, 2000). On the other figure 1 hand, nornicotine was shown to contain 70%�C96% (S)-nornicotine (Armstrong, Wang, Lee, & Liu, 1999; Carmella et al., 2000), and it has been previously noted that the enantiomeric composition of NNN in tobacco indicates that nornicotine, and not nicotine, is the major precursor of NNN in tobacco (Carmella et al., 2000). Our results further reinforce that hypothesis. Thus, removal of nornicotine from tobacco could be a potential strategy to reduce the levels of NNN in tobacco products (Gavilano et al., 2006). Even though the variation in percent (S)-NNN among brands and product types was not large, we found statistically significant differences among some product categories. This could be due to the differences in tobacco types and processing methods used.

For instance, the % contribution of (S)-NNN to NNN in novel tobacco products was found to be higher than that in conventional moist snuff (Table 1). Previous research showed that, compared with conventional U.S. moist snuff, Marlboro Snus and Camel Snus products are generally low in tobacco-specific nitrosamine content (Stepanov et al., 2008; Stepanov, Biener, et al., 2012). This is most likely due to differences in tobacco processing methods, with the tobacco used for the manufacturing of novel products undergoing pasteurization��a process known to inhibit TSNA formation in processed tobacco. However, due to the higher percentage of (S)-NNN, the absolute amount of this carcinogenic enantiomer in some novel products can be comparable to those found in conventional moist snuff (Table 1).

This is the first study to measure the enantiomeric composition of NNN in cigarette smoke. The measured percent contribution of (S)-NNN in smoke was similar to that in cigarette tobacco, indicating that no significant thermal racemization of NNN occurs during cigarette burning. The higher carcinogenic potential of (S)-NNN compared to (R)-NNN is indicated in both in vitro and in vivo studies. In vitro, cultured rat esophagus, a target tissue for NNN carcinogenicity, metabolizes (S)-NNN predominantly by 2��-hydroxylation, which is the major bioactivation pathway of NNN in rats (McIntee & Hecht, 2000). In vivo, the urine of rats treated with (S)-NNN contained higher levels of metabolites formed via 2��-hydroxylation than the urine of rats treated with (R)-NNN (McIntee & Hecht, 2000).

Furthermore, treatment of rats with (S)-NNN produced two to six times higher levels of DNA adducts in the esophagus and three to five higher levels of DNA adducts in oral tissue, compared with treatment with (R)-NNN (Lao et al., 2007; Zhang et al., 2009). In agreement with these results, the treatment of rats Dacomitinib with (S)-NNN in our recent study produced a 100% incidence of oral and esophageal tumors, compared with only 5 and 3 out of 24 rats developing oral and esophageal tumors, respectively, upon treatment with (R)-NNN (Balbo et al., 2012).