“it’s all about getting to the truth” – Karl Deisseroth. May 20th Nature Podcast

fMRI, it’s everywhere. From studies looking at how we separate fantasy from reality to identifying psychopaths. The technique offers a very powerful way analyse that most elusive of domains, the contents of another person’s brain. Even so the technology is not without it’s flaws, one flaw is the method of determining relevant data from noise. Another is that the most common signal studied with fMRI, blood oxygenation level-dependent (BOLD) signals, are only an indirect measure of neural activity. Further, it has been until now* an assumption with little empirical support.

To put this in proper perspective it is necessary to explain a little more fully what a BOLD signal is. Essentially the oxygenation level of the blood can be determined via the fMRI by discerning the difference in magnetic properties of oxygenated versus deoxygenated haemoglobin (the protein responsible for shuttling oxygen around our body and keeping us alive). Now, the link between blood oxygenation levels and brain activity is made like this: active neurons are performing cellular functions, cellular functions require energy, energy requires metabolism, metabolism requires oxygen, higher metabolism requires higher oxygen levels. Thus you can make a logical connection from an increase in neuron activity (representing brain activity) and an increase in oxygenated blood flow to an area of the brain.

Because neurons do not have an internal supply of either glucose or oxygen the chain of reasoning above is valid but not having direct empirical support is a potential weakness. To the rescue comes Optogenetics, this relatively new field concerns itself with engineering neurons in such a way as to allow them to be activated by pulses of light. I think you can see where this is going.

The paper “Global and local fMRI signals driven by neurons defined optogenetically by type and wiring” published in Nature details how this technique can help in fMRI work.  In this case rats were used as a model animal instead of humans. The procedure consisted of injecting the brains of the rats with a viral vector that was targeted to particular cell types in the brain, cortical neurons to be specific. This caused these cells to express a light sensitive protein that would in essence cause the cell to be activated**.

By specifically targeting the cell types of most interest this technique has shown that the BOLD signal detailed above really is correlated with neuronal activity. In reality, while this is an important result of the early implementation of this approach and provides a firmer foundation for the theoretical underpinnings of fMRI work it’s true power lies elsewhere.

In addition to activating the cells of a specific region of the brain this technique can also highlight larger networks that operate within the brain. Neurons do not fire in a vacuum, often one will trigger another which set off a third and so on in a cascading chain which results in thoughts or actions. By selectively activating areas of the brain researchers can then watch the downstream effects of those activations in other parts of the brain that were not directly stimulated. In this way we can effectively build a map of neuronal networks.

This approach will both stimulate new research and perhaps provide a method of validating conclusions drawn from previous work. This looks to be an important new tool for brain research, the full power of which we may not yet realise.

Now that’s exciting.


* For a given definition of “now”, which corresponds to earlier this year when the paper was published. I really need to get to this stuff faster.

** An ion channel protein, stimulation of which causes ions to flow through the cell membrane. Just like the process that occurs when a neuronal signal is initiated.

Lee, J., Durand, R., Gradinaru, V., Zhang, F., Goshen, I., Kim, D., Fenno, L., Ramakrishnan, C., & Deisseroth, K. (2010). Global and local fMRI signals driven by neurons defined optogenetically by type and wiring Nature, 465 (7299), 788-792 DOI: 10.1038/nature09108

Enhanced by Zemanta