High throughput liquid and gas chromatography-tandem mass spectrometry assays for tobacco-specific nitrosamine and polycyclic aromatic hydrocarbon metabolites associated with lung cancer in smokers

Abstract

We developed and applied high throughput liquid and gas chromatography-tandem mass spectrometry (LC-MS/MS and GC-MS/MS) methods for the cigarette smoking-associated biomarkers 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol (NNAL) and r-1,t-2,3,c-4-tetrahydroxy-1,2,3,4-tetrahydrophenanthrene (PheT), which are urinary metabolites of the carcinogenic tobacco-specific nitrosamine 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK) and the polycyclic aromatic hydrocarbon phenanthrene. NNAL and PheT levels have been linked to lung cancer in previous studies of smokers. Confirmation of these relationships will require further molecular epidemiology studies, necessitating improved methodology applicable to large numbers of small urine samples. Furthermore, NNAL is excreted in urine either unconjugated or as an N- or O-glucuronide, but little data are available on the amounts of each in urine. For the high throughput analysis of NNAL, 3 aliquots were processed from each urine sample, one for the analysis of free NNAL, one for free NNAL plus NNAL-N-Gluc, and one for total NNAL (the sum of free NNAL, NNAL-N-Gluc, and NNAL-O-Gluc). Ninety-six well plate technology was used for sample enrichment by supported liquid extraction plates, mixed mode reverse-phase/cation exchange solid-phase extraction, and LC-MS/MS analysis. For the analysis of PheT, the urine samples were cleaned up by solid-phase extraction on styrene-divinylbenzene sorbent, silylated, and analyzed by GC-MS/MS, both in 96-well format. The methods were validated analytically with respect to accuracy and precision, and applied in an ongoing molecular epidemiology study of smokers. The amount of total NNAL in smokers' urine was (mean ± SD) 1.65 ± 2.13 pmol/mL (N = 2641). Free NNAL, NNAL-N-Gluc, and NNAL-O-Gluc represented (mean ± SD) 31 ± 11%, 22 ± 14%, and 48 ± 15% of total NNAL, respectively. The amount of PheT in smokers' urine was (mean ± SD) 1.43 ± 2.16 pmol/mL (N = 2613). The methodology described here should be widely applicable in future studies of tobacco use and cancer.

Figure 1

Structures of the compounds discussed in this paper. NNK (1), NNAL (2), NNAL-O-Gluc (3), NNAL-N-Gluc (4), phenanthrene (5), anti-phenanthrene-1,2-diol-3,4-epoxide (6), and PheT (7).

Figure 2

LC-ESI-MS/MS chromatograms obtained upon analysis of Fraction 1 (Scheme 1) of a human urine sample for free NNAL(2), using the 50 mm column. A and B, monitoring of m/z 210 → m/z 93 for quantitation of NNAL and m/z 216 → m/z 98 for the internal standard [¹³C₆]NNAL. C and D, qualifier ions for NNAL, m/z 210 → m/z 180 and [¹³C₆]NNAL, m/z 216 → m/z 186.

Figure 3

LC-ESI-MS/MS chromatograms obtained upon analysis of Fraction 1 (Scheme 1) of a human urine sample for free NNAL(2), using the 150 mm column. A and B, monitoring of m/z 210 → m/z 93 for quantitation of NNAL and m/z 216 → m/z 98 for the internal standard [¹³C₆]NNAL. C and D, qualifier ions for NNAL, m/z 210 → m/z 180 and [¹³C₆]NNAL, m/z 216 → m/z 186. The E- (retention time 7.86 min) and Z- (retention time 8.07 min) rotamers of NNAL are partially separated.

Figure 4

Determination of the accuracy of the NNAL analysis. A non-smokers’ urine, containing no detectable NNAL, was spiked with 6 increasing amounts of NNAL and the analysis was carried out as described in Materials and Methods. The added and determined amounts correlated, R² = 1.00 and the accuracy was 94.1%.

Figure 5

GC-NICI-MS/MS chromatograms obtained upon analysis of a human urine sample for tetra-TMS derivatives of: (A) PheT (7), m/z 372.130 → m/z 210.07; (B) internal standard [¹³C₆]PheT, m/z 378.016 → m/z 216.070; and (C) injection and derivatization standard [D₁₀]PheT-TMS, m/z 382.160 → m/z 220.070.

Figure 6

Determination of the accuracy of the PheT analysis. A pooled smokers’ urine sample containing 2199 fmol/mL PheT was spiked with 6 increasing amounts of PheT and the analysis was carried out as described in Materials and Methods. Each point is the mean of 4 determinations. The added and determined amounts correlated, R² = 1.00, and the accuracy was 95%. The y-intercept agrees with the amount of PheT in the pooled smokers’ urine sample.

Scheme 1

Outline of the high throughput LC-MS/MS method for analysis of NNAL, NNAL-N-Gluc, and NNAL-O-Gluc in human urine. SLE, solid liquid extraction; SPE, solid-phase extraction.