The development of MRI to scan living brain for amyloid lags behind the impressive progress made recently with the PET-PIB approach (see ARF related news story), but interest remains high because MRI alone holds out the promise of high-resolution imaging of individual plaques. The use of MRI with novel contrast agents presents one promising approach being pursued by Takaomi Saido’s group (see ARF related news story). Taking another tack, last year Clifford Jack, Joseph Poduslo, and colleagues at the Mayo Clinic in Rochester, Minnesota, and the University of Minnesota Medical School described an in-vivo MRI microimaging technique that resolved single plaques in the brain of AD mice, in the absence of any contrast agents (Jack et al., 2004). This technique depends on the iron content of plaques to distinguish the deposits from surrounding tissue.

Now, Jack and colleagues have followed up with further validation of their imaging technique. In a new study just out in the Journal of Neuroscience, they show that MRI of live APP/PS1 double transgenic mice can reveal an age-related increase in plaque number that correlates with the gold standard of thioflavin S histological detection. Their results demonstrate the utility of MRI for the non-invasive tracking of amyloid plaques in mice, a technique that may speed evaluation of anti-amyloid therapies. For humans, though, this kind of MRI still faces several considerable hurdles.

In the new study, Jack gathered four imaging data sets on double transgenic AD mice at 3, 6, 9, 12, and 24 months of age. In-vivo MRI was compared to three in-vitro measures performed on fixed brain: ex-vivo MRI, thioflavin S staining, and tissue iron staining. Image analysis on spatially registered data allowed the researchers to compare the methods at the level of detecting individual plaques.

On MRI, amyloid plaques show up as dark spots in the cortex and hippocampus, and an increase in both the size and density of the spots with age was apparent. Individual spots could be matched precisely with thioflavin S reactive plaques, as well as iron staining. Although MRI detected fewer plaques than did thioflavin S staining, the increase in MRI signal with aging from 9 months on was proportional to plaque load detected by histological methods. The lower sensitivity of MRI meant that the plaques could first be detected between 6 to 9 months, much later than the 10 weeks of age or earlier for thioflavin S staining of postmortem brain, but after that time, the increase in plaque number was easily measured. MRI could detect plaques of roughly 35 micrometer diameter and larger.

MRI in fixed brain sections (ex-vivo MRI) was consistently more sensitive than the in-vivo measures, even though the imaging conditions were identical. One reason for this is that even very slight movement, like the heart beating and the animal’s breathing, hampers high resolution. Nonetheless, within its detection limits, in-vivo imaging should be useful for longitudinal studies, with the caveat that the imaging depends on the amount of iron, not amyloid, in plaques. The linear relationship between MRI signal and thioflavin S staining shown in this study might not hold under conditions where the iron/amyloid ratio in plaques was changing, for example, if amyloid was removed while iron stayed behind.

Extending this technique to humans would mean reducing the scan times (1 hour and 40 minutes in these experiments), and improving methods to reduce motion artifacts. In addition, sensitivity remains a big question. The method only visualized 20 percent of total plaques in the older APP mice, animals that carry much more brain amyloid than the average human. Other techniques that combine contrast agents with MRI may turn out to be more practical for clinical use.—Pat McCaffrey


  1. This paper explores in-vivo MRI plaque number quantitation in PS1/APP doubly transgenic mice and relates these studies to invasive techniques such as ex-vivo MRI and histologic staining. The correlations observed between in-vivo MR plaque load and ex-vivo correlative measures are very encouraging. This study is at the cutting edge and represents the best state of the art for the field of MR amyloid imaging. The study was carefully performed by expert investigators who have extensively published in both MRI (in AD and mice) and histologic evaluation of the PS1/APP mouse.

    It is an important contribution to the literature, and great care has been taken to present the data in a balanced way. That is, the authors have pointed out the issues that remain to be surmounted before human amyloid imaging with MR is a possibility, and they note that a “number of significant technical barriers must be solved for this technique to be viable in the living human subject.”

    Of most importance is the issue of limiting motion to the degree necessary to resolve 100-micron plaques. It is interesting that with the present state of the art in MR imaging, it is no longer the resolution of the neuroimaging technique itself that is the limiting factor. Resolution of inanimate objects at 100 microns is no great feat in MR imaging studies today. The current challenge is to limit the inherent movement of living objects so this optimal resolution can be realized. While this paper and a similar previous report by this group (Jack et al., 2004) represent an impressive tour de force for cardiac and respiratory gating in order to minimize motion artifacts in the mice, it remains a great challenge to apply these approaches to humans. The authors put this work into a realistic context by stating, “This study serves as a foundation for using in-vivo MRI of transgenic mice (or other animal models) as a surrogate measure of plaque burden in natural history studies of plaque biology as well as drug discovery studies….” Regardless of when or if this MR technology is extended to humans, this remains a very important contribution for the animal work alone.

    Although each approach has its advantages, this MR approach is less invasive and measures a larger brain area than comparable multiphoton imaging studies that have demonstrated very high-resolution images of plaques in living mice (Bacskai et al., 2003). MicroPET studies, in addition to having lower resolution, have failed to show plaque detection in transgenic mice to this point [Klunk et al. (in press) and Toyama et al., 2005]. Thus, the latest findings by Jack et al. represent a significant advance in imaging plaques in animal models of AD and present the field with a worthy challenge to extend this technology to human studies.


    . In vivo visualization of Alzheimer's amyloid plaques by magnetic resonance imaging in transgenic mice without a contrast agent. Magn Reson Med. 2004 Dec;52(6):1263-71. PubMed.

    . Four-dimensional multiphoton imaging of brain entry, amyloid binding, and clearance of an amyloid-beta ligand in transgenic mice. Proc Natl Acad Sci U S A. 2003 Oct 14;100(21):12462-7. PubMed.

    . PET imaging of brain with the beta-amyloid probe, [11C]6-OH-BTA-1, in a transgenic mouse model of Alzheimer's disease. Eur J Nucl Med Mol Imaging. 2005 May;32(5):593-600. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Pittsburgh Compound-B Zooms into View
  2. Visualizing Success with MRI of Amyloid Plaques in Live Mice

Paper Citations

  1. . In vivo visualization of Alzheimer's amyloid plaques by magnetic resonance imaging in transgenic mice without a contrast agent. Magn Reson Med. 2004 Dec;52(6):1263-71. PubMed.

Further Reading


  1. . Imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound-B. Ann Neurol. 2004 Mar;55(3):306-19. PubMed.

Primary Papers

  1. . In vivo magnetic resonance microimaging of individual amyloid plaques in Alzheimer's transgenic mice. J Neurosci. 2005 Oct 26;25(43):10041-8. PubMed.