, 1993). However, metabolism of nicotine to nornicotine, even though technical support potentially a contributing factor in the endogenous NNN synthesis in NRT users, is not likely to be a major determinant of the effectiveness of this process. If endogenous formation of NNN depended on the enzyme-regulated metabolism of nicotine to nornicotine, significant interindividual, but not intraindividual, differences in urinary excretion of NNN would be observed in our previous studies (Stepanov, Carmella, Briggs, et al., 2009; Stepanov, Carmella, Han, et al., 2009). The most likely sources of nornicotine in saliva are cigarette smoke or extraction from oral tobacco or NRT products. We here found that a single piece of nicotine gum or lozenge contains 100 to 200 times higher amount of nornicotine than the amount used in our incubation experiments.

Users of nicotine gum extract up to 70% of its nicotine content (Benowitz, Jacob, & Savanapridi, 1987), and the same is likely true for nornicotine. Thus, chewing several pieces of nicotine gum per day for prolonged periods of time can potentially expose NRT users to significant amounts of orally synthesized NNN, depending on dietary habits, oral health status, and other factors. Conditions in the stomach are even more favorable for the nitrosation reactions (Mirvish, 1975). There are no data available on intragastric NNN synthesis in humans. In summary, our results demonstrate that NNN can be formed in human saliva in the presence of nornicotine, supporting the hypothesis that NNN can be synthesized endogenously in humans.

Oral synthesis of NNN could potentially contribute to the overall intake of this carcinogen by some smokers and smokeless tobacco users, affecting their risk of developing cancer. Removal of nornicotine from NRT products should be considered in order to protect consumers from being exposed to this potent carcinogen. Funding This work was supported by the National Cancer Institute [CA-81301] and the National Institutes of Health [DA-13333]. Declaration of Interests There are no competing interests. Acknowledgment We thank Bob Carlson for editorial assistance.
Smoking during pregnancy poses a great threat to the unborn child and is associated with a range of negative pregnancy outcomes, such as low birth weight, stillbirth, and sudden infant death syndrome (Cnattingius, 2004; Shea & Steiner, 2008).

Although these risks are well documented and despite a general declining prevalence of smoking in Norway and most industrialized countries, studies typically show that as Brefeldin_A many as 13�C20% of women do smoke during pregnancy (Colman & Joyce, 2003; Helleve, Weis?th, & Lindbak, 2010; Kahn, Certain, & Whitaker, 2002; Martin et al., 2008). For many women, becoming pregnant is probably the strongest motivator for smoking cessation and a unique opportunity to modify their health behavior in order to protect their unborn child from harm (Ludman et al.