Do Fruit Flies Dream
of Electric Bananas?
Visualizing neuronal activity in small brains
over four dimensions
Some people are amazingly focused, energized, and attentive. Others always seem to have their heads in the clouds, dreaming the day away. One might think that our brains doze off and switch to a resting state when we let our minds wander, but research using functional magnetic resonance imaging (fMRI) to visualize brain activity has shown the reverse: brains are particularly active during these mind-wandering episodes.
One of the key brain networks active during the resting state is aptly termed the default-mode network (DMN). It is inactivated by task-related networks in a push-pull fashion. Perhaps not surprisingly, impairments in the ability to dynamically activate or inactivate the DMN have been found to be associated with attention deficits, depression, schizophrenia, Alzheimer’s, and many other psychiatric disorders. These findings are challenging one particular, dominant school of thought among neuroscientists—that brains are mainly reactive, computing adaptive behaviors from available sensory input.
The discoverer of the DMN, Marcus Raichle at Washington University in St. Louis, portrays the brain as an active organ, constantly producing spontaneous activity which is only modulated by incoming stimuli. It seems intuitive that brains are both reactive and active, depending on the circumstances. All brains react to external stimuli, and all brains anybody has ever looked at are also spontaneously active. Nobody knows what the function of the DMN is, and many different hypotheses are floating around. What if the push-pull relationship between the DMN and the task-related networks is exactly the way in which brains balance activity with reactivity? If we focus hard on a task, we suppress the DMN long enough to complete the task, but fluctuations in the suppressed DMN lead to variation in our task performance. In fact, just before we make mistakes, there often is an increase in DMN activity. One would imagine that optimizing the balance between external and internal demands is an extremely important evolutionary function. All animals must constantly weigh drives, motivations, plans, and expectations against the situation at hand.
Maybe the push-pull relationship between brain networks is a reflection of the pushing and pulling of internal and external demands made on the brain? We know experimentally that a DMN exists in other primates and in larger rodents, so it is not exclusive to humans. If the DMN is so important to our mental health because it is part of the balance between external and internal demands, one would expect to find a DMN in animals other than mammals. But limits in the spatial resolution of fMRI prevent us from looking for it in smaller brains in which we could manipulate the DMN—if we found it—to see how it worked. One way to test the universality of the push-pull hypothesis would be to look for the DMN in other animals with smaller brains, using a method analogous to fMRI, but one that provides a higher spatial resolution without sacrificing temporal resolution.
Optophysiology (imaging) with genetically encoded fluorescent dyes is an obvious candidate. This technique has been used for about twenty years now to study the responses of individual neuronal circuits, predominantly in Drosophila, but also in other genetically accessible model systems such as the mouse or the zebrafish. Optophysiologists genetically target a specific reporter molecule to neurons of interest. When the neuron is active such molecules emit light, which can be recorded with a camera on a microscope. So far, this technology has been used only in genetically defined circuits, and only three dimensions have been imaged: the two spatial dimensions of the focal plane, and time. To emulate fMRI, the fluorescent dye has to be expressed in all neurons of the brain, not just in individual circuits; and current imaging technology has to be expanded to include all four dimensions. Because of technical hurdles ranging from phototoxicity to light scattering to the need for sufficiently fast and sensitive optics, such a microscope does not exist—yet.
Using the Drosophila brain as a model for studying human brain function took a leap forward with the recent mapping of some 16,000 of the fly’s 100,000 neurons (Curr Biol, 21:1-11, 2011). With newer functional technology we would not only be able to look for a DMN in resting flies, we could also use the powerful genetic arsenal of this and other model systems to manipulate their brains and thereby find out how they balance their internal and external demands. Lacking this potential to manipulate—as we still do in all animals currently being studied with fMRI—largely precludes any functional assessment of DMN activity. It may be a long shot, but with the DMN playing so central a role not only in basic brain science but also in mental health, I propose that the next technological development in neuroscience will be 4-D imaging of the living brains of genetically accessible model organisms.
Björn Brembs is a Heisenberg Fellow (DFG) at the Freie Universität Berlin, Germany. He studies spontaneous behaviors in wild-type, mutant, and transgenic fruit flies and teaches them self-control and other completely useless things, just to find out how they do it. He has applied to the European Research Council for funding to develop 4-D optophysiology to study the Drosophila brain. He blogs at bjoern.brembs.net.