Question
Asked 15th Aug, 2020

How to depolarize cultured primary neurons with KCl without them dying?

Dear colleagues, I’ve started working with cultured primary neurons and came across a problem.
I need to depolarize neurons for different time intervals (up to 1 hour) and then use them for an assay 2 days after. The problem is that most of them are dying after depolarization.
I culture neurons in complete neurobasal medium (Neurobasal + 2% B27 supplement + 1%Glutamax +1% penstrep) with 1/3 media change every 3 days. I depolarize them on DIV11-14 by swapping media on Tyrode’s solution (45mM KCl) for up to an hour. Then I wash the cells with Tyrode’s solution (5mM KCl) twice and swap the collected media back.
45mM K+ Tyrode’s: 100 mM* NaCl; 45 mM KCl; 1 mM MgCl2; 1.8 mM CaCl2; 1.04 mM Na2HPO4; 26.2 mM NaHCO3; 10.9 mM HEPES; 10 mM D-glucose
* NaCl is used to adjust osmolarity of the solution, so concentration varies.
Since in the current setup I depolarize cells in 5%CO2 incubator I used buffering formula of Neurobasal media. I adjust the pH to Neurobasal’s pH=7.7 and checked that in CO2 incubator it equilibrates to pH=7.4. And I adjust solution’s osmolarity to match the current neuron’s medium too.
Also, I depolarized neurons in live cell imaging using GCaMP6s to monitor calcium elevation and after minutes I can see that some neurites are destroyed (Fig.1, attached) and after 0.5-1 hour cells don’t look good and most of them die afterwards (Fig.2).
At the same time, I keep seeing papers with no explicit details on solution and osmolarity where cultured primary neurons are stimulated with KCl for hours. For example, here 6 hours of 55mM KCl (https://www.nature.com/articles/nature09033).
I guess there are a lot of people routinely working with primary neuronal culture. Could you please help me, what am I missing?

Most recent answer

Felix Lange
Max Planck Institute for Biophysical Chemistry
Hi there,
maybe you already solved your problem. In case you did not, here is my thought. I did some depolarization experiments myself and encountered the same problem (rat hippocampal neurons). Later, I found this publication addressing the high cysteine concentration of Neurobasal formulation leading to excitatory death of neurons:
So what I did in my experiments is to split the supernatant of the cells and use 1 half spiked up with my desired final concentration of KCL and the second half to replace the depolarization solution after the intended period of "stimulation". In my case this did the trick! Also, for any refeeding or change of medium I only used the first 10 DiV to do so with Neurobasal medium. I hope this might be helpful
1 Recommendation

All Answers (14)

Manoj kumar Jaiswal
Turn Biotechnology Inc.
1-hour incubation is too long and effects in non-reversible. It completely depolarises the cell and possibility of reversal in none. We use for short acute exposure of KCl (30s-1min). See the paper here.
Max Karasev
University of Helsinki
Thank you Manoj kumar.
I agree with you, I did timelaps imaging with only 5 minutes depolarization, these effects started to occur but they were reversible in contrast to 30 min depolarization.
Well, I just wonder if longer depolarization is possible without neuron morbidity.
Max Karasev
University of Helsinki
Hi Marc,
Thank you for the advice!
I need this setup to develop a genetically encoded biosensor. I'd like to use the simplest method of pharmacological neuronal activation for this. So far, indeed, it seems that toxicity is caused by high potassium concentration. I try to use lower concentrations now and BIC/4-AP stimulation.
What wonders me is evidence in the literature that prolonged depolarization with high potassium concentration is not toxic to neurons. For example, in this paper (https://doi.org/10.1016/0006-8993(95)00552-2) authors specifically assessed viability 24h after treatment. 1hour of 50mM KCl was almost non-toxic.
1 Recommendation
Damian J Williams
Columbia University
Hi Max,
Have you done a control experiment using 5 mM K Tyrodes rather than the 45 mM one? It might be worth checking that the process of changing the media back and forth isn't damaging the neurons.
Max Karasev
University of Helsinki
Hi Damian,
Yes, I've observing neurons in 5mM K+ Tyrode's for 1h and they looked fine.
I'm not sure about media change. I know that substituting the significant part of media with fresh one is damaging to the neurons. Also, I recently found an interesting paper (doi.org/10.1371/journal.pone.0025633) where authors observed neuronal toxicity caused by neurobasal media itself. They claimed that this is excitotoxicity caused by L-cysteine which concentration in the commercially available media is, for some reason, 26x higher than in the originally described recipe.
I think the changing of conditioned neuronal media back and forth should inflict minimal damage. I can’t see any other option when using chemicals for stimulation.
Also, I think I’m making some progress. Lower potassium concentrations (25, 35mM) significantly increase calcium with much less toxicity. So probably that will do. I need to do more tests to be sure.
Damian J Williams
Columbia University
It's never easy is it :) Glad you are making some progress. I use 30 mM KCl when stimulating neurons for Ca imaging, and with even a 2 s exposure get a robust Ca transient. I guess you could estimate the required concentration by using GHK equation and activation potential of the Ca channels (if you are aiming for Ca influx). Good luck!
Federico Zampa
The Scripps Research Institute
As also suggested before by others I would definitely consider using Bicuculline (40uM), it's more physiological, way less toxic and reversible if you wash it out. It's also normally used to induce homeostatic downscaling, where it is left for 48h in the medium. A lot of people used KCl so you will find many protocols, but many good scientists advised me against it because of its toxicity. For depolarizing neurons people use also media change (see FLARE form Alice Ting's lab). Good luck!
Max Karasev
University of Helsinki
Thank you for the advice, Federico.
Mateus Silva
Yale University
Hi Max,
I am not fully aware of the details of your work, but have you tried to 1) incubate neurons with KCl plus antagonists of NMDA and non-NMDA receptors? Both are responsible for apoptotic signaling in neuronal cells, maybe you could prevent the damage by blocking them; 2) Add an antioxidant to your incubation medium, such as glutathione (GSH). Neurons are highly susceptible to oxidative damage and KCl-evoked depolarization enhances this. Maybe the addition of an antioxidant can protect your neurons.
Mike Watson
Neurona Therapeutics
I use a media similar to your's to test neurotransmitter release. 30 min 90 mM KCl treatment in the same growth media that the cells are natively in. They appear to be fine after switching back to normal growth media after though.
Mohamed Khashan
Freie Universität Berlin
Max Karasev is there a specific type of kcl to induce depolarization? In other words, in research used kcl for depolarization, they didn't mention any catalog number. When I looked for firms sell kcl I noticed there are many types of kcl so I got confused.
1 Recommendation
Felix Lange
Max Planck Institute for Biophysical Chemistry
Hi there,
maybe you already solved your problem. In case you did not, here is my thought. I did some depolarization experiments myself and encountered the same problem (rat hippocampal neurons). Later, I found this publication addressing the high cysteine concentration of Neurobasal formulation leading to excitatory death of neurons:
So what I did in my experiments is to split the supernatant of the cells and use 1 half spiked up with my desired final concentration of KCL and the second half to replace the depolarization solution after the intended period of "stimulation". In my case this did the trick! Also, for any refeeding or change of medium I only used the first 10 DiV to do so with Neurobasal medium. I hope this might be helpful
1 Recommendation

Similar questions and discussions

Why do my primary neurons disintegrate in my cultures?
Question
8 answers
  • Dan PiranerDan Piraner
Hello,
I'm trying to culture primary cortical neurons from C57J mice (E14-15). I get nice looking cultures (in my opinion - I'm new at this) for the first 5 days. Then the processes start to disintegrate.
I plate in 6 well format. For the prep, I wash the plates with HEPES and 1x PBS and then incubate overnight with 20 ug/mL PEI. I then wash the plates thrice with 1x PBS immediately before plating dissociated neurons. I tested PEI (as advised by other ResearchGate threads for optimizing neuronal adhesion), PDL, and laminin - PEI worked by far the best to reduce neurosphere formation in culture. I extract the cortices, treat with papain, dissociate via trituration/centrifugation, and plate 400,000 live cells per well as indicated by Trypan Blue. The media in each well is 1 mL NeuroBasal (LifeTech) / Gentamycin (20 ug/mL) / 1x Glutamax / 1x GS21. I change the media one day after plating, and every 2 days thereafter, keeping the volume 1 mL.
On the morning of DIV 2 the cells all look round or ovoid but by the end of the day they start reaching out to each other. By day 3 I have a really nice looking filamentous network with mostly well-spaced neurons and only a few clusters/neurospheres. However, on the evening of day 5 the neurons show hints of granulation. By morning of day 6 the neuronal processes are highly granulated (image attached) and by day 7 the connections between granules are lost, leaving behind disintegrated blebs.
Does anyone have any suggestions to prolong the life of the neurons? I could change my protocol to transfect on DIV2 and image on DIV4 but I'd like to be able to take time points farther out. Any tips?

Related Publications

Article
Rho family GTPases have been implicated in cytoskeletal reorganization during neuritogenesis. We have recently identified a new gene of this family, cRac1B, specifically expressed in the chicken developing nervous system. This GTPase was overexpressed in primary neurons to study the role of cRac1B in the development of the neuronal phenotype. Overe...
Article
Full-text available
Neurites, both dendrites and axons, are neuronal cellular processes that enable the conduction of electrical impulses between neurons. Defining the structure of neurites is critical to understanding how these processes move materials and signals that support synaptic communication. Electron microscopy (EM) has been traditionally used to assess the...
Got a technical question?
Get high-quality answers from experts.