I am not an expert in this topic, however it could be helpful my own experience.

I am using this device from multichannel systems with 350 microns hippocampal slices and it was difficult to obtain any response at the beggining. It is not necessary to increase voltage applied, with that range of voltage (500-3000mV) in the I/O curve I have obtained 1,45 mV response. Probably, the difficulty to obtain response is due to a bad slice adherence to the MEA surface or a bad slice viability. You could try to use a grid and put it over the slice by the following way:

At first take the slice with a glass pasteur pipette and put onto the MEA surface, then set the position with a paint brush and quikly remove all the solution avoiding any movement of the slice (I recommend use a plastic pasteur pipette with a 200 microliters pipette tip inserted, and remove slowly the solution putting the tip in contact with the MEA surface away from the slice), put the grid onto the slice and add solution over the grid to avoid its movement. If you see a wide propagation of your stimulus artifact, when you start your stimulation protocol, probably you will have problems to obtain any response because it could mean you don't have a good slice adherence over the MEA surface.

By my own experience, it is very important to avoid movement of the slice after removing the solution. Also adherence could be improved with the use of polyethylenimine or nitrocellulose for coating, however, I can ensure the cell debris accumulated by repetitive use is the best way to obtain better responses.

To ensure viability you could try to use protective cutting procedure using ACSF with partial substitution of Na with sucrose. See (http://www.brainslicemethods.com/).

About your doubt in relation with EPSP and population spikes, the main difference is the speed and also their origin, the population spikes appears over the EPSP and it changes the EPSP form. The EPSP observed is the resultant sumation of all the current changes which are taking place around the electrode, population spikes is only asociated with action potentials transmission. Also you can see isolated spikes over the EPSP indicating you have reached saturation. I show you some examples.

I hope it hepls you, good luck.

Dear David,

Sorry for my delayed response.

First of all, let me thank you for an elaborated explanation with pictures for my doubts and question.

It is very very helpful for us.

It is very difficult to find the people working on MEA's for acute hippocampal slice preparations.

We always, generally stimulate 1 or 2 electrodes nearby the SC region and look for response in CA3 or DG regions. Please let us know regarding your experience in stimulation protocol which will be very helpful to us.

Would you please share your experience regarding which region's electrode need to be stimulated and which one we nee to look into while performing I/O curves or the actual recording.

Also please let us know which anesthetic would be ideal for decapitation without any influence on electrophysiological recordings and EPSP amplitude.

We are getting slices made from vibrotome using sucrose cutting solution and they are showing spontaneous response with recordings on ACSF (isoflurane anesthesia). However, the spontaneous response is fading away over the time without any EPSP or population spikes following the stimulation at SC region.

Please give your suggestions in this regard.

Thank you so much for your help.

Best Regards,

Grandhi V Ramalingayya

Dear Dr. Ramalingayya

You can use only 1 electrode for stimulation at Schäffer Collateral region and observe the response in the CA1 region, however, you can see also response at the CA3 region next to the stimulation electrode. For I/O curve evaluation I recommend to analyze the electrode that shows the biggest response, it will be usually the nearest electrode to the stimulation one in the CA1 region. The classical trisynaptic circuit spread from CA3 to CA1 and there are no evidences of strong transmission in the retrograde way (for more information about circuit at the hippocampus you can see Strien et al 2009).

For stimulation you can use voltage adjustment and applying biphasic pulse of 100 microseconds per phase, firstly negative phase following by positive phase, and 1 minute between each stimulus and the following.

For decapitation procedure I don't know the effect of use any anesthetic procedure. In spite of my unknowledge of your work proposal, it could be helpful to know that LTP seems not to work with CO2 anesthesia and maybe halothane works better (by Dr. Frank Hofmann recommendation, from MultiChannel Systems).

In relation with spontaneous activity disappearance, you must take care with slice adherence and viability, that is the main point. You must also ensure carbogen suministration to the ACSF and an ACSF flux of 1-5ml/min to the slice. However, it is not necessary to observe spontaneous activity to obtain stimulus-associated response.

Best regards.

Dear David,

Thank you once again for a wonderful explanation of stimulation pathways with pictures.

Would you please let me know, which solution you use generally for slice cutting either ACSF slush or sucrose???

I have another query that, do we need to use osmometer to check the osmolarity of the ACSF solution??? We generally filter the solution following the pH adjustment and use the same without checking the osmolarity. .

Now we observed that, ACSF is working better than the sucrose cutting solution (for making slices) in view of persistent spontaneous activity and EPSP signals. We have confirmed the same by Dr. Frank Hofmann, MCS.

Isoflurane anaesthesia is working better for us. Also, P28 rat is giving better response than the adult rat.

Thanking you for an extended help...

Best Regards,

Grandhi V Ramalingayya

Dear Dr. Ramalingayya,

You can use ACSF with partial sustitution of Na with sucrose for cutting procedure, however, conventional ACSF must work fine too, because there are many people who use it. According to your question about the osmolarity checking, it would be necessary, however, you must ensure you have no problem with tissue adhesion because it is the main trouble to find response. You must remember that repetitive use of this device increases notably the adherence and the presence of response. As you refers, young animals give better response than olders, if it is not necessary the use of old rats for your experiments, I think it would be better continue with young animals.

Best regards.

Dear David,

Thank you very much for the information provided.

Have you analyzed any data recorded and LTP (EPSP / population spike slope etc.), please let me know. Also how many electrodes you are selecting for stimulus at a time and also for recording.

Would you please suggest the resources for basics in electrophysiology for understanding the fundamentals.

Would you don't mind, please send the soft copies (if you have any) of them to my mail id'

[email protected]

Thanking you,

Best Regards,

Grandhi V Ramalingayya

Dear Dr. Ramalingayya,

In relation with your last doubt in the use of MEAs, if you are using the new software Multichannel Experimenter and Multichannel Analyzer I can tell you this software is still not completely developed, so I recommend convert the files to other formats as CED with Multichannel Data Manager to analyze them.

In addition you will find many troubles with convert files to get a program that will be able to open them. However, CED files can be opened and analyzed by Spike2 software (you can try a free access demo downloadable here http://ced.co.uk/upgrades/spike2demodownload). This software also lets you evaluate EPSP slope or population spike slope. EPSP slope gives information about speed of synaptic transmission associated with synaptic efficiency, so it will give you relevant information.

To evaluate EPSP slope you must measure slope between the 10% and the 90% of the amplitude of the initial curve (Lein et al 2011). To evaluate LTP you must set a baseline (about 20-30 minutes with a variation minor than 10%) and then you apply LTP induction protocol. The baseline's mean gives your 100% response, your baseline response, so the increment that you observe after LTP induction protocol you must express it as % from the baseline response to compare with other experiments that you perform.

As I told you previously you can use just one electrode to stimulate and the others 58 to record. You will usually get response only at the nearest electrodes, however, all the rest could give you information about hippocampal circuit and signal transmission.

To obtain some information about electrophysiology and MEAs it could be interesting to read "Current Laboratory Methods in Neuroscience Research" part VII (http://link.springer.com/book/10.1007/978-1-4614-8794-4) and "Advances in Network Electrophysiology: Using MicroElectrode Array" (http://www.springer.com/la/book/9780387258577) both two are available from Springer. Also it could be useful these papers: Obien et al 2015, Bortolotto et al 2011, Buzsáki et al 2012, Mathis et al 2011, Spira and Hai 2013, Steidl et al 2006 and maybe Neves et al 2008. All of them give you relevant information about hippocampal electrophysiology study with MEA devices.

Best regards.

Dear David,

Thank you very much for the excellent information provided.

Very helpful resources for our electrophysiological studies using MEA's.

Thank you so much.

Best Regards,

Grandhi V Ramalingayya

Dear Dr. Ramalingayya,

I am glad to help you. I have no much expertise in this topic, however, for any question about this topic you can contact by mail and maybe I will be able to help you.

[email protected]

Best regards, and good luck.

Jesús D. Urbano Gámez

Dear David,

Very Good Morning...

Thanks for your help and support in validating the hippocampal slice electrophysiology studies in our laboratory.

Would you please share your experience of working with any cholinergic agonists (carbachol) or any other spike potentiating agents using hippocampal slice electrophysiology.

Also, I have noticed that, spontaneous response is visible some times specifically in CA3 region (with one experiment) and some times in CA1 region stratum pyramidale layers (with another experiment) .

Have you noticed similar findings in this regard? Would you please help us what specific information we can get with region specific spontaneous response.

Thanking you,

Best Regards,

Grandhi V Ramalingayya

Dear Dr. Ramalingayya

I haven’t worked with spike potentiating agents, so I can’t give any help in relation with this topic. About spontaneous activity, it is associated with viability of tissue, it could appear over all the hippocampus without any significant difference. In my experience CA3 and dentate gyrus usually show more spontaneous activity than CA1 region but it could be only due to viability. You may analyze the synchrony of spontaneous response between electrodes in order to associate it with specific circuit connections and also you can analyze frequency to discern specific patterns of shooting (rhythms), but I have no experience in this type of analysis. You must see specific bibliography about that. Sorry I can't provide any further help.

Best regards

Jesús D. Urbano.

Dear David,

Thanks for your response.

Would you please help me in MCS analysis part.

We tried to export data from recorded files, it is saving as .dat file format. We installed Spike2 (version 7.18) from CED.

Please, let me know which software, you are currently working for various analysis methods and whether it will come along with MCS system software or not.

Thanking you,

Best Regards,

Grandhi V Ramalingayya

Dear Dr. Ramalingayya,

If you are recording with Multichannel Experimenter software and you want use Spike 2 for analysis you must convert files to CED with Mutichannel Data Manager (export to CED). If you are using MC_rack software, I am not sure, but I think you can use the same software for analysis. You can also generate HDF5 files to open them with Matlab and if you export to ASCII you can use Microsoft Excel to open files. You can see that information in http://www.multichannelsystems.com/software/multi-channel-datamanager. For spiking analysis you could analyze it with Multichannel Analyzer software using spike explorer tool (https://www.youtube.com/watch?v=f8y5JP-9uFw).

Best regards.

Jesús D. Urbano

Dear David,

Thanks for the suggestions given.

We are recording the streams using MC_Rack software only. With Spike 2 software, .dat format files that were exported from MC_rack are not getting opened.

 We have not used Multichannel Experimenter tool yet. I heard that it is easy to use.

Would you please provide where we can get information on operating Multichannel experimenter tool to record the data streams as well as analysis.

Also, can we set different analysis parameters for on line and off line analysis of the same data streams?

Please let me know your experience in this regard.

Thanking you,

Best Regards,

Grandhi V Ramalingayya

Dear Dr. Ramalingayya,

Here you will get all the information available of Multichannel Experimenter software, http://www.multichannelsystems.com/software/multi-channel-experimenter the online analysis is being developed yet and in the version which I have worked with, there aren't a set of tools for online analysis as complete as MC_rack, probably MC_rack would work better. In relation with the problem opening .dat files with Spike 2 you can try to export with Multichannel Data Manager the files from MC_rack, I know these files generated with Multichannel Data Manager can be opened with Spike 2. About your question, in Multichannel experimenter you can record raw data and data from analysis tools, if you are recording raw data it should be no troubles to perform offline analysis different from the online one.

Best regards,

Jesús D. Urbano

Thank David....

Thanks a lot for the information provided and your help.

Best Regards,

Grandhi V Ramalingayya

Dear Grandhi,

Not sure if you still need help with the MEA system, but I'm using the same from Multi Channel Systems, for mouse hippocampal slices. See my recent paper (Schreurs et al. 2017, Sci Rep 7(5157)).

We use ice-cold, standard ACSF for dissection and cutting, but do this with a chopper; it's crude but very fast (

Dear An,

Thank you for sharing your experience with MEA system recordings with hippocampal slices and also your recent work published with helpful information.

We have high failure rate with slices made from tissue chopper as we could not get evoked signals. Then we started using Vibratome and got good healthy viable hippocampal slices from young Wistar rats.

We hardly find people (got one response on RG from Jesús D. Urbano almost an year ago) who are working on hippocampal slices using MEAs.

I have just gone through your publication. You have used patch clamp recordings as well to assess mIPSCs.

Hope you wont feel unhappy with my detailed questionaire. Sorry for a lot of following questions.

How many slices you are considering in general for statistical analysis?

Is it like, young animal's dorsal hippocamapal slices show better response than ventral ones' for LTP recordings?

Would you please let me know, whether you are using new software tools from MCS, i.e. Multichannel Experimenter or Multichannel Analyser to record and do the analysis of slice data streams.

We switched to these new tools as it has been informed by MCS that, they are not going to develop MC_Rack further.

Would you please let me know, whether you have tested any spike potentiating effects of any cholinergic agonists, like carbachol in CA1 region of hippocampal slices.

Would you please also share your experience with LTP experiments on hippocampal slices.

We are facing issues with slice placement in constant position facing MEAs as the slice size and shape may vary from day to day. would you please help me in this regard.

Thanking you once again, for the helpful information provided with your paper.

Best Regards,

Grandhi V Ramalingayya

Dear Grandhi,

Sorry for my late reply, I only saw your post now. I'm happy that my paper is useful to you! Here my suggestions for your other questions:

- For statistics, we usually aim to have results from at least 7 animals. I did a power analysis and 6-7 turned out to be enough for the typical big LTP differences that we observe across groups. That said, you're not always successful in obtaining a stable baseline recording and LTP induction, or due to technical problems the signal can suddenly be lost, so we usually start with ~10 animals per group. For analysis, each n is 1 animal, so I only included a single dorsal and/or ventral slice per animal. My recordings are long-lasting (LTP for at least 2-3 hours), so usually I can only do a full experiment for a single slice per animal anyway. This is different for patch clamp experiments, where you can record several cells/slices per animal. My colleague Victor Sabanov performed the patch clamp recordings and pooled results from multiple cells of the same animal, but used never more than 3 cells for dorsal/ventral per animal.

- Indeed, in my experience, it is easier to get signals from the dorsal slices and they are usually bigger (both in young and older animals) - at least in the DG. But my colleagues who work on CA1 (using sharp electrodes) also focus on dorsal slices. In chopped slices, the cell layers are more easily distinguishable in dorsal than in ventral HC, so that could also be one of the reasons. In the DG, I get a much higher LTP in ventral slices, but the opposite has been observed in CA1 (see the discussion and references in my paper).

- I'm not using the new MCS software tools, but simply MC_Rack and MC_DataTool to merge or convert files. Noteworthy, I'm focusing my analysis on a single channel (closest to the stimulation electrode). I have considered using more advanced analysis programs, as I'm sure there is more to get from the multichannel data, but so far I haven't found the real need nor time... I have also tried a demo of "LTP-Director", MCS's new software for LTP experiments, and it's more user-friendly but less flexible. Since it took me many months to understand MC_Rack and get it to work, and I'm so used to it now, I won't change the analysis software for the final year of my PhD :) How do the new MCS tools work for you?

- No, sorry, I haven't worked with any cholinergic agonists so cannot advice on that. I'm not doing any slice pharmacology; my main topic is dorsoventral synaptic plasticity and its impairment in Alzheimer's mouse models.

- What do you mean with placing the slices in constant position? Do you want to use the exact same stimulation and recording electrodes for each slice? This is indeed difficult. I place every slice in the same orientation in the recording chamber (turning it around with a pipette tip) and then add a harp slice grid (ALA Scientific Instruments) on top to immobilize it. When necessary, I very gently move it around until the region of interest is close(r) to an electrode. When the signals are quite small, I try alternative stimulation electrodes or move the slice a bit. If none of this works, I remove it and take another slice. In young mice, usually the first slice already gives me nice responses, while in older mice it can sometimes take several slices.

I hope this helps. Let me know if any other questions come up! :)

Best,

An