Aggrecan-based extracellular matrix is an integral part of the human basal ganglia circuit

Neuroscience 151 (2008) 489 –504

AGGRECAN-BASED EXTRACELLULAR MATRIX IS AN INTEGRAL PART OF THE HUMAN BASAL GANGLIA CIRCUIT G. BRÜCKNER,a,1* M. MORAWSKIa,b,1 AND T. ARENDTa

scaffold organized in region and cell type– dependent patterns (Delpech et al., 1989; Brückner et al., 1993; Asher et al., 1995; Yamaguchi, 2000; Viapiano and Matthews, 2006; Carulli et al., 2007). The most conspicuous form of extracellular matrix organization is the perineuronal nets (PNs) associated with many types of neurons in the brain and spinal cord. Different proportions of hyaluronan and aggregating proteoglycans, the predominance of certain glycosylation forms of proteoglycans, as well as the degree of sulfation may influence the physicochemical properties of the extracellular space, e.g. diffusion parameters, ion buffering capacity or binding properties for a variety of regulatory factors (Syková, 2004). It is therefore reasonable to assume that the cellular microenvironment is adjusted to major functional properties of neurons, such as their firing patterns, as well as to the plasticity of synapses and the sprouting capacity of axons retained in the adult state (Dityatev and Schachner, 2003; Rauch, 2004; Viapiano and Matthews, 2006; Galtrey and Fawcett, 2007). Moreover, the extracellular matrix may protect neurons against damage induced by oxidative stress and other environmental factors (Morawski et al., 2004, 2005). The highly specified extracellular matrix scaffold in the normal brain suggests that a disturbed matrix integrity can severely affect the physiological properties and the plastic capacity of individual neurons and complex neuronal circuits. It has been shown that dramatic changes can be related to the additional expression of matrix components in glial scars in response to acute brain injuries, such as stroke (Hobohm et al., 2005) and spinal cord injury (Silver and Miller, 2004; Galtrey et al., 2007; Massey et al., 2007). In contrast, a long-lasting decomposition of the extracellular matrix scaffold may occur in chronic pathological states such as viral infections (Belichenko et al., 1997, 1999; Medina-Flores et al., 2004; Vidal et al., 2006). An accelerating influence of extracellular matrix components, especially of heparan sulfate proteoglycans, in the Alzheimer’s disease amyloidopathy has also been suggested (van Horssen et al., 2003). Studies on the organization of the extracellular matrix in human brain regions most frequently affected in neurodegenerative diseases have been focused on the cerebral cortex (Belichenko et al., 1997, 1999; Brückner et al., 1999; Baig et al., 2005) and basal forebrain (Adams et al., 2001). In the substantia nigra complex, despite its central role in Parkinson’s disease (PD), knowledge is restricted to the detection of hyaluronan (Yasuhara et al., 1994). Our studies in the rat (Hobohm et al., 1998) and the Chilean mouse opossum (Brückner et al., 2006), using lectin staining and chondroitin sulfate proteoglycan (CSPG) antibod-

a

Paul Flechsig Institute for Brain Research, Department of Neuroanatomy, Medical Faculty, University of Leipzig, Jahnallee 59, D-04109 Leipzig, Germany

b

Interdisciplinary Centre for Clinical Research Leipzig (IZKF) Medical Faculty, University of Leipzig, Inselstraße 22, D-04103 Leipzig, Germany

Abstract—The extracellular matrix is known to be involved in neuronal communication and the regulation of plastic changes, and also considered to protect neurons and synapses against damage. The goal of this study was to investigate how major extracellular matrix components (aggrecan, link protein, hyaluronan) constitute the pathways of the nigral system in the human basal ganglia circuit affected by neurodegeneration in Parkinson’s disease. Here we show that aggrecan- and link protein–related components form clear regional distribution patterns, whereas hyaluronan is widely distributed in gray and white matter. Two predominant phenotypes of the aggrecan-based matrix can be discriminated: (1) perineuronal nets (PNs) and (2) axonal coats (ACs) encapsulating preterminal fibers and synaptic boutons. Clearly contoured PNs are associated with GABAergic projection neurons in the external and internal division of the globus pallidus, the lateral and reticular part of the substantia nigra, as well as subpopulations of striatal and thalamic inhibitory interneurons. Dopaminergic nigral neurons are devoid of PNs but are contacted to a different extent by matrix-coated boutons forming subnucleus-specific patterns. A very dense network of ACs is characteristic especially of the posterior lateral cell groups of the compact substantia nigra (nigrosome 1). In the subthalamic nucleus and the lateral thalamic nuclei numerous AC-associated axons were attached to principal neurons devoid of PNs. We conclude from the region-specific patterns that the aggrecan-based extracellular matrix is adapted to the fast processing of sensorimotor activities which are the therapeutic target of surgery and deep brain stimulation in the treatment of advanced stages of Parkinson’s disease. © 2008 IBRO. Published by Elsevier Ltd. All rights reserved. Key words: basal ganglia, perineuronal net, chondroitin sulfate proteoglycan, synapse, Parkinson’s disease.

The extracellular matrix is an integral part of the neural tissue. Aggregating proteoglycans, complexed with hyaluronan, tenascins and link proteins form a macromolecular 1

These authors contributed equally to this work. *Corresponding author. Tel: ⫹49-341-9725-732; fax: ⫹49-341-9725-749. E-mail address: [email protected] (G. Brückner). Abbreviations: AC, axonal coat; BHABP, biotinylated hyaluronic acidbinding protein; CRTL1, cartilage-related aggrecan-binding link protein; CSPG, chondroitin sulfate proteoglycan; DAB, diaminobenzidine; DAB-Ni, nickel-enhanced diaminobenzidine; GAD, glutamic acid decarboxylase; PBS, phosphate-buffered saline; PD, Parkinson’s disease; PN, perineuronal net; TH, tyrosine hydroxylase.

0306-4522/08$32.00⫹0.00 © 2008 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2007.10.033

489

490

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

ies, revealed region-specific patterns of extracellular matrix organization in the substantia nigra complex. Notably, the absence of PNs around dopaminergic neurons in the substantia nigra and the ventral tegmental area appeared to be a phylogenetically conserved principle. The general goal of the present study was to analyze whether the extracellular matrix contributes to region-specific properties of the basal ganglia circuit. Specifically, we intended to investigate if neurons known to be affected by degeneration in PD, the dopaminergic nigral neurons and cholinergic neurons of the tegmental pedunculopontine nucleus, differ with respect to extracellular matrix organization from neurons that are relatively resistant to degeneration in PD (Hirsch et al., 1987; van Domburg and ten Donkelaar, 1991; Hardman et al., 1996, 1997; Ma et al., 1996; Hirsch, 1999; Pahapill and Lozano, 2000; Jellinger, 2002; Halliday et al., 2005; Wakabayashi et al., 2006; DeLong and Wichmann, 2007). Therefore, we focused our study on major intrinsic components of the basal ganglia circuit, including the compact part of the substantia nigra, the ventral tegmental area, the lateral and basal part of the reticular substantia nigra, the external and internal division of the globus pallidus, the caudate nucleus and putamen, the subthalamic nucleus, as well as the thalamus and the tegmental pedunculopontine nucleus receiving basal ganglia output signals. The data may have possible clinical relevance to the pharmacological treatment of PD. They may also be considered in the discussion of functional benefits induced by invasive therapies, since the globus pallidus, the subthalamic nucleus, the motor thalamus and the pedunculopontine nucleus are the preferred interventional targets in severe cases of PD (Hamani et al., 2006; Moro and Lang, 2006; Temel and Visser-Vandewalle, 2006).

EXPERIMENTAL PROCEDURES Autopsied cases Human brains were provided by the brain bank of the Interdisciplinary Centre for Clinical Research at the University of Leipzig, Leipzig, Germany. All procedures of acquisition of the patient’s personal data, autopsy and the handling of the autopsy material have been approved by the Ethical Committee of Leipzig University. The brains were obtained from five individuals of both sexes aged 48 – 88 years and showed no signs of pathological alterations (Table 1). The brains were dissected and fixed in 4% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, for 3– 4 weeks at 4 °C.

Tissue preparation After fixation, 15 mm-thick slices were prepared in the frontal plane according to the atlas of the human brain (Mai et al., 1997). Tissue blocks containing the regions of interest were cryoprotected in 30% sucrose in 0.1 M phosphate-buffered saline, pH 7.4 (PBS). Series of 30 ␮m-thick sections were cut on a freezing microtome and collected in PBS containing 0.1% sodium azide.

Identification of anatomical regions and applied nomenclature Anatomical regions were identified using Nissl-stained sections and the atlas of the human brain (Mai et al., 1997). The nomenclature of brain regions was mainly adopted from brain atlases of human (Mai et al., 1997) and rhesus monkey (Paxinos et al., 2000). The identification of anatomical divisions and cell groups of the substantia nigra is based on studies of Braak and Braak (1986), Fearnley and Lees (1991), van Domburg and ten Donkelaar (1991), McRitchie et al. (1995), and Damier et al. (1999a). We additionally used the data of detailed anatomical descriptions of the basal ganglia complex (Morel et al., 2002; Waldvogel et al., 2004, 2007), the striatum (Waldvogel and Faull, 1993), the thalamus (Morel et al., 1997; Münkle et al., 2000), and the pedunculopontine nucleus (Mesulam et al., 1989).

Cytochemistry Antigen retrieval. All sections were pre-treated with an initial antigen retrieval step (Evers and Uylings, 1997) by boiling the sections three times for 10 s in a microwave (at 900 W) in 0.1 M citrate buffer, pH 2.5, followed by rinsing with PBS. Detection of extracellular matrix components. To abolish endogenous peroxidase activity cytochemical procedures except the fluorescence methods were started by the treatment with 1% H2O2 in PBS-T (0.05% Tween) for 30 min followed by rinsing with PBS-T. A subsequent blocking step with 2% bovine serum albumin, 0.3% casein and 0.1% gelatin in TBS-T was carried out for 1 h at room temperature to prevent non-specific antibody binding to the tissue. For demonstration of extracellular matrix components sections were incubated with the following antibodies: rabbit antiCSPG (Quartett, Berlin, Germany; 1:2000) raised against bovine nasal cartilage (Bertolotto et al., 1991) and introduced as stable marker for PNs in human brain (Bertolotto et al., 1991; Hausen et al., 1996; Brückner et al., 1999; Adams et al., 2001), goat antihuman CRTL1 polyclonal antibody (R&D Systems Inc., Minneapolis, MN, USA; 1:400) raised against recombinant human hyaluronan and proteoglycan link protein 1 (HAPLN1), known to stabilize the connection between proteoglycans (mainly aggrecan) and hyaluronan (Neame and Barry, 1994; Carulli et al., 2007), and monoclonal aggrecan antibody (clone HAG7D4, Acris, Hiddenhausen, Germany; 1:10) raised against human cartilage aggrecan. The presence and distribution of hyaluronan were detected by using biotinylated hyaluronic acid-binding protein (BHABP, Cape Cod Inc., East Falmouth, MA, USA; 1 ␮g/ml). To test the

Table 1. Profile of the human subjects used in this study Case

Gender

Age (yr)

Postmortem delay (h)

Cause of death

Neuron no. nigrosome 1

1 2 3 4 5

M M F M M

48 72 57 88 65

72 72 24 26 24

Pancreatitis Lymphoma Peritonitis Pancreatitis Myocardial infarction

19,800 30,233 33,350 21,533 23,750

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

491

Table 2. Cytochemical detection of extracellular matrix components Detected components

Reagent

Dilution

Source

References

Aggrecan, core protein CSPG core protein CRTL1 Hyaluronan

Mouse anti-human aggrecana Rabbit anti-CSPGb Goat anti-human HAPLN1c Biotinylated HABPd

1:5 1:2000 1:400 1 ␮g/ml

Acris Antibodies Quartett R&D Systems Cape Cod

Bertolotto et al. (1991) Carulli et al. (2007)

a

Clone HAG7D4, tissue culture supernatant. Antigen from chondroitinase ABC-digested bovine nasal cartilage proteoglycan. c Antigen from purified, NSO-derived, recombinant human hyaluronan and proteoglycan link protein. d Hyaluronic acid binding protein, isolated from bovine nasal cartilage proteoglycan. b

specificity of the BHABP-binding, free-floating sections were pretreated with hyaluronidase from Streptomyces hyalurolyticus (H1136, Sigma-Aldrich, Munich, Germany; 60 U/ml 0.1 M PBS, pH 5.0; Köppe et al., 1997). Binding of BHABP was at the background level after 12 h of enzymatic treatment at 37 °C. The methods used for the staining of matrix constituents are summarized in Table 2. For the demonstration of the different neurotransmitter systems sections were stained by applying a conventional peroxidase-anti-peroxidase (PAP) method based on monoclonal antibodies to tyrosine hydroxylase (TH; MAB318, 1:200; Chemicon, Schwalbach, Germany), parvalbumin (PARV; PV-28, 1:1000; Swant, Bellinzona, Switzerland), glutamic acid decarboxylase (GAD65/67; G5163, 1:10,000; Sigma) and choline acetyltransferase (ChAT; AB144, 1:200; Chemicon). Late endosomes were detected with a monoclonal antibody to man-

nose-6-phosphate receptor (2G11, 1:100; Abcam, Cambridge, MA, USA). Immunoreactivity was visualized by using a standard streptavidin– biotin technique including biotinylated secondary antibodies (Dianova; Hamburg, Germany) and diaminobenzidine (DAB) or nickel-enhanced diaminobenzidine (DAB-Ni) as chromogen. Controls. In control experiments, primary antibodies were omitted, yielding the unstained sections.

Light microscopy, confocal laser scanning microscopy and image processing Tissue sections were examined with a Zeiss Axioplan-AxioVision microscope (Carl Zeiss, Jena, Germany). Fluorescence labeling was examined with a Zeiss confocal laser scanning microscope (LSM 510). Photoshop 9.0 (Adobe Systems, Mountain View, CA,

Abbreviations used in the figures ac APul Aq Atg AV BM Cd Cd-Pu CL cp DpMe Di EC Eml Iml Gpe Gpi Ic Icj LD LHb MB MD MG Mt Mml Opt PC PPTg Pu Rt RiMLF

anterior commissure anterior pulvinar cerebral aqueduct (Sylvius) anterior tegmental nucleus anteroventral thalamic nucleus basal nucleus of Meynert caudate nucleus caudate-putamen centrolateral thalamic nucleus cerebral peduncle deep mesencephalic nucleus diffuse part of the substantia nigra external capsule external medullary lamina internal medullary lamina globus pallidus, external division globus pallidus, internal division internal capsule interstitial nucleus of Cajal lateral dorsal thalamic nucleus lateral habenular nucleus mammilary body mediodorsal thalamic nucleus medial geniculate nucleus mammillothalamic tract internal medullary lamina of the globus pallidus optic tract posterior commissure tegmental pedunculopontine nucleus putamen reticular thalamic nucleus rostral interstitial nucleus of the medial longitudinal fascicle

RRF RN SC Scp SNC SNCai

retrorubral field red nucleus superior colliculus superior cerebellar peduncle substantia nigra, compact part substantia nigra, compact part, anterointermediate subnucleus SNCal substantia nigra, compact part, anterolateral subnucleus SNCam substantia nigra, compact part, anteromedial subnucleus SNCpl substantia nigra, compact part, posterolateral subnucleus SNCpm substantia nigra, compact part, posteromedial subnucleus SNCps substantia nigra, compact part, posterosuperior subnucleus SNL substantia nigra, lateral part SNR substantia nigra, reticular part STh subthalamic nucleus Thal thalamus VAL ventral anterior thalamic nucleus, lateral part VAM ventral anterior thalamic nucleus, medial part VLL ventral lateral thalamic nucleus, lateral part VLM ventral lateral thalamic nucleus, medial part VPL ventral posterior lateral thalamic nucleus VPM ventral posterior medial thalamic nucleus VTA ventral tegmental area Xscp decussation of the superior cerebellar peduncle ZI zona incerta 3N oculomotor nucleus 3V third ventricle 4N trochlear nucleus

492

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

Fig. 1. Cytochemical characterization of the aggrecan-based extracellular matrix in human basal ganglia using antibodies to link protein (CRTL1), aggrecan (HAG7D4), and CSPG, as well as detection of hyaluronan with BHABP. (A, D, G) At the regional level, CRTL1, HAG7D4 and CSPG visualized with Ni-enhanced DAB show similar staining patterns. (A) CRTL1 revealed clear differential staining in all regions, including the STh, the SNC, SNL and SNR, as well as the ZI, the RiMLF and the RN. (D) HAG7D4 was weak in staining intensity. (G) CSPG typically showed a relatively high background. At the cellular level, double fluorescence labeling was identical for CRTL1 and HAG7D4. (B, E) Pallidal PN. (C, F) ACs associated with dendrites of a thalamic neuron containing accumulated lipofuscin (asterisk). (H) CSPG-immunoreactive pallidal PN. (I) CSPG-immunoreactive ACs contacting dendrites of a thalamic neuron (asterisk). (J) BHABP revealed the ubiquitous distribution of hyaluronan. (K) In the globus pallidus PNs appeared as perisomatic rim (arrows). (L) After pretreatment of sections with Streptomyces hyaluronidase the BHABP staining was clearly reduced. Pallidal neurons (arrow) were devoid of labeled PNs. Scale bars⫽2 mm in A, applies to D, G, J; 20 ␮m in B, applies to C, E, F, H, I, K, L.

G. Brückner et al. / Neuroscience 151 (2008) 489 –504 USA) was used to process the images with minimal alterations to the brightness, sharpness, color saturation and contrast.

Quantitative analysis of dopaminergic neurons in nigrosome 1 To exclude cases with obvious nigral pathology we analyzed the largest and most vulnerable part of substantia nigra pars compacta, the nigrosome 1 (Damier et al., 1999a,b). Quantitative analysis of TH-immunoreactive neurons was performed following Damier et al. (1999a) on all five cases used in this study. Dopaminergic neurons were identified by the occurrence of neuromelanin in Nissl-stained sections as well as in consecutive anti-TH-stained sections. Nigrosome 1 was identified by anti-CRTL-1 immunoreaction (see Fig. 4). Neurons were counted in Nissl-stained sections if a nucleus was visible and in anti-TH-stained sections if the soma and at least one dendrite could be identified. The outline of nigrosome 1 and the location of the neurons inside this subnucleus were assessed by the use of a Stereoinvestigator (Microbrightfield, Williston, VT, USA; including a Zeiss Axioskop 2 plus, Märzhäuser micropositioning system and a Ludl 5000 controller) with the Neurolucida software. Nigrosome 1 was outlined in three sections from rostral to caudal (largely comprising the whole nigrosome 1) in all cases. In these outlined sections, the number of TH-immunoreactive neurons was determined. The total number of dopaminergic neurons in the nigrosome 1 was estimated from the map of TH-positive neurons using the formula of Konigsmark (1970). The estimated number of dopaminergic neurons in nigrosome 1 is shown in Table 1.

RESULTS Detection of extracellular matrix components in the human brain To investigate the aggrecan-based extracellular matrix in autoptic brain tissue we first tested available reagents established as cytochemical markers in different mammals. The staining patterns revealed by peroxidase techniques with the aggrecan antibody HAG7D4 showed a large congruence with the antibody to cartilage-related aggrecan-binding link protein (CRTL1) in most brain regions (Fig. 1A, D). This result was confirmed using double fluorescence staining (Fig. 1B, C, E, F). However, a notable exception was the clear staining of neuropil zones with CRTL1 in the striatum and cortex (not shown) that was at low intensity only with HAG7D4. The antibody against CSPG (Fig. 1G–I) showed patterns resembling the patterns obtained after HAG7D4 immunoreaction. The detection of hyaluronan with BHABP (Fig. 1J–L) tested for binding specificity by treatment of the sections with Streptomyces hyaluronidase, indicated a ubiquitous distribution of hyaluronan different from the clear regional patterns of aggrecan and link protein immunoreactivity. PNs typically appeared as perisomatic rim. However, PNs distinctly exceeding the staining intensity of the surrounding neuropil occurred only in some regions, such as the globus pallidus. Wisteria floribunda agglutinin (WFA), a widely used marker for PNs in different animal species inclusive of nonhuman primates (Härtig et al., 1992; Adams et al., 2001) was not applied in our study because of inconsistent or negative results on the autoptic tissue.

493

The robust staining and the low background seen with CRTL1 antibody to aggrecan-binding link protein led us use this reagent as the most suitable tool throughout this study to characterize the organization of the aggrecanbased extracellular matrix. Architectural principles of aggrecan-based extracellular matrix At the regional level, immunoreactions for aggrecan and aggrecan-binding link protein revealed clear distribution patterns of the aggrecan-based matrix in the basal ganglia circuit. For mapping of anatomical structures six representative frontal sections stained for link protein are shown in Fig. 2. The different regional patterns were caused by two structural phenotypes of matrix assembly: the well-known PNs and axonal coats (ACs) associated with preterminal axons and boutons (Fig. 3). The detection of ACs was an unexpected result related to the aggrecan and link protein antibodies that have not been previously applied to the human brain. PNs. A great structural diversity of PNs specifically adapted to the types of neurons in the different regions was a consistent observation in the present investigation. This is in agreement with previous studies in human cortical and subcortical regions (Bertolotto et al., 1991; Hausen et al., 1996; Brückner et al., 1999; Adams et al., 2001; Morawski et al., 2004). In the basal ganglia circuit, lattice-like, clearly contoured PNs were associated especially with pallidal neurons (Fig. 3A–C), as well as neurons of the lateral and reticular part of the substantia nigra. The PN-associated neurons were additionally characterized by intracellular clusters of granular inclusions (aggrecan bodies, 0.5–3 ␮m) immunoreactive for aggrecan and link protein (Fig. 3A, C). These previously not described compartments may indicate synthetic processing of aggrecan and link protein, since endosomes labeled with an antibody to mannose-6-P receptor were not immunoreactive with HAG7D4 and CRTL1 antibodies (Fig. 3C). Periaxonal coats. Aggrecan and link protein–immunoreactive globular and varicose axon-like profiles were attached to neuronal somata and dendritic domains in region-specific patterns (Fig. 3B, D). GAD immunoreaction revealed the periaxonal assembly of matrix components in a fraction of GABAergic fibers most clearly in the subthalamic nucleus (Fig. 3E, F). The cellular origin, morphology and neurochemical type of the various AC-ensheathed presynaptic structures in the basal ganglia circuit remain to be elucidated. Region-specific features of aggrecan-based extracellular matrix Major characteristics of regions and the different types of neurons are summarized in Table 3 and in a schematic figure (Fig. 9). Globus pallidus. The detection of aggrecan and link protein revealed clearly stained PNs in the globus pallidus

494

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

Fig. 2. Regional distribution patterns of the aggrecan-based extracellular matrix in the basal ganglia circuit revealed by immunoreaction for CRTL1 link protein. The Cd-Pu, the globus pallidus, the subthalamic nucleus, the substantia nigra, the ventral tegmental area, the pedunculopontine tegmental nucleus, as well as motor, sensory and association thalamic nuclei are shown in the transverse plane in a rostral (A) to caudal (F) direction. (A) Cd, Pu, GPe and GPi division of the globus pallidus. (B) Cd, Pu, Gpe, Gpi; anterior thalamic nuclei (AV, VAL, VAM). (C) Anterior part of the SNC (subnuclei indicated by arrows), the VTA, the SNL and SNR, the STh at its full extent; VLL, VLM and MD. (D) SNCps of the compact substantia nigra adjacent to the RN at its full extent, SNL and SNR; VLM, CM, MD and APul, VPL, VPM.

(Fig. 2A, B). The internal medullary lamina of the globus pallidus contained PNs specifically expressing a diffuse phenotype (not shown). As a characteristic of both the external and the internal pallidal division, the neuropil surrounding the net-associated neurons was stained with low

intensity. A low number of ACs integrated into PNs contacted the perisomatic domains of neurons, but additionally occurred selectively stained in close proximity to profiles resembling distal dendritic domains in both divisions of the globus pallidus (Fig. 3A, C).

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

495

Fig. 2. (Continued.) (E) Posterior subnuclei SNCpl, SNCpm; posterior APul, VPL. (F) PPTg dorsal to the scp. For further explanation see list of abbreviations used in the figures. Scale bar⫽10 mm in F, applies to A–E.

Subthalamic nucleus. The subthalamic nucleus was characterized by a homogeneous staining of the aggrecanbased matrix outlining the elliptic shape of the nucleus at low magnification (Fig. 2C). A clear segregation of different compartments related to defined functional units expressing parvalbumin or calretinin (Parent et al., 1996; Augood et al., 1999; Morel et al., 2002; Lévesque and Parent, 2005) was not observed in the major territory of the nucleus. However, a tendency of weaker labeling existed in the ventromedial part known to contain a high proportion of calretinin-positive neurons. Typical PNs were not revealed around principal subthalamic neurons and GAD-positive interneurons. However, the principal neurons were contacted by numerous ACs (Fig. 3E). In many cases, these matrix-associated profiles completely covered the soma and dendritic domains, and thus may constitute a perineuronal structure resembling PNs. Large numbers of ACs were distributed in the neuropil. GAD immunoreaction clearly revealed that the proteoglycan matrix was associated with axonal varicosities in the neuropil and around neuronal somata (Fig. 3F). Striatum. In the caudate nucleus and putamen (Fig. 2A, B) a subpopulation of GAD and parvalbumin-positive interneurons was ensheathed by PNs immunoreactive for aggrecan and link protein (Fig. 4A). Cholinergic interneurons and the GABAergic projection neurons were not associated with PNs (Fig. 4B). This confirms the results of our previous work using the CSPG antibody (Adams et al., 2001). However, CRTL1 immunoreaction revealed ACs surrounding very thin axonal profiles sparsely distributed in the neuropil (Fig. 4C). The morphological correlate of the cloudy CRTL1 staining in the striatal neuropil (Fig. 2A, B), possibly indicating the striosomal patch-matrix organization (Graybiel and Ragsdale, 1978; Gerfen, 1992), remains to be identified.

Compact part of the substantia nigra, ventral tegmental area, and retrorubral field. The dopaminergic, pigmented neurons clustered in the substantia nigra, the ventral tegmental area, and the retrorubral field were devoid of PNs. In contrast to this uniform feature, the packing density of ACs was extremely different in the individual nuclei and in defined divisions of the compact part of the substantia nigra (Fig. 2C–E; Fig. 5). A very high density of ACs was characteristic of the posteromedial and posterolateral division, including the posterosuperior and the posterolateral subnucleus (defined by Braak and Braak, 1986). In these territories, TH-immunoreactive neurons were embedded in a matrix-associated fiber plexus (Fig. 5B, C, E). In the remaining parts of the ventral tier and in the anterior division, cell clusters showing moderate and low density of ACs, respectively, were observed (Fig. 5A–C, E, F). Cell groups with different density of ACs often occurred in close proximity (Fig. 5B). The VTA was virtually devoid of ACs (Fig. 5G). In the retrorubral field (Fig. 2E), both clearly contoured PNs and a moderate number of ACs were associated with TH-negative neurons (not shown). Lateral and reticular substantia nigra. Non-pigmented, TH-negative neurons in the lateral and reticular part of the substantia nigra (Fig. 2C, D) were ensheathed by PNs and additionally contacted by ACs (Figs. 5A, B; 6A, B). However, neurons in the lateral subnucleus were tightly covered by ACs, thereby clearly differentiating the lateral from the reticular territories of the substantia nigra. Thalamus. The thalamus showed distinct distribution patterns of aggrecan-related matrix components, often clearly indicating the borders between nuclei (Fig. 2B–E). These patterns were largely identical to the regional

496

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

Fig. 3. Structural phenotypes of aggrecan-based PNs and ACs in the basal ganglia circuit. Immunoreaction for CRTL1 link protein is visualized by DAB-Ni (black) in single-stained sections (A–C), and by DAB (brown) for double staining (D–F). (A) Neuron in the internal segment of the globus pallidus ensheathed by a PN. The asterisk denotes the axon initial segment. Arrows indicate intracellular aggrecan bodies of different size. (B) PN of a pallidal neuron shown in the tangential view. Arrowheads indicate ACs integrated in the PN. (C) ACs (arrowheads) aligned on a dendrite of a pallidal neuron. (D) Aggrecan body (arrow) different by its size from CRTL1-negative endosomes (violet grains) detected by immunoreaction for mannose-6-P receptor in a pallidal neuron. At the level of focus the PN and ACs can be identified. (E) Dense plexus of ACs associated with GAD-immunoreactive fibers and boutons (violet) in the subthalamic nucleus. The asterisk indicates a principal neuron contacted by coated axons. A GAD-positive interneuron (arrow) is devoid of PN. (F) ACs associated with boutons (arrows) of a GAD-immunoreactive varicose axon (violet) in the subthalamic nucleus. Scale bars⫽20 ␮m in A–E; 5 ␮m in F.

patterns described in monkeys using lectin histochemistry (Preuss et al., 1998). The lateral thalamic nuclei involved in motor functions, such as the ventral lateral nucleus, as well as the somatosensory thalamus were characterized by intense staining. In contrast, major association thalamic nuclei, such as the mediodorsal nucleus and the medial pulvinar nucleus, were largely devoid of immunoreactivity. At the cellular level, the aggrecan and link protein immunoreaction was related mainly to matrix-associated axons contacting principal thalamic neurons (Fig. 7A). These thalamic ACs typically contained clusters of fine

GAD-positive boutons collectively embedded in the extracellular matrix on dendritic domains (Fig. 7B). PNs were associated with a subpopulation of GAD-positive interneurons mainly located in territories containing ACs. The vast majority of thalamic GABAergic interneurons was devoid of PNs (Fig. 7B). Neurons in the reticular thalamic nucleus were surrounded by PNs, and, in contrast to interneurons, additionally showed a low number of intracellular aggrecan bodies (Fig. 7C). Pedunculopontine nucleus. In the pedunculopontine nucleus (Fig. 2F), neurons immunoreactive for choline

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

497

Table 3. Association of PNs, presynaptic ACs and intracellular aggrecan bodies with different types of neurons in the human basal ganglia circuit Region/neuronal transmitter

Perineuronal nets

Aggrecan bodies

Presynaptic coats

Caudate-putamen, interneurons/GABA Caudate-putamen, interneurons/acetylcholine Caudate-putamen, projection neurons/GABA Globus pallidus, external division/GABA Globus pallidus, internal division/GABA Substantia nigra, compact part/dopamine Ventral tegmental area/dopamine Retrorubral field/dopamine Substantia nigra, reticular part/GABA Substantia nigra, lateral part/GABA Subthalamic nucleus/glutamate Pedunculopontine nucleus/acetylcholine Pedunculopontine nucleus/gutamate Thalamus, ventral lateral nucleusb/glutamate Thalamus, subpopulation of interneurons/GABA Thalamus, reticular nucleus/GABA

⫹ O O ⫹ ⫹⫹ O O O ⫹ ⫹⫹ O O O O ⫹⫹ ⫹⫹

O O O ⫹⫹ ⫹⫹ O O O ⫹ ⫹⫹ O O O O O ⫹

O O O ⫹ ⫹ ⫾/⫹/⫹⫹a O/⫾ O/⫾ ⫹ ⫹⫹ ⫹⫹ O ⫹⫹ ⫹⫹ ⫹ O

Explanation of symbols: O absence, ⫾ low number, ⫹ moderate number, ⫹⫹ high number of immunoreactive structures. Highest density of presynaptic coats in the posterolateral division of the compact part of the substantia nigra. b Principal neurons. a

acetyltransferase were devoid of PNs in the compact and diffuse part of the nucleus (Fig. 8). These neurons were only rarely contacted by axons associated with ACs. Clearly contoured PNs and neurons densely covered by ACs occurred in close proximity to the cholinergic neurons in both pedunculopontine territories.

vious study (Horn et al., 2003). In addition to the oculomotor nucleus, PNs were located in the rostral interstitial nucleus of the medial longitudinal fascicle (Fig. 2C, D) and in the interstitial nucleus of Cajal (Fig. 2E). Net-associated neurons in oculomotor nuclei typically contained numerous aggrecan bodies (not shown).

Other regions. Among territories clearly labeled in aggrecan- and link protein–immunoreacted sections, the lateral habenular nucleus was characterized by a dense network of ACs in the neuropil (Fig. 2E). In contrast, nuclei related to oculomotor functions and rapid eye movements contained intensely stained, clearly contoured PNs. Using the CSPG antibody, PNs in these nuclei have been revealed in a pre-

DISCUSSION The present study revealed clear regional distribution patterns of aggrecan and the associated link protein in the human basal ganglia circuit. At the cellular level, the aggrecan-based matrix was postsynaptically associated with PNs around different types of neurons, and with ACs en-

Fig. 4. PNs and ACs in the putamen revealed by immunoreaction for CRTL1 link protein combined with detection of neuronal markers for interneurons. (A) Parvalbumin-immunoreactive interneuron (brown) associated with a PN. (B) Interneuron immunoreactive for choline acetyltransferase (brown) devoid of a PN. Arrows indicate PNs surrounding non-cholinergic interneurons. Asterisks denote projection neurons. (C) ACs (arrowheads) in the striatal neuropil. Parvalbumin-immunoreactive axon-like profiles (brown) are devoid of matrix coats. Scale bars⫽20 ␮m in A, B; 10 ␮m in C.

498

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

Fig. 5. Compartmentalization of the substantia nigra revealed by immunoreaction for CRTL1 link protein (black) and TH (brown). (A–C) Distribution patterns of the aggrecan-based extracellular matrix shown in frontal sections at three levels in a rostro-caudal direction. (A) Anterior subnuclei (am, ai, al) of the SNC expressing weak CRTL1 immunostaining. The SNL and SNR division are characterized by numerous PNs. (B) Substantia nigra ventrally adjacent to the RN in its full extent. The SNCpl, SNCps, and SNCpm are characterized by intense CRTL1 staining. The arrow denotes the position of subnuclei displayed in E, F. (C) Caudal part of the substantia nigra showing intense CRTL1 staining in SNCpl, SNCps and SNCpm. (D–G) At higher magnification the CRTL1 immunoreactivity is revealed to be associated with ACs in the different subnuclei of the SNC. Clear differences between subnuclei exist in the density of immunoreactive profiles contacting TH-positive neurons devoid of PNs. (D) Posterolateral subnucleus. (E, F) Neighboring cell groups in the ventral tier. (G) The ventral tegmental area. Scale bars⫽500 ␮m in A–C; 20 ␮m in D–G.

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

499

Fig. 6. PNs and ACs in the lateral part of the substantia nigra revealed by immunoreaction for CRTL1 link protein. (A) Double staining for CRTL1 (black) and TH (brown) shows the association of PNs with somata. Dendrites are contacted by numerous matrix-coated axons. Two TH-positive neurons (arrows) are devoid of PNs. The boxed regions are shown at higher magnification in B. Scale bars⫽20 ␮m.

sheathing preterminal parts of axons (Fig. 9). It appears that both the number and spatial arrangement of the postsynaptic and presynaptic matrix compartments essentially contribute to the structural and neurochemical organization of the individual regions. PNs are associated with different types of neurons Neurons ensheathed by PNs belong to morphologically and functionally different types (Brückner et al., 1993; Celio et al., 1998). In the human basal ganglia complex, clearly contoured PNs are associated with GABAergic projection neurons in the external and internal division of the globus pallidus, neurons in the lateral and reticular part of the substantia nigra, as well as striatal and thalamic GABAergic interneurons. These PN-ensheathed neurons

additionally express parvalbumin (Waldvogel and Faull, 1993; Parent et al., 1996; Holt et al., 1997; Morel et al., 2002; Waldvogel et al., 2004). Intracellular compartments indicate synthesis of aggrecan in net-associated neurons It was a new finding of this study that many neurons associated with PNs in the globus pallidus and the lateral and reticular part of the substantia nigra contained clusters of aggrecan- and link protein–immunoreactive granular inclusions. In contrast to ubiquitously distributed endosomes, these intracellular compartments indicate that the PN-ensheathed neurons are the producers of their extracellular aggrecan-based coat, as immunocytochemically revealed in neuronal cell cultures (Lander et

Fig. 7. PNs and ACs in the thalamus revealed by immunoreaction for CRTL1 link protein. (A) Principal neuron in the ventrolateral thalamic nucleus typically associated with aggrecan-based matrix on dendritic domains. The cell body (asterisk) is devoid of a PN and only sparsely contacted by immunoreactive profiles. (B) Double staining for CRTL1 (brown) and GAD (black) reveals varicose GABAergic axons collectively ensheathed by aggrecan-based matrix on dendritic domains related to a principal neuron (black asterisk). White asterisk labels a GAD-positive interneuron devoid of a PN. (C) PN associated with a neuron in the Rt. The net typically shows the diffuse perisomatic phenotype. Intracellular aggrecan bodies (arrows) can be identified. Scale bars⫽20 ␮m.

500

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

Fig. 8. PNs and ACs in the PPTg revealed by immunoreaction for CRTL1 link protein. Double staining for CRTL1 (black) and choline acetyltransferase (brown) shows the association of PNs with somata of noncholinergic neurons. (A) Diffuse part of the pedunculopontine nucleus at low magnification. Numerous ACs are distributed in the neuropil. (B) High magnification reveals ACs (arrowheads) in continuity with a PN. Scale bars⫽20 ␮m.

al., 1998), as well as by in situ hybridization in the rat and murine brain (Matthews et al., 2002; Carulli et al., 2006, 2007; McRae et al., 2007). Similar to the pericellular matrix of chondrocytes (Lou et al., 2000), aggrecan may be secreted and crosslinked with other matrix com-

ponents to form PNs on perisomatic domains, as well as ACs on preterminal and synaptic axonal domains of neurons. As suggested from previous studies, glial cells may provide matrix components supplementary to neuronal ag-

Fig. 9. Organization principles of the aggrecan-based extracellular matrix in the human basal ganglia circuit. Four major phenotypes may be distinguished. (A) GABAergic pallidal and nigral projection neurons are associated with PNs and ACs. The neurons contain clusters of aggrecan bodies. (B) GABAergic striatal interneurons are ensheathed by PNs. The neurons are devoid of aggrecan bodies and not associated with ACs. (C) Glutamatergic principal neurons of the subthalamic nucleus and motor thalamus are devoid of PNs but are contacted by numerous ACs. Matrixassociated boutons are preferentially localized on proximal dendrites in the thalamus. (D) Pigmented dopaminergic neurons of the substantia nigra are devoid of PNs. The number of ACs contacting the neurons is clearly different in the individual subnuclei. Especially the posterolateral subnuclei are characterized by a high density of ACs. The ventral tegmental area is virtually devoid of ACs. The latter phenotype is also representative of cholinergic neurons of the PPTg.

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

grecan during postnatal formation and the maintenance of PNs and ACs in the adult state (Brückner et al., 1993; Derouiche et al., 1996). Axons with aggrecan coats frequently contact neurons devoid of PNs The perisynaptic localization of extracellular matrix components surrounding synapses in typical PNs has been described in different CNS regions in a number of species (Hockfield and McKay, 1983; Brückner et al., 1993; Murakami and Ohtsuka, 2003). In the present study, the selective association of preterminal axonal profiles with an aggrecan-related matrix was a characteristic of regions devoid of clearly contoured PNs, such as the dorsolateral part of the compact substantia nigra, the subthalamic nucleus and the ventrolateral thalamus. A similar correlation has been demonstrated in rats for the brain-specific link protein Bral2 showing perisynaptic immunoreactivity in the target regions of axonal projections of neurons synthesizing Bral2 (Bekku et al., 2003). The GABAergic character of many of the axons associated with ACs in the basal ganglia circuit appears to be evident. However, the origin of axons ensheathed by ACs in the human basal ganglia circuit and the related output regions currently remains a matter of speculation. The axonal GAD-immunoreactive ‘pericellular nets,’ discussed to be formed by pallidal afferents in the human subthalamic nucleus (Lévesque and Parent, 2005), are reminiscent of AC-associated axonal profiles found in the present study. Aggrecan-based matrix clearly differentiates nigral territories The parcellation of the nigral complex has previously been revealed by established neuronal markers (Pearson et al., 1983; Braak and Braak, 1986; McRitchie et al., 1995, 1996; Damier et al., 1999a; Bolam et al., 2000; Smith and Kieval, 2000; Karachi et al., 2002). However, the aggrecan-based extracellular matrix further differentiates distinct classes of subnuclei. The dopaminergic nigral neurons devoid of PNs in mammals (Hobohm et al., 1998; Brückner et al., 2006) were clustered in subnuclei characterized by different density of AC-associated axons. In the posterolateral neuronal group of the pars compacta the density of ACs was very high, resembling the pattern seen in the subthalamic nucleus. This nigral subnucleus largely corresponds with nigrosome 1 described by Damier et al. (1999a). In contrast, the ventral tegmental area and the retrorubral field were virtually devoid of ACs. A specific pattern was also found in the GABAergic lateral part of the substantia nigra. Numerous PNs were contacted by AC-associated axons, indicating structural and functional properties of the lateral subnucleus different from all other parts of the nigral complex. This is in agreement with cytochemical and tract-tracing studies in monkeys and cats, describing differences in morphology, immunocytochemistry, afferents, and site of termination of axonal projections in the midbrain between pars lateralis and pars reticulata (May and Hall, 1986; Ma, 1989).

501

Functional relevance of extracellular matrix organization in the basal ganglia circuit Many recent investigations demonstrated the involvement of the extracellular matrix in mechanisms of synaptic plasticity and the regulation of regenerative capacity of neurons (Pizzorusso et al., 2002; Dityatev and Schachner, 2003; Rauch, 2004; Galtrey and Fawcett, 2007; Galtrey et al., 2007). It is therefore conceivable that PNs and ACs form an extracellular scaffold which hinders structural changes and the mobility of synaptic contacts. However, not excluding this stabilizing and plasticity-inhibiting function, our results strongly suggest the physiological relevance of PNs and ACs. The aggrecan-based extracellular matrix showed a largely identical distribution pattern with immunoreactivity for parvalbumin at the regional level. This is obvious especially in regions showing a dense network of AC-associated afferents, such as the subthalamic nucleus and the lateral thalamus. As discussed in previous studies (Härtig et al., 1999; Morris and Henderson, 2000; Horn et al., 2003; Dityatev et al., 2007), the coexistence of parvalbumin and PNs indicates a prevalent association of the aggrecan-based extracellular matrix with fast neuronal activity. Net-associated striatal interneurons have been characterized as fast-spiking neurons (Kawaguchi, 1997; Tepper and Bolam, 2004). Pallidal neurons and neurons of the reticular part of the substantia nigra are tonically active and fire rapidly (Gerfen, 1992; Atherton and Bevan, 2005; Hikosaka et al., 2006). In contrast, the dopaminergic neurons and the cholinergic neurons of the pedunculopontine nucleus have a modulatory function with variable activity patterns (Pahapill and Lozano, 2000; Marinelli et al., 2006). ACs may sustain the synaptic information processing of neurons associated with PNs. However, it is conceivable that ACs may also be specifically involved in presynaptic mechanisms, such as the modulatory function of presynaptic GABAB receptors shown in the rat subthalamic nucleus (Shen and Johnson, 2001; Chen and Yung, 2005). Pathophysiological aspects Previous studies indicated that PNs may protect neurons against damage (Brückner et al., 1999; Morawski et al., 2004, 2005). However, extracellular matrix components may also represent key factors in the initiation and progression of neurodegeneration and the related functional alterations. Severe matrix destruction has been shown as a result of viral infections in humans (Belichenko et al., 1997, 1999), monkeys (Medina-Flores et al., 2004) and sheep (Vidal et al., 2006), as well as after experimental stroke in rats (Hobohm et al., 2005). It is conceivable that the loss of the region-specific organization of the extracellular matrix may severely affect diverse forms of structural plasticity, such as retraction and formation of spines (Gerfen, 2006), but may result also in changes of the electrical activity of neuronal networks (Dityatev et al., 2007). In the basal ganglia circuit, the loss of extracellular matrix integrity may contribute to the increased neuronal activity shown to be related to degeneration of dopaminergic neu-

502

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

rons in PD (Blandini et al., 2000; Rodriguez-Oroz et al., 2001; Jellinger, 2002; Gerfen, 2006; Tan et al., 2006). The different vulnerability of nigral cell groups in PD may partly depend on the organization of extracellular matrix components. Interestingly, the posterolateral region, showing the highest density of ACs, largely corresponds with the highly vulnerable nigrosome 1 (Damier et al., 1999b).

CONCLUSIONS The results of our study indicate that the aggrecan-based extracellular matrix is a substantial chemoarchitectural component of the basal ganglia circuit. Potential changes of the extracellular matrix related to the degeneration of dopaminergic neurons may influence the pathophysiology of PD. Extracellular matrix organization in the substantia nigra and in functionally connected brain regions may be a factor to be considered if new therapeutic strategies are developed. Acknowledgments—The authors wish to thank Mrs. Hildegard Gruschka for excellent technical assistance and Jens T. Stieler for the help in macrophotography and image processing. This work was supported by Deutsche Forschungsgemeinschaft (Gert Brückner, BR 1208/3-4; Thomas Arendt, AR 200/6-1); Deutsche Forschungsgemeinschaft GRK 1097 “INTERNEURO” (Markus Morawski); Interdisciplinary Center of Clinical Research (IZKF) at the Faculty of Medicine of the Universität Leipzig (Markus Morawski, project C01) in the course of the MD/PhD program at the Universität Leipzig.

REFERENCES Adams I, Brauer K, Arélin C, Härtig W, Fine A, Mäder M, Arendt T, Brückner G (2001) Perineuronal nets in the rhesus monkey and human basal forebrain including basal ganglia. Neuroscience 108:285–298. Asher RA, Scheibe RJ, Keiser HD, Bignami A (1995) On the existence of a cartilage-like proteoglycan and link proteins in the central nervous system. Glia 13:294 –308. Atherton JF, Bevan MD (2005) Ionic mechanisms underlying autonomous action potential generation in the somata and dendrites of GABAergic substantia nigra pars reticulata neurons in vitro. J Neurosci 25:8272– 8281. Augood SJ, Waldvogel HJ, Münkle MC, Faull RLM, Emson PC (1999) Localization of calcium-binding proteins and GABA transporter (GAT-1) messenger RNA in the human subthalamic nucleus. Neuroscience 88:521–534. Baig S, Wilcock GK, Love S (2005) Loss of perineuronal net Nacetylgalactosamine in Alzheimer’s disease. Acta Neuropathol 110:393– 401. Bekku Y, Su WD, Hirakawa S, Fässler R, Ohtsuka A, Kang JS, Sanders J, Murakami T, Ninomiya Y, Oohashi T (2003) Molecular cloning of Bral2, a novel brain-specific link protein, and immunohistochemical colocalization with brevican in perineuronal nets. Mol Cell Neurosci 24:148 –159. Belichenko PV, Miklossy J, Celio MR (1997) HIV-I induced destruction of neocortical extracellular matrix components in AIDS victims. Neurobiol Dis 4:301–310. Belichenko PV, Miklossy J, Belser B, Budka H, Celio MR (1999) Early destruction of the extracellular matrix around parvalbumin-immunoreactive interneurons in Creutzfeldt-Jacob disease. Neurobiol Dis 6:269 –279. Bertolotto A, Rocca G, Canavese G, Migheli A, Schiffer D (1991) Chondroitin sulfate proteoglycan surrounds a subset of human and rat CNS neurons. J Neurosci Res 29:225–234.

Blandini F, Nappi G, Tassorelli C, Martignoni E (2000) Functional changes of the basal ganglia circuitry in Parkinson’s disease. Prog Neurobiol 62:63– 88. Bolam JP, Hanley JJ, Booth PAC, Bevan MD (2000) Synaptic organisation of the basal ganglia. J Anat 196:527–542. Braak H, Braak E (1986) Nuclear configuration and neuronal types of the nucleus niger in the brain of the human adult. Hum Neurobiol 5:71– 82. Brückner G, Brauer K, Härtig W, Wolff JR, Rickmann MJ, Derouiche A, Delpech B, Girard N, Oertel WH, Reichenbach A (1993) Perineuronal nets provide a polyanionic, glia-associated form of microenvironment around certain neurons in many parts of the rat brain. Glia 8:183–200. Brückner G, Hausen D, Härtig W, Drlicek M, Arendt T, Brauer K (1999) Cortical areas abundant in extracellular matrix chondroitin sulphate proteoglycans are less affected by cytoskeletal changes in Alzheimer’s disease. Neuroscience 92:791– 805. Brückner G, Pavlica S, Morawski M, Palacios AG, Reichenbach A (2006) Organization of brain extracellular matrix in the Chilean fat-tailed mouse opossum Thylamys elegans (Waterhouse, 1839). J Chem Neuroanat 32:143–158. Carulli D, Rhodes KE, Brown DJ, Bonnert TP, Pollack SJ, Oliver K, Strata P, Fawcett JW (2006) Composition of perineuronal nets in the adult rat cerebellum and the cellular origin of their components. J Comp Neurol 494:559 –577. Carulli D, Rhodes KE, Fawcett JW (2007) Upregulation of aggrecan, link protein 1, and hyaluronan synthases during formation of perineuronal nets in the rat cerebellum. J Comp Neurol 501:83–94. Celio MR, Spreafico R, De Biasi S, Vitellaro-Zuccarello L (1998) Perineuronal nets: past and present. Trends Neurosci 21:510– 515. Chen L, Yung W (2005) Tonic activation of presynaptic GABAB receptors on rat pallidosubthalamic terminals. Acta Pharm Sinica 26:10 –16. Damier P, Hirsch EC, Agid Y, Graybiel AM (1999a) The substantia nigra of the human brain. I. Nigrosomes and the nigral matrix, a compartmental organization based on calbindin D28K immunohistochemistry. Brain 122:1421–1436. Damier P, Hirsch EC, Agid Y, Graybiel AM (1999b) The substantia nigra of the human brain. II. Patterns of loss of dopamine-containing neurons in Parkinson’s disease. Brain 122:1437–1448. DeLong MR, Wichmann T (2007) Circuits and circuit disorders of the basal ganglia. Arch Neurol 64:20 –24. Delpech B, Delpech A, Brückner G, Girard N, Maingonnat C (1989) Hyaluronan and hyaluronectin in the nervous system. In: The biology of hyaluronan. Ciba Foundation Symposium 143 (Evered D, Whelan J, eds), pp 208 –232. Chichester: John Wiley & Sons. Derouiche A, Härtig W, Brauer K, Brückner G (1996) Spatial relationship of lectin-labelled extracellular matrix and glutamine synthetase-immunoreactive astrocytes in rat cortical forebrain regions. J Anat 189:363–372. Dityatev A, Schachner M (2003) Extracellular matrix molecules and synaptic plasticity. Nat Rev Neurosci 4:456 – 468. Dityatev A, Brückner G, Dityateva G, Grosche J, Kleene R, Schachner M (2007) Activity-dependent formation and functions of chondroitin sulfate-rich extracellular matrix of perineuronal nets. J Neurobiol 67:570 –588. Evers O, Uylings HBM (1997) An optimal antigen retrieval method suitable for different antibodies on human brain tissue stored for several years in formaldehyde fixative. J Neurosci Methods 72:197–207. Fearnley JM, Lees AJ (1991) Ageing and Parkinson’s disease: substantia nigra regional selectivity. Brain 114:2283–2301. Galtrey CM, Asher RA, Nothias F, Fawcett JW (2007) Promoting plasticity in the spinal cord with chondroitinase improves functional recovery after peripheral nerve repair. Brain 130:926 –939. Galtrey CM, Fawcett JW (2007) The role of chondroitin sulfate proteoglycans in regeneration and plasticity in the central nervous system. Brain Res Rev 54:1–18.

G. Brückner et al. / Neuroscience 151 (2008) 489 –504 Gerfen CR (1992) The neostriatal mosaic: multiple levels of compartmental organization. Trends Neurosci 15:133–139. Gerfen CR (2006) Indirect-pathway neurons lose their spines in Parkinson disease. Nat Neurosci 9:157–158. Graybiel AM, Ragsdale CW (1978) Histochemically distinct compartments in the striatum of human, monkeys and cat demonstrated by acetylcholinesterase staining. Proc Natl Acad Sci U S A 75: 5723–5726. Halliday GM, Macdonald V, Henderson JM (2005) A comparison of degeneration in the motor thalamus and cortex between progressive supranuclear palsy and Parkinson’s disease. Brain 128:2272– 2280. Hamani C, Dostrowsky JO, Lozano AM (2006) The motor thalamus in neurosurgery. Neurosurgery 58:146 –158. Hardman CD, McRitchie DA, Halliday GM, Cartwright HR, Morris JGL (1996) Substantia nigra pars reticulata neurons in Parkinson’s disease. Neurodegeneration 5:49 –55. Hardman CD, Halliday GM, McRitchie DA, Morris JGL (1997) The subthalamic nucleus in Parkinson’s disease and progressive supranuclear palsy. J Neuropathol Exp Neurol 56:132–142. Härtig W, Brauer K, Brückner G (1992) Wisteria floribunda agglutininlabelled nets surround parvalbumin-containing neurons. Neuroreport 3:869 – 872. Härtig W, Derouiche A, Welt K, Brauer K, Grosche J, Mäder M, Reichenbach A, Brückner G (1999) Cortical neurons immunoreactive for the potassium channel Kv3.1b subunit are predominantly surrounded by perineuronal nets presumed as a buffering system for cations. Brain Res 842:15–29. Hausen D, Brückner G, Drlicek M, Härtig W, Brauer K, Bigl V (1996) Pyramidal cells ensheathed by perineuronal nets in human motor and somatosensory cortex. Neuroreport 7:1725–1729. Hikosaka O, Nakamura K, Nakahara H (2006) Basal ganglia orient eyes to reward. J Neurophysiol 95:567–584. Hirsch EC (1999) Mechanism and consequences of nerve cell death in Parkinson’s disease. J Neural Transm 56:127–137. Hirsch EC, Graybiel AM, Duyckaerts C, Javoy-Agid F (1987) Neuronal loss in the pedunculopontine tegmental nucleus in Parkinson’s disease and in progressive supranuclear palsy. Proc Natl Acad Sci U S A 84:5976 –5980. Hobohm C, Günther A, Grosche J, Ro␤ner S, Schneider D, Brückner G (2005) Decomposition and long-lasting downregulation of extracellular matrix in perineuronal nets induced by focal cerebral ischemia in rats. J Neurosci Res 80:539 –548. Hobohm C, Härtig W, Brauer K, Brückner G (1998) Low expression of extracellular matrix components in rat brain stem regions containing modulatory aminergic neurons. J Chem Neuroanat 15:135–142. Hockfield S, McKay RDG (1983) A surface antigen expressed by a subset of neurons in the vertebrate central nervous system. Proc Natl Acad Sci U S A 80:5758 –5761. Holt DJ, Graybiel AM, Saper CB (1997) Neurochemical architecture of the human striatum. J Comp Neurol 384:1–25. Horn AK, Brückner G, Härtig W, Messoudi A (2003) Saccadic omnipause and burst neurons in monkeys and human are ensheathed by perineuronal nets but differ in their expression of calciumbinding proteins. J Comp Neurol 455:341–352. Jellinger KA (2002) Recent developments in the pathology of Parkinson’s disease. J Neural Transm 62:347–376. Karachi C, Francois C, Parain K, Bardinet E, Tande D, Hirsch E, Yelnik J (2002) Three-dimensional cartography of functional territories in the human striatopallidal complex using calbindin immunoreactivity. J Comp Neurol 450:122–134. Kawaguchi Y (1997) Neostriatal cell subtypes and their functional roles. Neurosci Res 27:1– 8. Konigsmark BW (1970) Methods for the counting of neurons. In: Contemporary research methods in neuroanatomy (Nauta WHJ, Ebbesson SOE, eds), pp 315–380. Berlin: Springer.

503

Köppe G, Brückner G, Härtig W, Delpech B, Bigl V (1997) Characterization of proteoglycan-containing perineuronal nets by enzymatic treatments of rat brain sections. Histochem J 29:11–20. Lander C, Zhang H, Hockfield S (1998) Neurons produce a neuronal cell surface-associated chondroitin sulfate proteoglycan. J Neurosci 18:174 –183. Lévesque JC, Parent A (2005) GABAergic interneurons in human subthalamic nucleus. Mov Disord 20:574 –584. Lou W, Guo C, Zheng J, Chen TL, Wang PY, Vertel BM (2000) Aggrecan from start to finish. J Bone Miner Metab 18:51–56. Ma TP (1989) Identification of the substantia nigra pars lateralis in the macaque using cytochrome oxidase and fiber stains. Brain Res 480:305–311. Ma SY, Rinne JO, Collan Y, Royotta M, Rinne UK (1996) A quantitative morphometrical study of neuron degeneration in the substantia nigra in Parkinson’s disease. J Neurol Sci 140:40 – 45. Mai JK, Assheuer J, Paxinos G, eds (1997) Atlas of the human brain. San Diego: Academic Press. Marinelli M, Rudick CN, Hu XT, White FJ (2006) Excitability of dopamine neurons: modulation and physiological consequences. CNS Neurol Dis Drug Targets 5:79 –97. Massey JM, Amps J, Viapiano MS, Matthews RT, Wagoner MR, Whitaker CM, Alilain W, Yonkof AL, Khalyfa A, Cooper NG, Silver J, Onifer SM (2007) Increased chondroitin sulfate proteoglycan expression in denervated brain stem targets following spinal cord injury creates a barrier to axonal regeneration overcome by chondroitinase ABC and neurotrophin-3. Exp Neurol, doi: 10.1016/ j.expneurol.2007.03.029. Matthews RT, Kelly GM, Zerillo CA, Gray G, Tiemeyer M, Hockfield S (2002) Aggrecan glycoforms contribute to the molecular heterogeneity of perineuronal nets. J Neurosci 22:7536 –7547. May PJ, Hall WC (1986) The sources of the nigrotectal pathway. Neuroscience 19:159 –180. McRae P, Rocco MM, Kelly G, Brumberg JC, Matthews RT (2007) Sensory deprivation alters aggrecan and perineuronal net expression in the mouse barrel cortex. J Neurosci 27:5405–5413. McRitchie DA, Halliday GM, Cartwrith H (1995) Quantitative analysis of the variability of substantia nigra pigmented cell clusters in human. Neuroscience 68:539 –551. McRitchie DA, Hardman CD, Halliday GM (1996) Cytoarchitectural distribution of calcium binding proteins in midbrain dopaminergic regions of rats and humans. J Comp Neurol 364:121–150. Medina-Flores R, Wang G, Bissel SJ, Murphey-Corb M, Wiley CA (2004) Destruction of extracellular matrix proteoglycans is pervasive in simian retroviral neuroinfection. Neurobiol Dis 16:604 – 616. Mesulam MM, Geula C, Bothwell, Hersh LB (1989) Human reticular formation: Cholinergic neurons of the pedunculopontine and laterodorsal tegmental nuclei and some cytochemical comparisons to forebrain cholinergic neurons. J Comp Neurol 281:611– 633. Morawski M, Brückner MK, Riederer P, Brückner G, Arendt T (2004) Perineuronal nets potentially protect against oxidative stress. Exp Neurol 188:309 –315. Morawski M, Meinecke CH, Reinert T, Dörffel AC, Riederer P, Arendt T, Butz T (2005) Determination of trace elements in the human substantia nigra. Nucl Inst Methods B 231:224 –228. Morel A, Loup F, Magnin M, Jeanmonod D (2002) Neurochemical organization of the human basal ganglia: Anatomofunctional territories defined by the distributions of calcium-binding proteins and SMI-32. J Comp Neurol 443:86 –103. Morel A, Magnin M, Jeanmonod D (1997) Multiarchitectonic and stereotactic atlas of the human thalamus. J Comp Neurol 387:588–630. Moro E, Lang AE (2006) Criteria for deep-brain stimulation in Parkinson’s disease: review and analysis. Expert Rev Neurother 6:1695– 1705. Morris NP, Henderson Z (2000) Perineuronal nets ensheath fast spiking, parvalbumin-immunoreactive neurons in the medial septum/ diagonal band complex. Eur J Neurosci 12:828 – 838.

504

G. Brückner et al. / Neuroscience 151 (2008) 489 –504

Münkle MC, Waldvogel HJ, Faull RLM (2000) The distribution of calbindin, calretinin, and parvalbumin immunoreactivity in the human thalamus. J Chem Neuroanat 19:155–173. Murakami T, Ohtsuka A (2003) Perisynaptic barrier of proteoglycans in the mature brain and spinal cord. Arch Histol Cytol 66:195–207. Neame PJ, Barry FP (1994) The link proteins. EXS 70:53–72. Pahapill PA, Lozano AM (2000) The pedunculopontine nucleus and Parkinson’s disease. Brain 123:1767–1783. Parent A, Fortin M, Côté PY, Cicchetti F (1996) Calcium-binding proteins in primate basal ganglia. Neurosci Res 25:309 –334. Paxinos G, Huang XF, Toga AW, eds (2000) The rhesus monkey brain in stereotaxic coordinates. San Diego: Academic Press. Pearson J, Goldstein M, Markey K, Brandeis L (1983) Human brainstem catecholamine neuronal anatomy as indicated by immunocytochemistry with antibodies to tyrosine hydroxylase. Neuroscience 8:3–32. Pizzorusso T, Medini P, Barardi N, Chierzi S, Fawcett JW, Maffei L (2002) Reactivation of ocular dominance plasticity in the adult visual cortex. Science 298:1248–1251. Preuss TM, Gray D, Cusick CG (1998) Subdivisions of the motor and somatosensory thalamus of primates revealed with Wisteria floribunda agglutinin histochemistry. Somatosens Mot Res 15:211–219. Rauch U (2004) Extracellular matrix components associated with remodeling processes in brain. Cell Mol Life Sci 61:2031–2045. Rodriguez-Oroz M, Rodriguez M, Guridi J, Mewes K, Chockkman V, Vitek J, DeLong MR, Obeso JA (2001) The subthalamic nucleus in Parkinson’s disease: somatotopic organization and physiological characteristics. Brain 124:1777–1790. Shen KZ, Johnson SW (2001) Presynaptic GABAB receptors inhibit synaptic inputs to rat subthalamic neurons. Neuroscience 108: 431– 436. Silver J, Miller JH (2004) Regeneration beyond the glial scar. Nat Rev Neurosci 5:146 –156. Smith Y, Kieval JZ (2000) Anatomy of the dopamine system in the basal ganglia. Trends Neurosci 23:S28 –S33. Syková E (2004) Diffusion properties of the brain in health and disease. Neurochem Int 45:453– 466. Tan SKH, Temel Y, Blokland A, Steinbusch HWM, Visser-Vandewalle V (2006) The subthalamic nucleus: From response selection to execution. J Chem Neuroanat 31:155–161.

Temel Y, Visser-Vandewalle V (2006) Targets for deep brain stimulation in Parkinson’s disease. Expert Opin Ther Targets 10:355–362. Tepper JM, Bolam JP (2004) Functional diversity and specificity of neostriatal interneurons. Curr Opin Neurobiol 14:685– 692. van Domburg PHMF, ten Donkelaar HJ, eds (1991) The human substantia nigra and ventral tegmental area. A neuroanatomical study with notes on aging and aging diseases. New York: Springer Verlag. van Horssen J, Wesseling P, van den Heuvel LPWJ, de Waal RMW, Verbeek MM (2003) Heparan sulphate proteoglycans in Alzheimer’s disease and amyloid-related disorders. Lancet Neurol 2: 482– 492. Viapiano MS, Matthews RT (2006) From barriers to bridges: chondroitin sulfate proteoglycans in neuropathology. Trends Mol Med 12:488 – 496. Vidal E, Bolea R, Tortosa R, Costa C, Domènech A, Monleón E, Vargas A, Badiola JJ, Pumarola M (2006) Assessment of calciumbinding proteins (parvalbumin and calbindin D-28K) and perineuronal nets in normal and scrapie-affected adult sheep brains. J Virol Methods 136:137–146. Wakabayashi K, Mori F, Takahashi H (2006) Progression patterns of neuronal loss and Lewy body pathology in the substantia nigra in Parkinson’s disease. Parkinsonism Relat Disord 12:S92–S98. Waldvogel HJ, Baer K, Allen KL, Rees MI, Faull RL (2007) Glycine receptors in the striatum, globus pallidus, and substantia nigra of the human brain: an immunohistochemical study. J Comp Neurol 502:1012–1029. Waldvogel HJ, Billinton A, White JH, Emson PC, Faull RLM (2004) Comparative cellular distribution of GABAA and GABAB receptors in the human basal ganglia: Immunohistochemical colocalization of the ␣1 subunit of the GABAA receptor, and the GABABR1 and GABABR2 receptor subunits. J Comp Neurol 470:339 –356. Waldvogel HJ, Faull RLM (1993) Compartmentalization of parvalbumin immunoreactivity in the human striatum. Brain Res 610:311–316. Yamaguchi Y (2000) Chondroitin sulfate proteoglycans in the nervous system. In: Proteoglycans. Structure, biology, and molecular interactions (Iozzo RV, ed), pp 379 – 402. New York: Marcel Decker. Yasuhara O, Akiyama H, McGeer E, McGeer OL (1994) Immunohistochemical localization of hyaluronic acid in rat and human brain. Brain Res 635:269 –282.

(Accepted 2 November 2007) (Available online 7 November 2007)