Brain, Vol. 123, No. 1, 19-30,
January 2000
© 2000 Oxford University Press
Immunohistochemical characterization of mossy fibre sprouting in the hippocampus of patients with pharmaco-resistant temporal lobe epilepsy
.
1 Rudolf Magnus Institute for Neurosciences, Utrecht University, 2 Departments of Neurosurgery and 3 Pathology, Academic Hospital Utrecht, The Netherlands
Correspondence to:
P. N. E. de Graan, Rudolf Magnus Institute for Neurosciences, Universiteitsweg 100, 3584 CG Utrecht, The Netherlands
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsSynaptophysin, GFAP and B-50...ResultsDiscussionReferences |
|---|
Hippocampal sclerosis (HS) is a common derangement in many patients with temporal lobe epilepsy. As a result of neuronal cell loss in the hilar region of the hippocampus, it is proposed that mossy fibres sprout and re-innervate new regions of the dentate gyrus. This sprouting may cause recurrent excitation that may lead to the generation of seizures. Here, we determined neuronal density, and synaptophysin and glial fibrillary acidic protein (GFAP) immunoreactivity in hippocampal specimens from patients with pharmaco-resistant temporal lobe epilepsy. Patients were classified into two groups: those with severe and those with no HS. Non-epileptic autopsy tissue served as controls. Mossy fibre sprouting was investigated in these two groups of epilepsy patients using Timm's staining and an immunohistochemical staining of the presynaptic growth-associated protein B-50 (also known as GAP-43, neuromodulin, F1). B-50 immunoreactivity in the different sub-areas of the hippocampus was quantified by image analysis. Our results show the following: (i) in both groups of temporal lobe epilepsy patients, there was a significant loss in cell number in all major hippocampal sub-areas compared with autopsy control tissue; (ii) in HS patients, when compared with non-HS patients, there was a further decline in the number of principal cells in all hippocampal sub-areas analysed, which was associated with an increase in GFAP immunoreactivity; (iii) the decline in cell density was accompanied by a reduced number of synaptic terminals; (iv) in the HS group, there were sprouted mossy fibres in the supragranular layer (SGL) of the dentate gyrus; (v) there was an increase in synaptophysin immunostaining in the SGL indicating that functionally active nerve terminals were formed; and (vi) B-50 immunoreactivity was also increased in the SGL in the HS group compared with the non-HS and control groups. These data showed that all temporal lobe epilepsy hippocampi investigated had severe neuronal cell loss which was most dramatic in the HS group, where it was accompanied by a severe loss of synapses. In the HS group, mossy fibre sprouting into the SGL was found. The increase in B-50 immunoreactivity in the SGL indicated that there was still active sprouting. This sprouting was accompanied by an increased density of synapses, indicating that mossy fibre terminals are not only anatomically present, but probably also functional. Thus, functional glutamatergic mossy fibre terminals are in the right position to synapse on to the dendrites of granule cells and thus may contribute to the onset of seizures.
temporal lobe epilepsy; hippocampal sclerosis; B-50/GAP-43; sprouting; mossy fibres
GFAP = glial fibrillary acidic protein; HS = hippocampal sclerosis; OD = optical density; SGL = supragranular layer
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsSynaptophysin, GFAP and B-50...ResultsDiscussionReferences |
|---|
Epilepsy can be defined as a chronic neurological disorder, characterized by recurrent seizures, which are caused by partial (focal) or generalized paroxysmal electric discharges in the brain (for reviews, see Swanson, 1995; Engel, 1996). Most patients with epilepsy respond well to treatment with anti-epileptic drugs, but there still remain a large number of patients that are pharmaco-resistant. In this pharmaco-resistant group, surgical removal of the epileptogenic zone is an alternative treatment option. In temporal lobe epilepsy, this resection involves the removal of the amygdalo-hippocampal complex and a varying part of the anterior temporal neocortex. In a survey of surgical results after anterior temporal lobe resection, ~80% of patients were seizure free or showed a worthwhile improvement (Primrose and Ojemann, 1991
).
The most frequent pathological finding in the resected tissue is hippocampal sclerosis (also known as mesial temporal sclerosis or Ammon's horn sclerosis). Patients with hippocampal sclerosis (HS) display a characteristic neuronal degeneration accompanied by pronounced astrogliosis (Babb and Brown, 1987). Neuronal cell loss is most prevalent in the hilar region, in the CA3, in the beginning of the CA1 pyramidal cells and, to a lesser extent, in the granule cell layer and the CA2 pyramidal cells (Babb and Brown, 1987). There are indications that some neuronal cell loss also occurs in hippocampi of tumour-associated temporal lobe epilepsy patients (Babb and Brown, 1987; Sloviter, 1994), a factor which may co-determine the extent of surgical resection in these patients. To what extent neuronal cell degeneration is accompanied by a decrease in synapse density remains to be determined. The precise pathogenesis of the neuronal loss observed in HS and its relationship to epileptogenesis still remain largely unknown.
One of the leading hypotheses concerning HS is the `sprouting hypothesis', in which it has been proposed that, as a result of hilar cell loss caused by an initial precipitating injury such as febrile convulsions, axons of the surviving dentate granule cells (mossy fibres) form new synapses. These newly sprouted mossy fibres are thought to form a local excitatory feedback circuit on the granule cells and probably contribute to the generation of seizures (Babb et al., 1991). In other words, the sprouting model proposes that HS is the cause, and not the consequence, of seizures. Indeed, studies on human resection tissue show that there is significantly greater mossy fibre punctated staining in the supragranular layer of the dentate gyrus (SGL) of temporal lobe epilepsy patients than in human autopsy controls (Sutula et al., 1989; Mathern et al., 1995b). Furthermore, Mathern and colleagues showed statistically greater mossy fibre puncta densities in the SGL of HS than in non-sclerotic controls (Mathern et al., 1995b). In another study (Isokawa et al., 1993), where dyes were used to trace individual mossy fibre projections, evidence was found of aberrant innervation to the SGL, confirming the results of Mathern and colleagues. Retrospective analysis of clinical histories in temporal lobe epilepsy patients indicated that a previous cerebral injury was often sustained prior to the onset of their habitual complex partial seizures (Mathern et al., 1995a). Other studies, however, indicate that sclerosis and mossy fibre sprouting are acquired as a consequence of repeated seizures (Dam, 1980; Cavazos et al., 1994; Salmenperä et al., 1998).
In the discussion as to whether sclerosis is the cause or the result of the seizures, it is important to know whether there is still an active sprouting process occurring in the epileptic hippocampus. A frequently used marker for neuronal sprouting is the presynaptic protein B-50 (also known as GAP-43, neuromodulin and F1) (for a review, see Oestreicher et al., 1997). It is expressed at high levels during nervous system development, and is then, after the establishment of synaptic contacts, downregulated to basal levels. In the normal adult human brain, relatively high B-50 mRNA expression and B-50 immunoreactivity are present in the hippocampal area (Neve et al., 1988; Mercken et al., 1992; Oestreicher et al., 1997). A dramatic increase in B-50 synthesis is observed in injured and regenerating tissue (Oestreicher et al., 1997).
In the kainic acid model of temporal lobe epilepsy, it has been demonstrated that sprouted mossy fibres show increased B-50 immunoreactivity in the SGL, which is preceded by an upregulation of B-50 mRNA in the granule cells (Bendotti et al., 1994; McNamara et al., 1995). To date, it is not known whether similar changes occur in the immunohistochemical distribution pattern of B-50 in the human epileptic hippocampus.
The first objective of the present study was to determine the degree of hippocampal neuron loss (Nissl staining) and astrogliosis [glial fibrillary acidic protein (GFAP) staining] in three groups: a non-epileptic control group and two groups of temporal lobe epilepsy patients, those with advanced sclerosis and those without signs of sclerosis. Furthermore, we investigated whether neuronal degeneration is accompanied by a reduction in synaptic density by means of immunohistochemical staining of synaptophysin, which is a good molecular marker for synaptic density (Masliah et al., 1990; Eastwood et al., 1994).
The second objective of this study was to determine the pattern of mossy fibre sprouting in the epileptic hippocampus. The localization of mossy fibres, which contain high levels of zinc, was visualized by means of Timm's staining (Danscher, 1981). To determine whether mossy fibre sprouting is an active process, we measured B-50 immunoreactivity in the hippocampus of all these patient groups.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsSynaptophysin, GFAP and B-50...ResultsDiscussionReferences |
|---|
Patients and tissues
We examined hippocampal specimens of patients with pharmaco-resistant chronic epilepsy who had undergone resection of the amygdalo-hippocampal complex and varying resection of the anterior part of the temporal lobe. In all cases, informed consent was obtained for use of any data for research studies. Surgical procedures were performed under general or local anaesthesia. The excision was based on clinical evaluations, interictal and ictal EEG studies (video EEG monitoring), MRI and intraoperative electrocorticography. After en bloc resection of the hippocampus, the tissue was cut into slices (0.5 cm thickness) perpendicular to the longitudinal hippocampal axis and immediately immersion-fixed in 4% paraformaldehyde in 0.1 M phosphate buffer for 24–48 h. Sections of post-mortem human hippocampal tissue from six patients without neurological disorders were used as controls. In all autopsy control tissue, post-mortem delay was <24 h. After fixation, several dehydration steps were carried out and the tissue was embedded in paraffin. Serial sections of 7 µm were cut and immediately mounted on slides, coated with 3-aminopropyltriethoxysilane.
In this study, two groups of temporal lobe epilepsy patients were selected: a non-HS group (no signs of hippocampal sclerosis) and an HS group (severe hippocampal sclerosis, Wyler grade 4). This diagnosis was made by the neuropathologist, who used the grading system according to Wyler (Wyler et al., 1992). Relevant clinical data of all patients used in this study are summarized in Table 1. The average age of temporal lobe epilepsy patients undergoing surgery was 27.6 ± 2.8 years, with an average time of seizure activity within this group of 11.2 ± 2.1 years. As a non-epileptic control group, post-mortem hippocampal tissue without any aberrations was used. In this group, the average age was 30.5 ± 2.2 years.
|
Neuron counting procedures
The distribution of neurons in the hippocampal formation was visualized with Nissl staining. Neuron density measurements were performed on two Nissl-stained sections per patient using an image-based analysis system. Neuron densities were calculated in the dentate gyrus (CA4, polymorphic layer and granular cell layers) and in the CA3, CA2 and the beginning of CA1 pyramidal cell layers. Each delineated area was divided by the computer into small, randomly selected squares, and images were digitized and stored. Using a x40 objective on the microscope, the number of nucleoli-containing profiles was counted within each square. Thus, we used a variation on the dissector method using nucleoli counting. This method has been described and validated for determining neuron densities in the human brain (e.g. Kremer et al., 1990, 1991).
In order to determine pathology-related tissue shrinkage, we calculated hippocampal volume by analysing MRI scans of the epileptic patients. The protocol included coronal images through the temporal lobes with a section thickness of 5 mm using a T2-weighted or FLAIR (fluid-attenuated inversion recovery) sequence on a Philips Easy Vision CT/MR system. This yielded about seven coronal hippocampal sections, which were delineated manually. Hippocampal volume was calculated from both the epileptogenic and non-epileptogenic side from each patient. Due to the fact that not all MRI scans were made through the whole anterior–posterior axis of the hippocampus, we could only perform volumetric analysis on six non-HS and nine HS patients. In the HS group, a 50% reduction in volume in the epileptogenic (1646 mm3) compared with the non-epileptogenic (3326 mm3 hippocampus) side was found, whereas in the non-HS group no differences were found between the epileptogenic (3458 mm3) and non-epileptogenic side (3331 mm3).
In addition, hippocampal area measurements using image-based analysis equipment were performed on the Nissl-stained hippocampal sections in all three patient groups. We found a small (~20%) statistically non-significant decline in hippocampal area in the HS compared with the non-HS group, which closely matched the 50% decrease in hippocampal volume. The results of the MRI and area measurements thus show pathology-related hippocampal shrinkage, which should be taken into account when analysing the cell density in these patient groups.
|
Synaptophysin, GFAP and B-50 immunohistochemistry |
|---|
|
TopAbstractIntroductionMaterial and methodsSynaptophysin, GFAP and B-50...ResultsDiscussionReferences |
|---|
Paraffin sections were processed for immunohistochemistry of synaptophysin, GFAP and B-50 according to standard procedures. In each experiment, autopsy control and epileptic tissues were stained simultaneously. In brief, sections were incubated overnight at room temperature with either the anti-synaptophysin (p38) antibody (mouse monoclonal antibody, Boehringer, diluted 1 : 20), anti-GFAP antibody (rabbit polyclonal antibody, Dako, diluted 1 : 800) or anti-B-50 antibody (mouse monoclonal antibody NM6, characterized by Mercken et al., 1992), diluted 1 : 150 000. Control sections were incubated in the same solution without the primary antibody. Sections were stained using the correct avidin–biotin complex method, and with 3,3'-diaminobenzidine used as chromogen. The negative control sections were devoid of staining.
Densitometric analysis of synaptophysin and B-50 immunoreactivity
For both antibodies, three sections per patient were stained and quantified. Optical density (OD) measurements were obtained using an image-based analysis system. Sections were viewed with a x10 objective and 1.25 intermediate lens. This yielded a spatial resolution of 1 µm = 1.6 pixels. In each delineated area, the average pixel density was measured. In each section, the OD was measured in the dentate gyrus (CA4, polymorphic layer and supragranular region) and the stratum radiatum and stratum oriens of CA3, CA2 and CA1. The different sub-structures in the hippocampus were defined according to the nomenclature of Amaral and Insausti (Amaral and Insausti, 1990).
Timm's staining
From the total group of temporal lobe epilepsy patients, 12 hippocampal specimens (seven with severe sclerosis and five without any signs of sclerosis) were collected in the operating theatre and used for Timm's staining (Danscher, 1981) (Table 1
). The resected tissue was immersed immediately in 0.4% sodium sulphide solution in 0.1 M phosphate buffer for 30 min and then fixed overnight in 0.1 M phosphate buffer containing 1% paraformaldehyde, 1.25% glutaraldehyde and 20% sucrose. After rapidly freezing, sections of 20 µm thickness were cut and immediately mounted on slides (SuperFrost plus, Menzel-gläser). Development of the sections was carried out in a 12 : 6 : 2 mixture of gum arabic, hydroquinone and citric acid–sodium citrate buffer, pH 4, with 0.5 ml of 34% silver nitrate solution. After the dehydration steps, a drop of malinol was placed on the sections, a coverslip was added and the sections were then dried at room temperature.
Statistical analysis
Regional cell counts and OD values of the B-50 and synaptophysin staining in the HS and non-HS patient groups and in the control group were tested using a one-way ANOVA (analysis of variance), together with a pairwise comparison (Student–Newman–Keuls). Data were considered significantly different at a minimum confidence level of P < 0.05. For certain patients, it was not possible to obtain OD measurements or cell density measurements for a number of sub-regions, typically CA3 and CA2, due to damage during the resection procedure.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsSynaptophysin, GFAP and B-50...ResultsDiscussionReferences |
|---|
Hippocampal neuron loss and astrogliosis
Neuron density in the hippocampus was visualized using Nissl staining. Figure 1A
shows a Nissl staining from a control hippocampus and a non-sclerotic (Fig. 1B) and sclerotic (Fig. 1C) epileptic hippocampus. No systematic differences could be observed between the control group and the non-HS group. Severe neuronal cell loss in all pyramidal cell layers was observed in the sclerotic when compared with the non-sclerotic epileptic hippocampus. The granule cells in the dentate gyrus and the pyramidal cells of the CA2 layer also show neuronal cell loss, but were relatively preserved compared with other hippocampal sub-layers. Many of the patients with severe sclerosis also revealed alterations in organization of the granule cells of the dentate gyrus. This so-called `granule cell dispersion' was observed in nine out of 12 sclerotic hippocampi. In the control and non-HS group, granule cells are densely packed. In most of our sclerotic specimens, however, the granule cells were more dispersed, appearing in the molecular layer and creating an irregular outer border and a widened cell layer (Fig. 1C). Calculated neuron densities in the different hippocampal sub-fields are displayed in Fig. 2A and B. An important finding is that the non-HS group, when compared with the control group, shows considerable, statistically significant, cell loss in almost all hippocampal sub-regions (average cell loss: 34.7 ± 6.84%). Another finding was the statistically significant decrease in neuronal density in all hippocampal sub-areas in the HS group versus the non-HS group, with an average cell loss of 68.4 ± 8.07%. Compared with the control group, the average cell loss in the HS group is 77.5 ± 7.5%.
|
|
Since granule cell dispersion is widely observed in the sclerotic hippocampus, two counting methods were used. First, we measured the average cell density of the granule cell layer (Fig. 2B
). Secondly, a more detailed quantification method was used, which accounts for the width of the granule cell layer. Combining these two methods, a correction factor was calculated for the increased granule cell layer area in the HS group (as compared with the non-HS group) and thus the total number of cells in the granule cell layer could be estimated (see Fig. 2C). In the HS group, both granule cell layers were increased in width compared with the non-HS group (correction factor: 1.82 and 1.75 for the lower and upper layers of the dentate gyrus, respectively). Therefore, the large decrease in granule cell density in the HS versus the non-HS group is diminished by granule cell dispersion, but the loss in granule cells is still highly significant.
In order to measure astrogliosis in the sclerotic hippocampus, we determined the distribution of astrocytic marker GFAP (Fig. 1D–F). Within sections of the control and non-HS group (Fig. 1D and E), high immunoreactivity was observed in the hilus. This area stained for numerous astrocytic cell bodies, with some processes branching in the granular cell layer. While the stratum oriens and the stratum radiatum of all CA areas were almost devoid of any immunoreactivity, weak GFAP-immunoreactive processes were seen in the stratum lacunosum moleculare. In the HS group, a more diffuse staining pattern was found than in the non-HS group. Increased diffuse immunoreactivity was observed in the hilus, which indicates that there is a greater number of astrocytic processes. In the inner molecular region of the dentate gyrus, a high density of GFAP-immunoreactive cell bodies was observed. The stratum oriens and stratum radiatum of all CA areas displayed the same diffuse staining as seen in the hilus, and this diffuse staining suddenly ended at the beginning of the subiculum.
Changes in synapse density
To measure changes in synaptic connections, we performed an immunohistochemical synaptophysin staining (Fig. 3A–C). This presynaptic marker showed dense staining in the neuropil layers, whereas no immunoreactivity was found in the dendrites and cell bodies. Overall, less synaptophysin immunoreactivity was found in the HS group than in the non-HS group. In the first group, there was light staining in the hilus of the dentate gyrus, as well as in all CA areas. The immunoreactivity in the CA2 area, which is known to be relatively resistant to damage, was higher than in the other CA areas. The SGL, however, showed a dense granular immunostaining. A striking feature concerning the synaptophysin staining pattern was the distinct border between the vulnerable CA1 area and the relatively resistant subicular region. In the subiculum, the staining pattern resembled the pattern seen in the non-HS group. Quantification of the synaptophysin immunoreactivity from OD measurements (Fig. 4) showed a clear loss of synaptic terminals in almost all hippocampal sub-regions in the HS group when compared with the non-HS group. The CA2 layer and the SGL remained unchanged between the three groups. OD measurements did not reveal any differences between the non-HS group and the control group in any of the sub-areas measured.
|
|
Mossy fibre sprouting: Timm's staining
Figure 5
shows, at low magnification, a typical example of the Timm's staining pattern in a non-sclerotic (Fig. 5A) versus a sclerotic hippocampus (Fig. 5B). In the non-sclerotic hippocampus, this Zn2+ staining specifically reveals the mossy fibres. A dark precipitate was present in the hilar region of the dentate gyrus and in the CA3 area, the two projection areas of the mossy fibres (Fig. 5A). In the non-HS group, the SGL shows background staining and some, limited, punctate staining. In all hippocampal specimens of the HS group, however, fewer mossy fibres were detected in the hilus. Moreover, a dense band of mossy fibres was observed in the SGL (Fig. 5B). At a higher magnification, this difference in staining pattern in the non-sclerotic hippocampus (Fig. 5C) versus the sclerotic hippocampus (Fig. 5D) is clearly demonstrated.
|
B-50 immunoreactivity
B-50 immunostaining showed rather homogeneous labelling with punctate deposits in the neuropil areas (Fig. 6
). The presynaptic marker showed little or no staining in the cytosol of the cell bodies, and dendrites were also immunonegative. At low magnification, the immunoreactivity pattern in the human hippocampal region showed no staining in the principal cell layers, whereas the stratum oriens, the stratum radiatum and the stratum lacunosum moleculare from all CA areas were immunopositive. In most cases, a characteristic dense band of B-50 immunoreactivity was noted in the supragranular layer, the region rich in axon terminals making synapses on the apical dendrites of the granule cells. Visual inspection revealed no obvious differences in the immunoreactivity pattern between the three groups observed (Fig. 6A–C). OD measurements of B-50 immunoreactivity confirmed that there were no differences between the HS, the non-HS and the control group in any of the hippocampal sub-areas (Fig. 7), except in the SGL. In this layer, the HS group showed a significantly higher B-50 immunoreactivity than the non-HS and the control group. In fact, B-50 immunoreactivity in the HS SGL was higher than in almost all other layers.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsSynaptophysin, GFAP and B-50...ResultsDiscussionReferences |
|---|
One of the major questions we addressed in this study is to what extent morphological alterations in the sclerotic hippocampus, such as neuronal cell loss, astrogliosis and mossy fibre sprouting, may contribute to the generation of seizures and not only be a consequence of them. To this end, we investigated these three parameters in a group of non-epileptic control patients, a group of temporal lobe epilepsy patients without sclerosis (non-HS) and a group with sclerosis-associated temporal lobe epilepsy (HS).
An important finding in our study is the amount of neuronal cell loss in the non-HS hippocampus compared with the control hippocampus. Although pathological analysis of non-HS hippocampi revealed a normal hippocampal structure, quantitative analysis showed an average neuronal cell loss of 35% (see also Babb and Brown, 1987; Sloviter, 1994). This finding emphasizes the importance of a non-epileptic control group, and clearly shows that the non-sclerotic hippocampus cannot be considered as normal, and displays so-called `dual pathology'. An important point of consideration in these patients is, not only to remove the temporal cortex, but also to include resection of the hippocampus.
In the sclerotic hippocampus, a further neuronal loss of ~68% was found compared with the non-sclerotic hippocampus, which is consistent with earlier studies (Sagar et al., 1987; Houser, 1992; Wyler et al., 1992; Hand et al., 1997). The loss of neurons was most evident in the CA1, CA3 and CA4 layers, while there was some sparing of the CA2 pyramidal cells and of the granule cells of the dentate gyrus. In view of the observed shrinkage of the sclerotic hippocampus (50% reduction in hippocampal volume in HS compared with the non-HS group), our measurements of neuronal cell loss may be a considerable underestimation. However, precise figures cannot be calculated because hippocampal volume determined by MRI could not be correlated directly with neuronal cell number.
In our study, we also found granule cell dispersion, which previously has been shown by Houser (Houser, 1990). Although the cause of this type of reorganization is unknown, it has been proposed that it may arise as a result of abnormal granule cell migration (Houser, 1992). Another hypothesis is that granule cell dispersion is induced by seizures during early childhood (Lurton et al., 1998). The effect of this aberrant distribution of granule cells on the circuitry of the dentate gyrus remains as yet unknown.
Another important feature of the sclerotic hippocampus is astrogliosis. To investigate the correlation between neuronal cell loss and astrogliosis, we used GFAP as an astrocytic marker. A striking finding was that, despite considerable cell loss in the non-HS group compared with the control group, no signs of astrogliosis were found. In the sclerotic hippocampus, however, many reactive astrocytes and astrocytic processes were present in the hilus, the region where a significant reduction in neuronal density was found. Conversely, in the subiculum, which is known to be relatively resistant, no GFAP-positive astrocytes were observed, indicating that astrogliosis is detectable mainly in severely affected regions in the hippocampus.
To investigate whether cell loss is accompanied by loss of synaptic connections, we studied synaptophysin immunoreactivity, an approach used by Masliah and colleagues (Masliah et al., 1989) in the Alzheimer brain. Synaptophysin, a 38 kDa calcium-binding glycoprotein which is located in the membrane of the small presynaptic vesicles (Wiedenmann and Franke, 1985), is thought to be essential for calcium-dependent transmitter secretion (Alder et al., 1992). In both the HS and the non-HS group, synaptophysin immunoreactivity (synaptic density) corresponded nicely with neuronal density. In the HS group, both parameters were lower than in the non-HS group. The decrease in synaptophysin immunoreactivity (synaptic density) corresponds with the decrease in neuronal density when comparing the HS and non-HS group. However, the percentage of neuronal cell loss seems more dramatic than the loss of synaptic connections (68% versus 40%, respectively). This difference in magnitude is probably due to the different techniques used to visualize neurons and synaptic terminals. For the determination of neuron density, a histological Nissl stain was performed and a computerized counting procedure was used, whereas for the visualization of the synaptic connections an immunohistochemical staining was followed by computerized OD measurements. Therefore, the comparison between these two parameters may be more qualitative than quantitative. The significant decrease in the granular neuropil reaction to synaptophysin in the sclerotic hippocampus is most probably a direct consequence of synaptic loss preceding neuronal death. Also striking is the detection of synaptophysin immunoreactivity in the CA2 area, whereas other areas are almost devoid of synaptophysin-positive terminals, further emphasizing the relative sparing of CA2 neurons and their connections. Even though a 35% loss of neurons was found in the non-HS group compared with controls, we found no difference in synapse density between the two groups using immunohistochemistry. This implies that either this method is not sensitive enough to pick up the loss, or that neuronal cell loss is associated with compensatory sprouting in other hippocampal sub-layers forming new synapses. Alternatively, the neuronal population, which is lost in the non-HS group, contained relatively few synapses.
To demonstrate sprouting of the mossy fibres in the epileptic hippocampus, Timm's technique (Danscher, 1981) was used. Comparing the sclerotic with the non-sclerotic hippocampus, a low density of mossy fibres was found in the hilus, probably as a result of the decreased number of granule cells. Moreover, Timm's granules were located in the SGL, suggesting a projection of mossy fibres into this region. This massive mossy fibre sprouting was not observed either in the control group or in the epileptic non-HS group. This indicates that either a 35% cell loss in the non-HS group was not enough to initiate mossy fibre sprouting or that a morphologically different group of neurons was affected in the non-HS group. It is noteworthy that also the cell loss in the non-HS hippocampus is not associated with astrogliosis. In other words, it seems that in this patient group, distinct morphological changes are found, which seem to be different from those found in hippocampal sclerosis. It seems likely that in the non-HS group a different mechanism underlies the changes. Probably the epileptogenic hippocampus is affected physically or biochemically by the lesion (e.g. tumours or vascular malformations), as has been postulated by Sloviter (Sloviter, 1994).
In the HS group, a high synapse density (as shown by the synaptophysin stain) was found in the SGL, which is in line with the Timm stain results. Taken together, these data indicate that these reorganized mossy fibres contain terminals loaded with synaptic vesicles, strongly indicating functional synapses. The formation of new functional synapses is supported by work of Babb and colleagues, who used quantitative Timm-stained electron microscopy and reported large, zinc-labelled vesicles in terminals with synapses on dendrites in the inner molecular layer and granule cell layers in HS (Babb et al., 1991).
Active sprouting of mossy fibres was visualized by means of immunohistochemistry for B-50. In the control, non-HS and HS groups, the general distribution of this presynaptic protein corresponded well with the findings of earlier studies (Benowitz et al., 1988; Mercken et al., 1992). In the HS group, a small, significant increase in B-50 immunoreactivity levels was found in the SGL compared with the non-HS and control groups. In fact, B-50 immunoreactivity in the SGL in the HS group was higher in this sub-area than in any other hippocampal sub-areas, except CA2. This shows that in the sclerotic hippocampus an active, on-going sprouting process takes place. This conclusion is in line with recent findings (Mikkonen et al., 1998) showing increased NCAM (neural cell adhesion molecule) immunoreactivity in the SGL of drug-resistant temporal lobe epilepsy patients.
To our surprise, no detectable loss in B-50 immunoreactivity was found between the groups in any of the CA areas, despite severe neuronal cell loss. To rule out that, as a result of neuronal damage in the sclerotic hippocampus, astrocytes start to synthesize B-50 (as shown by Yamada et al., 1994), we performed co-localization studies for B-50 and GFAP, and found no B-50/GFAP co-localization in a sclerotic hippocampus (data not shown). Possibly the surviving neurons in the sclerotic hippocampus express high levels of B-50 and in this way compensate for the loss of B-50 in the different hippocampal regions. To investigate such compensatory mechanisms, we currently are investigating B-50 mRNA expression in the epileptic hippocampus.
In summary, our data show that, in the sclerotic epileptic hippocampus, new mossy fibre collaterals are formed, which are present in the SGL, where they are in a position to form synapses on the proximal apical dendrites of granule cells. These sprouted fibres contain vesicles and thus appear to be functionally active. Moreover, this mossy fibre sprouting is an active, on-going process in the sclerotic hippocampus, even though these patients have been suffering from seizures for many years. In fact, the non-HS group, which also has a long history of seizures (and considerable neuronal cell loss), hardly shows any mossy fibre sprouting.
Thus, our data indicate that hippocampal cell loss followed by progressive mossy fibre sprouting might be an important factor contributing to the process of epileptogenesis, and not just a consequence of seizures. How important this progressive sprouting is relative to other factors remains to be determined. Nevertheless, it seems important to remove the sclerotic tissue as early as possible.
|
Acknowledgments |
|---|
We are grateful to the patients who allowed us the opportunity to study their epilepsy by consent for removal and study of surgical specimens, to Chris Pool and Joop van Heerikhuize from The Netherlands Institute for Brain Research for assistance in using the image-based analysis system and to Gerard de Kort from the Department of Radiology, University Hospital Utrecht, for performing volumetric analysis of MRI scans. This study is supported by the Epilepsy Fund of the Netherlands, grant no. 96-04.
|
Notes |
|---|
Deceased, September 29, 1997 
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsSynaptophysin, GFAP and B-50...ResultsDiscussionReferences |
|---|
Alder J, Lu B, Valtorta F, Greengard P, Poo MM. Calcium-dependent transmitter secretion reconstituted in Xenopus oocytes: requirement for synaptophysin. Science 1992; 257: 657–61.
Amaral DG, Insausti R. Hippocampal formation. In: Paxinos G, editor. The human nervous system. San Diego: Academic Press; 1990. p. 711–55.
Babb TL, Brown WJ. Pathological findings in epilepsy. In: Engel J Jr, editor. Surgical treatment of the epilepsies. New York: Raven Press; 1987. p. 511–40.
Babb TL, Kupfer WR, Pretorius JK, Crandall PH, Levesque MF. Synaptic reorganization by mossy fibers in human epileptic fascia dentata. Neuroscience 1991; 42: 351–63.[Web of Science][Medline]
Bendotti C, Pende M, Samanin R. Expression of GAP-43 in the granule cells of rat hippocampus after seizure-induced sprouting of mossy fibres: in situ hybridization and immunocytochemical studies. Eur J Neurosci 1994; 6: 509–15.[Web of Science][Medline]
Benowitz LI, Apostolides PJ, Perrone-Bizzozero N, Finklestein SP, Zwiers H. Anatomical distribution of the growth-associated protein GAP-43/B-50 in the adult rat brain. J Neurosci 1988; 8: 339–52.[Abstract]
Cavazos JE, Das I, Sutula TP. Neuronal loss induced in limbic pathways by kindling: evidence for induction of hippocampal sclerosis by repeated brief seizures. J Neurosci 1994; 14: 3106–21.[Abstract]
Dam AM. Epilepsy and neuron loss in the hippocampus. Epilepsia 1980; 21: 617–29.[Web of Science][Medline]
Danscher G. Histochemical demonstration of heavy metals. A revised version of the sulphide silver method suitable for both light and electron microscopy. Histochemistry 1981; 71: 1–16.[Web of Science][Medline]
Eastwood SL, Burnet PW, McDonald B, Clinton J, Harrison PJ. Synaptophysin gene expression in human brain: a quantitative in situ hybridization and immunocytochemical study. Neuroscience 1994; 59: 881–92.[Web of Science][Medline]
Engel J Jr. Introduction to temporal lobe epilepsy. [Review]. Epilepsy Res 1996; 26: 141–50.[Web of Science][Medline]
Engel J Jr, Van Ness PC, Rasmussen TB, Ojemann LM. Outcome with respect to epileptic seizures. In: Engel J Jr, editor. Surgical treatment of the epilepsies. 2nd edn. New York: Raven Press; 1993. p. 609–21.
Hand KS, Baird VH, Van Paesschen W, Koepp MJ, Revesz T, Thom M, et al. Central benzodiazepine receptor autoradiography in hippocampal sclerosis. Br J Pharmacol 1997; 122: 358–64.[Web of Science][Medline]
Houser CR. Granule cell dispersion in the dentate gyrus of humans with temporal lobe epilepsy. Brain Res 1990; 535: 195–204.[Web of Science][Medline]
Houser CR. Morphological changes in the dentate gyrus in human temporal lobe epilepsy. [Review]. Epilepsy Res 1992; Suppl 7: 223–34.
Isokawa M, Levesque MF, Babb TL, Engel J Jr. Single mossy fiber axonal systems of human dentate granule cells studied in hippocampal slices from patients with temporal lobe epilepsy. J Neurosci 1993; 13: 1511–22.[Abstract]
Kremer HP, Roos RA, Dingjan G, Marani E, Bots GT. Atrophy of the hypothalamic lateral tuberal nucleus in Huntington's disease. J Neuropathol Exp Neurol 1990; 49: 371–82.[Web of Science][Medline]
Kremer B, Swaab D, Bots G, Fisser B, David R, Roos R. The hypothalamic lateral tuberal nucleus in Alzheimer's disease. Ann Neurol 1991; 29: 279–84.[Web of Science][Medline]
Lurton D, El Bahh B, Sundstrom L, Rougier A. Granule cell dispersion is correlated with early epileptic events in human temporal lobe epilepsy. J Neurol Sci 1998; 154: 133–6.[Web of Science][Medline]
Masliah E, Terry RD, DeTeresa RM, Hansen LA. Immunohistochemical quantification of the synapse-related protein synaptophysin in Alzheimer disease. Neurosci Lett 1989; 103: 234–9.[Web of Science][Medline]
Masliah E, Terry RD, Alford M, DeTeresa R. Quantitative immunohistochemistry of synaptophysin in human neocortex: an alternative method to estimate density of presynaptic terminals in paraffin sections. J Histochem Cytochem 1990; 38: 837–44.[Abstract]
Mathern GW, Babb TL, Vickrey BG, Melendez M, Pretorius JK. The clinical–pathogenic mechanisms of hippocampal neuron loss and surgical outcomes in temporal lobe epilepsy. Brain 1995a; 118: 105–18.
Mathern GW, Pretorius JK, Babb TL. Quantified patterns of mossy fiber sprouting and neuron densities in hippocampal and lesional seizures. J Neurosurg 1995b; 82: 211–9.[Web of Science][Medline]
McNamara RK, Routtenberg A. NMDA receptor blockade prevents kainate induction of protein F1/GAP-43 mRNA in hippocampal granule cells and subsequent mossy fiber sprouting in the rat. Brain Res Mol Brain Res 1995; 33: 22–8.[Medline]
Mercken M, Lubke U, Vandermeeren M, Gheuens J, Oestreicher AB. Immunocytochemical detection of the growth-associated protein B-50 by newly characterized monoclonal antibodies in human brain and muscle. J Neurobiol 1992; 23: 309–21.[Web of Science][Medline]
Mikkonen M, Soininen H, Kälväinen R, Tapiola T, Ylinen A, Vapalahti M, et al. Remodeling of neuronal circuitries in human temporal lobe epilepsy: increased expression of highly polysialylated neural cell adhesion molecule in the hippocampus and the entorhinal cortex. Ann Neurol 1998; 44: 923–34.[Web of Science][Medline]
Neve RL, Finch EA, Bird ED, Benowitz LI. Growth-associated protein GAP-43 is expressed selectively in associative regions of the adult human brain. Proc Natl Acad Sci USA 1988; 85: 3638–42.
Oestreicher AB, De Graan PN, Gispen WH, Verhaagen J, Schrama LH. B-50, the growth associated protein-43: modulation of cell morphology and communication in the nervous system. [Review]. Prog Neurobiol 1997; 53: 627–86.[Web of Science][Medline]
Primrose DC, Ojemann GA. Outcome of resective surgery for temporal lobe epilepsy. In: Lüders HO, editor. Epilepsy surgery. New York: Raven Press; 1991. p. 601–11.
Sagar HJ, Oxbury JM. Hippocampal neuron loss in temporal lobe epilepsy: correlation with early childhood convulsions. Ann Neurol 1987; 22: 334–40.[Web of Science][Medline]
Salmenperä T, Kälväinen R, Partanen K, Pitkänen A. Hippocampal damage caused by seizures in temporal lobe epilepsy [letter]. Lancet 1998; 351: 35.[Web of Science][Medline]
Sloviter RS. The functional organization of the hippocampal dentate gyrus and its relevance to the pathogenesis of temporal lobe epilepsy. [Review]. Ann Neurol 1994; 35: 640–54.[Web of Science][Medline]
Sutula T, Cascino G, Cavazos J, Parada I, Ramirez L. Mossy fiber synaptic reorganization in the epileptic human temporal lobe. Ann Neurol 1989; 26: 321–30.[Web of Science][Medline]
Swanson TH. The pathophysiology of human mesial temporal lobe epilepsy. [Review]. J Clin Neurophysiol 1995; 12: 2–22.[Web of Science][Medline]
Wiedenmann B, Franke WW. Identification and localization of synaptophysin, an integral membrane glycoprotein of Mr 38,000 characteristic of presynaptic vesicles. Cell 1985; 41: 1017–28.[Web of Science][Medline]
Wyler AR, Dohan FC, Schweitzer JB, Berry AD. A grading system for mesial temporal pathology (hippocampal sclerosis) from anterior temporal lobectomy. J Epilepsy 1992; 5: 220–5.
Yamada K, Goto S, Oyama T, Inoue N, Nagahiro S, Ushio Y. In vivo induction of the growth associated protein GAP43/B-50 in rat astrocytes following transient middle cerebral artery occlusion. Acta Neuropathol 1994; 88: 553–7.[Medline]
Received April 27, 1999. Revised June 21, 1999. Accepted July 4, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  G. M. Wilczynski, F. A. Konopacki, E. Wilczek, Z. Lasiecka, A. Gorlewicz, P. Michaluk, M. Wawrzyniak, M. Malinowska, P. Okulski, L. R. Kolodziej, et al. Important role of matrix metalloproteinase 9 in epileptogenesis J. Cell Biol., March 5, 2008; 180(5): 1021 - 1035. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. K. Sharma, R. Y. Reams, W. H. Jordan, M. A. Miller, H. L. Thacker, and P. W. Snyder Mesial Temporal Lobe Epilepsy: Pathogenesis, Induced Rodent Models and Lesions Toxicol Pathol, December 1, 2007; 35(7): 984 - 999. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. G. E. Notenboom, D. R. Hampson, G. H. Jansen, P. C. van Rijen, C. W. M. van Veelen, O. van Nieuwenhuizen, and P. N. E. de Graan Up-regulation of hippocampal metabotropic glutamate receptor 5 in temporal lobe epilepsy patients Brain, January 1, 2006; 129(1): 96 - 107. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. S. van der Hel, R. G.E. Notenboom, I. W.M. Bos, P. C. van Rijen, C. W.M. van Veelen, and P. N.E. de Graan Reduced glutamine synthetase in hippocampal areas with neuron loss in temporal lobe epilepsy Neurology, January 25, 2005; 64(2): 326 - 333. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. B. Bausch and J. O. McNamara Contributions of Mossy Fiber and CA1 Pyramidal Cell Sprouting to Dentate Granule Cell Hyperexcitability in Kainic Acid-Treated Hippocampal Slice Cultures J Neurophysiol, December 1, 2004; 92(6): 3582 - 3595. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Duncan Seizure-induced neuronal injury: Human data Neurology, November 12, 2002; 59(90095): S15 - 20. [Abstract] [Full Text]
|
|
|
|
|
|
E. A. Proper, G. Hoogland, S. M. Kappen, G. H. Jansen, M. G. A. Rensen, L. H. Schrama, C. W. M. van Veelen, P. C. van Rijen, O. van Nieuwenhuizen, W. H. Gispen, et al. Distribution of glutamate transporters in the hippocampus of patients with pharmaco-resistant temporal lobe epilepsy Brain, January 1, 2002; 125(1): 32 - 43. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kokaia, K. Holmberg, A. Nanobashvili, Z.-Q. D. Xu, Z. Kokaia, U. Lendahl, S. Hilke, E. Theodorsson, U. Kahl, T. Bartfai, et al. Suppressed kindling epileptogenesis in mice with ectopic overexpression of galanin PNAS, October 31, 2001; (2001) 231496298. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Kokaia, K. Holmberg, A. Nanobashvili, Z.-Q. D. Xu, Z. Kokaia, U. Lendahl, S. Hilke, E. Theodorsson, U. Kahl, T. Bartfai, et al. Suppressed kindling epileptogenesis in mice with ectopic overexpression of galanin PNAS, November 20, 2001; 98(24): 14006 - 14011. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||