Microscopic Computed Tomography–Based Virtual Histology for Visualization and Morphometry of Atherosclerosis in Diabetic Apolipoprotein E Mutant Mice
Atherosclerosis is a progressive disease characterized by the accumulation of lipids and fibrous elements in the arteries and is a leading cause of heart disease and stroke in developed and developing countries.1 Animal models have become increasingly important tools for addressing key mechanistic and therapeutic questions that cannot be answered from human studies of atherosclerosis. However, the small-scale vascular structures in genetically engineered mice require labor-intensive histomorphometric techniques to quantify lesions.
Recently, a new technique has emerged to image ex vivo blocks of soft tissue by staining tissue with metal solutions, then scanning with a microscopic computed tomography (microCT) instrument (Figure I in the online-only Data Supplement).2 This technique was originally applied to the study of the developing heart in embryos3 and fetuses (Figure II in the online-only Data Supplement) but can also be applied to the en bloc imaging of the heart, great vessels, and lesions thereof. By this method, tissues are left intact, but one can employ image analysis to create “virtual” histological sections that allow the identification of an individual lesion in a 2-dimensional image. Furthermore, the 3-dimensional nature of microCT scans permits the volumetric assessment of multifocal atherosclerotic lesions.
Shown in Figure 1 are excised hearts from an apolipoprotein E knockout mouse (ApoE null), a well-established murine model for atherosclerosis, as well as a control mouse. Both mice are on a diabetic background strain, KK-Ay, maintained on a standard diet. ApoE null mice are spontaneously hypercholesterolemic and develop fatty streaks as well as advanced lesions at aortic sites where atherosclerosis is also typically seen in humans.4 In these coronal virtual histological sections, the ventricular myocardium and atria can be appreciated (Figure 1A through 1D), as well as atherosclerotic lesions of the aortic root (Figure 1B) and ascending aorta (Figure 1D). After microCT-based virtual histology scans, the ApoE null aortic root was sectioned by traditional histology for staining of atherosclerotic lesions with Oil Red O, which stains the fatty plaque red (Figure 2A), thereby confirming the atherosclerotic lesion seen by virtual histology. With higher-resolution (1.8 μm) microCT instrument scanning, these lesions are even more readily visualized (Figure III and Movies I through III in the online-only Data Supplement).
An advantage of microCT-based virtual histology is that the data are entirely digital and not subject to loss of individual sections in handling or improper plane of sectioning. Furthermore, freely available software tools exist for the visualization of natural edge boundary features of 3-dimensional tissues (Figure 3, left) as well as volume quantification of atherosclerotic lesions at multiple foci to microliter accuracy (Figure 3, right) (Seg3D, http://www.sci.utah.edu/cibc/software).
Thus, evaluation of mechanisms underlying atherosclerosis and interventions can be accelerated by this time-saving, higher-accuracy research imaging technique. In addition, because microCT-based virtual histology is inherently digital, data analysis for virtual histology is poised to take advantage of an emerging new generation of powerful software tools such as those mentioned above, which will further speed vascular disease research.
We thank Itzik Goldberger and Tiffany Fong for the scans graciously provided by Xradia Inc.
Sources of Funding
This research was supported by the Veterans Administration (Merit Review) and National Institutes of Health grant R01 AR 052755 to Dr Ahuja. microCT scans were performed at Numira Biosciences (Salt Lake City, Utah) through a reciprocal scanning agreement. This work was made possible in part by software from the National Institutes of Health/National Center for Research Resources Center for Integrative Biomedical Computing (5P41RR012553).
Dr Keller is cofounder of Numira Biosciences (http://www.numirabio.com), which licenses microCT-based virtual histology from the University of Texas Health Science Center at San Antonio for commercial applications. The other authors report no conflicts.