Jul 30 2007

Mentally Ill Mice

Takatoshi Hikida, of Johns Hopkins University in Baltimore, and his team have produced a strain of mice that they claim can serve as a useful animal model for schizophrenia. The press reports refer to the lab mice as “mentally ill,” which is not untrue but does tend to cause some confusion. At least they did try to convey the fact that is potentially a huge breakthrough.

Understanding the different kinds of research is necessary to put a new research finding in perspective. The lay press generally does a poor job at this, so it is something I try to address in my blog. This type of research is designed to improve research, rather than to achieve a specific end in itself. It’s like inventing a machine that will improve manufacturing – the machine itself is not something useful to the average consumer, but it may make the manufacture of many products more efficient, and therefore less expensive.

Likewise, research is an industry and it requires tools. Researchers get very excited, and rightly so, when a new research tool is developed, because it may mean that research will now be easier, cheaper, and progress more quickly.

One of the most useful tools for biomedical research is the animal model. It is much easier to conduct research on animals than on people, and you can learn much more from a biological system like a whole animal than you can from just looking at cells in a test tube. Therefore, animal models accelerate biomedical research significantly.

What is an animal model? It is a strain of some laboratory animal that had been modified so that it manifests a specific disease or some biological abnormality. This can be done through genetic modification, by injecting animals with a toxin or other biologically active substance, infecting them with an organism, transferring a tumor, modifying them anatomically through surgery, or subjecting them to some other process. The goal is to create a biological abnormality that as closely as possible behaves like a specific disease that you want to study.

The utility of an animal model is that you can test hypotheses and new treatments on the animals as a way of screening candidates for human research. You can also do basic science studies to gain insight into the mechanisms of the disease.

It is understood that animal models are not perfect. Studying an animal with a model of a disease is not the same thing as studying the actual disease in a human. But some animal models are excellent, while others are imperfect and therefore less useful. The more experience we have with a specific model the more we learn how good it is and how best to extrapolate from studies on the animal model. So animal models become more and more useful over time as we learn more about them.

One of the primary difficulties of psychiatric research is the complete lack of any useful animal model. The primary hurdle was the lack of clear markers in an animal for psychiatric disease. In other words – how do you know when a mouse is depressed? Like much of medical research, you can’t always directly see or measure the thing of interest so you have to find a proxy or a marker than can be measured. Over the years researchers have found clever ways to measure animal behavior to infer mental health – their interaction with other animals, their overall level of activity, and desire to engage in specific activities, the display of abnormal behaviors like self-mutilation, etc.

Still there was the hurdle of how to create a reproducible mental illness in an animal. Hikida and his colleagues have done this by taking a gene that has been associated with a high risk of developing schizophrenia and inserting it into a mouse genome. I am not sure how much validation they have done so far – but the next step will be to test this animal model to see if it produced reliable measurable behaviors to follow. It would also be nice to see if the model predicts response to medications that are known to treat schizophrenia. That would be very useful for then the model could be used to screen for potential new treatments.

It remains to be seen if this first animal model for schizophrenia will prove to be useful, but it sounds like a good start. I will be interested to see if it spawns a new generation of research, perhaps leading to other better animal models.

2 responses so far