Tobacco Documents Online

323 III. ISOLATION AND QUANTITATIVE ANALYSIS OF NICOTINE AND Y COTININE IN PHYSIOLOGICAL FLUIDS ~ M. P. Maskarinec, R. W. Harvey, J. E. Caton One of the major purposes of the collaborative dosimetry experiment with Borriston Research Laboratories has been to determine the efficacy of nicotine and/or cotinine concentrations in serum or urine as a quantitative indicator of the smoke dose which an animal retains. In spite of the wide- spread use of tobacco products, there has existed no simple quantitative method for the determination of nicotine or its major metabolite, cotinine, in body fluids. Previous methods for nicotine and cotinine in urine and serum have relied mainly on solvent partition schemes for isolation and purification, while the final determination is made by gas chromatography, using conventional flame ionization detection, nitrogen-selective detection or electron capture detection. While the sensitivity of these methods has been entirely sufficient, the isolation schemes have been generally non-reproducible and tedious. In addition, total analysis times have been rather long. The advent of High- Performance Liquid Chromatography (HPLC) as a rapid, reliable analytical tool has led to the development of a method for the estimation of nicotine and cotinine by HPLC using UV detection (1). This method also employed solvent partition for isolation. Earlier work on the screening of urine for basic drugs of abuse by adsorption onto Amberlite XAD-2 (2) led us to investigate the practicality of this approach to the rapid quantitative isolation of nicotine and its metab- olites from body fluids, while relying on HPLC for the final determination. Method The isolation procedure is the key to reproducibility of the analytical method. Urine (20 ml) is adjusted to pH 9 with saturated NH4C1 buffer (pH 10). Plasma (1-5 ml) is diluted to 20 ml using a 10°o saturated buffer.

324 14C-nicotine (0.27 pCi) was added to each sample to measure recovery. To equilibrate the XAD-2 column, 10 ml of l0l buffer solution is allowed to flow through it. The sample was then passed through the column at a flow rate of approximately 2.5 ml/min. After a second wash with 10% hN4Cl buffer (15 ml), the alkaloids were eluted with 1 ml acetone, followed by 20 ml methanol/ chloroform (1:3). The eluate was biphasic. The methanol layer was discarded, and the chloroform layer gently evaporated to dryness after removal of a 50 01 aliquot for liquid scintillation counting. The residue was redissolved in 200 ul dioxane/isopropanol/NHL,OH (80:3:.4), which was also used as the mobile phase. A 20 ul aliquot was applied to the liquid chromatogrpphic system and eluted at a flow rate of 1.0 ml/min on a 25 cm x 4.6 mm ID ZorbaxSil column. Discussion Six replicate analyses of urine samples indicated a mean recovery (after evaporation) of 85.7% for nicotine and 86.4111 for cotinine, with standard devia- tions of 4.8%) and 0.98%, respectively as shown in Table 111-3. Adsorption of both nicotine and cotinine on the XAD-2 was virtually quantitative. While the alka- loids partitioned between chloroform and methanol approximately 95:5, the urinary pigments (which would be expected to interfere in the final determination) were partitioned into the methanol phase. The accuracy of the method was determined by standard addition of nicotine to a urine sample in which no nicotine was detected (< 2 ng/ml). The results are shown in Figure 111-3. Over the range 10 ng/ ml - 10 ug/m1 all determinations were within 5% of the correct value. Even at a level of 10 ug/m1, no significant breakthrough of nicotine on the XAD-2 columns was observed. A comparison of the liquid chromatograms of the urinary extracts of a smoker and a non-smoker is shown in Figure 111-4. In addition to the nicotine and cotinine peaks, nornicotine was also tentatively identified (by retention time

TABLE 111-3 Recovery of Nicotine and Cotinine from Urine* Nicotine Cotinine # Determinations 6 6 % Unretained 3.69 ± 5.1 1.14 ± .28 % In Methanol 4.2 ± .47 5.87 ± .29 % Ch1 oroform 86.5 ± .48 86.8 ± 0.49 % Recovered After Evaporation 85. 7 ± 4.8 86,4 ± 0.98 *Units are i± I standard deviation.

326 FIGURE 111-3 STANDARD A'DD I T I ON OF N I C®T INE TO UR I NE to ® 0 iof io2 ~03 io' 105 NI CoT INiE RDD ED, NRr1oGRRNS/NI LL I L I Tr-_R

FIGURE 111-4 COMPARISON OF LIQUID CHROMATOGRAMS OF URINARY ALKALOIDS SMOKER (a) VS NON-SMOKER (b) ORNL-DWG 78-7050 (a) I 2 4 6 8 1 1 0 2 4 6 8 w z z ~ 0 c.~ r. I , i i i i i i 40 42 44 96 0 2 4 6 8 40 42 14 46 TIME (min) f 1 11 1 1 40 12 44 46 0 2 4 6 8 40 42 44 46 ELUTION VOLUME (ml) Sensitivity: 0.16 AU Full Scale ~89~~~652

328 comparison with an authentic standard) in the smoker`s urine. The levels in this individual (a heavy smoker of non-filter cigarettes) were 247 ng/ml nicotine and 521 ng/ml cotinine (after recovery correction). The nicotine level in the non-smoker was 7 ng/ml while the. cotinine level was below the detection limits of the method (2 ng/ml). The extracts were further analyzed by glass capillary gas chromatography on a 50 m x 0.26 mm ID OV-101 column using simultaneous nonselective (FID) and nitrogen-selective (PdPFID) detection (Perkin-Elmer Model 3920 Gas Chromato- graph). The results were within 5% of those obtained by HPLC. To further test the validity of the method, a series of urine samples from smoking baboons(obtained from Dr. Walter Rogers, SWRI) were analyzed blind for nicotine and cotinine and the results were in close agreement (p. < .05) with those obtained in a separate laboratory by gas chromatography- mas spectrometry (3). While urinary levels of nicotine and its metabolites are important in screening for cigarette smoking and in determining total smoke dose, it was of interest to examine the possibility of determining plasma nicotine levels by this method. Recovery of plasma nicotine was found to be similar to the urinary recovery. A typical liquid chromatogram obtained from a 5 ml plasma sample of a dog exposed to cigarette smoke in the BRLL chronic exposure is shown in Figure 111-5. The recovery corrected nicotine level in this case was found to be 112 ng/ml. However, subsequent studies have shown that due to the rapid clearance of nicotine from the plasma, the utility of this deter- mination may be limited. References 1. Ian D. Watson. Rapid analysis of nicotine and cotinine in the urine of smokers by high-performance liquid chromatography. J. Chromatoqr. ~ 143: 203-206 (1977).

329 2. M. P. Kullberg and C. W. Gorodetsky. Studies on the use of XAD-2 resin for detection of absued drugs in urine. Clin. Chem. 20(2): 177-183 (1974). 3. W. Rogers: Personal Communication (1978).

330 FIGURE 111-5 LIQUID CHROMATOGRAM OF NICOTINE IN DOG PLASMA -ORNL-DWG 78-7049 __F_1 2 4 6 8 40 12 44 46 TIME (min) I 0 2 4 6 8 40 12 44 16 ELUTION VOLUME (m! ) Sensitivity: 0.08 AU Full Scale