Quantitative Sar Qsar analysis in the safety assessment of constituents

If the available genetic toxicity data, or the qualitative SAR analysis, indicate potential concern for the carcinogenicity of a constituent, QSAR may be a feasible approach to safety assessment by estimating a worst-case LCR (Bailey et al., 2005). Analog identification is an empirical case-by-case process that uses overlapping approaches to identify structurally and physicochemically related groups of analogs. Historically, analogs have been identified by experience and searching of the toxicology literature. More recently, searches are greatly facilitated by the availability of electronic structure-searchable databases. Publicly available examples are the National Library of Medicine's (NLM) ChemlDPlus (NLM, 2005) and EPA's Distributed Structure-Searchable Toxicity (DSSTox) Public Database Network (EPA, 2005b). FDA also uses internally prepared structure-searchable version of the CPDB and other internal databases.

For transparency and reproducibility, the rules and assumptions used during analog selection should be clearly defined, both for manual analog searches and when substructure/similarity software is used. Usually, decisions to include or omit potential analogs are specific to the context of the relative wealth of candidates and their data.

Analog identification requires careful evaluation by the analyst. Not only is it important to identify an appropriate scaffold (e.g., a common substructure such as a phenanthrene ring system) that contains the same or similar functional groups (SAs) (e.g., nitroso or nitrosamines), but it is also important to consider the positioning of the functional groups on the scaffold and how this might alter toxicity. Examples of positioning differences include the steric hinderance of a SA of the constituent from metabolic activation, or electronic effects (such as conjugation) due to a functional group located within the molecule.

Relatively small differences in substructural features may have a dramatic effect on the metabolism of the constituent or the analog, or even alter the reactivity of the molecules with DNA (genotoxic versus nongenotoxic), thus changing the detoxification pathway or the molecule's mode of action. For example, metabolic activation of N-nitrosamines to potential carcinogenic metabolites is thought to proceed through hydroxylation at the carbon alpha to the nitroso group. Hydroxylation is followed by the release of the hydroxyalkyl moiety as an aldehyde and the generation of a primary nitrosamine (Klaassen, 2001). The primary amine ultimately forms a carbonium ion that can interact with DNA. Therefore, the use of nitrosamines containing aliphatic chains (see Fig. 7.1) may not be suitable analogs for N-nitrosodiphenylamine (see Fig. 7.2) since Fig. 7.2 does not contain a carbon alpha to the nitroso group that can undergo hydroxylation; thus altering the metabolic detoxification pathway for N-nitrosodiphenylamine compared to compounds in Fig. 7.1. One should also consider the physical properties of the molecule in question and the potential set of analogs. For example, large differences in molecular n = 0 - 4

Fig. 7.1 Structures of aliphatic nitrosamines 1-5.

Fig. 7.1 Structures of aliphatic nitrosamines 1-5.

Fig. 7.2 N-nitrosodiphenylamine 6.

Fig. 7.2 N-nitrosodiphenylamine 6.

weight, boiling point, pKa, and water/octanol solubility may affect toxicokinetic and toxicodynamic parameters.

0 0

Post a comment