Compatibility of Gd-DTPA perfusion and histologic studies of the brain Richard G. Spencer a, 4 , Kenneth W. Fishbein a , Aiwu Cheng b , Mark P. Mattson b a Nuclear Magnetic Resonance Unit, National Institute on Aging, Intramural Research Program, Baltimore, MD 21224, USA b Laboratory of Neurosciences, National Institute on Aging, Intramural Research Program, Baltimore, MD 21224, USA Received 30 June 2005; revised 16 October 2005; accepted 16 October 2005 Abstract Histology, including immunohistochemistry, and magnetic resonance imaging microscopy (AMRI) are complementary techniques for the analysis of brain structure. Therefore, AMRI analysis, often of formalin-fixed tissue, precedes histologic evaluation of the same experimental animal in many studies. However, the application of gadopentetate dimeglumine (Gd-DTPA), while of value for MRI studies, has an unknown effect on subsequent histology. We demonstrate here that for the mouse brain, histology with Nissl staining and immunostaining for microtubule-associated protein 2, using standard techniques for tissue preparation, are unaffected by prior perfusion of the tissue with Gd-DTPA. This conclusion was based on qualitative morphologic comparisons of stained sections, as well as quantification of mean immunofluorescence pixel intensities from Gd-treated (meanFS.D. = 131.2F28.4; n = 3) as compared to nontreated specimens (116.2F34.7; n =3, P =.7). Therefore, Gd-DTPA may be applied as a AMRI contrast agent in formalin-fixed brain tissue prior to histologic studies. D 2006 Elsevier Inc. All rights reserved. Keywords: Gd-DTPA; Brain imaging; MR microscopy; Histology; Immunohistochemistry 1. Introduction Magnetic resonance imaging microscopy (AMRI) has become an established technique for microstructural studies, with mouse phenotyping being one of the most important applications. It is of value for its ability to acquire data nondestructively, permitting subsequent invasive tissue analysis for correlative studies, or longitudinal experiments in live animals. Further, the variety of available AMRI pulse sequences and techniques allows for a number of distinct contrast modalities to be applied to the same tissue. Other important features of AMRI are the ability to collect true three-dimensional data, permitting accurate visualization and morphologic studies, and the ever-improving spatial resolu- tion available. Currently, isotropic resolution on the order of 20 Am is available in some instances. The central nervous system (CNS) is of outstanding interest in AMRI. This is due not only to its obvious central role in many animal studies, but also because of the ability to image this (largely) nonmoving structure with very high resolution, permitting visualization of its richly detailed tissue structure [1]. Limitations of AMRI for CNS studies include flat contrast, rendering structures difficult to discern. While different pulse sequences and pulse timings can be used to address this problem, there remain limitations related to increased acquisition time and potential image artifacts. In addition, imaging times can be very lengthy, particularly for three-dimensional acquisitions and studies at high spatial resolution. This problem is exacerbated by the longer T 1 relaxation times typically seen at the higher field strengths often used for AMRI studies. Sample degradation during lengthy imaging studies can be prevented by use of formalin as a fixative, which also constitutes a first step in tissue preparation for histology. The above considerations indicate that use of a contrast agent such as gadopentetate dimeglumine (Gd-DTPA) can be very useful in AMRI studies of formalin-fixed brain. First, it can result in greatly improved tissue contrast. Second, it significantly decreases tissue T 1 ’s, permitting more rapid image acquisition with higher signal-to-noise than in non- contrast studies [2]. In spite of the utility of optimized AMRI studies, histology and immunostaining remain the gold standard techniques for evaluating brain tissue. These methods have exquisite spatial resolution and can probe tissue at the 0730-725X/$ – see front matter D 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.mri.2005.10.017 4 Corresponding author. Tel.: +1 410 558 8226; fax: +1 410 558 8318. E-mail address: spencer@helix.nih.gov (R.G. Spencer). Magnetic Resonance Imaging 24 (2006) 27 – 31