I unabashedly overstate that to understand the brain we need a wiring diagram of it. This is the goal of the burgeoning field of connectomics – to generate an all encompassing brain wiring diagram or connectome. Lately though, after some heated discussions, I’m starting to realize that without information on how the cells in the network function, even the most detailed map of connections might not be all that useful. It was this caveat of connectomics that prompted a pair of papers published in Nature last month. Both papers studied the mouse visual system, one the retina, the other the primary visual cortex (where early visual processing occurs). I review these papers here and here, but to understand the game-changing combo of techniques used in these papers you may need a crash course on one or all of the following items:
1.) Calcium Indicators: Calcium indicators are special dyes that can be used to monitor neuronal activity. Neuronal activity results from ion fluctuations either into or out of the cell. One ion that enters the cell during activity is calcium. When a calcium indicator dye that has been loaded into cells encounters an influx of calcium, the dye glows brighter. Since calcium enters neurons when they are active, calcium dyes glow brighter when neurons are active, and so they allow you to track the activity of neurons.
Here is and example of calcium indicator imaging of waves of neuronal activity propagating through a neural network in a model of epilepsy:
2.) Electron Microscopy: Is an old (1930’s) but powerful type of microscopy that allows for very high resolution micrographs, much higher than traditional light microscopy. Rather than using a beam of light (photons), electron microscopes use a beam of electrons and special techniques for detecting that beam. The result is a microscope that, according to wikipedia, can magnify up to 10,000,000 x, while light microscopy only gets up to 1000’s of times magnification.
3.) Visual Stimulus Selectivity: This is the ability of neurons in the visual nervous system to detect specific visual stimuli, to exclusion of others. The most basic type of selectivity you will hear about is orientation selectivity. A neuron that is orientation selective will only respond when a line of a certain orientation is presented in a certain part of the visual field. This type of selectivity was discovered by David Hubel and Torsten Wiesel, in a rather ironic way that you can hear Hubel talk about below (the popping sound you hear in the background is the readout of neurons becoming excited). Another type of selectivity you’ll read about in the coming posts is direction selectivity, which refers to neurons that fire only when a moving line or dot is present in a specific part of the visual field, and only when it is moving in the right direction.
The general form that both of these papers take runs something like this: Both groups start with calcium indicator imaging in live neurons in either the brain (visual cortex) or the retina. The reason why these groups are working in the visual system is because neurons that deal with visual stimuli respond to very specific stimuli. For instance, in the retina, retinal ganglion cells respond to things like color, or points of light in a specific part of the visual field or lines moving in specific directions. Cells with similar preferred stimuli exist in the primary visual cortex as well. In both papers, the authors chose stimuli to present to the eye, and then using calcium indicator imaging they found neurons that responded to a range of preferred stimuli. They now had information on what role these neurons play – this constituted their functional data.
Once they had their functional data, both groups sliced the pieces of brain or retina that contained their functionally imaged neurons into extremely thin pieces and used electron microscopy to take images of each slice. They then they used the resulting huge data sets to make 3D reconstructions of the cells and circuitry in the volume that contained their functionally imaged cells. The result was a wiring diagram that included their functionally characterized cells as well as the cells received input from and made input into them. Using this diagram both groups – whether successfully or unsuccessfully – made conclusions about how the functionally imaged cells work in the network.
In specific, Bock and friends tried to determine whether orientation selective pyramidal neurons synapse onto inhibitory interneurons in such a way as to make those interneurons orientation selective too?
On the other hand, Briggman and friends kind of do the opposite, asking how input from inhibitory starburst amacrine cells makes direction selective retinal ganglion cells direction selective.
Both of these matters are rather complex, so if you are interested in how these groups undertook their endeavours, have a look at the In Depth critiques (linked directly above).
Regardless of the particulars though, both of these papers involved heroic efforts and a result, for the first time, we get to see functional data combined with electron microscopy in the central nervous system. Keeping in mind that solving the brain will rely on reconstructing cellular circuitry and integrating the resulting network diagram with knowledge of how the cells function in that circuitry, this new step toward functional connectomics is an extremely important one. You can count on seeing studies like this pop up more often in the near future.