Dear Giampaolo:

Cell attached recordings are easy and very straightforward. Hippocampal CA1 neurons in acute brain slices are typically quiescent or discharge at very rates. So if you do not see any activity, this could be part of the problem. Try to increase the temperature in the bath (+34 C) and see if that makes difference. Or lower the extracellular Ca2+. Also, you may try to use low concentrations of cadmium or cobalt to get your recordings optimized. All these should make your neurons fire.

Cell attached firing you can record in all modes: VC, CC or Io. Make sure that you have a low impedance patch electrode; compensate its capacitance well; set the sampling rate quite high, to make sure that you are not missing spikes >5kHz. To get your techniques right, I would suggest you to go to cerebellar slices and try to record the activity of Purkinje cells. The vast majority of them fire spontaneously in acute slices like crazy. If you can record there, this means that your settings are right and all that needs to be done is to get recording conditions optimized for hippocampal slices.

Good luck - 

Saak V. Ovsepian, Ph.D., M.D. 

Hi,

You could try using the techniques mentioned in this paper.

Cell-attached voltage-clamp and current-clamp recording and stimulation techniques in brain slices.  Katherine L. Perkins

http://www.sciencedirect.com/science/article/pii/S0165027006001221

Regards,

Kevin

Noone seems to have mentioned that if you voltage clamp you are stopping all spontaneous activity - that's the whole purpose of the technique: there are no spikes in voltage clamp mode other than those that are generated in regions of the neuron that are too distant for the current passing to control.  Thus, if you want to record spike activity the first thing you have to learn is current clamp.  Your amplifier will do this with just a flip of a switch - and you should expect to then see the voltage settle to the resting potential (determined both by the membrane permeability AND the solution you put in your pipette) and also spontaneous fluctuations in the membrane potential - including any due to synaptic inputs or currents from dendrites.

He was asking about cell-attached mode, not whole cell. When in cell-attached mode and recording spikes, you should put the amplifier in VC and set the potential to the potential that gives you zero pA baseline.  That will be zero mV if the seal is loose.  It will be more like -50 mV if the seal is tight. You are NOT voltage-clamping the neuron when you are in cell-attached mode even though the amplifier is in VC. If you think that you are not getting any spikes because the cell is not firing, try some picrotoxin or bicuculline or gabazine in the bath with 5 mM extracellular potassium in the ACSF.  That will give you interictal activity in pyramidal cells if your hippocampal slice is intact (includes CA3 region). And as said above, check out my methods paper.

Good luck,

Katherine

Katherine's point about VC and firing of neurons is correct. In cell attached mode, setting the amplifier in VC does not really clamp the membrane of neuron and as such does not suppress its firing. What is actually clamped in CA mode (if amplifier is set up in VC mode) is a small patch of the membrane that is at the tip of the pipette. Typically this does not affect the bio-electrical activity of the neuron, and because of this, CA recording configuration is the least invasive of all. 

As regards to setting of the potential, it makes perfect sens to set up the amplifier to the potential that gives you a zero current after loose seal formation. This however works fine if you have a very stable seal, which typically is not the case with CA patches. Is not the CA seal called 'loose'? The loose seal by definition is unstable and often fluctuates (becomes better or deteriorates) during recordings and over time the baseline current may change. Because of this, the current that amplifier (if set in VC and at certain voltage) passes will also change, which may interfere with measurements as well as with the seal. 

Thus, ideally you should set the amplifier in CC mode and let the cell do whatever it does. This will not effect your readouts if you use a threshold based software for counting your spikes. But as said earlier, the choice of the mode is flexible with cell attached experiments...     

As discussed in my methods paper referred to above, the signal-to-noise for recording spiking activity in cell-attached mode is better in VC than in CC.  The capacitance of the patch of membrane acts as a low-pass filter in CC mode, thus fast events like spikes are filtered and appear very small. Fig. 12 of that paper shows a comparison of cell-attached CC and VC for the same cell. Someone at Society for Neuroscience meeting once commented to me that when recording cell-attached in CC, his spikes got smaller and smaller in apparent amplitude as the seal got tighter and tighter. 

I am strongly supporting Elise Stanley suggestion, you should set a current clamp settings in amplifier as well as Pclamp (if you are using axopatch product). Spontaneous spikes you can see only if you inject current. 

I hope it will useful for you.

I record from tonically active cholinergic interneurons.  I use a loose patch technique and it's rough.  Its a lot of preference.  Hold the cell/pipette at 0 mV and you'll see lots of pA activity.  Hold the cell at 0 pA and you'll see perfectly shaped action potentials with relatively no noise.  I personally prefer current clamping, I get a much better visual.  Voltage clamp give you the control to keep that pipette sitting exactly where it needs to be though; 0 mV