7.817, 7.849, and 7.957 mm, respectively will be ... - Clinical Chemistry

1 downloads 0 Views 450KB Size Report
test and would, therefore, not be tested by the confir- mation method. However, in samples where both 9THCA and ritodrine metabolites coexist, false-negative.
Ion

31310

amu

from

Ion

395406

f

316.10

am,,

from

395406

228

400

I

7;57

1.

1.

o__ 7.5 Ion

357.10

8.0 am,,

8.5 from

372.10

.o amu

from

241j

7.5

395406

.. Ion

9.0

8.0

Ion

360.10

amu

Ion

375.10

9.0

395406

from

.A8L

15:

385406

8.5

9:0

10:1

amu

from

395406

.s

Fig. 1. GC/MS SIM chromatograms from a 9THCA analysis showing two interfering substances and the internal standard, d3-9THCA, in the mlz 316 ion track at retention times of 7.817, 7.849, and 7.957 mm, respectively ference and alters the procedure to remove it, the ion ratios of the internal standard will be outside the expected limits, and the assay will fail to meet quality-control criteria. Many laboratories prefer to limit quality-control failures and would report the result as

negative

(falsely

negative

in this

case). The coelution of ritodrine metabolite(s) is unlikely to produce a false positive, as was reported for an amphetamine/methamphetamine immunoassay (4). In drug-testing laboratories where each sample must be positive by two tests based on different principles, ritodrine metabolites would not be screened positive by the initial 9THCA immunoassay test and would, therefore, not be tested by the confirmation method. However, in samples where both 9THCA and ritodrine metabolites coexist, false-negative confirmations can result. Recognizing ritedrine and its behavior in a 9THCA analysis can assist efforts to eliminate its interference. Identifying it can also assist expert witnesses who may be asked to explain to finders of fact in legal proceedings the “unknown” substance appearing in a chromatogram that is similar to 9THCA and its deuterated internal standard. Presented in part at the American Academy of Forensic Sciences Annual Meeting, Philadelphia, February 15-20, 1988. The opinions expressed are those of the authors and not to be construed as official or as reflecting the views of the Department of Army or Department of Defense. References 1. Whiting

JD, Manders WW. Confirmation of 9-carboxy-THC in urine by gas chromatography/mass spectrometry. Aviat Space Environ Med 1983;54:1031-3. 1306

2. Brashear WT, Kuhnert BR, Wei R. Maternal and neonatal urinary excretion of sulfate and glucuronide ritodrine conjugates. Cm Pharmacol Ther 1988;44:63441. 3. Mills T, Roberson JC. Instrumental data for drug analysis, 2nd ed. New York: Elsevier, 1987:2052. 4. Nice A, Maturen

A. False-positive urine amphetamine screen with ritodrine. Clin Chem 1989;35:1542-.3.

Berndt-Ingo Podkowik Hoff mann-La!? oche AG CH4002 Base! Switzerland

Mark L. Repka Annapolis Annapolis,

OB-GYNAssoc., MD 21146

PA.

Michael L. Smith Div. of Forensic Toxicol. Armed Forces Institute of Pathol. Washington, DC 20306-6000

Blotin Radloilgand Assay with Polyethylene Glycol as Separation Reagent To the Editor: The clinical and biochemical importance of biotin is well known (1). A few years ago, we reported a very sensitive radioligand assay for the determination of this vitamin in human serum by using an ‘251-labeled biotin deriva-

tive as radiotracer, avidin, rabbit antiavidin antibody, and sheep anti-rabbit IgG in combination with polyethylene glycol (2). Here, we present a modification of this method based on polyethylene glycol (PEG, M 6000; Merck, Darmstadt, F.R.G.) as separation re-

CLINICALCHEMISTRY,Vol. 37, No. 7, 1991

agent, which eliminates the necessity of using the double-antibody separation step and expedites considerably the assay procedure. We synthesized the msIlabeled biotin derivative used in the assay as previously described (2). The standards were prepared in 40 mmol/L phosphate buffer (pH 6.5) containing 7.2 g of NaCl and 36 g of bovine serum albumin per

liter. The polyethylene glycol solution contained 200 g of PEG per liter of 50 mmol/L phosphate buffer (pH 7.4). The assay was performed by mixing 100 L of biotin standards (20-1250 ng/L) or patients’ serum (pretreated with p-hydroxymercuribenzoate to inhibit biotinidase activity), 200 L of radiotracer solution

(350 kCi/mol,

in 0.5 molfL

phosphate buffer, pH 6.5), and 50 L of avidin solution (35 ,.g/L, containing 1.6 g of human serum albumin per liter) in a 3-mL polystyrene tube and incubating for 15 mm at 20-22#{176}C. We then added 50 L of biotin-free, biotinidaseinhibited serum (as a source of protein) or 50 L of phosphate buffer (0.5 mol/L, pH 6.5) to the standards or the test sample tubes, respectively, followed by 2 mL of PEG solution. After centrifuging the samples at 3000 x g for 15 mm, we measured the radioactivity of the precipitate. Using this procedure (y), we estimated the biotin concentrations in 33 human sera and compared the values with those obtained by the biotin radioligand assay (x) previously described (2).The results obtained by the two methods correlated very well (r = 0.936, P = 0.01): y = 7.755 (± 38.89) + 0.966 (± 0.133)x (mean ± SD; standard error of the estimate = 62.2). The nonspecific binding of the proposed assay was 50 failed to

meet quality-control

criteria,

owing to

at relative retention time 1.012 containing a mlz 360 ion that interfered with the measurement of the ion ratios of the internal standard. A sample ion chromatogram in Figure a substance

1 shows the substance in the miz 360 ion track at a retention time of 8.300 mm. Careful chromatography in this case had nearly separated it from d39THCA, which had a retention time of 8.198 miii. When the certifying scien-

Ion

360.30

9.0

7.5

017195

amu

from

017195

26

8.0 mm,,

8.5 from

9.0

017195

I

I

1 1

8.300



0

Radioimmunochem. Lab. Inst. of Radioisotopes/Radiodiagnostic Products NCSR Demokritos” Aghia Paraskevi Attikis, 15310 Greece

316.30

0 7.5

S. A. Evangelatos S. E. Kakabakos E. Livaniou G. P. Evangelatos

Ion 944

-.--.---

7.5

Ion

372.30

8.0 mm,,

8.5 from

9.0

7.5

Ion

017185

1000

375.30

8.0

amu from

8.5

9.0

017195

209

-

0 7.5

8.0

8.5

9.0

7.5

8.0

9.5

9.0

Fig. 1. GCMS SIM ctiromatogran from a 9THCA analysis showing an interferingsubstance and the Internal standard, d3-9THCA, in the m/z 360 ion track at respective retention times of 8.300 and 8.198 mm. The interferingsubstance which often coeluted and caused assay interference,was identifiedas the methylatedformof the secondaryThC metabolite8-hydroxy-3’,4’,5’-trls-nor-delta-9-THC-2’-oic acid tists first noticed this recurring problem, they thought it was an extraction or chromatography artifact; subsequent observations showed that it was sample-specific. Therefore, we prospectively investigated three samples. The urine samples selected, donated for testing in a random urinalysis program, had given a positive screening result by radioimmunoassay (Roche Diagnostic Systems, Inc., Nutley, NJ), and failed to meet ion-ratio requirements for the internal standard as described. (All of the sample results are more than one year old and not currently involved in litigation.) The samples were re-extracted and the substance was separated by using either an instrument with better resolution or the modified chromatographic procedure described previously (1). A full mass spectral analysis produced the following characteristic mass ions (mlz): 360 (43%; M), 345(100%; M - CH3), 301 (40%;M COOCH3), 329 (18%; M - OCH3), 285(18%; M - CH3 - CH2OH - COHi, 302 (11%;M - CH3 CH2OH1, 59(9%; COOCH3i. The percentage of relative abundance and the possible assignment of fragments are given parenthetically. The chemical characteristics of the unknown were similar to those of an acidic metabolite of THC, as recently described in a review mentioning 19 of these metabolites (2). The full mass spectrum was consistent with only one of these, 8-hydroxy-3’,4’,5’-tris-nordelta-9-THC-2’-oic acid (dibenzopyran system of nomenclature). Nordqvist et al. (3) published a mass spectrum for this metabolite, compound IX, isolated from rabbit urine, but their derivative -

had trimethyl-silylated hydroxyl moieties and a methylated carboxylic acid group. Also, they used an MS electron impact voltage of 20 eV instead of the 70 eV used by our laboratory. These different conditions made comparison of mass fragmentation patterns unreliable. For these reasons, and because we were unable to obtain pure reference material for a positive identification, we used the combination of chemical properties and probable mass assignments to identify the metabolite. Being able to recognize the appearance of this secondary THC metabolite had several benefits. It helped the laboratory’s chemists direct their efforts to improve the quality of GCIMS confirmations. For example, they began to investigate the use of another internal standard and different chromatographic conditions instead of devoting their efforts to cleaning up sample extracts to remove the contaminant. Knowing the identity of the interfering substance also assisted expert witnesses responsible for explaining laboratory results to finders of fact in legal proceedings. The members of courts, usually laymen, became concerned when an “unknown” substance similar to 9ThCA was identified in the urine of the accused, even when this interference was removed, as in the sample in Figure 1. Explaining to them that the interference was from another metabolite of the drug of interest reduced their anxiety and strengthened the weight of the laboratory evidence. The opinions in this letter are those of the authors and not to be construed as official or as representing the views of the Department of Army or Department of Defense.

CLINICAL CHEMISTRY,

Vol. 37, No. 7, 1991 1307