Avian Chemical Repellency: A Structure-Activity Approach and Implications

Abstract

Until recently, the discovery of avian sensory repellents has been empirical (Mason, Adams and Clark 1989). However, recent studies in our laboratory have shown that many avian repellents have similar perceptual and structural properties (Mason et al. 1989; Mason Clark and Shah 1991; Clark and Shah 1991; Clark, Shah and Mason 1991; Shah, Clark and Mason 1991). For example, methyl anthranilate, which has a grapy odor, is repellent to birds (Kare and Pick, 1960). Ortho-aminoacetophenone has an odor and structure similar to that of methyl anthranilate, differing only in the substitution of a ketone for an ester group (Mason et al. 1991). Behavioral tests of this aminoacetophenone isomer showed that it is at least an order of magnitude more repellent to birds than methyl anthranilate (Clark and Shah 1991). These similarities in structure and function prompted us to undertake a series of studies to elucidate a predictive model of chemical structure-activity (Clark and Shah 1991; Clark et al 1991; Shah et al. 1991). As a consequence of these studies we hypothesize a model where the following structural features appear to be important: (1) A phenyl ring with an electron donating or a basic group is central to repellency; (2) An electron withdrawing group in resonance with a basic group decreases the repellency (as well as the toxicity) of a substance. These effects are pronounced when the groups are ortho to one another; (3) The presence of an acidic group decreases repellency; (4) The presence of an H-bonded ring or a covalently bonded fused ring that possesses the required features (e.g., electron donating and withdrawing groups ortho to each other) can enhance repellency, but is not essential; (5) Steric hinderance can overpower the features described above, and can weaken the effectiveness of potentially aversive substances.