Dual-component synaptic responses were evoked… , neurological procedures

c

d

Fig. 3. IA attenuates short but not prolonged current inputs. (a) Short-duration somatic current injections into a granule cell (2 ms, 150 to 300 pA)

elicited depolarizations that were greatly enhanced when IA was blocked by 4-AP. The magnitude of the 4-AP-effect (‘Difference;) was as much as 10 mV

and depended on whether the depolarization reached IA threshold. (b) Granule cell depolarizations in response to prolonged somatic current injections

(800 ms, +30 pA) were transiently enhanced by 4-AP (arrow), but control responses eventually converged to the same amplitude as responses in 4-AP

at spike threshold (dashed line). The difference in the lag to spiking in this experiment, plotted on the right as the cumulative probability of spiking, was

between 50–100 ms. (c) 4-AP also enhanced short-duration AMPA receptor (AMPAR)-mediated EPSPs in granule cells, measured in the presence of D, L-

AP5 (50 µM). 4-AP enhanced EPSPs only when the depolarization reached IA threshold (compare holding potentials, Vh, of –98 and –68 mV). EPSPs

were evoked by focal mitral stimulation. (d) Prolonged AMPAR/NMDAR-mediated synaptic responses in granule cells displayed an upward creep (inset,

short arrow in inset), leading to spiking after a 4-AP-sensitive lag. The lag to spiking, used for construction of the histogram, was taken as the time

between the stimulus and the peak of the evoked action potentials. Dual-component synaptic responses were evoked with glomerular stimulation.

Functions of IA for reciprocal and lateral inhibition

timing of granule cell spiking and the observed lateral IPSC

Glomerular stimulation-evoked dendrodendritic IPSCs have both

(Fig. 6b) showed a close match in the time to peak for the pre-

a reciprocal component (as shown in Fig. 1a) and a lateral compo-

dicted and observed IPSCs, consistent with a 4-AP-induced

nent reflecting inhibition between mitral cells through synaptical-

change in triggering of the lateral IPSC. The observed IPSCs were

ly ‘shared; granule cells. The mechanisms of generation of reciprocal

longer in duration than predicted, perhaps reflecting GABA

and lateral inhibition may differ. For example, the spread of volt-

release driven by a spike-induced residual calcium signal28.

age in granule cells required for lateral inhibition is expected to be

The reciprocal IPSC was isolated by direct stimulation

significantly augmented by action potentials, whereas reciprocal

(2–4 ms voltage steps to 0 mV) of a mitral cell (Fig. 6c). We

inhibition can be recorded in tetrodotoxin10, and thus requires only

first performed experiments in reduced extracellular magne-

local signaling within dendritic spines. Similarly, reciprocal and lat-

sium to enhance the amplitude of the reciprocal IPSC12. In

eral inhibition could be differentially regulated by IA.

100 µM Mg2+, 4-AP markedly changed the kinetics of the rec-

To isolate the lateral IPSC, we modified our standard glomeru-

iprocal IPSC from a slow monophasic decay (τ = 478 ± 28 ms;

lar stimulation protocol by including the sodium channel antag-

n = to a biphasic decay with fast and slow components

onist QX–314 (15 mM) in the patch pipet to block reciprocal

(τf =59 ± 11 ms, τs = 528 ± 49 ms, Af/As = 2.5 ± 0.6). 4-AP

inhibition (Fig. 6a). In 8 mitral cells, 4-AP did not affect the mag-

increased the total charge associated with the reciprocal IPSC

nitude of the lateral IPSC (Q4-AP/QControl =1.09 ± 0.15), but

(Q4-AP/QControl = 2.1 ± 0.3) by introducing a fast component, as

reduced its decay time constant by 20 ± 6% and time to peak

well as by causing a modest 1.4 ± 0.2-fold increase in the slow

from 60 ± 8 to 38 ± 4 ms (Fig. 6b). The kinetic effects of 4-AP

component. A large rapid component was also induced by

on the lateral IPSC were presumably due to the shortened laten-

4-AP in IPSCs measured in our standard extracellular solution

cy to synaptically evoked spiking in granule cells (Fig. 3d). Indeed,

containing 1 mM Mg2+ (τ =32 ± 7 ms; n = 8; Fig. 6d). Quan-

a direct comparison between the predicted IPSC based on the

titative analysis of the slow component of these reciprocal

• volume 2 no 12 • december 1999

1109

a

© 1999 . •

articles

b

Connected themes: neuroscience law, neuroscience leadership, neuroscience learning, Dual-component synaptic responses were evoked… , neurological procedures, neuroscience lecture, neuroscience lectures, neuroscience letters.