No abstract

The GABA(A) receptor is an important target for a variety of general anesthetics (Franks and Lieb, 1994) and for benzodiazepines such as diazepam. Specific point mutations in the GABA(A) receptor selectively abolish regulation by benzodiazepines (Rudolph et al., 1999; McKernan et al., 2000) and by anesthetic ethers (Mihic et al., 1997; Krasowski et al., 1998; Koltchine et al., 1999), suggesting the existence of discrete binding sites on the GABA(A) receptor for these drugs. Using anesthetics of different molecular size (isoflurane > halothane > chloroform) together with complementary mutagenesis of specific amino acid side chains, we estimate the volume of a proposed anesthetic binding site as between 250 and 370 A(3). The results of the "cutoff" analysis suggest a common site of action for the anesthetics isoflurane, halothane, and chloroform on the GABA(A) receptor. Moreover, the data support a crucial role for Leu232, Ser270, and Ala291 in the alpha subunit in defining the boundaries of an amphipathic cavity, which can accommodate a variety of small general anesthetic molecules.

SUMMARY Ribbon synapses mediate continuous release in neurons that have graded voltage responses. While mammalian retinas can signal visual flicker at 80-100 Hz, the time constant, τ, for refilling of a depleted vesicle release pool at cone photoreceptor ribbons is 0.7–1.1 s. Due to this prolonged depression, the mechanism for encoding high temporal frequencies is unclear. To determine the mechanism of high frequency signaling, we focused on an Off cone bipolar cell type in the ground squirrel, the cb2, whose transient postsynaptic responses recovered following presynaptic depletion with a τ of ~0.1 s, or 7-10-fold faster than the τ for presynaptic pool refilling. The difference in recovery time course is caused by AMPA receptor saturation, where partial refilling of the presynaptic pool is sufficient for a full postsynaptic response. By limiting the dynamic range of the synapse, receptor saturation counteracts ribbon depression to produce rapid recovery and facilitate high frequency signaling.

In daylight vision, parallel processing starts at the cone synapse. Cone signals flow to On and Off bipolar cells, which are further divided into types according to morphology, immunocytochemistry, and function. The axons of the bipolar cell types stratify at different levels in the inner plexiform layer (IPL), and can interact with costratifying amacrine and ganglion cells. These interactions endow the ganglion cell types with unique functional properties. The wiring that underlies the interactions between bipolar, amacrine, and ganglion cells is poorly understood. It may be easier to elucidate this wiring if organizational rules can be established. We identify 13 types of cone bipolar cells in the ground squirrel, 11 of which contact contiguous cones with the possible exception of short-wavelength sensitive cones. Cells were identified by antibody labeling, tracer filling, and Golgi-like filling following transduction with an adeno-associated virus encoding for GFP. The 11 bipolar cell types displayed two organizational patterns. In the first pattern, 8-10 of the 11 types came in pairs with partially overlapping axonal stratification. Pairs shared morphological, immunocytochemical, and functional properties. The existence of similar pairs is a new motif that may have implications for how signals first diverge from a cone to bipolar cells, and then re-converge onto a costratifying ganglion cell. The second pattern is a mirror symmetric organization about the middle of the IPL involving at least 7 bipolar cell types. This anatomical symmetry may be associated with a functional symmetry in On and Off ganglion cell responses.

No abstract