Strawberry Milkshakes Forever
Research Models
Regina Kelder

Strawberry Milkshakes Forever

Finding a mouse's sweet spot. How touchscreen technology is helping determine if a new therapy improves impaired memory function in AD mice.

Chocolate and vanilla may be the most popular ice cream flavors, but it takes a strawberry milkshake to get research mice to learn to use a tablet.

Touchscreen technology platforms represent a relatively new way of measuring cognitive functions in the studies of mice and rats, notably ones genetically engineered to express many of the features of the human Alzheimer's disease (AD) condition: amyloid plaques, neurofibrillary tangles composed of the tau protein, neuronal loss and behavioral deficits.

Many different tools, methods and models are being used to study AD pathogenesis and to evaluate AD drugs, of course, including different behavioral tests that measure learning and memory. The Morris water maze, one of the most common ones, measures how quickly mice, using visual cues, find their way to a submerged platform in order to escape from the pool of water. Another test uses electric foot shocks to see how well mice can remember the conditions of this unpleasant experience and adapt to their possible repeat.

But these high-stress tests aren't exactly translatable to how one measures signs of AD in humans, a progressive and complicated disease that usually becomes obvious when people can't acquire new memories and have a hard time holding on to the old ones. Touchscreen technology is a way to assess cognitive function in a low-stress environment and determine if a new therapy improves the impaired memory function in AD mice.

Pharmaceutical companies still largely rely on the Morris water maze and other less translational routines, which have been around for decades, but with an estimated 46 million people worldwide suffering from AD or some other form of dementia, and the failure rate of AD drugs extraordinarily high, fresh approaches to modeling AD in animals are also being explored. As more data emerges about the merits of using touchscreen technology in translational medicine, laboratories, including our Discovery Research Services site in Kuopio, Finland, which specializes in diseases of the central nervous system, are adding the service. Senior scientific writer Regina McEnery caught up with Maksym Kopanitsa, a PhD scientist at the Kuopio site, to talk more about how the behavioral tests are conducted  and how they can be used to bridge the translational divide. Here are his edited responses.

Q: What does the platform look like and how does it function?

MK: The touchscreen is built inside an operant chamber and connected to a computer. When different images appear on the screen, the mice poke them with their nose. It20160620-maksymkopanitsa-portrait-200x200 could be a plane and a flower, or a square or a triangle. Animals can be taught to discriminate between negative and positive stimuli by rewarding them with a strawberry milkshake when they pick the right one. Furthermore, other tests explore how quickly the mice react for transient short stimulus appearing in different parts of the screen, or for how long they persist in touching the stimulus after stopping the reward.

Q: Who knew mice liked the taste of strawberries.

MK: Strawberry milkshakes were used in the laboratory of Drs. Tim Bussey and Lisa Saksida at the University of Cambridge, where the technique was invented. However, the same brand wasn't available here in Finland, so we decided to just give the mice glucose and saccharine as a reward. During the pre-training, we noticed that the mice didn't learn things as quickly. So we found a different kind of strawberry shake and the mice started to do better.

Q: Is the screen miniaturized for mice?

MK: No, it is approximately of the size of an iPad. But the screen has to be slightly thicker. A mouse's touch is much lighter than a human's so the screen needs to be more sensitive.

Q: Do mice find it strange being in these chambers?

MK: Probably the most stressful part is that the mice have to be slightly hungry to work. They need to be at 85%–90% of their normal body weight.

Q: How long does it take to teach the animals the fundamentals of touchscreen testing?

MK: Around two weeks. Once they learn to nose-poke onto images appearing on the screen to receive their milkshake, more sophisticated tests can be administered. For example, we may ask how long does it take for the mice to learn that out of the two simultaneously displayed stimuli on the screen, one is correct (and rewarded), while the other one is incorrect (and not rewarded). If the animal pokes an incorrect image there is no punishment, but there are sounds and light signals that tell them the answer is wrong and there will be no food reward. Normal mice usually can correctly nose-poke in 24 out of 30 pairs (i.e., 80% level of accuracy) within 10–12 days.

Q: What is the biggest upside to using touchscreen systems?

MK: Translatability. When evaluating a drug that might be used in humans, you want any test of rodent cognition to be highly translatable. The behavioral tests in rodents have to provide reactions that are closely similar to the reactions that humans experience in their daily lives. The problem is that many animal tests currently used to study memory and learning—like Morris water maze or fear conditioning—are all quite intrusive. The mice will learn something, but how translational is that in the clinical or home setting? These are not measures that you would obtain from human patients. You don't ask old people to swim in a pool of opaque water to find a step and do not probe them with electric shocks in order to see whether they would memorize this manipulation upon their next visit to the clinic. But what you can do is ask them to solve certain tasks on an iPad.

Q: What are the downsides?

MK: Most touchscreen tasks usually take a lot longer to complete than other cognitive tests. It is because more complicated and subtle things have to be remembered. For example, during the touchscreen object location test called Paired Associates learning, mice have to remember that each of the three symbols that may appear on the screen may be correct, depending on its spatial location (left, right or center). This type of learning is more difficult to acquire than an associative freezing reaction upon exposure to the cage or sound that accompanied electric shock treatment. Many rapid cognitive tests create very strong motivations for the memory of simple adverse experiences, whereas touchscreen tests are oriented toward less stressful, gradual operant learning. Also, to be able to complete touchscreen tests, mice have to have good vision, which is a problem with some strains.

Q: So are identical touchscreen tests being used in mice and humans?

MK: There was a paper published last year by biologists and clinicians from Edinburgh and Cambridge Universities that looked at people with a loss-of-function mutation in a particular gene that predisposed them to schizophrenia and genetically engineered mice that had that gene deleted. Initially, they tested humans with a human version of the Paired Associates Learning touchscreen task and mice with a mouse version, but then, they offered humans the mouse task, and still found the humans with the mutations to be performing worse than controls.

Q: What are you learning about the cognitive function in AD mice?

MK: The biggest change in the AD line that we tested recently was revealed when in the Discrimination test, we switched the positive and negative stimuli. The mice had to "forget" the previous association and learn the new reversed relationship between the stimulus identity and reward. It took the mutant mice significantly longer than the controls to re-learn this, so it appeared that these AD mutants had lower cognitive flexibility.

Q: So how ultimately might this be used to evaluate drugs for AD?

MK: Impaired performance in such touchscreen tasks can be used as a gauge to measure the ability of novel anti-AD drugs to bring the altered cognitive parameters back to the level seen in control, non-mutant animals. In addition, it may be interesting to look if learning and attention basal parameters can be improved by drugs in aged control mice. Here, in Kuopio, we are fully equipped to combine touchscreen evaluations of cognition with other behavioral tests, as well as with powerful in vivo imaging approaches and post-experimental in vitro biomarker studies.

How to cite:

McEnery, Regina. Strawberry Milkshakes Forever. Eureka blog. June 20, 2016. Available: