increase in electrode resistance

r norman rsn_ at _comcast.net
Thu Mar 25 14:53:54 EST 2004


On 25 Mar 2004 10:47:48 -0800, jonesmat at physiology.wisc.edu (Matt
Jones) wrote:


>> That being said, is a 3x increase in impedance really so much?
>> Differences in electrode impedance are really only significant if
>> they're outside the range of the amplifier impedance. 
>
>So the problem with having too high electrode impedance (or
>resistance) is that the series resistance, Rs, and the cell's
>resistance, Rm, form a voltage divider (actually the capacitance Cm
>also contributes to this divider in a freq-dependent manner, but for
>simplicity....):
>
>Electrode ------/\/\/\/ (Rs) /\/\/\/ ----- Cell -----/\/\/\/ (Rm)
>/\/\/\/---------GND
>
>The voltage that one wants to measure and/or control is that between
>Cell and Ground, but between the amplifier and the cell is a voltage
>drop across Rs.  In order for that Rs drop to be negligible relative
>to the Rm drop, Rs must be small relative to Rm. Otherwise there's a
>series resistance error (i.e., a systematic error in both measuring Vm
>and in controlling Vm by injecting current across Rs). This error is
>directly proportional to the amount of current being injected. This
>issue is more serious in voltage clamp experiments than in current
>clamp, but it is still there in both cases.
>
>To put the above in perspective, in whole cell patch clamp recording,
>one typically uses a big pipette tip (small resistance, on order 1-5
>MOhm), whereas the cell membrane is much much higher resistance
>(~50-1000 MOhm). In Elizabeth's situation, she's using fairly high
>resistance pipettes to begin with (90-120 MOhm), and I'm guessing that
>her cells have membrane resistance not too much higher than that
>(maybe 200-500 MOhm). If her pipette resistance climbs to >300 MOhm,
>this is going to introduce a potentially very large series resistance
>error.
>
>Again, this situation would make voltage clamp experiments completely
>uninterpretable, but might not be so huge a problem for current clamp
>experiments, though it's still probably not great.
>
>With regard to Richard Norman's comments:
>
>r norman <rsn_ at _comcast.net> wrote in message news:<ek3460l8bmtoa9e0scv3d3utulj1ubmnei at 4ax.com>...
>
>> Does the resistance stay high even after withdrawing from the cell?  I
>> always thought it was "crud" in the electrode tip -- never thinking to
>> call it by a more elegant term -- endoplasmic resistance or nucleus.
>
>Oh yeah, sometimes it's just crud in the electrode tip. But in the
>present case it seems to be crud that's coming from the cell (the
>resistance is lower and stable in the bath). So if it were a whole
>cell recording, I'd try two things to ensure low series resistance: 1)
>if the Rs is high immediately upon breaking in, I'd assume that the
>patch of membrane had not been completely ruptured and was clogging
>the tip. Might try to improve it by additional suction, and b) if Rs
>was initially low, but gradually got higher, I'd assume it might be
>the patch falling back into the tip (try sucking) or that it was cell
>contents (ER or nucleus or something). Might try to improve it by
>blowing (this almost never works, usually lose the cell) or by
>withdrawing slightly (this also hardly ever works once the problem has
>manifested itself, but sometimes it does work if you do it immediately
>after breaking in - that is, if I do it routinely, I seem to have less
>problems with Rs later in the expt).
>
>A lot of this is voodoo, as I'm sure you know.
>
>While i totally agree with Richard Vickery that filtering pipette
>solutions is important in general, and that maintaining good
>chloriding on the wire and ground are important, my gut tells me that
>neither of those are responsible for the present problem (unfiltered
>solutions would probably cause clogs while still in the bath instead
>of witing to be in the cell, and likewise with increased resistance
>due to poor chloriding).
>
>Ahh. It warms the cockles of my heart to see discussion of
>neuroscience in bionet.neuroscience, as opposed to the typical fare
>for this group.
>
>Thanks Elizabeth!
>

I agree with Matt -- this type of question is really what the news
group was intended for.

My impression is that Elizabeth is doing classical microelectrode
recording, small diameter sharp electrodes impaling whole cells, not
patch clamp.  The opening at the tip might be in the tens of
nanometers size range.  There is no suction or blowing involved, just
advancing and withdrawing the electrode. The whole cell resistance
shouldn't be anywhere near hundreds of megohms, something you would
expect for a square micrometer of membrane area.

Things do stick to glass -- supposedly the cell membrane adhering to
the electrode is how the cell accepts the electrode penetration in the
first place.  My reference to crud really meant crud from outside the
electrode.  Matt is right, crud from inside would cause problems just
sitting in the bath. Is there something about the electrode
preparation or treatment that perhaps leaves a high charge density on
the surface that would predispose it to more clogging problems?

However, Elizabeth really needs practical advice, not speculation.
So, Elizabeth, is this a new problem where it has always worked in the
past or is it a problem that you find on trying to develop a new
technique?  What if you use lower resistance electrode -- down in the
20 M to 50 M range.  Of course, you will not get as good impalements,
but are they subject to the same problem?  What happens if you try
your electrodes in a different cell type?  Or if you use larger
electrodes on a different cell type?  What about people working in
your lab or in nearby labs?  What do they find?  If you can't work
with the system you have, one resolution is to start with easier and
simpler cells and electrodes to make sure you have the technique down
just right.  Then work on getting your electrodes down to the needed
size (up to the needed resistance) even with the "simpler" or "easier"
cells.  Only then go back to the real prep.








More information about the Neur-sci mailing list