Clinica Chimica Acta 436 (2014) 268–272
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Development of a liquid chromatography–mass spectrometry method for the determination of the neurotoxic quinolinic acid in human serum Andreas Meinitzer a,⁎, Andreas Tomaschitz b,c, Stefan Pilz d, Manfred Truber a, Gabriele Zechner a, Martin Gaksch d, Barbara Prietl d, Gerlies Treiber d, Michaela Schwarz e, Andreas Baranyi f a
Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Graz, Austria Department of Cardiology, Medical University of Graz, Graz, Austria Specialist Clinic for Rehabilitation PV Bad Aussee, Bad Aussee, Austria d Department of Internal Medicine, Division of Endocrinology and Metabolism, Medical University of Graz, Graz, Austria e Division of Surgical Research, Department of Surgery, Medical University of Graz, Graz, Austria f Department of Psychiatry, Medical University of Graz, Graz, Austria b c
a r t i c l e
i n f o
Article history: Received 9 May 2014 Received in revised form 13 June 2014 Accepted 13 June 2014 Available online 21 June 2014 Keywords: Quinolinic acid Mass spectrometry Kynurenin pathway
a b s t r a c t Background: Quinolinic acid (QA) is thought to be one of the most important metabolites of the kynurenine pathway with the highest biological activity in apoptotic responses and neurodegenerative diseases. The determination of QA might be of clinical relevance in different patient groups, but currently, only a few laborious methods with high levels of sample volume consumption are available. Methods: We developed and validated a simple liquid chromatography–tandem mass spectrometric (LC–tandem MS) method for the determination of QA in human serum with low sample volume requirements. Results: The presented method provides high sample throughput with 25 μL aliquots and works in the positive electrospray ionization (ESI) mode. A commercially available QA-d3 was used as internal standard. Specific transitions for QA and QA-d3 were m/z 280 → m/z 78 and m/z 283 → m/z 81, respectively. The intra- and inter-assay coefficients of variation (CVs) were all below 10%. Applying this method, in 50 healthy humans a mean serum concentration of QA of 350 ± 167 nmol/L (mean ± SD) was determined. Conclusion: The described method is suitable for large clinical trials, which is of potential clinical importance to elucidate the function of QA and its relationship to different disease patterns and may be applicable for clinical laboratory routine. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Quinolinic acid (QA) is generated within the kynurenine pathway resulting from L-tryptophan catabolism and functions as the precursor of nicotinamide adenine dinucleotide (NAD). Along with other acid compounds of the kynurenine pathway, kynurenic acid and anthranilic acid, QA is suggested to be the most important in terms of biological activity [1]. It is involved in complex interrelationships with inflammatory and apoptotic responses associated with neuronal cell damage and death in the central nervous system [2,3]. Production of QA in the brain is increased in inflammatory and neurodegenerative diseases [4, 5]. There is growing evidence that QA is involved in the development of brain diseases such as Alzheimer's disease [6–9], HIV-associated neurocognitive disorders [10–12], Parkinson's disease [13], motor neuron diseases [14], Huntington's disease [15], multiple sclerosis [16] and major psychiatric disorders [17]. High serum concentrations of QA were found in chronic hemodialysis patients. The accumulation of QA ⁎ Corresponding author. Tel.: +43 316 385 83988; fax: +43 316 385 13419. E-mail address: [email protected] (A. Meinitzer).
http://dx.doi.org/10.1016/j.cca.2014.06.010 0009-8981/© 2014 Elsevier B.V. All rights reserved.
in uremic blood may be involved in the pathogenesis of anemia, a suppressed immune system, and uremic encephalopathy [18]. An increase in serum levels leads to enhanced QA concentrations in brain tissue, probably due to an increase in blood–brain barrier transport [19]. Several methods for the quantification of QA in biological fluids, in which different separation and detection methods, have been described. Methods based on gas chromatography [20,21] and liquid chromatography [22] are time-consuming and need several different reagents with partly unpleasant properties. The detection sensitivity of QA, which has two carboxylic groups side by side and thus is not fluorescent nor electrochemically active, is rather low. Some authors esterified the carboxyl groups and measured QA by the use of gas chromatography and mass spectrometry [23–26]. Odo et al. refined a method of Mawatari [27] and converted QA in the presence of hydrogen peroxide and the catalytic activity of horseradish peroxidase into a fluorescent compound for determining trace amounts of QA [28]. Recently, Moeller et al. reported a method for underivatized QA measurement using liquid chromatography– and electrospray ionization (ESI)–tandem mass spectrometry in a sample volume of 800 μL [29]. Polar compounds like QA have lower ESI responses than less polar compounds. Mass
A. Meinitzer et al. / Clinica Chimica Acta 436 (2014) 268–272 Table 1 Characteristics of the quinolinic acid method. Quinolinic acid Working range [nmol/L] Calibration curve Slope Intercept Correlation r2 Intra-day precision n = 6 Mean [nmol/L] SD [nmol/L] Coefficient of variation [%] Inter-day precision n = 20 Mean [nmol/L] SD [nmol/L] Inter-day variability [%] Recovery (mean extraction efficiency) [%, ±SD] Limit of quantification (LOQ) [nmol/L] SD [nmol/L] Coefficient of variation [%] Limit of detection (LOD) [nmol/L] 3-fold SD of the baseline noise Stability in the matrix tested
50–5000 0.03725 0.00224 0.9998 Low 225 10.2 4.5
High 725 8.5 1.2
229 17.0 7.2 95.0 ± 7.7
752 47.3 6.3
elements were controlled by Xcalibur software, which also managed the Voyager TSQ Quantum triple quadrupole instrument (Thermo). LCquan software (Thermo) was used to calculate results from the raw data. To test the concentrations in humans we recruited 50 healthy subjects aged at least 18 years. Exclusion criteria were pregnancy, participation in other interventional clinical trials and any disease requiring medical treatment. Laboratory measurements for basic patient characteristics (LDL-cholesterol, HDL-cholesterol, glucose and triglycerides) were performed by routine methods. The study was performed at the Division of Endocrinology and Metabolism (Department of Internal Medicine) at the Medical University of Graz. The study adheres to the Declaration of Helsinki and ethical approval was obtained from the ethics committee at the Medical University of Graz, Austria. 3. Sample preparation
50 4.7 9.3 15 2 days
spectrometers connected to liquid chromatography systems are not able to create sufficient and stable ionization response to detect QA in small amounts of biological samples. Our aim therefore was to develop a simple derivatization method for QA to enhance the ionization response, which enables QA quantification in small volume portions of serum with liquid chromatography–mass spectrometers. 2. Materials and methods HPLC grade methanol and analytical grade sodium hydroxide were purchased from Merck (Darmstadt, Germany). Sodium tetraborate, QA (2,3 pyridinedicarboxylic acid), formic acid, 3 M hydrogen chloride-1butanol solution and activated charcoal all in analytical grade were obtained from Sigma-Aldrich (Austria). CHROMASOLV-type liquid chromatography–mass spectrometry water was purchased from Fluka (Hamburg, Germany). Sample clean-up was performed by solid-phase extraction on strong anion-exchange columns (Strata XA 30 mg/1 mL, Phenomenex Aschaffenburg, Germany) using QA-d3 (Toronto Research Chemicals, North York, Ontario, Canada) as internal standard. The cartridges placed in 10 mL plastic tubes (100 × 16 mm, Greiner Holding AG, Kremsmünster, Austria) were loaded and eluted with gravitation on an Eppendorf centrifuge 5810 R (Eppendorf, Hamburg, Germany) at different times (s) and g values. In the sample procedure the exact times and g values are given in parentheses. Chromatographic separation was done on a reversed phase column Synergi Hydro, RP 100A (100 × 3 mm, 2.5 μm) embedded in a Thermo Surveyor system (Thermo Instruments, San Jose, California, USA), which consists of a quaternary pump, an autosampler and a column oven. All chromatographic Table 2 Patients characteristics and quinolinic acid concentration. Number of patients Females [%] Age [yrs] Body mass index [kg/m2] Systolic blood pressure [mm Hg] Diastolic blood pressure [mm Hg] Glucose [mmol/L] HDL-cholesterol [mmol/L] LDL-cholesterol [mmol/L] Triglycerides [mmol/L] Quinolinic acid [nmol/L]
269
50 64 31 ± 8 23.3 ± 4.3 129 ± 15 84 ± 10 4.89 ± 0.92 1.84 ± 0.37 2.46 ± 0.57 1.08 ± 0.49 350 ± 167
Continuous variables are presented as means ± standard deviation (SD).
The Strata XA cartridges were initially washed with 1 mL methanol followed by 1 ml of borate buffer (25 mmol/l, pH = 9.5) (both steps 60 s, 1500 g). Twenty-five-microliter serum samples, calibrators or controls were mixed with 10 μL internal standard (3 μmol/L in 50% [v/v] methanol/water) and 950 μL borate buffer, and passed through the columns (90 s, 800 g). The loaded cartridges were then washed with 1 mL borate buffer, 1 mL water, and two times with 1 mL methanol (all four steps with 60 s, 1500 g). After the last washing step the columns were centrifuged for 60 s with 4000 g to remove most of the liquid. The cartridges were then dried for 20 s by applying 2 bar compressed air. Elution of the absorbed analytes in tapered glass tubes was performed with 500 μL 3 M hydrogen chloride in 1-butanol. The eluates were heated for 30 min at 80° for allowing esterification. Then the eluates were evaporated with a stream of compressed air at 80° on a Turbovap System (American Laboratory Trading, East Lyme, Connecticut, USA). The dried residue was reconstituted with 200 μL of mobile phase B, vortexed for few seconds, transferred to auto sampler vials and 50 μL were injected in the LC–MS/MS system. 4. Chromatographic conditions and MS/MS detection parameters Separation was done with the above-mentioned column with dual gradient elution. Mobile phase A consisted of 0.1% formic acid in water, mobile phase B was a mixture of 35% water, 65% methanol and 0.1% formic acid (v/v/v). After injection, a mixture of 30% A and 70% B with a flow rate of 300 μL/min was pumped through the column for 4 min. Then a linear gradient started to 100% B with a duration of 2 min followed by a period of 2 min with 100% B for washing out and a period of 2 min for equilibration with the initial condition. The mass instrument was set in multiple-reaction monitoring mode. The analytes were identified in the positive-ion mode and determined by means of characteristic product ions formed from protonated molecules by collision induced dissociation (CID) with the following fragments: m/z 280.1 → 78 for QA with qualifier m/z 280.1 → 106 and m/z 283.1 → 81 for QA d3 as internal standard with qualifier 283.1 → 109.14. Settings for the ion source were as follows: spray voltage 4000 V, sheath gas pressure 60 psi, ion sweep gas pressure 0 psi, aux gas pressure 15 psi, capillary temperature of 350 °C, source fragmentation energy 22 V, scan time 0.02 s, scan width m/z 0.02, peak width Q1 0.7, peak width Q3 0.7, skimmer offset 5 V, and collision gas pressure 1.5 mTorr. Validation was done according to FDA guidelines for bioanalytical method validation [30]. The tested validation parameters were linearity, intra-assay and inter-assay precision, matrix effects, recovery, carry over and stability. The within- and between-day imprecisions were assessed by performing 6 replicate analyses of quality control samples within one day and within 6 consecutive days, respectively. The limit of detection (LOD) and the limit of quantification (LOQ) were defined as the minimum concentration showing a signal of at least 3 times and 10 times respectively greater at the retention time of the analyte
270
A. Meinitzer et al. / Clinica Chimica Acta 436 (2014) 268–272
Fig. 1. Chromatogram of a patient with 540 nmol/L quinolinic acid. Every peak has over 30 measure points. Peaks are shown without smoothing. Left: Derivate Dibutyl 2,3 pyridinedicarboxylate with analyte and qualifier fragments. Right: internal standard derivate with analyte and qualifier fragments. NL: normalized scale; SRMms2, single reacting mode and transmission mass ranges; A, Area under the curve; S/N, signal to noise ratio.
than the average background noise of an unspiked blank (only containing IS).
5. Standardization and quality control To produce calibration curves and to determine the recovery of the method, a drug-free serum pool obtained from the local Department of Transfusion Medicine was used. To remove endogenous QA 100 mL of the serum was agitated overnight with 5 g of activated charcoal at 4 °C. After centrifugation at 25,000 g for 30 min at 4 °C, the supernatant was treated with a portion of 5 g Strata XA anion exchange material and agitated once more for 1 h. After the centrifugation of the matrix we
measured the QA concentration in the treated pool. Despite a second extraction step a small amount of QA in the range of 50 nmol/L remained in the human serum matrix. We determined this concentration with the standard addition method and included this blank in the calculation of the calibrator values. A stock solution of QA (1 mmol/L) was prepared in water and diluted with the above described matrix to the appropriate value. Standard series were prepared with concentrations of 0; 100; 250; 500; 1000; and 2000 nmol/L. For precision control, we produced two pooled serum samples: one from patients attending the gynecological clinic, because the levels in these patients are low (pooled control serum in the low range), and another one from patients of the Department for Angiology, which are mostly high (pooled control serum in the high range). Aliquots of these controls were analyzed in each run after ten specimens. Before the series were performed, the controls were analyzed 10 times and the mean value was taken as the reference value. Stability of samples: As specimens are usually handled and transported to the laboratory at room temperature and we investigated stability at room temperature. Instability is usually easier seen at room temperature, which is the worst stress condition for the specimens. We stored 3 patient samples at room temperature with daylight exposure and analyzed these after time intervals of 0, 4, 24 and 48 h. An analyte was considered stable when the measured concentration was in the range of the inter-day standard deviation of the method.
6. Results
Fig. 2. Representative ion suppression chromatograms. A solution of dibutyl 2,3 pyridinedicarboxylate (150 μmol/L) was mixed via a “tee” to the eluent after the column into the mass detector at a constant rate of 5 μL/min. Mass chromatograms were taken after injection of an extract of a serum pool from 10 patients. The upper chromatogram is a patient sample (2500 nmol/L) without post-infusion of QA and serves as reference. NL: normalized scale; SRMms2, single reacting mode and transmission mass ranges.
The characteristics of the analytical method as a result of the method evaluation are presented in Table 1. Calibration curves were linear throughout the selected ranges and the coefficient of determination (r2) exceeded 0.990. The standard deviations for imprecision and accuracy results were all highly acceptable. The extraction efficiencies from the biological matrix to a protein-free solution, were 95 ± 7.7%. Based on the examination of chromatograms of serum samples – after injection of pathologically high samples – without analytes, the calculated carryover effects were below 1%. The mean serum concentration of QA in healthy humans was 350 ± 167 nmol/L (n = 50). Clinical
A. Meinitzer et al. / Clinica Chimica Acta 436 (2014) 268–272
characteristics of the participants were in the reference range (Table 2). In women, QA concentrations (381 ± 198 nmol/L, n = 32) were higher compared to men (297 ± 68 nmol/L, n = 18), but the difference did not reach statistical significance (p = 0.090). There was no significant association between serum QA concentration and age. Fig. 1 provides a typical chromatogram of a patient and contains pure raw data without smoothing. All peaks show a symmetrical form and have more than 25 measurement points. Injection of an extracted serum pool in the post column-infusion experiment (Fig. 2) showed valleys at the solvent front (1.5 to 2.5 min) and minimum response valleys at 5.15 and 5.85 min, which indicates interactions with eluted matrix compounds or elution of extraction solution ions. At the retention times of QA (6.4 min) no influence of the signal was observed.
7. Discussion The aim of the present study was to develop a simple volume-saving method for the expeditious determination of QA in human serum. In contrast to a recently published LC–MS/MS method for QA determination in rat plasma with the use of 800 μL sample volume [29], the present method on hand needs only 25 μL. Our developed method would also work with smaller volumes, but due to better robustness and convenience 25 μL might be appropriate. One limitation of the present method is that it does not work in EDTA plasma. EDTA is highly concentrated in the tubes for taking blood, usually at 1 to 10 mg, depending on their volume. The molar ratio to QA in plasma is approximately 1:10.000. EDTA probably displaces QA from the binding sites of the anionic exchange column in the sample preparation. The recovery of QA decreases to zero. Interestingly, this effect is not observed during the measurement of monocarboxylic kynurenic acid, which can be easily extracted from plasma or serum with the same method. Two aspects should be considered in regard to the sample preparation: (1) The SPE cartridges must be highly dried before the eluting step is carried out; and (2) The derivatization agent must be free of water. Although a lot of literature is found about QA, there are only rare data of QA measured in humans. Using gas chromatography–mass spectrometry [32] Schwarz et al. [31] determined QA in twenty patients with Alzheimer's disease [median: 560 nmol/L (370–1060 nmol/L)] and in humans with subjective cognitive impairment [median: 730 nmol/L (390–1220 nmol/L)]. Pawlak et al. [33] separated QA on an ion exchange chromatography column and detected QA with an ultraviolet detector with the need of 2 mL of serum. In this investigation the median values of QA in patients with chronic kidney disease and 20 healthy controls were 2050 (550–8000) nmol/L and 150 (70–630) nmol/L, respectively. Markedly increased concentrations of QA in uremic patients have been already observed [18]. The mean values of QA in uremic patients and healthy controls were 9200 ± 5300 nmol/L (n = 54) and 600 (±300) nmol/L (n = 10), respectively. Using a not precisely described method Raison [34] in healthy controls and in patients with hepatitis C virus documented a mean QA value of 330 nmol/L (n = 27). Heyes [35] in patients living with human immunodeficiency virus (HIV; n = 6) found clearly elevated levels of QA (mean: 1393 ± 108 nmol/L as compared to QA values of healthy controls (432 ± 64 nmol/L, n = 11). The mean values of QA measured in healthy controls conform to the previously determined QA concentrations in healthy subjects. Interestingly, these groups demonstrated highly significant correlations between QA serum and QA cerebrospinal concentrations. However the QA levels in the cerebrospinal fluid are 10-fold lower than those in serum. It can therefore be speculated that circulating QA concentrations reflect QA concentrations in the CSF. In conclusion, we developed a LC–tandem MS method which allows the quantification of QA in only 25 μL of human serum. A commercially available stable isotope labeled QA served as internal standard. The method is suitable for large clinical trials, which is of crucial importance
271
to elucidate the status of QA in neurodegenerative diseases and may be applicable for clinical laboratory routine. References [1] Guillemin GJ. Quinolinic acid, the inescapable neurotoxin. FEBS J 2012;279(8):1356–65. [2] Stone TW. Endogenous neurotoxins from tryptophan. Toxicon 2001;39:61–73. [3] Stone TW, Perkins MN. Quinolinic acid: a potent endogenous excitant at amino acid receptors in CNS. Eur J Pharmacol 1981;72:411–2. [4] Heyes MP, Saito K, Crowley JS, et al. Quinolinic acid and kynurenine pathway metabolism in inflammatory and non-inflammatory neurological disease. Brain 1992; 115:1249–73. [5] Heyes MP, Saito K, Major EO, Milstien S, Markey SP, Vickers JH. A mechanism of quinolinic acid formation by brain in inflammatory neurological disease. Attenuation of synthesis from L-tryptophan by 6-chlorotryptophan and 4-chloro-3hydroxyanthranilate. Brain 1993;116:1425–50. [6] Ting KK, Brew BJ, Guillemin GJ. Effect of quinolinic acid on human astrocytes morphology and functions: implications in Alzheimer's disease. J Neuroinflammation 2009;6 [36–2094–6–36]. [7] Guillemin GJ, Brew BJ, Noonan CE, Takikawa O, Cullen KM. Indoleamine 2,3 dioxygenase and quinolinic acid immunoreactivity in Alzheimer's disease hippocampus. Neuropathol Appl Neurobiol 2005;31:395–404. [8] Guillemin GJ, Brew BJ. Implications of the kynurenine pathway and quinolinic acid in Alzheimer's disease. Redox Rep 2002;7:199–206. [9] Bullido MA. A polymorphism in the regulatory region of APOE associated with risk for Alzheimer's dementia. Nat Genet 1998;18:69–71. [10] Kandanearatchi A, Brew BJ. The kynurenine pathway and quinolinic acid: pivotal roles in HIV associated neurocognitive disorders. FEBS J 2012;279:1366–74. [11] Davies NW, Guillemin G, Brew BJ. Tryptophan, neurodegeneration and HIVassociated neurocognitive disorder. Int J Tryptophan Res 2010;3:121–40. [12] Guillemin GJ, Kerr SJ, Brew BJ. Involvement of quinolinic acid in AIDS dementia complex. Neurotox Res 2005;7:103–23. [13] Zinger A, Barcia C, Herrero MT, Guillemin GJ. The involvement of neuroinflammation and kynurenine pathway in Parkinson's disease. Park Dis 2011;2011:716859. [14] Chen Y, Stankovic R, Cullen KM, et al. The kynurenine pathway and inflammation in amyotrophic lateral sclerosis. Neurotox Res 2010;18:132–42. [15] Schwarcz R, Guidetti P, Sathyasaikumar KV, Muchowski PJ. Of mice, rats and men: revisiting the quinolinic acid hypothesis of Huntington's disease. Prog Neurobiol 2010;90:230–45. [16] Lim CK, Brew BJ, Sundaram G, Guillemin GJ. Understanding the roles of the kynurenine pathway in multiple sclerosis progression. Int J Tryptophan Res 2010;3:157–67. [17] Myint AM. Kynurenines: from the perspective of major psychiatric disorders. FEBS J 2012;279:1375–85. [18] Niwa T, Yoshizumi H, Emoto Y, et al. Accumulation of quinolinic acid in uremic serum and its removal by hemodialysis. Clin Chem 1991;37:159–61. [19] Furlanetto S, Pinzauti S, La Porta E, Chiarugi A, Mura P, Orlandini S. Development and validation of a differential pulse polarographic method for quinolinic acid determination in human plasma and urine after solid-phase extraction: a chemometric approach. J Pharm Biomed Anal 1998;17:1015–28. [20] Toseland PA. The determination of urinary quinolinic acid by gas–liquid chromatography. Clin Chim Acta 1969;25:185. [21] Chamberlin BA, Sweeley CC. Metabolic profiles of urinary organic acids recovered from absorbent filter paper. Clin Chem 1987;33:572–6. [22] Patterson JI, Brown RR. Determination of urinary quinolinic acid by highperformance liquid chromatography. J Chromatogr 1980;182:425–9. [23] Heyes MP, Markey SP. Quantification of quinolinic acid in rat brain, whole blood, and plasma by gas chromatography and negative chemical ionization mass spectrometry: effects of systemic L-tryptophan administration on brain and blood quinolinic acid concentrations. Anal Biochem 1988;174:349–59. [24] Naritsin DB, Boni RL, Markey SP. Pentafluorobenzylation method for quantification of acidic tryptophan metabolites using electron capture negative ion mass spectrometry. Anal Chem 1995;67:863–70. [25] Dobbie MS, Surtees RA. Concentrations of quinolinic acid in cerebrospinal fluid measured by gas chromatography and electron-impact ionisation mass spectrometry. Age-related changes in a paediatric reference population. J Chromatogr B Biomed Sci Appl 1997;696:53–8. [26] Smythe GA, Poljak A, Bustamante S, et al. ECNI GC–MS analysis of picolinic and quinolinic acids and their amides in human plasma, CSF, and brain tissue. Adv Exp Med Biol 2003;527:705–12. [27] Mawatari K, Oshida K, Iinuma F, Watanabe M. Determination of quinolinic acid by liquid chromatography with fluorimetric detection. Adv Exp Med Biol 1996; 398:697–701. [28] Odo J, Inoguchi M, Hirai A. Fluorometric determination of quinolinic acid using the catalytic activity of horseradish peroxidase. J Health Sci 2009;55:242–8. [29] Moller M, Du Preez JL, Harvey BH. Development and validation of a single analytical method for the determination of tryptophan, and its kynurenine metabolites in rat plasma. J Chromatogr B Anal Technol Biomed Life Sci 2012;898:121–9. [30] US Department of Health and Human Services Food and Drug Administration. ICH ICoH. Text on Validation of Analytical Procedures ICH-Q2A; March 1995 [http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/ Guidances/UCM073381.pdf; http://www.fda.gov/Drugs/Guidancecompliance RegulatoryInformation/Guidances/default.htm; last accessed 4 June 2014]. [31] Schwarz MJ, Guillemin GJ, Teipel SJ, Buerger K, Hampel H. Increased 3hydroxykynurenine serum concentrations differentiate Alzheimer's disease patients from controls. Eur Arch Psychiatry Clin Neurosci 2013;263:345–52.
272
A. Meinitzer et al. / Clinica Chimica Acta 436 (2014) 268–272
[32] Smythe GA, Braga O, Brew BJ, et al. Concurrent quantification of quinolinic, picolinic, and nicotinic acids using electron-capture negative-ion gas chromatography–mass spectrometry. Anal Biochem 2002;301:21–6. [33] Pawlak K, Mysliwiec M, Pawlak D. Kynurenine pathway — a new link between endothelial dysfunction and carotid atherosclerosis in chronic kidney disease patients. Adv Med Sci 2010;55:196–203.
[34] Raison CL, Dantzer R, Kelley KW, et al. CSF concentrations of brain tryptophan and kynurenines during immune stimulation with IFN-alpha: relationship to CNS immune responses and depression. Mol Psychiatry 2010;15(4):393–403. [35] Heyes MP, Brew BJ, Martin A, et al. Quinolinic acid in cerebrospinal fluid and serum in HIV-1 infection: relationship to clinical and neurological status. Ann Neurol 1991;29:202–9.
Contents lists available at ScienceDirect
Clinica Chimica Acta journal homepage: www.elsevier.com/locate/clinchim
Development of a liquid chromatography–mass spectrometry method for the determination of the neurotoxic quinolinic acid in human serum Andreas Meinitzer a,⁎, Andreas Tomaschitz b,c, Stefan Pilz d, Manfred Truber a, Gabriele Zechner a, Martin Gaksch d, Barbara Prietl d, Gerlies Treiber d, Michaela Schwarz e, Andreas Baranyi f a
Clinical Institute of Medical and Chemical Laboratory Diagnostics, Medical University of Graz, Graz, Austria Department of Cardiology, Medical University of Graz, Graz, Austria Specialist Clinic for Rehabilitation PV Bad Aussee, Bad Aussee, Austria d Department of Internal Medicine, Division of Endocrinology and Metabolism, Medical University of Graz, Graz, Austria e Division of Surgical Research, Department of Surgery, Medical University of Graz, Graz, Austria f Department of Psychiatry, Medical University of Graz, Graz, Austria b c
a r t i c l e
i n f o
Article history: Received 9 May 2014 Received in revised form 13 June 2014 Accepted 13 June 2014 Available online 21 June 2014 Keywords: Quinolinic acid Mass spectrometry Kynurenin pathway
a b s t r a c t Background: Quinolinic acid (QA) is thought to be one of the most important metabolites of the kynurenine pathway with the highest biological activity in apoptotic responses and neurodegenerative diseases. The determination of QA might be of clinical relevance in different patient groups, but currently, only a few laborious methods with high levels of sample volume consumption are available. Methods: We developed and validated a simple liquid chromatography–tandem mass spectrometric (LC–tandem MS) method for the determination of QA in human serum with low sample volume requirements. Results: The presented method provides high sample throughput with 25 μL aliquots and works in the positive electrospray ionization (ESI) mode. A commercially available QA-d3 was used as internal standard. Specific transitions for QA and QA-d3 were m/z 280 → m/z 78 and m/z 283 → m/z 81, respectively. The intra- and inter-assay coefficients of variation (CVs) were all below 10%. Applying this method, in 50 healthy humans a mean serum concentration of QA of 350 ± 167 nmol/L (mean ± SD) was determined. Conclusion: The described method is suitable for large clinical trials, which is of potential clinical importance to elucidate the function of QA and its relationship to different disease patterns and may be applicable for clinical laboratory routine. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Quinolinic acid (QA) is generated within the kynurenine pathway resulting from L-tryptophan catabolism and functions as the precursor of nicotinamide adenine dinucleotide (NAD). Along with other acid compounds of the kynurenine pathway, kynurenic acid and anthranilic acid, QA is suggested to be the most important in terms of biological activity [1]. It is involved in complex interrelationships with inflammatory and apoptotic responses associated with neuronal cell damage and death in the central nervous system [2,3]. Production of QA in the brain is increased in inflammatory and neurodegenerative diseases [4, 5]. There is growing evidence that QA is involved in the development of brain diseases such as Alzheimer's disease [6–9], HIV-associated neurocognitive disorders [10–12], Parkinson's disease [13], motor neuron diseases [14], Huntington's disease [15], multiple sclerosis [16] and major psychiatric disorders [17]. High serum concentrations of QA were found in chronic hemodialysis patients. The accumulation of QA ⁎ Corresponding author. Tel.: +43 316 385 83988; fax: +43 316 385 13419. E-mail address: [email protected] (A. Meinitzer).
http://dx.doi.org/10.1016/j.cca.2014.06.010 0009-8981/© 2014 Elsevier B.V. All rights reserved.
in uremic blood may be involved in the pathogenesis of anemia, a suppressed immune system, and uremic encephalopathy [18]. An increase in serum levels leads to enhanced QA concentrations in brain tissue, probably due to an increase in blood–brain barrier transport [19]. Several methods for the quantification of QA in biological fluids, in which different separation and detection methods, have been described. Methods based on gas chromatography [20,21] and liquid chromatography [22] are time-consuming and need several different reagents with partly unpleasant properties. The detection sensitivity of QA, which has two carboxylic groups side by side and thus is not fluorescent nor electrochemically active, is rather low. Some authors esterified the carboxyl groups and measured QA by the use of gas chromatography and mass spectrometry [23–26]. Odo et al. refined a method of Mawatari [27] and converted QA in the presence of hydrogen peroxide and the catalytic activity of horseradish peroxidase into a fluorescent compound for determining trace amounts of QA [28]. Recently, Moeller et al. reported a method for underivatized QA measurement using liquid chromatography– and electrospray ionization (ESI)–tandem mass spectrometry in a sample volume of 800 μL [29]. Polar compounds like QA have lower ESI responses than less polar compounds. Mass
A. Meinitzer et al. / Clinica Chimica Acta 436 (2014) 268–272 Table 1 Characteristics of the quinolinic acid method. Quinolinic acid Working range [nmol/L] Calibration curve Slope Intercept Correlation r2 Intra-day precision n = 6 Mean [nmol/L] SD [nmol/L] Coefficient of variation [%] Inter-day precision n = 20 Mean [nmol/L] SD [nmol/L] Inter-day variability [%] Recovery (mean extraction efficiency) [%, ±SD] Limit of quantification (LOQ) [nmol/L] SD [nmol/L] Coefficient of variation [%] Limit of detection (LOD) [nmol/L] 3-fold SD of the baseline noise Stability in the matrix tested
50–5000 0.03725 0.00224 0.9998 Low 225 10.2 4.5
High 725 8.5 1.2
229 17.0 7.2 95.0 ± 7.7
752 47.3 6.3
elements were controlled by Xcalibur software, which also managed the Voyager TSQ Quantum triple quadrupole instrument (Thermo). LCquan software (Thermo) was used to calculate results from the raw data. To test the concentrations in humans we recruited 50 healthy subjects aged at least 18 years. Exclusion criteria were pregnancy, participation in other interventional clinical trials and any disease requiring medical treatment. Laboratory measurements for basic patient characteristics (LDL-cholesterol, HDL-cholesterol, glucose and triglycerides) were performed by routine methods. The study was performed at the Division of Endocrinology and Metabolism (Department of Internal Medicine) at the Medical University of Graz. The study adheres to the Declaration of Helsinki and ethical approval was obtained from the ethics committee at the Medical University of Graz, Austria. 3. Sample preparation
50 4.7 9.3 15 2 days
spectrometers connected to liquid chromatography systems are not able to create sufficient and stable ionization response to detect QA in small amounts of biological samples. Our aim therefore was to develop a simple derivatization method for QA to enhance the ionization response, which enables QA quantification in small volume portions of serum with liquid chromatography–mass spectrometers. 2. Materials and methods HPLC grade methanol and analytical grade sodium hydroxide were purchased from Merck (Darmstadt, Germany). Sodium tetraborate, QA (2,3 pyridinedicarboxylic acid), formic acid, 3 M hydrogen chloride-1butanol solution and activated charcoal all in analytical grade were obtained from Sigma-Aldrich (Austria). CHROMASOLV-type liquid chromatography–mass spectrometry water was purchased from Fluka (Hamburg, Germany). Sample clean-up was performed by solid-phase extraction on strong anion-exchange columns (Strata XA 30 mg/1 mL, Phenomenex Aschaffenburg, Germany) using QA-d3 (Toronto Research Chemicals, North York, Ontario, Canada) as internal standard. The cartridges placed in 10 mL plastic tubes (100 × 16 mm, Greiner Holding AG, Kremsmünster, Austria) were loaded and eluted with gravitation on an Eppendorf centrifuge 5810 R (Eppendorf, Hamburg, Germany) at different times (s) and g values. In the sample procedure the exact times and g values are given in parentheses. Chromatographic separation was done on a reversed phase column Synergi Hydro, RP 100A (100 × 3 mm, 2.5 μm) embedded in a Thermo Surveyor system (Thermo Instruments, San Jose, California, USA), which consists of a quaternary pump, an autosampler and a column oven. All chromatographic Table 2 Patients characteristics and quinolinic acid concentration. Number of patients Females [%] Age [yrs] Body mass index [kg/m2] Systolic blood pressure [mm Hg] Diastolic blood pressure [mm Hg] Glucose [mmol/L] HDL-cholesterol [mmol/L] LDL-cholesterol [mmol/L] Triglycerides [mmol/L] Quinolinic acid [nmol/L]
269
50 64 31 ± 8 23.3 ± 4.3 129 ± 15 84 ± 10 4.89 ± 0.92 1.84 ± 0.37 2.46 ± 0.57 1.08 ± 0.49 350 ± 167
Continuous variables are presented as means ± standard deviation (SD).
The Strata XA cartridges were initially washed with 1 mL methanol followed by 1 ml of borate buffer (25 mmol/l, pH = 9.5) (both steps 60 s, 1500 g). Twenty-five-microliter serum samples, calibrators or controls were mixed with 10 μL internal standard (3 μmol/L in 50% [v/v] methanol/water) and 950 μL borate buffer, and passed through the columns (90 s, 800 g). The loaded cartridges were then washed with 1 mL borate buffer, 1 mL water, and two times with 1 mL methanol (all four steps with 60 s, 1500 g). After the last washing step the columns were centrifuged for 60 s with 4000 g to remove most of the liquid. The cartridges were then dried for 20 s by applying 2 bar compressed air. Elution of the absorbed analytes in tapered glass tubes was performed with 500 μL 3 M hydrogen chloride in 1-butanol. The eluates were heated for 30 min at 80° for allowing esterification. Then the eluates were evaporated with a stream of compressed air at 80° on a Turbovap System (American Laboratory Trading, East Lyme, Connecticut, USA). The dried residue was reconstituted with 200 μL of mobile phase B, vortexed for few seconds, transferred to auto sampler vials and 50 μL were injected in the LC–MS/MS system. 4. Chromatographic conditions and MS/MS detection parameters Separation was done with the above-mentioned column with dual gradient elution. Mobile phase A consisted of 0.1% formic acid in water, mobile phase B was a mixture of 35% water, 65% methanol and 0.1% formic acid (v/v/v). After injection, a mixture of 30% A and 70% B with a flow rate of 300 μL/min was pumped through the column for 4 min. Then a linear gradient started to 100% B with a duration of 2 min followed by a period of 2 min with 100% B for washing out and a period of 2 min for equilibration with the initial condition. The mass instrument was set in multiple-reaction monitoring mode. The analytes were identified in the positive-ion mode and determined by means of characteristic product ions formed from protonated molecules by collision induced dissociation (CID) with the following fragments: m/z 280.1 → 78 for QA with qualifier m/z 280.1 → 106 and m/z 283.1 → 81 for QA d3 as internal standard with qualifier 283.1 → 109.14. Settings for the ion source were as follows: spray voltage 4000 V, sheath gas pressure 60 psi, ion sweep gas pressure 0 psi, aux gas pressure 15 psi, capillary temperature of 350 °C, source fragmentation energy 22 V, scan time 0.02 s, scan width m/z 0.02, peak width Q1 0.7, peak width Q3 0.7, skimmer offset 5 V, and collision gas pressure 1.5 mTorr. Validation was done according to FDA guidelines for bioanalytical method validation [30]. The tested validation parameters were linearity, intra-assay and inter-assay precision, matrix effects, recovery, carry over and stability. The within- and between-day imprecisions were assessed by performing 6 replicate analyses of quality control samples within one day and within 6 consecutive days, respectively. The limit of detection (LOD) and the limit of quantification (LOQ) were defined as the minimum concentration showing a signal of at least 3 times and 10 times respectively greater at the retention time of the analyte
270
A. Meinitzer et al. / Clinica Chimica Acta 436 (2014) 268–272
Fig. 1. Chromatogram of a patient with 540 nmol/L quinolinic acid. Every peak has over 30 measure points. Peaks are shown without smoothing. Left: Derivate Dibutyl 2,3 pyridinedicarboxylate with analyte and qualifier fragments. Right: internal standard derivate with analyte and qualifier fragments. NL: normalized scale; SRMms2, single reacting mode and transmission mass ranges; A, Area under the curve; S/N, signal to noise ratio.
than the average background noise of an unspiked blank (only containing IS).
5. Standardization and quality control To produce calibration curves and to determine the recovery of the method, a drug-free serum pool obtained from the local Department of Transfusion Medicine was used. To remove endogenous QA 100 mL of the serum was agitated overnight with 5 g of activated charcoal at 4 °C. After centrifugation at 25,000 g for 30 min at 4 °C, the supernatant was treated with a portion of 5 g Strata XA anion exchange material and agitated once more for 1 h. After the centrifugation of the matrix we
measured the QA concentration in the treated pool. Despite a second extraction step a small amount of QA in the range of 50 nmol/L remained in the human serum matrix. We determined this concentration with the standard addition method and included this blank in the calculation of the calibrator values. A stock solution of QA (1 mmol/L) was prepared in water and diluted with the above described matrix to the appropriate value. Standard series were prepared with concentrations of 0; 100; 250; 500; 1000; and 2000 nmol/L. For precision control, we produced two pooled serum samples: one from patients attending the gynecological clinic, because the levels in these patients are low (pooled control serum in the low range), and another one from patients of the Department for Angiology, which are mostly high (pooled control serum in the high range). Aliquots of these controls were analyzed in each run after ten specimens. Before the series were performed, the controls were analyzed 10 times and the mean value was taken as the reference value. Stability of samples: As specimens are usually handled and transported to the laboratory at room temperature and we investigated stability at room temperature. Instability is usually easier seen at room temperature, which is the worst stress condition for the specimens. We stored 3 patient samples at room temperature with daylight exposure and analyzed these after time intervals of 0, 4, 24 and 48 h. An analyte was considered stable when the measured concentration was in the range of the inter-day standard deviation of the method.
6. Results
Fig. 2. Representative ion suppression chromatograms. A solution of dibutyl 2,3 pyridinedicarboxylate (150 μmol/L) was mixed via a “tee” to the eluent after the column into the mass detector at a constant rate of 5 μL/min. Mass chromatograms were taken after injection of an extract of a serum pool from 10 patients. The upper chromatogram is a patient sample (2500 nmol/L) without post-infusion of QA and serves as reference. NL: normalized scale; SRMms2, single reacting mode and transmission mass ranges.
The characteristics of the analytical method as a result of the method evaluation are presented in Table 1. Calibration curves were linear throughout the selected ranges and the coefficient of determination (r2) exceeded 0.990. The standard deviations for imprecision and accuracy results were all highly acceptable. The extraction efficiencies from the biological matrix to a protein-free solution, were 95 ± 7.7%. Based on the examination of chromatograms of serum samples – after injection of pathologically high samples – without analytes, the calculated carryover effects were below 1%. The mean serum concentration of QA in healthy humans was 350 ± 167 nmol/L (n = 50). Clinical
A. Meinitzer et al. / Clinica Chimica Acta 436 (2014) 268–272
characteristics of the participants were in the reference range (Table 2). In women, QA concentrations (381 ± 198 nmol/L, n = 32) were higher compared to men (297 ± 68 nmol/L, n = 18), but the difference did not reach statistical significance (p = 0.090). There was no significant association between serum QA concentration and age. Fig. 1 provides a typical chromatogram of a patient and contains pure raw data without smoothing. All peaks show a symmetrical form and have more than 25 measurement points. Injection of an extracted serum pool in the post column-infusion experiment (Fig. 2) showed valleys at the solvent front (1.5 to 2.5 min) and minimum response valleys at 5.15 and 5.85 min, which indicates interactions with eluted matrix compounds or elution of extraction solution ions. At the retention times of QA (6.4 min) no influence of the signal was observed.
7. Discussion The aim of the present study was to develop a simple volume-saving method for the expeditious determination of QA in human serum. In contrast to a recently published LC–MS/MS method for QA determination in rat plasma with the use of 800 μL sample volume [29], the present method on hand needs only 25 μL. Our developed method would also work with smaller volumes, but due to better robustness and convenience 25 μL might be appropriate. One limitation of the present method is that it does not work in EDTA plasma. EDTA is highly concentrated in the tubes for taking blood, usually at 1 to 10 mg, depending on their volume. The molar ratio to QA in plasma is approximately 1:10.000. EDTA probably displaces QA from the binding sites of the anionic exchange column in the sample preparation. The recovery of QA decreases to zero. Interestingly, this effect is not observed during the measurement of monocarboxylic kynurenic acid, which can be easily extracted from plasma or serum with the same method. Two aspects should be considered in regard to the sample preparation: (1) The SPE cartridges must be highly dried before the eluting step is carried out; and (2) The derivatization agent must be free of water. Although a lot of literature is found about QA, there are only rare data of QA measured in humans. Using gas chromatography–mass spectrometry [32] Schwarz et al. [31] determined QA in twenty patients with Alzheimer's disease [median: 560 nmol/L (370–1060 nmol/L)] and in humans with subjective cognitive impairment [median: 730 nmol/L (390–1220 nmol/L)]. Pawlak et al. [33] separated QA on an ion exchange chromatography column and detected QA with an ultraviolet detector with the need of 2 mL of serum. In this investigation the median values of QA in patients with chronic kidney disease and 20 healthy controls were 2050 (550–8000) nmol/L and 150 (70–630) nmol/L, respectively. Markedly increased concentrations of QA in uremic patients have been already observed [18]. The mean values of QA in uremic patients and healthy controls were 9200 ± 5300 nmol/L (n = 54) and 600 (±300) nmol/L (n = 10), respectively. Using a not precisely described method Raison [34] in healthy controls and in patients with hepatitis C virus documented a mean QA value of 330 nmol/L (n = 27). Heyes [35] in patients living with human immunodeficiency virus (HIV; n = 6) found clearly elevated levels of QA (mean: 1393 ± 108 nmol/L as compared to QA values of healthy controls (432 ± 64 nmol/L, n = 11). The mean values of QA measured in healthy controls conform to the previously determined QA concentrations in healthy subjects. Interestingly, these groups demonstrated highly significant correlations between QA serum and QA cerebrospinal concentrations. However the QA levels in the cerebrospinal fluid are 10-fold lower than those in serum. It can therefore be speculated that circulating QA concentrations reflect QA concentrations in the CSF. In conclusion, we developed a LC–tandem MS method which allows the quantification of QA in only 25 μL of human serum. A commercially available stable isotope labeled QA served as internal standard. The method is suitable for large clinical trials, which is of crucial importance
271
to elucidate the status of QA in neurodegenerative diseases and may be applicable for clinical laboratory routine. References [1] Guillemin GJ. Quinolinic acid, the inescapable neurotoxin. FEBS J 2012;279(8):1356–65. [2] Stone TW. Endogenous neurotoxins from tryptophan. Toxicon 2001;39:61–73. [3] Stone TW, Perkins MN. Quinolinic acid: a potent endogenous excitant at amino acid receptors in CNS. Eur J Pharmacol 1981;72:411–2. [4] Heyes MP, Saito K, Crowley JS, et al. Quinolinic acid and kynurenine pathway metabolism in inflammatory and non-inflammatory neurological disease. Brain 1992; 115:1249–73. [5] Heyes MP, Saito K, Major EO, Milstien S, Markey SP, Vickers JH. A mechanism of quinolinic acid formation by brain in inflammatory neurological disease. Attenuation of synthesis from L-tryptophan by 6-chlorotryptophan and 4-chloro-3hydroxyanthranilate. Brain 1993;116:1425–50. [6] Ting KK, Brew BJ, Guillemin GJ. Effect of quinolinic acid on human astrocytes morphology and functions: implications in Alzheimer's disease. J Neuroinflammation 2009;6 [36–2094–6–36]. [7] Guillemin GJ, Brew BJ, Noonan CE, Takikawa O, Cullen KM. Indoleamine 2,3 dioxygenase and quinolinic acid immunoreactivity in Alzheimer's disease hippocampus. Neuropathol Appl Neurobiol 2005;31:395–404. [8] Guillemin GJ, Brew BJ. Implications of the kynurenine pathway and quinolinic acid in Alzheimer's disease. Redox Rep 2002;7:199–206. [9] Bullido MA. A polymorphism in the regulatory region of APOE associated with risk for Alzheimer's dementia. Nat Genet 1998;18:69–71. [10] Kandanearatchi A, Brew BJ. The kynurenine pathway and quinolinic acid: pivotal roles in HIV associated neurocognitive disorders. FEBS J 2012;279:1366–74. [11] Davies NW, Guillemin G, Brew BJ. Tryptophan, neurodegeneration and HIVassociated neurocognitive disorder. Int J Tryptophan Res 2010;3:121–40. [12] Guillemin GJ, Kerr SJ, Brew BJ. Involvement of quinolinic acid in AIDS dementia complex. Neurotox Res 2005;7:103–23. [13] Zinger A, Barcia C, Herrero MT, Guillemin GJ. The involvement of neuroinflammation and kynurenine pathway in Parkinson's disease. Park Dis 2011;2011:716859. [14] Chen Y, Stankovic R, Cullen KM, et al. The kynurenine pathway and inflammation in amyotrophic lateral sclerosis. Neurotox Res 2010;18:132–42. [15] Schwarcz R, Guidetti P, Sathyasaikumar KV, Muchowski PJ. Of mice, rats and men: revisiting the quinolinic acid hypothesis of Huntington's disease. Prog Neurobiol 2010;90:230–45. [16] Lim CK, Brew BJ, Sundaram G, Guillemin GJ. Understanding the roles of the kynurenine pathway in multiple sclerosis progression. Int J Tryptophan Res 2010;3:157–67. [17] Myint AM. Kynurenines: from the perspective of major psychiatric disorders. FEBS J 2012;279:1375–85. [18] Niwa T, Yoshizumi H, Emoto Y, et al. Accumulation of quinolinic acid in uremic serum and its removal by hemodialysis. Clin Chem 1991;37:159–61. [19] Furlanetto S, Pinzauti S, La Porta E, Chiarugi A, Mura P, Orlandini S. Development and validation of a differential pulse polarographic method for quinolinic acid determination in human plasma and urine after solid-phase extraction: a chemometric approach. J Pharm Biomed Anal 1998;17:1015–28. [20] Toseland PA. The determination of urinary quinolinic acid by gas–liquid chromatography. Clin Chim Acta 1969;25:185. [21] Chamberlin BA, Sweeley CC. Metabolic profiles of urinary organic acids recovered from absorbent filter paper. Clin Chem 1987;33:572–6. [22] Patterson JI, Brown RR. Determination of urinary quinolinic acid by highperformance liquid chromatography. J Chromatogr 1980;182:425–9. [23] Heyes MP, Markey SP. Quantification of quinolinic acid in rat brain, whole blood, and plasma by gas chromatography and negative chemical ionization mass spectrometry: effects of systemic L-tryptophan administration on brain and blood quinolinic acid concentrations. Anal Biochem 1988;174:349–59. [24] Naritsin DB, Boni RL, Markey SP. Pentafluorobenzylation method for quantification of acidic tryptophan metabolites using electron capture negative ion mass spectrometry. Anal Chem 1995;67:863–70. [25] Dobbie MS, Surtees RA. Concentrations of quinolinic acid in cerebrospinal fluid measured by gas chromatography and electron-impact ionisation mass spectrometry. Age-related changes in a paediatric reference population. J Chromatogr B Biomed Sci Appl 1997;696:53–8. [26] Smythe GA, Poljak A, Bustamante S, et al. ECNI GC–MS analysis of picolinic and quinolinic acids and their amides in human plasma, CSF, and brain tissue. Adv Exp Med Biol 2003;527:705–12. [27] Mawatari K, Oshida K, Iinuma F, Watanabe M. Determination of quinolinic acid by liquid chromatography with fluorimetric detection. Adv Exp Med Biol 1996; 398:697–701. [28] Odo J, Inoguchi M, Hirai A. Fluorometric determination of quinolinic acid using the catalytic activity of horseradish peroxidase. J Health Sci 2009;55:242–8. [29] Moller M, Du Preez JL, Harvey BH. Development and validation of a single analytical method for the determination of tryptophan, and its kynurenine metabolites in rat plasma. J Chromatogr B Anal Technol Biomed Life Sci 2012;898:121–9. [30] US Department of Health and Human Services Food and Drug Administration. ICH ICoH. Text on Validation of Analytical Procedures ICH-Q2A; March 1995 [http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/ Guidances/UCM073381.pdf; http://www.fda.gov/Drugs/Guidancecompliance RegulatoryInformation/Guidances/default.htm; last accessed 4 June 2014]. [31] Schwarz MJ, Guillemin GJ, Teipel SJ, Buerger K, Hampel H. Increased 3hydroxykynurenine serum concentrations differentiate Alzheimer's disease patients from controls. Eur Arch Psychiatry Clin Neurosci 2013;263:345–52.
272
A. Meinitzer et al. / Clinica Chimica Acta 436 (2014) 268–272
[32] Smythe GA, Braga O, Brew BJ, et al. Concurrent quantification of quinolinic, picolinic, and nicotinic acids using electron-capture negative-ion gas chromatography–mass spectrometry. Anal Biochem 2002;301:21–6. [33] Pawlak K, Mysliwiec M, Pawlak D. Kynurenine pathway — a new link between endothelial dysfunction and carotid atherosclerosis in chronic kidney disease patients. Adv Med Sci 2010;55:196–203.
[34] Raison CL, Dantzer R, Kelley KW, et al. CSF concentrations of brain tryptophan and kynurenines during immune stimulation with IFN-alpha: relationship to CNS immune responses and depression. Mol Psychiatry 2010;15(4):393–403. [35] Heyes MP, Brew BJ, Martin A, et al. Quinolinic acid in cerebrospinal fluid and serum in HIV-1 infection: relationship to clinical and neurological status. Ann Neurol 1991;29:202–9.
