How to isolate GABAa currents?

jonesmat jonesmat at physiology.wisc.edu
Tue Apr 5 11:55:56 EST 2005


Hi,

Thomas Gabriel wrote:
> (since I am using a Axoclamp 2B amplifier i prefer
> current-clamp mode). I expected the remaining PSPs to be gabaergic
and
> thought that i should be able to block them with bicuculine.

I strongly suggest that you use voltage clamp for this. First, you are
asking about "GABAa currents", and currents can be directly measured in
voltage clamp, but can only be indirectly inferred in current clamp.

Second, you didn't specify whether you're doing whole-cell or sharp
electrode recording, or what your chloride gradient is, but it could be
that you are sitting too close to ECl to actually get any voltage
change, even if the GABA conductance is large. If you were sitting
exactly at ECl, you would see no PSP even if the synaptic conductances
were humongous.

The ideal experiment would be to use whole-cell voltage clamp, with
symmetrical chloride internally and externally (CsCl internal solution
would also help reduce noise through leak channels, but KCl would be
fine too), hold the cell around -60 mV, and use DNQX/APV. Under these
conditions you could normally record pretty small currents (e.g.,
mIPSCs as smal as 5 pA).

I think the worst scenario would be to use sharp electrodes filled with
1 M KCl and do current clamp, because then you have really no idea
where ECl is, or whether you're near it or not. But if you have to do
it this way, I would again use DNQX/APV, and try to measure PSPs over a
very large range of Vm, say from -120 right up to 0 mV if possible
(though the recording would get pretty wobbly up there).

If all you really want to do is to make sure your bicuculline works,
then you could also try pressure applying GABA directly to the cell
body (or bath applying it, but that will not work as well at all
because GABA receptors desensitize during long exposures). This should
activate pretty huge conductances (e.g., tens of nS in hippocampal
neurons), which you could detect even if you're already at ECl by
injecting periodic current pulses away from rest (i.e., if a
conductance opens, the voltage response to the pulses will get
smaller). Then bath apply bicuculline (10-50 uM) and see if the pulse
responses go back to their original size.

Good luck,

Matt




More information about the Neur-sci mailing list