Two mini microscopes watched a mouse’s brain move its body in real time

The NINscope will help researchers uncover how neurons in different regions of the brain interact with each other

Sruthi Sanjeev Balakrishnan

Cell Biology

National Centre for Biological Sciences

 Nearly every move we make, from binge-watching Netflix to debating whether to finally learn some baking, is controlled by the brain. Different parts of the brain have different functions, and we are still only a fraction of the way through to understanding what each of these parts do. The realization that parts of the brain work as a network forms the foundation of current neuroscience research.

To work out the networks that create specific behaviors, neuroscientists need live peeks into different parts of the brain, while also recording the actions performed by the animal attached to the brain they are peeking at. In a recent study published in eLife, a group of scientists from the Netherlands figured out how to simultaneously monitor two brain regions, using miniature microscopes, called NINscopes, that can be directly mounted on the head of a mouse. 

The miniscopes are connected to a camera, letting researchers see what’s going on in the mouse’s brain.These devices are so small and light that not one, but two of them can be mounted on a single mouse. This nifty capability allows brain researchers do something they have been trying to achieve for years: observe two brain regions at the same time, while making sure that the mice they are studying are free to move around and behave normally.

Using two NINscopes at once, the scientists behind the new study were able to see neurons glow in two parts of the mouse brain – the cerebellum and the cortex – while the mouse was moving. A glowing neuron is an active neuron, so the researchers could match movements made by the mouse to specific patterns of neuronal activity. From these patterns, they figured out that the cerebellum and cortex work with each other to control some aspects of movement, but not all of them. 

The ability to visualize neurons in freely moving mice has historically been a difficult feat to achieve. It is a tricky and delicate procedure even in immobilized live mice, requiring time and surgical finesse on the part of the scientist. Traditionally, this was done by sticking electrodes into the brain and recording the electrical signals generated by the neurons. While this technique gives precise readings of neuronal activity and timing, it does not allow researchers to see exactly which neurons are active. At best they can only assign activity to select regions, but not specific neurons. 

In the 1990s, a new technique to record brain activity arrived, where scientists could use microscopes and special molecules called calcium indicators to observe neurons in action. When brain cells are activated, their interiors are flooded with calcium. The indicators bind to calcium and glow, making active neurons light up like fireflies. This effect can be captured very easily with a microscope and camera. What’s more, a single microscopic field can capture hundreds of cells at a time. Unlike the previous approach of using electrodes, calcium imaging lets scientists directly point out which neurons are active and when, just by looking at them. 

Miniscopes also use calcium indicators, but they bypass a major shortcoming of traditional microscopes, which is that all imaging has to be done from a mouse that is restrained. While images from restrained mice are of good quality, you can only study a very small set of behaviors from such mice. But when using miniscopes, researchers no longer have to worry about restraining the mice. Once the miniscope is in place, the mouse is allowed some time to recover from the process and is then free to go about its daily routine. Miniscopes give researchers freedom to correlate a plethora of behaviors, ranging from sleep-wake cycles to social interactions, with brain activity. 

The invention of the miniscope solved the issue of imaging brain cells in freely moving mice, but it still limited scientists to observing only one region at a time. When studying neuronal ensembles, it is important to visualize multiple brain regions in actively behaving mice. 

This is where the NINscope offers a unique advantage over other miniscopes. It combines the dual-imaging capacity of traditional microscopes with the ability of miniscopes to track behaviors in unrestrained animals. The researchers custom-built electronic and optical components to build the NINscope, allowing them to shrink the instrument, house it in a 3D-printed shell, and place two NINscopes on a single mouse to study interactions between brain regions. 

The NINscope does have some limitations. Its main selling point of dual-imaging, meaning simultaneously visualizing two brain regions, is not actually possible for any two regions of choice. The NINscope design mandates that a minimum distance has to be maintained between two mounted devices, meaning that there is some physical constraint on the combination of regions that can be imaged. It is also not clear whether the NINscope has any long-term effects on the mice. This is important to assess since experiments that analyze behaviors often run into weeks, in this case requiring the mice to go about all their usual activities with two miniscopes on their heads.

Nonetheless, the NINscope sets the stage for a large range of experiments aiming to map out potential neural networks. Weighing a mere 1.6 g (about the weight of 1.5 paperclips), the device comes power-packed with a few extra features. Researchers can use it to manipulate brain cells and see how they react using a module for optogenetic stimulation, which is a method where one can use light to artificially activate a neuron. With the proper modifications, they can also use it to look deeper into the brain. The NINscope also comes with a built-in accelerometer, allowing scientists to track the movements of the mice. 

The NINscope is the culmination of years of smaller advances in technology. It is also a reflection of the multidisciplinary nature of science; the team needed expertise in the fields of optics and electronics to make the device, as well as in the field of neuroscience. An instrument like the NINscope offers an avenue into investigating neuronal ensembles, laying the groundwork for understanding how our brains function as a network. 

Comment Peer Commentary

We ask other scientists from our Consortium to respond to articles with commentary from their expert perspective.

Raj Rajeshwar Malinda

Cell Biology and Developmental Biology

Nice read! As you mentioned about the limited knowledge about the long-term effects of using NINScope, then how would they justify the data with long-term behavioural patterns in mice to check whether any significant effect of having the microscope attached is seen? And, you mentioned limitation of choices of brain regions, is it possible to study one region with NINscopes and other one with any other imaging methods?

Sruthi Sanjeev Balakrishnan responds:

Their behavioural data on the effect of the NINscope  comes from a period of 3-4 days, so there is limited data on the effects over weeks. Regarding the brain regions, it’s a matter of spatial constraints, so I’m guessing that a smaller miniscope could be potentially paired with the NINscope. That being said, the NINscope is one of the smallest so far, so I guess we’ll have to wait for a few more engineering marvels before we expand the number of regions that can be examined. 

Dori Grijseels


University of Sussex

This is a great article explaining an amazing new technique in neuroscience. I think an important point from this article is the importance of studying freely-moving mice. As Sruthi rightly points out, and I’ve written about this before as well, having to restrain the mice limits the studies you can do, as the mice don’t get natural inputs, as much as we try to mimic those with virtual reality systems.

This article has me wondering what the next step will be. As pointed out in the article the NINscope is a culmination of many smaller advances, and it seems like each year the head-mounted microscopes are getting a little better. At the same time, we see a massive improvement in electrophysiology approaches (for example with the Neuropixels probe, which can track 1000s of neurons at the same time). And people have even started combining the two, leveraging the advantages of both. I’m excited to see where neuroscience goes next!

Sruthi Sanjeev Balakrishnan responds:

Combining electrophysiology with microscopy will definitely go a long way in evening out the shortcomings of both techniques. I’m curious myself to see where things are headed in the future. 

Adithi Ramakrishnan

Developmental Neuroscience

College of William and Mary

Movement is an especially intriguing target for measuring brain activity because of the activity in the brain prior to the initiation of movement – the decision to move – as well as the activity from the cerebellum to fine-tune that movement.  I’d be curious as to whether the NINscope could be used in the future to track all the different brain areas involved in the planning, initiation, and refinement of movement, although the minimum distance limit is a factor. This article got me  excited about the future directions of brain imaging and about how small yet precise our imaging technologies can get! 

Sruthi Sanjeev Balakrishnan responds:

You’ve really hit the nail on the head regarding the decision to move. One of the proof-of-principle experiments the researchers did was to simultaneously observe the cerebellum and cortex in moving mice. They  were able to identify patterns in a small time window before peak movement, possibly reflecting the intent to move. This was very preliminary, of course, so lots more to be studied in the future! 

Kamila Kourbanova

Neuroscience and Molecular Biology

Johns Hopkins University

This article does an amazing job describing the limitations scientists face in recording the natural brain activity in mice and how NINscopes can be an extremely useful tool in getting around some of these obstacles. This finding is exceedingly useful for some areas of neuroscience that require as minimal interference to behavior as possible! For example, in sleep research scientists currently use electrodes that are similarly placed on the head of a mouse to record EEG activity of a freely moving mouse. This gives us a pretty good idea of the sleep patterns in mice but a major limitation is we cannot get a good read on deeper brain structures that we know are important in sleep simply because of the anatomical structure of the brain. NINscopes would be particularly useful to not only visualize these deeper brain structures but to record exactly when they are activated during behaviors like wake and sleep! It would be interesting to see in the future if they can be used in conjunction with EEG headmounts. 

Sruthi Sanjeev Balakrishnan responds:

The ability to visualise deep brain structures is definitely a big plus with the NINscope. They also did some early experiments by combining deep brain imaging with optogenetics and play around with neuronal activation in the dorsal striatum. Here’s hoping this technique takes  off and helps us answer a lot more questions in behavioural neuroscience. 

Interesting article that really highlights an important methodological advancement in the field of neuroscience! I wonder if these miniscopes can be used to study brains from other species and if not, why not? Logically, one would think that it would be easier to study multiple brain areas in rats because they’re bigger in size. Another question that comes to mind: Based on the timing of these calcium signals, what can the scientists say (or not say) about the relationship between the  neurons in these two brain regions? For example, can the timing data help determine if neurons within one area that’s being recorded project directly onto neurons within the other area? 

Vidya Ganapati

Swarthmore College

This well-researched article details the importance of applying progress in engineering to advance biology. Accelerometers, which are used in NINscopes to measure movement of the mice, were optimized for use in consumer electronics such as your smartphone. A key ingredient in making miniscopes mini is the use of gradient-index (GRIN) lenses. Traditional lenses are curved to bend and focus light, but GRIN lenses are flat, and allow for a smaller form factor. To bend light, GRIN lenses use changes in material properties rather than curvature. In the future, advances in machine learning may be used to process the data obtained from the NINscopes to reveal more biological insights.