Ethylation and methylation of hemoglobin in smokers and non-smokers

Two previous studies demonstrated elevated levels of 3-ethyladenine in smokers\u27 urine, suggesting that cigarette smoke may contain a DNA ethylating agent. We hypothesized that such an agent would also lead to elevated levels of N-terminal N-ethylvaline in hemoglobin. N-terminal N-alkylated valines in hemoglobin can be measured using a modified Edman degradation, which employs pentafluorophenyl isothiocyanate to produce a pentafluorophenylthiohydantoin. The latter is quantified by gas chromatography-negative ion chemical ionization-mass spectrometry (GC-NICI-MS). We modified the published method to increase its sensitivity and selectivity, thereby allowing quantification of N-terminal N-ethylvaline. Modifications included the use of a deuterated peptide as the internal standard, the introduction of an HPLC purification step, and the use of tandem mass spectrometry (MS/MS) for detection and quantification of the analyte, 1-ethyl-5-isopropyl-3-pentafluorophenyl-2-thiohydantoin. We also quantified N-terminal N-methylvaline in the same samples. The mean level of N-terminal N-ethylvaline in the hemoglobin of smokers was 3.76 ± 2.77 pmol/g globin (n = 39), significantly higher than in non-smokers, 2.50 ± 1.65 pmol/g globin (n = 28), P = 0.023. The difference remained significant after correction for gender and age. The mean level of N-terminal N-methylvaline in smokers was 997 ± 203 pmol/g globin (n = 45) compared with 904 ± 149 pmol/g globin in non-smokers (n = 29); these values were not significantly different when corrected for gender and age. As levels of hemoglobin and DNA adducts often correlate, the results of this study support the proposal that cigarette smoke contains an as yet unidentified ethylating agent, which might be involved in DNA damage and tumor initiation