For years, neuroscientists have gained new revelations about how the brain works by separately studying genes, neurons, or animal behavior. Yet they are surely far from understanding the entirety of brain function. Today, neuroscientists are making strides by developing new tools and combining existing techniques to yield fresh insights.
One strategy is to use behavior as a specific entryway into learning how the brain operates. A simple behavior—like the patellar or knee-jerk reflex your doctor tests by tapping the patellar tendon in your knee—indicates the healthy functioning of a specific region of your central nervous system. But studying behaviors that involve multiple neural and motor pathways and different brain areas—such as eating and physical activity—remains challenging. And researchers encounter even more obstacles when trying to analyze complex behaviors like social interactions. Neuroscientists are now harnessing new technologies and advanced computing capabilities that analyze both simple and complex behaviors to increase our knowledge of brain function.
Body in motion, mind in motion
As any neurologist can tell you, small movements speak volumes about what’s going on inside the brain. For example, supination—turning the hand and forearm to the palm-up position—can indicate the health of the corticospinal tract, a band of nerve fibers in the brain that ferries information about limb movements to and from the spinal cord.
Neurologists can detect brain or spinal cord damage—sustained during a stroke, for example—by evaluating a patient’s supination ability. “This is a classic neurological sign for corticospinal tract injury,” says Jason Carmel, director of the Motor Recovery Laboratory at the Burke Medical Research Institute. Patients with damage to the corticospinal tract usually show arm drift when asked to hold their arms out palms-up, as if holding up a pizza box. But researchers can’t ask rats to pretend to hold a pizza box. So Carmel, also an attending neurologist, developed a supination task for rats to quantify the effectiveness of therapies—ultimately to find the best therapies to help his patients regain motor function.
Carmel worked with scientists at Vulintus, whose forelimb assessment system MotoTrak was designed to replace the near-constant human supervision required for traditional motor assessments with precision sensors and computer monitoring for behavioral reinforcement. Pretraining animals for motor assessment tasks can take weeks or months, via traditionally one-on-one training, which can monopolize the lab’s time. MotoTrak automates almost all of an animal’s training and testing. Now, says Carmel, “we can get studies done more quickly and use a larger cohort of animals.”
MotoTrak uses an adaptive training protocol, automatically adjusting the task parameters required for rewards to keep the animal motivated. In MotoTrak’s supination module, “animals are trained to reach for a doorknob-like manipulandum [object] and turn the knob a specified number of degrees to trigger a [food pellet] reward,” says Andrew Sloan, director at Vulintus. Once rats are fluent in the supination task, MotoTrak automatically tests how well they supinate before and after different therapies for corticospinal tract injury. These therapies include standard neurorehabilitation exercises, and future work may incorporate electrical stimulation of the brain and/or spinal cord. Carmel hopes to use this new device to develop therapies for corticospinal tract injuries that will benefit his patients.
Home sweet home
Rodents are the most common models for behavioral studies, and the advent of knockout technology has made mice especially interesting subjects. Traditionally, lab mice live in home cages—not their natural habitat, but as “normal” as possible for mice who live in a lab—with researchers removing them temporarily to test devices such as mazes or social recognition chambers. But increasingly, researchers are recognizing the advantages of testing animals in their home cages whenever possible, as it reduces stress on the animals. Testing mice in a home cage environment is especially important when measuring behavioral parameters that might be more easily confounded by the effects of human handling or changes in environment. Several companies offer home cages that double as behavioral testing chambers.
The PhenoTyper home cage from Noldus Information Technology (NIT) was specifically developed for behavioral tracking—a video camera and lights on top are optimized for video recording, eliminating shadows and illuminating the cage floor uniformly. “In the PhenoTyper, animals can move about, burrow in bedding, eat and drink, dig holes, and play—[it’s cozier] than the older test cages, which are bare plastic and far from a natural environment,” says Lucas Noldus, founder and managing director of NIT.
Increasingly, researchers are recognizing the advantages of testing animals in their home cages whenever possible, as it reduces stress on the animals.
PhenoTyper is currently used in many behavioral studies of knockout mice. NIT is a partner in the Innovative Medicines Initiative project EU-AIMS (European Autism Interventions—A Multicentre Study for Developing New Medications), a consortium of academia and pharmaceutical companies searching for therapeutic drugs for autism spectrum disorder. NIT is providing the PhenoTyper cage and software for preclinical studies at Utrecht University Medical Center, the Netherlands. Hoffman-La Roche, Pfizer, and other partners are testing the effects of drugs on behaviors in mouse models for autism.
Putting the “move” in movie
In conjunction with the PhenoTyper cage, NIT offers EthoVision XT automatic behavior recognition software, which monitors rodents in their normal environment around the clock by video tracking. Unlike other systems that track an animal as one center-of-mass point, EthoVision finds the animal’s contour and follows it as three points—nose point, center of mass, and tail base—connected by two vectors, and tracks many other body features extracted from the video image.
Tracking an animal using points and vectors allows EthoVision to derive much more information automatically. For example, the angle formed by the two vectors (which intersect at the center of mass) tells you whether the animal’s body is stretched out straight, curled up, or turned to the left or right. The direction of the front vector indicates the direction an animal is heading (if moving), and its field of view. If the back vector is stationary and the front vector is moving side-to-side, then the animal is moving its head back and forth, scanning its environment.
EthoVision, which automatically recognizes 10 different behaviors, has been used to develop animal models for Parkinson’s and Alzheimer’s diseases, and to screen for therapeutic drugs by looking for behavioral changes in the animal models, says Noldus. NIT also works with other technology companies, such as Data Sciences International, which manufactures wireless telemetry sensors. A small microchip-like sensor injected into an animal can transmit vital signs such as body temperature and heart rate to EthoVision, which synchronizes the physiological vital signs to the behavioral data. “In a resident–intruder paradigm, this lets you ask if the heart rate is elevated when the intruder is confronted by an angry resident animal that is defending its territory,” says Noldus, “and whether the stress on the animal is suppressed by anxiolytic drugs.”
Mind over food: Latest behavior tracking technology
Researcher Catherine Kotz, a professor in the Department of Food Science and Nutrition at the University of Minnesota and the Minneapolis VA Medical Center, along with collaborators Charles Billington and Jennifer Teske, are also applying new behavior tracking technology in obesity research. The team applies multiple techniques in studying the role of the neuropeptide orexin in the brain’s regulation of eating behaviors and energy expenditure in rodents.
Kotz uses Promethion system home cages from Sable Systems, because they are identical to the animals’ home cages and do not require an adaptation period—which she emphasizes is especially important for energy balance measurements. “A characteristic sign of stress for these animals during acclimatizing to a new environment is weight loss, so you can’t study weight loss,” says Kotz. “All the assumptions that go into measuring energy being used by metabolic rate and physical activity must be met, and this requires a stable energy balance.” And the Promethion system offers more sensitive measurements of food and water intake, body weight, physical activity, and energy expenditure than any system she has previously encountered. In addition, it allows for measurement of total energy expenditure instead of energy expended due to metabolic rate only. “This system enables us to get much better data,” she says.
Orexin affects eating behavior as well as physical activity level, so Kotz wants to measure the energy expended during particular activities. “Other systems are OK for measuring basic metabolic rate, but if you want to measure the calories burned during a certain behavior, the other systems don’t work,” she says. The Promethion system lets her do this, because it can return synchronized behavioral and metabolic information in real time. It uses light beams in the X, Y, and Z planes so that beam breaks record the animal’s movement path inside the cage. In addition, the cage contains a balance to measure body weight, a running wheel, and sensors on the food and water dispensers. This information can be paired with time-stamped energy expenditure data to determine the calories burned during a certain behavior.
Orexin is produced in the hypothalamus and projects throughout the brain. To study its effects, Kotz is applying the technique of optogenetics, using mice engineered to express light-sensitive ion channels in orexin-producing neurons. When light is pulsed onto the brains of these mice via a small cable attached to the head, the neurons release this peptide. Kotz is also applying the pharmacogenetic DREADD (designer receptor exclusively activated by designer drugs) approach to define orexin action for physical activity energy expenditure. Kotz’s lab is verifying the electrophysiology of the engineered mice in collaboration with University of Minnesota neuroscientist Mark Thomas, and it soon plans to study their behaviors in response to light stimulation using the Promethion system.
Kotz is especially interested in the behavioral outcomes of the optogenetics and DREADD approaches because in previous work, they introduced orexin into the brain by injection. The hypothalamic neurons that release orexin also release other substances at the same time, possibly leading to different behaviors than would be attributable to orexin alone. With optogenetics and DREADD, “we’ll be able to look at the physiology of the whole orexin neuron and everything it releases when stimulated—and the behaviors that follow,” she says.
Of mice and men: Social behavioral monitoring
Researchers from MRC Harwell, a Medical Research Council institute in the United Kingdom, are analyzing behaviors of knockout mice in conjunction with the International Mouse Phenotyping Consortium. Their goal is to look at the pleiotropic effects of knocking out every protein-coding gene in mice—a feat that involves characterizing over 20,000 knockouts with many hours of behavioral observation. “We are hoping that 24-hour monitoring will allow us to see subtle changes to the behavior, feeding patterns, activity, and social interactions of many of our mutant lines,” says Sara Wells, director of the Mary Lyon Centre at MRC Harwell.
MRC Harwell recently adopted the Actual Analytics system for behavioral studies of knockout mice, using the ActualHCA home cage environment that fits into their high-density cage racks. According to Patrick Nolan, neurobehavioral group leader at MRC Harwell’s Mammalian Genetics Unit, the system has two advantages: the ability to monitor behaviors in an undisturbed home cage environment, and the opportunity for group housing. Social animals by nature, mice are healthier and less stressed in a group-house setting than when housed alone, which also allows researchers to study social interactions.
Automatic behavioral tracking of more than one animal has always been tricky, because the monitoring system cannot always distinguish animals. Actual Analytics makes this possible using radio frequency identification (RFID) technology, in which a microchip is placed just under the skin of each animal (just like microchipping your pets at the vet). Underneath the ActualHCA cage is an RFID reader that sends information to a computer to help track the identity of each animal. An array of detectors and high-definition video also tracks the animals’ behaviors. “Our analytical tools measure a range of spontaneous behaviors such as locomotion, circadian rhythms, climbing, and interactions between animals,” says Douglas Armstrong, chief science officer of Actual Analytics. “These are under constant development, with new behaviors regularly being added.”
For Nolan’s group, which studies the genetic basis of neuropsychiatric disorders, the ActualHCA (home cage analysis) system lets them record social behaviors more accurately, and tests whether mixing genotypes in a single cage influences behaviors. He hopes that studying mouse mutants will yield insight into behaviors that might “provide benefits not only for our specific research domains, but also for neurological and neurodegenerative disease research in general.”
Touchscreen technology taps into brain behavior
At the forefront of cognitive testing in mice is touchscreen technology, which is generating excitement because such testing is easily translated from mouse models to humans. Mice are trained in a Bussey-Saksida touchscreen chamber, named for its developers Tim Bussey and Lisa Saksida, both researchers at Cambridge University in the United Kingdom. Inside the chamber is a touch-sensitive video screen that displays images. Mice learn that when they touch the correct image(s) on the screen with their noses (so called “nose-pokes”), they are rewarded. Each correct nose-poke earns them something tasty, like a droplet of strawberry milkshake. Several companies offer Bussey-Saksida chambers, including Campden Instruments, Lafayette Neuroscience, Med Associates, and Stoelting.
At Charles River Discovery (part of Charles River Laboratories), head of translational biology Maksym Kopanitsa is using these touchscreen chambers to study cognition in mouse models of brain disorders such as Alzheimer’s and Huntington’s diseases. Kopanitsa’s group wants to know which cognitive parameters change the most in mouse disease models, so that Charles River’s clients can then use those parameters to test novel drug treatments—perhaps finding a drug that reverses the altered parameter back to that of normal mice.
The touchscreen tests used with mouse models of Alz-heimer’s disease resemble those used in human Alzheimer’s patients, making translation easier. “Touchscreen technology allows one to evaluate mouse models in tasks that are directly relevant to human learning and memory, and to test how well drugs improve performance on those tasks,” says Kopanitsa. He hopes to eventually perform the touchscreen experiments simultaneously with electrophysiological measurements of brain activity, in order to observe neuronal circuits engaged during cognition. “Having these measurements will essentially close the circle by providing evidence that a mouse model reproduces a human brain disorder behaviorally and physiologically.”
Armed with touchscreen tests and other techniques, today’s neuroscientists are opening new behavioral windows into how the brain functions. As researchers come to better understand the link between function and behavior, discoveries may be translated into better therapeutics for some of our most devastating medical disorders.