peaks. We found several unknown intense peaks with three known peaks in urine of the Niemann–Pick type C patient. In the patient with 3β-hydroxysteroid dehydrogenase deficiency, intense peaks that were tentatively identified as 5-cholenoic acid sulphates and their glycine and taurine conjugates were present.

Conclusion

The method should lead to the discovery of new urinary biomarkers for these disturbances of cholesterol catabolism and transport.

Keywords

Conjugated cholesterol metabolites focused metabolomics 3β-hydroxysteroid dehydrogenase deficiency liquid chromatography/electrospray ionization tandem mass spectrometry Niemann–Pick disease type C

Introduction

Cholesterol is converted to bile acids, vitamin D, and steroid hormones. Although cholesterol and its esters are lipid-soluble substances, their metabolites are much more hydrophilic because of the addition of hydroxyl, carbonyl, and/or carboxyl groups. The hydroxyl and carboxyl groups may be in turn modified by ‘conjugation’ with amino acids such as glycine or taurine,¹ sulfuric acid,² glucuronic acid,³ N-acetylglucosamine (GlcNAc),⁴ glucose,⁵ and galactose.⁶ These conjugations have profound effects on the fate of such molecules. Conjugates may be less tightly bound to albumin, and therefore enter the glomerular filtrate, increasing renal excretion. In addition, if conjugates are accumulated in the hepatocyte because of inadequate canalicular transport into bile, they may regurgitate into plasma and be eliminated in urine. Thus, the presence of conjugates of cholesterol metabolites in urine could indicate inborn defects of cholesterol metabolism or transport and serve as biomarkers for these genetic diseases. Inborn defects of cholesterol metabolism include 3β-hydroxysteroid-Δ⁵-C₂₇-steroid dehydrogenase (3β-HSD) deficiency,^7,8 Δ⁴-3-oxosteroid 5β-reductase deficiency,^9,10 oxysterol 7α-hydroxylase deficiency,¹¹ and sterol 27-hydroxylase deficiency.¹² In patients of 3β-HSD deficiency, severe cholestasis, pruritus, jaundice, and cirrhosis appear as clinical symptoms. Giant cell change is shown as a pathological finding.⁷ On the other hand, Niemann–Pick disease, type C (NPC),¹³ is known as inborn defect of cholesterol transport, and patients are common in Europe, the United States, and Japan. In patients of NPC, intrahepatic cholestasis, splenohepatomegaly, and neurological and psychiatric symptoms appear, and accumulation of cholesterol, sloughing off of Purkinje cell, and swelling of axon appear as pathological features.¹⁴

Liquid chromatography/electrospray ionization tandem mass spectrometry (LC/ESI-MS/MS) appears to be the best technique for analysis of urinary conjugated cholesterol metabolites.^15–19 If we have preliminary information on the nature of disease-specific molecules, we should be able to analyse target molecules with high sensitivity by using selected reaction monitoring mode. However, there are many unknown biomarker candidates including molecules for which synthetic reference standards are unavailable. Taguchi et al. reported that precursor ion scan and neutral loss scan were effective for focused analysis of lipids possessing common partial structures,²⁰ and the technique is reviewed by Griffiths and Sjövall.²¹

We have previously investigated the characteristics of fragmentation patterns obtained by low-energy collision-induced dissociation (CID) of various conjugated cholesterol metabolites.²² According to conjugation types, characteristic product ions and/or neutral losses were observed in the product ion spectra. Thus, precursor ion scan and neutral loss scan have the potential to be effective for focused analysis of conjugated cholesterol metabolites.

In this study, we investigated the usefulness of precursor ion scan and neutral loss scan using characteristic fragment pattern of various conjugated cholesterol metabolites for focused analysis. We then applied the technique to the analysis of urine samples from a patient with NPC, a disease of cholesterol transport, and another with 3β-HSD deficiency, a genetic defect in bile acid biosynthesis, to validate the usefulness of the method.

Material and methods

Chemicals and materials

Cholic acid (CA), chenodeoxycholic acid (CDCA), deoxycholic acid (DCA), ursodeoxycholic acid (UDCA), and lithocholic acid (LCA) were purchased from Nakalai Tesque, Inc. (Kyoto, Japan). Glycine- and taurine-conjugated bile acids,²³ bile acid 3-sulphates,²⁴ bile acid 3-glucuronides,²⁵ bile acid 24-glucuronides,²⁶ bile acid 7β-GlcNAc conjugates,²⁷ bile acid 3-glucosides,²⁸ bile acid 24-galactosides,²⁹ 3β-sulfooxy-7β-N-acetylglucosaminyl-5-cholen-24-oic acid (SNAG-Δ⁵-CA), and its glycine (SNAG-Δ⁵-CG) and taurine conjugates (SNAG-Δ⁵-CT) were synthesized in our laboratories using previously reported methods.³⁰ 17β-Estradiol 3-sulphate was purchased from Sigma-Aldrich Corp. (St. Louis, MO, USA). 17β-Estradiol 3-glucuronide, 17β-estradiol 17-glucuronide, estrone 3-glucuronide, 17α-estradiol 3-sulphate, 17β-estradiol 17-sulphate, estriol 3-sulphate, estriol 17-sulphate, dehydroepiandrosterone sulphate, epiandrosterone sulphate, pregnenolone sulphate, pregnanolone sulphate, corticosterone 21-sulphate, cortisone 21-sulphate, dehydroepiandrosterone glucuronide, pregnenolone glucuronide, cortisol glucuronide, cholesterol sulphate, cholestanol sulphate, epicoprostanol sulphate, and cholesterol glucuronide were purchased from Steraloids, Inc. (Newport, RI, USA). 17β-Estradiol 3-sulphate 17β-glucuronide was purchased from Toronto Research Chemicals, Inc. (Toronto, Ontario, Canada).

All standard specimens were dissolved in water/ethanol (1:1, v/v) at a concentration of 100 mg/L as stock solutions. A mixture of all reference compounds was prepared, with a concentration of 2.0 µmol/L for each compound. Ultrapure water was prepared using a PURELAB ultra apparatus (Organo Co., Ltd, Tokyo, Japan). All other chemicals were analytical grade and solvents were HPLC or LC/MS grade.

Urine from an NPC patient was collected at Tottori University Hospital. The patient is a male 3-month-old infant. The patient was receiving oral UDCA when the urine sample was collected. Urine from a patient with 3β-HSD deﬁciency (a male 6-month-old infant) and a normal infant (a female 3-month-old infant) were donated by the Junshin Clinic Bile Acid Institute (Tokyo). A healthy adult is a 21-year-old man. All experiments were carried out according to the protocol approved by the ethical committee of the Graduate School of Medicine in Tohoku University.

Focused analysis conditions for conjugated cholesterol metabolites

Mass spectrometric analysis was performed using an API 5000 mass spectrometer (AB Sciex Pte. Ltd, Framingham, MA, USA) equipped with an ESI probe. It was coupled with a Nanospace SI-2 LC system (Shiseido Co., Ltd, Tokyo). Ion spray voltage, declustering potential, and turbospray gas probe were set at –4500, –80 V, and 700℃, respectively. Curtain gas, ion source gas1, ion source gas2, and collision gas were nitrogen and their flow was set to 25, 40, 60, and 6 units, respectively. A 100 µL of water and 100 µL of standard solution (2.0 µmol/L each) were mixed, and 50 µL of aliquots were injected into the LC/ESI-MS/MS system. Analytes were concentrated at the inlet edge of a trapping column (Shim-pack MAYI-C8, 4.6 mm i.d. × 10 mm, Shimadzu, Corp., Kyoto) using a mixture of 20 mmol/L ammonium acetate solution (pH 7.0) and methanol (9:1, v/v) as a mobile phase for trapping at a flow rate of 1 mL/min. The six-port valve position was switched 3 min after and the analytes were separated on a YMC-Pack Pro C18 (5 µm, 2.0 mm i.d.× 250 mm, YMC, Co., Ltd, Kyoto) as an analytical column. Compounds were eluted by a linear gradient elution mode at a flow rate of 0.2 mL/min with the following condition: Solvent A, 20 mmol/L ammonium acetate solution (pH 7.0) and methanol (19:1, v/v); solvent B, (1:19, v/v); 50% solvent B was linearly changed to 80% solvent B against A over 120 min and then the rate of solvent B was kept at 80% for 20 min. The rate of solvent B was then switched to 50% and kept for 40 min. Collision voltages were set at –50, –60, –50, –80, and –55 V on precursor ion scans of m/z 74, 97, 113, 124, and m/z 202, respectively. Neutral loss scan of 80, 162, and 221 Da were set using a collision voltage of –50, –50, and –50 V, respectively. In analysis of GlcNAc conjugates, neutral loss spectra were measured at a scan range from m/z 230 to 900, and in other cases, precursor ion spectra and neutral loss spectra were measured at a scan range from m/z 200 to 900 with a scan time of 2 s. Accuracy of retention time and m/z were calculated as coefficient of variation (CV%). Inter-day accuracy of retention time was within 1.6% and intra-day accuracy was within 0.2%. Both inter-day and intra-day accuracy of m/z were within 0.03%. Data were collected and processed using Analyst 1.4.1 data collection and integration software.

Results and discussion

Focused approach for analysis of conjugated cholesterol metabolites

First, we analysed a mixture of authentic conjugated cholesterol metabolites using the Q1-scan mode. Since we know retention times and m/z of all standard compounds, we can identify every peak by ion extraction (data not shown). Most authentic standards were found at the retention times from 8.55 to 126 min. However, C₂₇ compounds (cholesterol 3-sulphate, cholestanol 3-sulphate, epicoprostanol 3-sulphate, and cholesterol 3-glucuronide) were not eluted even at 140 min under the present analytical conditions. For more rapid elution of those compounds, more lipophilic mobile phases containing dichloromethane or chloroform are likely to be required.

Figure 1(a) is a total ion current (TIC) chromatogram using a precursor ion scan of m/z 74 for extraction of conjugated cholesterol metabolites possessing a glycine moiety. We were able to find 17 metabolites in the chromatogram and detect all metabolites possessing a glycine moiety, which we used as reference compounds in this study. The detected precursor ions included doubly charged deprotonated molecules because they have two acidic groups (carboxylic acid and sulfuric acid) in their molecules. In the TIC chromatogram using precursor ion scan of m/z 74, some pairs, such as peaks 8 and 9, peaks 10 and 11, and peaks 15 and 16, were not separated from each other; however, individual compounds can be distinguished by their m/z values. The TIC chromatogram in Figure 1(b) is obtained using precursor ion scan of m/z 124. The peaks in this chromatogram were derived from all taurine-conjugated metabolites used as standards in this study. Although their chromatographic separations were incomplete, they were able to be distinguished by their m/z values as was the case with glycine-conjugated metabolites.

Figure 1.

Total ion current chromatograms of a mixture of authentic conjugated cholesterol metabolites: (a) precursor ion scan of m/z 74, (b) precursor ion scan of m/z 124, (c) precursor ion scan of m/z 97, (d) neutral loss scan of 80 Da, (e) precursor ion scan of m/z 113, (f) neutral loss scan of 162 Da, (g) neutral loss scan of 221 Da, and (h) precursor ion scan of m/z 202. All tandem mass spectrometric and chromatographic conditions were as shown in ‘Material and methods’ section. The detected standard substances were numbered in each chromatograms. The peak list was shown in Table 1. e.g. The peak 2 in chromatogram (a) corresponds to the peak No. A2, GUDCA 3-sulphate, and the peak 15 in chromatogram (b) corresponds to the peak No. B15, TLCA.

Next, we performed precursor ion scan of m/z 97 for extraction of sulphate conjugates; the TIC chromatogram is shown in Figure 1(c). Peaks in this chromatogram included sulphate conjugates with the sulphate moiety being esterified with the aliphatic, secondary hydroxyl groups such as a 3α-hydroxyl (or 7α-hydroxyl group) of bile acids and a 17β-hydroxyl group on estrogens. Sulphate groups on phenolic hydroxyl groups do not generate an m/z 97 fragment. Therefore, although estradiol 17-sulphate was clearly detected in the TIC chromatogram on precursor ion scan of m/z 97, estradiol 3-sulphate did not give a signal in the same chromatogram. Sulphate conjugates on the phenolic hydroxyl group were found in the TIC chromatogram on neutral loss scan of 80 Da as shown in Figure 1(d). Neutral loss of 80 Da was also able to extract amino acid-conjugated bile acid 3-sulphates in addition to estrogen C-3 sulphates. No non-amidated bile acid 3-sulphates were identified as precursors for the neutral loss of 80 Da, suggesting that somehow interaction between the side chain amino acid and the hydroxyl group at C-3 occurs. Thus, precursor ion scan of m/z 97 is effective for group-specific extraction of sulphate conjugates in which the sulphate group is linked by an ester bond to aliphatic alcohol substituents. On the other hand, neutral loss of 80 Da is effective for extraction of sulphate conjugates in which the sulphate is linked to a phenolic hydroxyl group although double conjugates with glycine or taurine can be found in the TIC chromatogram on neutral loss of 80 Da even in the case of sulphate conjugates linked to aliphatic alcohols. This fact is useful for identifying N-acyl amidates containing glycine or taurine.

Figure 1(e) shows a TIC chromatogram using the precursor ion scan of m/z 113. This product ion has been found in product ion spectra of monosaccharide-conjugated compounds other than GlcNAc conjugates, although acyl galactosides exhibit relatively lower intense peaks on the TIC chromatogram.²² According to our previous paper, a precursor ion scan of m/z 175 and neutral loss scan of 176 Da aids in the identification of the mode of conjugation of glucuronides, in ester linkage with the carboxyl group, in ether linkage with an aliphatic hydroxyl group, or a phenolic hydroxyl group. Thus, this approach will work well for discrimination of conjugation types of glucuronides possessing the same molecular masses.²²

Figure 1(f) shows the TIC chromatogram on neutral loss scan of 162 Da. Precursors include 15 different bile acid 3-glucosides. We were not able to find acyl galactosides because of their lower intensities. Peak intensities of amino acid conjugates are extremely lower than unconjugated bile acid 3-glucosides, and this can be attributed to the fact that product ions corresponded to amino acid moieties at m/z 74 and 124 are easy to produce under low-energy CID conditions. In addition, both neutral loss scan of 221 Da and precursor ion scan of m/z 202 were useful for extraction of GlcNAc conjugates as shown in Figure 1(g) and (h). In Figure 1(g), a neutral loss of 221 Da generated by SNAG-Δ⁵-CA, SNAG-Δ⁵-CG, and SNAG-Δ⁵-CT was found; however, UDCA 7β-GlcNAc and its amino acid conjugates were not present in this chromatogram. Conversely, they were found in the TIC chromatogram on precursor ion scan of m/z 202 as shown in Figure 1(h) although TUDCA 7β-GlcNAc was not found because of its low intensity. This distinction may be due to presence or absence of a double bond adjacent to 7β-GlcNAc moiety. Table 1 gives m/z values for precursor ions and retention times of all reference standards shown in Figure 1.

Table 1.

List of m/z of precursor ion and retention times of all standard substances detected in Figure 1.

No.	Compound	Composition	m/z	t_R (min)	The same compounds
Precursor ion scan of m/z 74
A1	SNAG-Δ⁵-CG	C₃₄H₅₄N₂O₁₃S	364.5 ([M-2H]^2–), 729.7 ([M-H]^–)	11.6	C2, G2
A2	GUDCA 3-sulphate	C₂₆H₄₃NO₈S	264.0 ([M-2H]^2–), 528.6 ([M-H]^–)	16.6	C8, D4
A3	GCA 3-sulphate	C₂₆H₄₃NO₉S	271.9 ([M-2H]^2–), 544.4 ([M-H]^–)	23.3	C11, D7
A4	GUDCA 3-glucoside	C₃₂H₅₃NO₁₀	610.7 ([M-H]^–)	26.8	F2
A5	GUDCA 7β-GlcNAc	C₃₄H₅₄N₂O₁₀	651.7 ([M-H]^–)	29.5	H1
A6	GCDCA 3-sulphate	C₂₆H₄₃NO₈S	264.0 ([M-2H]^2–), 528.6 ([M-H]^–)	34.7	C16, D9
A7	GDCA 3-sulphate	C₂₆H₄₃NO₈S	264.0 ([M-2H]^2–), 528.6 ([M-H]^–)	37.6	C17, D10
A8	GUDCA	C₂₆H₄₃NO₅	448.6 ([M-H]^–)	39.3
A9	GCA 3-glucoside	C₃₂H₅₃NO₁₁	626.7 ([M-H]^–)	40.9	F5
A10	GLCA 3-sulphate	C₂₆H₄₃NO₇S	256.1 ([M-2H]^2–), 512.7 ([M-H]^–)	51.2	C21, D12
A11	GCA	C₂₆H₄₃NO₆	464.7 ([M-H]^–)	53.1
A12	GCDCA 3-glucoside	C₃₂H₅₃NO₁₀	610.6 ([M-H]^–)	57.6	F8
A13	GDCA 3-glucoside	C₃₂H₅₃NO₁₀	610.7 ([M-H]^–)	61.2	F10
A14	GCDCA	C₂₆H₄₃NO₅	448.6 ([M-H]^–)	73.7
A15	GLCA 3-glucoside	C₃₂H₅₃NO₉	594.7 ([M-H]^–)	77.6	F14
A16	GDCA	C₂₆H₄₃NO₅	448.6 ([M-H]^–)	79.0
A17	GLCA	C₂₆H₄₃NO₄	432.6 ([M-H]^–)	97.4
Precursor ion scan of m/z 124
B1	SNAG-Δ⁵-CT	C₃₄H₅₆N₂O₁₄S₂	389.5 ([M-2H]^2–)	11.6	C3, G1
B2	TUDCA 3-sulphate	C₂₆H₄₅NO₉S₂	289.1 ([M-2H]^2–), 578.6 ([M-H]^–)	16.2	C7, D3
B3	TUDCA 3-glucoside	C₃₂H₅₅NO₁₁S	660.5 ([M-H]^–)	27.8	F1
B4	TUDCA 7β-GlcNAc	C₃₄H₅₆N₂O₁₁S	701.6 ([M-H]^–)	29.3
B5	TCDCA 3-sulphate	C₂₆H₄₅NO₉S₂	289.1 ([M-2H]^2–), 578.6 ([M-H]^–)	33.8	C15, D8
B6	TUDCA	C₂₆H₄₅NO₆S	498.6 ([M-H]^–)	38.6
B7	TCA 3-glucoside	C₃₂H₅₅NO₁₂S	676.5 ([M-H]^–)	39.8	F4
B8	TLCA 3-sulphate	C₂₆H₄₅NO₈S₂	281.0 ([M-2H]^2–), 562.6 ([M-H]^–)	50.2	C20, D11
B9	TCA	C₂₆H₄₅NO₇S	514.6 ([M-H]^–)	51.6
B10	TCDCA 3-glucoside	C₃₂H₅₅NO₁₁S	660.5 ([M-H]^–)	56.3	F7
B11	TDCA 3-glucoside	C₃₂H₅₅NO₁₁S	660.6 ([M-H]^–)	60.4	F9
B12	TCDCA	C₂₆H₄₅NO₆S	498.6 ([M-H]^–)	72.0
B13	TLCA 3-glucoside	C₃₂H₅₅NO₁₀S	644.5 ([M-H]^–)	75.7	F13
B14	TDCA	C₂₆H₄₅NO₆S	498.6 ([M-H]^–)	77.2
B15	TLCA	C₂₆H₄₅NO₅S	482.4 ([M-H]^–)	95.2
Precursor ion scan of m/z 97
C1	Estriol 17-sulphate	C₁₈H₂₄O₆S	367.1 ([M-H]^–)	11.4
C2	SNAG-Δ⁵-CG	C₃₄H₅₄N₂O₁₃S	364.5 ([M-2H]^2–), 729.7 ([M-H]^–)	11.7	A1, G2
C3	SNAG-Δ⁵-CT	C₃₄H₅₆N₂O₁₄S₂	389.6 ([M-2H]^2–)	11.8	B1, G1
C4	Cortisone 21-sulphate	C₂₁H₂₈O₈S	439.5 ([M-H]^–)	12.6
C5	SNAG-Δ⁵-CA	C₃₂H₅₁NO₁₂S	672.7 ([M-H]^–)	15.5	G3
C6	17β-Estradiol 17-sulphate	C₁₈H₂₄O₅S	351.0 ([M-H]^–)	16.6
C7	TUDCA 3-sulphate	C₂₆H₄₅NO₉S₂	289.1 ([M-2H]^2–)	17.0	B2, D3
C8	GUDCA 3-sulphate	C₂₆H₄₃NO₈S	264.0 ([M-2H]^2–), 528.6 ([M-H]^–)	17.2	A2, D4
C9	Corticosterone 21-sulphate	C₂₁H₃₀O₇S	425.0 ([M-H]^–)	20.6
C10	UDCA 3-sulphate	C₂₄H₄₀O₇S	471.5 ([M-H]^–)	24.3
C11	GCA 3-sulphate	C₂₆H₄₃NO₉S	272.1 ([M-2H]^2–), 544.5 ([M-H]^–)	24.2	A3, D7
C12	Dehydroepiandrosterone 3-sulphate	C₁₉H₂₈O₅S	367.1 ([M-H]^–)	26.0
C13	Epiandrosterone 3-sulphate	C₁₉H₃₀O₅S	369.2 ([M-H]^–)	29.7
C14	CA 3-sulphate	C₂₄H₄₀O₈S	487.5 ([M-H]^–)	31.4
C15	TCDCA 3-sulphate	C₂₆H₄₅NO₉S₂	289.1 ([M-2H]^2–)	35.1	B5, D8
C16	GCDCA 3-sulphate	C₂₆H₄₃NO₈S	264.0 ([M-2H]^2–), 528.6 ([M-H]^–)	35.8	A6, D9
C17	GDCA 3-sulphate	C₂₆H₄₃NO₈S	264.0 ([M-2H]^2–), 528.6 ([M-H]^–)	38.7	A7, D10
C18	Pregnenolone 3-sulphate	C₂₁H₃₂O₅S	395.2 ([M-H]^–)	45.2
C19	DCA 3-sulphate	C₂₄H₄₀O₇S	471.5 ([M-H]^–)	49.9
C20	TLCA 3-sulphate	C₂₆H₄₅NO₈S₂	281.0 ([M-2H]^2–)	51.1	B8, D11
C21	GLCA 3-sulphate	C₂₆H₄₃NO₇S	256.1 ([M-2H]^2–), 512.7 ([M-H]^–)	52.3	A10, D12
C22	Pregnanolone 3-sulphate	C₂₁H₃₄O₅S	397.2 ([M-H]^–)	56.8
Neutral loss scan of 80 Da
D1	Estradiol 3-sulphate 17β-glucuronide	C₂₄H₃₂O₁₁S	263.3 ([M-2H]^2–), 527.2 ([M-H]^–)	8.50
D2	Estriol 3-sulphate	C₁₈H₂₄O₆S	367.3 ([M-H]^–)	9.52
D3	TUDCA 3-sulphate	C₂₆H₄₅NO₉S₂	578.4 ([M-H]^–)	18.8	B2, C7
D4	GUDCA 3-sulphate	C₂₆H₄₃NO₈S	528.4 ([M-H]^–)	19.0	A2, C8
D5	17β-Estradiol 3-sulphate	C₁₈H₂₄O₅S	351.0 ([M-H]^–)	19.6
D6	17α-Estradiol 3-sulphate	C₁₈H₂₄O₅S	351.0 ([M-H]^–)	21.8
D7	GCA 3-sulphate	C₂₆H₄₃NO₉S	544.4 ([M-H]^–)	26.5	A3, C11
D8	TCDCA 3-sulphate	C₂₆H₄₅NO₉S₂	578.4 ([M-H]^–)	38.8	B5, C15
D9	GCDCA 3-sulphate	C₂₆H₄₃NO₈S	528.4 ([M-H]^–)	39.5	A6, C16
D10	GDCA 3-sulphate	C₂₆H₄₃NO₈S	528.6 ([M-H]^–)	42.7	A7, C17
D11	TLCA 3-sulphate	C₂₆H₄₅NO₈S₂	562.6 ([M-H]^–)	55.8	B8, C20
D12	GLCA 3-sulphate	C₂₆H₄₃NO₇S	512.6 ([M-H]^–)	57.0	A10, C21
Precursor ion scan of m/z 113
E1	17β-Estradiol 3-glucuronide	C₂₄H₃₂O₈	447.6 ([M-H]^–)	12.7
E2	17β-Estradiol 17-glucuronide	C₂₄H₃₂O₈	447.6 ([M-H]^–)	12.7
E3	Estrone 3-glucuronide	C₂₄H₃₀O₈	447.6 ([M-H]^–)	12.7
E4	Cortisol 21-glucuronide	C₂₇H₃₈O₁₁	537.5 ([M-H]^–)	14.9
E5	Dehydroepiandrosterone 3-glucuronide	C₂₅H₃₆O₈	463.6 ([M-H]^–)	19.0
E6	UDCA 3-glucoside	C₃₀H₅₀O₉	553.2 ([M-H]^–)	32.5	F3
E7	Pregnenolone 3-glucuronide	C₂₇H₄₀O₈	491.6 ([M-H]^–)	33.0
E8	DCA 3-glucuronide	C₃₀H₄₈O₁₀	567.7 ([M-H]^–)	45.0
E9	CA 3-glucoside	C₃₀H₅₀O₁₀	569.6 ([M-H]^–)	53.4	F6
E10	LCA 3-glucuronide	C₃₀H₄₈O₉	551.6 ([M-H]^–)	62.7
E11	CDCA 3-glucoside	C₃₀H₅₀O₉	553.7 ([M-H]^–)	67.1	F11
E12	DCA 3-glucoside	C₃₀H₅₀O₉	553.3 ([M-H]^–)	70.7	F12
E13	LCA 3-glucoside	C₃₀H₅₀O₈	537.7 ([M-H]^–)	88.4	F15
Neutral loss scan of 162 Da
F1	TUDCA 3-glucoside	C₃₂H₅₅NO₁₁S	660.5 ([M-H]^–)	26.0	B3
F2	GUDCA 3-glucoside	C₃₂H₅₃NO₁₀	610.7 ([M-H]^–)	26.4	A4
F3	UDCA 3-glucoside	C₃₀H₅₀O₉	553.2 ([M-H]^–)	32.1	E6
F4	TCA 3-glucoside	C₃₂H₅₅NO₁₂S	676.5 ([M-H]^–)	39.5	B7
F5	GCA 3-glucoside	C₃₂H₅₃NO₁₁	626.7 ([M-H]^–)	40.4	A9
F6	CA 3-glucoside	C₃₀H₅₀O₁₀	569.6 ([M-H]^–)	47.0	E9
F7	TCDCA 3-glucoside	C₃₂H₅₅NO₁₁S	660.5 ([M-H]^–)	55.7	B10
F8	GCDCA 3-glucoside	C₃₂H₅₃NO₁₀	610.6 ([M-H]^–)	57.2	A12
F9	TDCA 3-glucoside	C₃₂H₅₅NO₁₁S	660.6 ([M-H]^–)	59.7	B11
F10	GDCA 3-glucoside	C₃₂H₅₃NO₁₀	610.7 ([M-H]^–)	61.0	A13
F11	CDCA 3-glucoside	C₃₀H₅₀O₉	553.7 ([M-H]^–)	66.1	E11
F12	DCA 3-glucoside	C₃₀H₅₀O₉	553.5 ([M-H]^–)	69.7	E12
F13	TLCA 3-glucoside	C₃₂H₅₅NO₁₀S	644.5 ([M-H]^–)	75.2	B13
F14	GLCA 3-glucoside	C₃₂H₅₃NO₉	594.7 ([M-H]^–)	76.7	A15
F15	LCA 3-glucoside	C₃₀H₅₀O₈	537.7 ([M-H]^–)	87.3	E13
Neutral loss scan of 221 Da
G1	SNAG-Δ⁵-CT	C₃₄H₅₆N₂O₁₄S₂	779.5 ([M-H]^–)	11.7	B1, C3
G2	SNAG-Δ⁵-CG	C₃₄H₅₄N₂O₁₃S	729.4 ([M-H]^–)	11.7	A1, C2
G3	SNAG-Δ⁵-CA	C₃₂H₅₁NO₁₂S	672.4 ([M-H]^–)	15.2	C5
Precursor ion scan of m/z 202
H1	GUDCA 7β-GlcNAc	C₃₄H₅₄N₂O₁₀	651.7 ([M-H]^–)	30.7	A5
H2	UDCA 7β-GlcNAc	C₃₂H₅₁NO₉	594.5 ([M-H]^–)	36.9

CA: cholic acid; CDCA: chenodeoxycholic acid; DCA: deoxycholic acid; GCA: glycine-conjugated cholic acid; GCDCA: glycine-conjugated chenodeoxycholic acid; GDCA: glycine-conjugated deoxycholic acid; GLCA: glycine-conjugated lithocholic acid; GlcNAc: N-acetylglucosamine; GUDCA: glycine-conjugated ursodeoxycholic acid; LCA: lithocholic acid; SNAG-Δ⁵-CA: 3β-sulfoxy-7β-N-acetylglucosaminyl-5-cholen-24-oic acid; SNAG-Δ⁵-CG: glycine-conjugated 3β-sulfoxy-7β-N-acetylglucosaminyl-5-cholen-24-oic acid; SNAG-Δ⁵-CT: taurine-conjugated 3β-sulfooxy-7β-N-acetylglucosaminyl-5-cholen-24-oic acid; TCA: taurine-conjugated cholic acid; TCDCA: taurine-conjugated chenodeoxycholic acid; TDCA: taurine-conjugated deoxycholic acid; TLCA: taurine-conjugated lithocholic acid; TUDCA: taurine-conjugated ursodeoxycholic acid; UDCA: ursodeoxycholic acid.

Focused analysis of urine from NPC and 3β-HSD deficiency patients

In hepatobiliary disease such as a cholestasis, the output of urinary bile acids increases significantly, and sulphate conjugates are known to constitute the majority of urinary bile acids. Therefore, we analysed urine samples obtained from a healthy volunteer, a healthy infant, a patient with NPC (ingesting UDCA), and a patient with 3β-HSD deficiency using a precursor ion scan of m/z 97. The results are summarized in Table 2 and Supplementary Table 1. All detected peaks were enumerated in Supplementary Table 1, and intense peaks over 10⁸ counts were extracted and listed in Table 2. We found few peaks in TIC chromatograms obtained from urine samples obtained from both a healthy infant and a healthy volunteer. In contrast, the urine samples of an NPC patient and a 3β-HSD deficiency patient contained many peaks corresponding to sulphate conjugates. In addition, there were intense peaks in the NPC sample that differed from those in the urine of the 3β-HSD deficiency patient, suggesting that detection and/or quantitation after developing a quantitative method of their characteristic peaks should permit the two diseases to be distinguished. However, many peaks could not be identified.

Table 2.

Peak list of conjugated cholesterol metabolites detected in abundance in urine of a patient with NPC and a patient with 3β-HSD deficiency.

			Precursor ion scan of m/z 97		Precursor ion scan of m/z 74		Precursor ion scan of m/z 124
			NPC	3β-HSD deficiency	NPC	3β-HSD deficiency	NPC	3β-HSD deficiency
No	m/z	Composition	t_R (Counts)	t_R (Counts)	t_R (Counts)	t_R (Counts)	t_R (Counts)	t_R (Counts)
27	364.6, 729.6	C₃₄H₅₄N₂O₁₃S	11.8 (3.4 E+08)	–	12.6 (2.5 E+08)	11.9 (1.8 E+07)	–	–
28	389.5	C₃₄H₅₆N₂O₁₄S₂	11.8 (4.9 E+08)	–	–	–	12.4 (3.0 E+07)	–
45	672.6	C₃₂H₅₁NO₁₂S	15.4 (2.2 E+08)	–	–	–	–	–
50	295.9	C₂₆H₄₃NO₁₀S₂	–	16.0 (2.9 E+09)	–	–	–	–
51	453.5	C₂₄H₃₈O₆S	16.1 (1.0 E+09)	–	–	–	–	–
52	270.8, 542.4	C₂₆H₄₁NO₉S	–	16.2 (2.9 E+09)	16.4 (6.9 E+08)	–	–	–
53	467.6	C₂₄H₃₆O₇S	–	16.2 (3.2 E+09)	–	–	–	–
63	289.1	C₂₆H₄₅NO₉S₂	16.8 (8.5 E+08)	–	–	–	–	–
64	264.1, 528.6	C₂₆H₄₃NO₈S	17.0 (1.5 E+09)	–	18.0 (5.4 E+08)	–	–	–
69	469.6	C₂₄H₃₈O₇S	21.0 (1.5 E+09)	–	–	–	–	–
70	485.6	C₂₄H₃₈O₈S	–	21.2 (3.3 E+09)	–	–	–	–
73	467.6	C₂₄H₃₆O₇S	23.4 (2.5 E+09)	–	–	–	–	–
74	471.6	C₂₄H₄₀O₇S	24.0 (7.7 E+08)	–	–	–	–	–
76	288.0	C₂₆H₄₃NO₉S₂	–	24.4 (3.3 E+09)	–	–	–	–
79	263.1, 526.6	C₂₆H₄₁NO₈S	–	24.8 (1.4 E+09)	–	25.1 (3.9 E+08)	–	–
95	469.6	C₂₄H₃₈O₇S	–	33.5 (4.7 E+09)	–	–	–	–

Peak numbers (No) were identical with those in Supplementary Table 1.

Underline: doubly charged ion.

Next, we performed precursor ion scans of m/z 74 and 124 for extracting glycine and taurine conjugates, respectively; the results also were put into the same tables. Taurine and sulphate double conjugates provided doubly charged deprotonated molecule [M-2H]^2– without a singly charged one. Glycine-conjugated and sulphate double conjugates provided both signals with parallel intensity. This information also helped to distinguish sulphated glycine and taurine conjugates.

The urine sample obtained from an NPC patient provided many characteristic intense peaks although unconjugated UDCA 3-sulphate, glycine-, and taurine-conjugated UDCA 3-sulphates were found at the retention times of 24.0 min (No. 74), 17.0 min (No. 64), and 16.8 min (No. 63), respectively, presumably because the patient was ingesting UDCA as possible therapy. The peak (No. 51) at the retention time of 16.1 min, a precursor ion at m/z 453.5, has C₂₄H₃₈O₆S as its elemental composition corresponding to a cholenoic acid sulphate conjugate possessing a hydroxyl group and one double bond, e.g. monohydroxy-5-cholenoic acid sulphate or possibly a sulphate of a hydroxyl, oxo-bile acid. The peak (No. 69) showing m/z 469.6 as a precursor ion at the retention time of 21.0 min corresponds to a dihydroxycholenoic acid sulphate or a hydroxy oxo bile acid sulphate, but the position and conformation of functional groups is not known. In addition, the peak (No. 73) with a retention time of 23.4 min, m/z 467.6, corresponds to a monohydroxy-monooxo-5-cholenoic acid sulphate. Alvelius et al. reported abnormal accumulation of SNAG-Δ⁵-CT, SNAG-Δ⁵-CG, and SNAG-Δ⁵-CA, in a patient with NPC.³¹ We have synthesized these compounds as noted earlier. The retention time of SNAG-Δ⁵-CT was 11.8 min (No. 28, m/z 389.5), that of SNAG-Δ⁵-CG was 11.8 min (No. 27, m/z 364.5 and 729.6), and that of SNAG-Δ⁵-CA was 15.4 min (No. 45, m/z 672.6). Although their intensities were weaker than unknown three cholenoic acid sulphate conjugates, they were specific to our NPC patient.

In urine from 3β-HSD deficiency patient, the peaks at the retention times of 16.0 min (No. 50), 16.2 min (No. 52, 53), 21.2 min (No. 70), 24.4 min (No. 76), 24.8 min (No. 79), and 33.5 min (No. 95) were specific to our 3β-HSD deficiency patient. HSD catalyses the shift of the double bond at C-5 to C-4 and at the same time oxidizes the hydroxyl group at C-3. Therefore, a deficiency of this enzyme should result in the accumulation of 3β-hydroxy, Δ⁵ bile acids that in principle could be either C₂₄ or C₂₇, depending on whether peroxisomal oxidation of the C₈ side chain occurs. Indeed, the occurrence of 3β,7α-dihydroxy-5-cholenoic acid and 3β,7α,12α-trihydroxy-5-cholenoic acid in urine has been reported, indicating that side chain oxidation proceeds.^7,32 Thus, it seems likely that the peaks having the retention times of 16.0 min (No. 50, m/z 295.9), 16.2 min (No. 52, m/z 270.8 and 542.4), and 21.2 min (No. 70, m/z 485.6) are taurine conjugated, glycine conjugated, and unconjugated 3β,7α,12α-trihydroxy-5-cholenoic acid 3-sulphate, respectively. The peaks with retention times of 24.4 min (No. 76, m/z 288), 24.8 min (No. 79, m/z 263.1 and 526.1), and 33.5 min (No. 95, m/z 469.6) are perhaps taurine conjugated, glycine conjugated, and unconjugated 3β,7α-dihydroxy-5-cholenoic acid 3-sulphate, respectively. We did not detect the C₂₇ homologues of these compounds.

Conclusion

In this study, we demonstrated the utility of focused metabolomics analysis for urinary conjugated cholesterol metabolites using precursor ion and neutral loss scans. Our work shows the usefulness of precursor ion scan and neutral loss scan analysis using characteristic fragmentations for conjugation moieties in extracting the desired peaks from very complicated matrices.

Using precursor ion scans analysis of urine from two patients with different diseases, NPC and 3β-HSD deficiency, we found some characteristic peaks specific for an NPC patient (No. 51, 69, and 73) and 3β-HSD deficiency (No. 50, 52, 70, 76, 79, and 95). The present method gives promise of identifying new biomarkers for early diagnosis of diseases characterized by urinary excretion of conjugated cholesterol and/or bile acid derivatives.

Footnotes

Declaration of conflicting interests

None declared.

Funding

This study was supported in part by KAKENHI (Grant-in Aid for Challenging Exploratory Research: 22659007).

Ethical approval

The ethical committee of the Graduate School of Medicine in Tohoku University approved this study (REC number: 2013-1-535).

Guarantor

NM.

Contributionship

MM collected data and wrote the manuscript. MS, JG, and HY contributed to the discussion. KO, MT, and HN collected urine sample. TI synthesized authentic standards. AFH contributed to the discussion and reviewed/edited the manuscript. NM designed the study, contributed to the discussion, and reviewed/edited the manuscript.

References

Falany

Johnson

Barnes

. Glycine and taurine conjugation of bile acids by a single enzyme. J Biol Chem 1994; 269: 19375–19379.

Palmer

. The formation of bile acid sulfates: a new pathway of bile acid metabolism in humans. Proc Natl Acad Sci 1967; 58: 1047–1050.

Matern

Farthmann

. Hepatic and extrahepatic glucuronidation of bile acids in Man. J Clin Invest 1984; 74: 402–410.

Marschall

Egestad

Matern

. N-Acetylglucosaminides. J Biol Chem 1989; 264: 12989–12993.

Matern

Gerok

. Formation of bile acid glucosides by a sugar nucleotide-independent glucosyltransferase isolated from human liver microsomes. Proc Natl Acad Sci 1984; 81: 7036–7040.

Goto

Shibata

Sasaki

. Identification of a novel conjugate in human urine: bile acid acyl galactosides. Steroids 2005; 70: 185–192.

Clayton

Leonard

Lawson

. Familial giant cell hepatitis associated with synthesis of 3β,7α-dihydroxy- and 3β,7α,12α-trihydroxy-5-cholenoic acids. J Clin Invest 1987; 79: 1031–1038.

Schwarz

Wright

Davis

. The bile acid synthetic gene 3β-hydroxy-Δ⁵-C₂₇-steroid oxidoreductase is mutated in progressive intrahepatic cholestasis. J Clin Invest 2000; 106: 1175–1184.

Setchell

KDR

Suchy

Welsh

. Δ⁴-3-Oxosteroid 5β-reductase deficiency described in identical twins with neonatal hepatitis. J Clin Invest 1988; 82: 2148–2157.

10.

Lemonde

Custard

Bouquet

. Mutation in SRD5B1 (AKR1D1), the gene encoding Δ⁴-3-oxosteroid 5β-reductase, in hepatitis and liver failure in infancy. Gut 2003; 52: 1494–1499.

11.

Setchell

KDR

Schwarz

O’connell

. Identification of a new inborn error in bile acid synthesis: mutation of the oxysterol 7α-hydroxylase gene causes severe neonatal liver disease. J Clin Invest 1998; 102: 1690–1703.

12.

Cali

Hsieh

Francke

. Mutations in bile acid biosynthetic enzyme sterol 27-hydroxylase underlie cerebrotendious xanthomatosis. J Biol Chem 1991; 266: 7779–7783.

13.

Carstea

Morris

Coleman

. Niemann-Pick C1 disease gene: homology to mediators of cholesterol homeostasis. Science 1997; 277: 228–231.

14.

Vanier

Suzuki

. Recent advances in elucidating Niemann-Pick C disease. Brain Pathol 1998; 8: 163–174.

15.

Ikegawa

Yanahihara

Murao

. Separatory determination of bile acid 3-sulfates by liquid chromatography/electrospray ionization mass spectrometry. J Mass Spectrom 1997; 32: 401–407.

16.

Goto

Myint

Sato

. LC/ESI-tandem mass spectrometric determination of bile acid 3-sulfates in human urine 3β-sulfooxy-12α-hydroxy-5β-cholanoic acid is an abundant nonamidated sulfate. J Chromatogr B 2007; 846: 69–77.

17.

Mitamura

Sogabe

Sakanashi

. Analysis of bile acid glutathione thioesters by liquid chromatography/electrospray ionization-tandem mass spectrometry. J Chromatogr B 2007; 855: 88–97.

18.

Muto

Takei

Unno

. Detection of Δ⁴-3-oxo-steroid 5β-reductase deficiency by LC-ESI-MS/MS measurement of urinary bile acids. J Chromatogr B 2012; 900: 24–31.

19.

Higashi

Shibayama

Kawasaki

. Determination of salivary 17-ketosteroid sulfates using liquid chromatography-electrospray ionization-mass spectrometry. J Pharm Biomed Anal 2007; 43: 1782–1788.

20.

Taguchi

Houjou

Nakanishi

. Focused lipidomics by tandem mass spectrometry. J Chromatogr B 2005; 823: 26–36.

21.

Griffiths

Sjövall

. Bile acids: analysis in biological fluids and tissues. J Lipid Res 2010; 51: 23–41.

22.

Maekawa

Shimada

Iida

. Tandem mass spectrometric characterization of bile acids and steroid conjugates based on low-energy collision-induced dissociation. Steroids 2014; 80: 80–91.

23.

Goto

Hasegawa

Kato

. A new method for simultaneous determination of bile acids in human bile without hydrolysis. Clin Chim Acta 1978; 87: 141–147.

24.

Goto

Kato

Hasegawa

. Synthesis of monosulfates of unconjugated and conjugated bile acids. Chem Pharm Bull 1979; 27: 1402–1411.

25.

Goto

Suzaki

Nambara

. Synthesis of 3-glucuronides of unconjugated and conjugated bile acids. Chem Pharm Bull 1980; 28: 1258–1264.

26.

Goto

Murao

Oohashi

. Synthesis of bile acid 24-acyl glucuronides. Steroids 1998; 63: 180–185.

27.

Niwa

Koshiyama

Goto

. Synthesis of N-acetylglucosaminides of unconjugated and conjugated bile acids. Steroids 1992; 57: 522–529.

28.

Iida

Nishida

Yamaguchi

. Potential bile acid metabolites. 23. Syntheses of 3-glucosides of nonamidated and glycine- and taurine-conjugated bile acids. J Lipid Res 1995; 36: 628–638.

29.

Kakiyama

Sadakiyo

Iida

. Chemical synthesis of 24-β-D-galactopyranosides of bile acids: a new type of bile acid conjugates in human urine. Chem Phys Lipids 2005; 134: 141–150.

30.

Iida

Kakiyama

Hibiya

. Chemical synthesis of the 3-sulfooxy-7-N-acetylglucosaminyl-24-conjugated conjugates of 3β,7β-dihydroxy-5-cholen-24-oic acid, and related compounds: unusual, major metabolites of bile acid in a patient with Niemann-Pick disease type C1. Steroids 2006; 71: 18–29.

31.

Alvelius

Hjalmarson

Griffiths

. Identification of unusual 7-oxygenased bile acid sulfates in a patient with Niemann-Pick disease, type C. J Lipid Res 2000; 42: 1571–1577.

32.

Ichimiya

Egestad

Nazer

. Bile acids and bile alcohols in a child with hepatic 3β-hydroxy-Δ⁵-C₂₇-steroid dehydrogenase deficiency: effects of chenodeoxycholic acid treatment. J Lipid Res 1991; 32: 829–841.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.38 MB