- Email: [email protected]
Molecular profiling of temporal lobe epilepsy: comparison of data from human tissue samples and animal models
Epilepsy Research 60 (2004) 173–178
Mini-review
Molecular profiling of temporal lobe epilepsy: comparison of data from human tissue samples and animal models Michael Majoresa , J¨urgen Eilsb , Otmar D. Wiestlera,b , Albert J. Beckera,∗ a
Department of Neuropathology, University of Bonn Medical Center, Sigmund-Freud Street 25, D-53105 Bonn, Germany b Deutsches Krebs Forschungszentrum (DKFZ), Heidelberg, Germany Received 1 January 2004; received in revised form 26 May 2004; accepted 1 July 2004 Available online 20 August 2004
Abstract The advent of gene chip technology and the era of functional genomics have initially been accompanied by huge anticipations to quickly unravel the molecular pathogenesis of multifactorial diseases. Expectations have, today, given way to some concerns about this non-hypothesis driven approach. However, the careful and controlled application of expression microarrays in concert with refined bioinformatic tools may provide novel insights in major disorders particularly of highly complex organs such as the central nervous system (CNS). Epilepsies are among the most frequent CNS disorders affecting approximately 1.5% of the population worldwide. In temporal lobe epilepsy (TLE), the seizure origin typically involves the hippocampal formation, a structure located in the mesial temporal lobe. Many TLE patients develop pharmacoresistance, i.e. seizures can no more be controlled by antiepileptic drugs. In order to achieve seizure control, surgical removal of the epileptogenic focus has been established as successful therapeutic strategy. Hippocampal biopsy tissue of pharmacoresistant TLE patients represents an excellent substrate to analyze molecular mechanisms related to structural and cellular reorganization in epilepsy. The complexity of alterations in TLE hippocampi suggests numerous genes and signaling cascades to be involved in the pathogenesis. By microarrays, genome wide expression profiles can be constituted from TLE tissues. However, hippocampi of pharmacoresistant TLE patients represent an advanced stage of the disease. Early stages of epilepsy development are not available for functional genome analysis in humans. Animal models of TLE appear particularly helpful to study molecular mechanisms of highly dynamic processes such as the development of hyperexcitability and pharmacoresistance. In this review, we summarize recent data of gene expression profiles in human and experimental TLE and discuss the relevance of novel tools for bioinformatic analysis and data mining. © 2004 Elsevier B.V. All rights reserved. Keywords: Microarray; Species comparison; Temporal lobe epilepsy
1. Neuropathology of human temporal lobe epilepsy ∗ Corresponding author. Tel.: +49 228 287 1352; fax: +49 228 287 4331. E-mail address: albert [email protected] (A.J. Becker).
0920-1211/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2004.07.002
Hippocampal biopsies obtained during surgery from pharmacoresistant temporal lobe epilepsy (TLE)
174
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
patients reveal two major neuropathological entities (Kral et al., 2002). The most frequent finding in approximately 2/3 of patients is Ammon’s horn sclerosis (AHS) (Fig. 1). Histopathologically, AHS hippocampi exhibit substantial neuronal cell loss in the areas CA1, CA3 and CA4, whereas the CA2 subfield and dentate gyrus granule cells (DG) appear less vulnerable to seizure-induced neurodegeneration (Bl¨umcke et al., 1999). The areas strongly affected by neuronal cell loss exhibit pronounced fibrillary astrogliosis and sclerosis. Further alterations in AHS hippocampi comprise dispersion and focal bilamination of the DG, persistence of Cajal–Retzius like interneurons and mossy fiber sprouting (Bl¨umcke et al., 2002). In approximately 1/3 of pharmacoresistant patients, TLE is associated with focal lesions in the temporal lobe. These lesions include highly differentiated malformations as well as benign glial and glioneuronal neoplasms such as oligodendrogliomas, pilocytic astrocytomas, pleomorphic xanthoastrocytomas (PXA) gangliogliomas, dysembryoplastic neuroepithelial tumors (DNTs) and the recently described long term epilepsy associated tumors (LEAT) (Tassi et al., 2002; Luyken et al., 2003). In contrast to AHS, lesionassociated TLE hippocampi do not exhibit pronounced segmental neuronal cell loss or concomitant astrogliosis/sclerosis. Recurrent axonal sprouting appears substantially attenuated (Bl¨umcke et al., 1999). The availability of human, non-epileptic control hippocampi constitutes a major obstacle for molecular studies of human TLE tissue. Non infiltrated hippocampal biopsy samples after surgery of tumors located in the temporal lobe adjacent to the Ammon’s horn in non-epileptic patients are rare. With respect to key alterations of AHS such as segmental neuronal cell loss, structural and cellular reorganization, lesion-associated hippocampi can be used as controls. Animal models of TLE constitute an important alternative to address this problem. A further advantage of animal TLE models is the opportunity to shed more light on the dynamic pro-
175
cesses of TLE development, a process referred to as epileptogenesis.
2. Experimental temporal lobe epilepsy Currently used TLE animal models can be separated into two major groups: induction of status epilepticus (SE) by application of an excitotoxic compound (kainic acid, pilocarpine) followed by development of a spontanous epilepsy condition and electrically stimulated chronic recurrent seizures referred to as kindling induced TLE (Mello et al., 1993; Ben-Ari and Cossart, 2000; Sutula, 2001). In the kainic acid and pilocarpine models, segmental neuronal cell loss and astrogliosis of hippocampi intriguingly reflect human AHS. Vice versa, neuropathological findings in lesion associated TLE are in line with hippocampi of animals exhibiting chronic seizures after electrical kindling of limbic structures (L¨oscher, 2002). An important feature of experimental TLE is the possibility to examine the dynamic development of recurrent seizures. After SE elicited by pilocarpine administration, animals undergo a silent period of approximately 2 weeks before the onset of recurrent epileptic seizures (Cavalheiro et al., 1991). Novel molecular biological strategies involving functional genomics and proteomics approaches provide the opportunity to study molecular alterations in these well defined models.
3. Expression profiling in temporal lobe epilepsy At present, genomic profiling is technically addressed by basically two different strategies, i.e. cDNA versus oligonucleotide microarrays (Brown and Botstein, 1999; Lockhart and Barlow, 2001; Kuo et al., 2002). For hybridizing mRNA derived from TLE specimens, these are characterized by distinct inherent
Fig. 1. Overview on comparative expression profiling between human and experimental TLE. cRNA of microdissected human and rat TLE hippocampal subfields is subjected to microarray analysis. For hybridization, HG U133 for human specimens and RG U34 microarrays for rat samples are used. Comparison of expression data between these species is carried out using the Netaffx portal, which provides a computerassisted tool for assignment of rat genes to corresponding human genes. The highest number of genes are co-expressed between human AHS and animals in the chronic epileptic stage compared to acute and intermediate stages, i.e. days 3 and 14 after status epilepticus induced by pilocarpine, respectively (Becker et al., 2003). Further information on differentially expressed genes can be found at http://www.meb.unibonn.de/neuropath/neurogenomics.html.
176
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
advantages with respect to the amount of starting material, sequence/signal specificity and reproducibility (for review see: Becker et al., 2002b). Both strategies have been used for studies in epileptic tissue. Pentylenetetrazol (PTZ) induced SE was applied to determine expression changes in response to an excitotoxic episode by oligonucleotide arrays in two different mouse strains (Sandberg et al., 2000). These microarrays were also used to identify transcript alterations in the parietal cortex after SE induced by kainic acid and compare the expression patterns to those induced by ischemic stroke, intracerebral haemorrhage, hypoglycemia, and hypoxia (Tang et al., 2002). This study showed that kainateassociated genomic responses were not distinct from those elicited by the other excitotoxic events. A recent study compared overlapping gene expression alterations in DG granule cells during development and in the epileptic brain (Elliott et al., 2003). Only 37 genes were co-expressed in both states. These could be attributed to cellular key functions such as structural integrity, axonal outgrowth and proliferation. Pitk¨anen and coworkers have analyzed epileptogenesis related expression profiles in hippocampal and temporal lobe specimens after electrically induced TLE (Lukasiuk et al., 2003). Many of the differentially expressed genes were associated with signal transduction, synaptic and axonal plasticity as well as cellular metabolism. Recently, first evidence has been found for specific gene expression signatures in response to pharmacotreatment in experimental TLE (Gu et al., 2003). Such approaches may open new avenues to further understand the molecular background of susceptibility for pharmacotreatment and pharmacoresistance. Our group studied gene expression patterns using mRNA from human AHS versus control/lesion associated hippocampi. For the first generation of cDNA arrays, significant amounts of mRNA from entire hippocampal specimens were used for hybridization. This approach was combined with a second real time PCR expression analysis of single cells to reach cellular resolution for selected expression signals (Becker et al., 2002a). Differentially expressed genes could be attributed to gene transcription control, calcium homeostasis and neuronal signaling. A comparative expression analysis between human and experimental TLE data represents a particular technical challenge for
bioinformatics analysis. Several important aspects have to be taken into consideration.
4. Comparative expression patterns in human and experimental TLE In a recent study we have used expression profiling to compare genomic responses in distinct hippocampal subfields, i.e. CA1 versus DG, in human as well as pilocarpine induced TLE (Becker et al., 2003). The use of microarray technology in both, human specimens and animal models recruits analysis tools for comparison of expression data between these species. We have therefore turned to oligonucleotide microarrays (Affymetrix), since sequences of approximately 8000 rat genes on the U34A microarray can be assigned to corresponding human genes on the U133A microarray with more than 20.000 human sequences using the Netaffx portal (http://www.NetAffx.com) (Fig. 1). It has to be taken into account that data mining procedures, which employ of direct comparative calculations between species-different data sets may bear methodical problems. Within one species, developmentally and regionally different gene expression patterns have to be considered and underline the importance of group homogeneity. A priori differential gene expression due to variation in genomic background heterogeneity is of particular relevance in humans compared to experimental animals and has to be carefully controlled. In addition, signal intensities depend on experimental conditions including the nucleotide sequence of the hybridization probes. Differences between the expression patterns in selected brain structures may be assumed to be similar in different species. These considerations prompted us to create ratios between different subfields from the animal data and to compare them to the ratios of the orthologous genes found in the human TLE samples. The comparative analysis resulted in 18 genes with consistent expression changes between human AHS and animals with pilocarpine induced TLE in the chronic epileptic stage. Interestingly, the numbers of co-expressed genes between human AHS and experimental TLE during the acute stage after SE as well as in the latency stage were significantly lower (Fig. 1) (Becker et al., 2003). The ontology analysis allowed a rapid stratification of interesting genes into distinct functional groups.
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
These include transcripts linked to key cellular functions such as cell-cell/matrix interaction, i.e. stromal cell derived receptor 1 (Langnaese et al., 1997) and neuronal growth and signaling, i.e. synapsin-II (Rosahl et al., 1995). Real time RT-PCR was used to confirm individual genes found to be differentially expressed on the microarrays. Detailed descriptions of differentially expressed genes can be found at http://www.meb.unibonn.de/neuropath/neurogenomics.html.
5. Data mining Recent expression profiling studies in hippocampi of pharmacoresistant TLE patients have yielded sig-
177
nificant variability between individual specimens. It will be an important future challenge to carry out such analyses in homogeneous patient samples standardized for distinct pathogenetic aspects, e.g. type of seizures, onset of the disease as well as resistance to specific modes of pharmacotherapy. Highly structured dynamic expression databases have been developed to cope with the wealth of expression profiling data and allow (a) the integration of expression and clinical data as well as (b) the automatic exchange and comparison of data from different groups. An additional challenge will arise by correlations of RNA expression and proteomics data. To fulfill these demands, an Oracle based database scheme (iCHIP) has been adapted to clinical and molecular data of TLE patients
Fig. 2. iCHIP-scheme for TLE patients and experimental TLE. The database integrates clinical information of TLE patients such as seizure type, duration of the disease and pharmacotreatment with the results of the neuropathological examination of hippocampal biopsy specimens and expression profiling data. It allows a high degree of standardization and dynamic data analysis. Also the comparisons between human and experimental TLE data, which is also integrated in the database, can be automatized.
178
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
as well as experimental TLE (Fig. 2). The underlying database scheme design follows the specifications developed by the microarray gene expression database group (MIAME standards) for electronic data publication (Brazma et al., 2001). Future efforts will have to focus on the functional relevance of the observed expression alterations. Novel molecular biological tools such as small interference RNA technology may allow the development of high throughput strategies to functionally validate gene expression profiles.
References Becker, A.J., Chen, J., Paus, S., Normann, S., Beck, H., Elger, C.E., Wiestler, O.D., Bl¨umcke, I., 2002a. Transcriptional profiling in human epilepsy: expression array and single cell real-time qRTPCR analysis reveal distinct cellular gene regulation. Neuroreport 13, 1327–1333. Becker, A.J., Chen, J., Zien, A., Sochivko, D., Normann, S., Schramm, J., Elger, C.E., Wiestler, O.D., Blumcke, I., 2003. Correlated stage- and subfield-associated hippocampal gene expression patterns in experimental and human temporal lobe epilepsy. Eur. J. Neurosci. 18, 2792–2802. Becker, A.J., Wiestler, O.D., Bl¨umcke, I., 2002b. Functional genomics in experimental and human temporal lobe epilepsy: powerful new tools to identify molecular disease mechanisms of hippocampal damage. Prog. Brain. Res. 135, 161–173. Ben-Ari, Y., Cossart, R., 2000. Kainate, a double agent that generates seizures: two decades of progress. Trends Neurosci. 23, 580– 587. Bl¨umcke, I., Beck, H., Lie, A.A., Wiestler, O.D., 1999. Molecular neuropathology of human mesial temporal lobe epilepsy. Epilepsy Res. 36, 205–223. Bl¨umcke, I., Thom, M., Wiestler, O.D., 2002. Ammon’s horn sclerosis: a maldevelopmental disorder associated with temporal lobe epilepsy. Brain Pathol. 12, 199–211. Brazma, A., Hingamp, P., Quackenbush, J., Sherlock, G., Spellman, P., Stoeckert, C., Aach, J., Ansorge, W., Ball, C.A., Causton, H.C., Gaasterland, T., Glenisson, P., Holstege, F.C., Kim, I.F., Markowitz, V., Matese, J.C., Parkinson, H., Robinson, A., Sarkans, U., Schulze-Kremer, S., Stewart, J., Taylor, R., Vilo, J., Vingron, M., 2001. Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat. Genet. 29, 365–371. Brown, P.O., Botstein, D., 1999. Exploring the new world of the genome with DNA microarrays. Nat. Genet. 21, 33–37. Cavalheiro, E.A., Leite, J.P., Bortolotto, Z.A., Turski, W.A., Ikonomidou, C., Turski, L., 1991. Long-term effects of pilocarpine in rats: structural damage of the brain triggers kindling and spontaneous recurrent seizures. Epilepsia 32, 778–782.
Elliott, R.C., Miles, M.F., Lowenstein, D.H., 2003. Overlapping microarray profiles of dentate gyrus gene expression during development- and epilepsy-associated neurogenesis and axon outgrowth. J. Neurosci. 23, 2218–2227. Gu, J., Lynch, B., Anderson, D., Liang, F., Nocka, K., Fisher, R., Lu, S., Elashoff, M., Ebert, U., Potschka, H., L¨oscher, W., Klitgaard, H., 2003. The antiepileptic drug levetiracetam (keppra) selectively modifies kindling-induced alterations in gene expression in the temporal lobe of rats [abstract]. Epilepsia 44, 55. Kral, T., Clusmann, H., Urbach, J., Schramm, J., Elger, C.E., Kurthen, M., Grunwald, T., 2002. Preoperative evaluation for epilepsy surgery (Bonn Algorithm). Zentralbl. Neurochir. 63, 106–110. Kuo, W.P., Jenssen, T.K., Butte, A.J., Ohno-Machado, L., Kohane, I.S., 2002. Analysis of matched mRNA measurements from two different microarray technologies. Bioinformatics 18, 405–412. Langnaese, K., Beesley, P.W., Gundelfinger, E.D., 1997. Synaptic membrane glycoproteins gp65 and gp55 are new members of the immunoglobulin superfamily. J. Biol. Chem. 272, 821–827. Lockhart, D.J., Barlow, C., 2001. Expressing what’s on your mind: DNA arrays and the brain. Nat. Rev. Neurosci. 2, 63–68. L¨oscher, W., 2002. Animal models of epilepsy for the development of antiepileptogenic and disease-modifying drugs. A comparison of the pharmacology of kindling and post-status epilepticus models of temporal lobe epilepsy. Epilepsy Res. 50, 105–123. Lukasiuk, K., Kontula, L., Pitkanen, A., 2003. cDNA profiling of epileptogenesis in the rat brain. Eur. J. Neurosci. 17, 271–279. Luyken, C., Bl¨umcke, I., Fimmers, R., Urbach, H., Elger, C.E., Wiestler, O.D., Schramm, J., 2003. The spectrum of long-term epilepsy-associated tumors: long-term seizure and tumor outcome and neurosurgical aspects. Epilepsia 44, 822–830. Mello, L.E., Cavalheiro, E.A., Tan, A.M., Kupfer, W.R., Pretorius, J.K., Babb, T.L., Finch, D.M., 1993. Circuit mechanisms of seizures in the pilocarpine model of chronic epilepsy: cell loss and mossy fiber sprouting. Epilepsia 34, 985–995. Rosahl, T.W., Spillane, D., Missler, M., Herz, J., Selig, D.K., Wolff, J.R., Hammer, R.E., Malenka, R.C., Sudhof, T.C., 1995. Essential functions of synapsins I and II in synaptic vesicle regulation. Nature 375, 488–493. Sandberg, R., Yasuda, R., Pankratz, D.G., Carter, T.A., Del Rio, J.A., Wodicka, L., Mayford, M., Lockhart, D.J., Barlow, C., 2000. Regional and strain-specific gene expression mapping in the adult mouse brain. Proc. Natl. Acad. Sci. USA 97, 11038–11043. Sutula, T.P., 2001. Secondary epileptogenesis, kindling, and intractable epilepsy: a reappraisal from the perspective of neural plasticity. Int. Rev. Neurobiol. 45, 355–386. Tang, Y., Lu, A., Aronow, B.J., Wagner, K.R., Sharp, F.R., 2002. Genomic responses of the brain to ischemic stroke, intracerebral haemorrhage, kainate seizures, hypoglycemia, and hypoxia. Eur. J. Neurosci. 15, 1937–1952. Tassi, L., Colombo, N., Garbelli, R., Francione, S., Lo Russo, G., Mai, R., Cardinale, F., Cossu, M., Ferrario, A., Galli, C., Bramerio, M., Citterio, A., Spreafico, R., 2002. Focal cortical dysplasia: neuropathological subtypes, EEG, neuroimaging and surgical outcome. Brain 125, 1719–1732.
Mini-review
Molecular profiling of temporal lobe epilepsy: comparison of data from human tissue samples and animal models Michael Majoresa , J¨urgen Eilsb , Otmar D. Wiestlera,b , Albert J. Beckera,∗ a
Department of Neuropathology, University of Bonn Medical Center, Sigmund-Freud Street 25, D-53105 Bonn, Germany b Deutsches Krebs Forschungszentrum (DKFZ), Heidelberg, Germany Received 1 January 2004; received in revised form 26 May 2004; accepted 1 July 2004 Available online 20 August 2004
Abstract The advent of gene chip technology and the era of functional genomics have initially been accompanied by huge anticipations to quickly unravel the molecular pathogenesis of multifactorial diseases. Expectations have, today, given way to some concerns about this non-hypothesis driven approach. However, the careful and controlled application of expression microarrays in concert with refined bioinformatic tools may provide novel insights in major disorders particularly of highly complex organs such as the central nervous system (CNS). Epilepsies are among the most frequent CNS disorders affecting approximately 1.5% of the population worldwide. In temporal lobe epilepsy (TLE), the seizure origin typically involves the hippocampal formation, a structure located in the mesial temporal lobe. Many TLE patients develop pharmacoresistance, i.e. seizures can no more be controlled by antiepileptic drugs. In order to achieve seizure control, surgical removal of the epileptogenic focus has been established as successful therapeutic strategy. Hippocampal biopsy tissue of pharmacoresistant TLE patients represents an excellent substrate to analyze molecular mechanisms related to structural and cellular reorganization in epilepsy. The complexity of alterations in TLE hippocampi suggests numerous genes and signaling cascades to be involved in the pathogenesis. By microarrays, genome wide expression profiles can be constituted from TLE tissues. However, hippocampi of pharmacoresistant TLE patients represent an advanced stage of the disease. Early stages of epilepsy development are not available for functional genome analysis in humans. Animal models of TLE appear particularly helpful to study molecular mechanisms of highly dynamic processes such as the development of hyperexcitability and pharmacoresistance. In this review, we summarize recent data of gene expression profiles in human and experimental TLE and discuss the relevance of novel tools for bioinformatic analysis and data mining. © 2004 Elsevier B.V. All rights reserved. Keywords: Microarray; Species comparison; Temporal lobe epilepsy
1. Neuropathology of human temporal lobe epilepsy ∗ Corresponding author. Tel.: +49 228 287 1352; fax: +49 228 287 4331. E-mail address: albert [email protected] (A.J. Becker).
0920-1211/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.eplepsyres.2004.07.002
Hippocampal biopsies obtained during surgery from pharmacoresistant temporal lobe epilepsy (TLE)
174
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
patients reveal two major neuropathological entities (Kral et al., 2002). The most frequent finding in approximately 2/3 of patients is Ammon’s horn sclerosis (AHS) (Fig. 1). Histopathologically, AHS hippocampi exhibit substantial neuronal cell loss in the areas CA1, CA3 and CA4, whereas the CA2 subfield and dentate gyrus granule cells (DG) appear less vulnerable to seizure-induced neurodegeneration (Bl¨umcke et al., 1999). The areas strongly affected by neuronal cell loss exhibit pronounced fibrillary astrogliosis and sclerosis. Further alterations in AHS hippocampi comprise dispersion and focal bilamination of the DG, persistence of Cajal–Retzius like interneurons and mossy fiber sprouting (Bl¨umcke et al., 2002). In approximately 1/3 of pharmacoresistant patients, TLE is associated with focal lesions in the temporal lobe. These lesions include highly differentiated malformations as well as benign glial and glioneuronal neoplasms such as oligodendrogliomas, pilocytic astrocytomas, pleomorphic xanthoastrocytomas (PXA) gangliogliomas, dysembryoplastic neuroepithelial tumors (DNTs) and the recently described long term epilepsy associated tumors (LEAT) (Tassi et al., 2002; Luyken et al., 2003). In contrast to AHS, lesionassociated TLE hippocampi do not exhibit pronounced segmental neuronal cell loss or concomitant astrogliosis/sclerosis. Recurrent axonal sprouting appears substantially attenuated (Bl¨umcke et al., 1999). The availability of human, non-epileptic control hippocampi constitutes a major obstacle for molecular studies of human TLE tissue. Non infiltrated hippocampal biopsy samples after surgery of tumors located in the temporal lobe adjacent to the Ammon’s horn in non-epileptic patients are rare. With respect to key alterations of AHS such as segmental neuronal cell loss, structural and cellular reorganization, lesion-associated hippocampi can be used as controls. Animal models of TLE constitute an important alternative to address this problem. A further advantage of animal TLE models is the opportunity to shed more light on the dynamic pro-
175
cesses of TLE development, a process referred to as epileptogenesis.
2. Experimental temporal lobe epilepsy Currently used TLE animal models can be separated into two major groups: induction of status epilepticus (SE) by application of an excitotoxic compound (kainic acid, pilocarpine) followed by development of a spontanous epilepsy condition and electrically stimulated chronic recurrent seizures referred to as kindling induced TLE (Mello et al., 1993; Ben-Ari and Cossart, 2000; Sutula, 2001). In the kainic acid and pilocarpine models, segmental neuronal cell loss and astrogliosis of hippocampi intriguingly reflect human AHS. Vice versa, neuropathological findings in lesion associated TLE are in line with hippocampi of animals exhibiting chronic seizures after electrical kindling of limbic structures (L¨oscher, 2002). An important feature of experimental TLE is the possibility to examine the dynamic development of recurrent seizures. After SE elicited by pilocarpine administration, animals undergo a silent period of approximately 2 weeks before the onset of recurrent epileptic seizures (Cavalheiro et al., 1991). Novel molecular biological strategies involving functional genomics and proteomics approaches provide the opportunity to study molecular alterations in these well defined models.
3. Expression profiling in temporal lobe epilepsy At present, genomic profiling is technically addressed by basically two different strategies, i.e. cDNA versus oligonucleotide microarrays (Brown and Botstein, 1999; Lockhart and Barlow, 2001; Kuo et al., 2002). For hybridizing mRNA derived from TLE specimens, these are characterized by distinct inherent
Fig. 1. Overview on comparative expression profiling between human and experimental TLE. cRNA of microdissected human and rat TLE hippocampal subfields is subjected to microarray analysis. For hybridization, HG U133 for human specimens and RG U34 microarrays for rat samples are used. Comparison of expression data between these species is carried out using the Netaffx portal, which provides a computerassisted tool for assignment of rat genes to corresponding human genes. The highest number of genes are co-expressed between human AHS and animals in the chronic epileptic stage compared to acute and intermediate stages, i.e. days 3 and 14 after status epilepticus induced by pilocarpine, respectively (Becker et al., 2003). Further information on differentially expressed genes can be found at http://www.meb.unibonn.de/neuropath/neurogenomics.html.
176
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
advantages with respect to the amount of starting material, sequence/signal specificity and reproducibility (for review see: Becker et al., 2002b). Both strategies have been used for studies in epileptic tissue. Pentylenetetrazol (PTZ) induced SE was applied to determine expression changes in response to an excitotoxic episode by oligonucleotide arrays in two different mouse strains (Sandberg et al., 2000). These microarrays were also used to identify transcript alterations in the parietal cortex after SE induced by kainic acid and compare the expression patterns to those induced by ischemic stroke, intracerebral haemorrhage, hypoglycemia, and hypoxia (Tang et al., 2002). This study showed that kainateassociated genomic responses were not distinct from those elicited by the other excitotoxic events. A recent study compared overlapping gene expression alterations in DG granule cells during development and in the epileptic brain (Elliott et al., 2003). Only 37 genes were co-expressed in both states. These could be attributed to cellular key functions such as structural integrity, axonal outgrowth and proliferation. Pitk¨anen and coworkers have analyzed epileptogenesis related expression profiles in hippocampal and temporal lobe specimens after electrically induced TLE (Lukasiuk et al., 2003). Many of the differentially expressed genes were associated with signal transduction, synaptic and axonal plasticity as well as cellular metabolism. Recently, first evidence has been found for specific gene expression signatures in response to pharmacotreatment in experimental TLE (Gu et al., 2003). Such approaches may open new avenues to further understand the molecular background of susceptibility for pharmacotreatment and pharmacoresistance. Our group studied gene expression patterns using mRNA from human AHS versus control/lesion associated hippocampi. For the first generation of cDNA arrays, significant amounts of mRNA from entire hippocampal specimens were used for hybridization. This approach was combined with a second real time PCR expression analysis of single cells to reach cellular resolution for selected expression signals (Becker et al., 2002a). Differentially expressed genes could be attributed to gene transcription control, calcium homeostasis and neuronal signaling. A comparative expression analysis between human and experimental TLE data represents a particular technical challenge for
bioinformatics analysis. Several important aspects have to be taken into consideration.
4. Comparative expression patterns in human and experimental TLE In a recent study we have used expression profiling to compare genomic responses in distinct hippocampal subfields, i.e. CA1 versus DG, in human as well as pilocarpine induced TLE (Becker et al., 2003). The use of microarray technology in both, human specimens and animal models recruits analysis tools for comparison of expression data between these species. We have therefore turned to oligonucleotide microarrays (Affymetrix), since sequences of approximately 8000 rat genes on the U34A microarray can be assigned to corresponding human genes on the U133A microarray with more than 20.000 human sequences using the Netaffx portal (http://www.NetAffx.com) (Fig. 1). It has to be taken into account that data mining procedures, which employ of direct comparative calculations between species-different data sets may bear methodical problems. Within one species, developmentally and regionally different gene expression patterns have to be considered and underline the importance of group homogeneity. A priori differential gene expression due to variation in genomic background heterogeneity is of particular relevance in humans compared to experimental animals and has to be carefully controlled. In addition, signal intensities depend on experimental conditions including the nucleotide sequence of the hybridization probes. Differences between the expression patterns in selected brain structures may be assumed to be similar in different species. These considerations prompted us to create ratios between different subfields from the animal data and to compare them to the ratios of the orthologous genes found in the human TLE samples. The comparative analysis resulted in 18 genes with consistent expression changes between human AHS and animals with pilocarpine induced TLE in the chronic epileptic stage. Interestingly, the numbers of co-expressed genes between human AHS and experimental TLE during the acute stage after SE as well as in the latency stage were significantly lower (Fig. 1) (Becker et al., 2003). The ontology analysis allowed a rapid stratification of interesting genes into distinct functional groups.
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
These include transcripts linked to key cellular functions such as cell-cell/matrix interaction, i.e. stromal cell derived receptor 1 (Langnaese et al., 1997) and neuronal growth and signaling, i.e. synapsin-II (Rosahl et al., 1995). Real time RT-PCR was used to confirm individual genes found to be differentially expressed on the microarrays. Detailed descriptions of differentially expressed genes can be found at http://www.meb.unibonn.de/neuropath/neurogenomics.html.
5. Data mining Recent expression profiling studies in hippocampi of pharmacoresistant TLE patients have yielded sig-
177
nificant variability between individual specimens. It will be an important future challenge to carry out such analyses in homogeneous patient samples standardized for distinct pathogenetic aspects, e.g. type of seizures, onset of the disease as well as resistance to specific modes of pharmacotherapy. Highly structured dynamic expression databases have been developed to cope with the wealth of expression profiling data and allow (a) the integration of expression and clinical data as well as (b) the automatic exchange and comparison of data from different groups. An additional challenge will arise by correlations of RNA expression and proteomics data. To fulfill these demands, an Oracle based database scheme (iCHIP) has been adapted to clinical and molecular data of TLE patients
Fig. 2. iCHIP-scheme for TLE patients and experimental TLE. The database integrates clinical information of TLE patients such as seizure type, duration of the disease and pharmacotreatment with the results of the neuropathological examination of hippocampal biopsy specimens and expression profiling data. It allows a high degree of standardization and dynamic data analysis. Also the comparisons between human and experimental TLE data, which is also integrated in the database, can be automatized.
178
M. Majores et al. / Epilepsy Research 60 (2004) 173–178
as well as experimental TLE (Fig. 2). The underlying database scheme design follows the specifications developed by the microarray gene expression database group (MIAME standards) for electronic data publication (Brazma et al., 2001). Future efforts will have to focus on the functional relevance of the observed expression alterations. Novel molecular biological tools such as small interference RNA technology may allow the development of high throughput strategies to functionally validate gene expression profiles.
References Becker, A.J., Chen, J., Paus, S., Normann, S., Beck, H., Elger, C.E., Wiestler, O.D., Bl¨umcke, I., 2002a. Transcriptional profiling in human epilepsy: expression array and single cell real-time qRTPCR analysis reveal distinct cellular gene regulation. Neuroreport 13, 1327–1333. Becker, A.J., Chen, J., Zien, A., Sochivko, D., Normann, S., Schramm, J., Elger, C.E., Wiestler, O.D., Blumcke, I., 2003. Correlated stage- and subfield-associated hippocampal gene expression patterns in experimental and human temporal lobe epilepsy. Eur. J. Neurosci. 18, 2792–2802. Becker, A.J., Wiestler, O.D., Bl¨umcke, I., 2002b. Functional genomics in experimental and human temporal lobe epilepsy: powerful new tools to identify molecular disease mechanisms of hippocampal damage. Prog. Brain. Res. 135, 161–173. Ben-Ari, Y., Cossart, R., 2000. Kainate, a double agent that generates seizures: two decades of progress. Trends Neurosci. 23, 580– 587. Bl¨umcke, I., Beck, H., Lie, A.A., Wiestler, O.D., 1999. Molecular neuropathology of human mesial temporal lobe epilepsy. Epilepsy Res. 36, 205–223. Bl¨umcke, I., Thom, M., Wiestler, O.D., 2002. Ammon’s horn sclerosis: a maldevelopmental disorder associated with temporal lobe epilepsy. Brain Pathol. 12, 199–211. Brazma, A., Hingamp, P., Quackenbush, J., Sherlock, G., Spellman, P., Stoeckert, C., Aach, J., Ansorge, W., Ball, C.A., Causton, H.C., Gaasterland, T., Glenisson, P., Holstege, F.C., Kim, I.F., Markowitz, V., Matese, J.C., Parkinson, H., Robinson, A., Sarkans, U., Schulze-Kremer, S., Stewart, J., Taylor, R., Vilo, J., Vingron, M., 2001. Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nat. Genet. 29, 365–371. Brown, P.O., Botstein, D., 1999. Exploring the new world of the genome with DNA microarrays. Nat. Genet. 21, 33–37. Cavalheiro, E.A., Leite, J.P., Bortolotto, Z.A., Turski, W.A., Ikonomidou, C., Turski, L., 1991. Long-term effects of pilocarpine in rats: structural damage of the brain triggers kindling and spontaneous recurrent seizures. Epilepsia 32, 778–782.
Elliott, R.C., Miles, M.F., Lowenstein, D.H., 2003. Overlapping microarray profiles of dentate gyrus gene expression during development- and epilepsy-associated neurogenesis and axon outgrowth. J. Neurosci. 23, 2218–2227. Gu, J., Lynch, B., Anderson, D., Liang, F., Nocka, K., Fisher, R., Lu, S., Elashoff, M., Ebert, U., Potschka, H., L¨oscher, W., Klitgaard, H., 2003. The antiepileptic drug levetiracetam (keppra) selectively modifies kindling-induced alterations in gene expression in the temporal lobe of rats [abstract]. Epilepsia 44, 55. Kral, T., Clusmann, H., Urbach, J., Schramm, J., Elger, C.E., Kurthen, M., Grunwald, T., 2002. Preoperative evaluation for epilepsy surgery (Bonn Algorithm). Zentralbl. Neurochir. 63, 106–110. Kuo, W.P., Jenssen, T.K., Butte, A.J., Ohno-Machado, L., Kohane, I.S., 2002. Analysis of matched mRNA measurements from two different microarray technologies. Bioinformatics 18, 405–412. Langnaese, K., Beesley, P.W., Gundelfinger, E.D., 1997. Synaptic membrane glycoproteins gp65 and gp55 are new members of the immunoglobulin superfamily. J. Biol. Chem. 272, 821–827. Lockhart, D.J., Barlow, C., 2001. Expressing what’s on your mind: DNA arrays and the brain. Nat. Rev. Neurosci. 2, 63–68. L¨oscher, W., 2002. Animal models of epilepsy for the development of antiepileptogenic and disease-modifying drugs. A comparison of the pharmacology of kindling and post-status epilepticus models of temporal lobe epilepsy. Epilepsy Res. 50, 105–123. Lukasiuk, K., Kontula, L., Pitkanen, A., 2003. cDNA profiling of epileptogenesis in the rat brain. Eur. J. Neurosci. 17, 271–279. Luyken, C., Bl¨umcke, I., Fimmers, R., Urbach, H., Elger, C.E., Wiestler, O.D., Schramm, J., 2003. The spectrum of long-term epilepsy-associated tumors: long-term seizure and tumor outcome and neurosurgical aspects. Epilepsia 44, 822–830. Mello, L.E., Cavalheiro, E.A., Tan, A.M., Kupfer, W.R., Pretorius, J.K., Babb, T.L., Finch, D.M., 1993. Circuit mechanisms of seizures in the pilocarpine model of chronic epilepsy: cell loss and mossy fiber sprouting. Epilepsia 34, 985–995. Rosahl, T.W., Spillane, D., Missler, M., Herz, J., Selig, D.K., Wolff, J.R., Hammer, R.E., Malenka, R.C., Sudhof, T.C., 1995. Essential functions of synapsins I and II in synaptic vesicle regulation. Nature 375, 488–493. Sandberg, R., Yasuda, R., Pankratz, D.G., Carter, T.A., Del Rio, J.A., Wodicka, L., Mayford, M., Lockhart, D.J., Barlow, C., 2000. Regional and strain-specific gene expression mapping in the adult mouse brain. Proc. Natl. Acad. Sci. USA 97, 11038–11043. Sutula, T.P., 2001. Secondary epileptogenesis, kindling, and intractable epilepsy: a reappraisal from the perspective of neural plasticity. Int. Rev. Neurobiol. 45, 355–386. Tang, Y., Lu, A., Aronow, B.J., Wagner, K.R., Sharp, F.R., 2002. Genomic responses of the brain to ischemic stroke, intracerebral haemorrhage, kainate seizures, hypoglycemia, and hypoxia. Eur. J. Neurosci. 15, 1937–1952. Tassi, L., Colombo, N., Garbelli, R., Francione, S., Lo Russo, G., Mai, R., Cardinale, F., Cossu, M., Ferrario, A., Galli, C., Bramerio, M., Citterio, A., Spreafico, R., 2002. Focal cortical dysplasia: neuropathological subtypes, EEG, neuroimaging and surgical outcome. Brain 125, 1719–1732.