Morphology. The dominant component… , neurological diseases

• volume 2 no 12 • december 1999

1107

er and calcium-activated types of potassium chan-

nels18.

However,

4-aminopyridine

(4-AP;

5–10 mM), an antagonist of transient A-type potas-

sium channels (IA) caused a dramatic recovery of

spiking (0.66 ± 0.17 spikes per trial; n = 7; Fig. 1b).

The rescue of granule cell excitation by 4-AP implies

that IA severely limits the effectiveness of AMPA

receptor-mediated synaptic inputs. 4-AP can

increase transmitter release by broadening the presy-

naptic action potential19,20. Therefore, we tested

whether the effect of 4-AP on granule cell excita-

tion could be explained by enhanced glutamate

release from mitral cells. To eliminate effects of 4-

AP on the electrical responsiveness of mitral cells, we used a sat-

urating stimulus intensity. 4-AP did broaden the mitral cell action

potential (Fig. 1c, inset) but did not increase the granule cell

EPSC (3 ± 7% increase in the EPSC charge; n = 7) nor EPSPs at

holding potentials below threshold for IA (see Fig. 3c). Thus, the

enhancement of granule cell firing by 4-AP reflects increased

excitability of the postsynaptic granule cell membrane rather than

a presynaptic effect on mitral cells.

Granule cells had a prominent transient IA blocked by mil-

limolar concentrations of 4-AP (Fig. 2a; 51 ± 4% block, 1 mM

4-AP, +12 mV; n = 5). IA was unaffected by low concentrations of

4-AP (100 µM; n = 4) or dendrotoxin (1 µM; n = 3) and was only

slightly reduced by TEA (28 ± 10% block, 20 mM; n = 4). IA mea-

sured in nucleated outside-out patches in the presence of TEA

(10–20 mM; Fig. 2b) had a threshold activation voltage of

–44 ± 2 mV (n = and a midpoint voltage for steady-state inac-

tivation of –66 ± 4 mV (n = 5). The activation threshold for IA

was near the threshold for granule cell spiking (–47 ± 2 mV;

n = 9), indicating that IA is capable of affecting spike initiation. IA

inactivated with a voltage-independent decay time constant

(24 ± 5 ms at –38 mV; n = 5), and recovery from inactivation (at

–108 mV) was approximated by a single exponential with

τ =62± 10 milliseconds (n = 3). Granule cells also had a delayed,

a

b

c

Fig. 1. Blockade of IA with 4-AP enhances granule cell

excitation by AMPA receptor-mediated synaptic inputs.

(a) Morphology. The dominant component of inhibition

in the olfactory bulb occurs at dendrodendritic synapses

between the secondary dendrites of mitral cells (M) and

the dendrites of granule cells (G). The cartoon on the

right shows that glutamate released from the shaft of the

mitral cell dendrite elicits excitation (+) of the granule

spine, which in turn leads to recurrent GABAergic inhibi-

tion (–) of the activated mitral cell as well as lateral inhibi-

tion of other mitral cells. (b) Glomerular stimulation (as

shown in a) elicited spike firing in granule cells that was

greatly reduced by D, L-AP5 (50 µM), but then recovered

with subsequent addition of 4-AP (6 mM). Four glomeru-

lar stimulation-evoked responses are shown from one

cell for each condition on the left. The summary plot on

the right reflects pooled data from responses elicited by

glomerular stimulation (three cells) or by focal mitral cell

stimulation (four cells). (c) The granule cell EPSC, as

measured by its total charge (see Methods), was not

increased in size by 4-AP, indicating that 4-AP did not

enhance glutamate release from the mitral cell. The inset

shows that 4-AP did broaden the mitral cell presynaptic

action potential, accounting for the somewhat longer

onset delay and duration of the EPSC in 4-AP. The sum-

mary on the right reflects the same experiments as in (b).

articles

non-inactivating potassium current IK blocked by TEA

(10–20 mM; Fig. 2a) but not 4-AP (6 ± 4% reduction in

5 mM 4-AP; n = 4). However, IK had a threshold for activation

(–33 ± 5 mV; n = 5) that was above the spike initiation voltage of

granule cells, consistent with the lack of effect of TEA on spiking.

IA specifically attenuates brief depolarizations

The attenuation of AMPA receptor-mediated synaptic inputs by

IA might be explained by the short duration of the AMPA recep-

tor-mediated depolarization. Indeed, we found that granule cell

depolarizations in response to short (2–4-ms) somatic current

injections that mimicked an AMPA receptor-mediated EPSP

(Fig. 3a) were significantly enhanced when IA was blocked by 4-

AP (by 7.2 ± 2.6 mV; n = 5). As expected, the effect of 4-AP

depended on whether the injected current elicited a depolariza-

tion that reached the activation threshold for IA. Longer somat-

ic current injections (800 ms, +30 pA; Fig. 3b) elicited spiking

in granule cells in the absence or presence of 4-AP, but 4-AP

reduced the lag to spiking by 47 ± 15 milliseconds (n = 6). Close

examination of the voltage traces showed that 4-AP enhanced

the early portion of the responses, whereas all the responses con-

verged to the same voltage at spike initiation, consistent with

transient attenuation by IA.

© 1999 . •

Connected themes: neuroscience journal, neuroscience journal articles, neuroscience journal ranking, Morphology. The dominant component… , neurological diseases, neuroscience journals, neuroscience journals list, neuroscience kids.