Advanced Imaging in Neuroscience Research

Neuroscience is a vast field, or perhaps a vast collection of fields. No single person, or even a single lab or institution, can be an expert in it all. Those imaging human subjects may have little overlap with those whose primary focus is on animal models or in vitro studies, for example—although discoveries and insights are shared, borrowing from and cross-pollinating each other. This was inevitably the case among the more than 20,000 attendees at the Society for Neuroscience’s annual gathering in Chicago last month.

Rather than attempting to summarize advancements in imaging methods allowing the field to move forward—a Herculean task that would certainly have to include discussion of the newly mapped complete fruit fly brain connectome and the emergence of the neuroimaging databases that enabled it—this article will discuss a select few examples from practitioners in the field of addiction research, along with an innovation in cell culture that provides some solutions for in vitro imaging.

PET/MRI

Dardo Tomasi, a physicist and a Staff Scientist in the Laboratory of Neuroimaging (LNI) of the National Institute of Alcohol Abuse and Alcoholism, uses imaging to see the neurological basis of psychiatric disorders, in particular addiction. “We focus on dopamine and on functional connectivity, and also brain activation, concentrating on the reward pathway. For that we use fMRI and PET.”

Functional magnetic resonance imaging (fMRI) looks at the changes in magnetic field produced by, in this case, changes in blood flow. It is a safe procedure that allows researchers to penetrate through the skull deep into the brain. Changes in blood flow can be used, in turn, as an indicator of brain activity resulting from a challenge such as administration of a stimulant.

Tomasi uses a standard 3 Tesla (3T) instrument, with a resolution of up to a few millimeters across. In some emerging technologies, magnetic fields of 7T or even higher are being used. An 11.7T instrument in France—purported to be the world’s most powerful—can resolve images down to 0.2 mm. “But we are not using these techniques. They are not fully developed. There are some challenges to overcome, for instance susceptibility to artifacts and things like that,” says Tomasi.

Positron emission tomography (PET) typically uses naturally occurring or synthetic substances, such as metabolic precursors, pharmaceuticals, street drugs, or neurotransmitters, that are radiolabeled to allow their tracking. Distinct radiotracers can be followed simultaneously to examine, for example, how the presence of drugs like methylphenidate (Ritalin) affects dopamine D1 and D2 receptors over time.

“We try to use ultra-fast techniques that can help provide this high spatio-temporal resolution in MRI. At the same time we use hybrid PET/MRI scanners to map neurotransmitter activity and brain activation or brain connectivity within the same individual at the same time,” explains Tomasi.

He has the “luxury” to conduct research at the clinical center of the NIH, that affords access to simultaneous MRI and PET scanning, “along with a radiochemistry lab with a cyclotron to produce the radiotracers like ¹¹C that we are using,” he says. “So only a few places in the world would be able to do this type of research.”

Animal models

There are imaging techniques used in neuroscience research that are unavailable or impractical—even beyond obvious ethical and financial considerations—for garnering information directly from humans.

For example, many optical solutions depend on the detection of fluorescently labeled markers, whether genetically encoded or added exogenously. “With optical imaging you can get sub-cellular resolution, but you’re only looking a penetration depth of about 100 micrometers, traditionally,” points out Yingtian Pan, Professor, Department of Biomedical Engineering at Stony Brook University in New York. Such penetration will allow superficial imaging of a mouse cortex (beneath a transparent window), but is not applicable to humans.

Multi-photon imaging (also known as 2-photon excitation imaging, or TPI), requires an elaborate setup, but offers penetration of about 600 micrometers by limiting out-of-focus light signals. “A lot of people are using this as a tool to look at awake animal models, how the vascular system, the blood flow, changes under associated brain diseases like Alzheimer’s Disease. Now people even use it in addiction research,” Pan says. Yet, he points out two limitations to using TPI to look at blood flow. First, a tracker is necessary, which can cause complications if staining the red blood cells. Second, the field of view is very small—on the order of one or two vessels, which does not allow for averaging of blood flow.

Functional ultrasound is beginning to play an important role in imaging of neural-vascular couplings (NVCs), with a resolution between that of MRI and optical techniques. “But they have great depth that optical images are not able to get.”

Sometimes research requires differentiating between different types, or sub-types, of cells. Pan’s Stony Brook collaborator Congwu Du used viral vectors encoding red and green Ca2+ indicators to specifically target astrocytes and neurons, respectively. They used a newly developed hybrid instrument—capable of multi-channel fluorescence and Doppler optical coherence microscopy, alongside interrogating vascular interactions—to simultaneously image blood flow change, metabolic change (like oxygen/de-oxygen change), and also the cellular change, in each cell type, in vivo, Pan explained. This allowed them, for example, “to see how second messenger is going to be affected by the psychostimulant, such as cocaine.”

Don’t use restraint

Many in vivo imaging studies—of cerebral blood flow (CBF), for example, which is widely used to assess brain function—are affected by motion of the subject. Therefore most preclinical studies of CBF are conducted on anesthetized animals, or perhaps on awake but restrained animals. Yet this introduces confounding factors, such as the interaction between the anesthesia a drug being tested, or the stress invoked by being unable to move.

Pan and his colleagues have developed a solution in the form of a self-supervised deep learning algorithm that is able to de-noise the image and remove motion artifacts in ultra-high-resolution Doppler optical coherence tomographic studies of awake mice.

In vitro

In vitro studies of individual neurons and NVCs are still a fundamental part of neuroscience, with imaging an important component of the research.

Bio-Blocks™ are a modular hydrogel made from a proprietary natural polymer that allows cells—monocultures and co-cultures alike—to “operate as they do in vitro,” producing and growing in their own scaffold, says Scott Leigh, Chief Business Development Officer at the parent company Ronawk. “Once they're in there, the cell will start communicating, migrating, propagating in 3D, and the substrate does not interfere with the development of the cell culture.”

The 3D block can be imaged as a Z-stack. It can also be fixed and sectioned “just like a tissue,” he points out.

With so much to learn about, and so many ways to study the nervous system, its components, its development and its maladies, innovation in neuroscience imaging will continue apace into the foreseeable future.