Accepted January 24, 2003.
Expression and Cell Distribution of Group I and Group II Metabotropic Glutamate Receptor Subtypes in Taylor-type Focal Cortical Dysplasia
Article first published online: 9 JUN 2003
Volume 44, Issue 6, pages 785–795, June 2003
How to Cite
Aronica, E., Gorter, J. A., Jansen, G. H., Van Veelen, C. W. M., Van Rijen, P. C., Ramkema, M. and Troost, D. (2003), Expression and Cell Distribution of Group I and Group II Metabotropic Glutamate Receptor Subtypes in Taylor-type Focal Cortical Dysplasia. Epilepsia, 44: 785–795. doi: 10.1046/j.1528-1157.2003.54802.x
- Issue published online: 9 JUN 2003
- Article first published online: 9 JUN 2003
- Metabotropic glutamate receptors;
- Focal cortical dysplasia;
- Balloon cells;
Summary: Purpose: Focal cortical dysplasia (FCD) is known to be a major cause of intractable epilepsy. The cellular mechanism(s) underlying the epileptogenicity of FCD remain largely unknown. Because recent studies indicate that metabotropic glutamate receptor subtypes (mGluRs) play a role in epileptogenesis, we investigated the expression and cellular distribution pattern of mGluRs in FCD specimens.
Methods: Immunocytochemical expression of group I and group II mGluR subtypes was investigated in 15 specimens of human FCD obtained during epilepsy surgery.
Results: Strong mGluR1α and mGluR5 (group I mGluRs) immunoreactivity (IR) was observed in the majority of FCD specimens in dysplastic as well as in heterotopic neurons. mGluR1α was expressed in a subpopulation of neurons (mainly large dysplastic cells), whereas mGluR5 was represented in a higher percentage of dysplastic neuronal cells. Group II mGluRs (mGluR2/3) IR was observed less frequently than that in group I mGluRs and generally appeared in <10% of the dysplastic neurons. IR for all three mGluR subtypes was observed in balloon cells. mGluR2/3 appeared to be most frequently expressed in glial fibrillary acidic protein (GFAP)-positive balloon cells (glial type), and mGluR1α, in microtubule-associated protein (MAP)2-positive cells (neuronal type). mGluR5 was present in the majority of balloon cells. Occasionally glial mGluR1α IR was observed in bizarre glial cells with di- or multinuclei. Reactive astrocytes were intensively stained, mainly with mGluR5 and mGluR2/3.
Conclusions: The cellular distribution of mGluR subtypes, with high expression of mGluR1α and mGluR5 in dysplastic neurons, suggests a possible contribution of group I mGluRs to the intrinsic and high epileptogenicity of dysplastic cortical regions.
Focal cortical dysplasia (FCD) is a developmental disorder known to be a major cause of pediatric intractable epilepsy. It is histologically characterized by regions of abnormal cortical cytoarchitecture, with dyslaminated cortical layers (1,2). Characteristic cell types in FCD include large bizarre neurons within the cortex (dysplastic neurons), heterotopic neurons in white matter, and balloon cells with pleomorphic nuclei and ballooned opalescent eosinophilic cytoplasm (1–4). The intrinsic and high epileptogenicity of FCD is clearly supported by electrocorticographic, surgical, and immunocytochemical results, indicating the presence of a hyperexcitable neuronal component functionally integrated with excitatory pathways (2,5–7).
The cellular mechanism(s) underlying the epileptogenicity of FCD remain largely unknown (8). Recent work supports the role of developmental alterations of the balance between excitation and inhibition in the pathogenesis of epileptic focal discharges in pediatric patients (9). In particular, attention has been focused on the local pathways of excitatory amino acid synaptic transmission in the dysplastic cortex. Recently, changes in the expression, molecular composition, and functional properties of glutamate receptors have been described in human dysplastic neurons, as well as in animal models of cortical dysplasia [reviewed in (9)]. These previous studies focused on the ionotropic glutamate receptors (GluRs), which are ligand-gated ion channels including N-methyl-d-aspartate (NMDA) and non–NMDA-receptor subtypes (10–16). The cell-specific distribution and the role of metabotropic glutamate receptors (mGluRs) (17) in the epileptogenicity of FCD have not been defined.
mGluRs consist of at least eight subtypes that regulate a variety of intracellular signalling systems via activation of guanosine triphosphate (GTP)-binding proteins (18,19). They have been subdivided into three main groups. Group I includes mGluR1 and mGluR5, which are coupled to phosphoinositide hydrolysis. Group II (mGluR2 and mGluR3) as well as group III (mGluR4, -6, -7, -8) mGluRs are negatively coupled to adenylyl cyclase (19). Progress in the study of mGluRs indicates a role for these receptors in epileptogenesis and suggests that mGluRs may be considered a potential target for treatment of epilepsy (20,21). Several studies support the anticonvulsant effects of group II and III mGluR activation as opposed to the convulsant action of group I, which may induce seizure discharges and epileptogenesis in a variety of experimental models (22–29).
In the present study, immunocytochemistry with antibodies (Abs) specific for mGluR1, mGluR2/3, and mGluR5 was performed in surgical specimens of patients with severe FCD (Taylor type) and pharmacoresistant epilepsy. Our major aim was to provide data that may help to define the distribution of mGluRs within the heterogeneous cell population of the dysplastic region and may provide better insights into the mechanisms underlying the intrinsic and high epileptogenicity of FCD.
MATERIALS AND METHODS
The 15 cases included in this study were obtained from the files of the Departments of Neuropathology of the Academic Medical Center (University of Amsterdam) and the University Medical Center in Utrecht. Patients underwent resection of FCD for medically intractable epilepsy. Informed consent was obtained for the use of brain tissue and for access to medical records for research purposes. Two neuropathologists reviewed all patients independently and confirmed the diagnosis of FCD by following the classification system proposed by Mischel et al. (1) for grading the degree of FCD. Table 1 summarizes the clinical features (derived from patients' medical records) with particular attention to the characteristics of seizures (type and frequency of seizures, age at seizure onset, postoperative seizure outcome). The predominant type of seizure pattern was that of complex partial seizures, which were resistant to maximal doses of antiepileptic drugs (AEDs). All patients underwent presurgical evaluation (30). In all patients, the lesion was localized by brain magnetic resonance imaging (MRI); electroencephalographic (EEG) recordings were performed to detect the epileptogenic area. We classified the postoperative seizure outcome according to Engel (31). Class I consisted of patients who remained completely seizure free. We based our evaluation on a review of the patient files and/or by telephone interview with the patient and family. The follow-up period ranged from 2 to 9 years.
|Patient/age (yr)/sex||Lesion type||Age at onset (yr)||Duration of epilepsy (yr)||Seizure type||Location||Engel Class|
Paraffin-embedded tissue was sectioned at 5–6 μm and mounted on organosilane (3-aminopropylethoxysilane; Sigma-Aldrich, Zwijndrecht, the Netherlands)-coated slides. Representative sections of all specimens were processed for hematoxylin–eosin (H&E) and Nissl stains, as well as for immunocytochemical reactions by using a number of neuronal and glial markers described in Table 2.
|Antigen||Primary antibody||Source||Microwave treatment Sodium citrate (pH 6.0) 10 min, 650 W||Dilution|
|mGluR5||Polyclonal R||Upstate Biotech.||Required||1:100|
|NF||Mouse clone 2F11||Dako||Not required||1:500|
|NeuN||Mouse clone MAB377||Chemicon||Required||1:1,000|
|MAP2||Mouse clone HM2||Sigma||Required||1:100|
|GFAP||Polyclonal R||Dako||Not required||1:2,000|
|Vimentin||Mouse clone V9||Dako||Required||1:400|
Neuron specific enolase (NSE), neurofilament protein, synaptophysin (Dako, Glostrup, Denmark), neuronal nuclear protein (NeuN), microtubule-associated protein (MAP2), glial fibrillary acidic protein (GFAP), and vimentin were used in the routine immunocytochemical analysis to document the presence of a heterogeneous population of cell types, within the dysplastic cortex, including neuronal, glial, and glia–neuronal elements. For the detection of mGluRs, we used Abs specific for the mGluR subtypes 1α, 5, and 2/3. We used an anti-mGluR1 polyclonal Ab (AB 1551; Chemicon, Temecula, CA, U.S.A.), raised in rabbit against a 20-amino-acid peptide (PNVTYASVILRDYKQSSSTL), corresponding to the C-terminus of mGluR1α. mGluR2/3 rabbit polyclonal Ab (Chemicon) was raised against a C-terminus peptide of mGluR2 (NGREVVDSTTSSL); it recognizes both mGluR2 and mGluR3. mGluR5 rabbit polyclonal Ab (Upstate Biotech., U.S.A.) was raised against a 21-residue peptide (KSSPKYDTLIIRDYTNSSSSL) corresponding to the C-terminal of mGluR5. Characterization of these Abs (Table 2) was recently reported in human brain tissue. The Ab specificity was tested by preincubating the Abs with 100-fold excess of the antigenic peptides and by Western blots of the total homogenates of human control brain (32,33).
The sections were deparaffized in xylene and, after rinses in ethanol (100% and 95%), were incubated with 1% H2O2 diluted in methanol for 20 min. Slides were then washed with phosphate-buffered saline (PBS; 10 mM, pH 7.4). For the Abs indicated in Table 2, the slides were placed into sodium citrate buffer (0.01 M, pH 6.0) and heated in a microwave oven (650 W for 10 min). The slides were allowed to cool for 20 min in the same solution at room temperature (RT) and then washed in PBS. They were incubated with a mixture of 10% normal goat serum (NGS), for 1 h before the incubation with the primary Ab at dilutions specified in Table 2 (30 min at RT and at 4°C overnight). Sections were then washed with PBS and incubated at RT for 1 h with the appropriate biotinylated secondary Ab diluted in PBS [1:400 goat–anti-rabbit immunoglobulin (Ig) or 1:200 goat–anti-mouse or goat–anti-rabbit Ig (Dako, Glostrup, Denmark)]. Single-label immunocytochemistry was carried out by using the avidin–biotin peroxidase method (Vector Elite, Burlingame, CA, U.S.A.) and 3,3'-diaminobenzidine as chromogen. Sections were counterstained with hematoxylin, dehydrated in alcohol and xylene, and coverslipped. Sections incubated without the primary Ab or with preimmune sera were essentially blank. As positive controls for immunocytochemical staining, paraffin-embedded autopsy specimens of normal human cortex from four age-matched patients (one female, three male patients; mean age, 25 years) with no history of seizures or other neurologic diseases were used. In the analysis we also included four FCD specimens (patients 10, 11, 13, 14) that contained a sufficient amount of perilesional zone (histologically normal-appearing cortex/white matter adjacent to the lesion). Histologically normal tissue adjacent to the lesional zone represents ideal disease control tissue, because it has been exposed to the same seizure activity and drugs, had the same fixation time and, of course, is of the same age or sex.
For the double labeling, sections (after the incubation with the primary Abs; GFAP-monoclonal mouse; Chemicon; 1:500, MAP-2 or NeuN and mGluR1α, mGluR5, or mGluR2/3) were incubated for 2 h with a 1:200 dilution of ALEXA 488 goat–anti-mouse (Molecular Probes, Eugene, OR, U.S.A.) and CY3 goat–anti-rabbit IgG antisera (Zymed, San Francisco, CA, U.S.A.). To block autofluorescence due to the presence of lipofuscin pigment in the tissue, sections were stained with Sudan Black B (Merck, Darmstadt, Germany) for 10 min, as previously described (34). Sections were then analyzed by means of a laser scanning confocal microscope (Biorad, MRC1024) equipped with argon-ion laser.
Evaluation of immunostaining
Labeled tissue sections were examined by two observers with respect to the presence or absence of various histopathologic parameters and specific IR for the different markers in neurons and glial or glia–neuronal balloon cells. Two representative paraffin sections per case were stained with each Ab and assessed by two investigators independently; a consensus score was obtained. We rated the degree of mGluR staining on a semiquantitative 3-point scale in which immunoreactivity was defined as –, not present; +, moderate; ++, strong; Table 3). Moreover, the labeling indices (number of labeled neurons or balloon cells per total number of each cell type in the dysplastic cortex) were assigned semiquantitatively to four categories: (1) <1%; (2) 1–10%; (3) 11–50%; and (4) >50%. Neuronal cell bodies were differentiated from glia and glia–neuronal balloon cells on the basis of morphology, and cells were counted from 10 representative fields of two labeled sections of each tumor at a magnification of ×250, by using an ocular grid, as previously described (33). Only neurons in which the nucleus could be clearly identified were included. Balloon cells have eccentric nuclei and ballooned opalescent eosinophilic cytoplasm. Astrocytes were differentiated from neuronal and microglial cells on the basis of morphology and by staining of serial sections with Abs against GFAP and vimentin. Normal astrocytes were differentiated from reactive astrocyte cells on the basis of morphology and the absence of vimentin IR [as previously reported (35,36)]. Sections stained with NeuN and MAP-2 adjacent to those used for the mGluR staining also were studied.
|Focal cortical dysplasia|
|Dysplastic neurons (n = 15)||Heterotopic neurons (n = 15)||Glial cells (n = 15)||Balloon cells (n = 11)|
To analyze the percentage of mGluR-positive dysplastic neurons and balloon cells, sections labeled with mGluR1α, mGluR5, or mGluR2/3 and GFAP, MAP2, or NeuN were digitized by using an Olympus Vanox microscope equipped with a DP-10 digital camera (Olympus, Japan). Images (magnification, ×200) from representative fields of the lesion in two double-labeled sections of four FCD cases with abundant dysplastic neurons and balloon cells (patients 6, 13, 14, and 15) were collected with an Apple Macintosh PPC 82 computer. The total number of balloon cells stained with GFAP (or MAP2), as well as the number of balloon cells double-labeled for GFAP (or MAP2) and mGluR1α (or mGluR5 or mGluR2/3) was counted visually. We calculated the percentages of balloon cells IR for GFAP or MAP2 that also contained mGluR subtype IR. We also calculated the percentages of NeuN-positive dysplastic neurons IR for the different mGluR subtypes.
Case material and histologic features
The clinical features of the cases included in this study are summarized in Table 1. All 15 FCD patients had a history of chronic pharmacoresistant epilepsy. Postoperatively, 10 patients with FCD were completely seizure free (Engel's class I). The follow-up period ranged from 2 to 9 years. In this study we excluded patients with a mild degree of cortical dysplasia and no detectable lesion on the MRI, which could represent a nonspecific pathologic change associated with long-term seizure activity. The FCD cases included in this study have all the previously described histopathologic features of severe (Taylor-type) FCD (1,37), including cortical laminar disruption and presence of giant dysmorphic neurons and/or balloon cells (Table 1).
Metabotropic glutamate receptor expression in focal cortical dysplasia
Expression of both group I and II mGluR subtypes was observed in FCD specimens (Figs. 1 and 3). Differences in the expression level, as well as in the cell-specific distribution of the different subtypes (mGluR1α, mGluR5, and mGluR2/3) within the dysplastic region were found (Tables 3 and 4).
|mGluR subtypes||Dysplastic neurons NeuN+||Balloon cells (neuronal type) MAP2+||Balloon cells (glial type) GFAP+|
|mGluR1α||40.1 ± 7.8||90 ± 8.3||12.7 ± 3.4|
|mGluR5||83 ± 14||95 ± 2.4||78.1 ± 11.6|
|mGluR2/3||7.7 ± 1||10.8 ± 4||93.6 ± 4.9|
mGluR1α in neuronal and glial cells
In agreement with previous observations (33,38), mGluR1α IR was seen mainly in neurons with a diffuse distribution (and occasionally cell-membrane punctate labeling) in human control specimens, as well as in the normal-appearing cortex adjacent to the dysplastic region (Fig. 1A). mGluR1α IR was not detectable in resting glial cells from control cortex (Fig. 1A) and subcortical white matter [not shown (32,33)].
mGluR1α IR was observed within the dysplastic cortex (Fig. 1C), as well as in the subcortical white matter (Fig. 1G). The labeling was localized mainly in neuronal cells. The neuronal mGluR1α IR pattern (with somatodendritic labeling) was similar to that of control cortex, with the exception of selective enrichment in a population of large dysplastic neurons (Fig. 1D) and heterotopic neurons in the white matter (Fig. 1E). Dysplastic neurons, as well as heterotopic neuronal cells, displayed a strong mGluR1α somatodendritic IR in a large percentage of cases (Table 3). The neuronal labeling index in the dysplastic cortex usually ranged from 11 to 50% (Fig. 2). The percentage of neuronal mGluR1α expression was quantified in four FCD specimens with a high number of dysplastic neurons (Table 4). In the majority of cases, mGluR1α IR was not observed in glial cells (Table 3). mGluR1α often highlighted a group of dysplastic balloon cells in the white matter (Table 3; Fig. 1G), whereas glial markers (GFAP or vimentin) showed a more diffuse IR and, in some cases, a variable degree of reactive astrocytes, which however did not appear to express mGluR1α (Fig. 1H and I). mGluR1α was occasionally observed in bizarre glial cells with di- or multinuclei (Table 3; Fig. 1J).
mGluR5 in neuronal and glial cells
Immunocytochemistry performed with Abs directed against mGluR5 in control autopsy brain as well as in normal-appearing cortex adjacent to the dysplastic region (Fig. 3A) showed mainly neuronal labeling within the cytoplasm and at the neuronal membrane, as previously described (33,39,40). Only weak mGluR5 labeling was detectable in a few astrocytic processes within the control white matter [not shown (33)].
mGluR5 IR was highly represented in the neuronal as well as glial component of dysplastic regions (Table 3; Fig. 3). Apart from a strong expression in a population of large dysplastic neurons (Fig. 3B and C) and heterotopic neurons (Fig. 3D) within the white matter, the neuronal mGluR5 IR pattern in FCD was similar to that in control cortex. In a large majority of cases, neuronal cells (large dysplastic neurons and heterotopic neurons) were strongly stained within the cytoplasm and at the neuronal membrane; neuropil staining also was observed (Table 3; Fig. 3 C and D). A high percentage of neurons (labeling indices often >50%) showed mGluR5 IR in the dysplastic cortex (Fig. 2; Table 4). No significant differences in mGluR5 (or mGluR1α) protein neuronal expression (in terms of intensity of staining and immunoreactive cell numbers) were observed between patients with relatively short (<10 years) or long duration of epilepsy (>10 years). Strong mGluR5 IR also was observed in glial cells, in particular in regions where GFAP or vimentin highlighted reactive astrocytosis (Table 3; Fig. 3F).
mGluR2/3 in neuronal and glial cells
The polyclonal Ab mGluR2/3 recognizes both subtypes of group II mGluRs. As previously reported in normal control cortex, mGluR2/3 IR was demonstrated in neuropil as well as in neuronal cell bodies and associated process (33,38,41,42) (Fig. 3G). Glial cells in control brain were only lightly labeled (33).
mGluR2/3 IR in the dysplastic cortex was observed less frequently than group I mGluRs and generally appeared in <10% of the neurons (Fig. 2; Tables 3 and 4). In the majority of cases, neurons were only moderately labeled, and weaker neuropil staining was detected compared with the expression in the normal cortex (Table 3; Fig. 3G–J). mGluR2/3 labeling was observed mainly in the glial component of the dysplastic region (Fig. 3K).
mGluR expression in balloon cells
Expression of both group I (mGluR1α and mGluR5; Fig. 1K–O; Fig. 3E and F) and group II mGluRs (mGluR2/3; Fig. 3L and M) was observed in large balloon cells in FCD specimens containing this cell type. In the majority of the cases, moderate to strong IR was detected with mGluR1α and mGluR5 in a variable proportion of balloon cells (usually ranging from 11 to 50% for mGluR1α; >50% for mGluR5). mGluR2/3-positive balloon cells displayed mainly a moderate IR (labeling index, 11–50%); however, strong staining was observed in three cases (Table 3; Fig. 3L and M). The origin of these cells, showing a large number of morphologic changes, is still unclear. GFAP and MAP2, two proteins linked to the glial and neuronal cytoskeleton, respectively, in the mature central nervous system, have been previously shown to be useful for the detection of two types of balloon cells [the glial and the neuronal types (43)]. In agreement with this report (43), we found balloon cells with prominent nucleoli and IR for MAP2, whereas others were reactive for GFAP (not shown). Thus to investigate the specific population of the balloon cells that are IR for a specific mGluR protein, double labeling of mGluRs with GFAP or MAP2 (Fig. 1M–O) was performed in four FCD specimens with a high number of mGluR-positive balloon cells. Qualitatively, it appeared that GFAP- and MAP2-positive balloon cells are differentially coexpressed with the different mGluR subtypes. The percentage of colocalization of mGluR subtypes with the two cytoskeletal markers (detecting distinct cell populations) was quantified and supported the qualitative observation. mGluR2/3 appeared to be most frequently expressed in GFAP-positive balloon cells and mGluR1α in MAP2-positive cells, whereas mGluR5 was present in the majority of balloon cells (both GFAP- and MAP2-positive cells; Table 4).
Although FCD is a major cause of pediatric medically intractable epilepsy, the putative cellular mechanisms underlying the epileptogenicity of these lesions are still unclear. It has been postulated that chronic epilepsy in these lesions is associated with increased excitability that may result from abnormal glutamatergic transmission, involving ionotropic glutamate receptors (44–46). In the present study, we showed that also mGluR subtypes are expressed in FCD and may thus contribute to the abnormal excitatory transmission within the dysplastic epileptic tissue. The cell-specific distribution in relation to the epileptogenicity of FCD is discussed later.
mGluRs in dysplastic neurons: prominent expression of group I subtypes
Neuronal mGluRs are known to have important roles in synaptic plasticity and in different pathophysiologic processes, including epilepsy (21,47). Recent evidence points in particular to group I mGluRs (mGluR1 and mGluR5), which may actively contribute to seizure discharges and epileptogenesis (20,48,49). In our study we showed that both mGluR1α and mGluR5 are highly represented in neuronal cells within the dysplastic cortex and the subcortical white matter. Although mGluR5 was observed in a higher proportion of neuronal cells, compared with mGluR1α, large dysplastic neurons often display strong IR for both receptor subtypes. We do not have a definitive explanation for the presence of the intense group I mGluR IR in the neuronal population of FCD specimens. It is, however, possible that the increased excitatory synaptic network, with a high number of excitatory neurons in FCD (2), requires high turnover and synthesis of glutamate-receptor proteins, particularly in highly branched hypertrophic excitatory neurons. Previous studies suggest that dysplastic neurons more efficiently synthesize or posttranslationally modify receptor proteins in both cell bodies and in dendritic processes (13,46). Accordingly, dysplastic neurons were labeled with mGluR1 and mGluR5 subtypes in their somatodendritic compartment.
The neuronal expression of mGluR1α and mGluR5 in FCD is in line with our previous observations, showing strong expression of these subtypes in the neuronal component of glioneuronal tumors (gangliogliomas and dysembryoplastic epithelial tumors) associated with intractable epilepsy in children (33). Glioneuronal tumors are characterized by the presence of neurons with dysmorphic features of unclear origin. One possible hypothesis is that they originate from still immature or multipotent dysplastic cells (50–54). Thus the prominent expression of group I mGluRs in the neuronal component of glioneuronal tumors, as in FCD, supports the malformative nature of these lesions and points to the expression of mGluR1 and mGluR5 as common features of epileptogenic developmental lesions, sharing a similar neuronal phenotype. Whether the strong neuronal expression of group I mGluRs in these developmental lesions is constitutive or induced is still unclear. Both glioneuronal tumors and FCD are associated with epilepsy, which has been shown to regulate the expression and function of neuronal mGluRs (55–59). Thus we cannot exclude that chronic seizure activity could contribute to the strong neuronal expression of group I mGluRs. However, no significant differences in the neuronal group I mGluR IR were observed in our study in patients with different durations of epilepsy, as well as in normal cortex adjacent to the dysplastic region compared with control tissue from patients with no history of seizures.
Whether the detection of group I mGluRs in the neuronal epileptogenic component of FCD may affect future therapeutic approaches of patients with FCD is unknown. However, specific antagonists of mGluR subtypes have been recently shown to display an effective AED activity in different experimental models (24).
In contrast to the convulsant action of group I, activation of group II mGluRs decreases epileptiform activity in several experimental models (26,28,60–62). Compared with normal cortex, a lower expression of mGluR2/3 (group II) was observed in neuronal cells within the dysplastic cortex.
Because neuroprotective effects also have been described, the role of neuronal mGluRs is still ambiguous. Evidence supporting a link between group I mGluRs and cell survival has been reported for different neuronal cells (63,64). Thus the possibility that the strong expression of group I mGluR subtypes may be critical with respect to the survival of dysplastic neurons also must be taken in consideration.
mGluR expression in balloon cells
Recently, attention has been focused on giant bizarre balloon cells for their role in the epileptogenicity of FCD (65). In the present study, we report expression of both group I (mGluR1α and mGluR5) and group II mGluRs (mGluR2/3) in large balloon cells in FCD specimens containing this cell type. The question whether these cells are glial or neuronal in nature has been controversial (4,66). Recent studies support the idea that at least some of these balloon cells are of neuronal origin (2). These “neuronal types” of balloon cells often have prominent nucleoli, are positive to different neuronal markers (e.g., MAP2), and show a neurochemical profile similar to that observed in dysplastic neuronal cells (33,43,67). In our series, “glial type” GFAP-positive balloon cells are mainly immunoreactive for mGluR2/3, whereas “neuronal type” balloon cells display prominent mGluR1α labeling, in agreement with the predominant mGluR1α neuronal expression within the dysplastic tissue. The presence of mGluR5 in the majority of balloon cells (both GFAP and MAP2 positive) reflects the known expression of this subtype in both glial and neuronal cells (68,69).
The expression of mGluR subtypes in specific populations of balloon cells suggests that not only dysplastic neurons but also balloon cells may critically contribute to the mGluR-mediated transmission in the dysplastic region and may represent a potential target for the treatment of epilepsy in FCD.
mGluR expression in glial cells
Besides neuronal cells, glial cells also are an important component of FCD. A variable degree of reactive astrocytosis is present within the dysplastic cortex and the subcortical white matter (2,4). Expression of both group I (mGluR5) and group II (mGluR2/3) mGluR in glial cells with the morphology of reactive astrocytes was observed in all the FCD epileptic specimens included in our study. This is in agreement with previous reports showing mGluR2/3 and mGluR5 overexpression in reactive glial cells of epileptic brain (33,36,70,71). Activation of mGluRs can critically regulate glial cell functions, including their interaction with neurons (72) and the control of brain microcirculation (73). The production and release of different growth factors in astrocytes is also under the control of group II mGluRs (74,75).
The physiologic role of group I mGluRs in glia is still not clear. However, it is known that these receptors mediate phosphoinositol hydrolysis and intracellular calcium increase. This can result in the occurrence of calcium waves that are locally initiated and can spread throughout neighboring cells, probably through gap junctions (76). In addition, calcium waves also may induce calcium signals in the neighboring neurons (77,78), contributing to create a highly coupled electrical and biochemical syncytium involved in the generation of a seizure focus. Activation of group I mGluRs in cultures of rat and human astrocytes also has been shown to enhance glial cell proliferation (79) and to regulate glial glutamate transporter protein expression in both mouse, as well as human astrocytes in culture (80,81). Therefore glial mGluRs within the FCD tissue may represent an additional target to regulate the extracellular levels of glutamate, limiting the excitability within these abnormal brain regions.
Together, these observations support the possible involvement of mGluRs in the complex excitatory network of FCD and indicate the need for future investigation of pharmacologic modulation of specific mGluR subtypes (e.g., in slices of human neocortical tissue obtained during surgery for the treatment of seizures associated with FCD).
Acknowledgment: This work was supported by the “Christelijke Vereniging voor de Verpleging van Lijders aan Epilepsie,” the Stichting AZUA-funds (E. Aronica), the National Epilepsy Fund “Power of the Small” and Hersenstichting Nederland (J.A. Gorter; E. Aronica, NEF grant 02-10). We thank W.P. Meun for expert photography and Dr. M.J.B. van de Hoff for his kind assistance with confocal microscopy.
- 31Outcome with respect to epileptic seizures. In: EngelJEJr, ed. Surgical treatment of epilepsy. New York : Raven Press, 1987: 553–71..
- 43Abnormal cytoarchitecture of cortical dysplasia verified by immunohistochemistry. Clin Neuropathol 1998;17: 10–9., , , et al.
- 73Central role of neuron-to-astrocyte signalling in the control of brain microcirculation. Neuropharmacology 2002;43: 180..
- 81Expression and functional role of mGluR3 and mGluR5 in human astrocytes and glioma cells: opposite regulation of glutamate transporter proteins. European Journal of Neuroscience (in press)., , , et al.