What you've described is basically every electrophysiologist's nighmare. If it's any comfort, we've all been there -- or will be, if we stick around long enough.

It sounds like you've tried nearly every likely thing (make new solutions, check osmolarity, switch up your slice preparation protocol). You didn't mention pH, but I expect you've check that too, multiple times.

My suggestion: throw out all your reagents and order fresh. Or if that is too expensive, throw out your calcium and magnesium bottles and order those two fresh. Both ions are very important for getting good seals and for healthy recordings in general. If anything is amiss with them, things aren't going to work.

I offer this suggestion because of recent experience. For a couple of weeks, we struggled to get gigaohm seals from basolateral amygdala neurons in mouse brain slices, even though for many preceding months everything had worked just fine. Like you, we ran through the gamut: pH, osmolarity, new solutions, new silver wire, etcetera and so forth. Nothing worked. Finally, we checked our reagents. We found that on the shelf were two magnesium chloride bottles, used to make the magnesium chloride stock solution we pipetted into our ACSF every day. The magnesium chloride in the older bottle looked and felt like what MgCl2 always does (large-ish crystals, clumpy, sticky), but the stuff in the newer bottle was a fine powder. I don't know what it was or where it came from (nobody had written anything on the bottle), but it definitely wasn't MgCl2. We remade our stock solution with the old stuff, and everything worked again. Moral of the story: check your reagents and especially check magnesium and chloride.

Good luck.

What you've described is basically every electrophysiologist's nighmare. If it's any comfort, we've all been there -- or will be, if we stick around long enough.

It sounds like you've tried nearly every likely thing (make new solutions, check osmolarity, switch up your slice preparation protocol). You didn't mention pH, but I expect you've check that too, multiple times.

My suggestion: throw out all your reagents and order fresh. Or if that is too expensive, throw out your calcium and magnesium bottles and order those two fresh. Both ions are very important for getting good seals and for healthy recordings in general. If anything is amiss with them, things aren't going to work.

I offer this suggestion because of recent experience. For a couple of weeks, we struggled to get gigaohm seals from basolateral amygdala neurons in mouse brain slices, even though for many preceding months everything had worked just fine. Like you, we ran through the gamut: pH, osmolarity, new solutions, new silver wire, etcetera and so forth. Nothing worked. Finally, we checked our reagents. We found that on the shelf were two magnesium chloride bottles, used to make the magnesium chloride stock solution we pipetted into our ACSF every day. The magnesium chloride in the older bottle looked and felt like what MgCl2 always does (large-ish crystals, clumpy, sticky), but the stuff in the newer bottle was a fine powder. I don't know what it was or where it came from (nobody had written anything on the bottle), but it definitely wasn't MgCl2. We remade our stock solution with the old stuff, and everything worked again. Moral of the story: check your reagents and especially check magnesium and chloride.

Good luck.

Thanks for the suggestion Niraj! I ordered all chemicals including MgCl2 and CaCl2 new and they should be here in few days. Also, this afternoon I asked one of the post Docs in the department to record from my slice on his rig. Bath solution was mine, internal was mine and also the slice. He got giga seal within few seconds and started recording almost perfect synaptic currents in gap-free mode. So, does this suggest that nothing is wrong with my solutions and also my slice? Now I can only think of my recording pipettes and pressure system (to get seal and break in). He uses pipettes of about 5.3-5.5 megaOhms when mine is about 2.5. I fire polish mine but he doesn't. And, the surprising thing is that his pipettes reads 1.8 megaOhms on his system (Axograph) while it reads 5.3-5.5 in mine (Patchmaster, HEKA). I wonder why is that. I am also not sure what should I do with my suction/pressure system. There is no leak as I checked them yesterday. I wonder if I am applying too much positive pressure and as a result when I release the pressure the suction force through the pipette is little bit on the high side. These are all just speculations but I got the same result with pipettes pulled from his puller as well. This really makes me wonder if my pressure system has a problem. But, I don't know what could be done about it since its a pretty simple system with a tube connected to a 5 ml syringe with a stopcock.

If your pressure system was the problem, I think it would be pretty obvious. As the pipette touches the slice, you ought to be able to see tissue being pushed away. If you have a leak or the pipette is clogged, you won't see that. Likewise, as the pipette approaches a cell, you ought to be able to see a dimple form (I'm assuming you're not doing blind patch). If your pressure is so much that you're damaging the cell, I'd think that would be quite noticeable.

Smaller pipettes might help with seal formation, though it's harder to break through and they're a bit noisier and more apt to seal over during a recording. I don't know what to make the of the difference in the tip resistance measurement between your rig and your friend's; I'd check with an oscilloscope.

Thanks for the suggestions once again Niraj and Michel!

Sorry for this late update. I was trying different things over the past week or so and changed few things in the rig to see if that makes things better. I did get several giga seal and also got few good recordings. However, the same problem comes back and I thought recording a video of the problem will help you guys understand it more clearly. The video is attached here.

The video is after releasing the positive pressure, applying some suction and holding at -10 mV to ease in getting giga seal. But you can see it is going towards getting a giga seal but then there is a sudden upward shift and then its a gone case. This has been the problem for months and the case with almost 99% of cells I have patched over the last two months. Very frustrating but investigating this has also helped me narrow down where the source of the problem. It is certainly not the slice health (that I initially thought), not the external and internal solutions or issues with pipette (however smaller pipette, about 5 Megs, works better).I wonder if one of these three things is causing the problem

1. Loosing pressure (positive and negative). However, I have put in a new pipette holder with new O-ring, changed the tubing and checked for leaks by dunking the pipette holder in a beaker full of water. I don't think there's a leak, but I have not used a mamometer.

2. I may be applying a little too much negative pressure. I have seen the upward shift as seen in the video happen simultaneously with cell organelles getting pulled inside the pipette.

3. I may be releasing the positive pressure either a little above the cell surface or too deep into the cell. (The system and the optics I am using makes visualizing cell surface pretty difficult).

Any thoughts?

Thanks Michel!

Yeah, I ramp up the potential to -70 mV starting from Zero. And, the cell that doesn't need any negative pressure were the cells where recordings were the best and they stayed stable for long time.

I haven't changed the stopcock but the syringe. I should do that. And, the resistance change with pipette tip close to the cell is something I haven't paid attention to. That was the way I did things when patching on cultured cells but with the current rig (upright objective) and the way how its set up both my hands are occupied, one on lowering the objective and other lowering the electrode, and my eyes are looking at the cell on the monitor screen. That makes it hard to look at the resistance change on the other screen but it is certainly doable and I need to get that in my trial next time. Will update you how it goes.

Thanks a lot again.

Michel had good suggestions: slowly ramp from 0 mV to -70 mV to help with seal formation, make sure you really have positive pressure as you approach the cell, and check for a resistance change as your electrode gets into patching distance.

I'd only add one thing about the last point: the size of the resistance change depends on how much positive pressure you're applying. If the positive pressure is very high, your pipette will be pushing membrane away as it approaches, effectively clearing a path to ground and reducing the effect on the resistance of the proximity of the membrane. In the course of doing something else a few years ago, I actually quantified this effect a little bit. See part B of attached figure. For context, 0.6 psi is roughly the pressure when I blow into the mouthpiece as hard as I can and "dimple" is the position I would consider optimal for patching (where the positive pressure just starts to dimple the membrane). So, I would say ideal is a gentle positive pressure (0.2 psi, just enough to keep the tip clean) and a resistance change of ~20%. If you use more positive pressure, the resistance change will be smaller -- still present but a bit harder to pick out.

It's possible to poke a cell in a slice a few times to check where your pipette tip is in relation to the membrane, meaning move it back and forth diagonally. So, if your visual display and computer monitor aren't both within a single sight line, you could position the electrode visually and then turn your head to see how much the resistance changes as you move the tip back and forth.

Good luck,

Niraj

P.S. The figure I attached is from here: Article MATLAB-based automated patch-clamp system for awake behaving mice

Thanks Niraj for the detailed answer. I see in your paper that 20% resistance change roughly corresponds to the dimple formation. I think this may hold true for slice electrophysiology as well. I recorded yesterday and noticed my resistance change is not 20% but lot less. I know it doesn't always have to be 20% but the change in my case is very minimal. For example, if I have a pipette resistance of 3 megs I will see it change to 3.2-3.3 megs when the pipette makes a dimple. This looks very little to me and I am saying that from my experience of patching in cultured cells. The changes were much bigger in the cultured cells. I think the sample, whether slice or culture, shouldn't matter for this, right? How much change do you see when recording from slice?

I did significantly better yesterday in getting giga seals. I changed three things different. First, as you recommended and I have also practiced, I released the positive pressure when I see the dimple. Second, I immediately applied gentle mouth suction after releasing the positive pressure. I was applying stronger suction previously and also was a little slow on applying suction after releasing the positive pressure (slow in few seconds). Third, I patiently waited for giga seal instead of getting restless and applying more negative pressure to get seal faster. I think all of these helped. One of the post docs in my department told something that stuck with me. He said 'When things go wrong after a long period of good recordings we try many different things to get it right and in that process we forget what we were doing right'. I think it holds true in my case with respect to applying gentle suction and not releasing suction until the dimple appears. I thought I had a leak in my pressure system and was so paranoid of loosing positive pressure that I was releasing positive pressure prematurely. May be paying more attention to these things that I may have ignored will make things better.

Thanks a lot again for all the suggestions. I will update you when I record next.

In slices, when using visualized patch (as opposed to blind patch), I usually use a positive pressure much higher than 0.2 psi and get a resistance change closer to 10% than 20%. So, for a 3 MOhms electrode, 3.3 MOhms is about right.

That advice you got -- "don't forget what you were doing right" -- is really good, something for us all to remember.