Recessive Symptomatic Focal Epilepsy and Mutant Contactin-Associated Protein-like 2
http://www.100md.com
《新英格兰医药杂志》
SUMMARY
Contactin-associated protein-like 2 (CASPR2) is encoded by CNTNAP2 and clusters voltage-gated potassium channels (Kv1.1) at the nodes of Ranvier. We report a homozygous mutation of CNTNAP2 in Old Order Amish children with cortical dysplasia, focal epilepsy, relative macrocephaly, and diminished deep-tendon reflexes. Intractable focal seizures began in early childhood, after which language regression, hyperactivity, impulsive and aggressive behavior, and mental retardation developed in all children. Resective surgery did not prevent the recurrence of seizures. Temporal-lobe specimens showed evidence of abnormalities of neuronal migration and structure, widespread astrogliosis, and reduced expression of CASPR2.
Most epileptic disorders can be traced to an abnormality of cortical architecture, channel-mediated currents, neuronal growth and differentiation, or cerebral metabolism.1,2 In most cases, however, the underlying biologic complexity of epilepsy precludes the identification of the genetic cause, and 65 to 79 percent of recurrent seizure syndromes remain unexplained.3 Microarray analysis of DNA samples can be a powerful tool for revealing a genetic lesion in well-defined families. We have used this approach in Old Order Amish families, some members of which have a clinical and neuropathological phenotype that we designate as the cortical dysplasia–focal epilepsy (CDFE) syndrome. We identified a genetic variation in the gene encoding CASPR2 in affected patients, a finding that suggests that CASPR2 influences brain development.
Methods
The study was approved by the Western Institutional Review Board of Olympia, Washington, and written informed consent was obtained from all participating parents. Phenotype information is based on clinical data from nine patients between the ages of two and nine years. Clinical investigations were routine. Methods are described briefly here; details are included in the Supplementary Appendix, which is available with the full text of this article at www.nejm.org.
Four affected children and their six parents were used for analysis of single-nucleotide polymorphisms (SNPs) with the use of the GeneChip Human Mapping 10K assay kit (Affymetrix). Genotype data were analyzed with Varia software (Silicon Genetics), which assumes mutation homogeneity and scans for regions that are autozygous (identical by descent) among affected persons. Target gene sequencing was performed as previously described.4
Serial 8-μm sections from paraffin-embedded temporal-lobe specimens were stained with hematoxylin and eosin, Luxol fast blue–cresyl violet, and routine immunoperoxidase methods with antibodies against NeuN, neurofilament, synaptophysin, and glial fibrillary acidic protein. Additional sections were stained with primary antibodies against CASPR2 (Santa Cruz Biotechnology), Kv1.1 alpha subunit, and type II sodium channel (Nav1.2) alpha subunit (both purchased from Upstate Cell Signaling Solutions). Samples were stained in parallel, as slide pairs, with control hippocampal and temporal neocortical tissue obtained from a donor-tissue bank of adult volunteers not known to have neurologic disease (Sun Health Research Institute).
Results
Phenotype of the CDFE Syndrome
Detailed clinical information was available for nine patients with CDFE and is summarized in Table 1. During infancy, all patients had mild gross motor delay and subtle limitations in skills that required imitation, concentration, or motor planning. In general, language comprehension was good before the onset of seizures, and cognitive and social development were age-appropriate in eight of the nine children by the age of 18 months. Patients had no distinguishing physical features, and growth trajectories were normal, although all patients had relatively large heads and diminished or absent deep-tendon reflexes (Table 1).
Table 1. Clinical Features and Seizure Activity of Nine Patients with the Cortical Dysplasia–Focal Epilepsy Syndrome.
Seizures exhibiting simple, partial, or complex partial semiology5 began in early childhood and were frequent and intractable between two and seven years of age. Learning ability and social behavior deteriorated after the onset of seizures. All patients with CDFE who were more than three years of age had language regression, aberrant social interactions, and a restricted behavioral repertoire. Scores on the Griffiths Scales of Mental Development6 for three presurgical patients with CDFE (ages 32, 40, and 73 months) revealed global mental ages of 21, 17, and 13 months, respectively (Table 1 of the Supplementary Appendix). The most common interictal neuropsychiatric disturbances were hyperactivity, inattention, and aggression.
Results of 24-hour video electroencephalography (EEG) available from seven patients showed normal background rhythms, with seizures arising from temporal and occasionally from frontal regions as unilateral high-amplitude spike–slow-wave discharges or focal slowing lasting 20 to 110 seconds. Interictal spikes were observed frequently and always from regions of the electroencephalographic onset of seizures. Transsphenoidal recording revealed that discharges starting within inferior and medial frontotemporal fields could rapidly spread to ipsilateral frontal and parietal regions, cross to a mirror focus, or occur with no surface EEG correlate. Intraoperative electrocorticography of temporal lobes targeted for resection demonstrated multiple seizure-onset zones distributed over the amygdalohippocampal complex, parahippocampal gyrus, and lateral temporal gyri.
Magnetic resonance imaging (MRI) of the brain showed focal malformations in three of seven patients studied. Two patients had unilateral dysplasia of the anterior temporal lobe, and one had a malformation of the left striatum (Figure 2 of the Supplementary Appendix).
Timed-injection single-photon-emission computed tomography7 in four patients showed postictal hypoperfusion of the epileptogenic temporal lobe and adjacent ipsilateral structures (e.g., striatum and frontal cortex).
Long-term developmental outcomes were uniformly poor. Patients who were four years of age or older with CDFE were impulsive, had autistic characteristics, and were fully dependent on others for daily living, with projected adult mental ages ranging from one to three years. Three patients underwent electrocorticography-guided epilepsy surgery8 for disabling complex partial seizures and were temporarily seizure-free, but all had a recurrence of seizures from 6 to 15 months after surgery.
Molecular Genetic Studies of CDFE
A large block of putative autozygosity on chromosome 7q36 spanned 7.1 Mb and comprised 18 contiguous SNPs delimited by rs721124 and rs756438 (Figure 1A and Figure 1B). The linked region contained 83 known or hypothetical genes. The sequencing of the first candidate gene, CENTG3, revealed no pathogenic variants in patients with CDFE. However, two patients were heterozygous for a synonymous SNP in exon 13 (1389TC), excluding all genes distal to CENTG3 and reducing the size of the autozygous interval from 7.1 to 3.8 Mb (Figure 1B).
Figure 1. Molecular Genetics of the Cortical Dysplasia–Focal Epilepsy Syndrome.
In Panel A, location scores were maximized on chromosome 7q36, which provided substantial evidence of identity by descent among all eight affected haplotypes. In Panel B, all four patients were homozygous for the same alleles at 18 SNPs in the linked interval. The region spanned 7.1 Mb and contained 83 known or hypothetical genes. The entire shaded region delimits the putative autozygous region. The darker blue shading shows the boundaries of the region after detection of heterozygosity for a synonymous SNP in exon 13 of CENTG3. Panel C shows the structure of the CNTNAP2 locus, the extent of autozygosity within the locus for the four patients, and the three genotypes with respect to the 3709delG mutation in CNTNAP2 (arrows). In Panel D, the 3709delG mutation in CNTNAP2 is predicted to cause a frameshift, which would result in the misincorporation of 16 amino acids beginning at position 1237. Premature termination of translation occurs at codon 1253. This lesion is predicted to result in the loss of the transmembrane and intracellular domains of CASPR2.
A second candidate gene (CNTNAP2) encodes CASPR2, a transmembrane scaffolding protein involved in the clustering of Kv1.1 at the nodes of Ranvier.9 CNTNAP2 straddled the proximal boundary of the autozygous region in such a way that its 5' portion (exons 1 to 8) fell outside the linked region, with exons 9 to 24 encompassing approximately 30 percent of the remaining 3.8-Mb autozygous interval. Sequence analysis of CNTNAP2 exons 9 to 24 in patients with CDFE revealed a single-base deletion at nucleotide 3709 (coding sequence 3709delG) in exon 22 (Figure 1C). All patients were homozygous for the mutation, and their parents were heterozygous. The frameshift mutation results in a premature stop codon and is predicted to yield a nonfunctional protein, owing to a lack of transmembrane and cytoplasmic domains (Figure 1D).
Genotype analysis of 105 healthy Old Order Amish controls revealed none who were homozygous for 3709delG but identified four carriers. We then sequenced CNTNAP2 in 18 additional Old Order Amish patients with complex partial seizures. This analysis identified nine additional patients who were homozygous for 3709delG from seven sibships who had the characteristic clinical features of the CDFE syndrome.
Neuropathology of CDFE
Three patients had surgery in an attempt to control their seizures. Two of these patients had temporal lobe abnormalities (Figure 2A) that were visible on MRI with a 1.5-tesla system, and the other patient did not. Nevertheless, resected brain samples from all three patients showed similar histologic abnormalities diffusely distributed throughout the tissues studied.
Figure 2. Neuropathological Features on MRI and in Histologic Specimens from Three Patients.
In Panel A, coronal T2-weighted MRI of a three-year-old patient with the CDFE syndrome shows an area of increased tissue volume and elevated T2 signal in the right anterior medial temporal lobe, as well as thickening of the lateral temporal cortical gray matter, with blurring of the junction between the gray matter and white matter. Panels B through H show histologic changes in resected temporal lobes (delineated by the white border in Panel A). In Panel B, neurons are arranged in clusters or radial columns, with the dashed line showing a common radial migration pathway in deep layers of the temporal neocortex (NeuN stain). Panel C shows rare cortical neurons that are very large, with abnormally oriented processes and dense neurofilament staining. In Panel D, cortical neurons are rounded and some are binucleate (arrow). Multiple small spherical bodies, perhaps glial nuclei, are shown adjacent to neuronal-cell membranes (arrowheads). In Panel E, staining with Luxol fast blue–cresyl violet shows numerous ectopic neurons (arrows) in subcortical white matter. In Panel F, a large reactive astrocyte extends processes to nearby clusters of hippocampal CA4-sector neurons (arrows) (glial fibrillary acidic protein stain). In Panel G, the hippocampal dentate granule-cell layer has increased cellularity and dispersed margins (NeuN stain). In Panel H, GFAP staining of the section shown in Panel G reveals a dense field of reactive astrocytes abutting and scattered within the granule-cell layer (arrows point to identical anatomical regions in Panel G and Panel H).
Gross inspection of resected anterior and mesial temporal cortex revealed irregular areas of cortical thickening and blurring of the junction between gray matter and white matter. There was no microscopic evidence of neoplasm, inflammation, or vascular dysplasia. In multiple neocortical areas, neurons were abnormally organized into tightly packed columns or clusters (Figure 2B). In both the hippocampus and temporal neocortex, neuron density was increased, and many neurons had a rounded, rather than pyramidal, morphology. A few neurons were binucleate, and some were very large, with an abnormal dendritic structure, inappropriate orientation, and neurofilament tangles (Figure 2C and Figure 2D). Most neocortical neurons had multiple discrete spherical bodies of unclear identity adjacent to the plasma membrane; these may have been nuclei of satellite glia (Figure 2D). Numerous ectopic neurons populated the subcortical white matter (Figure 2E).
The hippocampal dentate granule-cell layer was hyperplastic, had indistinct margins, and was bordered by reactive astrocytes (Figure 2G and Figure 2H). CA1 and CA2 sectors showed diffuse gliosis but no definite pyramidal neuron loss. In the amygdala, there were eccentric microscopic islands of partially matured neuronal precursors in tight clusters surrounded by astrogliosis.
Astrocyte density was increased throughout the temporal cortex, amygdala, and hippocampus. In most brain sections, hypertrophied astrocytes stained darkly for glial fibrillary acidic protein and had abundant cytoplasm, eccentric nuclei, and occasional nuclear duplication. However, numerous cell processes distinguished these cells from classic balloon cells. Reactive astrocytes had a close physical relationship to dysplastic and ectopic neurons; long astrocyte extensions rich in glial fibrillary acidic protein enveloped nerve-cell bodies and made contact with neuronal processes (Figure 2F). Similar astrocyte processes extended to cerebral vessels and the pial layer.
Immunoperoxidase staining for CASPR2 was reduced in brain sections from patients with CDFE (Figure 3 of the Supplementary Appendix). In control hippocampus, there was anti-Kv1.1 staining in efferent projection fibers from the CA region to the perirhinal–entorhinal cortex and the terminal axons within the CA-sector neuropil. In CDFE specimens, there was intense anti-Kv1.1 stain in the cell body of some (but not all) CA sector neurons, and Kv1.1 expression on axons was sparse, suggesting abnormal localization of the protein. In control hippocampus and temporal neocortex, staining for anti-Nav1.2 was evident in cell bodies of large pyramidal neurons and neurons of the dentate granule-cell layer. In hippocampal specimens from patients with CDFE, anti-Nav1.2 staining was completely absent within the hyperplastic dentate granule-cell layer and diminished and irregular in the CA sector.
Discussion
In addition to its scaffolding role at the nodes of Ranvier,9 CASPR2 also appears to be involved in human cortical histogenesis and may mediate intercellular interactions during latter phases of neuroblast migration and laminar organization (Figure 2B and Figure 2E). In mice, targeted disruption of Caspr2 or its chief binding partner, Tag1, eliminates spatial clustering of axonal inwardly rectifying potassium channels but does not result in overt cortical dysplasia or spontaneous seizures.10,11 However, CASPR2 may have different biologic functions in humans and mice, and the precise nature of a genetic lesion (in this case, a null mutation in mice and a nonsense mutation in humans) may be an important determinant of its biologic effects.12,13
In contrast to the discrete lesions often seen in focal epileptic conditions such as tuberous sclerosis,14 tissue abnormalities in brain specimens from patients with CDFE were diffusely distributed throughout the hippocampus, amygdala, neocortex, and subcortex, indicating that cerebral abnormalities of the syndrome could be widespread (Figure 2 of the Supplementary Appendix). Addressing this question directly will require a combination of more sensitive imaging methods and postmortem studies. Also, the investigation of patients over a narrow age range limited our ability to determine the long-term neuropathological consequences of the CNTNAP2 3709delG mutation. At least during childhood, none of the patients had evidence of a neurodegenerative process. Neither generalized cerebral atrophy nor selective neuronal loss was seen in temporal-lobe tissue from the eldest patient studied (who was eight years old at the time of surgery).
Seizures showed a developmental pattern in all the patients with CDFE we studied; all seizure activity began after the age of 14 months and tended to abate spontaneously several years after onset. There was a close temporal association between the onset of repetitive seizures and cognitive and behavioral deterioration, suggesting that seizures contributed directly to mental regression. In support of this hypothesis, we found signs of cellular remodeling in the dentate granule-cell layer and astrocyte compartment (Figure 2F, 2G, and 2H). Such changes may become epileptogenic15,16,17 and also interfere with the normal development and function of limbic networks mediating cognitive and emotional behavior.18,19
An intriguing finding of this study was the altered expression of Kv1.1 and Nav1.2 channels in resected brain samples (Figure 3 of the Supplementary Appendix). Only a few surgical specimens were available for immunostaining, so these results should be interpreted with caution. However, our provisional observations suggest that disturbances of both cortical architecture and ion-channel expression arose from the CNTNAP2 mutation. Kv1.1 (encoded by KCNA1) interacts with CASPR2 and is expressed at high levels in the hippocampus, amygdala, and basal ganglia.20 In humans, KCNA1 mutations can cause childhood-onset temporal-lobe epilepsy,21 and mice with engineered truncating mutations in Kcna1 have large brains, neuronal hypertrophy, astrocytosis, and limbic-system seizures beginning after birth.2,22,23 Intractable partial epilepsy and severe mental regression began at the age of two years in the single reported patient with a recessive truncating mutation of SCN2A.12
It remains to be seen whether mutations affecting CASPR2 or its associated proteins will be identified as causes of symptomatic childhood-onset epilepsy outside of the Old Order Amish community. Lesions that have been described as microdysgenesis or focal cortical dysplasia3,24 are similar to the histologic abnormalities we observed. However, the clinical and etiologic heterogeneity among the many reported patients with symptomatic focal epilepsy precludes meaningful comparisons, and the electroencephalographic, radiologic, and neurodevelopmental features of the CDFE phenotype were nonspecific. Clinical findings that may prompt sequence analysis of CNTNAP2 include seizure onset in early childhood after a period of relatively normal development, mental and behavioral regression temporally linked to the onset of seizures, relative macrocephaly, hypoactive or absent tendon reflexes, and a family history consistent with autosomal recessive inheritance.
No potential conflict of interest relevant to this article was reported.
We are indebted to Jean-Pierre Farmer, M.D., of Montreal Children's Hospital for providing outstanding neurosurgical care; to Steffen Albrecht, M.D., and Peter Crino, M.D., Ph.D., for reviewing the histologic data and making helpful comments on the manuscript; and to the staffs of Lancaster General Hospital and Montreal Children's Hospital for providing outstanding care for patients and families and technical expertise for imaging and electroencephalographic studies.
Source Information
From the Clinic for Special Children, Strasburg, Pa. (K.A.S., E.G.P., D.H.M.); the Translational Genomics Research Institute, Phoenix, Ariz. (M.J.H., J.M.P., D.A.S.); Lancaster General Hospital, Lancaster, Pa. (S.G.); and the Center for Human Genetics, Marshfield Clinic Research Foundation, Marshfield, Wis. (S.E.D.).
Drs. Strauss and Puffenberger contributed equally to this article.
Address reprint requests to Dr. Strauss at the Clinic for Special Children, 535 Bunker Hill Rd., Strasburg, PA 17579, or at kstrauss@clinicforspecialchildren.org or to Dr. Stephan at the Translational Genomics Research Institute, 445 N. 5th St., Phoenix, AZ 85004, or at dstephan@tgen.org.
References
Robinson R, Gardiner M. Molecular basis of Mendelian idiopathic epilepsies. Ann Med 2004;36:89-97.
Noebels JL. The biology of epilepsy genes. Annu Rev Neurosci 2003;26:599-625.
Honovar M, Meldrum BS. Epilepsy. In: Graham DI, Lantos PL, eds. Greenfield's neuropathology. 7th ed. London: Arnold, 2002:899-942.
Puffenberger EG, Hu-Lince D, Parod JM, et al. Mapping of sudden infant death with dysgenesis of the testes syndrome (SIDDT) by a SNP genome scan and identification of TSPYL loss of function. Proc Natl Acad Sci U S A 2004;101:11689-11694.
Fogarasi A, Jokeit H, Faveret E, Janszky J, Tuxhorn I. The effect of age on seizure semiology in childhood temporal lobe epilepsy. Epilepsia 2002;43:638-643.
Luiz DM, Foxcroft CD, Stewart R. The construct validity of the Griffiths Scales of Mental Development. Child Care Health Dev 2001;27:73-83.
Avery RA, Spencer SS, Spanaki MV, Corsi M, Seibyl JP, Zubal IG. Effect of injection time on postictal SPET perfusion changes in medically refractory epilepsy. Eur J Nucl Med 1999;26:830-836.
Wass CT, Grady RE, Fessler AJ, et al. The effects of remifentanil on epileptiform discharges during intraoperative electrocorticography in patients undergoing epilepsy surgery. Epilepsia 2001;42:1340-1344.
Poliak S, Peles E. The local differentiation of myelinated axons at nodes of Ranvier. Nat Rev Neurosci 2003;4:968-980.
Poliak S, Salomon D, Elhanany H, et al. Juxtaparanodal clustering of Shaker-like K+ channels in myelinated axons depends on Caspr2 and TAG-1. J Cell Biol 2003;162:1149-1160.
Traka M, Goutebroze L, Denisenko N, et al. Association of TAG-1 with Caspr2 is essential for the molecular organization of juxtaparanodal regions of myelinated fibers. J Cell Biol 2003;162:1161-1172.
Kamiya K, Kaneda M, Sugawara T, et al. A nonsense mutation of the sodium channel gene SCN2A in a patient with intractable epilepsy and mental decline. J Neurosci 2004;24:2690-2698.
Yamakawa K. Epilepsy and sodium channel gene mutations: gain or loss of function? Neuroreport 2005;16:1-3.
Crino PB. Malformations of cortical development: molecular pathogenesis and experimental strategies. Adv Exp Med Biol 2004;548:175-191.
Morimoto K, Fahnestock M, Racine RJ. Kindling and status epilepticus models of epilepsy: rewiring the brain. Prog Neurobiol 2004;73:1-60.
Tian GF, Azmi H, Takano T, et al. An astrocytic basis of epilepsy. Nat Med 2005;11:973-981.
de Lanerolle NC, Kim JH, Robbins RJ, Spencer DD. Hippocampal interneuron loss and plasticity in human temporal lobe epilepsy. Brain Res 1989;495:387-395.
Holmes GL, Ben-Ari Y. The neurobiology and consequences of epilepsy in the developing brain. Pediatr Res 2001;49:320-325.
Johnston MV. Clinical disorders of brain plasticity. Brain Dev 2004;26:73-80.
Ashcroft FM. Ion channels and disease. San Diego, Calif.: Academic Press, 2000.
Zuberi SM, Eunson LH, Spauschus A, et al. A novel mutation in the human voltage-gated potassium channel gene (Kv1.1) associates with episodic ataxia type 1 and sometimes with partial epilepsy. Brain 1999;122:817-825.
Petersson S, Persson AS, Johansen JE, et al. Truncation of the Shaker-like voltage-gated potassium channel, Kv1.1, causes megencephaly. Eur J Neurosci 2003;18:3231-3240.
Smart SL, Lopantsev V, Zhang CL, et al. Deletion of the K(V)1.1 potassium channel causes epilepsy in mice. Neuron 1998;20:809-819.
Bocti C, Robitaille Y, Diadori P, et al. The pathological basis of temporal lobe epilepsy in childhood. Neurology 2003;60:191-195.(Kevin A. Strauss, M.D., E)
Contactin-associated protein-like 2 (CASPR2) is encoded by CNTNAP2 and clusters voltage-gated potassium channels (Kv1.1) at the nodes of Ranvier. We report a homozygous mutation of CNTNAP2 in Old Order Amish children with cortical dysplasia, focal epilepsy, relative macrocephaly, and diminished deep-tendon reflexes. Intractable focal seizures began in early childhood, after which language regression, hyperactivity, impulsive and aggressive behavior, and mental retardation developed in all children. Resective surgery did not prevent the recurrence of seizures. Temporal-lobe specimens showed evidence of abnormalities of neuronal migration and structure, widespread astrogliosis, and reduced expression of CASPR2.
Most epileptic disorders can be traced to an abnormality of cortical architecture, channel-mediated currents, neuronal growth and differentiation, or cerebral metabolism.1,2 In most cases, however, the underlying biologic complexity of epilepsy precludes the identification of the genetic cause, and 65 to 79 percent of recurrent seizure syndromes remain unexplained.3 Microarray analysis of DNA samples can be a powerful tool for revealing a genetic lesion in well-defined families. We have used this approach in Old Order Amish families, some members of which have a clinical and neuropathological phenotype that we designate as the cortical dysplasia–focal epilepsy (CDFE) syndrome. We identified a genetic variation in the gene encoding CASPR2 in affected patients, a finding that suggests that CASPR2 influences brain development.
Methods
The study was approved by the Western Institutional Review Board of Olympia, Washington, and written informed consent was obtained from all participating parents. Phenotype information is based on clinical data from nine patients between the ages of two and nine years. Clinical investigations were routine. Methods are described briefly here; details are included in the Supplementary Appendix, which is available with the full text of this article at www.nejm.org.
Four affected children and their six parents were used for analysis of single-nucleotide polymorphisms (SNPs) with the use of the GeneChip Human Mapping 10K assay kit (Affymetrix). Genotype data were analyzed with Varia software (Silicon Genetics), which assumes mutation homogeneity and scans for regions that are autozygous (identical by descent) among affected persons. Target gene sequencing was performed as previously described.4
Serial 8-μm sections from paraffin-embedded temporal-lobe specimens were stained with hematoxylin and eosin, Luxol fast blue–cresyl violet, and routine immunoperoxidase methods with antibodies against NeuN, neurofilament, synaptophysin, and glial fibrillary acidic protein. Additional sections were stained with primary antibodies against CASPR2 (Santa Cruz Biotechnology), Kv1.1 alpha subunit, and type II sodium channel (Nav1.2) alpha subunit (both purchased from Upstate Cell Signaling Solutions). Samples were stained in parallel, as slide pairs, with control hippocampal and temporal neocortical tissue obtained from a donor-tissue bank of adult volunteers not known to have neurologic disease (Sun Health Research Institute).
Results
Phenotype of the CDFE Syndrome
Detailed clinical information was available for nine patients with CDFE and is summarized in Table 1. During infancy, all patients had mild gross motor delay and subtle limitations in skills that required imitation, concentration, or motor planning. In general, language comprehension was good before the onset of seizures, and cognitive and social development were age-appropriate in eight of the nine children by the age of 18 months. Patients had no distinguishing physical features, and growth trajectories were normal, although all patients had relatively large heads and diminished or absent deep-tendon reflexes (Table 1).
Table 1. Clinical Features and Seizure Activity of Nine Patients with the Cortical Dysplasia–Focal Epilepsy Syndrome.
Seizures exhibiting simple, partial, or complex partial semiology5 began in early childhood and were frequent and intractable between two and seven years of age. Learning ability and social behavior deteriorated after the onset of seizures. All patients with CDFE who were more than three years of age had language regression, aberrant social interactions, and a restricted behavioral repertoire. Scores on the Griffiths Scales of Mental Development6 for three presurgical patients with CDFE (ages 32, 40, and 73 months) revealed global mental ages of 21, 17, and 13 months, respectively (Table 1 of the Supplementary Appendix). The most common interictal neuropsychiatric disturbances were hyperactivity, inattention, and aggression.
Results of 24-hour video electroencephalography (EEG) available from seven patients showed normal background rhythms, with seizures arising from temporal and occasionally from frontal regions as unilateral high-amplitude spike–slow-wave discharges or focal slowing lasting 20 to 110 seconds. Interictal spikes were observed frequently and always from regions of the electroencephalographic onset of seizures. Transsphenoidal recording revealed that discharges starting within inferior and medial frontotemporal fields could rapidly spread to ipsilateral frontal and parietal regions, cross to a mirror focus, or occur with no surface EEG correlate. Intraoperative electrocorticography of temporal lobes targeted for resection demonstrated multiple seizure-onset zones distributed over the amygdalohippocampal complex, parahippocampal gyrus, and lateral temporal gyri.
Magnetic resonance imaging (MRI) of the brain showed focal malformations in three of seven patients studied. Two patients had unilateral dysplasia of the anterior temporal lobe, and one had a malformation of the left striatum (Figure 2 of the Supplementary Appendix).
Timed-injection single-photon-emission computed tomography7 in four patients showed postictal hypoperfusion of the epileptogenic temporal lobe and adjacent ipsilateral structures (e.g., striatum and frontal cortex).
Long-term developmental outcomes were uniformly poor. Patients who were four years of age or older with CDFE were impulsive, had autistic characteristics, and were fully dependent on others for daily living, with projected adult mental ages ranging from one to three years. Three patients underwent electrocorticography-guided epilepsy surgery8 for disabling complex partial seizures and were temporarily seizure-free, but all had a recurrence of seizures from 6 to 15 months after surgery.
Molecular Genetic Studies of CDFE
A large block of putative autozygosity on chromosome 7q36 spanned 7.1 Mb and comprised 18 contiguous SNPs delimited by rs721124 and rs756438 (Figure 1A and Figure 1B). The linked region contained 83 known or hypothetical genes. The sequencing of the first candidate gene, CENTG3, revealed no pathogenic variants in patients with CDFE. However, two patients were heterozygous for a synonymous SNP in exon 13 (1389TC), excluding all genes distal to CENTG3 and reducing the size of the autozygous interval from 7.1 to 3.8 Mb (Figure 1B).
Figure 1. Molecular Genetics of the Cortical Dysplasia–Focal Epilepsy Syndrome.
In Panel A, location scores were maximized on chromosome 7q36, which provided substantial evidence of identity by descent among all eight affected haplotypes. In Panel B, all four patients were homozygous for the same alleles at 18 SNPs in the linked interval. The region spanned 7.1 Mb and contained 83 known or hypothetical genes. The entire shaded region delimits the putative autozygous region. The darker blue shading shows the boundaries of the region after detection of heterozygosity for a synonymous SNP in exon 13 of CENTG3. Panel C shows the structure of the CNTNAP2 locus, the extent of autozygosity within the locus for the four patients, and the three genotypes with respect to the 3709delG mutation in CNTNAP2 (arrows). In Panel D, the 3709delG mutation in CNTNAP2 is predicted to cause a frameshift, which would result in the misincorporation of 16 amino acids beginning at position 1237. Premature termination of translation occurs at codon 1253. This lesion is predicted to result in the loss of the transmembrane and intracellular domains of CASPR2.
A second candidate gene (CNTNAP2) encodes CASPR2, a transmembrane scaffolding protein involved in the clustering of Kv1.1 at the nodes of Ranvier.9 CNTNAP2 straddled the proximal boundary of the autozygous region in such a way that its 5' portion (exons 1 to 8) fell outside the linked region, with exons 9 to 24 encompassing approximately 30 percent of the remaining 3.8-Mb autozygous interval. Sequence analysis of CNTNAP2 exons 9 to 24 in patients with CDFE revealed a single-base deletion at nucleotide 3709 (coding sequence 3709delG) in exon 22 (Figure 1C). All patients were homozygous for the mutation, and their parents were heterozygous. The frameshift mutation results in a premature stop codon and is predicted to yield a nonfunctional protein, owing to a lack of transmembrane and cytoplasmic domains (Figure 1D).
Genotype analysis of 105 healthy Old Order Amish controls revealed none who were homozygous for 3709delG but identified four carriers. We then sequenced CNTNAP2 in 18 additional Old Order Amish patients with complex partial seizures. This analysis identified nine additional patients who were homozygous for 3709delG from seven sibships who had the characteristic clinical features of the CDFE syndrome.
Neuropathology of CDFE
Three patients had surgery in an attempt to control their seizures. Two of these patients had temporal lobe abnormalities (Figure 2A) that were visible on MRI with a 1.5-tesla system, and the other patient did not. Nevertheless, resected brain samples from all three patients showed similar histologic abnormalities diffusely distributed throughout the tissues studied.
Figure 2. Neuropathological Features on MRI and in Histologic Specimens from Three Patients.
In Panel A, coronal T2-weighted MRI of a three-year-old patient with the CDFE syndrome shows an area of increased tissue volume and elevated T2 signal in the right anterior medial temporal lobe, as well as thickening of the lateral temporal cortical gray matter, with blurring of the junction between the gray matter and white matter. Panels B through H show histologic changes in resected temporal lobes (delineated by the white border in Panel A). In Panel B, neurons are arranged in clusters or radial columns, with the dashed line showing a common radial migration pathway in deep layers of the temporal neocortex (NeuN stain). Panel C shows rare cortical neurons that are very large, with abnormally oriented processes and dense neurofilament staining. In Panel D, cortical neurons are rounded and some are binucleate (arrow). Multiple small spherical bodies, perhaps glial nuclei, are shown adjacent to neuronal-cell membranes (arrowheads). In Panel E, staining with Luxol fast blue–cresyl violet shows numerous ectopic neurons (arrows) in subcortical white matter. In Panel F, a large reactive astrocyte extends processes to nearby clusters of hippocampal CA4-sector neurons (arrows) (glial fibrillary acidic protein stain). In Panel G, the hippocampal dentate granule-cell layer has increased cellularity and dispersed margins (NeuN stain). In Panel H, GFAP staining of the section shown in Panel G reveals a dense field of reactive astrocytes abutting and scattered within the granule-cell layer (arrows point to identical anatomical regions in Panel G and Panel H).
Gross inspection of resected anterior and mesial temporal cortex revealed irregular areas of cortical thickening and blurring of the junction between gray matter and white matter. There was no microscopic evidence of neoplasm, inflammation, or vascular dysplasia. In multiple neocortical areas, neurons were abnormally organized into tightly packed columns or clusters (Figure 2B). In both the hippocampus and temporal neocortex, neuron density was increased, and many neurons had a rounded, rather than pyramidal, morphology. A few neurons were binucleate, and some were very large, with an abnormal dendritic structure, inappropriate orientation, and neurofilament tangles (Figure 2C and Figure 2D). Most neocortical neurons had multiple discrete spherical bodies of unclear identity adjacent to the plasma membrane; these may have been nuclei of satellite glia (Figure 2D). Numerous ectopic neurons populated the subcortical white matter (Figure 2E).
The hippocampal dentate granule-cell layer was hyperplastic, had indistinct margins, and was bordered by reactive astrocytes (Figure 2G and Figure 2H). CA1 and CA2 sectors showed diffuse gliosis but no definite pyramidal neuron loss. In the amygdala, there were eccentric microscopic islands of partially matured neuronal precursors in tight clusters surrounded by astrogliosis.
Astrocyte density was increased throughout the temporal cortex, amygdala, and hippocampus. In most brain sections, hypertrophied astrocytes stained darkly for glial fibrillary acidic protein and had abundant cytoplasm, eccentric nuclei, and occasional nuclear duplication. However, numerous cell processes distinguished these cells from classic balloon cells. Reactive astrocytes had a close physical relationship to dysplastic and ectopic neurons; long astrocyte extensions rich in glial fibrillary acidic protein enveloped nerve-cell bodies and made contact with neuronal processes (Figure 2F). Similar astrocyte processes extended to cerebral vessels and the pial layer.
Immunoperoxidase staining for CASPR2 was reduced in brain sections from patients with CDFE (Figure 3 of the Supplementary Appendix). In control hippocampus, there was anti-Kv1.1 staining in efferent projection fibers from the CA region to the perirhinal–entorhinal cortex and the terminal axons within the CA-sector neuropil. In CDFE specimens, there was intense anti-Kv1.1 stain in the cell body of some (but not all) CA sector neurons, and Kv1.1 expression on axons was sparse, suggesting abnormal localization of the protein. In control hippocampus and temporal neocortex, staining for anti-Nav1.2 was evident in cell bodies of large pyramidal neurons and neurons of the dentate granule-cell layer. In hippocampal specimens from patients with CDFE, anti-Nav1.2 staining was completely absent within the hyperplastic dentate granule-cell layer and diminished and irregular in the CA sector.
Discussion
In addition to its scaffolding role at the nodes of Ranvier,9 CASPR2 also appears to be involved in human cortical histogenesis and may mediate intercellular interactions during latter phases of neuroblast migration and laminar organization (Figure 2B and Figure 2E). In mice, targeted disruption of Caspr2 or its chief binding partner, Tag1, eliminates spatial clustering of axonal inwardly rectifying potassium channels but does not result in overt cortical dysplasia or spontaneous seizures.10,11 However, CASPR2 may have different biologic functions in humans and mice, and the precise nature of a genetic lesion (in this case, a null mutation in mice and a nonsense mutation in humans) may be an important determinant of its biologic effects.12,13
In contrast to the discrete lesions often seen in focal epileptic conditions such as tuberous sclerosis,14 tissue abnormalities in brain specimens from patients with CDFE were diffusely distributed throughout the hippocampus, amygdala, neocortex, and subcortex, indicating that cerebral abnormalities of the syndrome could be widespread (Figure 2 of the Supplementary Appendix). Addressing this question directly will require a combination of more sensitive imaging methods and postmortem studies. Also, the investigation of patients over a narrow age range limited our ability to determine the long-term neuropathological consequences of the CNTNAP2 3709delG mutation. At least during childhood, none of the patients had evidence of a neurodegenerative process. Neither generalized cerebral atrophy nor selective neuronal loss was seen in temporal-lobe tissue from the eldest patient studied (who was eight years old at the time of surgery).
Seizures showed a developmental pattern in all the patients with CDFE we studied; all seizure activity began after the age of 14 months and tended to abate spontaneously several years after onset. There was a close temporal association between the onset of repetitive seizures and cognitive and behavioral deterioration, suggesting that seizures contributed directly to mental regression. In support of this hypothesis, we found signs of cellular remodeling in the dentate granule-cell layer and astrocyte compartment (Figure 2F, 2G, and 2H). Such changes may become epileptogenic15,16,17 and also interfere with the normal development and function of limbic networks mediating cognitive and emotional behavior.18,19
An intriguing finding of this study was the altered expression of Kv1.1 and Nav1.2 channels in resected brain samples (Figure 3 of the Supplementary Appendix). Only a few surgical specimens were available for immunostaining, so these results should be interpreted with caution. However, our provisional observations suggest that disturbances of both cortical architecture and ion-channel expression arose from the CNTNAP2 mutation. Kv1.1 (encoded by KCNA1) interacts with CASPR2 and is expressed at high levels in the hippocampus, amygdala, and basal ganglia.20 In humans, KCNA1 mutations can cause childhood-onset temporal-lobe epilepsy,21 and mice with engineered truncating mutations in Kcna1 have large brains, neuronal hypertrophy, astrocytosis, and limbic-system seizures beginning after birth.2,22,23 Intractable partial epilepsy and severe mental regression began at the age of two years in the single reported patient with a recessive truncating mutation of SCN2A.12
It remains to be seen whether mutations affecting CASPR2 or its associated proteins will be identified as causes of symptomatic childhood-onset epilepsy outside of the Old Order Amish community. Lesions that have been described as microdysgenesis or focal cortical dysplasia3,24 are similar to the histologic abnormalities we observed. However, the clinical and etiologic heterogeneity among the many reported patients with symptomatic focal epilepsy precludes meaningful comparisons, and the electroencephalographic, radiologic, and neurodevelopmental features of the CDFE phenotype were nonspecific. Clinical findings that may prompt sequence analysis of CNTNAP2 include seizure onset in early childhood after a period of relatively normal development, mental and behavioral regression temporally linked to the onset of seizures, relative macrocephaly, hypoactive or absent tendon reflexes, and a family history consistent with autosomal recessive inheritance.
No potential conflict of interest relevant to this article was reported.
We are indebted to Jean-Pierre Farmer, M.D., of Montreal Children's Hospital for providing outstanding neurosurgical care; to Steffen Albrecht, M.D., and Peter Crino, M.D., Ph.D., for reviewing the histologic data and making helpful comments on the manuscript; and to the staffs of Lancaster General Hospital and Montreal Children's Hospital for providing outstanding care for patients and families and technical expertise for imaging and electroencephalographic studies.
Source Information
From the Clinic for Special Children, Strasburg, Pa. (K.A.S., E.G.P., D.H.M.); the Translational Genomics Research Institute, Phoenix, Ariz. (M.J.H., J.M.P., D.A.S.); Lancaster General Hospital, Lancaster, Pa. (S.G.); and the Center for Human Genetics, Marshfield Clinic Research Foundation, Marshfield, Wis. (S.E.D.).
Drs. Strauss and Puffenberger contributed equally to this article.
Address reprint requests to Dr. Strauss at the Clinic for Special Children, 535 Bunker Hill Rd., Strasburg, PA 17579, or at kstrauss@clinicforspecialchildren.org or to Dr. Stephan at the Translational Genomics Research Institute, 445 N. 5th St., Phoenix, AZ 85004, or at dstephan@tgen.org.
References
Robinson R, Gardiner M. Molecular basis of Mendelian idiopathic epilepsies. Ann Med 2004;36:89-97.
Noebels JL. The biology of epilepsy genes. Annu Rev Neurosci 2003;26:599-625.
Honovar M, Meldrum BS. Epilepsy. In: Graham DI, Lantos PL, eds. Greenfield's neuropathology. 7th ed. London: Arnold, 2002:899-942.
Puffenberger EG, Hu-Lince D, Parod JM, et al. Mapping of sudden infant death with dysgenesis of the testes syndrome (SIDDT) by a SNP genome scan and identification of TSPYL loss of function. Proc Natl Acad Sci U S A 2004;101:11689-11694.
Fogarasi A, Jokeit H, Faveret E, Janszky J, Tuxhorn I. The effect of age on seizure semiology in childhood temporal lobe epilepsy. Epilepsia 2002;43:638-643.
Luiz DM, Foxcroft CD, Stewart R. The construct validity of the Griffiths Scales of Mental Development. Child Care Health Dev 2001;27:73-83.
Avery RA, Spencer SS, Spanaki MV, Corsi M, Seibyl JP, Zubal IG. Effect of injection time on postictal SPET perfusion changes in medically refractory epilepsy. Eur J Nucl Med 1999;26:830-836.
Wass CT, Grady RE, Fessler AJ, et al. The effects of remifentanil on epileptiform discharges during intraoperative electrocorticography in patients undergoing epilepsy surgery. Epilepsia 2001;42:1340-1344.
Poliak S, Peles E. The local differentiation of myelinated axons at nodes of Ranvier. Nat Rev Neurosci 2003;4:968-980.
Poliak S, Salomon D, Elhanany H, et al. Juxtaparanodal clustering of Shaker-like K+ channels in myelinated axons depends on Caspr2 and TAG-1. J Cell Biol 2003;162:1149-1160.
Traka M, Goutebroze L, Denisenko N, et al. Association of TAG-1 with Caspr2 is essential for the molecular organization of juxtaparanodal regions of myelinated fibers. J Cell Biol 2003;162:1161-1172.
Kamiya K, Kaneda M, Sugawara T, et al. A nonsense mutation of the sodium channel gene SCN2A in a patient with intractable epilepsy and mental decline. J Neurosci 2004;24:2690-2698.
Yamakawa K. Epilepsy and sodium channel gene mutations: gain or loss of function? Neuroreport 2005;16:1-3.
Crino PB. Malformations of cortical development: molecular pathogenesis and experimental strategies. Adv Exp Med Biol 2004;548:175-191.
Morimoto K, Fahnestock M, Racine RJ. Kindling and status epilepticus models of epilepsy: rewiring the brain. Prog Neurobiol 2004;73:1-60.
Tian GF, Azmi H, Takano T, et al. An astrocytic basis of epilepsy. Nat Med 2005;11:973-981.
de Lanerolle NC, Kim JH, Robbins RJ, Spencer DD. Hippocampal interneuron loss and plasticity in human temporal lobe epilepsy. Brain Res 1989;495:387-395.
Holmes GL, Ben-Ari Y. The neurobiology and consequences of epilepsy in the developing brain. Pediatr Res 2001;49:320-325.
Johnston MV. Clinical disorders of brain plasticity. Brain Dev 2004;26:73-80.
Ashcroft FM. Ion channels and disease. San Diego, Calif.: Academic Press, 2000.
Zuberi SM, Eunson LH, Spauschus A, et al. A novel mutation in the human voltage-gated potassium channel gene (Kv1.1) associates with episodic ataxia type 1 and sometimes with partial epilepsy. Brain 1999;122:817-825.
Petersson S, Persson AS, Johansen JE, et al. Truncation of the Shaker-like voltage-gated potassium channel, Kv1.1, causes megencephaly. Eur J Neurosci 2003;18:3231-3240.
Smart SL, Lopantsev V, Zhang CL, et al. Deletion of the K(V)1.1 potassium channel causes epilepsy in mice. Neuron 1998;20:809-819.
Bocti C, Robitaille Y, Diadori P, et al. The pathological basis of temporal lobe epilepsy in childhood. Neurology 2003;60:191-195.(Kevin A. Strauss, M.D., E)