‘Dynamic’ QSAR For Semicarbazide-induced Mortality in Frog Embryos

Abstract

The conventional quantitative–structure–activity relationship (QSAR) provides only a single three‐dimensional (3D) model for a given two‐dimensional (2D) item in the modeling process. However, in complex reaction environments with solvents of different polarity, especially biological systems, the molecules can take the form of different conformers depending on the particular interaction. Therefore, chemical behavior, e.g. toxicity, may be considered the integral effect of a set of conformers rather than the property of a single 3D isomer. The ‘dynamic’ QSAR method is unique in that it provided for the calculation of a set of conformers for 2D representation of each chemical of the series under investigation. Moreover, these conformers can be selected interactively according to the hypothesized mechanism of toxic action. The acute lethality of 36 semicarbazides and thiosemicarbazides, evaluated using the Frog Embryo Teratogenesis Assay: Xenopus (FETAX), was modeled by using the ‘dynamic’ QSAR method. The assumed mode of action, osteolathyrism, was defined by the failure of connective tissue to polymerize properly due to interference with lysyl oxidase. Conformer screenings were based on parameter distribution according to the frontier orbital energies and volume polarizability, conditioning their reactivity and hydrophobicity, respectively. The best results were obtained by the selection of conformers providing prevailing values of electron acceptor properties. Moreover, the best two‐parameter QSARs encompassing all the evaluated compounds incorporate a geometric parameter, the geometric analog of the Wiener topological index, and the local electronic characteristics of the C==O or C==S group, superdelocalizabilities and charges.

Quantitative Structure–activity Relationships for Polychlorinated Hydroxybiphenyl Estrogen Receptor Binding Affinity: An Assessment of Conformer Flexibility

“…The applicability of the dynamic QSAR method was evaluated by comparing correlation samples of optimized and nonoptimized conformers. As in most of our previous studies [21,33,34], the results obtained with 1SCF calculations (i.e., nonoptimized conformers) were statistically equivalent to those obtained using the more time-consuming optimization procedures. Through the use of the dynamic QSAR method, the most polarizable nonoptimized and optimized conformers were found to be most closely associated with ER binding.…”

“…Based on the assumption that conformational flexibility is of critical importance in ligand-receptor interactions, we employed a dynamic QSAR method [21] to model the relative ER binding affinity of a series of PCHBs. As opposed to conventional QSAR methods, the approach used in this study does not assume a priori that 3-D molecular structures represented by single lowest-energy gas-phase conformers are appropriate to model physicochemical properties, toxicological end points, or the interactions of xenobiotics with receptors in a solvated environment [21,33,34]. Use of this method permits a systematic assessment of the conformational space associated with the compounds of interest, either in the context of testing specific hypotheses or as a means of exploring possible ligandreceptor interactions.…”

“…Recent dynamic QSAR studies on the toxicity of unsaturated alcohols [21], semicarbazides [33], and ␣-terthienyls [34], as well as the binding of PCBs to the Ah receptor (O.G. Mekenyan et al, unpublished manuscript), indicated that models based on nonoptimized conformer geometries were not qualitatively different from those based on optimized structures.…”

The Development of New Concepts for Assessing Reproductive Toxicity Applicable to Large Scale Toxicological Programmes

“…The uncertainty of the origin of data should be taken into account when integrating these models into testing strategies. For example several QSARs that claim to predict developmental toxicity of various chemical classes are based on Xenopus embryos [53][54][55] or on Hydra Attenuata [56][57][58]. The relevance of these species for predictive human health effects needs to be carefully considered since e.g.…”

The role of ligand flexibility in predicting biological activity: Structure–activity relationships for aryl hydrocarbon, estrogen, and androgen receptor binding affinity

“…The identification of proposed “active” conformers is based on an evaluation of the regression statistics associated with QSARs derived using the different conformation samples. Finally, recent dynamic QSAR studies on the toxicity of unsaturated alcohols [19], semicarbazides [21], and α‐terthienyls [22] indicated that models based on nonoptimized conformer geometries are not qualitatively different from those based on optimized structures when nonstrained chemicals are analyzed.…”