This article describes a detailed methodology to obtain flattened tangential sections from mammalian cortices and visualize cortical modules using histochemical and immunohistochemical methods.
The cortex of mammalian brains is parcellated into distinct substructures or modules. Cortical modules typically lie parallel to the cortical sheet, and can be delineated by certain histochemical and immunohistochemical methods. In this study, we highlight a method to isolate the cortex from mammalian brains and flatten them to obtain sections parallel to the cortical sheet. We further highlight selected histochemical and immunohistochemical methods to process these flattened tangential sections to visualize cortical modules. In the somatosensory cortex of various mammals, we perform cytochrome oxidase histochemistry to reveal body maps or cortical modules representing different parts of the body of the animal. In the medial entorhinal cortex, an area where grid cells are generated, we utilize immunohistochemical methods to highlight modules of genetically determined neurons which are arranged in a grid-pattern in the cortical sheet across several species. Overall, we provide a framework to isolate and prepare layer-wise flattened cortical sections, and visualize cortical modules using histochemical and immunohistochemical methods in a wide variety of mammalian brains.
Some of the most significant changes in the brain structure across phylogeny can be observed in the cerebral cortex. Despite significant differences, the cortex of animals follows a common pattern and can be broadly divided in two distinct ways, by layers and areas1. Cortical layers lie parallel to the surface of the brain and vary in number from 3 layers in reptilian cortices2 to 6 layers in mammalian cortices1. Cortical areas on the other hand are distinct regions of the cortex which largely correspond to distinct functionalities, e.g., the somatosensory cortex is involved in the sensation of touch or the visual cortex in processing visual inputs. These cortical areas can often be subdivided into patches or modules3, which are regularly repeating anatomical structures, essentially found parallel to the pial surface of the brain. Cortical modules may be confined to a particular layer4, or extend across several layers5.
Standard sectioning methods of the brain involve sections normal to the surface of the brain, like coronal or sagittal. While these methods can be used to visualize cortical modules, a multitude of interesting features can be revealed when the cortical modules are visualized tangentially, in a plane parallel to the surface of the brain. For instance, somatosensory modules in the rodent brain representing whiskers, appear as barrels when visualized normal to the brain surface, and thus the regions derive the name barrel cortex. However, on visualizing the barrels in a tangential orientation, they reveal a whisker-map, with the barrels being laid out in a topographic orientation mirroring the exact layout of the whiskers on the external body surface. In certain cases, modular arrangement has even escaped detection for considerable periods, when visualized in a non-tangential manner. The medial entorhinal cortex, is known for the presence of grid cells, neurons which fire in a regular hexagonal pattern when an animal is traversing an environment. Even though it is a heavily investigated area, until recently, the presence of patches or modules of cells in the medial entorhinal cortex, which are physically laid out in a hexagonal pattern6, had escaped detection. The presence and arrangement of these modules, in the rat brain, was facilitated by making tangential sections of the medial entorhinal cortex and investigating the cytoarchitecture in a layer-wise manner.
Subsequent to sectioning, the particular aspect of visualization of cortical modules can also be realized in multiple ways. Classically, studies have delineated modules based on cell density or fiber layout1. Another popular approach is the use of cytochrome oxidase histochemistry, which reveals areas of higher activity8. Newer approaches include looking at genetically determined cell types, distinguished on the basis of their protein expression profiles6,8.
In this study, we highlight methods to isolate the cortex from mammalian brains, obtain flattened tangential sections, and visualize cortical modules based on cytochrome oxidase histochemistry and immunohistochemistry of cell-type specific proteins.
All experimental procedures were performed according to the German guidelines on animal welfare under the supervision of local ethics committees (LaGeSo). Human and bat brain data were derived from Naumann et al.5 The following procedure is performed on a male adult Wistar rat (strain: RJHan:WI).
1. Perfusion and Brain Extraction
NOTE: In order to obtain a homogenously fixed and blood-free brain, transcardial perfusion of the animal is highly encouraged, as residual blood increases unspecific background signal during staining. Nevertheless, it is also possible to obtain flattened sections from un-perfused specimen and stain them. The ease of handling the specimen varies with the concentration of fixative used. Too little fixation increases the risk of handling damage to the brain during flattening and cutting, while too high concentrations lower the flexibility for flattening and the quality of the staining signal.
2. Brain Dissection and Flattening
Figure 1: Schematic representation of workflow for flattening of a rat cortical hemisphere and visualization of modules in somatosensory cortex. Subsequent to the transcardial perfusion, the brain of a mouse was dissected (A). Subcortical structures were removed and cortex was flattened between two glass slides in phosphate buffer (B). Flattened hemisphere (C) was post-fixed, tangentially sectioned, and stained for cytochrome oxidase activity (D). Scale bars = 1 cm. R: Rostral, C: Caudal, L: Lateral, M: Medial. Figure adapted from Lauer et al.23 Please click here to view a larger version of this figure.
3. Cutting Tangential Sections
NOTE: Depending on the requirements of the staining protocols, the cutting procedure and thickness can be adapted. A vibratome was used to cut the hemispheres for further histochemical processing (step 3.2) at 80-150 µm. However, for immunohistochemical processing, thinner sections are desired and a freezing microtome was used for sectioning (step 3.3) at 10-60 µm. See Video 1.
Video 1: Schematic video of tangential sectioning from a rat medial entorhinal cortex and layout of parasubicular and entorhinal modules. The medial entorhinal cortex of a rodent brain is situated at the posterior end of the cortex and is tilted towards the medial and ventral side. Tangential sections are obtained by orienting a knife along this angle. Consequently, appropriate cell-type specific staining reveals modular structures in the medial entorhinal cortex and adjoining parasubiculum. Video adapted from Ray and Brecht8. Please click here to view this video. (Right-click to download.)
4. Visualization of Cortical Modules Using Cytochrome Oxidase Staining
NOTE: Different staining protocols have been developed for histochemical detection of cytochrome oxidase activity, e.g., first by Wong-Riley17 and later modified by Divac et al.18 This protocol is based on the one by Divac et al.18, since the use of nickel-ammonium sulfate (NiAS) results in a higher contrast and better defined modules in stained cortical areas.
5. Visualization of Cortical Modules Using Immunohistochemical Staining
NOTE: Multiple protocols are available for immunohistochemistry, optimized for the specimen and the type of probe. Adaptations can be made as required, by varying concentrations of antibodies, permeabilizing agents, and incubation times. The following protocol leads to good results for detecting a large range of antibodies and visualization by fluorescent probes.
We obtained flattened cortical sections of the somatosensory cortex in a variety of brains, and processed them for cytochrome oxidase histochemistry to visualize the somatotopic modules representing different body parts. This comparative approach allows studying the evolutionary forces that shape cortex, e.g., showing highly conserved representation of mystacial vibrissae in rodents and lagomorpha as barrels21 (Figure 2). In contrast, other body parts such as paws and genitals show variations in their relative size and reflect the specialization to an ecological niche or to sexual selection22,23.
To understand the architecture of the medial entorhinal cortex, we obtained sections parallel to the pial surface. This was mainly achieved by tangential sectioning of the medial entorhinal cortex in mice, rats, and Egyptian fruit bats. In humans, because of the considerably larger size and more undulations in the entorhinal cortex, we gently flattened the cortex subsequent to making a tangential cut of the entorhinal cortex. Subsequently, all brains were cryopreserved and sectioned on a cryostat at 60 µm. Immunohistochemistry was performed on the obtained sections with an anti-calbindin antibody, to visualize the calbindin-positive pyramidal cell modules in the medial entorhinal cortex5 (Figure 3). The calbindin-modules in the entorhinal cortex show a remarkable periodicity across all these brains, and vary in size by only a factor of 10 across ~ 20,000-fold variation in brain sizes5.
Figure 2: Topographical layout of the barrel cortex modules across mammals identified by cytochrome oxidase histochemistry. Tangential sections of layer IV from the somatosensory cortex of (A) mouse, (B) Mongolian gerbil, (C) rat, (D) degu, (E) hamster, (F) rabbit, showing barrels as a highly conserved somatotopic representation of the mystacial vibrissae. Scale bars = 500 µm. M: Medial, L: Lateral, R: Rostral, C: Caudal; Orientation in A also applies to B-E. Please click here to view a larger version of this figure.
Figure 3: Periodic layout of medial entorhinal cortex modules across mammals identified by calbindin immunoreactivity. Tangential sections from layer II of the medial entorhinal cortex of (A) mouse, (B) rat, (C) Egyptian fruit bat, and (D) human, showing a conserved periodic layout of calbindin-positive pyramidal cell modules. Scale bars = 250 µm. M: Medial, L: Lateral, D: Dorsal, V: Ventral, R: Rostral, C: Caudal; Orientation in D also applies to B, C. Please click here to view a larger version of this figure.
Modularity in the cerebral cortex has been identified using a variety of techniques. The earliest studies typically identified cortical modules by either visualizing cell dense regions, or an absence of fibers1. Subsequent methods have utilized the presence of dendritic bundles24, afferents from a particular region25, or enrichment of neurotransmitters26. Here we demonstrate two techniques, (i) cytochrome oxidase histochemistry and (ii) immunohistochemical staining.
Cytochrome oxidase staining has been one of the most popular methods to visualize cortical modules, and has been widely used in the primary sensory cortices27,28. It visualizes modules by staining darker for areas of higher mitochondrial activity17, thereby acting as a proxy for anatomical modules which elicit substantial functional responses. This method has been used to visualize somatosensory cortical modules (Figure 2) and cortical modules in the visual cortex27.
One of the drawbacks of traditional histochemical techniques is that their typical visualization by bright-field light microscopy limits the number of modules that can be visualized simultaneously. However, immunohistochemical methods, in conjugation with fluorescent visualization probes can also be utilized to view particular proteins and identify cortical modules. For instance, thalamic afferents to the sensory cortices project in a somatotopic manner. Thus, visualizing these afferents, using VGluT2 immunoreactivity can also be used to visualize the body maps in the primary somatosensory cortex which we have visualized using cytochrome oxidase activity (Figure 2). Selective proteins can also be used as cellular markers to identify groups of genetically defined cells. Using immunohistochemistry, we identified clusters of pyramidal cells in the medial entorhinal cortex expressing the calcium binding protein, calbindin6 (Figure 3) and determined that clusters of this particular genetically determined group of cells are conserved across evolution5, indicating common microcircuit principles underlying their function. Using multiple fluorophores, we also delineated complementary modules in the medial entorhinal cortex26, demonstrating parallel microcircuits which underlie spatial memory.
Historically, the brain has been sectioned in a plane normal to its position in the body, in either sagittal, coronal, or horizontal orientations. The identification of cortical modules, like the barrel fields in the somatosensory cortex4 also prompted sectioning along a tangential and subsequently in flattened cortical preparations29, though isolated cases of such sectioning were observed much earlier30,31. Subsequent to the identification of the barrel cortex, other cortical modules were also identified in the primary visual cortex27, as well as the presence of a complete body map32 with a somatotopic representation of the body in primary somatosensory cortex. Flattened sections of the cat visual cortex also revealed the topographic organization of orientation columns without the need for additional reconstructions33. Cortical modularity has also been demonstrated in parahippocampal cortices, including the presubiculum and parasubiculum26,34 and medial entorhinal cortex6. Typically flattening a curved object induces distortions in laminar topography – as can be observed by the multitude of projections of the curved earth surface on a flat map35. A relatively simplistic method to partially compensate for these distortions is by introducing reference points at defined distances before flattening, and then measuring the distance between them after flattening. These marks then not only subserve in quantitatively estimating distortions introduced by flattening, but also provide an easy reference point to align consecutive sections. Further crucial aspects for preserving laminar fidelity include ensuring homogenous flattening across the section, and finally sectioning exactly tangential to the laminar layout to obtain optimized laminar cortical sections.
The visualization of cortical modules in flattened tangential sections has played an important part in revealing fascinating structure-function relationships in the cerebral cortex. The topographic representation of whisker and body parts in the primary somatosensory cortex and the presence of an anatomical grid in the microcircuits generating functional grid cells in flattened tangential cortical sections reveal intricate microcircuits details which would be hard to comprehend from other orientations. However, our current techniques provide only a two-dimensional slice of these essentially three-dimensional modules. Using immunofluorescence techniques and tissue clearing methods36,37, it would be possible to visualize these cortical modules as a whole and perhaps reveal further insights into our understanding of cortical maps and modules.
The authors have nothing to disclose.
This work was supported by Humboldt Universität zu Berlin, the Bernstein Center for Computational Neuroscience Berlin, the German Center for Neurodegenerative Diseases (DZNE), the German Federal Ministry of Education and Research (BMBF, Förderkennzeichen 01GQ1001A), NeuroCure, and the Gottfried Wilhelm Leibniz prize of the DFG. We thank Shimpei Ishiyama for excellent graphic design and Juliane Diederichs for excellent technical assistance.
Cytochrome oxidase staining | |||
Cytochrome c from equine heart | Sigma-Aldrich | C2506 | |
3,3'Diaminobenzidine tetrahydrochloride hydrate | Sigma-Aldrich | D5637 | |
D(+)-Saccharose | Carl Roth | 4621.1 | |
Ammonium nickel(II) sulfate hexahydrate | Sigma-Aldrich | A1827 | |
HEPES | Carl Roth | 9105.4 | |
Name | Company | Catalog Number | Comments |
Antigen retrieval | |||
Trisodium citrate dihydrate | Sigma-Aldrich | S1804 | |
Citric acid monohydrate | Sigma-Aldrich | C1909 | |
Name | Company | Catalog Number | Comments |
Phosphate buffer/phosphate-buffered saline/prefix/PFA | |||
Potassium dihydrogen phosphate | Carl Roth | 3904.2 | |
Sodium chloride | Carl Roth | 9265.1 | |
Di-Sodium hydrogen phosphate dihydrate | Carl Roth | 4984.3 | |
Paraformaldehyde | Carl Roth | 0335.3 | |
TRITON-X 100 | Carl Roth | 3051.3 | |
Name | Company | Catalog Number | Comments |
Immunohistochemistry | |||
Calbindin D-28k puriefied from chicken gut, Mouse monoclonal | Swant | RRID: AB_10000347 | |
Calbindin D-28k from recombinant rat calbindin D-28k, Rabbit polyclonal | Swant | RRID: AB_10000340 | |
Albumin Fraction V, biotin free | Carl Roth | 0163.4 | |
Name | Company | Catalog Number | Comments |
Mounting or freezing media | |||
Fluoromount (immunofluorescence) | Sigma-Aldrich | F4680 | |
Eukitt (histochemistry) | Sigma-Aldrich | 03989 | |
Tissue freezing medium | Leica Biosystems | NC0696746 | |
Name | Company | Catalog Number | Comments |
Alcohol dehydration | |||
Ethanol 100% | Carl Roth | 9065.3 | |
Ethanol 96% | Carl Roth | P075.3 | |
2-Propanol | Carl Roth | 6752.4 | |
Xylene substitute | Fluka | 78475 | |
Name | Company | Catalog Number | Comments |
Devices/tools | |||
Microm HM 650V | Thermo Scientific | ||
Jung RM2035 | Leica Biosystems | ||
Dumont #55 Forceps – Inox | Fine Science Tools | 11255-20 | |
Dumont #5 Forceps – Inox Biology Tip | Fine Science Tools | 11252-30 | |
Dumont #5SF Forceps – Inox Super Fine Tip | Fine Science Tools | 11252-00 | |
Bone Shears – 24 cm | Fine Science Tools | 16150-24 | |
Friedman Rongeur | Fine Science Tools | 16000-14 | |
Blunt Scissors | Fine Science Tools | 14000-18 | |
Surgical Scissors – Large Loops | Fine Science Tools | 14101-14 | |
Surgical Scissors – Sharp-Blunt | Fine Science Tools | 14001-13 | |
Fine Iris Scissors | Fine Science Tools | 14094-11 |