HBM2004

Receptor fingerprinting of cortical areas in the human brain

Karl Zilles

Institute of Medicine, Research Center Juelich, and C. & O. Vogt Institute of Brain Research, Heinrich-Heine University Duesseldorf, Germany

The most influential map of the human cerebral cortex is that of Korbinian Brodmann (1909). Brodmann’s map still dominates present concepts on the microstructural organization of the cerebral cortex, and, via the atlas of Talairach and Tournoux (1988) , the topographical interpretation of functional imaging data.

Brodmann’s research was based on the working hypothesis that the cerebral cortex is composed of multiple cortical areas, each of them characterized by a distinct cytoarchitecture and function. Cytoarchitecture should be more or less constant within a cortical area, and changes abruptly at its border. E.g., Brodmann’s area (BA) 4 was conceptualized as the anatomical equivalent of the primary motor cortex, which guides voluntary movements. Brodmann did not argue for an extreme localizational concept. I.e., he did not try to relate each complex function to one distinct cytoarchitectonic area. He created a “neutral” nomenclature by numbering different cytoarchitectonic areas mainly according to their dorso-ventral sequence. Brodmann’s map and cytoarchitectonic analyses constitute an impressive scientific achievement and they have influenced research on the structural and functional organization of the human cerebral cortex for many decades.

In the context of increasing spatial and functional resolution of data provided by recent imaging techniques, however, several drawbacks of these classical architectonic maps became critical for their practical use as anatomical references. One of the disadvantages is the restriction of the earlier studies to Nissl- or myelin-stained sections, which do not reflect functionally highly relevant information. In contrast, the mapping of receptor binding sites has recently become a powerful tool to reveal the (receptor-) architectonic organization of the cerebral cortex ( Geyer et al., 1996; 1997; 1998; Zilles, 1992; Zilles and Schleicher, 1995; Zilles and Clarke, 1997; Zilles and Palomero-Gallagher, 2001; Zilles et al., 1995, 1998, 1999, 2002a, b) .

- introduce the principles and methodological basis of quantitative in vitro mapping of neurotransmitter receptors in human and non-human primate cortex
- demonstrate advantages and power of receptorarchitectonic brain mapping which allows the definition of complex neurochemical profiles of cortical areas (“receptor fingerprints”) by using a multireceptor approach
- demonstrate the function-related organization of the cerebral cortex on the basis of receptor autoradiographic data.

The cerebral cortex and can be subdivided into isocortex (neocortex) and allocortex (paleo- and archicortex). The isocortex has a thickness of 2 mm (e.g., visual cortex) to 5 mm (e.g., primary motor cortex). It shows a laminar organization which consists of 6 horizontal layers (parallel to the cortical surface) and vertical cell columns. The allocortex contains less than 6 layers in most regions, but can also present more than 6 layers in a some areas (e.g., entorhinal cortex).

The layers of the isocortex are defined according to the packing density of neuronal cell bodies, the proportion and spatial arrangement of different neuronal cell types as well as cell sizes and shapes. The layers of the isocortex are:

- Lamina I (molecular layer). It occupies the most superficial parts of the cortex and contains only a few scattered, small neuronal cell bodies.

- Lamina II (external granular layer). It shows many densely packed, small granular cell bodies which look like granules when Nissl-stained sections are viewed at a low magnification.

- Lamina III (external pyramidal layer) is characterized by small and medium sized pyramidal cells. Their apical dendrites reach Lamina I. In some cortical regions, very large pyramids are found in the lower third of Lamina III. The axons of layer III-pyramids project mainly to other cortical regions of the ipsi- and contralateral (via corpus callosum) hemisphere.

- Lamina IV (internal granular layer). Numerous, densely packed small granular cells can be found in this layer. Its thickness varies considerably between the different isocortical regions, reaching the largest width in primary sensory areas. Whereas layer IV is not recognizable in the adult motor cortex (BA 4 and BA 6), it consists of 3 sublayers (which can be even further subdivided) in the primary visual cortex (BA 17, = area V1). Layer IV is the major target of the thalamo-cortical input (afferent fibers from the dorsal thalamus and the metathalamus).

- Lamina V (internal pyramidal layer) consists mainly of pyramidal cells of large but variable size. In BA 4, extremely large pyramidal cells (=Betz cells, giant pyramids; cell body height reaching up to 130 mm) are present in this layer. The axons of the pyramidal cells reach other cortical areas in the ipsi- or contralateral hemisphere via short and long fiber tracts of the white matter, or they descend to the basal ganglia, brain stem and spinal cord in the corticostriatal, corticonuclear and corticospinal tracts, respectively.

- Lamina VI (multiform layer). It contains cell bodies of different shapes, e.g., spindle-like, polygonal. Their axons project mainly into subcortical brain regions.

Cytoarchitectonics, as discussed in this talk is defined on the basis of the horizontal, laminar characteristics of the cortex.

Quantitative in vitro receptor autoradiography combines a considerable degree of spatial resolution with a high sensitivity, therefore permitting the anatomical identification of receptor localization as well as the visualization of low receptor concentrations. Furthermore, in combination with computerized image analysis, it enables accurate and reproducible quantification of receptor densities .

The degree of spatial resolution obtainable with in vitro receptor autoradiography is based mainly on the type of isotope with which the ligands are labeled. The results discussed in the present talk were obtained using tritiated ligands, since this isotope permits a better local resolution compared with other isotopes such as ¹²⁵I and ¹⁴C. The structures of interest should exceed 50 µm in their smallest dimension to be resolvable with receptor autoradiography.

Diverse methodological aspects deserve special attention when using quantitative receptor autoradiography to analyze the distribution and densities of neurotransmitter binding sites in the cerebral cortex: pre- and postmortem conditions, tissue processing, labeling procedure, and automated image analysis.

Autoradiographical analysis of postmortem human brain tissue or biopsies raises a series of methodological problems, some of which are also relevant for studies in animal models; others, however, are unique in experiments carried out with human brain sections and include both pre- and post-mortem conditions.

The most important premortem factor, which has been described in numerous reports, is the effect that neurological diseases can have on the distribution, density, and affinity of specific neurotransmitter receptors (Blows, 2000; Mann et al., 2001; Mihailescu and Drucker-Colin, 2000; Tedroff, 1999) . Therefore, only brains obtained from patients who died without a history of neurological or psychiatric disorders were used for the chemoarchitectonic mapping of the human cerebral cortex. Furthermore, binding site density and affinity can also be affected by aging depending on the receptor type under consideration. A consistent finding is the age-related decrease in the density of glutamatergic NMDA receptors, which seems to be accompanied by regionally specific changes in the interaction between glutamate and other neurotransmitters such as dopamine and GABA (Adams et al., 2001; Segovia et al., 2001; Wenk and Barnes, 2000) .

The effect of postmortem delay in the freezing of brain tissue as well as of prolonged storage of the frozen tissue prior to analysis on receptor binding assays are potential artifacts that may limit interpretation of the effects of disease on receptor populations. However, only a relatively small number of reports discuss the problems caused by these circumstances. Some receptors show a surprisingly high stability, with binding site densities remaining constant up to 70-80 hours postmortem (e.g., NMDA (Kornhuber et al., 1988) , GABA (Lloyd and Dreksler, 1979) , M₁ (Burke and Greenbaum, 1987) , D₂ (Kontur et al., 1994) and 5-HT₂ receptors (Gross-Isseroff et al., 1990; Kontur et al., 1994) ). Decreases in densities and affinities were described for D₁ and 5-HT_1A receptors (Kontur et al., 1994) , as well as for [³H]N-methylscopolamine binding sites (Rodriguez-Puertas et al., 1996) . Conversely, the density of benzodiazepine binding sites increased with increasing postmortem delay to freezing (Whitehouse et al., 1984) .

Prolonged storage of deep frozen tissue is inevitable when analyzing a statistically significant sample of human brains. Storage of tissue for up to three years resulted in stable [³H]N-methylscopolamine binding site densities (Rodriguez-Puertas et al., 1996; Whitehouse et al., 1984) . Furthermore, no significant changes were observed when [³H]GABA and [³H]prazosin binding properties were examined after a storage period of at least six years (Faull et al., 1988; Lloyd and Dreksler, 1979) . Although it cannot be ruled out that other receptors may be vulnerable, current knowledge provides evidence that a storage time of up to six years has no influence on the stability of most receptors.

Both the quality of receptor autoradiographs and the preservation of histological stainings are highly dependent on the handling of the brain immediately after autopsy. Although fixation of the brain before deep freezing and cutting clearly improves the quality of histological stainings, it also impairs the structure of receptor proteins and, consequently, leads to changes of specific and non-specific binding, altering the ratio between both parameters to different degrees (Herkenham, 1988; Rotter et al., 1979; Zilles and Schleicher, 1995) . Therefore, we only use unfixed, deep frozen brains for receptor autoradiography.

Immediately after autopsy, the brains were photographed, the hemispheres and brainstem were separated and stored in plastic bags on crushed ice before further dissection. The meninges and blood vessels were not removed, since this process, however carefully it is carried out, causes damage to the brain surface and leads to a partial loss of cortical layer I. Each hemisphere was cut into coronal, sagittal or horizontal slabs (1.5 – 3.0 cm thick), which were placed on a sheet of strong aluminum foil to preserve a flat sectioning surface and to avoid distortions. Each slab comprises parts, or the complete circumference of a hemisphere. The foil with the tissue was slowly immersed in N-methylbutane at –50°C for 10 - 15 minutes. This method enabled fast freezing of the brain tissue, avoiding freeze-artifacts such as the appearance of ice crystals, which would destroy cellular morphology. The tissue was then stored in a deep freezer at –70°C in air-tight plastic bags to protect it from freeze-artifacts.

The brain tissue was serially sectioned in a cryostat microtome for large sections in 20µm sections at -20°C. The sections were thaw-mounted on gelatin-coated glass slides and freeze-dried overnight. Alternating sections were incubated with tritiated ligands alone (total binding), with the tritiated ligands and a receptor type-specific displacing agent (non-specific binding), or stained with modified silver methods that produce Nissl-like images (Merker, 1983) or visualize myelinated fibers (Gallyas, 1979) . The latter histologically stained sections enable a precise microscopical identification of architectonically defined areas and layers.

The autoradiographical labeling method is carried out following standardized protocols (Zilles and Palomero-Gallagher, 2001; Zilles et al. 2002a, b) . In short, it consists of three steps: a preincubation, a main incubation and a rinsing step.

The aim of the preincubation step is to re-hydrate the sections and to wash out endogenous substances which bind to the examined receptor and thus block the binding site for the tritiated ligand. In the main incubation step, adjacent sections are incubated in a buffer solution containing the tritiated ligand (in nM concentrations) or the tritiated ligand (in nM concentrations) plus a non-labeled specific displacer (in µM concentrations). Since the incubation of a brain section with a labeled ligand demonstrates the total binding of this ligand, the incubation with the tritiated ligand in the presence of a specific displacer is necessary to determine what proportion of the total binding sites is occupied by non-specific, and thus non-displaceable binding. Specific binding is the difference between total and non-specific binding. Non-specific binding is only taken into consideration when it amounts to more than 10% of the total binding sites marked by the ligand. Finally, the rinsing step stops the binding procedure and eliminates surplus tritiated ligand as well as buffer salts, thus preventing artifacts on the film emulsion during exposure.

The radioactively marked sections were co-exposed with plastic tritiated standards of varying but known concentrations of radioactivity against b-sensitive films for four to ten weeks. After development of the film, the spatial distribution of optical densities in the autoradiograph indicates the local concentration of radioactivity present in the brain tissue, and thus represents a measure of the local binding site concentrations.

The first step in the evaluation of autoradiographs is the image acquisition. Autoradiographs are digitized as binary files with a spatial resolution of 512x512 pixels and 8-bit gray value resolution (shades of gray ranging from 0=black to 255=white). Digitization is carried out by means of a KS400^® image analyzing system (Zeiss, Germany) connected to a CCD camera (Sony, Tokyo) with an S-Orthoplanar 60-mm macro lens (Zeiss, Germany), which is corrected for geometric distortions. For each exposed film, a blank area (= reference field) is placed on an illumination box, and the intensity of the light source and aperture of the macro lens are adjusted to obtain a mean gray value of 220, which is calculated as the mean gray value of all the pixels in the reference field. A reference gray value well below 255 is chosen in order to avoid saturation effects in the camera. Furthermore, gray values below 20, which cannot be resolved by a CCD camera, should not occur at any place on the film. Therefore, exposure time of the film is determined for each ligand according to this constraint. Additional prerequisites for correct densitometry are reduction of stray light, sufficient warm-up of the light source as well as of the camera to avoid shifts in the system, and a homogeneous light intensity , which is obtained by means of a light box fitted with a double opal glass diffusor. To permit full-contrast densitometry, the smallest structures of interest in an image must be covered by more than one pixel in the x and y directions (Ramm et al., 1984) in order to avoid biasing by the point spread function of neighboring pixels, which may belong to other anatomical structures. Since image acquisition with a video system involves an inevitable noise component caused by non spatially correlated discrete isolated pixel variations, eight images of the same autoradiograph are averaged during acquisition in order to improve the signal to noise ratio. During image acquisition, a shading correction is also carried out. Shading is the variation of gray values within an image of a homogeneous object, and is caused by the video target, camera electronics, the illumination source, and the camera lens . Shading causes a dependency of the gray values of an object on its position in the measuring field, and leads, therefore, to a decrease in the number of resolvable gray values, and, if neglected, induces considerable artifacts in the measurements. Shading correction is achieved by using the above mentioned reference image [R(x,y)], which contains a homogeneous and empty (without any brain tissue) film area, and its mean gray value (C), to transform each pixel of the digitized autoradiographs [A(x,y)] into corrected values [SA(x,y)] by means of equation 1:

Following these preparatory steps, an autoradiograph can be visualized as a black and white image. These images, however, only represent gray values, and not concentrations of radioactivity. Therefore, a gray value scaling is carried out, in which the gray values are transformed into fmol binding sites/mg protein (Schleicher and Zilles, 1988; Zilles and Schleicher, 1995) . This scaling is performed in two stages: firstly, the gray value images of co-exposed tritium-standards are used to compute a calibration curve by non-linear, least-squares fitting, thus defining the relationship between the gray values of the autoradiographs and concentrations of radioactivity. Each of the standards has a known amount of radioactivity, which is determined in an adjacent standard section by liquid scintillation counting. For each of the standards, the amount of radioactivity (R) is converted to the concentration of binding sites (C_b) using equation 2:

Where E is the efficiency of the scintillation counter, B is the number of decays per unit of time and radioactivity, W_b is the protein weight of a standard, S_a is the specific activity of the ligand, K_D is the dissociation constant of the ligand, and L is the free concentration of the ligand during incubation. The nonlinear correlation between gray values and increasing concentrations of binding sites must be emphasized. Due to this nonlinear correlation, it is imperative that the range of gray values in the standards covers the gray value range found in the digitized autoradiograph, since no valid extrapolation can be carried out. Secondly, the gray value of each pixel in an image is converted into a corresponding concentration of radioactivity by interpolation into the calibration curve, and subsequently linearly transformed into new gray values in order to create an image in which the gray values are a linear function of the concentration of radioactivity.

Now the mean ligand concentration of an anatomically defined brain region can be quantified. Two different strategies can be applied depending on whether mean laminar or regional densities are to be extracted. The laminar distribution of neurotransmitter receptors within a given area can be characterized by means of density profiles.

In order to extract numerical values for the mean regional receptor densities in an anatomically defined area, comparison with an adjacent cell-body stained histological section is necessary, since not all anatomical structures are regularly associated with clear-cut differences in receptor densities. Therefore, in order to precisely identify and delineate the regions of interest in an autoradiograph, a print of the digitized autoradiograph and an adjacent cell-body stained section are superimposed by means of a microscope equipped with a drawing tube (Schleicher and Zilles, 1988; Zilles and Schleicher, 1995) . Both the histological section and the hard copy of the autoradiograph are visible simultaneously in the microscope, and the cytoarchitectonic borders are ink-traced on the hard copy. This traced contour of a brain region is then used as a template on a digitizer connected with a computer in which the original data matrix of the digitized autoradiograph is stored. All pixels values of a given structure are automatically selected from the stored image, and the mean receptor concentration per unit protein (fmol/mg protein) contained in a specific region over a series of three to five sections is calculated by weighting the means from single sections by their respective areal size.

Color coding of autoradiographs is carried out solely to provide a clear visual impression of regional and laminar receptor distribution patterns. Since the complete range of available gray values is not necessarily used by the frequency distribution of actually occurring receptor densities, images are linearly contrast enhanced, thus preserving the absolute scaling between gray values and receptor densities, while improving the optical presentation of the images. The full range of 256 gray values (0-255) obtained after contrast enhancement is then pseudo-color-coded. The assignment of colors to the density ranges can be done in an arbitrary fashion, but the spectral arrangement of eleven colors to equally spaced density ranges results in the best visualization of the density pattern of the autoradiographs (Zilles and Schleicher, 1995) .

Receptors for glutamate, GABA, acetylcholine, noradrenaline, serotonin, and dopamine are heterogeneously distributed throughout the human cerebral cortex. They show clear regional differences both in their mean densities and in their laminar distribution patterns. Furthermore, these variations are also present between different receptor types for a single neurotransmitter, i.e. glutamatergic AMPA and NMDA receptors, or muscarinic cholinergic M₂ and nicotinic cholinergic receptors. Although each receptor does not indicate all areal borders, there is a perfect agreement in the location of those borders which are displayed by several receptors.

The M₂ receptor antagonist [³H]oxotremorine-M selectively emphasizes primary sensory areas. The primary somatosensory, auditory and visual cortices contain conspicuously higher M₂ receptor densities than any other cortical region of the human brain. Contrary to the M₂ receptors, which by all means clearly visualize further cortical parcellations, i.e. auditory and inferior temporal association cortices, the nicotinic receptors exclusively accentuate the family of sensorimotor areas.

Receptors for classical neurotransmitters faithfully reflect the complex laminar structure of the primary visual cortex. Furthermore, their regional heterogeneity also enables the parcellation of extrastriate visual areas (Zilles and Clarke, 1997) .

The hippocampus has a clear laminar structure, which is associated with segregated input, output and intrinsic fiber systems, and is therefore a favorable model for comparisons of anatomical structures with functionally and neurochemically identified neuronal systems. The regional and laminar distribution patterns of receptors for classical neurotransmitters in the hippocampus are highly correlated with its anatomical structure (Zilles et al., 1993) .

It is important to stress the fact that changes in receptor densities should not be interpreted as being a mere reflection of variations in the degree of cell packing density in a given region or cortical layer. I.e., a high receptor density does not necessarily imply a high cell packing density, and vice versa. One and the same cytoarchitectonically defined region may contain highest densities of one receptor and lowest of another. E.g., the primary auditory cortex shows extremely high M₂ receptor densities, whereas it contains one of the lowest a₁ binding site concentrations measured in the human brain. This lack of correlation between receptor density and cell packing density is plausible, since by far the majority of transmitter receptors demonstrated by receptor autoradiography are located on dendrites, which represent a major proportion of the cell body-free neuropil compartment. Thus, receptor concentration is not correlated with cell packing density, which is defined as volume density of cell bodies. It is interesting to note, that this situation is in contrast to immunohistochemical receptor studies, which demonstrate single protein subunits of a receptor and not the native receptor complex. These subunits are frequently accumulated in the cell body, and thus their local concentrations are clearly associated with cell packing density.

The complex co-distribution patterns of various receptors in architectonically defined brain regions stimulated the introduction of a new analytical procedure, the so-called receptor fingerprint. Receptor fingerprints of cortical areas are polar coordinate plots of the mean regional densities of several different receptors over all cortical layers in a single, architectonically defined brain region. They demonstrate the site-specific balance between different receptor types and transmitter systems. These fingerprints may differ between regions by their shapes and/or sizes, thus representing the locally specific neurochemical organization at the receptor level. The shapes of the fingerprints differ between the motor, the unimodal sensory and the associative isocortices, as well as the allocortex, indicating the functionally specific balances between the different receptors in these different areas. Fingerprints may also define families of several cortical areas which are similar regarding the balance between different receptors. Differences in areal size of fingerprints may represent different hierarchical levels within a functional system, i.e. a larger receptor fingerprint for the primary visual and auditory areas than for their respective secondary cortices. An analysis of receptor fingerprints in the mesial motor areas (primary, supplementary and pre-supplementary motor areas) of macaque monkeys clearly demonstrates identical shapes (i.e., all three areas belong to the “motor” family), but increasing sizes (i.e., proportionally increasing receptor densities from the primary motor through the supplementary to the pre-supplementary motor area) of their receptor fingerprints (Geyer et al., 1998) .

The distribution patterns of receptors for classical neurotransmitters reveal a more detailed cortical parcellation than that described by classical brain maps, i.e. the cytoarchitectonic map of (Brodmann, 1909) . This lack of congruence does not imply a total lack of correspondence between classical parcellation schemes and the areas revealed by receptor autoradiography. In some cases, a total correspondence exists, i.e., in BA 17. In other instances, receptor autoradiography leads to a further parcellation of a region originally described by Brodmann. Such is the case of BA 4, which, according to Brodmann is not further subdivided based on cytoarchitectonical features. A recent combined receptor autoradiographical and functional imaging study demonstrated that BA 4 can be subdivided at least into an anterior (4a) and a posterior (4b) component based both on differential receptor distribution patterns and on functional activations (Geyer et al., 1996) .

Adams, M. M., Smith, T. D., Moga, D., Gallagher, M., Wang, Y., Wolfe, B. B., Rapp, P. R., and Morrison, J. H. (2001). Hippocampal dependent learning ability correlates with N-methyl-D-aspartate (NMDA) receptor levels in CA3 neurons of young and aged rats. J. Comp. Neurol. 432, 230-243.

Blows, W. T. (2000). Neurotransmitters of the brain: serotonin, noradrenaline (norepinephrine), and dopamine. J. Neurosci. Nurs. 32, 234-238.

Brodmann, K. (1909). "Vergleichende Lokalisationslehre der Großhirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues". Barth JA, Leipzig.

Burke, R. E. and Greenbaum, D. (1987). Effect of postmortem factors on muscarinic receptor subtypes in rat brain. J. Neurochem. 49, 592-596.

Faull, K. F., Bowersox, S. S., Zeller-DeAmicis, L., Maddaluno, J. F., Ciaranello, R. D., and Dement, W. C. (1988). Influence of freezer storage time on cerebral biogenic amine and metabolite concentrations and receptor ligand binding characteristics. Brain Res. 450, 225-230.

Gallyas, F. (1979). Silver staining of myelin by means of physical development. Neurol. Res. 1, 203-209.

Geyer, S., Ledberg, A., Schleicher, A., Kinomura, S., Schormann, T., Bürgel, U., Klingberg, T., Larsson, J., Zilles, K., and Roland, P. E. (1996). Two different areas within the primary motor cortex of man. Nature 382, 805-807.

Geyer, S., Schleicher, A., and Zilles, K. (1997). The somatosensory cortex of human: Cytoarchitecture and regional distributions of receptor-binding sites. NeuroImage 6, 27-45.

Geyer, S., Matelli, M., Luppino, G., Schleicher, A., Jansen, Y., Palomero-Gallagher, N., and Zilles, K. (1998). Receptor autoradiographic mapping of the mesial and premotor cortex of the macaque monkey. J. Comp. Neurol. 397, 231-250.

Gross-Isseroff, R., Salama, D., Israeli, M., and Biegon, A. (1990). Autoradiographic analysis of age-dependent changes in serotonin 5-HT₂ receptors of the human brain postmortem. Brain Res. 519, 223-227.

Herkenham, M (1988). Influence of tissue treatment on quantitative receptor autoradiography. In "Molecular Neuroanatomy" (F.W.van Leeuwen, R.M.Buijs, C.W.Pool, and O.Pach, Eds.), pp. 111-120. Elsevier, Amsterdam.

Kontur, P. J., al-Tikriti, M., Innis, R. B., and Roth, R. H. (1994). Postmortem stability of monoamines, their metabolites and receptor binding in rat brain regions. J. Neurochem. 62, 282-290.

Kornhuber, J., Retz, W., Riederer, P., Heinsen, H., and Fritze, J. (1988). Effect of antemortem and postmortem factors on [3H]glutamate binding in the human brain. Neurosci. Letters 93, 312-317.

Lloyd, K. G. and Dreksler, S. (1979). An analysis of [³H]gamma-aminobutyric acid (GABA) binding in the human brain. Brain Res. 163, 77-87.

Mann, J. J., Brent, D. A., and Arango, V. (2001). The neurobiology and genetics of suicide and attempted suicide: A focus on the serotonergig system. Neuropsychopharmacology 24, 467-477.

Merker, B. (1983). Silver staining of cell bodies by means of physical development. J. Neurosci. 9 , 235-241.

Mihailescu, S. and Drucker-Colin, R. (2000). Nicotine and brain disorders. Acta Pharmacol. Sin. 21, 97-104.

Ramm, P., Kulick, J. H., and Farb, D. H. (1984). Video and scanning microdensitometer-based imaging systems in autoradiographic densitometry. J. Neurosci. Methods 11, 89-100.

Rodriguez-Puertas, R., Pascual, J., and Pazos, A. (1996). Effects of freezing storage time on the density of muscarinic receptors in the human brain postmortem: An autoradiographic study in control and Alzheimer's disease brain tissues. Brain Res. 728, 65-71.

Rotter, A., Birdsall, N. J. M., Burgen, A. S. V., Field, P. M., Hulme, E. C., and Raisman, G. (1979). Muscarinic receptors in the central nervous system of the rat. I. Technique for autoradiographic localization of the binding of [³H]propylbenzilylcholine mustard and its distribution in the forebrain. Brain Res. Rev. 1, 141-166.

Schleicher, A and Zilles, K. (1988). The use of automated image analysis for quantitative receptor autoradiography. In "Molecular Neuroanatomy" (F.W.Leeuwen van, R.M.Buijs, C.W.Pool, and O.Pach, Eds.), pp. 147-157. Elsevier, Amsterdam.

Segovia, G., Porras, A., Del Arco, A., and Mora, F. (2001). Glutamatergic neurotransmission in aging: A critical perspective. Mech. Aeging Dev. 122, 1-29.

Talairach, J. and Tournoux, P. (1988). "Coplanar stereotaxic atlas of the human brain". Thieme, Stuttgart.

Tedroff, J. M. (1999). Functional consequences of dopaminergic degeneration in Parkinson's disease. Adv. Neurol. 80, 67-70.

Wenk, G. L. and Barnes, C. A. (2000). Regional changes in the hippocampal density of AMPA and NMDA receptors across the lifespan of the rat. Brain Res. 885, 1-5.

Whitehouse, P. J., Lynch, D., and Kuhar, M. J. (1984). Effects of postmortem delay and temperature on neurotransmitter receptor binding in a rat model of the human autopsy process. J. Neurochem. 43, 553-559.

Zilles, K. (1992). Neurotransmitter receptors in the forebrain: regional and laminar distribution. Progr. Histochem. Cytochem. 26, 229-240.

Zilles, K and Clarke, S. (1997). Architecture, connectivity and transmitter receptors of human extrastriate visual cortex. Comparison with non-human primates. In "Cerebral Cortex. Vol 12" (Rockland et al., Ed.), pp. 673-742. Plenum Press, New York.

Zilles, K., Palomero-Gallagher, N.: Cyto-, myelo- and receptor architectonics of the human parietal cortex. NeuroImage 14, 8-20 (2001)

Zilles, K and Schleicher, A. (1995). Correlative imaging of transmitter receptor distributions in human cortex. In "Autoradiography and correlative imaging" (W.Stumpf and H.Solomon, Eds.), pp. 277-307. Academic Press, San Diego.

Zilles, K., Schlaug, G., Matelli, M., Luppino, G., Schleicher, A., Qü, M., Dabringhaus, A., Seitz, R., and Roland, P. E. (1995). Mapping of human and macaque sensorimotor areas by integrating architectonic, transmitter receptor, MRI and PET data. J. Anat. 187, 515-537.

Zilles, K., Qü, M., Schleicher, A., and Luhmann, H. J. (1998). Characterization of neuronal migration disorders in neocortical structures: Quantitative receptor autoradiography of ionotropic glutamate, GABA_A and GABA_B receptors. Europ. J. Neurosci. 10, 3095-3106.

Zilles, K., Qü, M. S., Köhling, R., and Speckmann, E.-J. (1999). Ionotropic glutamate and g-aminobutyric acid receptors in human epileptic neocortical tissue: Quantitative in vitro receptor autoradiography. Neuroscience 94, 1051-1061.

Zilles, K., Schleicher, A., Palomero-Gallagher, N., Amunts, K.: Quantitative analysis of cyto- and receptorarchitecture of the human brain, pp. 573-602. In: Brain Mapping: The Methods, 2^nd edition (A.W. Toga and J.C. Mazziotta, eds.). Academic Press (2002a)

Zilles, K., Palomero-Gallagher, N., Grefkes, C., Scheperjans, F., Boy, C., Amunts, K., Schleicher, A.: Architectonics of the human cerebral cortex and transmitter receptor fingerprints: Reconciling functional neuroanatomy and neurochemistry. Europ. Neuropsychopharmacol. 12, 587-599 (2002b)