Whenever neuroscientists want to talk about dendrites — the part of a neuron that typically is used for receiving information from other neurons — neuroscientists will almost always show a picture of a Purkinje cell. I’m sure you’ve seen it before, you know, those big cells with the giant, branching beautiful dendrites that look like this:
And that’s pretty much where things end. You never hear about what these peacocky cells do, where they live, how they act. But in fact they do a lot and are very unique among neurons.
Purkinje cells were discovered in 1837 by Bohemian (as in Czech, not as in Kerouac) anatomist Jan Evangelista Purkyně, who among other things, discovered that you could get high on nutmeg. As you can see from the picture above, Purkinje cells have a very large and branchy and practically two-dimensional (ie. flat) dendrtic arbor, studded with little things called spines. The dendritic arbor stems from a primary dendrite which emerges from a roundish cell body which has a single axon emerging from the other end. Purkinje cells are in a part of the brain known as the cerebellum, more specifically the cerebellar cortex, and in fact they are the only source of output for the entire cerebellar cortex. Purkinje cells send their axons to a region of the cerebellum know as the deep cerebellar nuclei, a dark and obscure place where you don’t want to go, but that’s another story. Purkinje cells sit nicely aligned to each other in one layer, with their dendrites forming these parallel arrays along the cerebellar cortex. This architecture is basically maintained in most vertebrates, and constitutes an evolutionarily old part of the brain.
One of the things that makes Purkinje cells unusual is that they are inhibitory, meaning that when they are active, their targets stop being active, or are inhibited. In the nervous system, for the most part, neurons that connect one area of the brain or nervous system to another, also known as projection neurons, tend to be excitatory. Meaning that they activate their targets. In contrast, neurons that connect to nearby neurons in the same brain region can be either excitatory or inhibitory; these are known as interneurons. Purkinje cells, are inhibitory projection neurons, which is unusual.
Another nifty thing about Purkinje cells, is that they have active dendrites. Meaning dendrites that can generate electrical impulses, or action potentials. Most basic neuroscience textbooks will tell you that axons generate action potentials to communicate with target cells, while dendrites passively integrate information from other neurons. However, Purkinje cell dendrites can also generate action potentials. Actually it turns out almost all dendrites can generate action potentials, but this was first discovered in Purkinje cells by neuroscientist Rodolfo Llinás and colleagues. Since Purkinje cells have such big-ass dendrites, they were able to record electrical activity from directly inside the dendrite, as well as from the cell body. What they observed was that the dendrites could generate action potentials, but in contrast to action potentials generated by axons, which are usually mediate by influxes of sodium ions into the cell, the dendritic ones were mediated by calcium. This was a remarkable finding that changed the way we thought of neuron function, and way cool. You can see their results below.
So why the big dendrites? Typically cells have big dendrites in order to receive information from many other neurons. The larger your dendritic tree, the more cells will tend to project to it. Purkinje cells recieve two types of excitatory inputs in their dendrites. Most of them come from a shitload of so-called granule cells. Granule cells are another type of cerebellar neuron which constitute half, yes half, of the total number of neurons in the brain. They have these axons called parallel fibers that make contacts throughout the arrays of Purkinje cells. Each Purkinje cells receive inputs from up to 200,000 parallel fibers, but each input by itself is very weak. The other type of excitatory input comes from what are known as climbing fibers. Climbing fibers are the axons from neurons which reside in another brain region known as the inferior olive. What’s cool is that each Purkinje cell receives input from only one climbing fiber, but this input is incredibly powerful. These two types of inputs thus activate Purkinje cells very differently.
When parallel fibers are active, Purkinje cells will fire action potentials at a rate proportional to the number of parallel fiber activated, yet when the climbing fiber is active, the Purkinje cell will generate a burst of action potentials occurring at a really high rate, and then will stop firing altogether. This is called a complex spike. If this is sounding a bit like a computer is because that’s what it is. The cerebellum is basically an error correction machine. Whenever you perform a motor action, such as picking your nose, your cerebellum is continuously adjusting your arm movement to make sure that your finger expertly reaches your nostril, and doesn’t, for example, poke your eye. Thus your cerebellum is constantly comparing the movement that you intended to do with what you are actually doing and makes little adjustments such that the two match as best as possible. Parallel fibers continuously set the gain of this adjustment, while climbing fibers are active when an error is detected. So without a cerebellum, you’d be mostly fine, but would be highly uncoordinated to say the least. And at the center of this computation is the Purkinje cell. This type of adjustment is also useful for mediating certain types of associative learning, such as classical conditioning, but that’s the topic for another post.
So to summarize: Purkinje cells are nifty because they have cool dendrites, are inhibitory projection neurons, generate action potentials in their dendrites, and allow you to pick your nose without poking your eyes out. And be thankful for that.