Monday, January 30, 2012

Listening to cells spike with your phone

In 1992, Linder and Palka proposed using a very simple cockroach leg preparation to teach students about neuroscience; specifically to teach them about the most visible electrical property of the neuron, the action potential.  The company Backyard Brains expanded and modernized this idea by creating the SpikerBox, an inexpensive, portable way to record action potentials. And now you can download their app for your smart phone!

A quote from their download site explains it all:
"When was the first time you heard the sound of a brain cell? Was it in elementary school, high school, or college? Chances are your answer is "never," as we neuroscientists have typically had to go to major research universities and use expensive equipment to learn about and investigate the brain.
With the SpikerBox (available open-source from you can now listen to living neurons from insects. With this android app, you can view what the spikes of neurons look like and record it to your phone for later analysis. What used to take up a whole room you now have in your hand, and for cheap. Think of it!"
I have not tried this app for the phone or the spikerbox, so I cannot speak from experience, but I did visit their poster (the most interactive 'poster' I have ever been to) at 2011 Society for Neuroscience meeting. There they demonstrated the cockroach leg preparation and the simple use of the spikerbox. 
Here is their commercial that shows how the spikerbox works:

I am so on board with the mission of this company, to make basic, cellular-level neuroscience accessible for everyone.

One more thing: I have never blogged about a company before, so I just want to let everyone know that I am in no way affiliated with, and have no commercial interest in Backyard Brains, I just think that they are really cool. 

Readers, I am curious have any of you tried this? seen it? what did you think? Linder TM, & Palka J (1992). A student apparatus for recording action potentials in cockroach legs. The American journal of physiology, 262 (6 Pt 3) PMID: 1616063

Thursday, January 26, 2012

The "Human Neuron", not so special after all?

Von Economo neurons, a set of neurons classified by their elongated, 'spindle-like' shape, were once thought to belong only to humans and great apes. This uniqueness, as you might imagine, encouraged extensive speculation about what this neuron does.  Do they make you smart? Do they process emotions? social cues? future planning?

Not that extensive speculation is a bad thing, it's just that it is easy to jump into the deep end and assume that because something is unique to humans, it is what makes humans unique.

You may have guessed where this is going.  It turns out that this type of cell is not unique to humans and great apes.  It is found in whales, hippos, zebras, manatees, and elephants.

Figure 8. Butti et al., 2011

This doesn't mean that the cells don't process emotions or social cues, it just means that they are not unique to humans.

There are studies showing that Von Economo neurons are reduced in post-mortem human brains afflicted with certain diseases, such as schizophrenia and autism (reviewed in Butti et al., 2011)

But according to Butti et al.
"It is important to note that, despite the previously discussed hypotheses on the possible functional role and implication of VENs in neuropsychiatric disorders, direct evidence has yet to be found."
These cells are obviously difficult to study because no 'research animal' has them.  As far as I am aware, no one knows where they project in the brain or what inputs feed onto them.  And I don't think anyone has been able to investigate their spiking patterns. 

The real question that these cells bring up for me is 'what does it mean to be a class of cell'? why does it matter that these cells have elongated spindle-like bodies, when they are chemically similar to some pyramidal cells in the mouse brain. Butti et al. point out that VENs have specific peptide expression of GRP and NMB, and that pyramidal cells in the mouse cortex also express  GRP and NMB.

If we define type of neuron by chemical signature, then these are not necessarily  unique cells. It is only their specific shape which is special and even that is common to many large animals. Maybe this type of neuron is expressed in all mammals, but in the large ones it needs to have a large spindle shape to work more efficiently. 
source Butti C, Santos M, Uppal N, & Hof PR (2011). Von Economo neurons: Clinical and evolutionary perspectives. Cortex; a journal devoted to the study of the nervous system and behavior PMID: 22130090

Wednesday, January 25, 2012

Neurons are like Power Cords

Neurons are like power cords because the work by communicating electrical signals from one place to another. 

A neuron has 3 main parts, the dendrites (B), the cell body, or soma (C), and the axon (A).
 The dendrites receive the input from other cells.  This input causes an electrical change in the cell, either pushing the cell's voltage upwards (exciting the cell) or downwards (inhibiting the cell).  That signal has to travel down the dendrite to the soma (C).
The dendrites act like leaky power cords, and the strength of the signal decays during its journey to the soma. 
In certain cells, the shape of the dendrite contributes to how it integrates multiple signals and how strongly it sends these signals to the soma.

The axon is responsible for sending out information from the neuron. It is usually longer that the dendrites because it often is carrying a signal from one brain region to another, from one hemisphere to the other, or even from the sensory neurons (like touch receptors) to the spinal cord.
When the axon is very long, it can't work if it is leaky. It will often have a myelin sheath that insulates* it and allows it to propagate its signal with great efficacy and speed.

Some ways that neurons are NOT like power cords are:
  1. Power cords conduct electricity through the jumping of electrons, the current in neurons is due to the movement of ions (like sodium (Na+), potassium (K+), calcium (Ca2+), and chloride (Cl-)), and is not as fast as direct electron movement.
  2. Power cords conduct huge amounts of electricity compared to neurons. The potentials relevant for neurons are on the millivolt scale, and the currents are measured in picoAmps. You could never be shocked by touching a neuron. 

*it is more complicated than just acting as insulation, see Nodes of Ranvier if you really want to know how it works.

Monday, January 23, 2012

The cells that make us eat: Part 2

In the last post, we discussed the finding that stimulating the AgRP neurons in the hypothalamus directly causes mice to eat.  You can see the video of the mouse eating with the light stimulation here.

Today we will look at a follow up paper by the same group.  This paper looks at the mechanisms that might naturally stimulate these neurons.  As the authors mention in the discussion, the origin of the pathways that naturally cause these neurons to fire is not known. (as in, the part of the brain that sends the main signals to these neurons is still a mystery)  However, they can investigate what is happening at the junction between these pathways and the AgRP neurons, that is, at the synapse.

Using brains from mice that are either hungry or full, the researchers found that in hungry mice, the AgRP neurons recieve more synaptic input. 

But why?

There are a few ways this could happen.  One possibility is that the upstream neurons are firing more frequently because they are receiving more input from other pathways. But since it is not known where these neurons are, that is difficult to test.  Another possibility is that the output ends of these neurons, the part which arrives at the synapse, releases neurotransmitter more easily in hungry mice than in full mice. 

The research in this paper supports option number 2, that the hunger signal modulation occurs right at the synapse.

The researchers find that through a complex molecular pathway, involving ghrelin, leptin, and opioid signalling, the neurotransmitter release at the synapse is regulated by the animal's hunger state. Although too detailed to fully summarize here, the paper presents a satisfying and thorough explanation for how the AgRP neurons (the ones that cause eating when stimulated) could be modulated by hunger signals from the body. 

One of the nice parts about this explanation is that it avoids the 'never ending chain of neurons' problem, where activity in one neuron is caused by a neuron stimulating it, which in turn is caused by a neuron stimulate that one, which is caused by... you get the point. 
This paper strongly suggests that dynamic modulation of eating behavior can happen right at the synapse. Yang Y, Atasoy D, Su HH, & Sternson SM (2011). Hunger states switch a flip-flop memory circuit via a synaptic AMPK-dependent positive feedback loop. Cell, 146 (6), 992-1003 PMID: 21925320

Friday, January 20, 2012

The cells that make us eat: Part 1

It is always exciting when a specific behavior can be directly linked to particular neurons. 
In this case, eating.  In March 2011, a paper came out from the Sternson lab at Janelia Farm explaining that when certain neurons (AgRP) in the mouse hypothalamus were stimulated with light, the mouse would spontaneously start eating.  The mouse would pretty much keep eating (except for water breaks) until the stimulation stopped.  What's even more interesting is that the neurons right next to these (POMC) had pretty much the opposite effect.  When they were stimulated, the mouse didn't eat much and over time lost weight.

This may seem like the basis for the next miracle diet,  find a way to stimulate POMC neurons and suppress AgRP neurons, right?

Unfortunately it's not that easy.  There is a lot more work to be done.  In fact this is just the beginning of understanding the hunger circuit.  Sure stimulating the neurons directly with light causes eating, but what naturally stimulates those neurons? 

Neurons fire when they recieve signals from other neurons which in turn fire when they receive signals from other neurons.... and so forth in a never ending chain.

So, where does the 'must eat' signal ultimately come from?
where is the beginning of that neural chain?

We will investigate this a little more in Part 2, when we look at a follow up paper. Aponte Y, Atasoy D, & Sternson SM (2011). AGRP neurons are sufficient to orchestrate feeding behavior rapidly and without training. Nature neuroscience, 14 (3), 351-5 PMID: 21209617

Monday, January 16, 2012

Neurons activated by light

Optogenetics is a relatively new technique that inserts a light-sensitive channel or pump into a cell that would normally not be light sensitive. This channel can be inserted into the neurons of a mouse, worm, fly, rat, or whatever (has not been tried on humans yet to my knowledge).
Then, when a light is flashed onto that neuron, it activates, much like it would if it had just been stimulated by electricity or the normal neuronal pathways leading to it. 

What is so great about this is that with tricks of genetics, a researcher can get this channel to express only in one kind of cell, making it so that cell1 is activated when the light comes on, but cell2 is not. 

Previously if you wanted to stimulate neurons in a living brain you had to insert metal electrodes which would stimulate all the cells within a certain radius.  Optogenetics gives cellular specificity, where metal electrodes don't. 

Some great discoveries are already being made using this technique:

Which neurons cure the effects of Parkinson's Disease during deep brain stimulation (Gradinaru et al., 2009)?

Which neurons make a mouse eat (Aponte et al., 2011)?

And many more.

Though this is not a technique that is meant to be used on humans for disease treatment, metal stimulating electrodes are already being used to treat some disorders (parkinson's see here and here. and depression).  How would the ethics of optogenetic treatments differ? Instead of a metal electrode, a genetic alteration would have to take place in some neurons and a fiber optic cable would have to be implanted. 

Saturday, January 7, 2012

Neurons are like Snowflakes

One of the first things that people noticed when looking at the brain on the cellular scale is that neurons are shaped differently from other cells.
Drawings like this one by Ramon y Cajal show that neurons are not only shaped differently from say, blood cells, but also shaped differently from each other. (I am not going to give you a history lesson, but you can read all about Ramon y Cajal and his famous drawings here.) And if you are really lost and don't know what a neuron is, check out neuroscience for kids.

Now there are much more sophisticated ways to analyze the shape of neurons. Software such as Neuromantic allow scientists to digitally reconstruct neurons and quantitatively analyze their shape. Once a neuron is digitally reconstructed, it can be deposited in a database for everyone to use. You can browse at least 7,000 neurons at, visualize them in 3D, and even analyze information about how long the dendrites are, how many dendrites branch off, and how much area is in each compartment.

The interesting thing about the shape of neurons is that though they might all be unique (like snowflakes), they can be classified into any number of categories based on their details.  There are certain classes of neuron that could never be confused with one another.  for example, the Purkinje cell of the cerebellum has a distinct sea-coral shape...
purkinje cell
sea coral

cortical pyramidal cells
 ... while the Pyramidal cells of the cortex have a bushy basal arbor of dendrites and a long vertical apical dendrite. 

There are two main types of question currently being investigated by researchers:

1. what genes and molecules make neurons grow into a certain shape
 (review: Libersat and Duch, 2004)? Similarly, how are neuron shapes altered in neurological diseases?


2. What does a neuron's shape have to do with it's function?
What sort of information it can receive and communicate when it has a huge flat dendritic arbor (like the Purkinje cell), or when it has a long narrow dendrite (like the Pyramidal cell)? Or could many different neuron shapes function the same way and these differences are just accidents of evolution (as a snowflake's shape is not really for a purpose, but an accident of condensation)?

Feel free to leave answers, questions, and wild hypotheses below in the comments section.

Tuesday, January 3, 2012


Hello and welcome to The Cellular Scale.

This is my first post, and I would like to explain why I am creating this blog.

There are some great blogs out there that discuss neuroscience for both scientists and lay people, and I am planning to have links to them soon. I thoroughly enjoy these blogs, but find that they generally focus on the higher level of neuroscience, such as behavioral studies and human fMRI scans. While this level of neuroscience is interesting to everyone, some people might want to focus a little closer in. 

With The Cellular Scale, I am hoping to bring cellular level neuroscience to the blog world and to start discussions between scientists and non-scientists regarding neurons and the way they work.