A wake-up call for biomedical research on mouse models

15 December 2015
Tom Vanden Berghe and colleagues from the Peter Vandenabeele Lab (VIB/UGent) have demonstrated how the side effects of genetic modification of mice can complicate the interpretation of biomedical research. Downloaded no less than 15.000 times in one month, the landmark paper caused quite a stir in de biomedical community. What makes the research so remarkable? Peter and Tom shed light on the issue.

By deactivating a specific gene in mouse strains, investigators can study its effect on the development of a disease. However, mouse models alone can’t deliver irrefutable scientific conclusions. Clinical studies with human cells remain essential to validate the results. Moreover, clinical studies often produce different conclusions.

Why is there a ‘problem’ with mouse models?
Peter: Our analysis revealed that each mouse strain contains approximately one thousand genes that result in an abnormal protein. About a hundred of those could actually be attributed to a functional defect. In the first generation of a genetically modified mouse strain, the so-called recombinant congenic mice, we almost always see various other defective genes close to the inactivated gene. This means that in certain cases, researchers cannot be certain whether the inactivated gene, the dysfunctional neighbouring genes, or a combination of both are responsible for the observed effect. Geneticists are familiar with the phenomenon, but it is very often overlooked.

How would you explain the enormous attention for this research?
Tom: I guess it is mainly caused by the shock effect. After all, our study reveals that nearly all congenic knockout mice are stuffed with passenger mutations that might affect the phenotypic outcome. This implies that any researcher working with congenic transgenic mice is potentially affected by our findings. The case studies we provided, have certainly amplified the effect. Some top research institutes have already changed their transgenic procedure to avoid these issues.

"Since the paper was published in July our tool has been used by more than 3800 scientists." Liesbet Martens

What will your research change?
Tom: Our findings will have a profound retro-active impact on the interpretation of a great deal of scientific research. In addition, it will also aid in explaining controversies in scientific literature surrounding certain disease models. Finally, in the long term, these results can contribute to an improved translation of findings from lab animals to humans.

WEB TOOL TO PREDICT ABNORMALITIES
In an effort to tackle the side effects of genetic modification of mice, the team developed a web tool that allows scientists to estimate the impact of this phenomenon more accurately. Tom Vanden Berghe worked with bio-informaticists Liesbet Martens and Paco Hulpiau to develop a tool that gives researchers insight into the presence of possible abnormalities in the vicinity of the targeted gene and their potential effect on research results. Paco Hulpiau: "By combining publicly available
data we were able to pinpoint the tremendous impact of passenger mutations in all congenic knockout mice. Making this accessible in a webtool allows everyone working with mouse models to easily interpret the consequences of passenger mutations in their own research field."

Vanden Berghe et al., Immunity 2015


Back row from left to right: Peter Vandenabeele and Paco Hulpiau. Front row from left to right: Tom Vanden Berghe and Liesbet Martens
© VIB