Image courtesy of R. Mark Henkelman,
Mouse Imaging Centre, Hospital for Sick
Children, Toronto
Magnetic resonance imaging (MRI) isn't just for humans anymore. Using MRI, mouse model phenotypes can be determined in live animals, without the loss of valuable research subjects. Compared to MRI in the human, larger magnetic fields and longer imaging times are necessary to obtain high-resolution images of such small subjects. Therefore, these experiments can be both expensive and time consuming. A new technique that allows the imaging and analysis of multiple mice at once is described in a recent study in the Journal of Neuroscience.
In this study, the authors used wild-type and cerebellar deficient folia (cdf) mutant mice. The Catna2 gene, which encodes the structural protein αN-catenin, is truncated in mice homozygous for the cdf mutation. Catenins are thought to link the actin cytoskeleton with cadherins, which are cell adhesion molecules. The cdf mutation causes hypoplasia and abnormal lobe formation in the cerebellum.
The authors injected mice with a contrast agent and administered anesthesia. To ensure each mouse was in a fixed position, the authors secured the mice in sleds. The sleds were placed into centrifuge tubes modified to allow the constant flow of anesthesia and oxygen. The authors then placed the tubes inside the 'hive,' a centrifuge rotor-shaped, MRI machine adaptor with slots for up to 16 tubes. The authors used a 7 Tesla magnetic field, much larger than the 1.5 Tesla field commonly used for humans. The imaging session lasted approximately 3 h and obtained images with a resolution of approximately 150 μm.
The authors averaged the images taken at approximately the same level for the wild-type mice to create an atlas image. A mutant atlas image was also created and compared to the wild-type atlas image. Structures that differed by more than 300 μm between the atlas images were defined as structures of interest. The authors used image analysis software to manually select or 'paint' structures of interest in the wild-type atlas image. The authors then transformed the painted structures from the atlas image back to the original images regardless of genotype, allowing fast and objective selection of anatomical structures in the individual mice.
Image analysis software determined the volume of the 3D painted structures in each brain. Consistent with previous studies, the cerebellum was 18% smaller in cdf/cdf compared to wild-type mice. The authors also detected previously unreported volumetric differences. The inferior colliculi and olfactory bulbs were 38% and 14% smaller, respectively, in the mutant compared to the wild-type mice. Histological studies suggested that the olfactory bulb deficit was due to a defect in lamination.
Therefore, this new technique uncovered previously unknown phenotypes for a well-studied mutant. The authors admit that their method is unlikely to detect subtle cellular defects. However, this technique shows strong potential as a screening method in mutagenesis studies in which founder animals are precious.