Bulletin of Experimental Biology and Medicine, Vol. 155, No. 3, July, 2013
399
MORPHOLOGY AND PATHOMORPHOLOGY Brain Morphology Imaging by 3D Microscopy and Fluorescent Nissl Staining
A. A. Lazutkin, N. V. Komissarova, D. M. Toptunov, and K. V. Anokhin Translated from Byulleten’ Eksperimental’noi Biologii i Meditsiny, Vol. 155, No. 3, pp. 381-384, March, 2013 Original article submitted April 4, 2012 Modern optical methods (multiphoton and light-sheet fluorescent microscopy) allow 3D imaging of large specimens of the brain with cell resolution. It is therefore essential to refer the resultant 3D pictures of expression of transgene, protein, and other markers in the brain to the corresponding structures in the atlas. This implies counterstaining of specimens with morphological dyes. However, there are no methods for contrasting large samples of the brain without their preliminary slicing. We have developed a method for fluorescent Nissl staining of whole brain samples. 3D reconstructions of specimens of the hippocampus, olfactory bulbs, and cortex were created. The method can be used for morphological control and evaluation of the effects of various factors on the brain using 3D microscopy technique. Key Words: light planar fluorescent microscopy; ultramicroscopy; neuroimaging; Nissl staining; brain Methods of 3D imaging of organs and tissues are in great demand in modern experimental biology and medicine. The development of new approaches to microscopic studies has been rapidly unfolding during recent years. For example, optical methods have been developed for imaging of large (several millimeters in size) optically transparent samples to an appreciable depth with the resolution of several microns [14]. Light-sheet fluorescent microscopy (LSFM) is one of the most promising technologies of this kind. It actively develops in several modifications: selective plane illumination microscopy [7], ultramicroscopy [1,4], etc. The gist of LSFM consists in imaging of optical “sections” of a transparent tissue specimen due to selective excitation of fluorescence in a narrow layer by means of a fine flat laser beam and recordP. K. Anokhin Institute of Physiology, the Russian Academy of Medical Sciences, Moscow, Russia. Address for correspondence:
[email protected]. A. A. Lazutkin
ing of the signal in the direction perpendicular to the illumination plane. By now, LSFM (ultramicroscopy) was used in neurobiology for detection of EGFR in the brain of transgenic mice [4,6], autofluorescence [4,10], and FITC-lectin-labeled vascular bed [8]. We previously used ultramicroscopy for 3D imaging of whole brain specimens stained by the immunohistochemical method [1]. The sphere of LSFM application in neurobiology is extending, and therefore it is essential to refer the resultant pictures of the expression of transgenes, proteins, and other markers in the brain to the corresponding structures in the atlas. Due to the use of white laser, modern ultramicroscopes allow detection of fluorescence at various wavelengths [6], which permits basal staining of the studied samples with morphological stains. However, there are no available methods for contrasting bulky specimens of the brain without preliminary slicing.
0007-4888/13/15530399 © 2013 Springer Science+Business Media New York
400
Bulletin of Experimental Biology and Medicine, Vol. 155, No. 3, July, 2013 MORPHOLOGY AND PATHOMORPHOLOGY
Nissl staining is a common neurohistological method for detection of the nuclei of neural and glia cells and the basophilic substance (Nissl substance and granules) in the neuronal cytoplasm. This method best fits for contrasting the whole brain specimens. First, modern atlases of the brain of laboratory animals are compiled on the basis of Nissl-stained preparations. Second, though the classical variant of Nissl staining implies the use of aniline, thionin, or cresyl violet, fluorescent stains have been synthesized for this method [13] and now they are manufactured commercially. We have developed a protocol of fluorescent staining of whole specimens of the brain by the Nissl method and created their 3D reconstructions by means of LSFM.
MATERIALS AND METHODS The study was carried out on adult (50-day-old) male C57Bl/6 mice. The animals were kept and all manipulations with them were carried out in accordance with the Order No. 267 of the Ministry of Health of Russia of 19.06.2003 “On Laboratory Practice Regulations in the Russian Federation” and “Regulations of Studies with the Use of Laboratory Animals” (P. K. Anokhin Institute of Physiology, Protocol No. 1, 03.09.2005). The animals were kept in standard cages (36×21×13.5 cm) with free access to water and food at natural day/night cycle. Brain specimens from adult mice not exposed to external factors of any kind were used for staining. Fluorescent staining of the brain by the Nissl method was carried out using NeuroTrace 500/525 Nissl dye (N-21480, Invitrogen) at 500 nm excitation wavelength. The staining protocol was developed on the basis of manufacturer’s instructions, some modificatons improving brain tissue permeability were added [9]. The brain was fixed in 1% paraformaldehyde (24 h, 4oC). After fixation, the preparations of isolated structures were prepared from the whole brain: the hippocampus, hemispheric cortex, and olfactory bulbs. The samples were then postfixed in Dent fixative (100% methanol and DMSO, 4:1) for 4 h at ambient temperature. After postfixation, the samples were treated with Dent bleach (methanol, DMSO, H2O2, 4:1:1) at bright light for 3 h at ambient temperature in order to minimize autofluorescence. The samples were then washed (2×1 h) in 100% methanol at ambient temperature and constant stirring (650 rpm), cooled to -70oC, and passed through 5 freezing/defrosting cycles for increasing brain tissue permeability for the dye. The samples were processed in descending methanols 100-70-50-25-0% (30 min each at constant stirring at 650 rpm) with gradual replacement of methanol with 0.1 M phosphate buffer (pH 7.4) and then washed in 0.1 M phosphate buffer (2×1 h) and Tris-buffer
(0.05 M Tris-HCl, pH 7.5; 0.15 M NaCl, 0.1 M Triton X-100; 30 min with constant stirring at 650 rpm). The samples were then incubated in Tris-buffer with NeuroTrace 500/525 Nissl stain (1:200) at 4oC in darkness at stirring (650 rpm) for 14 h. The incubation solution also contained 5% DMSO and 0.01% sodium azide. The samples were then washed in Tris-buffer (4×1 h) at constant stirring and ambient temperature and in a solution containing 0.01% sodium azide for 14 h at 4oC. The samples were then dehydrated in ascending ethanols (30, 50, 80, 96, 100%, all solutions in 0.1 M phosphate buffer) at constant stirring (1 h in each solution), washed (2×1 h) in 100% ethanol, and plunged in butoxyethanol, in which they could be stored at 4oC. Directly before photographing, the samples were clarified in Murray’s clarifying agent (benzylbenzoate and benzyl alcohol, 2:1) for 14 h. Stained samples were visualized using a ultramicroscopy device of our design [1] fitted with planapochromatic objective lens (M=1×, NA=0.25) and 3.2-megapixel cooled Alta U32 P2X camera (Apogee Inc.). Mathematical processing and 3D reconstruction of the sample images was carried out using Imaris software (Bitplane Inc.).
RESULTS The preparations of the mouse whole olfactory bulb, hemispheric cortex, and hippocampus were wholly stained by the Nissl fluorescent method. Stacks of 300400 images for each sample (optical “sections”) were obtained with an ultramicroscope [1] and used for 3D reconstruction. Qualitative analysis of the preparations was carried out on optical sections and 3D images. Stained cells were characterized by highly intense signal. The basal staining was slight. High contrast (base/signal proportion) was observed in two-dimensional optical and in tomographic sections and 3D projection images (Figs. 1, 2). The cytoarchitecture of the olfactory bulbs was clearly seen and the glomeruli, mitral and granular layers were detected in optical sections (Fig. 1), obtained by direct scanning with a flat laser beam, and in virtual topographic sections in the 3D reconstruction arbitrary plane. In the hippocampus, the pyramidal layer exhibited intense fluorescence. Because of high density of cells in this layer, individual cells could not be discerned at magnifications (up to 4-fold) used in scanning. In the cortical preparations, the borderline of the visual and somatosensory cortex could be detected by higher cell density in layer 4. However, individual barrels in the somatosensory cortex could be detected only in virtual sections (Fig. 2, a) but not in 3D reconstructions because of high cell density in the upper layers 2 and 3 (Fig. 2, b).
A. A. Lazutkin, N. V. Komissarova, et al.
401
Fig. 1. Olfactory bulb preparation wholly stained after Nissl by NeuroTrace 500/525 Nissl Stain (×4). a) optical “section”; b) 3D reconstruction built from 298 “sections”. Gr: granular layer; Mi: mitral layer; Gl: glomerular layer.
Fig. 2. Preparation of the brain cortex, wholly stained after Nissl by NeuroTrace 500/525 Nissl Stain (×1.25). a) optical “section”; b) 3D reconstruction built from 401 “sections”. S1: primary somatosensory cortex. Representation zones: J: jaw; FL: fore limbs; HL: hind limbs; BF: barrel field; DZ: agranular zone; S2: secondary somatosensory cortex.
The level of autofluorescence in our preparations was low. However, sometimes intense fluorescence was recorded in the blood vessels (Fig. 2, b), which decreased significantly after long (up to 24 h) exposure of the preparations in Dent’s bleach in bright light. Despite the fact that dehydration fluorescent-labeled conjugates was not recommended, we dehydrated the samples, as optical clarification was an obligatory condition for ultramicroscopy. We showed that dehydration and subsequent replacement of liquid media by benzyl alcohol and benzyl benzoate was inessential for the fluorescence of NeuroTrace 500/525 Nissl Stain. Analysis of the depth of Nissl’s staining in optical sections showed complete staining of the bulb, cortical, and hippocampal preparations. Stained preparations were up to 7.6 mm long, up to 5.5 mm wide, and 1.5-3.0 mm thick. Hence, the depth of the stain penetration was at least 1.5 mm. Thicker samples could not be stained by Nissl’s method.
Nissl’s staining is the main instrument for not only detection of the borderline between brain compartments (for example, in immunohistochemical detection of various antigens [2]). It is also used for qualitative visual evaluation of morphological and neuroanatomical changes in nerve tissue structure, caused by external factors, for example, injection of neurotoxins, traumas, tumor development, etc. [12,15], or for verification of the cellular neuronal phenotype [5], or in comparative neuroanatomical studies [3]. Our method allows studies of this kind using 3D microscopy methods. Though the samples in this study were visualized by an ultramicroscope, the preparations could be documented by confocal and multiphoton microscopy [11]. The study was supported by the Russian Foundation for Basic Research (grant No. 06-04-08353-ofi) and the Ministry of Education and Science of the Russian Federation (SC No. 02.522.11.2002).
402
Bulletin of Experimental Biology and Medicine, Vol. 155, No. 3, July, 2013 MORPHOLOGY AND PATHOMORPHOLOGY
REFERENCES 1. A. N. Morozov, A. A. Lazutkin, I. V. Turchin, et al., Kvant. Elektron., 40, No. 9, 842-846 (2010). 2. J. Bagley, G. LaRocca, D. A. Jimenez, and N. N. Urban, BMC Neurosci., 8, 92 (2007). 3. K. J. Burman, S. M. Palmer, M. Gamberini, and M. G. Rosa, J. Comp. Neurol., 495, No. 2, 149-172 (2006). 4. H. U. Dodt, U. Leischner, A. Schierloh, et al., Nat. Methods, 4, No. 4, 331-336 (2007). 5. K. D. Dunlap, E. A. McCarthy, and D. Jashari, Dev. Neurobiol., 68, No. 12, 1420-1428 (2008). 6. A. Ertürk, C. P. Mauch, F. Hellal, et al., Nat. Med., 18, No. 1, 166-171 (2011). 7. J. Huisken and D. Y. Stainier, Development, 136, No. 12, 19631975 (2009).
8. N. Jährling, K. Becker, and H. U. Dodt, Organogenesis, 5, No. 4, 227-230 (2009). 9. S. H. Kim, P. Che, S. H. Chung, et al., Curr. Protoc. Neurosci., Ch. 2, Unit 2.10 (2006). 10. U. Leischner, A. Shierloh, W. Zieglgänsberger, and H. U. Dodt, PLoS One, 5, No. 4, e10391 (2010). 11. S. G. Parra, T. H. Chia, J. P. Zinter, and M. J. Levene, J. Biomed. Opt., 15, No. 3, 036017 (2010). 12. N. Pilati, M. Barker, S. Panteleimonitis, et al., J. Histochem. Cytochem., 56, No. 6, 539-550 (2008). 13. B. Quinn, A. W. Toga, S. Motamed, and C. A. Merlic, Neurosci. Lett., 184, No. 3, 169-172 (1995). 14. P. A. Santi, J. Histochem. Cytochem., 59, No. 2, 129-138 (2011). 15. M. Zhou and M. Baudry, J. Neurosci., 26, No. 11, 2956-2963 (2006).
