Alcohol and Alcoholism Advance Access originally published online on May 28, 2007
Alcohol and Alcoholism 2007 42(4):321-325; doi:10.1093/alcalc/agm039
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Biomarkers to disclose recent intake of alcohol: potential of 5-hydroxytryptophol glucuronide testing using new direct UPLC-tandem MS and ELISA methods
1 Department of Medicine, Division of Clinical Pharmacology, Karolinska Institute and Karolinska University Hospital, Stockholm, Sweden
2 Arztpraxis für Medizinische Mikrobiologie und Labordiagnostik, Dessau, Germany
3 Department of Psychiatry, University of Mainz, Germany
4 Department of Clinical Neuroscience, Alcohol Laboratory, Karolinska Institute and Karolinska University Hospital, Stockholm, Sweden
* Author to whom correspondence should be addressed at: Karolinska University Hospital Solna, L7:05 Clinical Pharmacology, SE-171 76 Stockholm, Sweden. Tel: +46 8 51773026; Fax: +46 8 331343; E-mail: olof.beck{at}karolinska.se
Received 21 January 2007; in revised form 23 February 2007; in revised form 26 March 2007; accepted 30 March 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT IntroductionExperimentalResults and DiscussionReferences |
|---|
Aims: This study compared two new methods for direct determination of 5-hydroxytryptophol glucuronide (GTOL) in urine, a biomarker for detection of recent alcohol consumption. Methods: Urine samples were collected from ten alcoholic patients during recovery from intoxication. A direct injection ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) method for measurement of the urinary GTOL to 5-hydroxyindoleacetic acid (5-HIAA) ratio, and an ELISA assay for direct measurement of GTOL, were used. Comparison was made with the urinary ethanol and ethyl glucuronide (EtG) concentrations. Results: The breath ethanol concentration on admission ranged between 1.0–3.1 g/l. The UPLC-MS/MS method showed a median detection time of 39 h for an elevated urinary GTOL/5-HIAA ratio, while EtG was detected for a median of 65 h. Determination of GTOL by the ELISA assay showed 87% sensitivity in detecting positive samples at a 44% specificity, as compared with the UPLC-MS/MS method. Conclusions: The lower sensitivity of the urinary GTOL/5-HIAA ratio compared with EtG for recent drinking may be clinically useful, in cases where the EtG test provides an unwanted high sensitivity for intake of only small amounts of alcohol or unintentional ethanol exposure.
|
Introduction |
|---|
|
TOPABSTRACTIntroductionExperimentalResults and DiscussionReferences |
|---|
In situations where self-reports of alcohol consumption are not considered valid, there may be a need to verify abstinence or drinking by an objective biochemical measure (Helander, 2003
). This has most commonly been done by testing for ethanol in blood, urine or breath, which has the limitation of offering a very short detection window (Jones, 2006). The recent interest in 5-hydroxytryptophol (5-HTOL) and ethyl glucuronide (EtG) has demonstrated the need for more sensitive measures of acute alcohol consumption (Helander et al., 1996; Wurst et al., 2005; Jones, 2006).
EtG was discovered as a potential alcohol biomarker already 10 years ago (Schmitt et al., 1995, 1997) and has found an increasing use ever since. The EtG test has been applied in clinical, forensic and workplace testing, as a complement to self-reports of drinking, and in combination with screening for illicit drugs in urine. The interest in EtG relates to the longer detection time of about 24–48 h compared with ethanol (Dahl et al., 2002; Borucki et al., 2005). The standard methodology for measuring EtG in urine involves mass spectrometric detection, which has an inborn high sensitivity. However, the high sensitivity and absence of an established cut-off limit could represent a problem, as intake of very small amounts of ethanol, and perhaps even unintentional intake, may also be detected. It was recently shown that use of ethanol containing mouthwash and hand sanitizers could lead to a positive EtG test, if a low cut-off limit is applied (Constantino et al., 2006; Rohrig and Ross, 2006). This and other observations in workplace testing have lead to a debate on the reliability of this alcohol biomarker in some situations. A medical advisory (Center for Substance Abuse Treatment, 2006) stated that legal or disciplinary action taken solely on a positive EtG test is inappropriate and scientifically unsupported at the moment.
An alternative biomarker focusing more on recent moderate-to-high drinking levels is 5-HTOL, which is excreted in urine predominantly as 5-hydroxytryptophol glucuronide (GTOL) (Beck and Helander, 2003). As for EtG, the 5-HTOL level remains increased in the urine for several hours after ethanol is no longer measurable in body fluids or breath (Helander et al., 1993, 1999; Bendtsen et al., 1998), but the detection time is shorter (Sarkola et al., 2003). A widespread use of the 5-HTOL marker has been hindered mainly by methodological problems, because it normally occurs at low concentrations (Beck and Helander, 2003). In routine clinical use, 5-HTOL is reported as a ratio to 5-hydroxyindoleacetic acid (5-HIAA) instead of creatinine, because this practice compensates for interferences from variations in urine dilution as well as serotonin turnover (Helander et al., 1992).
The common analytical approach for urinary 5-HTOL and 5-HIAA has been to employ separate methods for each compound; typically gas chromatography-mass spectrometry (GC-MS) for 5-HTOL and high-performance liquid chromatography (HPLC) for 5-HIAA (Beck and Helander, 2003). A recent evaluation of a commercial ELISA kit for GTOL showed promising results and offered a viable alternative to GC-MS (Dierkes et al., 2006), but a second analytical method was still required for 5-HIAA. This problem was overcome by the recent development of a direct liquid chromatographic-mass spectrometric (LC-MS) procedure for simultaneous measurement of GTOL and 5-HIAA in urine (Stephanson et al., 2007).
The aims of this study was to evaluate the new LC-MS and ELISA methods on a clinical material, and to compare the urinary GTOL/5-HIAA results with those obtained for ethanol and EtG.
|
Experimental |
|---|
|
TOPABSTRACTIntroductionExperimentalResults and DiscussionReferences |
|---|
Urine samples
Urine samples from randomly selected patients (meeting DSM IV criteria for alcohol dependence) being hospitalized for recovery from acute alcohol intoxication were collected on admission to the hospital, the next morning, and then every
24 h for several days. Breath ethanol was measured in parallel using a Dräger Alcotest model 7410 (Dräger Safety AG, Lübeck, Germany). All patients involved in this study showed a positive breath alcohol concentration on admission and a negative test at the third sampling at the latest. The urine specimens were collected in plastic Urine Monovettes without preservatives (Sarstedt AG, Nürnbrecht, Germany) and stored at –20°C until analysis. The ethics committee at the University of Mainz approved this study.
Analytical methods
GTOL/5-HIAA by LC-MS/MS
Urinary GTOL and 5-HIAA were analysed by an ultra-performance LC-tandem MS method (UPLC-MS/MS) (Stephanson et al., 2007). The LC-MS/MS system consisted of a Waters Acquity UPLC with a vacuum degasser, binary pump, and sample manager connected to a Quattro Premier XE tandem mass spectrometer with MassLynx/Target Lynx Software version 4.1 (Waters Co, Milford, MA, USA). An electrospray interface was used with the instrument operating in the negative ion mode. The chromatographic column was a 1.8-µm 100 x 2.1 mm (inner diameter) high strength silica (HSS) trifunctional C18 column (Waters Co), and gradient elution was used with a flow rate of 400µl/min. Solvent A consisted of 0.1% (26.5 mmol/l) formic acid (pH 2.85) and Solvent B was 100% acetonitrile. The total run time of the method was 3.6 min. Aliquots of urine (25µl) were added to autosampler vials together with 25µl of a water solution containing internal standards (GTOL-D4 and 5-HIAA-D2), and 10µl of 1% ammonia. The injection volume was 4.5µl and the column temperature 60°C. Mass spectrometric detection was performed using selected reaction monitoring (SRM) of 2 product ions for GTOL and 5-HIAA (m/z 352.2/131.0;175.9 and m/z 190.0/146.2;116.2, respectively) and 1 product ion for each internal standard (m/z 356.2/131.0 and m/z 192.0/148.2, respectively) from the respective deprotonated molecules. The GTOL and 5-HIAA concentrations of unknown samples were determined from the peak area ratio by reference to the calibration curve between GTOL and GTOL-D4, and between 5-HIAA and 5-HIAA-D2, respectively. The criteria for identification was a relative ion intensity between qualifier and target ions within ±20% of the target value and a relative retention time between analyte (both ions) and deuterated internal standard within ±1% of the target value. The measuring ranges of the method are 6.7–10000 nmol/l for GTOL and 0.07–100µmol/l for 5-HIAA. The intra- and inter-assay coefficients of variation (CV) for GTOL were < 6% and < 7% at levels 110 and 2066 nmol/l (N = 15), respectively, and for 5-HIAA < 3.3% and < 5% at levels 15 and 31 nmol/l (N = 15), respectively. A cut-off limit of the GTOL/5-HIAA molar ratio of 0.015 was used (Beck and Helander, 2003).
GTOL by ELISA A commercial ELISA kit in standard microtitre-plate format was used for direct measurement of GTOL after 6-point calibration according to the manufacturers instructions (AlcoDia Co, Stockholm, Sweden). The kit is based on a monoclonal mouse antibody, requires 25µl urine sample, and has a measuring range of 100–4500 nmol/l GTOL. Besides urine calibrators, the kit contains a normal range control QC1 (target value = 522 nmol/l GTOL) and a positive urine control QC2 (target value = 2700 nmol/l). Every series was fully calibrated with duplicates (6 points) and both controls were run twice. Washing of the plates and pipetting was conducted with multi-channel pipettes and optical density reading at 450 nm and curve fitting was performed with a DAS microplate reader (Palombara, Sabina, Italy). The quality control data (inter-assay) from 8 batches were as follows: QC1 (nominal value = 357–596 nmol/l) mean = 530 nmol/l, CV = 17.4%, N = 8; QC2 (nominal value = 2173–3622 nmol/l) mean = 2861 nmol/l, CV = 12.9%, N = 8.
5-HIAA by HPLC 5-HIAA was analysed on an isocratic HPLC system consisting of a Model 1350 HPLC pump, an AS-100 autosampler, a column heater (Bio-Rad, Munich, Germany) and a CLC 100 electrochemical detector (Chromsystems, Munich, Germany) with a glassy-carbon working electrode and a stainless-steel auxiliary electrode. The potential was maintained at 0.62–0.70 V versus a Ag/AgCl electrode. The detector sensitivity was set at 50 nA/V. Preparation of the sample (200µl) and chromatography was carried out using the sample preparation kit, analytical column and mobile phase from Chromsystems (Munich, Germany) according to the manufacturers instructions. According to the manufacturer the intra-assay CV was < 4.1%, the inter-assay CV <7% and the limit of quantification (LOQ) 2.6µmol/l. The flow rate was 1.0 ml/min and a 10-µl sample with internal standard was injected onto the column conditioned at 25°C. Every run was started by a 1-point matrix calibration (Urine Calibration Standard) followed by a quality control sample (Urine Endocrine Control, normal and pathological range). A quality control sample was injected for every 10th patient sample. The quality control data (inter-assay) from nine batches were as follows: QC1 (nominal value = 29.7µmol/l) mean = 30.8µmol/l, CV = 7.5%, N = 17; QC2 (nominal value = 113µmol/l) mean = 105µmol/l, CV = 6.1%, N = 17).
EtG by LC-MS
Urinary EtG was analysed by a negative ion electrospray LC-MS method (Stephanson et al., 2002). The ions monitored were m/z 221 for EtG and m/z 226 for EtG-D5 (internal standard). The intra- and inter-assay CV of the method were < 12%, the measuring range was 0.5–200 mg/l (extended by dilution in water), and the limit of detection (LOD) 0.05 mg/l. All positive results were verified by LC-MS/MS by the presence of the product ions of EtG (m/z 113, 85, and 75).
Urinary ethanol Ethanol was determined using the ADH method on an Olympus AU640 (Olympus, Hamburg, Germany) with reagents from Microgenics (Passau, Germany).
|
Results and Discussion |
|---|
|
TOPABSTRACTIntroductionExperimentalResults and DiscussionReferences |
|---|
After income in a state of alcohol intoxication, as confirmed by positive breath ethanol tests (mean = 1.98 g/l; range = 0.97–3.10), the ten alcohol dependent patients were followed by serial urine sampling over several days. Clinical observations and breath ethanol testing were used, to control for abstinence from alcohol during the study period. The individual urine ethanol elimination curves are shown in Fig. 1. As sampling occurred rather infrequently, it should be noted that these curves overestimate the elimination time for ethanol in the body, because the last point for several individuals were at times when ethanol was no longer detected in urine. Accordingly, the blood ethanol concentration may have reached zero many hours before (Helander et al., 1996
; Jones, 2006). The median detection time for ethanol in urine obtained in this way was about 17 h.
|
For all patients, urinary EtG remained positive for a much longer time than ethanol, but there was no correlation between the detection time for EtG and the initial breath ethanol concentration (r = 0.278, P = 0.439). However, there is a marked inter-individual variability in the EtG concentration, even after correcting the values for urine dilution (Dahl et al., 2002
; Goll et al., 2002; Bergström et al., 2003). The shortest detection time for EtG was between 10–20 h and the longest > 90 h after admission (Fig. 1) with a median value of about 65 h. These detection times are considerably longer than observed in experiments where healthy volunteers were administered a single low-to-medium ethanol dose (Dahl et al., 2002; Borucki et al., 2005). Because prolonged drinking does not cause accumulation of EtG in the body (Sarkola et al., 2003), the longer detection times observed in this study most likely relate to the much higher doses of ethanol ingested. The detection times were also longer than those observed in a recent study of alcoholic patients (Wurst et al., 2004), but in better agreement with reported detection times of up to 84 h in another study (Wurst et al., 2002).
An increased urinary GTOL/5-HIAA ratio was observed for between 9 and > 96 h after admission, with a median value of 39 h (Fig. 1). The shorter detection time of GTOL/5-HIAA compared with EtG is in accordance with observations on alcohol patients and from experiments in healthy controls (Dahl et al., 2002; Sarkola et al., 2003; Borucki et al., 2005). Nonetheless, the shorter detection time (i.e. lower sensitivity) of GTOL/5-HIAA compared with EtG may be useful, for applications aiming at detecting intentional intake of a more substantial (moderate-to-heavy) dose of alcohol but not very small amounts or unintentional intake. The work of Kroke and co-workers (Kroke et al., 2001) demonstrated the GTOL marker to disclose alcohol intake above 0.1 g/kg the preceding day.
A major obstacle for the routine application of the GTOL (5-HTOL) test has been the low concentrations in urine without previous drinking, and the relatively complicated analytical work required. Two solutions for this were explored in the present study. The available ELISA method gave results correlating with the UPLC-MS/MS method (Fig. 2), albeit with a marked scatter at all concentrations. The ELISA method was used together with a commercial HPLC assay for urinary 5-HIAA and, if applying the established molar ratio of 0.015 as cut-off limit (Beck and Helander, 2003), the overall agreement of qualitative GTOL/5-HIAA results with the UPLC-MS/MS results was 56%, with 87% sensitivity for finding samples with elevated levels at a specificity of 44%. However, in terms of diagnostic sensitivity and specificity, the results of the present study favour the UPLC-MS/MS method, which could either be used as a stand-alone test, or for confirmation of positive (screening) results by the ELISA method. A possibility to improve the specificity of the ELISA method would be to adjust the cut-off level. The difference between the methods may be due to an unwanted cross-reactivity of the ELISA method, as a positive bias was observed at low levels (Fig. 2(a)). The UPLC-MS/MS method utilizes the same analytical technology as for EtG testing, and is consequently an alternative for those laboratories that already perform EtG analysis.
|
In conclusion, the present study demonstrated that two new methods are available for measurement of urinary GTOL. This might facilitate the use of this alcohol biomarker of recent drinking in routine laboratories. In cases where the EtG test provides an unwanted high sensitivity for intake of very low amounts of ethanol or even unintentional ethanol exposure, the GTOL test may offer a useful alternative as it focuses on moderate-to-heavy drinking.
|
References |
|---|
|
TOPABSTRACTIntroductionExperimentalResults and DiscussionReferences |
|---|
Beck O., Helander A. 5-Hydroxytryptophol as a marker for recent alcohol intake. Addiction (2003) 98(Suppl. 2):63–72.[CrossRef][ISI][Medline]
Bendtsen P., Jones A. W., Helander A. Urinary excretion of methanol and 5-hydroxytryptophol as biochemical markers of recent drinking in the hangover state. Alcohol and Alcoholism (1998) 33:431–438.
Bergström J., Helander A., Jones A. W. Ethyl glucuronide concentrations in two successive urinary voids from drinking drivers: relationship to creatinine content and blood and urine ethanol concentrations. Forensic Science International (2003) 133:86–94.[CrossRef][ISI][Medline]
Borucki K., Schreiner R., Dierkes J., et al. Detection of recent ethanol intake with new markers: comparison of fatty acid ethyl esters in serum and of ethyl glucuronide and the ratio of 5-hydroxytryptophol to 5-hydroxyindole acetic acid in urine. Alcoholism-Clinical and Experimental Research (2005) 29:781–787.[CrossRef]
Center for Substance Abuse Treatment. The role of biomarkers in the treatment of alcohol use disorders. Substance Abuse Treatment Advisory (2006) 5:1–8.
Constantino A., DiGregorio E. J., Korn W., et al. The effect of mouthwash on ethylglucuronide concentrations in urine. Journal of Analytical Toxicology (2006) 30:659–662.[ISI][Medline]
Dahl H., Stephanson N., Beck O., et al. Comparison of urinary excretion characteristics of ethanol and ethyl glucuronide. Journal of Analytical Toxicology (2002) 26:201–204.[ISI][Medline]
Dierkes J., Wolfersdorf M., Borucki K., et al. Determination of glucuronidated 5-hydroxytryptophol (GTOL), a marker of recent alcohol intake, by ELISA technique. Clinical Biochemistry (2006) 40:128–131.[ISI][Medline]
Goll M., Schmitt G., Ganssmann B., et al. Excretion profiles of ethyl glucuronide in human urine after internal dilution. Journal of Analytical Toxicology (2002) 26:262–266.[ISI][Medline]
Helander A. Biological markers in alcoholism. Journal of Neural Transmission (2003) 66:15–32.
Helander A., Beck O., Jones A. W. Laboratory testing for recent drinking: comparison of ethanol, methanol, and 5-hydroxytryptophol. Clinical Chemistry (1996) 42:618–624.
Helander A., Wikström T., Löwenmo C., et al. Urinary excretion of 5-hydroxyindole-3-acetic acid and 5-hydroxytryptophol after oral loading with serotonin. Life Sciences (1992) 50:1207–1213.[CrossRef][ISI][Medline]
Helander A., Beck O., Jacobsson G., et al. Time course of ethanol-induced changes in serotonin metabolism. Life Sciences (1993) 53:847–855.[CrossRef][ISI][Medline]
Helander A., von Wachenfeldt J., Hiltunen A., et al. Comparison of urinary 5-hydroxytryptophol, breath ethanol, and self-report for detection of recent alcohol use during outpatient treatment: a study on methadone patients. Drug and Alcohol Dependence (1999) 56:33–38.[CrossRef][ISI][Medline]
Jones A. W. Urine as a biological specimen for forensic analysis of alcohol and variability in the urine-to-blood relationship. Toxicological Reviews (2006) 25:15–35.[CrossRef][Medline]
Kroke A., Klipstein-Grobusch K., Hoffman K., et al. Comparison of self-reported alcohol intake with urinary excretion of 5-hydroxytryptophol:5-hydroxyindoleacetic acid, a biomarker of recent alcohol intake. British Journal of Nutrition (2001) 85:621–627.[ISI][Medline]
Rohrig T. P., Ross W. Detection of ethyl glucuronide in urine following the application of Germ-X. Journal of Analytical Toxicology (2006) 30:703–704.[ISI][Medline]
Sarkola T., Dahl H., Eriksson C. J. P., et al. Urinary ethyl glucuronide and 5-hydroxytryptophol levels during repeated ethanol ingestion in healthy human subjects. Alcohol and Alcoholism (2003) 38:347–351.
Schmitt G., Aderjan R., Keller T., et al. Ethyl glucuronide: an unusual ethanol metabolite in humans. Synthesis, analytical data, and determination in serum and urine. Journal of Analytical Toxicology (1995) 19:91–94.[ISI][Medline]
Schmitt G., Droenner P., Skopp G., et al. Ethyl glucuronide concentration in serum of human volunteers, teetotallers, and suspected drinking drivers. Journal of Forensic Sciences (1997) 42:1099–1102.[ISI][Medline]
Stephanson N., Dahl H., Helander A., et al. Direct quantification of ethyl glucuronide in clinical samples by liquid chromatography-mass spectrometry. Therapeutic Drug Monitoring (2002) 24:645–651.[CrossRef][ISI][Medline]
Stephanson N., Helander A., Beck O. Determination of 5-hydroxytryptophol glucuronide and 5-hydroxyindoleacetic acid by direct injection of urine using ultra-performance liquid chromatography-tandem mass spectrometry. Journal of Mass Spectrometry (2007) In press.
Wurst F. M., Seidl S., Ladewig D., et al. Ethyl glucuronide: on the time course of excretion in urine during detoxification. Addiction Biology (2002) 7:427–434.[CrossRef][ISI][Medline]
Wurst F. M., Wiesbeck G. A., Metzger J. W., et al. On sensitivity, specificity, and the influence of various parameters on ethyl glucuronide level in urine—results from the WHO/ISBRA study. Alcoholism-Clinical and Experimental Research (2004) 28:1220–1228.[CrossRef]
Wurst F. M., Alling C., Aradottir S., et al. Emerging biomarkers: new directions and clinical applications. Alcoholism-Clinical and Experimental Research (2005) 29:465–473.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||