Action Potential Of Nerve Cell

Understanding Action Potential Of Nerve Cell: The Practical Side

The action potential is just a voltage spike that travels down an axon. It happens because sodium and potassium ions move across the membrane through specialized protein channels. That's it. But getting that simple description right enough to actually work with electrophysiology data, simulate it, or troubleshoot a prep that won't fire properly is where things get complicated. When I first started working with intracellular recordings, I assumed the resting membrane potential was a fixed number. It's not. In a real neuron, it drifts. I spent three weeks trying to figure out why my patched hippocampal cells kept hyperpolarizing over the course of an experiment. Turned out the internal solution in the pipette was slightly different in composition from what the literature said, and the Donnan effect was slowly shifting the equilibrium. Not dramatic, but enough to throw off spike timing measurements if you're looking at sub-millisecond precision.

How the Action Potential Of Nerve Cell Actually Works

Start with the resting state. The membrane sits around minus 65 millivolts in a typical mammalian neuron. Potassium leak channels keep it there because potassium wants to leave the cell, and the sodium-potassium pump maintains the concentration gradients that make that possible. About three sodium ions get pumped out for every two potassium ions brought in. ATP dependent. Continuous. If you block that pump with ouabain, the whole thing falls apart over minutes, not seconds. Then a stimulus arrives. Whatever excites the neuron sends some sodium ions through voltage-gated channels. If the depolarization reaches threshold — usually somewhere between minus 50 and minus 40 millivolts depending on the cell type — those voltage-gated sodium channels flip open in large numbers. Sodium rushes in. The membrane potential climbs rapidly toward the sodium equilibrium potential, which is around plus 60 millivolts. You see the upstroke of the action potential here. The sodium channels don't stay open forever. They inactivate. There's a separate inactivation gate, sometimes called the h-gate, that swings shut about a millisecond after activation. Meanwhile, voltage-gated potassium channels — the delayed rectifiers — are opening more slowly. They don't respond as fast to the depolarization. By the time they're fully open, the sodium channels have already inactivated. Potassium flows out. The membrane repolarizes. Often overshoots into a brief hyperpolarization, which is the afterhyperpolarization phase you see on a trace.

The membrane returns to rest through the leak channels and the sodium-potassium pump restoring ionic balance. The refractory period follows, first absolute — no amount of stimulus can trigger another spike because the sodium channels are inactivated — then relative, where a stronger than normal stimulus is needed because the potassium channels are still open and the membrane is hyperpolarized.

The Myelin Question Most People Get Wrong

I see this all the time in grad seminars. Someone explains saltatory conduction and says myelin makes the action potential "travel faster." That's true but incomplete. Myelin doesn't just speed things up. It changes the fundamental biophysics of how the signal propagates. Without myelin, you need voltage-gated sodium channels distributed along the entire axon. With myelin, they're concentrated almost exclusively at the nodes of Ranvier. The internodal segments are essentially capacitors — they store charge but don't generate action potentials. Here's the part people miss: the diameter of the axon and the thickness of the myelin sheath matter more than the number of nodes. A thick myelinated axon can conduct at 120 meters per second. An unmyelinated axon of the same diameter conducts at maybe two meters per second. The difference isn't just about resistance. It's about the space constant and the time constant of the membrane. Myelin increases membrane resistance and decreases capacitance, which means the depolarizing current spreads farther before it decays. It reaches the next node fast enough to trigger another action potential before the previous one has fully dissipated. If you're building a model of neural conduction, don't just scale the conduction velocity linearly with diameter. The Hodgkin-Huxley equations assume unmyelinated cable. You need the modified version for myelinated fibers, and even then the node spacing has to be empirically determined for your specific preparation. I once ran a simulation with default node spacing from a textbook paper and got conduction blocks at frequencies above 200 hertz. The model wasn't wrong. The node spacing was off by about 15 micrometers from what the actual tissue had.

Common Pitfalls in Recording and Measurement

Space clamp errors are the most frustrating thing you'll encounter in voltage clamp experiments on neurons. When you try to hold the membrane potential at a fixed level across an entire neuron, you're assuming the neuron is isopotential. It's not. Dendrites, axons, and the soma all have different electrical properties and different distances from the recording electrode. The command voltage you set might be perfect at the soma but completely wrong at the distal dendrites. Ion channels there are seeing different voltages than what you think. I had a setup where I was measuring sodium currents in a dissociated cortical neuron. The peak current looked clean, but the inactivation kinetics were weird. Slow. I checked everything — solutions, temperature, drug contamination. Nothing. Then I realized the series resistance was too high. Not enough to blow the clamp outright, but enough to cause a voltage error during the large sodium influx. The command voltage was actually dropping by about eight millivolts during the peak current. Eight millivolts is enough to shift sodium channel availability significantly. Lowering the series resistance by improving the seal took care of it immediately. Another thing: temperature matters a lot more than most protocols account for. Channel kinetics are temperature dependent with a Q10 of about 2.5 to 3 for most voltage-gated channels. Recording at room temperature instead of physiological temperature will make your action potentials look slower and broader than they actually are in vivo. If you're comparing your data to literature values from warmed preparations, you're not measuring the same thing.

What Happens When It Goes Wrong

There are conditions where the action potential mechanism itself breaks down, and they reveal something important about how fragile this process actually is. Hypocalcemia is a good example. Low extracellular calcium reduces the surface charge screening on voltage-gated sodium channels, which shifts the voltage dependence of activation to more negative potentials. Neurons become hyperexcitable. You see this in tetany — spontaneous, uncontrolled firing because the threshold is lower than it should be. The action potential fires more easily, but it's not necessarily healthier. It's just running at the wrong operating point. Hypercapping has the opposite effect on excitability through direct acidosis of the extracellular space. Protons block sodium channels from the outside. The action potential threshold shifts positive. Neurons become harder to fire. This is one reason why prolonged hyperventilation followed by CO2 retention can cause neurological symptoms — the pH shift affects channel function directly. Local anesthetics work by binding to the inner pore of voltage-gated sodium channels in their inactivated state. They preferentially block fibers that are firing rapidly because the channels spend more time in the inactivated conformation. This is why pain fibers, which fire at high frequencies during nociception, are blocked before motor fibers. It's use-dependent blockade. If you're simulating anesthetic effects, you can't just reduce sodium conductance uniformly. You need to model the state-dependent binding.

Why Beginners Shouldn't Start With the Classic Hodgkin-Huxley

The original 1952 squid giant axon model is elegant and foundational, but it's also narrow. It describes one specific cell type under very specific conditions. The ion channel densities, the kinetics, the resting potential — they're all calibrated for the squid giant axon at 6 to 16 degrees Celsius. Apply those parameters to a mammalian cortical pyramidal neuron at 37 degrees and the model produces garbage. Not interesting garbage. Boring garbage. The spikes don't look wrong in a way that teaches you anything. They just don't happen at all because the kinetics are completely mismatched. If you're learning to build or interpret action potential models, start with a modern multicompartment model from a database like ModelDB. Pick a neuron type you actually care about. Adjust one parameter at a time and observe what breaks. The parameter that, when changed by ten percent, causes the most dramatic shift in firing pattern is the one you should understand first. In most central neurons, that's the persistent sodium current or the slow potassium current, not the transient sodium current everyone focuses on. The transient sodium current generates the spike. Everyone knows that. But the persistent sodium current, which is a small non-inactivating fraction of the total sodium conductance, determines whether the neuron fires tonically or in bursts. Miss that and you'll spend months confused about why your simulated neuron either never fires or fires continuously regardless of input. I learned that the hard way with a thalamic relay neuron model. The burst mode depended entirely on a single conductance parameter that had a note in a 1998 paper saying "adjust to match in vivo observations." I ignored that note for two weeks.

Practical Notes for Working With Real Tissue

If you're doing patch clamp or extracellular recordings, the quality of your action potential data depends heavily on the health of the preparation. Older neurons, neurons from ischemic tissue, neurons that have been in vitro for more than a few hours — they all show degraded action potentials. The spikes get narrower, the afterhyperpolarization shrinks, and eventually the neuron stops firing altogether. This isn't just a technical artifact. It reflects real changes in channel expression and membrane composition that happen during stress and degeneration. When troubleshooting a bad prep, check these in order: the integrity of the bathing solution, the pH, the osmolarity, the glucose concentration, the temperature, and then the age of the slice or culture. The bath solution is the most common culprit. Even a one millimolar deviation in potassium concentration can change resting potential enough to alter spike threshold and frequency. One millimolar sounds small. It's not. The Nernst equation is logarithmic, but in the physiological range, small changes in external potassium have outsized effects on membrane potential because the membrane is relatively permeable to potassium at rest. The action potential is a well-understood mechanism at the biophysical level. What's not well understood is how it varies between cell types, how it changes with development and disease, and how the same basic mechanism produces such different firing patterns across the nervous system. That's where the actual work happens, and it's nowhere near as clean as the textbook diagram.