Detection of Amyloid Plaques in Mouse Models of Alzheimer’s Disease by Magnetic Resonance Imaging Jiangyang Zhang, 1,2 Paul Yarowsky, 3 Marcia N. Gordon, 4 Giovanni Di Carlo, 3 Sanjay Munireddy, 3 Peter C.M. van Zijl, 1,5 and Susumu Mori 1,5 * We performed three-dimensional, high-resolution magnetic resonance imaging (MRI) of fixed mouse brains to determine whether MRI can detect amyloid plaques in transgenic mouse models of Alzheimer’s disease. Plaque-like structures in the cortex and hippocampus could be clearly identified in T 2 - weighted images with an image resolution of 46 m  72 m  72 m. The locations of plaques were confirmed in coregistra- tion studies comparing MR images with Congo red-stained histological results. This technique is quantitative, less labor- intensive compared to histology, and is free from artifacts re- lated to sectioning process (deformation and missing tissues). It enabled us to study the distribution of plaques in the entire brain in 3D. The results of this study suggest that this method may be useful for assessing treatment efficacy in mouse mod- els of Alzheimer’s disease (AD). Magn Reson Med 51: 452– 457, 2004. © 2004 Wiley-Liss, Inc. Key words: MRI; Alzheimer’s disease; mouse; amyloid; plaque; microimaging Transgenic mouse models of human Alzheimer’s disease (AD), such as that involving overexpression of mutant human amyloid precursor protein (mAPP), have been shown to possess behavioral and pathological features re- sembling those found in AD, i.e., progressive learning and memory impairment, and increasing amounts of A de- posits and amyloid plaques (1). Previous studies also found that the addition of mutant presenilin-1 transgenes to mice transgenic for mutant APP caused an accelerated rate of A deposition (2– 6), with a strong correlation be- tween the extent of A pathology and behavioral impair- ment in specific water maze tasks (7). The successful es- tablishment of mouse AD models is a powerful step to- ward the development of therapeutic measures. For example, vaccination with A peptide results in a dra- matic reduction of the amyloid deposition (8,9). In previous studies with mouse models, the extent of amyloid deposition (plaque burden) was quantified based on contrast in multiple histology sections. While histology remains an important process for characterizing the chem- ical and cellular properties of plaques, it is not an ideal technique for efficient screening of 3D plaque distribution and changes in this distribution resulting from gene alter- ations, aging, and interventions. Although it is theoreti- cally possible to obtain continuous histology slices to cover the entire brain, the process is extremely labor- intensive and time-consuming. Tissue deformation and damage during sectioning further compound the problems inherent in histology-based assays. Magnetic resonance imaging (MRI) technology is espe- cially promising for assessing the global anatomical status of mouse brains (10 –12) because it can provide a 3D data set of the entire brain without sectioning, and is free of artifacts related to brain deformation or missing tissue due to imperfect sectioning. The data are intrinsically digi- tized, and quantification is more straightforward. Com- pared to histological methods, MRI-based techniques have two significant limitations. First, the image resolution is much lower than that in histology, and cellular informa- tion cannot be obtained. The limitation in resolution is dictated by translational motion of water molecules, which is on the order of 2–20 m. The second limitation is the limited availability of imaging contrasts. Similarly to the use of staining techniques in histology, different types of contrasts can be produced by MRI. These contrasts are based on the different chemical or physical environments of water molecules inside different tissue compartments, which may not guarantee sensitive detection of abnormal- ities or anatomical structures of interest. In order to apply MRI to the measurement of amyloid plaque burdens, one must determine whether it can pro- vide sufficient resolution and contrast to identify the plaques. Benveniste et al. (13) showed that amyloid plaques can be identified using T * 2 contrast in postmortem human hippocampus tissues. In this work, we report the first identification of amyloid plaques in ex vivo murine models using T 2 -weighted images. With the use of a 3D imaging technique, we were able to visualize the 3D dis- tribution of the plaques. METHODS AND MATERIALS Brain Preparation All of the experiments and procedures in this study were approved by the Animal Research Committee of the Uni- versity of South Florida. Fixed mouse half-brain samples of transgenic APP+PS1 mice (N = 2), a transgenic APP mouse (N = 1), and wild-type control mice (N = 2) were 1 Department of Radiology, Division of NMR Research, Johns Hopkins Uni- versity School of Medicine, Baltimore, Maryland. 2 Department of Biomedical Engineering, Johns Hopkins University School of Medicine, Baltimore, Maryland. 3 Department of Pharmacology and Experimental Therapeutics, University of Maryland School of Medicine, Baltimore, Maryland. 4 Department of Pharmacology and Therapeutics, University of South Florida, Tampa, Florida. 5 F.M. Kirby Research Center for Functional Brain Imaging, Kennedy Krieger Institute, Baltimore, Maryland. Grant sponsor: National Institutes of Health; Grant numbers: RO3 HD37931; AG15490; Grant sponsor: NIH/NCRR; Grant number: P41 RR15241. *Correspondence to: Susumu Mori, Ph.D, Department of Radiology, Johns Hopkins University School of Medicine, 217 Traylor Bldg., 720 Rutland Ave., Baltimore, MD 21205. E-mail: susumu@mri.jhu.edu Received 23 December 2002; revised 28 October 2003; accepted 30 October 2003. DOI 10.1002/mrm.10730 Published online in Wiley InterScience (www.interscience.wiley.com). Magnetic Resonance in Medicine 51:452– 457 (2004) © 2004 Wiley-Liss, Inc. 452