Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry

G Model

ARTICLE IN PRESS

CHROMA-357029; No. of Pages 7

Journal of Chromatography A, xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma

Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry Zoltán Pataj, Gerhard Liebisch, Gerd Schmitz, Silke Matysik ∗ Institute of Clinical Chemistry and Laboratory Medicine, University Hospital Regensburg, Franz-Josef-Strauß-Allee 11, D-93053 Regensburg, Germany

a r t i c l e

i n f o

Article history: Received 30 July 2015 Received in revised form 4 November 2015 Accepted 5 November 2015 Available online xxx Keywords: HR-MS (high-resolution mass spectrometry) Orbitrap Oxysterols Human plasma Red blood cells Niemann–Pick type C

a b s t r a c t Oxysterols are important intermediates in numerous metabolic and catabolic pathways and their biological significance is also proven. The present paper describes a reliable and short liquid chromatography–high-resolution mass spectrometry method (LC–MS/HR-MS) for the quantification of 8 different oxysterols (24(S)-hydroxycholesterol, 25-hydroxycholesterol, 27-hydroxycholesterol, 4␤hydroxycholesterol, 7␣-hydroxycholesterol, 7␤-hydroxycholesterol, 7-ketocholesterol and cholestan3␤,5␣,6␤-triol) in human plasma and red blood cells. Oxysterols were extracted with iso-octane after saponification of esterified sterols. Due to the poor ionization efficiency of the target compounds in electrospray ionization (ESI) derivatization of the molecules has been performed with N,N-dimethylglycine (DMG). Within less than 8 min we were able to achieve baseline separation of the isobaric 24(S)-hydroxycholesterol, 25-hydroxycholesterol, 27-hydroxycholesterol, 4␤-hydroxycholesterol, 7␣-hydroxycholesterol and 7␤-hydroxycholesterol. Moreover, high mass resolution was advantageously applied to resolve quasi-isobaric interferences. The method was validated based on the recommendations of US Food and Drug Administration and the European Medicines Agency guidelines. Oxysterol concentrations were determined in human plasma and red blood cells from healthy volunteers. Furthermore, the applicability for clinical use has been proven by the analysis of oxysterols as biomarkers in Niemann–Pick type C or cerebrotendinous xanthomatosis patients. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Sterols are a biologically important class of organic molecules which contain a free OH group on their polycyclic steroid backbone. These compounds play essential roles as elements of cellular membranes, signaling, regulation and metabolism in almost all living systems [1,2]. From the clinical point of view, the most noteworthy classes of sterols in biological fluids are cholesterol and cholesteryl esters, cholesterol precursors, oxysterols, bile acids, and steroid hormones [3]. Oxysterols are oxygenated derivatives of cholesterol that are formed by enzymatic oxidation or by reactive oxygen species. Oxysterols are intermediates in the conversion of cholesterol to hormonal steroids and bile acids. They are involved in neurogenesis, protein prenylation and may activate the liver X receptor (LXR) [4,5]. Moreover, oxysterols are suspected to play a role in several diseases and pathological processes (e.g. neurodegeneration, atherosclerosis, apoptosis, necrosis, inflammation) [6–10].

∗ Corresponding author. E-mail address: [email protected] (S. Matysik).

For example, 7-ketocholesterol and cholestane-3␤,5␣,6␤-triol have been shown to be significantly elevated in various tissues and in blood plasma of patients with Niemann–Pick type C disease (NPC) [11–14]. Oxysterols, like 24(S)-hydroxycholesterol in combination with total cholesterol and 27-hydroxycholesterol are discussed as potential biomarkers in plasma and liquor for neurodegenerative diseases [15]. In addition, these side chain oxysterols together with 25-hydroxycholesterol are associated to Huntington’s and Alzheimer’s diseases [16]. 7-Ketocholesterol, 7␣- and 7␤-hydroxycholesterol, and cholestane-3␤,5␣,6␤-triol possess cytotoxic and inflammatory properties in vitro [6]. The analysis of oxysterols is challenging due to their low concentrations in biological fluids (lower ng/ml) accompanied by 103 –106 -fold excess of cholesterol. Secondly, some of the oxysterols may be generated by autoxidation of cholesterol during storage and sample preparation. This process might be partially reduced by adding antioxidants (e.g. butylated hydroxytoluene [BHT]). Several chromatographic methods coupled to MS for sterol analysis have been proposed and an extensive review about recent methodologies is given in [4]. GC-MS methods are most commonly used to detect sterols exhibiting good peak shape and high chromatographic resolution but often suffer from long run times and

http://dx.doi.org/10.1016/j.chroma.2015.11.015 0021-9673/© 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: Z. Pataj, et al., Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.11.015

G Model CHROMA-357029; No. of Pages 7

ARTICLE IN PRESS Z. Pataj et al. / J. Chromatogr. A xxx (2015) xxx–xxx

2

the necessity of large sample volumes. The separation and quantification of the target compounds are done usually after silylation [5,17–20]. However, in clinical laboratories instrumental equipment is often restricted to liquid chromatography (LC). LC coupled to tandem MS (LC–MS/MS) methods without [21,22] or with derivatization procedures have also been reported [13,23–27]. Chemical derivatization can enhance analytical sensitivity in mass spectrometry, e.g. derivatization into N,N-dimethylglycine (DMG) esters improves the mass spectrometric detection of oxysterols [28]. The derivatization with DMG is straightforward, adoptable for automatization and suitable for high throughput analysis. Successful applications after DMG derivatization have been shown only for the quantification of 7-ketocholesterol and cholestane-3␤,5␣,6␤-triol [12,14]. Here, we used DMG derivatization to analyze 8 oxysterols, including more of the biochemically relevant species. Essential requirements for application in large cohort studies are short LC run times. LC–MS approaches below 10 min run time that cover a panel of both side-chain and ring substituted oxysterols can be hardly found in literature. In LC–MS a chromatographic separation is essential since many of the target oxysterols are isobaric (with same transition masses in most cases) and separation by mass spectrometry is not possible. In the work described here, we present for the first time a baseline LC separation of the isobaric species 24(S)-, 25-, 27-, 4␤-, 7␣-, and 7␤-hydroxycholesterol to ensure reliable quantitation. Furthermore, high resolution tandem mass spectrometry is applied to enhance specificity because co-eluting compounds undergoing a similar transition can distort quantification. We illustrate the use of our LC–MS/HR-MS method for the analysis of 8 oxysterols in human plasma and red blood cells (RBCs). 2. Experimental 2.1. Chemicals and reagents Acetonitrile, ammonium acetate analytical grade, formic acid analytical grade, ethanol absolute EMSURE, and N,Ndimethylpyridin-4-amine (DMAP) were purchased from Merck (Darmstadt, Germany). Methanol (MeOH) LC–MS Chromasolv was from Fluka (Buchs, Switzerland). Chloroform ROTISOLV® was from Carl Roth GmbH (Karlsruhe, Germany). Standards and isotopically labeled standards as 24(S)-hydroxycholesterol, 4␤-hydroxycholesterol-D7, 7-ketocholesterol-D7 and cholestan3␤,5␣,6␤-triol-D7 were obtained from Toronto research chemicals (Toronto, Canada) while 25-hydroxycholesterol, 27-hydroxycholesterol, 4␤-hydroxycholesterol, 7␣-hydroxycholesterol, 7␤-hydroxycholesterol, 7-ketocholesterol, 25hydroxycholesterol-D6 and 7␤-hydroxycholesterol-D7 were purchased from Avanti Polar Lipids (Alabaster, Alabama, USA). 24(S)-Hydroxycholesterol-D10 and 27-hydroxycholesterol-D6 were from Sugaris GmbH (Munster, Germany). Cholestan-3␤,5␣,6␤-triol, butylated hydroxytoluene (BHT), N,N-dimethylglycine (DMG), N-(3-dimethylaminopropyl)-N ethylcarbodiimide (EDC), and iso-octane ACS reagent were from Sigma–Aldrich (Munich, Germany). 2.2. Sample preparation 2.2.1. Calibration samples Pooled plasma samples from healthy volunteers were diluted with 0.1 M sodium chloride in water in a ratio of 1:10. Aliquots of this diluted plasma were supplemented with a combined standard solution to obtain five calibrators in appropriate concentration ranges. The standard solution was prepared from stock solutions of oxysterol standards by dilution with MeOH. The concentration

range of the calibration samples used in this method was estimated based on our previous oxysterol studies [5,17] to cover the endogenous levels of oxysterols in human plasma and red blood cells. The concentration ranges of the oxysterols in the calibration solutions are given in Suppl. Table 1. 2.2.2. Plasma samples EDTA blood samples were collected from healthy volunteers followed by centrifugation at 4000 × g for 10 min and addition of BHT to the supernatant at a concentration of 50 ␮g/ml. The samples were stored in aliquots at −80 ◦ C. Sample preparation was based on the method described by Dzeletovic et al. [29] and the methods used in our research group in previous studies [5,17]. Briefly, 100 ␮l of human plasma or calibrator was transferred to a screw-capped vial. To this sample 10 ␮l of isotopically labeled internal standard mixture (24(S)hydroxycholesterol-D10 (2 ␮g/ml), 25-hydroxycholesterol-D6 (1 ␮g/ml), 27-hydroxycholesterol-D6 (2 ␮g/ml), 4␤hydroxycholesterol-D7 (1 ␮g/ml), 7␤-hydroxycholesterol-D7 (1 ␮g/ml), 7-Ketocholesterol-D7 (3 ␮g/ml) and Cholestan3␤,5␣,6␤-triol-D7 (1 ␮g/ml) in methanol) and 500 ␮l of freshly prepared 1 M potassium hydroxide in ethanol were added. Unless stated otherwise alkaline hydrolysis was conducted at 25 ◦ C in a water bath for 60 min under Argon atmosphere. Afterwards neutralization was performed with phosphoric acid solution. To extract lipids, 250 ␮l of 2 M sodium chloride solution was added to the sample, vortex-mixed with 1 ml of iso-octane and then centrifuged for 5 min at 4000 × g. The supernatant (iso-octane layer) was transferred by a Tecan Genesis (Männedorf, Switzerland) liquid handler to a new tube. The extraction step with 1 ml iso-octane was repeated followed by evaporation to dryness. The residue was dissolved in a mixture of 25 ␮l DMG (0.5 M) and DMAP (2 M) in CHCl3 and 25 ␮l EDC (1 M) in CHCl3 . This mixture was allowed to react for 1 h at 45 ◦ C. The excess of the derivatizing agent was deactivated with the addition of 50 ␮l MeOH at 45 ◦ C for 30 min. Evaporation was performed until dryness. The residue was redissolved in MeOH and transferred to vials for direct injection. 2.2.3. Red blood cells EDTA blood was centrifuged for 5 min at 2000 × g. The RBCs were separated from plasma and buffy coat and washed three times with physiological NaCl solution. After washing the RBCs were dispersed in NaCl solution and the concentration of RBCs was measured by a hematology analyzer (Sysmex 5000). For oxysterol analysis 100 ␮l of this suspension underwent the same preparation steps as plasma samples (as described above). 2.3. Apparatus and conditions The analysis of different oxysterols was performed on a liquid chromatography-high resolution tandem mass spectrometry (LC–MS/HR-MS) system consisted of an UltiMate 3000 XRS quaternary UHPLC pump, an UltiMate 3000 RS column oven and an UltiMate 3000 isocratic pump (Thermo Fisher Scientific Waltham, MA USA). The system was equipped with a PAL HTS-xt autosampler (CTC Analytics, Zwingen, CH) and a hybrid quadrupole-orbitrap mass spectrometer QExactive (Thermo Fisher Scientific, Bremen, Germany) equipped with a heated electrospray ionization source. 5 ␮l of the resolved samples were injected and separated on a KinetexTM 2.6 ␮m Biphenyl (50 × 2.1 mm, Phenomenex, Aschaffenburg, Germany) column at a column temperature of 30 ◦ C. The separation was achieved under gradient elution conditions. Mobile phase A consisted of methanol/water (5/95; v/v), mobile phase B was methanol/acetonitrile (10/90; v/v), both containing 0.1% formic acid and 2 mM ammonium acetate. Gradient elution started at 42% B with a flow rate of 500 ␮l/min, a linear increase to 50% B in

Please cite this article in press as: Z. Pataj, et al., Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.11.015

G Model CHROMA-357029; No. of Pages 7

ARTICLE IN PRESS Z. Pataj et al. / J. Chromatogr. A xxx (2015) xxx–xxx

3

Table 1 MS parameters of the compounds. Compound

PRM (quantifier; qualifier)*

Analyte/IS

24(S)-Hydroxycholesterol-DMG 24-Hydroxycholesterol-D10-DMG 25-Hydroxycholesterol-DMG 25-Hydroxycholesterol-D6-DMG 27-Hydroxycholesterol-DMG 27-Hydroxycholesterol-D6-DMG 4␤-Hydroxycholesterol-DMG 4␤-Hydroxycholesterol-D7-DMG 7␣-Hydroxycholesterol-DMG 7␤-Hydroxycholesterol-DMG 7␤-Hydroxycholesterol-D7-DMG 7-Ketocholesterol-DMG 7-Ketocholesterol-D7-DMG Cholestan-3␤,5␣,6␤-triol-DMG Cholestan-3␤,5␣,6␤-triol-D7-DMG

488.4 → 367.3359; 385.3465 498.5 → 395.4093; 377.3987 488.4 → 367.3359; 385.3465 494.4 → 373.3736; 391.3842 488.4 → 385.3465; 367.3359 494.4 → 391.3842; 373.3736 488.4 → 470.3993; 488.4098 495.5 → 477.4432; 495.4538 488.4 → 385.3465; 367.3359 488.4 → 385.3465; 367.3359 495.5 → 392.3904; 374.3799 486.4 → 383.3308; 486.3942 493.4 → 390.3748; 493.4381 506.4 → 506.4204; 367.3359 513.5 → 513.4643; 374.3799

Target IS Target IS Target IS Target IS Target Target IS Target IS Target IS

*

NCE

15

13 15 26 20

RT (min) 2.90 2.87 2.72 2.68 3.14 3.11 5.45 5.42 4.85 4.62 4.60 5.12 5.08 4.39 4.36

The precursor ion corresponds to the proton adduct ([M+H]+ ) of the derivatives.

3.0 min, followed by linear increase to 68% B in 2.6 min. For column cleaning the mobile phase B percentage increased to 100% within 0.1 min and maintained for 1 min. In the last equilibration step the solvent composition was changed back to 58% A within 0.1 min and hold for 1 min. To minimize contamination of the mass spectrometer, the column flow was directed only from 2.2 to 6.2 min into the mass spectrometer using a divert valve. Otherwise methanol with a flow rate of 200 ␮l/min was delivered into the mass spectrometer. The ion source was operated in the positive ion-mode using the following settings: Ion spray 3500 V, sheath gas 53, aux gas 14, sweep gas 3 and aux gas heater temperature of 438 ◦ C. Capillary temperature was set to 225 ◦ C and the S-lens RF level to 55. Data were acquired in the parallel reaction monitoring (PRM) with the following settings: Resolution 35,000, AGC target: 5e5, maximum IT 100 ms with a multiplex of 2 and quadrupole isolation window of 0.8 m/z. Data analysis was performed with TraceFinder 3.1 Clinical (Thermo Fisher Scientific), a software module which extracts target ions within ±10 ppm mass window, generates calibration lines, checks quality controls and ion ratios of quantifier to qualifier ions.

3. Results and discussion 3.1. Method optimization The aim of the current study was to develop a reliable and fast method for the quantification of oxysterols as DMG esters in biological samples by LC–MS/HR-MS. As we used a hybrid quadrupole-orbitrap instrument we applied full scan, selected ion monitoring (SIM) and parallel reaction monitoring (PRM) MS/HR-MS methods. Highly abundant signals of the protonated mono-derivatives [M+H]+ were obtained in case 24(S)-, 25-, 27, 4␤-, 7␣-, 7␤-hydroxycholesterol and cholestan-3␤,5␣,6␤-triol. Mono- and diprotonated double derivatives appeared in the spectra as well but their abundance was in contrast to other studies less significant [28]. Neither full scan nor SIM gave convincing signals due to baseline noise and unspecific signals. However, MS/HR-MS resulted in highly abundant signals with virtually no baseline noise for low concentrations. As described in Table 1, m/z 486.4, 488.4 and 506.4 were selected as precursor ions in PRM mode for 7-ketocholesterol, hydroxycholesterols, and cholestan-3␤,5␣,6␤-triol derivatives respectively. These compounds generate abundant product ions for both the sterol backbone and the DMG which has to be considered as unspecific. Cholestan-3␤,5␣,6␤-triol-DMG showed only a very low abundant fragment ion for the sterol backbone. Therefore, we optimized the collision energy in such a way to get the highest intensity for the unfragmented precursor ion.

It was not possible to separate DMG-derivatives of the isobaric oxysterols by specific mass transitions. Therefore, a chromatographic method was developed for baseline separation of isobaric derivatives to achieve reliable quantification of the target compounds. We tested different stationary phases (C18 and biphenyl core–shell) and optimized column temperature, and mobile phase composition to accomplish a fast and reliable separation (data not shown). The biphenyl core–shell stationary phase at a temperature of 30 ◦ C showed the best performance and allowed a baseline separation of the hydroxycholesterols within 8 min run-time (Fig. 1). To our knowledge this is the shortest method with baseline separation of the most relevant isobaric oxysterols [13,14,30–33]. 3.2. Method validation Method validation was performed on the basis of the recommendations of FDA [34] and EMA [35] guidelines on bioanalytical method validation. 3.2.1. Linearity Linearity was investigated by a 5-point calibration curve prepared in diluted plasma. Calibration functions were fitted by a linear model (y = mx + b) with correlation coefficients >0.99 for all oxysterols (Suppl. Table 1) and back calculated concentrations were found within ±11% of the nominal values (two examples are shown in Suppl. Fig. 1). Calibration experiments were done with triplicates at each concentration level. 3.2.2. Specificity Ion ratios of quantifier and qualifier ions were calculated in 6 different patient samples and compared to authentic standards. Patient samples were analyzed both without and with standard addition in two concentrations. The ion ratios (quantifier/qualifier) correspond to those of authentic standards closely with a maximum deviation of ±12% for all compounds. Furthermore, the high mass resolution (35.000 at m/z 200, ions were extracted with a ±10 ppm mass tolerance) provided an increased specificity. For example, plasma samples displayed peaks quasi-isobaric to cholestan-3␤,5␣,6␤-triol-DMG (m/z 506.4207) and its internal standard cholestan-3␤,5␣,6␤-triol-D7-DMG (m/z 513.4650) which were separable only by high mass resolution (Fig. 2). 3.2.3. Matrix effects To determine the matrix factors (MFs), eight plasma samples were analyzed without and with a spiked mixture of authentic standards (at a medium concentration level). The MFs were calculated as ratio of the peak area in the presence of matrix (which were

Please cite this article in press as: Z. Pataj, et al., Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.11.015

G Model CHROMA-357029; No. of Pages 7 4

ARTICLE IN PRESS Z. Pataj et al. / J. Chromatogr. A xxx (2015) xxx–xxx

Fig. 2. Separation of quasi-isobaric interferences of cholestan-3␤,5␣,6␤-triol-DMG (m/z: 506.4207); cholestan-3␤,5␣,6␤-triol-D7-DMG (m/z: 513.4650) by HR-MS. The mass spectrum extracted at 4.39 min displays interfering peaks from the matrix at m/z: 506.3213 and 513.3675.

CV of ≥20% (see one example in Suppl. Fig. 2). LOQ for all compounds were found at approximately 1 ng/ml or below except 25-hydroxycholesterol (4.5 ng/ml; Suppl. Table 1). This functional testing results in slightly higher LOQs compared to previous methods [23,24,39]. However, compared to triple-quadrupole MS/MS, LC–MS/HR-MS signals at the LOQ exhibit an infinite signal/noise ratio which allows accurate identification and integration of peaks (see Fig. 3 as example: cholestan-3␤,5␣,6␤-triol-DMG signal of a heterozygous family member of an NPC patient at LOQ level). Most importantly, the experimentally determined LOQ values were far below the normal oxysterol concentration range in human plasma samples with one exception. Cholestan-3␤,5␣,6␤-triol concentrations have been around the LOQ or slightly higher in healthy volunteers. Although LC–MS/HR-MS does not reach the LOQ of triple-quadrupole instruments it allows precise diagnosis of NPC patients since their level is significantly above LOQ. Fig. 1. Selected PRM chromatograms for the separation of oxysterol standard derivatives. Peaks: (1) 25-hydroxycholesterol-DMG; (2) 24(S)-hydroxycholesterolDMG; (3) 27-hydroxycholesterol-DMG; (4) 7␤-hydroxycholesterol-DMG; (5) 7␣hydroxycholesterol-DMG; (6) 4␤-hydroxycholesterol-DMG; (7) 7-ketocholesterolDMG; (8) cholestan-3␤,5␣,6␤-triol-DMG; for chromatographic conditions see Section 2.3.

corrected for the endogenous signal) to the peak area in absence of matrix. The IS normalized MF was calculated by dividing the MF of the analyte by the MF of the IS. The CV of the IS normalized MF was ≤20% except cholestan-3␤,5␣,6␤-triol-DMG for which a CV of 20.8% was calculated just above this criterion. (Suppl. Table 1). 3.2.4. LOQ In instrumental analytical chemistry the determination of the limit of detection (LOD) and quantification (LOQ) is usually based on the calculation of signal to noise ratios. This approach is not applicable in case of HR-MS because there is almost no baseline noise due to the high specificity as observed for 25-hydroxy vitamin D analysis [36]. Therefore, LOQ was determined by functional testing by means of repetitive measurements of serial dilutions of control and calibrator samples [37,38]. The CVs of these repetitions were determined, plotted against the concentrations and fitted by a power function. LOQ was calculated representing a

3.2.5. Carry over Carryover was assessed by injecting blank samples after five consecutive injections of the highest calibration standard. Carry over was not detected neither for the analytes nor for the internal standards.

3.2.6. Precision The precision of the method was determined in nonsupplemented plasma and spiked samples. Oxysterol standard solutions were added to native plasma prior to sample preparation to achieve two different levels. Intraassay and interassay CVs are presented in Suppl. Table 2. The CVs show good precision (<15%) except for 7-ketocholesterol, which can be formed during the preparation procedure by autoxidation of cholesterol.

3.2.7. Stability Sample stability assay was performed to investigate whether the concentration of the derivatized oxysterols is stable over time. We did not observe a decrease of the concentration in sealed vials kept at 10 ◦ C more than 15% during the investigated time (1 week).

Please cite this article in press as: Z. Pataj, et al., Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.11.015

G Model

ARTICLE IN PRESS

CHROMA-357029; No. of Pages 7

Z. Pataj et al. / J. Chromatogr. A xxx (2015) xxx–xxx

5

Fig. 3. PRM chromatogram of cholestan-3␤,5␣,6␤-triol-DMG in a heterozygous family member of an NPC patient at LOQ level.

Table 2 Concentrations of oxysterols in plasma and red blood cells (RBCs) samples. Compound

24-Hydroxycholesterol-DMG 25-Hydroxycholesterol-DMG 27-Hydroxycholesterol-DMG 4␤-Hydroxycholesterol-DMG 7␣-Hydroxycholesterol-DMG 7␤-Hydroxycholesterol-DMG 7-Ketocholesterol-DMG Cholestan-3␤,5␣,6␤-triol-DMG

Concentration in plasma samples (n = 32)

Concentration in erythrocyte samples (n = 32)

Median (ng/ml)

2.5–97.5 percentile

Median (pg/1 × 106 RBCs)

2.5–97.5 percentile

38.1 29.7 74.1 17.8 29.3 6.9 18.8
19.1–75.5
7.4 7.1 2.0 12.2 34.0 18.2 138.7


3.3. Application 3.3.1. Oxysterols in human plasma and red blood cells The oxysterol plasma concentrations of 32 healthy volunteers are summarized in Table 2. These concentrations are comparable with the literature and to previous own data analyzed by GC-MS/MS [17,21,40]. However, absolute values can show some variation depending on different methods of sample preparation and instrumental analysis. A direct comparison to other cohort studies must be interpreted with care as also pointed out by McDonald et al. [21]. A serious problem of correct quantitation of oxysterols is autoxidation of cholesterol. The addition of BHT and sample preparation under argon atmosphere should have mitigated the formation of oxidation products. However, the method allows relative comparisons between study groups, for example to assist in diagnosis of NPC or CTX. From our experience mild conditions like hydrolysis at room temperature diminish the degree of autoxidation and are preferable for analysis of oxidation sensitive compounds. Furthermore, the oxysterol content in red blood cells was evaluated in the same study group. All oxysterol derivatives were found above the LOQ except for cholestan-3␤,5␣,6␤-triol-DMG which was not detectable in most RBC samples. The analysis of oxysterols in human red blood cells is not very common. Nevertheless, the analysis of RBCs is rather straightforward. The necessary amount of cells is in most cases available from residual material after blood cell counting. Here we have shown that 7-ketocholesterol is the most prominent oxysterol in RBCs. RBCs are the fraction of human blood


cells with the highest amount of free cholesterol [41] and a high level of 7-ketocholesterol may result from oxidation of cholesterol. A recent study has suggested that oxysterols enhance eryptosis, especially 7-ketocholesterol and cholestan-3␤,5␣,6␤-triol might induce oxidative stress, Ca++ influx and hemolysis [42]. Finally, the method could be a valuable tool to provide deeper insights into the role of oxysterols in intracellular processes of RBCs. 3.3.2. Oxysterols in Niemann–Pick type C disease Recent reports demonstrated the clinical relevance of oxysterols in Niemann–Pick type C disease [13,14,43]. Therefore, LC–MS/HR-MS was applied to determine oxysterols in four confirmed NPC patients (age 30 ± 4) and their heterozygous family members (age 53 ± 10). BHT stabilized EDTA plasma was used which was stored prior to analysis at −80 ◦ C. We found for the NPC patients 90 ± 13 ng/ml, 173 ± 24 ng/ml, and 66 ± 22 ng/ml for 7␤-hydroxycholesterol, 7-ketocholesterol and cholestan-3␤,5␣,6␤-triol, respectively. The concentrations of the heterozygous family members were 14 ± 3 ng/ml, 41 ± 8 ng/ml, and around the LOQ (1.13 ng/ml) for 7␤-hydroxycholesterol, 7-ketocholesterol and cholestan-3␤,5␣,6␤-triol. Typical chromatograms of an NPC patient and a heterozygous family member are given in Fig. 4. The concentrations of 7-ketocholesterol and cholestan-3␤,5␣,6␤-triol in the plasma of the NPC patients are significantly elevated and in the range of those published recently [12,14]. Furthermore we showed that also 7␤-hydroxycholesterol levels are significantly higher in NPC patient’s plasma.

Please cite this article in press as: Z. Pataj, et al., Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.11.015

G Model CHROMA-357029; No. of Pages 7

ARTICLE IN PRESS Z. Pataj et al. / J. Chromatogr. A xxx (2015) xxx–xxx

6

Fig. 4. Selected PRM chromatograms for the separation of oxysterol derivatives in NPC patients. (A) Heterozygous family member of the patient; (B) Niemann–Pick type C patient; elution order: (1) 25-hydroxycholesterol-DMG; (2) 24(S)-hydroxycholesterol-DMG; (3) 27-hydroxycholesterol-DMG; (4) cholestan-3␤,5␣,6␤-triol-DMG; (5) 7␤hydroxycholesterol-DMG; (6) 7␣-hydroxycholesterol-DMG; (7) 7-ketocholesterol-DMG; (8) 4␤-hydroxycholesterol-DMG; for chromatographic conditions see Section 2.3.

3.3.3. Oxysterols in cerebrotendinous xanthomatosis (CTX) Cerebrotendinous xanthomatosis (CTX) is a rare metabolic disease caused by mutations in the CYP27A1 gene encoding sterol 27-hydroxylase [44,45]. This catalytic step is crucial for the conversion of cholesterol into bile acids by the liver. Its deficiency leads to storage of intermediate metabolites in the nervous system and other tissues. Diagnostic strategies might benefit from a combination of biochemical profiling and genetic analysis in potential affected individuals. The oxysterol profiling in plasma could play a pivotal role in the diagnosis of this disease since it offers a cheap and short option. Here, we analyzed EDTA plasma of two CTX patients. In accordance to the literature [46] 27-hydroxycholesterol was not detectable in these samples. 4. Conclusion A LC–MS/HR-MS method for separation and quantification of 8 oxysterols from human plasma and red blood cell samples has been presented. The method was optimized to support extensive investigation of these compounds in rare diseases such as NPC, CTX and neurodegenerative or cardiovascular diseases. This method has been solved the challenging task of the separation of structural isomers such 24(S)-, 25-, 27- and 7␣-, 7␤-hydroxycholesterols. The short run time (7.8 min) allows the quantification of a large number of samples daily which is important for clinical studies. Further applications might be seen in oxysterol analysis of cerebrospinal fluid, tissues and cell lysate samples. Acknowledgements Z.P. wishes to express his thanks for the Alexander von Humboldt Foundation to support his work. The authors are grateful to Jolanthe Aiwanger for her expert technical assistance. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chroma.2015.11. 015.

References [1] D.W. Russell, Fifty years of advances in bile acid synthesis and metabolism, J. Lipid Res. 50 (2009) S120–S125. [2] M.S. Brown, J.L. Goldstein, Cholesterol feedback: from Schoenheimer’s bottle to Scap’s MELADL, J. Lipid Res. 50 (2009) 1255. [3] W.D. Nes, Biosynthesis of cholesterol and other sterols, Chem. Rev. 111 (2011) 6423–6451. [4] W.J. Griffiths, P.J. Crick, Y.Q. Wang, Methods for oxysterol analysis: past, present and future, Biochem. Pharmacol. 86 (2013) 3–14. [5] S. Matysik, G. Schmitz, Application of gas chromatography–triple quadrupole mass spectrometry to the determination of sterol components in biological samples in consideration of the ionization mode, Biochimie 95 (2013) 489–495. [6] G.J. Schroepfer, Oxysterols: modulators of cholesterol metabolism and other processes, Physiol. Rev. 80 (2000) 361–554. [7] G. Poli, F. Biasi, G. Leonarduzzi, Oxysterols in the pathogenesis of major chronic diseases, Redox Biol. 1 (2013) 125–130. [8] N.J. Spann, C.K. Glass, Sterols and oxysterols in immune cell function, Nat. Immunol. 14 (2013) 893–900. [9] I. Bjorkhem, Do oxysterols control cholesterol homeostasis? J. Clin. Investig. 110 (2002) 725–730. [10] I. Bjorkhem, U. Diczfalusy, Oxysterols – friends, foes, or just fellow passengers? Arterioscler. Thromb. Vasc. Biol. 22 (2002) 734–742. [11] F.D. Porter, D.E. Scherrer, M.H. Lanier, S.J. Langmade, V. Molugu, S.E. Gale, D. Olzeski, R. Sidhu, D.J. Dietzen, R. Fu, C.A. Wassif, N.M. Yanjanin, S.P. Marso, J. House, C. Vite, J.E. Schaffer, D.S. Ory, Cholesterol oxidation products are sensitive and specific blood-based biomarkers for Niemann–Pick C1 disease, Sci. Transl. Med. 2 (2010). [12] X.T. Jiang, R. Sidhu, F.D. Porter, N.M. Yanjanin, A.O. Speak, D.T.T. Vruchte, F.M. Platt, H. Fujiwara, D.E. Scherrer, J. Zhang, D.J. Dietzen, J.E. Schaffer, D.S. Ory, A sensitive and specific LC–MS/MS method for rapid diagnosis of Niemann–Pick C1 disease from human plasma, J. Lipid Res. 52 (2011) 1435–1445. [13] S. Boenzi, F. Deodato, R. Taurisano, D. Martinelli, D. Verrigni, R. Carrozzo, E. Bertini, A. Pastore, C. Dionisi-Vici, D.W. Johnson, A new simple and rapid LC-ESI-MS/MS method for quantification of plasma oxysterols as dimethylaminobutyrate esters. Its successful use for the diagnosis of Niemann–Pick type C disease, Clin. Chim. Acta 437 (2014) 93–100. [14] G. Klinke, M. Rohrbach, R. Giugliani, P. Burda, M.R. Baumgartner, C. Tran, M. Gautschi, D. Mathis, M. Hersberger, LC–MS/MS based assay and reference intervals in children and adolescents for oxysterols elevated in Niemann–Pick diseases, Clin. Biochem. 48 (2015) 596–602. [15] V. Leoni, T. Masterman, F.S. Mousavi, B. Wretlind, L.O. Wahlund, U. Diczfalusy, J. Hillert, I. Bjorkhem, Diagnostic use of cerebral and extracerebral oxysterols, Clin. Chem. Lab. Med. 42 (2004) 186–191. [16] G. Lorbek, M. Lewinska, D. Rozman, Cytochrome P450s in the synthesis of cholesterol and bile acids – from mouse models to human diseases, FEBS J. 279 (2012) 1516–1533. [17] S. Matysik, H.H. Klünemann, G. Schmitz, Gas chromatography-tandem mass spectrometry method for the simultaneous determination of oxysterols, plant sterols, and cholesterol precursors, Clin. Chem. 58 (2012) 1557–1564. [18] M. Menendez-Carreno, C. Garcia-Herreros, I. Astiasaran, D. Ansorena, Validation of a gas chromatography-mass spectrometry method for the analysis

Please cite this article in press as: Z. Pataj, et al., Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.11.015

G Model CHROMA-357029; No. of Pages 7

ARTICLE IN PRESS Z. Pataj et al. / J. Chromatogr. A xxx (2015) xxx–xxx

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

[31]

of sterol oxidation products in serum, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 864 (2008) 61–68. H.S.M. Ahmida, P. Bertucci, L. Franzo, R. Massoud, C. Cortese, A. Lala, G. Federici, Simultaneous determination of plasmatic phytosterols and cholesterol precursors using gas chromatography-mass spectrometry (GC-MS) with selective ion monitoring (SIM), J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 842 (2006) 43–47. J. Acimovic, A. Lovgren-Sandblom, K. Monostory, D. Rozman, M. Golicnik, D. Lutjohann, I. Bjorkhem, Combined gas chromatographic/mass spectrometric analysis of cholesterol precursors and plant sterols in cultured cells, J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 877 (2009) 2081–2086. J.G. McDonald, D.D. Smith, A.R. Stiles, D.W. Russell, A comprehensive method for extraction and quantitative analysis of sterols and secosteroids from human plasma, J. Lipid Res. 53 (2012) 1399–1409. C. Helmschrodt, S. Becker, J. Schroter, M. Hecht, G. Aust, J. Thiery, U. Ceglarek, Fast LC–MS/MS analysis of free oxysterols derived from reactive oxygen species in human plasma and carotid plaque, Clin. Chim. Acta 425 (2013) 3–8. A. Honda, K. Yamashita, H. Miyazaki, M. Shirai, T. Ikegami, G.R. Xu, M. Numazawa, T. Hara, Y. Matsuzaki, Highly sensitive analysis of sterol profiles in human serum by LC-ESI-MS/MS, J. Lipid Res. 49 (2008) 2063–2073. A. Honda, T. Miyazaki, T. Ikegami, J. Iwamoto, K. Yamashita, M. Numazawa, Y. Matsuzaki, Highly sensitive and specific analysis of sterol profiles in biological samples by HPLC-ESI-MS/MS, J. Steroid Biochem. Mol. Biol. 121 (2010) 556–564. W.J. Griffiths, Y.Q. Wang, G. Alvelius, S. Liu, K. Bodin, J. Sjovall, Analysis of oxysterols by electrospray tandem mass spectrometry, J. Am. Soc. Mass Spectrom. 17 (2006) 341–362. P.J. Crick, T.W. Bentley, J. Abdel-Khalik, I. Matthews, P.T. Clayton, A.A. Morris, B.W. Bigger, C. Zerbinati, L. Tritapepe, L. Iuliano, Y.Q. Wang, W.J. Griffiths, Quantitative charge-tags for sterol and oxysterol analysis, Clin. Chem. 61 (2015) 400–411. K. Karu, M. Hornshaw, G. Woffendin, K. Bodin, M. Hamberg, G. Alvelius, J. Sjovall, J. Turton, Y.Q. Wang, W.J. Griffiths, Liquid chromatography–mass spectrometry utilizing multi-stage fragmentation for the identification of oxysterols, J. Lipid Res. 48 (2007) 976–987. X.T. Jiang, D.S. Ory, X.L. Han, Characterization of oxysterols by electrospray ionization tandem mass spectrometry after one-step derivatization with dimethylglycine, Rapid Commun. Mass Spectrom. 21 (2007) 141–152. S. Dzeletovic, O. Breuer, E. Lund, U. Diczfalusy, Determination of cholesterol oxidation products in human plasma by isotope dilution-mass spectrometry, Anal. Biochem. 225 (1995) 73–80. V.V.R. Bandaru, N.J. Haughey, Quantitative detection of free 24Shydroxycholesterol, and 27-hydroxycholesterol from human serum, BMC Neurosci. 15 (2014) 9. H. Roberg-Larsen, C. Vesterdal, S.R. Wilson, E. Lundanes, Underivatized oxysterols and nano LC–ESI–MS: a mismatch, Steroids 99 (2015) 125–130.

7

[32] R. Karun, I. Christen, A.W. Sailer, F. Bitsch, J. Zhang, Detection of dihydroxycholesterols in human plasma using HPLC-ESI-MS/MS, Steroids 99 (2015) 131–138. [33] H.F. Schott, D. Lutjohann, Validation of an isotope dilution gas chromatographymass spectrometry method for combined analysis of oxysterols and oxyphytosterols in serum samples, Steroids 99 (2015) 139–150. [34] Guidance for Industry Bioanalytical Method Validation, U.S. Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research, Center for Veterinary Medicine, 2001. [35] Guideline on Bioanalytical Method Validation, European Medicines Agency – Committee for Medicinal Products for Human Use, 2011. [36] G. Liebisch, S. Matysik, Accurate and reliable quantification of 25-hydroxyvitamin D species by liquid chromatography–high-resolution tandem mass spectrometry, J. Lipid Res. 56 (2015) 1234–1239. [37] D.A. Armbruster, T. Pry, Limit of blank, limit of detection and limit of quantitation, Clin. Biochem. Rev./Aust. Assoc. Clin. Biochem. 29 (Suppl. 1) (2008) S49–S52. [38] I.D. Hay, M.F. Bayer, M.M. Kaplan, G.G. Klee, P.R. Larsen, C.A. Spencer, L. Wartofsky, American-thyroid-association assessment of current free thyroidhormone and thyrotropin measurements and guidelines for future clinical assays, Clin. Chem. 37 (1991) 2002–2008. [39] A. Honda, K. Yamashita, T. Hara, T. Ikegami, T. Miyazaki, M. Shirai, G.R. Xu, M. Numazawa, Y. Matsuzaki, Highly sensitive quantification of key regulatory oxysterols in biological samples by LC–ESI–MS/MS, J. Lipid Res. 50 (2009) 350–357. [40] I. Bjorkhem, A. Lovgren-Sandblom, F. Piehl, M. Khademi, H. Pettersson, V. Leoni, T. Olsson, U. Diczfalusy, High levels of 15-oxygenated steroids in circulation of patients with multiple sclerosis: fact or fiction? J. Lipid Res. 52 (2011) 170–174. [41] K. Leidl, G. Liebisch, D. Richter, G. Schmitz, Mass spectrometric analysis of lipid species of human circulating blood cells, Biochim. Biophys. Acta: Mol. Cell Biol. Lipids 1781 (2008) 655–664. [42] L. Tesoriere, A. Attanzio, M. Allegra, A. Cilla, C. Gentile, M.A. Livrea, Oxysterol mixture in hypercholesterolemia-relevant proportion causes oxidative stressdependent eryptosis, Cell. Physiol. Biochem. 34 (2014) 1075–1089. [43] G. Klinke, D. Mathis, M. Hersberger, Oxysterols as biomarkers for Niemann–Pick type C disease, Clin. Biochem. 47 (2014) 772. [44] J.J. Cali, D.W. Russell, Characterization of human sterol 27-hydroxylase: a mitochondrial cytochrome-P-450 that catalyzes multiple oxidation reactions in bile-acid biosynthesis, J. Biol. Chem. 266 (1991) 7774–7778. [45] J.J. Cali, C.L. Hsieh, U. Francke, D.W. Russell, Mutations in the bile-acid biosynthetic enzyme sterol 27-hydroxylase underlie cerebrotendinous xanthomatosis, J. Biol. Chem. 266 (1991) 7779–7783. [46] B.J. Koopman, J.C. Vandermolen, B.G. Wolthers, J.B. Vanderpas, Determination of some hydroxycholesterols in human-serum samples, J. Chromatogr.: Biomed. Appl. 416 (1987) 1–13.

Please cite this article in press as: Z. Pataj, et al., Quantification of oxysterols in human plasma and red blood cells by liquid chromatography high-resolution tandem mass spectrometry, J. Chromatogr. A (2015), http://dx.doi.org/10.1016/j.chroma.2015.11.015