There is not enough information here to answer your question. First, what do you mean by 'baseline'? Do you mean holding current? Membrane potential? If current, does "up' mean holding current increased? If voltage, is "up' hyperpolarization or depolarization? None of these factors: the amplifier, the software, or the mode (voltage clamp) are likely to be your problems. What is the composition (in mM) of your internal and external solution. Why are you using KMeSO4 and not gluconate as a counter ion? We have routinely used KMeSO4 as a sharp-microelectrode filling solution, but never for a whole-cell internal solution. A k-gluconate based solution is more standard for most whole-cell applications where you don't want to reverse the chloride gradient. What is happening to your input resistance during the change in 'baseline'? Is it increasing or decreasing? This will tell you whether you are looking at an increase or decrease in resting current.

Hi Minghua,

It sounds as though you are using potassium methyl sulfate (K-MeSO4) for your internal recording solution. I believe this is different from potassium methanesulfonate (K-MeSO3) and this is a common point of confusion with the similarity in name. Methanesulfonate is supposed to be an inert anion and we have had luck using a more common Cs-MeSO3 formulation for whole cell patch recordings. However, I fully agree with Daniel that K-gluconate is a more common choice for a wide range of patch recording applications and in my hands it is much easier to work with in terms of forming rapid seals onto neurons in slices. This is particularly notable when your neurons are not perfectly healthy. I think you may have more luck with stability of the recordings using K-gluconate internal formulation.

Thank you guys for the suggestions. I will definitely try the K-gluconate based internal solution

Dr. Madison, “baseline” does mean holding current. Like you said, KMeSO4 –based internal solution was used to keep the internal chloride concentration. I did not pay much attention on the changes in the input resistance. However, when the holding current goes positive (baseline goes up), I found that the resting membrane potential is hyperpolarized.

So the change in baseline is either an outward current turning on or an inward current turning off. It could just be a sealing of the leak around the electrode (leaks, being ion non-selective, are essentially inward currents at hyper polarizing hold potentials). To tell the difference, you need to know whether the change in holding current (and/or hyper polarization) is accompanied by a decrease or increase in the input resistance. In voltage clamp, the instantaneous current jump (when making step changes in voltage) will increase or decrease in amplitude respectively. Sorry if I'm telling you something obvious.

Thank you very much, Dr. Madison. I will monitor the input resistance as you suggested.

Am I missing something here? If you are holding the cell in voltage clamp at only -45 mV and the normal resting membrane potential is much more hyperpolarized (like -60 mV), shouldn't you expect to have a positive holding current? In my experience, when using a well selected internal solution and appropriate holding potential (intended to be close to the normal RMP of a healthy neuron of the type you are recording), it is generally a good thing to have a positive holding current. This should indicate that the neuron is more hyperpolarized that the selected holding potential level, which in most cases would mean the cell is particularly healthy. We normally worry mostly about high holding currents in the negative direction...which conversely would indicate the neuron is depolarized relative to the selected holding potential and may be unhealthy.

The junction potential of KMeSO4-based internal is around -15mV

It still seems that perhaps the cells you are patching have a more hyperpolarized RMP than you anticipate (which may be a good thing) or that the LJP is not accurately estimated for your internal. Another benefit of the K-gluconate is that the LJP should be closer to -10 mV instead of -15 mV. Do you see the same problem when using KCl internal and holding at -60mV? KCl should have no significant LJP.

Because I do not want to change internal Cl concentration and want to see GIRK current during whole-cell recording, I prefer to use K-based internal. BTW, I use K-gluconate internal today. I feel it is hard to make seal (i am doing RVM slice recording).

I think Jonanthan was just suggesting trying KCl in order to test where your RMP actually rests.

I am only suggesting to try KCl for the diagnostic purpose of checking holding current under conditions with no LJP and I understand this would not be your choice for your real experiments.

You should try to see if your colleagues around the lab/ Dept have any 'good' K-gluconate that you can try...something that is tested to work and relatively fresh. There should be no issue using K-gluconate even in the cell type you are trying to patch. Here are two papers I found by a quick search and both are patching RVM neurons in slice using K-gluconate:

Rostral Ventromedial Medulla Neurons That Project to the Spinal Cord Express Multiple Opioid Receptor Phenotypes

Marinelli et al (2002)

Cannabinoid receptor activation inhibits GABAergic neurotransmission in rostral ventromedial medulla neurons in vitro

Vaughan et al (1999)

Any difficulty in your initial testing is more likely to be related to (a) improper formulation, pH, osmolarity of the internal or (b) the slice prep itself rather than the use of gluconate as the anion.

I hope this helps.

Thank you guys for good suggestions.

We have problems with 'batches' of K-gluconate internal solution - once we get a good batch, everyone wants to use that one. When we have a 'bad' batch, the problem usually manifests as trouble in the getting of seals. The issue usually seems to be osmolarity and/or cleanliness (filtering). It puzzles me that osmolarity should be so different when the recipe is the same every time, but it seems necessary to actually measure it every time.

Could you please tell me the Cat# of K-gluconate? My is 129K0231(Sigma). I have same problem (trouble in getting of seals and the cleanliness) even I adjust pH and osmolarity and filter twice.

We usually use Fluka brand potassium D-gluconate #60245. Sometimes we use gluconic acid titrated against KOH.

Our recipe is:

in Fisher HPLC grade water (80 ml)

HEPES 0.953g

MgATP 0.11g

NaGTP 0.018g

K-gluconate 2.81g

MgCl2 (1M stock) 0.102 ml

pH to 7.2 with 1M KOH

bring to 100 ml with HPLC grade water

Dr. Madison, Thank you very much for the detailed information

Dr. Madison, can you please tell me the junction potential of this internal solution?

Thanks again

About 15 mV. I recommend getting the LJP calculator at: http://web.med.unsw.edu.au/phbsoft/LJP_Calculator.htm

or, there is one built into P-Clamp. It is also fairly easy to just measure it, although this has some problems because of issues with the reference electrode used.

Sometimes with slight damage you can get K-ATP currents, so if you had a blocker of that you could try it. We've found that that can occur with K-gluc K-MeSO4 etc internals. Especially with slices from medulla/midbrain.

Thank you, Dr. Vaughan. Yes, it is K current as BaCl2 blocked this outward currents. Does this mean the slice was not healthy or the whole-cell clamping was not good?