Magnetic resonance microscopy of hamster olfactory bulbA histological correlation.код для вставкиСкачать
THE ANATOMICAL RECORD 242:132-135 (1995) Magnetic Resonance Microscopy of Hamster Olfactory Bulb: A Histological Correlation CHEN CHANG AND TAICHANG JANG Institute of Biomedical Sciences, Academia Sinica, Taipei, ROC (C.C.) and Worcester USA Foundation for Experimental Biology, Shrewsbury, Massachusetts 01545 (T.J.), ABSTRACT Background: Magnetic Resonance Imaging (MRI) has been widely used as a noninvasive diagnostic tool for obtaining morphological, metabolic, and functional information from tissue. However, its potential application in observing detailed structure comparable to that of the light microscope has not yet been fully explored. In order to evaluate the usefulness of MR microscopy, a high resolution three-dimensional (3-D) technique was applied to observe the laminar structure of the mammalian olfactory bulb (OB). Methods: Adult male hamsters (Mesocrecitus auratus) were used as an animal model. Hamster OB and the attached anterior olfactory nucleus were removed from the skull for the MRI examinations. The images were performed with a Bruker AMX-400 system equipped with microimaging accessories. T2 weighted 3-D spin echo sequence was used with a field of view of 9 mm and data matrix of 128*128*128.The in-plane resolution was 70*70*70 pm. Histological preparation, including vibratome sectioning at 40 pm and Nissl staining, were used for light microscopic evaluations and comparisons. Results: Five distinct layers from the superficial to the center of the OB were distinguished in the MR images of coronal, sagittal and horizontal slices. As compared to the histological sections at the corresponding cutting planes, the laminar structure of the OB displayed in the MR microscopic images correlated well with its counterparts. Conclusions: MR microscopy is capable of detecting cellular variation of unsectioned and unstained tissue. It can also be easily applied to obtain spatial information with good resolution. It appears to provide a great potential for diagnostic pathology. o 1995 Wiley-Liss, Inc. Key words: Magnetic resonance imaging, MR microscopy, Hamster, 01factory bulb The applications of the magnetic resonance imaging (MRI) have progressed rapidly in recent years. MRI became a n established diagnostic tool to reveal morphological information of tissues and organs for its superior soft tissue contrast in truly non-invasive and three-dimensional (3-D) manner (Morris, 1986; Stark and Bradley, 1992). It draws research interests for its ability to obtain biological parameters such as oxygen consumption, energy metabolism, and blood flow and volume of tissues (Shulman, 1993). In addition, MR microscopy with potential to visualize the cellular structures gradually emerges as a major imaging technique and is expected to have a considerable impact on biomedical sciences (Aquayo et al., 1986; Callaghan, 1991). MR microscopy has several advantages over the conventional histological preparation, including 1) its 3-D data are able to visualize tissue slices in multiplane and -orientation, e.g., coronal, sagittal, and horizontal planes, 2) isotropic resolution, and 3) non-destructive specimen prepara5on. These competitive 0 1995 WILEY-LISS, INC edges drive the present study to focus on the feasibility of MR microscopy by comparing the cytoarchitecture of a biological structure revealed by MRI to that of being processed with conventional histology procedures. In the present study, 9.4 Tesla (T), 8.9 cm vertical-bore superconducting magnet was used and isotropic 3-D data acquisition was applied to evaluate the microscopic structure of the olfactory bulb (OB). The OB was chosen to illustrate the use of MR microscopy for its distinctively layered structure as well as its apparent variations in the cellular organization of each layer (Allison, 1953; Mori, 1987; Shepherd, 1990). These microscopic features provide a good pattern and criteria to validate the sub-millimeter spatial resolution of the MRI. Received September 22, 1994; accepted January 3, 1995. Address reprint requests to Dr. Chen Chang, Ph.D. Institute of Biomedical Sciences, Academia Sinica, Taipei, Taiwan. HAMSTER OLFACTORYBULB 133 MATERIAL AND METHODS Olfactory bulb sampling and histology preparations Male adult Syrian hamsters (Mesocricetus aurutus) were used. Animals were deeply anesthetized with intraperitoneal injection of sodium pentobarbital (80 mg/kg body weight) followed by perfusion through the heart with either 10% phosphate buffered formalin, or with the mixture of 2% paraformaldehyde and 1%formalin, at pH7.4 for 30 min. At the completion of perfusion, the brain was removed immediately and immersed in the same fixative for at least 3 additional h. The pair of OB, including a part of the frontal lobe, was incised coronally a t a level caudal to the anterior olfactory nucleus. The OB was then positioned rostra1 end up and glued onto a stage made of a role of parafilm. The parafilm stage was approximately 1 cm in height. The OB together with the parafilm stage was inserted into a 10 mm NMR tube for further MR imaging. In the histological procedure, the animals were perfused and their OB were removed in the same manner as for the MRI preparation. Serial sections in the coronal, sagittal, and horizontal planes of the OB were cut with a Vibratome a t a thickness of 40 pm and were stained with Nissl (thionin) stain. MR Imaging The MR imaging experiments were performed with a Bruker AMX-400 spectrometer (8.9 cm Oxford verticalbore superconducting magnet) equipped with microimagining accessory. Spin echo version of isotropic Three Dimensional Fourier Transform (3DFT) pulse sequence was used (Edelstein, et al., 1980; Suddarth and Johnson, 19911, consisting of a hard 90" radiofrequency (rf) pulse and hard 180" refocussing rf pulse with two phase encoding gradients between the period of 90" and 180" rf pulses. The 3DFT isotropic volume images were obtained by using the acquisition parameters of an echo time (TE, time between the center of 90" rf pulse and the center of echo) of 80 ms to enhance the spin-spin relaxation effect, a repetition time of 2.5 Fig. 1. Coronal sections of the hamster OB. a: The MRI coronal slice sec, and a field of view of 9 mm with data matrix of showing the cytoarchitectural layers of the OB as repeated bright and 128*128*128. The resulting inplane resolution was dark bands surround the OV-PVZ. b: Nissl stained section of the OB 70*70*70 pm which is comparable to the histological at approximately the same plane as (a). sections. A saddle type rf coil with a diameter of 10 mm was used and its longitudinal axis parallel t o the static magnetic field. The raw data were processed on a Bruker X32 data station with the 3-D image display tology sections cut a t approximately the same plane. Coronal, sagittal, and horizontal sections of the hamand analysis package. ster OB processed with MRI and conventional histology RESULTS are illustrated in Figures 1, 2, and 3, respectively. The The cytoarchitectural layers of the hamster OB ob- three cutting planes of the OB prepared with histologserved in the MRI slices correlated well to that of his- ical procedure (Figs. lb, 2b, and 3b) were obtained from three animals, whileas the MRI of the three corresponding planes (Figs. l a , 2a, and 3a) were done with a single pair of OB a t one performance of 3DFT imaging acquisition. Abbreviations: The MRI contrast was based on the differences in AOB accessory olfactory bulb EPL external plexiform layer spin-lattice (Tl) and spin-spin (T2) relaxation times, GL glomerular layer and proton density of water in different tissues. NuGRL granule cell layer clear relaxation is due to several different mechanisms IPL internal plexiform layer combined mitral cell body and external plexiform layers and has been reviewed extensively in the literature EPL MBL mitral cell body layer (Bloembergen et al., 1948; Bottomley et al., 1984; EdelOB olfactory bulb stein et al., 1983; Solomon, 1955; Stark and Bradley, ONL olfactory nerve layer 1992). No attempt was made in the present study to ov olfactory ventricle PVZ periventricular zone specify the density-weighted, TI weighted, or T2 134 C. CHANG AND T. JANG Fig. 2. Sagittal sections of the hamster OB. a: MRI slice of the OB at the sagittal plane. b: Nissl stained OB through the same plane. Note the AOB located at the dorso-posterior region (parenthesis) in both (a) and (b). Fig. 3. Horizontal sections of the hamster OB. a: MRI horizontal slice of the OB. The photograph was selected at rather dorsal aspect to show the AOB (arrow). b Nissl stained OB at the same plane. AOB is indicated by a n arrow. weighted image. The acquisition parameters were chosen to highlight the best contrast among the different organizations of the OB a t the field strength of 9.4 T. The OB is well known for its neat layering and orderly repetitive cytoarchitecture (Mori, 1987; Shepherd, 1990). Figure l b shows the Nissl stained coronal section of the hamster OB. From the superficial to deep region, five layers were recognized based on their cellular organization, olfactory nerve layer (ONL), glomerular layer (GL), external plexiform layer (EPL), mitral cell body layer (MBL), internal plexiform layer (IPL), and granule cell layer (GRL). The center or the innermost area of the OB was occupied by a narrow band composed of the olfactory ventricle (OV) and periventricular zone (PVC, Mori, 1987; Shepherd, 1990). Figures 2b and 3b are the sagittal and horizontal sections of the OB, respectively. The same laminar arrangement was recognized in these sections. Since the OB has greater length in both dorso-ventral and antero-posterior axes than in the medio-lateral axis, the OV as well as each surrounding layer in the coronal and horizontal sections exhibited roughly a n ovalshaped profile (Figs. 1 and 3). In addition, the accessory olfactory bulb (AOB) at the postero-dorsal region of the OB was visible in the sagittal and horizontal section (parentheses in Fig. 2a,b, and arrows in Fig. 3a,b). Figures l a , 2a, and 3a are 3DFT MRI slices of the OB in the coronal, sagittal, and horizontal planes, respectively. As compared to the histology sections, the MR images displayed a n equally apparent layered-configuration. With the exception of the MBL and IPL, each layer observed in the histological sections was easily distinguished and assigned to a corresponding layer in the MRI slices. The MBL was a single-cell layer, and the IPL was a very thin layer of dendrites and axons (less than 100 pm in the hamster). Other layers found in the MRI slices of the OB appeared to be orderly repeated bright and dark bands surround the narrow 135 HAMSTER OLFACTORY BULB OV and PVZ. This characteristic lamination is particularly obvious in the coronal and horizontal MRI slices (Figs. l a and 3a). The OV and PVZ were located in the center of the OB, and together they represented a narrow and elongated dark band in the MRI slices. The GRL and EPL could respectively be recognized as a bright and a dark band that surrounded the OV and PVZ. The GL was found as the sequentially outer bright layer, and the ONL as the outermost, bright, and discontinued layer. With a similar layer-to-layer corresponding assignment, each layer of the OB in the MRI sagittal slice matched its counterpart in the histological sections very well (Fig. 2a,b). In general, the layer with more cellular elements was represented as a brighter area in the MRI slice. The GL and GRL possessed numerous small periglomerular cells and granule cells, respectively, and they were recognized as a bright bands in the MRI slices. On the contrary, the layer dominated by dendrites and axons such as the EPL was represented by a dark band. Owing to the postero-dorsal location of the AOB, it was shown only in the sagittal and horizontal planes (parenthesis in Fig. 2b and arrow in Fig. 3b). As seen in these histology sections, the AOB also displayed lamination. The most superficial layer of the AOB was the vomeronasal nerve layer (VNL), immediately bellowed was the GL, followed by the combined mitral cell and external plexiform layers (MBEPL). Deep to the MBEPL was the IPL. The deepest cellular region was occupied by the GRL (Mori, 1987; Shepherd, 1990). Despite the small size, the AOB was very evident in the MRI slices as well (parenthesis in Fig. 2a and arrow in Fig. 3a). In the MRI sagittal and horizontal slices, the most striking feature of the AOB was the IPL. The IPL is predominantly composed of myelinated axons and therefore was represented as a very dark band in the MRI slices. DISCUSSION The present study utilized rodent OB, a well-known layered structure as a model to validate the suitability of MR microscopy to visualize the different cellular components of brain structure. Our data have shown that the high resolution MRI is able to distinguish cytoarchitectural variations in unsectioned and unstained brain tissue. As compared to the histology sections of the OB, the cellular differences displayed with MRI correlated well with that of light microscopy. Furthermore, the MR isotropic 3-D volume acquisition is able to acquire one data set visualizing the entire sample and its serial sections a t any selected planes or angles, and to obtain various specimen volume measurements all in one single acquisition. The same principle can be applied to other brain regions. A layer-tolayer assignment on the MRI slices of the hamster OB in all three different cutting planes, as described in our study, has demonstrated that the MR microscopy was comparable to histological sections. The MRI can “slice” the brain tissue and collect its serial section images without involving tedious histological procedures. Most importantly, it provides images with quality to reveal sub-millimeter structure differences. Although as where detailed sub-cellular structure differences. Although where detailed subcellular organization of a tissue is required, an autopsy using conventional histological processing with light microscopy still is the procedure of choice to characterize pathological changes. Improvement of MR hardware and imaging processing software in the future would enhance the resolution and availability of MR microscopy. In conclusion,the capabilities of MR microscopy complement conventional histology and enable investigators to avoid massive animal sacrifice and time consuming tissue preparation procedures. It provides a quick and convenient way to visualize loci of interest buried in the brain and at the same time make morphological observations in serial sections as well as in 3-D structure. ACKNOWLEDGMENTS This work is supported by a grant from the National Science Council of the Republic of China (NSC83-0203B001-102). LITERATURE CITED Allison, A.C. 1953 The morphology of the olfactory system in the vertebrates. Biol. Rev., 28t195-224. Aquayo, A,, S. Blackband, J . Schoeniger, M. Mattingly, and M. Hintermann 1986 Nuclear magnetic-resonance imaging of a single cell. Nature, 322~190-191. Bloembergen, N., E.N. Rurcell, and R.V. Pound 1984 Relaxation effects in nuclear magnetic resonance absorption. Phys. Rev., 73: 679-710. Bottomley P.A., T.H. Froster, and R.E. Arershinger, and L.M. Pfeifer 1984 A review of normal tissue hydrogen NMR relaxation times and relaxation mechanisms from 1-100 MHz. Med. Phys., 11: 425-428. Callaghan, P.T. 1991 Principles of Nuclear Magnetic Resonance Microscopy. Clarendon Press, Oxford, pp. 173-227. Edelstein, W.A., P.S. Bottomley, H.R. Har, and L.S. Smith 1983 Signal, noise, and contrast in nuclear magnetic resonance (NMR) imaging. J . Comput. Assist. Tomogr., 7:391-401. Edelstein, W.Q., J.M.S. Huchison, G. Johnson, and T. Redpath 1980 Spin warp NMR imaging and applications to human whole body imaging. Phys. Med. Biol., 25t751-756. Mori, K. 1987 Membrane and synaptic properties of identified neurons in the olfactory bulb. Prog. Neurobiol., 29.275-320. Morris, P.G. 1986 Nuclear Magnetic Resonance Imaging Medicine and Biology. Clarendon Press, Oxford, pp. 123-188. Shepherd, G.M., and C. Greer 1990 Olfactory Bulb. In: The Synaptic Organization of the Brain. Shepherd, G. ed., Oxford University Press, Oxford, pp. 133-169 Shulman, R.G., A.M. Blamire, D.L. Rothman, and G. McCarthy 1993 Nuclear magnetic resonance imaging and spectroscopy of human brain function. Proc. Natl. Acad. Sci. U.S.A., 9Ot3127-3133. Solomon, I. 1955 Relaxation processes in a system of two spins. Phys. Rev., 99.559-569. Stark, D.D., and W.G. Bradley, Jr. 1992 Fourier Imaging. In: Magnetic Resonance Imaging. Patterson, A.S., ed. Mosby-Year Book, Inc., St. Louis, pp. 21-66, Suddarth, S.A., and G.A. Johnson 1991 Three-dimensional MR microscopy with large arrays. Mag. Resonance Med., 18:132-141.