In Vitro Electrophysiology - Science topic

There are those who take fancy precautions, like enclosing their experiment rooms in Faraday cages, or planting dedicated copper stakes for grounding in their basements. In my opinion, that isn't really necessary. I think it is most practical to just set the rig up, and de-noise using standard methods. The basic idea is to tie every piece of equipment to a common ground, avoiding ground loops. For example, if the air table serves as the common ground, then all devices ground to that, not to each other. We do use Faraday cages to shield our in vivo and in vitro rigs. They help a lot. It takes fiddling. Best of luck.

-Matthew

Then probably your best options are either to attend a school/workshop that also teaches ephys (but you would gain only limited experience), or to look for a postdoc in which you can combine both ephys and also the techniques in which you are already trained. Good luck! :)

Depends on your budget. If that's not a problem, get yourself an Axopatch 700B. You can do more than just field potentials. You can, when needed, do whole-cell experiments (say, when buffering internal Calcium with EGTA or Bapta, blocking a channel from the cytosolic side...etc.,) and unitary activity channel recordings (all configurations: C-A, I-O or O-O). Furthermore, you can perform , ion-selective electrode recording, amperometry/voltammetry and bilayer recordings.

I agree with Max Anstötz in general. Mammalian brain tissue requires continuous temperature control, oxygenation, nutrient supply, and waste removal if you want to study electrical activity that resembles physiological. On the other hand, there are ways to keep brain slices in culture for up to several months, through the tissue "flattens out," and the neuronal phenotypes may drift over time as some synaptic connections are lost and others change. You may want to look at the work by the Gähwiler laboratory in Switzerland.

Hello Diego Fernández-Fernández

I am also trying to get some single channel recordings of the Nav.

The solutions look fine in my opinion, but the protocol in my opinion is not correct.

The holding that you have in figure 2 is 0mV and not +100mV. If I am not mistaken in the Clampex software when you create a new protocol you decide the holding potential under the option outputs and there you put the desired holding.

Afterwards in the option waveform you can decide the first step and the delta level.

And yes the current you expect to see in the raw recording should be "positive".

About the regular spikes in the Patch mode, I was also wondering and decided to ask the Axon support and this was their reply "Instead of resistor, a feedback capacitor is used in the Patch mode in Axopatch 200B amplifier. The voltage across the feedback capacitor cannot ramp in one direction forever. At some point the capacitor voltage will approach the supply limits and the integrator must be reset to start again near zero volts. Thus the current record must be interrupted for 50 us while the integrator and differentiator reset. The frequency of resets depends on the current passing through the headstage, with the larger current requiring more frequency resets."

Hello Lingdi!

Of these two drugs, gabazine is more convenient to use, as it is water soluble and requires a smaller concentration to block the evoked IPSC (in rat cortical slices we use 5-10 uM of gabazine or 50-100 uM of PTX). However, gabazine is more expensive than PTX.

Bicuculline methiodide also blocks the GABAaR-mediated IPSC (it is water soluble and is cheaper than gabazine). Note that it affects the SK-channels.

If the IPSC in your preparation is mediated by both GABAaRs and GABAbRs you should also apply CGP-55845 or use the pipette solution with cesium salts.

Good luck with your experiments!

Thanks very much, we will try this and see what we will get

We did have photocopy manual for biologic 400.

Dear Maaike,

Consistency of the pull results, or rather lack thereof, is the most common issue in puller tech support. It can have a variety of reasons, either mechanical issues, or inappropriate choice of glass or the pull parameters. For some of the mechanical issues, I made a diagnostic video a few years ago: https://www.youtube.com/watch?v=75y3dfGwP58 You also want to make sure your drierite still has some blue left: https://www.youtube.com/watch?v=BLvYLLJmcnY

As for the pull parameters, be sure to follow the Pipette Cookbook. The latest revision can be downloaded from the Sutter web site. In a nutshell, you start with the program recommendation in the Cookbook, find the midpoint velocity (p. 30), then write the program out into individual lines (using line repeats on the P-1000) and fine tune your tip size on the last line. If that does not get you to consistent pull results, fill out the 15 questions worksheet on the support page (https://www.sutter.com/PDFs/15questions.pdf) and send it to adair@sutter.com. She will help you get to better pipettes and can also answer questions regarding the P-2000 Puller.

In general, heavy use in itself will not make the pull results inconsistent. Only if those many users constantly bump the filament and then don't realign it and run a ramp test, you will obviously see variability. Wear parts like the filament and drierite will also age faster with frequent use. You should not need to come to the same conclusion as Paula! Sorry to hear about your unsatisfactory experience, Paula. Did you contact Sutter about your issues? We do provide support for our products, and in virtually all cases we can get the customer to the point they want to be.

Both Niraj and Harald correctly state that the P-2000 Puller is not optimized for capillaries with more than 1.2 mm diameter. That simply has to do with the diameter of the laser beam. Too thick capillaries absorb most of the beam on the first pass, and it becomes increasingly difficult to reflect and focus the remainder of the beam on the back of the capillary. Off-center taper and tip are the results. If you already own a P-1000 and do not want to use quartz for low-noise single-channel recordings, you should try to optimize that before spending the money on a P-2000.

I hope this helps,

Jan

Frankly speaking, I dont have any experience in MEA recording. But let me clarify. Do you apply any stimulation protocol except for picrotoxin?

I mean most of neurons in slice/culture won't fire untill they are not stimulated, cause their main afferent paths have been interrupted during prep.

Although in some places you can catch pacemaker neurons activity (see 10.3389/neuro.01.1.1.009.2007 for further info). But most of neurons in slice/culture would make one single spontaneous spike in 5-10 minutes without any stimulation (eather chemical or electrical). Most of chemical stimulation protocols (low Mg2+, high K+, 4-AP, picrotoxin, etc.) are widely used as ex vivo epilepsy models. (see 10.3791/57548).

Hope this will help. Good luck!

here is my recipe for choline chloride cutting and slicing if you want to try it.

As I anderstand you wish to see Schaffer collaterals. You can't see them

All you need to see is cell layers and stratum radiatum, good anouth to be able to place electrodes. First place stimulation electrode in strat. rad. clouser to cell layer, were SC more dense, and start to place recording electrode somewhere in the middle of strat. rad., where dendrits is, (this is for field potential recording).

Than try to get EPSP (field potential) by increasing stimulation intensity. You may need to replace recording electrode several times to get good shaped potential.

Dear Zhang,

In adiition to above comments, I would like to add that one can use high Cl- concentrations (500 mM to 3M KCl) with high impedance electrodes (60-100 megaOhm) during sharp electrode intracellular recordings. For examples, you can see my earlier publication.

Best wishes,

Refik

Our company develops 3D cell barrier culture systems for in vitro study of virtually any cell barrier, and the system comes equipped with both a Fluid Perfusion Unit and a Trans-Endothelial/Epithelial Electrical Resistance (TEER) Measurement Unit! The use of a vessel-shaped 3D environment has been proven more effective than 2D systems like Transwell (

Article Santaguida S, Janigro D, Hossain M, Oby E, Rapp E, Cucullo L...

), and our advanced TEER Measurement Unit allows for frequency sweeps between 0.1-1,000 Hz, automated time point sampling, logging of data to Excel, and additional measurement of phase angle for cell capacitance calculations! For those that are committed to Transwell use, we also have a TEER Measurement Unit that is compatible with nearly all Transwell products (Endohm cell cup chamber, STX2 "chopstick" electrodes, etc) and allows the user to greatly expand their testing capabilities with Transwell equipment. Additionally, we have smaller modular systems that connect to microscope-friendly cell culture modules: i.e., a miniaturized 3D cell culture system that you can use right under your microscope! If you or anybody on this thread is interested in learning more, I encourage you to visit www.flocel.com or email me at djanigro@flocel.com, thank you!

For reconstructions from 3D image stacks, we have been quite satisfied with the Neuromantic freeware:

https://www.reading.ac.uk/neuromantic/body_index.php

It does what it is supposed to, with no bells and whistle that complicate matters more than they help. With Neuromantic, I have put dozens of undergraduate students on the manual reconstruction work, they get their lab experience, we get our reconstructions, everybody's happy. We've done >150 reconstructions in the lab this way. Works great, but support is nil and updates rare.

I have heard good things about Fiji/Simple Neurite Tracer but never tried it. There is also Herman Cuntz's TREES toolbox (http://www.treestoolbox.org), which is more advanced. Finally, you might benefit from the option not to reconstruct at all (Ferreira et al: Neuronal morphometry directly from bitmap images. Nature Methods 2014 DOI:10.1038/nmeth.3125.).

All of this requires 3D image stacks though, so you do not work "live" like people often do with Neurolucida. However, working offline comes with advantages too.

For our work with Neuromantic, see below, in particular Blackman et al:

Blackman AV, Grabuschnig S, Legenstein R, & Sjöström PJ: A comparison of manual neuronal reconstruction from biocytin histology or 2-photon imaging: morphometry and computer modeling. Frontiers in Neuroanatomy (2014) 8:65, DOI: 10.3389/fnana.2014.00065.

Lalanne T, Abrahamsson T, & Sjöström PJ: Using Multiple Whole-Cell Recordings to Study Spike-Timing-Dependent Plasticity in Acute Neocortical Slices. Cold Spring Harb Protoc (2016) 10.1101/pdb.prot091306

Buchanan KA, Blackman AV, Moreau AW, Elgar D, Costa RP, Lalanne T, Tudor Jones AA, Oyrer J, & Sjöström PJ: Target-Specific Expression of Presynaptic NMDA Receptors in Neocortical Microcircuits. Neuron (2012) 75:451-466.

Lalanne T, Oyrer J, Mancino A, Gregor E, Chung A, Huynh L, Burwell S, Maheux J, Farrant M, and Sjöström PJ: Synapse-specific expression of calcium-permeable AMPA receptors in neocortical layer 5. The Journal of Physiology (2016) 594(4):837-861.

Thank you very much Kelvin Broad for the help.

I would test to make sure you're getting stable EPSCs before trying to measure the NMDA component of the EPSC. If you're not reliably evoking stable EPSCs, playing with the conditions to isolate the NMDA currents will just be frustrating. Using a Cs based internal solution with normal aCSF and measuring at +40 mV would be one way, and would probably be the easiest. That way, you can ensure you have a stable EPSC at -60mV and then block AMPA-R's with NBQX and shift to +40mV to measure the NMDA currents. Alternatively, reducing the Mg in your bath and measuring at -60 mV is another method people have used. You might also consider adding a low concentration of glycine to the bath to help facilitate the NMDA currents, even though it shouldn't be limiting.

Dear Ercan,

Nice recordings.

From what I understand here you need to plot dV (voltage difference in mV) in the y axis and the dT (time difference-ms) on the x axis. It all seems to happen within 7 ms or so, as slowers are not able to generate any action potential-but only similar subthreshold potential changes. For example, 1ms long ramp at 1nA current is giving you about 70 mV difference (dV = 70 mV and dT = 1ms); 2 ms long 1 nA ramp giving you 65 mV (dV =65 and dT =2 ms); say 8 ms long ramp is giving you dV of 30 mV-the longer ramps are pretty much staying at that level and therefore you may need all of the longer ramps. So you plot them that way it should give you an exponential decline. If you like you can fit those with an exponential(s). These values you may compare among individual cells or groups of cells depending on what you are aiming at. For example some people use it/these values for comparing excitability.

Best wishes,

Refik

Do you want to perform patch clamp in slices or whole mount retina? For whole mount you can use small pieces of filter paper containing a small hole in the center. You can then attach the pieces of retina by applying gentle pressure (for more info see Marla Feller's papers, UC Berkeley). For slices, I think a grid would work fine, but be careful with the size of the slice. If it is too big, probabilities are that they will "fly away". 

Inn addition to blocking all non-sodium currents, shorten your test pulse to 10 ms. 100 ms is too long and is suitable for calcium and/or potassium currents.

Dear Jerome, thank you for your interest.

Yes, you are right, the idea of building our chamber was based on need of reproducibility, easy handling  and pretty big working surface (we search a lot of devices availible on market or published). We have before in our lab also setup with agar bridges and did nice observation of cell migration, but it is really not easy for handling, especially for long (2-3 weeks) -term stimulation of some samples in parallel.  

After 150 (200) mV with our setup we start having problem with medium and we were pretty happy with the effect of 100 mV. 

Not too late at all Matt, many thanks for your help!

Yes. Sometimes, under current clamp conditions, seal resistance may be unstable.  The baseline level of resting potential may fluctuate and hence generate action potential if potential change reaches the threshold level. "Spontaneous" action potentials could be virtually overestimated.

De Vanja,

Please see the below given link.

 http://persweb.wabash.edu/facstaff/torbertn/papers/torbert05.pdf

Regards

Krishna

DMSO can seriously damage  membranes and I think it could contribute to what you experienced. Are you using > 0.1% DMSO in the final solution? Some of DMSO soluble drugs can be dissolved in EtOH. Have you tried that?

It is mandatory to ascertain the osmolarity of the solution, otherwise the concentrtion may not support neuron survival. This article may help:Slice preparation Solutions ACSF: (in mM) 125 NaCl, 2.5 KCl, 2 CaCl2 ...

Hi Ruoqi,

The answer is yes. It is definitely doable to patch PV+ Fast-spiking interneurons(FSIN) without fluorescence.But the difficulty is depend on the brain region. In dentate gyrus, you can Identify FSIN by there soma size, which is much larger than granule cell. However, in CA1 region, the FSIN soma size and shape are very similar to pyramidal cell, so it is much harder to find them with DIC. You should find out whether FSIN have special shape/size/location in your target region.

About the age, PV is not expressed in certain age. You can try to use lhx6 or GAD as promoter to drive GFP. So you can identify interneuron first, and further identify PV neurons with their intrinsic electrophysiological properties.

Cs stands for the membrane capacitance of neurons; when asked for compensation, the amplifier uses series of small currents to get the Cs values based on tau and membrane resistance. Based on these measurements, extra current is passed during recordings to load the Cs.

This automatic procedure however is run in a way to make sure that Cs is not over-compensated as the latter can cause damage of the membrane and deterioration of the patch quality. Thus, typically upon request for Cs compensation, the Cs is underestimated and under-compensated. This is especially the case with larger neurons.

If your settings are right, you should be able to repeat Cs compensation several times (with no harm) to get the Cs properly compensated yielding good recordings. Hope this helps, SV     

You could try using multi electrode arrays embedded in glass bottom Petri dishes for electrical activation

Just  CO2 to the medium

Nat Methods. 2007 Apr;4(4):353-8.

Unitary permeability of gap junction channels to second messengers measured by FRET microscopy.

Hernandez VH1, Bortolozzi M, Pertegato V, Beltramello M, Giarin M, Zaccolo M, Pantano S, Mammano F.

Nat Cell Biol. 2005 Jan;7(1):63-9. 

Impaired permeability to Ins(1,4,5)P3 in a mutant connexin underlies recessive hereditary deafness.

Beltramello M1, Piazza V, Bukauskas FF, Pozzan T, Mammano F.

There might be many reasons for that, it is quite common. I suppose you are talking about stimulated activity, how are you stimulating ? And what is the behavior of the spontaneous activity ?

What is your chloride reversal potential ? If you are not blocking GABA receptors, it is possible that you are recording mainly GABA responses, and with the diffusion of your intracellular solution you slowly shift them to shunting.

If you are stimulating too strongly, you may just exhaust your synapse. Try to be in minimal conditions.

It could be a too high temperature as well, exhausting your synapse. 31 - 32 °C is fine, higher can be trickier.

For acute slices it might be better to use a vibratome rather than McIlwain tissue chopper (more intersting for organotypic hippocampal slices from p5-7 pups). Is the ACSF in which you perform dissection and slicing cold enough to get the tissue dense to avoid crushing when you cut with the chopper ?

please check - http://www.ncbi.nlm.nih.gov/pubmed/23318870

You might consider to convert your anti-NKG2A antibody of choice (and the respective isotype control) into Fab/F(ab)2-fragments - there are several commercial kits for that purpose that are easy to use and reliable. By using blocking Fab/F(ab)2-fragments you circumvent the potential problem of Fc-receptor-mediated activation. Good luck!

Roberto, test with a voltmeter wherther you have a passing current through the vibratome when slicing (one pole on the blade, the other one in solution)....

this happened to me in the past.... we were electrocuting the slices in the slicing chamber.... cheers

a ground got disconnected inside the slicer... it was supposed to ground the chasis..... yeah crazy shit!

Dear Sunney,

In aged mice, by replacing Na+ with sucrose, NMDG or choline, you are increasing the chances of neurons surviving the insult (dissection, slicing etc). I have tried most of these solutions to obtain good motoneurons, which are arguably the hardest neurons to obtain in adult mice/rat. For details of my ACSF solutions that I used for cutting brain slices at different stages of development, see my following paper. If you are cutting hippocampal, cortical, cerebellar or amygdala slices the modified sucrose solution works well with the old mice. 

best wishes, Refik

Well, I've tested in hippoc neurons from E17.5 mice also at DIVs 14-21 and in my best attempt the cells were "viable" only for 2h, but I'd say that after 1h the cells lose their smooth surface.

I've to do more tests and see if I can improve cell viability.

Thank you for your comments Alberto.

Regards!

It's worth having a look at this: http://www.ncbi.nlm.nih.gov/pubmed/24184743

The advantage of the kainate model of epilepsy is that, after the first KA-induced seizure, there is a latent period after the initial insult which precedes the development of recurrent seizures in the animal. The kainate model also reproduces anatomical features of epilepsy, such as mossy fibre sprouting in the DG. These features of the in vivo KA model require time and anatomical changes in the brain, so are not appropriate to use in an in vitro slice model. Acute activation of KA receptors (don't forget that KA is also an agonist of AMPA receptors) in slices is quite different from the excitotoxic insult used in the in vivo model of TLE.

In injecting the oxytocin neurons, where did you inject?  Not all OT neurons are similar in terms of the connectivity.

Out of the many possible explanations, I think the one that is most likely (and least satisfactory!) is that it's just a poor health region that you're recording from. This to a large extent depends on the age of animal you're slicing from - the younger, the healthier the slices would typically be. But also fEPSP recordings, can take a lot of "hunting around" for regions in the slice that will produce nice/healthy responses; sometimes you'll find that by moving your recording electrode a few micrometers laterally and/or deeper in the slice will give you dramatically improved post-synaptic response to pre-synaptic fibre volley ratio.

The other thing to consider is you mentioned you're using sagittal slices, which I don't think is the most common configuration (most people use whole brain coronal, or transverse cuts of isolated hippocampus), so it may be that in this configuration, it is less likely to obtain large post-synaptic responses.

There are lots of excellent resources in the literature that may be of use to you, this is an example of one that I found particularly helpful:

Bortolotto, Z.A., Anderson, W.W., Isaac, J.T., and Collingridge, G.L. 2001. Curr. Protocol Neurosci. Chapter 6: Unit 6.13.

Best of luck!

You could have at least a couple different things going on.

1) In the presence of bicuculline, you can get increased excitability and thus a larger fiber volley.

2) More likely, by blocking the excitatory response with APV + CNQX, you are unmasking the fiber volley from underneath the postsynaptic response. These can get tangled, especially when responses aren't as clean, like in a place like SLM.

A couple other caveats about PP-SLM recording. Your responses can also be contaminated by dentate gyrus pop. spikes because of your close proximity or via ring-around through CA3 and into Schaffer-Collateral. Some people place remove DG place a cut between CA3 and CA1 (see PubMed ID: 10085331, Figure 1 and Methods) . Also, make sure you check to make sure there isn't contamination of the response by the Schaffer Collaterals. Without moving your stimulating electrode, place your recording electrode into stratum radiatum. If you have a positive field potential then you're good. If you're field potential is negative, you're stimulating electrode is exciting Schaffer Collaterals as well. You'll have to move things around until the negative deflection in stratum radiatum goes away.

Dear Chandra,

In Figure 3, I measured the current at steady state or at the end of 1s long voltage step. The current at hyperpolarized levels was primarily a tertiapin sensitive K (Kir3.1 and Kir3.2) current.

best wishes, Refik

I really appreciate your useful tips!!

Then, now I report results so far and that  I was trying to.

RESULT: Although I seemed that slices were fixed stronger than before that, the motion drift was detected regretfully. The slices pressed by new harp's strings were DENTING, therefore I think that the pressure was sufficiently applied.

RESULT: This was not effective in the problem.

RESULT: I set very slow flow rate (~ 0.5 ml/min), then the artifact was improved but not perfectly gone... This flow rate is too slow in my experimental conditions (acute slices from adult animals, keep the temp. of ACSF at ~32 degree celsius and keep a whole-cell condition over a hour, etc...). Needless to say, the condition of neurons was going to be poor visibly.

...Next I'll try to replace a glass at bottom of the chamber with an adhesive-coating glass as mentioned by Refik.

Dear Jérôme, I agree with you that if one uses constant current stimulation mode the voltage at the stimulation electrode will change due to electrode tissue impedance variation. Therefore one will not get a constant and reproductive potential at the stimulation electrode. This would only be the case if the electrode-tissue impedance always would be constant. Anyway I would recommend the following review paper for details: Tehovnik EJ. Electrical stimulation of neural tissue to evoke behavioral responses. J Neurosci Methods, 1995; 65: 1-17. Also please have a look at the paper published by Histed and others: Histed MH, Bonin V, Reid RC. Direct Activation of Sparse, Distributed Populations of Cortical Neurons by Electrical Microstimulation. Neuron, 2009; 63: 508-22. The authors reported that microstimulation sparsely activates neurons around the electrode, sometimes millimeters away, even at low currents. This very interesting paper raises several issues to consider for the design and interpretation of microstimulation studies! I hope my answer was helpful. Best wishes, Dirk

I used Campden vibratome for spinal cord slices/patch clamp recordings. It was great. Maybe you need to adjust some of the cutting parameters for optimal results ( the speed, amplitude, frequency), fine-tune the z-axis oscillations, check the quality of the blade etc.

Dear Chandra,

In order to obtain Ca2+ current in cardiomyocytes first what you can do - keep them at holding potential of -40 mV and step them to 0 or +10 mV with rectangular pulse of 200-300 ms duration.

Then, to validate, that this is calcium current you can add dihydropyridine - verapamil in order to block ICa2+, or BayK8644 to increase it.

Patching at holding potential of -40 mV should help you to get rid of sodium current. In case if it does not happen, you can add 1 micromolar of TTX to your bath solution or reduce concentration of NaCl partially replacing it, for instance, with NMDG.

Best regards,

Miroslav

Dear Agata,

If I were you I would do the single channel recordings in expression systems that overexpress only your specific channel. And you can use native cells for studying whole-cell patch clamp properties of the channel. This way you can save your alot time and avoid frustration.

best wishes, Refik

ATP is actually quite stable in water, especially at pH 6.8-7.4 - adding freshly prepared ATP/GTP, or keeping the internal solution on ice is more like an habit which actually has little influence. But like always, better to play safe ;-)

You should definitely explore optimizing your cutting and incubation conditions for slices from older animals.  Some prefer sucrose or NMDG-based substitution (see, for example, www.brainslicemethods.com for recipes).  We have found that cutting in NMDG-based aCSF, then storing in a modified higher (20 mM) HEPES aCSF can improve tissue viability dramatically.

Hi,  question is  how  fast and how accurately you want to do this and what equipment you have to hand ?  There are now K+ dyes available  for extracellular recording but they are not very accurate if your extracellular space is less than the Abbe limit and can't be resolved (less than 250 nm). There are of course a few 'traditional methods to do this. Depending on what you have availably in terms of equipment:

1. Extracellular ion sensitive micro electrodes. Pros good point accuracy measurement.  cons poor time resolution (response time slow)  and they tend to 'make' a several micron space so don't really reflect the real K concentration. Also there are iononophores availably for Na as well.

2.  If your cell has well characterised K currents and you have good whole cell voltage clamp,  you can use the change in reversal potential of the K current to measure the resting and dynamic changes in extracellular K concentration.  this is very accurate and has good time resolution but requires a bit of nifty calculation.  If they have prominent Na currents you could use them as well to determine the extracellular Na concentration. 

Hope that helps (I can send you refs for the 2nd option if you are interested.

good luck!

Euan

make sure that you are using the strong Collagenase (e.g. Worthington Biochem type 2) and try slicing the aorta so you get more surface area for digestion. You should also either use a fungicide during arota prep or do your prep in a biosafety cabinet to minimize fungal infection of the cells. Also, make sure to wash the cells in media at least twice and allow some recovery time in an incubator after digestion.

Hi Kathleen,

I've not ordered anything yet, but I'm reasonably certain I'm going to go for the A-M systems 1800 amplifier. I've not decided on an ADC yet, but it'll probably be either one of the NI systems, or an ITC16/18, with acquisition done through Igor Pro.

Although I've just got back from a collaboration doing in vivo field recordings, and I'm trying to work out if there's a way I could adapt the Intan RHD2000 evaluation system to double up as both an in vivo and in vitro amplifier, as these are cheap but really powerful, and I could dispense with the ADC altogether.

I hope this helps!

Mick

Perhaps my phrasing was unclear - I meant that TEA only blocks N-type inactivation when it is applied intracellularly as opposed to extracellularly, not that TEA blocks N-type inactivation and nothing else when applied intracellularly. You are correct that it blocks many K+ channels, I was assuming this would be in culture if the goal is to test a specific type of inactivation. 

Hi Abhilasha,

I agree with both Teresa and Victor, and they both raised important points. In my experience I have always noticed difference in responses, as I think it was primarily due to intrinsic membrane and handling properties of individual cells.

best wishes,

Refik

Hi Jon,

That age should be no problem, though I would not go any younger as the hippocampus still developing.

For recording, I would recommend the Kerr Scientific Instruments mobile set up. It comes in a small suitcase including manipulators stimulating and recording electrodes and everything that you need. Extremely practical, cool, economic and highly recommended. I love it. You can check their products and all the details on line at:

cheers, Refik

Hello MiaoMiao

Our lab also uses this vibratome and recently we have had the same problem. In our case, the spring at the bottom of the blade swing arm had become dislodged and resulted in the blade fitting becoming misaligned, hence the screw sticking up. Cutting under these conditions can put a lot of strain on the whole mechanism and result in further very expensive damage (as was our case). We rang our local Leica tech and had the machine serviced as soon as it was possible. As Refik suggested, best to have the company come out and have a look. Good luck!

Cheers Michelle

I agree with the above, you should make your pipettes have a wider aperture. Also of  note : your Cl ion concentration is going to be lower in your CsF internal compared with CsCl and as chloride ions are the mediators between ionic current in the bath and electrical current in your AgCl electrode the ability of it to pass current is diminished. What is your Cl internal concentration? I don't know why your reversal potential is so positive, what is the Nearsnt equation predicted Na reversal for your Na Gradient ? Perhaps if you post your internal and external solutions it might reveal something else.

True, ionic concentrations of media used for suspension culture may be the possible reason for such high value. You need to wash the cells maintaining it's osmolarity and then measure the EL.

Dear Martina:

What you see in your recordings is normal. As mentioned above, TEA blocks voltage-gated K+ channels and the end result of this is a drop of the membrane potential. Typically, after short exposure, TEA can be washed out quite well, which leads to  partial recovery of the membrane potential. Note, that when applied briefly, it blocks K+ channels from outside and only in their open state, i.e. if cells are clamped at  negative potentials, the potency of TEA drops dramatically. Although TEA is thought to be a non-permeable K+ channel blocker, during longer exposure some of it gets inside of cells (especially when the seal is not good).  When this happens, it also blocks K+ channels from inside, and as a result, its washout becomes very problematic. Hope this helps.  

Good luck,

SV

The reversal potential can be calculated using the Nernst equation:

E_K = (RT)/F * ln ([K]_o/[K]_i)

where E_K is the Nernst (or equilibrium or reversal) potential for potassium, R is the universal gas constant, T is the temperature (in Kelvin), F is the Faraday constant, and [K]_o/i refers to the external and internal potassium concentration, respectively. Just plug in the numbers. For a HEK cell experiment, I keep my internal K concentration at 152 mM. If you use 4 mM external potassium (and we assume room temperature to be 25 C), then the E_K comes to -93.4 mV. For 40 mM extracellular K, you get an E_K of -34.3 mV.

To simplify: Potassium is high on the inside, which creates a chemical gradient toward the outside. Brownian motion, simply by statistics, will drive K toward the outside. As K leaves the cell, it takes with it one positive charge, making the inside more negative. With every K leaving the cell, the cell gets more polarized (negative on the inside); in other words, an electrical gradient is established that is opposite in direction compared to the chemical gradient (because the positive K+ is attracted to the negative inside). At some point, when the electrical and the chemical gradient are equal but opposite in sign, you have reached E_K. There is no more *NET* movement of K toward the outside; there is still K exchange, but for each K leaving, one comes in.

When you raise your extracellular [K], then potassium has a harder time leaving the cell. Therefore, the reversal potential is not as negative.

For reliable recording it is important to use K-gluconate. In addition to the problem of juction potential,  Cs blocks K channels and causes decrease of resting potential. So, parts of the cell, which you can't properly clamp (this is usual problem at whole-cell recording from neurons) will be artificially depolarised. 

Dear Caner, I agree with Henrique and Rogier as long as we consider astrocytes as the only glial cell you may encounter when patching in the CNS... Unfortunatelly for neurocentric oriented patch-clampers, but not for the health of your brain, you have also other types of glial cells that have quite different properties than astrocytes in terms of electrophysiological behavior. An easy one to distinghish is the microglia. In physiological conditions (even in slices) the input resistance of these cells is usually in the order of 1gigaohm. They express, no voltage activated currents and thus no action potential. The resting membrane potential is somewhere in between -20 and -70 mV but not so easy to measure due to the comparable membrane and seal resistances. They are not coupled through gap junctions. If you are in pathological conditions or during normal postal development, things may change (voltage activated potassium channels, chloride channels etc...) but still no action potential!. You can check the papers of Helmutt Kettenmann, Claudia Eder, or mine, (although none of us recorded these cells in current clamp I think). The most difficult ones are the NG2 cells of oligodendrocyte progenitors (OPCs). Those cells express voltage activated sodium channels and can fire small action portentials. David Attwell published a paper some years ago showing that some of these cells can fire repetitive action potentials just like neurons but I am not aware of any other report of this kind. What is sure though is that they can fire one or two-three small action potentials. And they also have actual synaptic inputs, so you will detect IPSPs and EPSPs from these cells, but usually at lower frequency than in most neurons. Also, the properties of NG2 cells change during postnatal development. You should check the papers of Dwight Bergles, Christian Steinhauser, Maria-Cecilia Angulo, David Attwell, Maria Kukley for instance to get a better idea of the behavior of these cells.  Finally, when NG2/OPCs differentiate into oligodendrocytes, they loose their ability to fire action potential, they don't show anymore synaptic currents and they express different potassium channels. Much easier to distinguish from neurons than NG2 cells.

I know this seems to be an awfull situation for beginers with an interest only in neurons but you will learn rapidly to distinguish neurons from glia. Using good optics (DIC or Dodt contrast) also helps to recognize them in slices where they usually have not the same aspect as neurons.

Check your signal reversing stimulus polarity. If anything looks just mirrored vs. 0 level, it is all artifact. 

Will not work with lockin amplifier with pClamp 10.  Problem is that pCalmp 10 will not  run lockin function (If anyone knows how to I'd like to know as well).  Axon (Molecular devices) suggest using the sealtest function in pClamp 10 and this works OK.  When you say 'fine' what range are you thinking of sub fF? and at what sample rate?  

Hi Rajesh,

I am assuming here that you have pClamp 10 and Axopatch 200B as per the question you submitted on LinkedIn. One last bit of useful information is, which Digidata model are you using?

The suggestion of Olivier to telegraph information from the Axopatch to the Digidata and Clampex is very good. I am just not sure that you can telegraph from the Axopatch 200B. That's a question you can ask Molecular Devices directly.

To configure pClamp for current clamp recording with an Axopatch without telegraphing I would use either Clampex or Axoscope software (both parts of pClamp). Historically I have used Axoscope because I didn't telegraph information from the amplifier (you can't receive telegraphed information on Axoscope whereas on Clampex you can) which in my case was also an Axopatch 200B. Then I would proceed as such:

One potential source of error might be if the scale factor you entered is not accepted by the software for some reason. In this case try using the "Scale Factor Assistant" (still in the Lab Bench window).

I hope it helps.

I presume you mean decreased current intensity within the slice? This sounds like simple habituation. If you're interested in cellular mechanism, the likely target would be voltage-gated ion channels. If voltage-gated ion channels are responsible, you could attempt to block them with a compound to "resolve" this.

Okay then please check

Sajikumar and Frey, 2003, 2004

Synaptic tagging and cross-tagging: the role of protein kinase Mzeta in maintaining long-term potentiation but not long-term depression.

Sajikumar S1, Navakkode S, Sacktor TC, Frey JU.

Best

Appreciate the replies y'all. 

Agreed that tissue health is critical, Steven. I've found that keeping a low 'time from decapitation' seems to best guarantee that (in addition to the stated prep techniques). This not only leaves more cells alive from which to choose, but makes them better able to withstand the needed disruption of the matrix. 

Shockingly, this week I've had very good luck sending in a first pipette and *actually attempting a patch* on a cell of interest, then pulling that pipette back such that the cell is very slightly stretched and applying the short burst of suction typical of breakthrough at the same time as rapidly pulling the pipette back with the micromanipulator. This was initially done as frustration after a failed GOhm seal, but it seemed to rupture both the extracellular matrix and peri-neuronal net without killing the cell allowing another pipette a successful patch. Extremely resilient little dudes, these cells are.

I've also managed to break the matrix by lowering a relatively large (3-4MOhm) pipette that is unfilled into the bath and allowed it to passively attach to the matrix as it front-fills. An alteration between light suction and slow, low amplitude, but long repeated movement picks up a lot of the matrix - rinse and repeat until the cell is clean. For a particularly stubborn matrix I pulled a pipette to a point and jousted diagonally (it helps if you're playing medieval English music and fretting about the amount to which your glass is doped with chivalry) into the matrix until I made a hole and then proceeded with the above.

Certainly going to try enzymes in the future, but deadlines approach! I'd also love to try organotypic slice culture, but the same applies.

(continue) @Chenglong Yu

(roughly R is ~500-1000 Ohm, but it depends). I could not find the design map of the Iso-flex from website, then the internal resistance Ri could not be determined. But for the Max Output Power (MOP), we can estimate based on two parameters: Voltage output: 0-90V, Current output: 0-10mA, thus MOP is roughly 90Vx10mA =900mW. If you set the output voltage 90V, the expected current through electrodes will be around 90V/1KOhm = 90mA, then the power between electrode will be 90Vx90mA=8100mW>>900mW, means the Load is too high for the MOP, or say it is overloaded. In this case, the actual voltage between electrodes would be much smaller than 90V, which might not be adequate for inducing vesicle turnover.

2) If the liquid surface is very high, then the equivalent R of solution in chamber is very small, let's say 100-300 Ohm (but it depends). In this case, we could not neglect the internal resistance Ri of the isolator. The actual voltage between electrodes depends on the ratio of 90V x R/(R+Ri), which might be much smaller than 90V.

It should be noted that in both cases the key problem is the OVERLOAD of power.

In addition, the Iso-flex can only give monophasic but not biphasic pulse, leading to the ionization of solution near electrodes during high-frequency repetitive stimulation, such as 1200APs 10Hz stimulus. This might cause the change of local R and subsequently affect the local vesicle turnover.

Finally, small tip: always double-check the input signal generated by the stimulator and make sure the voltage of signal is > 5V, otherwise one can not even trigger the TTL circuit of isolator and no (or very weak) output pulse will be elicited.

Good luck! 

We tend to use Retiga 200R cameras from QImaging for patching in slices. Thorlabs also sells a few less expensive cameras. Anecdotally I've heard of people even using the DCC1545M (about $350) for slice patching - these are supported by micromanager.

Hi Shakil,

The exponential standard is not the same equation as the n4j equation of the H&H model. 

The approach suggested by Oscar is a convenient solution, but remember to follow his advise in carefully choosing the region to fit. 

The choice of the equation to use for fittings is very important and you must remember that it represents a model of how the physical phenomenon works. The n4j equation implies 4 activation gates and 1 inactivation gate. It must be fitted to the whole current to get the kinetics of both phenomena. 

Since you are interested only in the kinetics of the activation you may use a single or double standard exponential to fit the activation phase of your traces.

I would also use a P/N leak subtraction protocol in voltage-clamp experiments to get rid of the leak current.

For an explanation of the fitting to the activation phase of potassium currents you may refer to Liu & Bean (2014).

Best,

Maximiliano