The classical tests would be to

1) make fresh pipette solution (with fresh ATP)

2) make sure no wire is creating any mechanical tension on your table.

good luck !

what about the osmolarity of your solutions? Could be that the reason? By the way, does the cell morphology changes? Did you try the perforated patch?

Hello. One of the biggest problems in maintaining whole cell recordings for a long time have to do with the size of the recording pipette: small tips allow you to get an easy seal, but the probability of resealing is very high. On the other hand, pipette tips that are too large will make it harder to get a good seal and once you breakthrough may cause a large change in holding current because the seal comes off a little. For cells that are about 15-20 micon in diameter I like tips between 3 and 5 megaOhms. Mitral cells are larger, correct? So you could try tips in the 2-4 mega Ohms range.

Another issue is the difference in osmolarity between your extracellular and intracellular solutions. In neocortex I tend to like a fair difference (320 mOsm extracellular; 295-300 mOsm internal) because this allows membranes to smooth out under the pipette, though not so much that the membrane becomes fragile.

Then of course pH and freshness of the intracellular solution matters very much. I am sure you have ATP and GTP, do you add phosphocreatine as well? It helps quite a bit for long recordings.

We are routinely doing whole cell recordings lasting for over an hour.  One of the common problems with patch stability during long recordings is drift of the pipette.  This issue is quite variable between different micromanipulators.  However, the "symptoms" are similar to what you described (sometimes cells last for long time while now and then you see leak developing within 5-15 mins).  Try to watch closely position of the pipette relative to the cell.  If you see a slow drift, try to adjust (very carefully!) position of your pipette and see if the recording quality improves.  If there is no drift, then you have another problem.  This trick works well for relatively round cells and relatively big currents.  Other comments are also valid.

Thanks for all the quick and detailed answers!

I now believe that the pipette solution might be at the root of the problem, because I do see some cell morphology changes (though it happens to everyone in my lab). I have some follow up questions about that:

- Osmolarity: Usually the lower osmolarity of the internal solution relative to the external is said to be important for the facilitation of gigaseal creation in the first place. I have absolutely no problem with that - is it possible that the osmolarity of my solution is too low, causing cell deformation over time, or should I make it even lower? Is there a reason to believe the osmolarity changed over time when to internal solution was sitting in the freezer?

- ATP, GTP, phosphocreatine: These are the major components that seem sensitive to degradation over time. I had them all in my solution, but they might be broken down by now. Do the lack of any of them might actually cause such an abrupt change in recording conditions? I always assumed that it would cause some slow deterioration or change in the activity of the cell, rather than some sudden breakdown like I get.

Hi Asaph,

what sort of cell are you investigating? You might be able to "simplify" your solutions because ATP is not necessary for many cell lines and channels, depending on your research project of course..

Also, how does your protocol approx. look like? If you pulse too often, your cell will be in a bad shape sooner. Your observations about the holding voltage are totally right. For most cells it represents significant stress to have an unnatural hyperpolarized holding voltage. See if you can make the voltage more positive without introducing considerable artefacts, f.i. due to steady-state inactivation regarding voltage-gated channels.

Also crucial: I hope that you have a constant perfusion system set up! If you measure for more than 20 minutes, the osmolarity of your bath solution can change significantly due to evaporation!

Best regards, Martin

Hi Asaph,

the full exchange of your internal with the cytoplasm occurs over few seconds after breakthrough, so a bad internal can change the condition of the recording quite rapidly. ATP, GTP and phsphocreatine are needed to maintain long term recordings, especially if you are interested in any sort of plastic change, but they do not necessarily affect the condition of the whole cell early on, unless something is very far off. To keep them from breaking down too quickly keep your internal solution cold.

Early on, pH and osmolality matter more. Also pipette shape, size, stability of the seal and possible drifts in the manipulator matter quite a bit.

Hi Asaph,

Based on your information I would gather that you are holding your cells at negative membrane potentials for long periods of time. These are only anecdotal observations of mine over the years, but I always found that holding a neuron with negative current caused inevitable problems with access resistance. I reasoned that it had to do with plugging up the pipette similar to forming a seal with negative current. It is likely pulling the nucleus and other cell constituents into the pipette and sealing against it. I also always found a persistent increase in holding current as well. What you may want to do is in between data collection, return the membrane potential to a value where a positive current is required to maintain Vm. Also, I don't know what you are using as your intracellular solution, but I would stay away from gluconate and use methylsulfate as your counter anion. Gluconate complexes with calcium (see Woehler et al., 2014 J. Physiol. in press) and as a result can cause changes in access resistance. I always had more luck with methylsulfate for long term recordings. Good luck.

All good suggestions as far as keeping your ICF at slightly lower osmolarity than your ecf and making sure you keep your ICF aliquot on ice.  One thing not mentioned in the suggestions above is the mechanical force you are putting on the cell when you seal and break in.  You can try and drop the pressure you are using to seal on and/or break in.  That pressure doesn't dissipate immediately and is compounded by pipette shape.  The net result is that you may be applying a constant force that will eventually disrupt the seal.  Sometimes you'll see events of seal breaks and then resealing, then seal breaks again.  These are usually indications that you've got membrane "creeping" up the  pipette.

Good luck!

Interestingly, in our lab, we have noticed that while ECS with higher osmolarity helps in attaining a giga-seal, cells tend to loose it's normal shape slowly but gradually. However, when we increased the ICS osmolarity slightly higher (~5 mOsm), that helped in retaining the patch for longer. There is probably an optimal level between attaining and retaining the patches, that has to be found for individual system by trial and error. 

Hi Asaph,

All suggestion above are v.good. Anyway it would be useful to know what you do have in your internal solution to be able to pinpoint the problem. Because you can record for long time from a cell it is quite obvious that the cell is not an issue. Problem is with your recording settings.

Thanks,

Ewa

Thanks again for all the detailed answers, I now have a lot of directions to try.

In case someone might find some clues in it, this is the composition of my internal solution (in mM):

K-gluconate (130), Na-gluconate (10), HEPES (10), phosphocreatine (10), MgATP (4), Na2GTP (0.3), NaCl (4), set to pH 7.25 with KOH. It is kept cold during the experiment.

One thing that sets it apart is the lack of calcium ions and that might have something to do with my problem (divalent ions are supposed to help with the pipette adhesion). The thing is - colleagues are using the same solution with the same cells without much trouble.

Hi Asaph,

I went back to your first post. If your goal is to record in voltage-clamp mode I would add some Cs+ to block potassium leak current. I see that you do not have any EGTA or BAPTA to buffer calcium influx to your cell. Also it is quite common to use QX-314 to prevent direct activation of the cell. These may not be exactly what you are looking for because I do not know exactly what your experimental goal is. In the end if it is possible ask for an aliquot of working internal solution from your friend and try it out.

There are so many things that may go wrong with internal. The ingredients may be contaminated in their original bottles. Thus, it is a good practice to have those separate from ingredients that you use for your ACSF. Also it is a good approach to buy new ingredients every year. These precautions should be taken under consideration especially in situation where there are several experimenters in the lab.

Last thing that comes to me is asking you to double check osmolarity of your recording aCSF. Few year ago I witnessed a long-term recording with a huge leak. It turned out that ACSF was hypoosmotic with 150 mOsm.

Please let me know if I could be of more help,

Ewa

Dear Asaph: There are indeed so many problems involved in maintenance of whole-cell recordings for a longer period of time. For example, fine-tuned manipulator used to hold the electrode needs to be very stable. Vibration-free table also needs to remain functional with caution as the cell examined is attached to the glass bottom.

TV TX