Gain Paradox

Abstract

Data obtained early on suggested that the chemotactic response is proportional to the change in receptor occupancy, with that occupancy characterized by a fixed dissociation constant, K _d, the concentration of ligand at which the probability of receptor occupancy is 1/2 (Berg and Tedesco, 1975; Mesibov et al., 1973). Then it became evident that the dissociation constant increases (i.e., cells become less sensitive) at higher concentrations of ligand, as receptors are methylated (Borkovich et al., 1992; Bornhorst and Falke, 2000; Dunten and Koshland 1991; Li and Weis, 2000). However, even at these higher concentrations (e.g., for the nonmetabolizable aspartate analog α-methylaspartate at an ambient concentration of 0.16 mM) the gain is prodigious: a step increase in concentration from 0.16 to 0.16 + 0.0027 mM (a change of about 1.7%) transiently increases the probability that the motor spins counterclockwise (CCW) by 0.23 (Segall et al., 1986). Computer simulations of the chemotaxis system (e.g., Bray et al., 1993; reviewed by Bray, 2002) fail to predict the necessary gain. Two recent findings appear to resolve the paradox. First, there is an amplification step at the beginning of the signaling pathway: the fractional change in kinase activity is some 35 times larger than the fractional change in receptor occupancy (Sourjik and Berg, 2002a).