Forgetfulness at the touch of a button: Teleopto Wireless Optogenetics in Science Magazine and The New York Times.

Forgetfulness at the touch of a button: Teleopto Wireless Optogenetics in Science Magazine and The New York Times.

Melanin-concentrating hormone (MCH) neurons are unlike most neurons: they are most active during sleep. Scientists have studied their role in regulating sleep and feeding behavior for some time, but the Yamanaka lab at Nagoya University in Japan has found that they may also have a role in preventing the consolidation of memories during sleep.  

Prof. Yamanaka (a co-developer of Teleopto) found that MCH neurons can suppress neurons in the hippocampus responsible for memory consolidation. His lab confirmed this role by using 

Teleopto Wireless Optogenetics: blue light and channelrhodopsin 2 were used to activate MCH neurons; green light/archaerhodopsin were used to inhibit them. This was done bilaterally during both memory consolidation (REM sleep) and awake periods. Teleopto was used so that the animals were able to move freely and interact naturally with objects during Novel Object Recognition (NOR) tests.

When mice had MCH neurons activated during sleep, their ability to remember events decreased: they forgot which objects they had encountered before sleeping and treated them the same way they treated novel objects. Conversely, when their MCH neurons were inhibited they were able to remember which objects they had already interacted with. They ignored the familiar objects and explored the novel objects instead.

When Teleopto was used to illuminate MCH neurons during awake periods, there was no effect on hippocampal-dependent memory.

 When interviewed by The New York Times, Prof. Yamanaka explained

“These results suggest that hypothalamic M.C.H. neurons help the brain actively forget new information that is not important.” And because the neurons are most active during R.E.M. sleep, they may explain why humans usually do not remember their dreams when they wake up. “The neurons may be clearing up memory resources for the next day,” Dr. Yamanaka said.

The article, “REM sleep–active MCH neurons are involved in forgetting hippocampus-dependent memories.” is available at:

OptoDroplets: Organelle formation controlled by Teleopto LED arrays

OptoDroplets: Organelle formation controlled by Teleopto LED arrays

What are Biomolecular condensates?

Biomolecular condensates are a unique class of organelles: they have no membranes. They can form, merge, split, and disappear in minutes, temporarily creating local incubators and assembly lines with properties very different from the bulk of the cells surrounding them. The local high concentrations of proteins and polynucleotides inside these condensates can both speed up and interfere with reactions, challenging the researchers trying to understand the rules of cell biology.  

Some condensates, such as the nucleolus and Cajal bodies, were first observed over a century ago, but others such as processing bodies, PML bodies, and paraspeckles were only discovered recently. It is only within the past few years that researchers have begun to understand that these organelles all share a common organizing principle: protein association drives the formation of gels which coalesce into the organelles themselves, which then behave according to the classic rules of phase separation and phase transition. These organelles condense in much the same way water vapor condenses into droplets on a window.

Why study biomolecular condensates?

This new understanding has led to condensates becoming a target for drug design. Dewpoint Therapeutics launched earlier this year, based on studies of stress granules. They seek to prevent temporary condensates of FUS protein from congealing into permanent aggregates, a driving force in amyotrophic lateral sclerosis (ALS). 

Liquid-liquid phase separation also has a role in gene expression: transcription factors have been found to rely on segregation inside condensates to initiate and control RNA production, yielding new targets for cancer therapies. The kinetics of ribosomal RNA processing is also proving to be dependent on the extent of gelation of the nucleolus.

Teleopto LED arrays and Biomolecular condensates

Clifford Brangwynne, Macarthur Fellow and Assoc. Prof. at Princeton University uses light to control the formation of condensates. Once activated by light, proteins like Cry2olig1 oligomerize within seconds. By fusing Cry2olig to an RNA binding protein that drives condensation in the nucleus (NPM1), the BW lab created optoDroplets: light activated condensates held together by a meshwork of protein and nucleic acids.

Blue light from a Teleopto LEDA array causes these CRY2 fusions (opto-NPM1) to coalesce into a meshwork of proteins capable of turning the nucleolus of a cell into a tightly linked gel2. The lab tunes the properties of the optoDroplets by adjusting the brightness: more light leads to more self-association and smaller pores in the meshwork. As the pores shrink, small proteins can still move through the hydrogel but larger molecules and complexes become trapped. This model allows the Brangwynne lab to study the effect of viscoelasticity on the formation of ribosomes and the processing of rRNA with just the press of a button. In a recent PNAS paper, they found that increasing the gelation of the nucleolus leads to the accumulation of larger rRNA precursors, while smaller precursors are depleted.

After the light is turned off, the condensates typically degenerate within 5 minutes. Fixing the cells while they are still illuminated allows the optoDroplets to be imaged and studied later, as shown in the figure below:

Incubator-compatible Teleopto LED arrays are tools designed for doing in-vitro optogenetics on 96 well plates. The arrays are available in wavelengths from UV to infrared and can be controlled by most pulse generators.

Postdoc Jorine Eeftens said that the Brangwynne lab used to use microscope mounted lasers to make condensates, but that only let them focus on a few cells at a time. The LEDA array allows them to activate many cells at once, greatly improving throughput in the lab. “We use it routinely, every day. We love working with it, the [LEDA] array allows us to use lots of cells, and then fix them for study. It’s our high throughput system.”

Biomolecular Condensates and Teleopto at the Woods Hole Physiology Course

The Woods Hole Marine Biology Laboratory discovery courses are intense, full-immersion summer courses for graduate students and postdocs. Students brainstorm, design and carry out their own projects – which frequently lead to publications. Ten years ago during a course led by Anthony Hyman and Brangwynne, then a postdoc in the Hyman lab, a project showed that P-granules behave like oil droplets when shearing forces are applied. The initial result from the Woods Hole class was followed up by Hyman and Brangwynne at Max Planck Institute, leading to a publication for both the students and the instructors. The paper shows that p-granule behavior follows the classic rules of phase separation and hinted at how this process could be involved in many more aspects of cellular behavior than previously thought3.

Coming full circle, this past summer Prof. Brangwynne and his postdocs led one of the Woods Hole course rotations and focused on the role of condensates in the nucleolus. They brought a LEDA array so that students could form optoDroplets in incubators during the class.

You can learn more about optoDroplets at the Brangwynne website.

Teleopto LED arrays are also being used in the development of new optogenetic switches, cardiovascular and nervous system developmental biology, ophthalmology, and photobiochemistry. 

(1) Taslimi, A., Vrana, J. D., Chen, D., Borinskaya, S., Mayer, B. J., Kennedy, M. J., & Tucker, C. L. (2014). An optimized optogenetic clustering tool for probing protein interaction and function. Nature communications, 5, 4925.

(2) Zhu, L., Richardson, T. M., Wacheul, L., Wei, M. T., Feric, M., Whitney, G., … & Brangwynne, C. P. (2019). Controlling the material properties and rRNA processing function of the nucleolus using light. Proceedings of the National Academy of Sciences, 116(35), 17330-17335. 

(3)Brangwynne, C. P., Eckmann, C. R., Courson, D. S., Rybarska, A., Hoege, C., Gharakhani, J., … & Hyman, A. A. (2009). Germline P granules are liquid droplets that localize by controlled dissolution/condensation. Science, 324(5935), 1729-1732.


Uncovering Neural Circuits Involved in Motor Learning

Uncovering Neural Circuits Involved in Motor Learning

Tanaka and colleagues in Dr. Matsuzaki’s lab at the University of Tokyo have been researching the role of thalamocortical axonal activity in motor learning using the TaskForcer. 

Brain regions involved in voluntary movement

The thalamus is a central hub through which neuronal signals are transmitted through the cortex and other subcortical structures including the basal ganglia, the pons, and the cerebellum.

Brain regions involved in voluntary motor control (Adapted from Waxman, SG. Clinical Neuroanatomy 26th edition, 2009).

Together, these structures are involved in controlling voluntary movements like manual skills. In animals, manual skills are learned and refined through repetitive motor learning, which instigates neuronal plasticity in the brain structures involved in these processes.

Measuring axonal activity in vivo

Using two-photon calcium imaging of GCaMP expressing thalamocortical axons in the mouse motor cortex in combination with the TaskForcer restraint operant chamber, Tanaka, et al., ascertained the role of thalamocortical axonal activity in skilled motor learning.

The TaskForcer operant chamber fits under the 2P microscope, enabling precise neural imaging during operant training. The task used was a self-initiated lever-pull task, where mice were trained to pull a lever in order to receive a water reward.

By recording calcium activity of GCaMP expressing thalamocortical axons in the motor cortex during learning, they were able to track the temporal dynamics of thalamocortical activity associated with each stage of the learning process.

Linking neuronal activity to coordinated movements

The authors found that thalamocortical activity was time-locked to both initiation and execution of the lever pull task and that this activity stabilized over time after the initial learning. As proof of concept to verify the thalamus’ role in motor learning, when the authors lesioned the thalamus, lever pull behavior significantly decreased. These results indicated that thalamocortical axonal activity is necessary for motor skill learning, and is more involved during the initial stages of motor skill learning.

Example of the lever-pull task using the TaskForcer. (Adapted from Tanaka et al., 2018)

To learn more about this product, click here.

Check out the full article in Cell here.

Part V – Tokyo Medical and Dental University

Part V – Tokyo Medical and Dental University

Upon my return back to Tokyo, I had one final visit with Dr. Isomura at Tokyo Medical and Dental University. He originally developed the TaskForcer for rats with O’Hara over 8 years ago!

Dr. Isomura’s research focuses on understanding information processing in Motor Cortex during motor skill learning. To do this, he performs in vivo whole-cell patch clamp recordings in Motor Cortex as animals learn the lever pull task that was specifically designed for the TaskForcer.

What makes simultaneous neural recording during operant behaviors possible with the TaskForcer is the unique spout-lever. This was specially designed by Dr. Isomura and O’Hara such that the reward (liquid from the spout) and operandum (lever) are combined into one. In this way, the animal can still obtain a reward for pulling the lever even while its body is restrained, allowing for operant learning during simultaneous neurophysiological recording.

Dr. Isomura explains, “Since the animals must learn to perform the lever pull task while under head fixation, we wanted to make sure that the animal could access the reward with minimal head movement, but still be motivated to perform the task.”

Isomura also explains, “We were surprised that rats started pulling the lever the very first day that we put them in the chamber. The lever pull task is very robust. We don’t see animal attrition from failure of animals to learn the task.”

The TaskForcer with a stereotaxic setup in a sound attenuating box.

Me with Dr. Takahashi at Doshisha University

“With the TaskForcer, we can reliably get extremely precise single unit recordings during motor behaviors which allows us to examine causal links between neural activity and behavior in great detail.” – Dr. Isomura

Me with O’Hara team members alongside Dr. Isomura (left).

Part IV – We Visit Doshisha University in Kyoto

Part IV – We Visit Doshisha University in Kyoto

While in Kyoto, I traveled to Doshisha University to visit the lab of Dr. Takahashi, who worked with O’Hara to design the Free Maze, a reconfigurable maze for learning and memory tests.

Me with Dr. Takahashi at Doshisha University

Dr. Takahashi studies hippocampal place cell activity in mice and rats, and wanted to build a maze system that he could easily change in order to understand how place cells adapt to changes in environments.

According to Dr. Takahashi, “The Free Maze was designed to be flexible, reliable, and repeatable.”
“We built this maze in order to design a system where users could build their own tasks to their own specifications, change maze designs rapidly, and reconfigure previous designs easily.”

For his research, Dr. Takahashi records population activity of hippocampal place cells in freely moving rats as they navigate through the Free Maze.

The Free Maze is an extremely unique product. It’s like legos for scientists!

The original paper detailing the Free Maze is currently under review and should be available soon.

The Free Maze for Rats

Next stop – Tokyo Medical and Dental University
Mapping Motor Circuit Mechanisms During Voluntary Movement

Mapping Motor Circuit Mechanisms During Voluntary Movement

Several users of our O’Hara behavioral testing systems are presenting their research at SfN.

Matsuzaki and colleagues at the University of Tokyo are investigating the role of primary and secondary motor cortices in information processing during self-initiated versus externally triggered movements. To do this they are using the TaskForcer for mice in combination with in vivo widefield two-photon imaging. Below is a summary of what they plan to present at SFN.

Voluntary motor movements can either be self-initiated, or externally triggered. Neuronal ensembles in the primary (M1) and secondary (M2) both play a role in information processing during voluntary movement, but the relative contribution of each remains unclear. Furthermore, how each region processes information when the same movement is self-initiated (SI) versus externally triggered (ET) remains unknown. Terada and colleagues in the Matsuzaki lab examined whether the pattern of activation differed in M2 compared to M1 during SI and ET movements. They hypothesized that the presence of external stimuli would be sufficient to alter neural activity patterns in M2 when the same movement was self-initiated versus externally triggered. To test this, they trained head-fixed mice to perform a self-initiated lever-pull task (SI) and an external cue-triggered lever-pull task (ET) using the TaskForcer. During task performance, they conducted calcium imaging of GcAMP infected layer 2/3 neurons concurrently in M2 and M1 using super-wide-field two-photon microscopy (Terada et al., 2018) in mice implanted with large cranial windows.
They found that the proportion of neurons that responded to movement-related activity specific to either learning type was greater in M2 compared to M1. Furthermore, calcium activity in M2 was differed significantly between the self-initiated and externally triggered trials, indicating that external stimuli are sufficient to drive differential neuronal responses in M2. These results also suggested that M2 can distinguish between learning trials even when the same body part is initiated.

To learn more about Terada and colleagues application of the TaskForcer check out their poster at SFN or visit our booth # 1502!

Abstract Citation

1The Univ. of Tokyo, Tokyo, Japan;2Natl. Inst. For Physiological Sci., Okazaki, Japan. Neural dynamics in the mouse secondary and primary motor cortices during self-initiated and externally triggered movements. Program No. 081.05. 2019 Neuroscience Meeting Planner. Chicago, IL: Society for Neuroscience, 2019. Online.

TaskForcer: Restraint Chamber for Operant Conditioning