Journal of Analytical Toxicology 2013;37:391 –394 doi:10.1093/jat/bkt052 Advance Access publication July 10, 2013
Article
Detection of In Utero Marijuana Exposure by GC –MS, Ultra-Sensitive ELISA and LC –TOF –MS Using Umbilical Cord Tissue A. Chittamma1, S.J. Marin2*, J.A. Williams1, C. Clark1 and G.A. McMillin1,3 1 3
ARUP Laboratories, Inc., Salt Lake City, UT, USA, 2ARUP Institute for Clinical and Experimental Pathology, Salt Lake City, UT, USA and Department of Pathology, University of Utah School of Medicine, Salt Lake City, UT, USA
*Author to whom correspondence should be addressed. Email:
[email protected]
Smoking marijuana during pregnancy can cause health problems in the neonate. The detection of exposure can guide treatment to meet the short- and long-term medical and social needs. Umbilical cord tissue was analyzed for 11-nor-delta-9-carboxy-tetrahydrocannabinol (THC-COOH) by gas chromatography– mass spectrometry (GC –MS), and compared with ultra-sensitive enzyme-linked immunosorbent assay (ELISA) and liquid chromatography time-of-flight mass spectrometry (LC –TOF–MS). Fortified extracts of drug-free cord tissue were used to determine the sensitivity and specificity of the LC– TOF–MS and ELISA assays, and 16 de-identified patient specimens previously analyzed by GC–MS were tested for THC-COOH by both methods. The cutoffs were 0.050 ng/g for the GC–MS assay, 0.1 ng/g for the ELISA assay and 1 ng/g for the LC–TOF–MS assay. Twelve specimens were negative by all three methods. Seven specimens were positive by GC –MS with concentrations from 0.066 to 6.095 ng/g. ELISA and LC –TOF–MS did not detect one specimen that was positive by GC–MS. LC–TOF–MS missed one specimen that was detected by GC –MS and ELISA. Five positive specimens were detected by all three methods. These results were consistent with the cutoff for each method. No false positives were detected by LC–TOF– MS or ELISA. Umbilical cord tissue is a viable specimen for the detection of in utero marijuana exposure. ELISA and GC–MS were more sensitive than LC–TOF–MS for the detection of THCCOOH in cord tissue, with the GC –MS method providing superior sensitivity.
Introduction Cannabis is one of the most widely used illicit drugs in the world, with the estimated annual use in 2010 ranging from 2.6 to 5% of the adult population. In the USA, the annual prevalence of cannabis use among the general population (15–64 years of age) continued to increase from 13.7% in 2009 to 14.1% in 2010 (1). Among pregnant women, the percent of women who admit to cannabis use varies from 4.6% in the first trimester to 1.4% in the third trimester (2). The major psychoactive component of cannabis is D9-tetrahydrocannabinol (THC), which can pass through the placenta, and has profound but variable effects in several areas of brain development and cognitive function (3). Prenatal exposure to cannabis has been associated with low birth weight, prematurity, shorter birth length and an increased rate of admission to the intensive care unit (4). Early detection of in utero cannabis exposure can guide neonatal treatment and social management. Because maternal self-report of drug use is unreliable, prenatal cannabis exposure is typically detected by drug testing. The traditional approach to drug testing is a two-step process, starting
with a drug screen (typically an immunoassay), followed by confirmation and quantitation using a mass spectrometry method (5). Immunoassays for detecting cannabis abuse determine D9-THC and its metabolites in biological specimens such as maternal urine, maternal blood, cord blood, neonate urine or meconium. These methods are adopted as a preliminary test because they are sensitive and cost-effective; however, false positive or false negative results may occur from variations in the relative percentages and forms of D9-THC metabolites present in the samples and variability in the cross-reactivity of the antibodies used in the assays (6). Although immunoassays developed for urine can be adapted for alternative specimens, the crossreactivity characteristics selected for urine cannabinoids screening may not be optimal for other biological matrices. Therefore, cannabinoids immunoassays for each type of biological specimen should be designed and interpreted based on testing strengths and limitations (5, 6). Gas chromatography and liquid chromatography are routinely coupled with mass spectrometry (GC–MS, LC–MS or LC–MS– MS) to support the confirmation and quantification of a prominent marijuana metabolite, 11-nordelta-9-carboxy-tetrahydrocannabinol (THC-COOH). Recently, LC–MS or LC–MS –MS have increased in popularity, in part because they offer greater sensitivity than GC–MS, without derivatization, for some drug analytes (7). Of the specimens routinely collected for drug testing, blood and urine provide the shortest window for detection. False negative and false positive results are common for neonatal urine, due to relatively low concentration and a documented failure to confirm (8). To widen the window of detection, meconium, the first stool of the newborn, has become the specimen of choice. Meconium reflects mostly the last trimester of drug exposure from maternal use (5, 9). The collection of meconium from diapers is noninvasive; however, the meconium can be expelled prior to, or during, delivery and become unavailable for testing. Further, a drug-exposed infant may not expel meconium for days to weeks after birth, and it is often of limited quantity, which may limit the extent of testing that can be performed (9). The meconium matrix is also very complex. Extraction and specimen preparation prior to instrumental analysis may prove difficult and time-consuming with meconium. Montgomery et al. and others (10 –13) have demonstrated that umbilical cord tissue is an alternative to meconium for detecting fetal drug exposure. The major advantages of umbilical cord tissue are that the sample is available immediately at birth, for all neonates, and there is typically a large amount of tissue (a typical cord is 22 inches long) available for testing. The aim of this study was to compare THC-COOH results in umbilical cord tissue using high-resolution accurate mass liquid
# The Author [2013]. Published by Oxford University Press. All rights reserved. For Permissions, please email:
[email protected]
chromatography time-of-flight mass spectrometry (LC – TOF – MS), ultra-sensitive enzyme-linked immunosorbent assay (ELISA) and gas chromatography–mass spectrometry (GC–MS).
Methods Samples A segment of umbilical cord was collected, rinsed in saline and patted dry prior to testing. Blank (drug-free) umbilical cords were provided by the Intermountain Medical Center (Murray, UT, USA) after collection and deidentification, from healthy births, clinically assessed to be low-risk for drug exposure, and qualified to be drug-free by LC–TOF–MS. Residual umbilical cord specimens (n ¼ 16), analyzed initially by an outside laboratory (USDTL, Des Plaines, IL, USA) using GC –MS, were acquired and deidentified according to approved protocols of the University of Utah Institutional Review Board, and stored at 2808C until subsequent analysis by LC– TOF –MS and ELISA.
Reagents for LC –TOF– MS THC-COOH was obtained from Cerilliant (Round Rock, TX, USA). All analytical reagents and solvents were LC–MS or HPLC grade and purchased from VWR (West Chester, PA, USA) or Thermo Fisher Scientific (Waltham, MA, USA). A Barnstead NanoPure Infinity ultra-pure water system (Thermo Fisher Scientific) was used for obtaining Type 1 (nanopure) water. The drug standard solution was prepared in methanol:nanopure water (75:25, v/v) at 1 ng/mL and stored at 2808C.
Sample preparation and extraction for LC–TOF–MS Each umbilical cord tissue sample (1.00 + 0.1 g) was weighed into a bag (Covaris, Inc., Woburn, MA, USA) attached to a 5-mL labeled culture tube. After placing the bag in liquid nitrogen to freeze the tissue, the tissue specimens were pulverized by the CryoPrep (Covaris Inc.) and homogenized with 2 mL of cold acetonitrile (2208C) while vortexing, then mixed for 1 min. Homogenates were centrifuged for 10 min at 64g and 08C. Each supernatant was transferred to a 10-mL labeled culture tube and 2 mL of nanopure water and 400 mL of 0.5 M NaOH were added. These samples were then applied to individual CEREXw hpspeTM THC SPE columns (SPEware, Inc., San Pedro, CA, USA) that had been conditioned with 1 mL of methanol followed by 1 mL of nanopure water, then placed in a SYSTEM-48TM CEREXw PRESSURE PROCESSOR (SPEware, Inc.). The SPE method, with two elution steps, was originally designed to recover both acidic and basic compounds. Next, the specimens were washed with 1 mL of nanopure water, acetonitrile and NH4OH (85:15:1, v/v/v) and air dried for 10 min at 80 psi, then eluted with 3 mL of hexane and ethyl acetate (50:50, v/v). One milliliters of methanol followed by 2 mL of ethyl acetate were used to wash the columns before being eluted with 2 mL of hexane, ethyl acetate and acetic acid (90:10:2, v/v/v). The extracts were evaporated to dryness at 408C under a gentle nitrogen stream with a Turbo Vapw LV (Biotage, Charlotte, NC, USA). The residues were reconstituted in 50 mL of methanol:nanopure water (75:25, v/v) and transferred onto 1.5 mL maxrecovery autosampler vials for analysis. Positive controls were 392 Chittamma et al.
prepared from drug-free umbilical cord tissue spiked with stock solutions in methanol and extracted in the same manner as patient samples. A negative control was also prepared and analyzed, and an unextracted control (no matrix or extraction) was analyzed to check instrument function before and at the end of sample analysis.
LC– TOF–MS conditions The LC was carried out by the Agilent 1260 series system (Agilent Technologies, Santa Clara, CA, USA) consisting of an autosampler, degasser, two binary pumps and a column oven with a 10 port switching valve. Separation was performed at 558C on a Poroshell 120 C8, 3.0 100 mm, 2.7-mm particle size column (Agilent Technologies) using isocratic elution with 25% of mobile Phase A (5 mM ammonium formate, pH 3.5) and 75% of mobile Phase B (methanol) for 10 min. The flow rate was set at 0.5 mL/min, and the injection volume was 40 mL. A 1-min post run time was added after each analysis. An Agilent 6230 TOF mass spectrometer equipped with a Jet Stream ESI source (Agilent Technologies) was used for TOF analysis. The mass spectrometry parameters were previously described and are shown in Table I (14). THC-COOH was identified using the Agilent MassHunter Qualitative Analysis software B.05.00I, based on the compound’s molecular formula, exact mass, retention time and chemical structure. The retention time was determined by a running standard solution at 1 mg/mL. The criteria of mass tolerance and retention time were set to +25 ppm and +0.1 min, respectively. Sensitivity was determined by analyzing spiked samples and choosing the lowest concentration that could be reliably detected with a signal-to-noise ratio of the 5:1 and the data meeting the TOF criteria above. Drug-free cord spiked at the cutoff was analyzed with the patient specimens to verify sensitivity with each batch of samples. A negative drug-free cord tissue control was also included.
Ultra-sensitive enzyme-linked immunosorbent assay Reagents for ELISA The ultra-sensitive cannabinoid ELISA kit was purchased from Immunalysis Corporation ( part number 230-0480, Immunalysis Corporation, Pomona, CA, USA). Phosphate buffered saline (PBS) was prepared in-house. Table I LC –TOF –MS parameters Instrument mode
Fast polarity switching (1,700 amu, 2 GHz, extended dynamic range)
Nebulizer (psi) Drying gas flow (L/min) Drying gas temperature (8C) Sheath gas flow (L/min) Sheath gas temperature (8C) Fragmentor voltage (V) Capillary voltage (V) Nozzle voltage Scan range (amu) Scan speed Reference masses positive ion:
30 6 350 12 400 125 3,500 V, positive ion; 2,500 V, negative ion 2000 105 – 1000 1.7 scans/s (5,000 transients/scan) Purine m/z 121.050873, HP-921 m/z 922.009798
Sample preparation and extraction for ELISA The pulverized cord tissue samples (calibrators and controls) and patient specimens (1.00 + 0.1 g) were prepared as described above and homogenized twice for 5 min after adding 0.9 –1.5 mm stainless steel beads and 2 mL of 0.1% Triton X-100 by using a Bullet Blender (Next Advance, Averill Park, NY, USA). Homogenates were incubated for 1 h at room temperature and then centrifuged for 15 min at 150g and 08C. Two hundred microliters of each supernatant were transferred to a deep well plate and 800 mL of PBS were added. ELISA method procedure Each umbilical cord extract was evaluated for THC-COOH using the ultra-sensitive ELISA kit (Immunalysis) following the manufacturer’s instructions. Briefly, a 200-mL aliquot of each extracted sample was transferred to individual wells of the ELISA plates and incubated for 1 h in the dark. After a washing step, 200 mL of horseradish peroxidase-labeled THC derivative were added to each well and incubated for 1 h in the dark. The wells were washed and a chromogenic substrate was added. The reaction was stopped with 1 N HCl and the plate was read at 450 nm. The intensity of the color development was inversely proportional to the concentration of drug in the sample. The calibrator was THC-COOH. Sensitivity was determined by running a negative sample prepared from drug-free cord tissue and analyzing spiked drug-free cord tissue samples at concentrations between 0.10 and 0.50 ng/g and evaluating b/b0 ( percent binding, the optical
density of spiked sample divided by the optical density of negative sample) curves to determine the cutoff. The cutoff concentration was 0.10 ng/g (100 pg/g). Precision was determined by running a negative specimen, and specimens spiked at the cutoff and twice the cutoff (0.10 and 0.20 ng/g, respectively) in triplicate over 5 days. These were all prepared from a single cord specimen. Average imprecision (% CV) was ,10% for the negative samples and cutoff calibrators. Imprecision was ,20% for the 0.20 ng/g samples, except for 1 day when the value was 21%. A negative control, cutoff calibrator and a positive control spiked at 0.20 ng/g were run with patient specimens. Results and discussions We present a comparison of THC-COOH analysis in umbilical cord tissue by GC–MS, LC–TOF–MS and ultra-sensitive ELISA methods. Results from all 3 assays agreed with each other, respective to their established cutoffs. Figure 1 illustrates the extracted ion chromatogram (EIC) and mass spectra of THC-COOH obtained from Patient 3 by the LC–TOF–MS method. The retention time of THC-COOH was 3.564 min. The cutoff concentration was 1.0 ng/g, obtained by analyzing spiked umbilical cord samples at various concentrations. The cutoff for the GC– MS assay was reported to be 0.05 ng/g (50 pg/g, USDTL), and that for the ELISA assay was determined to be 0.1 ng/g (100 pg/g, Immunalysis). Of the 16 patient cord segments, 12 specimens were negative by all three methods (Table II). One of the positive samples (#2, 0.066 ng/g) was not detected by ELISA or
Figure 1. The EIC (A) and mass spectrum (B) obtained by the LC–TOF –MS method for Patient 3 whose THC-COOH concentration is near to the cutoff at 1 ng/g. The insert is a magnified view of the molecular ion and isotopes for THC-COOH.
Cord Tissue LC– TOF– MS Drug Exposure 393
References
Table II LC –TOF–MS results compared with ELISA and GC– MS methods Sample
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Total positives
Methods and cutoffs LC –TOF –MS (cutoff 1 ng/g)
ELISA (cutoff 0.1 ng/g)
GC – MS concentration (cutoff 0.05 ng/g)
þ 2 þ 2 þ 2 2 þ 2 þ 2 2 2 2 2 2 5
þ 2 þ þ þ 2 2 þ 2 þ 2 2 2 2 2 2 6
3.156 0.066 1.075 0.255 2.878 2 2 2.020 2 6.095 2 2 2 2 2 2 7
LC–TOF–MS. Another specimen (#4, 0.255 ng/g) was detected by ELISA, but not by LC–TOF–MS. Five positive results (ranging from 1.075 to 6.095 ng/g) quantitated by GC–MS were detected by LC–TOF–MS and ELISA. These results are consistent with the established cutoffs for each method. Agreements for LC–TOF– MS and ELISA were 87.5 and 93.8%, respectively, compared with GC–MS. Sensitivity (cutoff ) for THC-COOH was most likely affected by the difference of sample preparation protocols. Because THC-COOH is extensively glucuronidated, an assay that is designed to only detect free THC-COOH, or one that does not include a hydrolysis reaction as part of the sample preparation, may have compromised sensitivity (15, 16). El Sohly and Feng (15) found significant increases in cannabinoid concentrations following enzymatic hydrolysis in meconium. Hydrolysis was not performed in the sample preparation for LC–TOF –MS and ELISA methods described here. In addition, the THC-COOH may not be the most appropriate analytical target for umbilical cord tissue. Indeed, reports have shown that 11-hydroxy-tetrahydrocannabinol (11-OH-THC) and 8b-11-diOH-D9-THC are also abundant biomarkers of THC in meconium, particularly following hydrolysis. The compositions of drugs and metabolites that deposit in umbilical cord tissue are not fully known or understood at this time. Further, there is no known correlation between umbilical cord concentration of THC-COOH and exposure, or impairment in the neonate. As such, a clinically relevant cutoff concentration remains to be established. Our data showed that umbilical cord tissue is a good biological specimen for detecting in utero cannabinoid exposure by LC– TOF–MS, ELISA and GC –MS methods. Although the cutoff for each method was possibly affected due to different sample preparations, good agreement was found for all three assays at concentrations that exceeded the cutoff for each of the assays. The GC –MS assay provided the best sensitivity, followed by ultrasensitive ELISA and then LC– TOF –MS.
394 Chittamma et al.
1. Results from the 2010 National Survey on Drug Use and Health: Summary of National Findings. United States of America, Department of Health and Human Services, Substance Abuse and Mental Health Services Administration, 2010. http://www.samhsa.gov/data/ NSDUH/2k10NSDUH/2k10Results.pdf (24 February 2013, date last accessed). 2. Substance Abuse and Mental Health Services Administration: Substance Use among Women During Pregnancy and Following Childbirth, the NSDUH Report. The Office of Applied Studies, Substance Abuse and Mental Health Services Administration, 2009. http://www.samhsa.gov/data/2k9/135/PregWoSubUseHTML.pdf (24 February 2013, date last accessed). 3. Jutras-Aswad, D., Dinieri, J.A., Harkany, T., Hurd, Y.L. (2009) Neurobiological consequences of maternal cannabis on human fetal development and its neuropsychiatric outcome. European Archives of Psychiatry and Clinical Neuroscience, 259, 395–412. 4. Hayatbakhsh, M.R., Flenady, V.J., Gibbons, K.S., Kingsbury, A.M., Hurrion, E., Mamun, A.A. et al. (2012) Birth outcomes associated with cannabis use before and during pregnancy. Pediatric Research, 71, 215–219. 5. Cotton, S.W. (2012) Drug testing in the neonate. Clinics in Laboratory Medicine, 32, 449–466. 6. Tsai, J.S. (2007) Immunoassays for the detection of cannabis abuse. In El Sohly, MA (ed). Marijuana and the Cannabinoids, 1st edition, Chapter 7. New York, NY: Humana Press, 145– 178. 7. Foltz, R.L. (2007) Mass spectrometric methods for determination of cannabinoids in physiological specimens. In El Sohly, MA (ed). Marijuana and the cannabinoids, 1st edition, Chapter 8. New York, NY: Humana Press, 179– 204. 8. Barakauskas, V.E., Davis, R., Krasowski, M.D., McMillin, G.A. (2012) Unresolved discrepancies between cannabinoid test results for infant urine. Clinical Chemistry, 58, 1364– 1367. 9. Lozano, J., Garcı´ a-Algar, O., Vall, O., de la Torre, R., Scaravelli, G., Pichini, S. (2007) Biological matrices for the evaluation of in utero exposure to drugs of abuse. Therapeutic Drug Monitoring, 29, 711–734. 10. Montgomery, D., Plate, C., Alder, S.C., Jones, M., Jones, J., Christensen, R.D. (2006) Testing for fetal exposure to illicit drugs using umbilical cord tissue vs meconium. Journal of Perinatology, 26, 11– 14. 11. Montgomery, D.P., Plate, C.A., Jones, M., Jones, J., Rios, R., Lambert, D.K. et al. (2008) Using umbilical cord tissue to detect fetal exposure to illicit drugs: a multicentered study in Utah and New Jersey. Journal of Perinatology, 28, 750– 753. 12. de Castro, A., Jones, H.E., Johnson, R.E., Gray, T.R., Shakleya, D.M., Huestis, M.A. (2012) Methadone, cocaine, opiates, and metabolite disposition in umbilical cord and correlations to maternal methadone dose and neonatal outcomes. Therapeutic Drug Monitoring, 33, 443–452. 13. Buchi, K.F., Suarez, C., Varner, M.W. (2013) The prevalence of prenatal opioid and other drug use in utah. American Journal of Perinatology, 30, 241– 244. 14. Marin, S.J., Hughes, J.M., Lawlor, B.G., Clark, C.J., McMillin, G.A. (2012) Rapid screening for 67 drugs and metabolites in serum or plasma by accurate-mass LC-TOF-MS. Journal of Analytical Toxicology, 36, 477–486. 15. El Sohly, M.A., Feng, S. (1998) Delta 9-THC metabolites in meconium: identification of 11-OH-delta 9-THC, 8 beta,11-diOH-delta 9-THC, and 11-nor-delta 9-THC-9-COOH as major metabolites of delta 9-THC. Journal of Analytical Toxicology, 22, 329– 335. 16. Coles, R., Clements, T.T., Nelson, G.J., McMillin, G.A., Urry, F.M. (2005) Simultaneous analysis of the Delta9-THC metabolites 11-nor-9-carboxy-delta9-THC and 11-hydroxy-delta9-THC in meconium by GC-MS. Journal of Analytical Toxicology, 29, 522–527.
