SfN 2012—Opto- and Pharmacogenetics: New Ways to Study Memory and AD
Adapted from a story that originally appeared on the Alzheimer Research Forum.
This is Part 2 of a two-part series. See also Part 1.
15 November 2012. By allowing researchers to turn on or off specific classes of neurons in mice at will, optogenetics and pharmacogenetics promise to revolutionize the field of neuroscience (see Part 1 of this series). In a satellite conference held 11-12 October 2012 before the Society for Neuroscience annual meeting in New Orleans, Louisiana, several researchers discussed how they have used the techniques to analyze basic memory mechanisms and to study Alzheimer’s disease.
Triggering and synthesizing memories: real-world inception?
In the Christopher Nolan movie Inception, Leonardo DiCaprio’s character inserted false memories into a man’s brain. Sound like science fiction? Perhaps not. Optogenetics might allow scientists to create fake memories, said Susumu Tonegawa from MIT in his talk. If that could be done, it would demonstrate that scientists understand how memories are made, Tonegawa noted. As a first step to figuring out how the brain stores experiences, Tonegawa wondered if he could artificially trigger a memory by activating the proper engram, which is a network of synaptically connected neurons believed to record experiences. To demonstrate this, he needed a way to label only the neurons involved in the memory, and only at the time of memory formation, as well as a means to selectively stimulate those neurons later. He turned to optogenetics.
Tonegawa planned to specifically label the memory engram neurons with channelrhodopsin-2, a light-activated cation channel (see Part 1 of this series). That would allow him to reactivate those neurons later. The challenge was to induce expression of the opsin only in the memory engram cells. As a first step, the researchers focused on hippocampal memories formed in a contextual fear conditioning paradigm. In mice, they expressed channelrhodopsin-2 in the dentate gyrus under control of an inducible TetO promoter. That promoter kept the gene silent as long as the mice received doxycycline in their diet. With the gene off, the researchers habituated the mice to a safe environment. Then they stopped doxycycline treatment, de-repressing channelrhodopsin synthesis in cells that also expressed the TetO activator tTA. Meanwhile, the tTA gene under the control of the c-fos promoter only turned on in response to activity.
The researchers waited two days to allow the doxycycline to clear from the mice, then placed the mice in a second context, where they received a shock. In theory, the cells that fired during this experience would turn on c-fos, thus activating tTA, which would cause channelrhodopsin to be made only in those cells. To test this, the researchers resumed doxycycline treatment, put the mice back in the safe context, and then activated channelrhodopsin-labeled cells with a pulse of blue light. The mice froze, showing that they were re-experiencing a fear memory without any external trigger (see Liu et al., 2012).
Numerous control experiments demonstrated that this response occurred only when channelrhodopsin labeled a fear memory. For example, if the mice did not get a shock in the fear context, they did not freeze later in response to light, even though their memories of the new environment got labeled with the opsin. The findings validate the theory of memory engrams, showing that a network of cells does store particular experiences, Tonegawa said.
In the future, Tonegawa will investigate whether there are multiple engrams in different brain regions for the same memory, and if each is sufficient to recall the memory. Examination of labeled cells will also help map engrams and their connections. Tonegawa wondered if scientists could go further and implant a false memory by using optogenetics. Though this sounds more like Inception sci-fi, Tonegawa believes it could be done by light-activating a previously formed memory engram in conjunction with exposure to a new, unconditioned stimulus, e.g., an odor. The animal would then have a false memory of receiving a foot shock along with the smell. Although one previous study found this method does not work (see Garner et al., 2012), Tonegawa suggested that study activated too broad a swath of brain to really test the theory.
New approaches in Alzheimer’s disease
Optogenetic and pharmacogenetic techniques are also being used to examine hypotheses and develop treatments for Alzheimer’s disease. Sylvie Claeysen at INSERM, France, discussed a pharmacogenetic strategy that shows therapeutic potential in AD mice. She introduced a point mutation into the serotonin receptor 5-HT4 to make it a Receptor Activated Solely by Synthetic Ligands, or RASSL (see also Part 1). The receptor also shows some basal activity in the absence of any ligand, Claeysen noted (see Claeysen et al., 2003). Claeysen first used this receptor to study Parkinson’s disease, but in recent studies found it beneficial in AD models as well.
When the researchers expressed this RASSL in mouse cortical neurons, the cells released more sAPPα, a product of the non-amyloidogenic α cleavage of amyloid precursor protein (APP). Levels of sAPPα climbed further in the presence of the synthetic ligand. Claeysen reported that 5-HT4 RASSL physically interacts with mature α-secretase ADAM10. The interaction probably occurs in the plasma membrane and serves to stimulate the secretase, Claeysen said. Knocking down ADAM10 levels sharply curtailed sAPPα release in this system. She is currently looking for more interacting proteins that might form a complex with ADAM10, APP, and the 5-HT4 RASSL.
In theory, stimulation of ADAM10 should reduce amyloidogenic processing of APP and could ameliorate pathology. To test this idea, the researchers expressed the RASSL in 5xFAD mice, which have particularly aggressive AD-like pathology, and treated them with the synthetic agonist at one or two months of age for several weeks. This stage corresponds to the prodromal phase of the disease, as the animals have amyloid plaques but do not yet show behavioral deficits, Claeysen said. Treated mice had fewer plaques and less inflammation compared to controls, with better effects the longer the treatment continued. Claeysen has not yet looked at behavior but plans to do that next.
In contrast to this pharmacogenetic approach, Li-Huei Tsai at MIT used optogenetics to look at the role of the cholinergic system in a different mouse AD model. Cholinergic neurons in the medial septum project to the hippocampus, and cholinesterase inhibitors improve cognitive function in mild AD. As cholinergic neurons are thought to mediate hippocampal ϑ rhythms, which have been linked to cognition, Tsai wondered if improved rhythms might explain some of the cognitive benefit of these drugs (see Yener et al., 2007).
To explore this, Tsai used CK-p25 mice, which accumulate amyloid plaques and neurofibrillary tangles, and exhibit profound neurodegeneration, and, in common with the 5xFAD mouse model, show dampened hippocampal ϑ and γ rhythms. She expressed channelrhodopsin-2 in cholinergic neurons of the medial septum, activated the cells with light, and measured oscillations in the hippocampus. A single pulse of light ramped up oscillations for the next 10 minutes. The effects on cognitive performance were dramatic. These animals perform poorly in associative learning tasks, but one minute of light stimulation improved their learning skills to wild-type levels. The boost lasted up to one week, Tsai reported. She also saw improvements in spatial memory and long-term potentiation.
What lies ahead
Presenters emphasized that current tools and methodologies for optogenetics and pharmacogenetics have only scratched the surface. Many talks highlighted technical advances. For example, Feng Zhang, previously in Karl Deisseroth’s lab at Stanford University, Palo Alto, California, and now at MIT, talked about finding novel opsins that respond to different wavelengths of light. He noted that channelrhodopsin-1 from the green algae Volvox activates with green or yellow light, while rhodopsin-3 from the alga Guillardia theta inhibits neurons in response to blue light (see Mei and Zhang, 2012). As green fluorescent protein and related proteins from jellyfish and other marine organisms have revolutionized the study of cell and molecular biology, these and other opsins will broaden the tools available to researchers and permit finer control of neuronal kinetics, Zhang said. Likewise, Ed Boyden at MIT noted that red light penetrates deeper into brain than other wavelengths, making opsins that respond to this wavelength particularly useful. He is developing several red light receptors, as well as other optogenetic tools such as a fiber array that can bend light 90 degrees for better targeting of specific brain regions (see website).
Other researchers focused on hardware. One limitation in recording electrical activity from mouse brain is that these small animals cannot support heavy equipment on their heads. Chris Moore at Brown University, Providence, Rhode Island, described the creation of an UltraLight FlexDrive, weighing less than two grams. The device allows recording from 32 or 64 channels and can also independently control several optical fibers, Moore said. In collaboration with Josh Siegle and Jakob Voigts at MIT, he is also building a system for real-time detection of electrical activity, which he said would be cheaper than current slower systems (see website for details of other hardware).
Optical fibers that protrude from the head and tether mice, interfering with free movement and behavior, present another problem for optogenetic studies. Jordan McCall, a doctoral student in the lab of Michael Bruchas at Washington University, St. Louis, Missouri, said providing light through an implantable micro-ILED (inorganic light-emitting diode) could circumvent that issue. The 20-micron-thick device (smaller than many neurons) can be inserted into the brain with a thin needle. The device causes less gliosis than a standard metal cannula does, McCall reported. The micro-ILED can emit multiple, independently controlled wavelengths of light in numerous directions. Researchers control the device by radio signals, which the animal receives through a tiny, removable antenna placed on the head. The system allows mice to scurry around freely while their neurons are stimulated.
Numerous resources exist to help researchers carry out optogenetic studies. These include The Jackson Lab websites for optogenetics and useful transgenic mouse strains, and a “cookbook” that describes how to combine optogenetics with functional MRI to look at the network impact of activating specific cell types (see Desai et al., 2011). In concluding remarks, Gary Aston-Jones at the Medical University of South Carolina, Charleston, praised the “amazing new technologies and devices presented” at the satellite meeting. He also invited attendees to submit papers to a special issue of Brain Research dedicated to optogenetics and pharmacogenetics that will appear next year.—Madolyn Bowman Rogers.
This is Part 2 of a two-part series. See also Part 1.