In article 3tn at eldborg.rhi.hi.is, thoreys at rhi.hi.is (Thor Eysteinsson) writes:
>In <2ubis9$d6m at aragorn.unibe.ch> larkum at optolab.unibe.ch (Matthew Larkum) writes:
>>>It strikes me that a study of membrane potential would be hard
>>because one would introduce a variable leak just by impaling a neuron
>>and the average would always be skewed in the positive direction, never
>>the more hyperpolarized direction.
>>This I don«t think is a problem if the methods are consistent. My experience
>with intracellular recordings of Vm (retinal neurons) with impalement is
>that if you get consistently good electrodes, and the set up is in shape,
>then the Vm is very consistent from cell to cell. This also would be a
>problem that should be independent of embryogenesis (or not?)
I agree. If the recordings are consistently at the same Vm then the worst
case is that one is introducing a constant leak with each impalement. My
experience is that one tends to "throw away" results from cells which look
"doubtful". I.e. a Vm of, say, -40 mV on entering the cell which then
proceeds to tend towards 0. Maybe I've come across a bad crop of
scientists, but you're the first person I heard of that never misses an
impalment.
> (By the way, I find I consistently
>>get a slightly better Vm by patching rather than impaling a neuron).
>>What do you mean by "better"? More stable?
I shouldn't have used "better". I should have said lower (more
negative), although there's no reason to call this better - I guess
this stems for calling impalements "bad" when then cell is obviously
injured by the process leading to a more positive membrane potential.
Being more stable is also a criterion.
>It should be pointed that
>if you are getting your cells for whole-cell clamping by enzymatic
>dissociation, then there is the danger that that treatment may change
>membrane properties (e.g. reveal "new" channels!). So, for this
>project I would prefer as intact a preparation as possible and
>of course then you would have to impale. Recordings that are unstable
>are suspect and should be discarded. A good impalement means you can
>hold the cell and record a stable Vm for at least an hour.
I find that recordings with patch electrodes give a much higher
value for input resistance which implies that the sharp electrodes
introduce some leak greater than the patch electrodes. Even very
stable recordings (longer than 2 hours) with microelectrodes seem
to have smaller values for input resistance.
>>Also, if the dendrites are slightly depolarized (or hyperpolarized)
>>relative to the soma due to constant synaptic activity, then the Vm at
>>the soma might be dependent on the dendritic structure of the neuron
>>which might also undergo changes during maturation (although, then
>>I, at least, would have expected Vm to increase not decrease).
>>But according to the original post (above) it DID increase with
>embryogenesis. But these points are still important.
I meant that I expected Vm would go from a more negative value to
a more positive value. The original poster measured it going in
the other direction.
Matthew.
larkum at optolab.unibe.ch
