I am doing voltage-clamp experiments in CA1 pyramidal cells (mEPSCs; AMPA-NMDA ratio; LTP; etc) from acute hippocampal slices in mice. I want to make sure I am only recording from excitatory pyramidal cells, but since I am doing voltage-clamp and not current-clamp I cannot rely on the usual test of action potential firing after current injection.

I have a lengthy explanation/review of approaches below, but my main question is: In voltage-clamp in CA1 specifically, is there a reliable way to distinguish excitatory pyramidal cells from interneurons by electrophysiological properties?

--------

I have come across each of the following methods, but each seems to have a downside or isn't foolproof---except for post-hoc staining (#5). This is of course the most time consuming, but is what I am leaning towards.

1) Resting membrane potential (I=0 mode immediately after break-through)

- Interneurons should have a more positive resting membrane potential. However, inclusion of Cesium in the internal solution for voltage-clamp will affect potassium channels and therefore the RMP. I typically check RMP by switching to zero-current-injection (I=0) shortly after breakthrough, but how quickly will my internal affect the RMP? At least one paper (Article How cesium dialysis affects the passive properties of pyrami...

2) Membrane time constant

- In recordings of cortical neurons, differences in the membrane time constant can be used to distinguish classes of neurons (https://www.researchgate.net/post/What_are_the_electrophysiological_differences_between_excitatory_and_inhibitory_neurons)

- However, in CA1, membrane time constant does not appear to be a good indicator, at least according to values from NeuroElectro.org. The Allen Brain Institute does not have these values for CA1 neurons.

3) Location (central location within stratum pyramidale)

- I have heard that interneurons are primarily located on the borders (or further) from the center of the stratum pyramidale layer. However, IHC of interneurons in several papers makes it look like they are well represented throughout (e.g. Article Distribution of interneurons in the CA2 region of the rat hippocampus

- Additionally, although I can do recordings on transverse and coronal sections, the "magic cut" (Bischofberger et al., 2006) spreads out the SP a bit and greatly increases my efficiency since cells are much easier to access. Therefore, I'm not sure that I can apply Location to "magic cut" sections.

4) Morphology (nice pyramidal shape)

- Of course, the first thing I look for is a nice pyramidal shape. But now I'm not sure how reliable this is, as it seems that interneurons can also assume a pyramidal shape (e.g. Fig1A https://www.jneurosci.org/content/23/3/826). I worry that this is more common than I originally thought.

5) Imaging (biocytin-fill and post-hoc IHC)

- This seems to be the most reliable but of course is the most time-consuming. I plan to do these, but I also already have many recordings for which I did not include biocytin in my internal.

Are there any tests that I am missing, or any downsides I am exaggerating?

One other consideration/possible test is any differences in response to stimulation of the Schaffer collaterals. It would be ideal if the resulting EPSCs were different, but I haven't come across that in the literature.

Thanks!