Recent developments of sequencing technologies and computational methods have given the opportunity to detect natural variation underlying complex traits in crops. For this purpose, association mapping can be used to identify the link between tobacco genes or molecular markers to smoke constituents. Such association can support the development of new tobacco varieties suitable for future potential regulatory constraints.

In our study, a panel of 161 tobacco varieties, composed of flue-cured, Burley, dark air-cured and Oriental types, was grown in open field. Because of its large genome, RNA-Seq based sequencing was chosen to capture differences of gene expression together with SNPs variation in the 161 varieties. Three different tissues at two growth stages were used to do a comprehensive analysis of the transcriptome. After curing, cigarettes were made with each variety and were mechanically smoked according to the Canadian Intense smoking conditions. The mainstream harmful or potentially harmful constituents (HPHCs) listed in the FDA abbreviated HPHC list^[1] were determined using in-house and internationally recommended methods.

Taking into account multiple factors linked to the growing environment and weight of tobacco actively burnt during puffing, our first investigations showed significant differences of several smoke constituents between varieties, thus demonstrating the potential of association mapping for the development of future varieties. For some constituents, differences were however not significant. A description of the approach and preliminary results obtained from the smoking of a sub-group of flue-cured varieties is reported here.

[1] U.S. Food and Drug Administration, 2012. Reporting Harmful and Potentially Harmful Constituents in Tobacco Products and Tobacco Smoke under Section 904(a)(3) of the Federal Food, Drug, and Cosmetic Act.
http://www.fda.gov/downloads/TobaccoProducts/GuidanceComplianceRegulatoryInformation/UCM297828.pdf