Adapted from a story that originally appeared on the Alzheimer Research Forum.
This is Part 1 of a two-part series. See also Part 2.
15 November 2012. Have neuroscientists entered the era of the programmable brain? Thanks to the related techniques of optogenetics and pharmacogenetics, which allow scientists to turn the activity of neurons on or off using light or designer drugs, scientists can now control specific subtypes of neurons in research animals with millisecond precision. This allows them to functionally dissect complex brain circuitry and determine which cells cause specific behaviors. In a satellite conference held 11-12 October 2012 in New Orleans, Louisiana, just before the Society for Neuroscience 2012 annual meeting, over 300 neuroscientists from around the world gathered to share data obtained with these methods and showcased new technical advances in this young field. Talks were greeted with enthusiasm by a highly engaged audience, and more than one presenter commented on the community spirit and buzz they felt. “These talks have been truly inspirational,” said Brad Lowell at Beth Israel Deaconess Medical Center, Boston, Massachusetts, summing up the group mood. Presentations ranged from basic science and discussions of new methodology to disease research. Only a few talks touched on neurodegenerative diseases such as Alzheimer’s and Parkinson’s, but speakers noted the potential the techniques hold for elucidating pathological mechanisms.
“We’re experiencing a revolution in neuroscience. New methods are allowing us to do things we couldn’t have dreamed of five to 10 years ago,” said Gary Aston-Jones at the Medical University of South Carolina, Charleston, expressing a common theme. Striking another popular note, Karl Deisseroth at Stanford University praised the willingness of the community to share unpublished data, reagents, and new protocols. Presenters noted that the field has taken off in the last five years, with hundreds of published papers and thousands of labs worldwide now using these methods.
News flash: how it works...
Deisseroth first demonstrated optical control of neurons in 2005 (see Boyden et al., 2005). The technique involves expressing light-sensitive transmembrane proteins called opsins in specific neurons in rodent brains. These opsins come from simple organisms such as microbes, fungi, or algae, and form ion channels that open when activated by light. The two most commonly used are halorhodopsin, which responds to yellow light, and channelrhodopsin, which opens under blue light. Halorhodopsin acts as an off switch for neurons, as it pumps in chloride ions and hyperpolarizes cells, while channelrhodopsin boosts neural activity by letting in cations and depolarizing the cells. Because light penetrates only about one millimeter through brain tissue, researchers implant optic fibers into rodent brains to reach the target regions. Deisseroth noted that the limited penetrance of light is a positive, giving the technique fine spatial control on top of millisecond temporal control. In his keynote address, he pointed out that a huge variety of opsins exist in the microbial world, potentially providing researchers with a “vast number of tools.” Molecular engineering improves the properties of these proteins and makes them more useful for research, he added.
...Even in the dark
The related field of pharmacogenetics complements these techniques. This approach, pioneered by researchers such as Bryan Roth at the University of North Carolina, Chapel Hill, involves engineering receptors to respond to specific drugs that have no other biological activity (see Armbruster et al., 2007, and Pei et al., 2008). These proteins, known as “Designer Receptors Exclusively Activated by Designer Drugs” (DREADDs) and “Receptors Activated Solely by Synthetic Ligands” (RASSLs), couple to specific G proteins that trigger downstream second messenger pathways that can have a variety of effects on cells.
In contrast to optogenetics, pharmacogenetics gives a longer-lasting response and can target a much broader brain area (since drugs penetrate better than light). John Neumaier at the University of Washington, Seattle, noted that designer receptors seem to modulate plasticity, rather than immediate behavior as optogenetics do. Optogenetic and pharmacogenetic approaches can be combined in the same animal to provide multiple levels of control, presenters noted. In the future, “Multiplexing will be de rigueur,” Roth suggested. He also pointed out that pharmacogenetic techniques have the potential to translate to people, since the drugs have no innate biological activity and therefore should have few, if any, side effects. However, just as in animals, therapeutic DREADDs would have to be introduced through a viral vector and be targeted to a particular neuron class.
Clarifying circuitry: behavioral and disease applications
The techniques could greatly advance understanding of brain circuitry. Scientists have struggled to map and decode circuits amongst the billions of neurons in the brain, partly because of the difficulty of functionally isolating specific types of neurons intermixed among many others. By injecting virally encoded opsins or designer receptors under the control of cell type-specific promoters, researchers can now single out neuronal classes. Moreover, they can manipulate the neurons to determine their direct effect on connecting cells. As one example, the hypothalamus, with its numerous anatomic nuclei and functions, has been a particular challenge to study, Lowell noted. With these new techniques, “the hypothalamus is now the land of opportunity,” he said. Although the hypothalamus is known to regulate appetite and feeding behavior, the wiring diagram remains a puzzle. Lowell used pharmacogenetic techniques to demonstrate an essential role for Agouti-related protein neurons found in the hypothalamus’ arcuate nucleus. Stimulation of these neurons triggered voracious appetite in rodents, while silencing them dropped feeding below normal, he reported.
Presenters described the use of optogenetics and pharmacogenetics to study a variety of complex processes such as depression, drug addiction, schizophrenia, satiety, arousal, and behavioral conditioning. “The level of sophistication in the coding of behaviors is stunning,” said Ann Graybiel of MIT. In many cases, researchers revisited classic studies that showed a role for a neuron type in a particular process but could not demonstrate causality. Using optogenetics, they determined which neurons directly act on others and gained new insight into complex circuitry.
For example, previous studies generated conflicting data about the role of cholinergic neurons in Parkinson’s disease. Some research showed that these neurons become hyperactive due to a loss of dopaminergic control (see, e.g., Pisani et al., 2003). In support of this, anticholinergic drugs were one of the first treatments for PD (see Brocks, 1999). However, other studies noted a loss of cholinergic neurons as the disease advances (see Bohnen and Albin, 2011) and reported that cholinesterase inhibitors, which increase cholinergic activity, improve PD symptoms (see, e.g., Aarsland et al., 2004).
To clarify the issue, Corinne Beurrier at Aix-Marseille University, France, selectively expressed channelrhodopsin and halorhodopsin in the striatal cholinergic neurons of mice, allowing her to activate or silence the cells with different-colored lights. Her preliminary data, presented in a poster session, suggests that turning on cholinergic neurons inhibits the majority of GABAergic medium spiny neurons (MSNs) in the striatum, while quieting cholinergic cells sparks MSNs, although she noted that more experiments are needed to confirm these results. She induced PD-like symptoms in the mice by either administering haloperidol, which blocks dopamine receptors, or lesioning the substantia nigra with the chemical 6-hydroxydopamine. In both models, silencing cholinergic neurons improved motor function. By contrast, activating these cells had no effect on motor deficits. Hyperactive cholinergic neurons do play a role in worsening motor symptoms in these models, Beurrier concluded. For data on how optogenetics and pharmacogenetics are being applied in AD and memory research, see Part 2.—Madolyn Bowman Rogers.
This is Part 1 of a two-part series. See also Part 2.