Pediatric Neurology 48 (2013) 259e270
Contents lists available at ScienceDirect
Pediatric Neurology journal homepage: www.elsevier.com/locate/pnu
Review Article
Radial Microcolumnar Cortical Architecture: Maturational Arrest or Cortical Dysplasia? Harvey B. Sarnat MS, MD, FRCPC a, b, c, *, Laura Flores-Sarnat MD a, c a
Department of Paediatrics (Neurology), University of Calgary Faculty of Medicine and Alberta Children’s Hospital Research Institute, Calgary, Alberta, Canada Department of Pathology (Neuropathology) and Laboratory Medicine, University of Calgary Faculty of Medicine and Alberta Children’s Hospital Research Institute, Calgary, Alberta, Canada c Department of Clinical Neurosciences, University of Calgary Faculty of Medicine and Alberta Children’s Hospital Research Institute, Calgary, Alberta, Canada b
article information
abstract
Article history: Received 18 July 2012 Accepted 10 October 2012
The fetal neocortical plate, from initiation of radial migration at 5 weeks’ gestation until midgestation, exhibits radial microcolumnar architecture. Horizontal histologic layering or lamination becomes superimposed in the second half of gestation, although residua of the columnar pattern persist postnatally, particularly where the cortex bends: at the crowns of gyri and in the depths of sulci. Columnar architecture of the cortical plate in the first half of gestation mostly results from radial migration of neuroblasts, but the Cajal-Retzius neurons and GABAergic neuroblasts from tangential migration regulate a transition to horizontal lamination of the mature cortex. In children and adults, prominent columnar architecture is a feature of many focal cortical dysplasias and is now recognized as a distinctive pattern of focal cortical dysplasias in the new International League Against Epilepsy classification. It also occurs, however, in many genetic syndromes and chromosomopathic conditions, including 22q12 deletions (DiGeorge syndrome), in several primary cerebral malformations, in the contralateral cingulate gyrus in hemimegalencephaly, in cortical tubers of tuberous sclerosis, in the margins of porencephalic cysts resulting from prenatal infarcts, and in some inborn metabolic defects such as methylmalonic acidemia. Synaptophysin demonstrates both radial and horizontal lamination of synaptic layers. Persistent fetal cortical architecture is potentially epileptogenic. We conclude that columnar architecture is a maturational arrest in histogenesis of the neocortical plate and becomes a component of cortical dysplasia in the perinatal period. An initially physiological process thus becomes pathologic by virtue of advancing age, but traces of it persist in normal mature brains. It also occurs in many genetic and inborn metabolic diseases and after acquired ischemic insults of the fetal brain. Ó 2013 Elsevier Inc. All rights reserved.
Introduction
Malformations of the brain can only be fully understood in the context of neuroembryology and as disorders of normal ontogenesis. Both generalized and focal cerebral cortical dysplasias are well defined morphologically and, in many cases, genetically as well. But, if put in the context of abnormal development, additional insight is revealed
by comparison with normal developmental processes that might become arrested or altered by a genetic defect or lesion acquired in prenatal life. In this review our goal is to demonstrate that focal cortical dysplasia type I and also several genetic malformations and metabolic diseases of the fetal nervous system can be better understood as persistence of an initially normal stage of maturation. Cerebral cortical architecture in normal ontogenesis
* Communications should be addressed to: Dr. H.B. Sarnat; Alberta Children’s Hospital; 2888 Shaganappi Trail NW; Calgary; Alberta T3B 6A8 Canada. E-mail address:
[email protected] 0887-8994/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.pediatrneurol.2012.10.001
The cytoarchitecture of all regions of the brain may be divided into fundamental nuclear and cortical arrangements of neurons. Nuclear structures are exemplified by
260
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
of which already exhibit proteins marking not only their lineage but also the particular types of neurons. A single radial glial fiber can conduct multiple migratory neuroblasts, simultaneously forming the initial migratory microcolumns. A specialized medial portion of this germinal matrix neuroepithelium is called the ganglionic eminence and is the origin of tangentially migratory neuroblasts that will become the 20% of cortical neurons that are inhibitory GABAergic interneurons. These interneurons occur in all layers but are most numerous in receptive layers 2 and 4. This migration of the inhibitory interneurons is initiated before radial migrations begin, and they form the marginal zone of the preplate plexus, including Cajal-Retzius and subsequent molecular zone and subplate neurons from 4.5 weeks’ gestation, at the time of prosencephalic cleavage into paired telencephalic hemispheres [2,5,6]. From 8-16 weeks any type of architecture of the cortical plate is difficult to discern histologically because the immature neurons appear as tightly packed nuclei with a thin rim of cytoplasm and sparse neuropil separating cells because there are few neurites and even fewer glial cells and processes. Nevertheless, a predominantly radial, rather than horizontally laminated architecture can be detected, becoming more apparent from 16 to 22 weeks because of increased cytoplasmic volume as neuroblasts mature and an increased amount of neuropil separates individual cells (Fig 1). Ramón y Cajal [7] was the first to recognize this
print & web 4C=FPO
cranial nerve nuclei, thalamus, and basal ganglia, in which functional synaptic relations of neurons still occur. A cortex, by contrast, is defined by horizontal lamination, each layer composed mainly of neurons of similar type, such as a majority population of glutamatergic excitatory neurons mixed with a minority of GABAergic inhibitory interneurons. Layers are oriented parallel to the brain surface, with each layer separated by neuropil in which horizontal layers of synaptic circuits integrate the neurons of that layer for a functional unit; perpendicular synaptic relations with other layers also occur by radially oriented dendrites and axons or axonal collateral vessels. Such laminated architecture occurs not only in the mammalian cerebral neocortex, but also in the cerebellar cortex, hippocampus, olfactory bulb, superior colliculus, part of the amygdala, lateral geniculate body, and other sites within the central nervous system. The familiar six-layered cerebral cortex of the adult human brain becomes histologically evident only in the second half of gestation, with maturation of the cortical plate and cytomaturation of its individual neurons. Radial neuroblast migration from the subventricular zone of the lateral hemispheres begins at 7-8 weeks’ gestation and is >90% complete by 16 weeks [1-3]. The germinal matrix that gives origin to migratory neuroblasts is not yet properly subependymal because ependyma does not fully line the lateral ventricles until 22 weeks’ gestation [4]. The germinal matrix is composed of postmitotic, premigatory cells, some
Figure 1. Cerebral cortical plate at (A) 9 weeks’ gestation, (B) 13 weeks’ gestation, (C) 18 weeks’ gestation, and (D) 20 weeks’ gestation. Radial columnar architecture predominates but is more difficult to discern in young fetuses because the cytoplasm of the immature neurons is a thin layer surrounding a relatively large, round nucleus, and intercellular neuropil is sparse because of the lack of neurites and few glial cells. At 20 weeks, superimposed horizontal lamination is first detected, coincidental with increased neuronal cytoplasm and intercellular neuropil; it becomes more clearly evident by 22 weeks. This microcolumnar histologic pattern is normal in the cortical plate in the first half of gestation. Although neuronal nuclear antigen (NeuN) better demonstrates pathologic columnar patterns postnatally, its immunoreactivity is not expressed in immature neurons; hence, it cannot be used to show cortical architecture in the early fetus. (E) Full-term (39-week) neonatal temporal neocortex, showing predominantly horizontal lamination of neurons that persists to adult life, although partial microcolumns often still are evident, particularly in the deep layers of cortex. Note the thin layer at the surface of the brain consisting of transitory glial cells of the subpial granular layer of Brun; this layer will complete its disappearance postnatally and identifies this cortex as neonatal and not adult, even though the horizontal lamination is nearly mature. Terms of reference of cortical architecture: radial means perpendicular to the pial surface; horizontal means parallel to the pial surface. Hematoxylin & eosin staining. (A-D) Original magnification 250; (E) original magnification 100.
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
histologic arrangement of the cortical plate, his meticulous observations and insightful interpretations forming a basis of modern neuroembryology and remaining as valid today as a century ago. The arrangement of a column of single neurons is now called microcolumnar to distinguish it from columnar arrangements of narrow clusters of neurons projecting from the ventricular zone to the pial surface, as occurs in some transmantle dysplasias or the aggregates of migratory neuroblasts and glioblasts in perivascular spaces of radially oriented small parenchymal vessels of the fetal cerebrum. After mid-gestation, horizontal lamination appears superimposed on this radial columnar pattern of the maturing cortex, so that the two patterns coexist, with the horizontal pattern eventually predominating. In the normal cerebral cortex of the mature brain, and even more in the neonatal and infant brain, traces of the original radial architecture can still be detected in areas where the cortex is tightly curved: in the margins of the crowns of gyri and in the depths of sulci, being much less in the straight walls of gyri within the sulci and at the tops of broad gyral crowns. This residual radial pattern is more evident in temporal and insular neocortices than in other lobes. This shift from radial to horizontal architecture of neurons is accompanied by a parallel shift in the synaptic layers from radial orientation to horizontal layering between laminae of neurons and is well demonstrated in fetal brain by synaptophysin immunoreactivity. Despite a lack of histologically distinctive layers until the second half of gestation, immunocytochemical and mRNA studies identify cells committed to a particular type of neuronal differentiation within the cortical plate before the histologically layering becomes evident [8,9]. Ninety percent of neurons of the cortex arrive by radial migration from the subventricular zone (germinal matrix), in an inside-out pattern with the earliest wave of migration forming neurons of layer 6 and later arrivals displacing the earlier ones from the surface of the plate, so that layer 2 is the most recent wave of neuroblast migration. The other 20% of cortical neurons are inhibitory interneurons that arrive by tangential, rather than radial, migration and are distributed in all layers, but predominantly in layers 2 and 4, among receptive rather than motor neurons. Early identification of specific types of neuronal progenitors is sometimes demonstrated even in postmitotic, premigratory neuroblasts of the germinal matrix and show a specific pattern in maturing neurons within periventricular nodular heterotopia [10]. GABAergic neurons of the cortical plate that arrive by tangential migration are immunoreactive for calretinin and other calcium-binding protein markers while still in the ganglionic eminence before migration and during migration before reaching their destination [6]. The sequential growth of afferent axons into the cortical plate from the thalamus and other subcortical structures may produce transitory patterns of horizontal cortical lamination even while the histologic columnar pattern of the cortical plate predominates [11]. Diffusion tensor tractography of postmortem fetuses (without motion artifact) reveals that at 17 weeks’ gestation a radial organization of afferent axons to and within the cortical plate dominates, which is gradually replaced by dominant tangential (horizontal) organization during the late second and third trimesters [12].
261
The importance of the tangentially migrating GABAergic neurons arising from the ganglionic eminence in the formation of the cortical plate cannot be overemphasized, apart from their mature function as inhibitory interneurons intrinsic within the mature cortex. GABAergic synapses are formed earlier than glutaminergic synapses and initially serve an excitatory rather than inhibitory function [13,14]. They constitute the first functional synapses in the developing hippocampus [13], as well as the first intrinsic cerebral cortical circuits between Cajal-Retzius neurons and layer 6 pyramidal cells [5]. GABAergic circuits determine the temporal and spatial aspects of synaptic integration as intrinsic synapses form within the cortical plate [14]. GABA is secreted by Cajal-Retzius neurons that also express the gene Reelin (RLN) essential for the arrangement of neurons within the cortical plate; they form the preplate plexus before the arrival of the first wave of radial migratory neuroblasts [5]. The most severe disorganization in cortical lamination occurs in those lissencephalies in which tangential migration of GABAergic neuroblasts is impaired by specific genetic mutations [15]. Finally, scattered subcortical white matter neurons are often assumed to be from incomplete radial migrations, but these heterotopic cells actually may have arrived from tangential migration [17]. Another gene more recently discovered to be primordial in the reorganization and maturation of the fetal cortical plate is GPR56 that programs a family of cell-adhesion G-proteinecoupled receptors with a large extracellular domain, which regulates the integrity of the pial membrane and cortical lamination, the defects of which may result in polymicrogyria with abnormal microscopic lamination [18,19]. Examples of abnormal tangential migration in cerebral malformations are demonstrated more recently than the well-known examples for decades of abnormal radial migrations. Defective tangential migration occurs in lissencephaly caused by DCX, LIS1, ARX genetic mutations [15] and also in holoprosencephaly [16]. The cortical plate is poorly organized and very abnormal in these dysgeneses, but the pathogenesis of these malformations is beyond the scope of this present treatise. In sum, in the normal ontogenesis of the cerebral cortex, the radial migrations of mainly glutaminergic neuroblasts provide the initial columnar architecture of the fetal cortical plate in the first half of gestation, but it is the tangential migration of GABAergic neuroblasts that regulate the transition to the mature laminar architecture of the cortical plate, including intrinsic cortical synaptic circuitry. It is important to bear in mind that a minor degree of radial microcolumnar architecture persists in normal mature cortex; the neuropathologist must be alert to not overinterpret such residual normal findings. New international league against epilepsy classification of cortical dysplasias
Until recently, the various classifications schemes proposed to define categories of focal cortical dysgeneses in partial epilepsy all addressed the issue of poor lamination, displaced and disoriented neurons, and, in some cases, cytologically abnormal and cytomegelic neurons and balloon cells (e.g., “Taylor-type”; cortical tubers in tuberous sclerosis). None of these initial schemes recognized radial columnar
262
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
architecture as a distinctive type of focal cortical dysgenesis, including the original prototype by Taylor et al. [20] and the widely cited prototype classification established by Palmini et al. [21]. This feature of focal cortical dysplasias was acknowledged by several authors in the past decade (see below), most notably by Blümcke et al. [21,22] and now, for the first time, it is incorporated into a classification scheme as a feature of cerebral cortical dysgenesis. A recent 2011 classification scheme published under the auspices of the International League Against Epilepsy (ILAE) is a consensus statement, with contributions by multiple European, North American, Japanese, and Australian neuropathologists who have special interest and long experience in epilepsy surgery [22,23] and has been validated by an international consortium of neuropathologists with experience in epilepsy [24]. This scheme has a built-in flexibility for future revisions as new data become available and interpretations evolve. It includes both focal isolated dysplasias of the cortex and also those associated with other lesions. It is expected that this scheme will become the template for uniform criteria in the neuropathologic diagnosis of cortical dysplasias, analogous to the World Health Organization of the United Nations’ classification of nervous system tumors, another consensus statement now accepted by neuropathologists throughout the world, who make note of World Health Organization grading when preparing reports of cerebral neoplasms in surgical resections and postmortem examinations. Another strength and modern approach of the new ILAE classification is that, although based primarily on neuropathologic features, it also considers and incorporates aspects of neuroimaging and clinical aspects to be integrated with interdisciplinary correlations. This first scheme of the ILAE Classification of Focal Cortical Dysplasias is reproduced in Table 1 to further disseminate and promote its recognition. Type 2 differs from type 1 because of abnormal growth and structure of individual cells in addition to the abnormal arrangement of neurons; this cytologic dysmorphism includes atypical neurons with abnormal shape, excessive growth (megalocytic forms), abnormal neuritis, and balloon cells. The atypical cells in type 2 dysplasias also exhibit expression of primitive proteins characteristic of progenitor cells, such as
nestin and vimentin [25]. Type 3 is not a distinctive or unique focal dysgenesis, as are types 1 and 2, but represents type 1 associated with other lesions. There is no provision in this scheme for type 2 associated with other lesions, although we have observed two cases of dysembryoplasic neuroepithelial tumors with a transitional zone between the dysembryoplasic neuroepithelial tumors and normal hexalaminar cortex that corresponds to type 2 dysplasia (H.B. Sarnat, unpublished data, 2013). This neuropathologic scheme should not be confused with another ILAE scheme designed for classifying clinical outcomes after epilepsy surgery, which does not include neuropathologic criteria [26] or with still other ILAE schemes on the basis of strictly clinical and electrophysiological (EEG) criteria of seizures. Isolated focal cortical dysplasias in epilepsy
Columnar architecture as one component of focal cortical dysplasias is noted by multiple recent authors [21,27-41]. The most thorough and systematic of the early classification schemes of focal cortical dysplasias were those of Mischel et al. [29] and Palmini et al. [21]. More recently, Professors Ingmar Blümcke (Germany), Roberto Spreafico (Italy), Mary Thom (United Kingdom), Eleonora Aronica (The Netherlands), and Harry Vinters (United States) have been the most influential epilepsy neuropathologists, as participants in the international consortium, in bringing recognition to the histologic pattern of columnar architecture, incorporating it as a reliable feature of some focal cortical dysplasias in surgical resections of epileptic foci in children, and including it in the ILAE classification [22,23,29,40,41]. This pattern had not been included by Palmini et al. [21] in their original, widely cited and useful preceding classification scheme, although Palmini remains an important contributor in the consortium creating the new ILAE scheme. A genetic basis is not evident in most cases and the focal nature of the lesion is more suggestive of an acquired focal disturbance in fetal life than of a genetic defect. Figs 2 and 3 illustrate representative cases of type Ia focal cortical dysplasia, in which radial columnar architecture is the most prominent feature, without dysplastic or megalocytic neurons or balloon cells as occur in type 2 focal cortical
Table 1. 2011 ILAE classification of focal cortical dysplasias [10]
FCD type 1 (isolated) 1a: FCD with abnormal radial cortical lamination* 1b: FCD with abnormal tangential cortical lamination 1c: FCD with abnormal radial and tangential cortical lamination FCD type 2 (isolated) 2a: FCD with dysmorphic neurons 2b: FCD with dysmorphic neurons and balloon cells FCD type 3 (type 1 associated with principal lesions) 3a: Cortical lamination abnormalities in the temporal lobe associated with hippocampal sclerosis 3b: Cortical lamination abnormalities adjacent to a glial or glioneuronal tumor or any other cerebral tumor 3c: Cortical lamination abnormalities adjacent to a vascular malformation 3d: Cortical lamination abnormalities adjacent to any other lesion acquired during early life (e.g. trauma; ischemic injury; infarct; encephalitis) Abbreviations: FCD ¼ Focal cortical dysplasias ILAE ¼ International League Against Epilepsy At present, there is no provision in this scheme for type 2 dysplasias associated with other (principal) lesions of the brain. * The term “lamination” used in this scheme potentially could introduce some ambiguity and requires clarification. Many neuropathologists prefer the term architecture of the tissue because lamination traditionally implies horizontal layers of cells of the same type, because the term was originally used by the early neuroanatomists of the late nineteenth and early twentieth centuries. For example, Ramón y Cajal [6], who first described the radial columns during development, did not use the term lamination to denote this architecture.
263
print & web 4C=FPO
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
Figure 2. Surgical resection of focal cortical dysplasia, type 1a, in the right frontal lobe of a 2 ½-year-old boy with refractory partial epilepsy and an active discrete epileptiform focus in the right frontal lobe near the Rolandic sulcus. The child had no neurologic deficits or evidence of developmental delay. The case was considered “nonlesional” because the MRI did not disclose any lesions. Macroscopic views of the fresh tissue show (A) well-formed gyri at the cortical surface and (B) well-demarcated gray and white matter in cut sections. (C, D) Some gyri show a prominent microscopic feature of radial microcolumnar architecture with more subtle superimposed discontinuous horizontal lamination. Dysplastic neurons and balloon cells are absent. The number and distribution of GABAergic inhibitory interrneurons, demonstrated with calretinin antibody (not shown), was normal, indicating a disturbance of radial, but not tangential, neuroblast migration. (E) Another gyrus shows disorganized cortex with neither columnar nor horizontal lamination, although both are seen in small focal regions. Many neurons are disoriented. (C-E) Neuronal nuclear antigen (NeuN), (C, E) original magnification 100; (D) original magnification 250.
dysplasia. The microcolumnar pattern is most evident at the crowns of gyri, less in the cortex at the depths of sulci, and least along the cortex bordering sulci (Fig 3). Radial columnar architecture associated with other conditions Porencephalic cyst margin
print & web 4C=FPO
Porencephalic cysts are due to occlusion of a major cerebral artery, usually the middle cerebral artery, in
midgestation. The porencephalic cyst is, thereby, a large parenchymal infarct in fetal life that includes the wall of the lateral ventricle [42]. In neuroimaging studies the cyst appears to communicate with not only the ventricular lumen but also the subarachnoid space over the surface of the brain, but careful neuropathologic examination postmortem demonstrates that the pial membrane is nearly always preserved and thus separates the two fluid compartments but is too thin for the limits of resolution of MRI or CT. The rim of porencephalic cysts usually consist of
Figure 3. Surgical resection of focal cortical dysplasia, type 1a, in the parietooccipital region of a 7-year-old boy with refractory partial epilepsy, learning disabilities, and behavioral problems. MRI showed T2 hyperintensity at the site of the paroxysmal EEG focus, corresponding to this resection. (A) At the crown of a gyrus, radial microcolumns are seen in the cortex without cytologic alterations of neurons or balloon cells. (B) At the depth of the sulcus of this gyrus, focal microcolumns are still identified but are not as prominent as at the crown, and the cellular architecture generally is less well organized. Along the margin of the sulcus (not shown) the microcolumns were least evident, and a more normal horizontal lamination was demonstrated. Neuronal nuclear antigen (NeuN), original magnification 100.
264
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
DiGeorge syndrome (22q11.2 microdeletion)
Microdeletion of more than 32 contiguous genes on chromosome 22 are demonstrated in human beings, most frequently at loci at 21q11.1 and especially 21q11.2 (DiGeorge syndrome, velo-cardio-facial syndrome) [43,44]. Clinical neurologic features range from minimal cognitive deficits and learning disabilities to a high incidence of psychosis, schizophrenia, or autism [45]. MRI studies may be nondiagnostic, although either hyperplasia or absence of the corpus callosum, large cavum septi pellucidi are other mild midline malformations are described in more than 90% of patients [46,47]. Congenital cardiac defects are frequent, and these same patients exhibit abnormal cerebral convolutions and cortical microscopic architecture [48]. Neuropathologic studies have been performed, mainly in a few adults, in whom microscopic evidence of a neuroblast migratory disorder, excessive subcortical heterotopic neurons, small periventricular nodular heterotopia, and poorly laminated cerebral cortex are the principal lesions [43]. Deviant trajectories of axons in long tracts are consistent with abnormal cerebral cortical organization [49]. In children, neuropathologic examination reveals similar findings and also includes expression as abnormal neuronogenesis, synaptogenesis, and mitochondrial function [44]. A 3-month-old infant with 22q11 deletion and
print & web 4C=FPO
polymicrogyria, and the histologic architecture is arrested radial columnar without lamination as shown in Fig 4, or a sheet of disorganized neurons and glial cells with no recognizable architecture. The pathophysiology of this migratory arrest is probably ischemic, with the margin of the cyst being barely perfused enough to preserve the viability of the cells but not enough to enable the normal developmental processes to proceed, yet not included in the zone of frank infarction. Porencephaly is sometimes confused with schizencephaly in neuroimaging, and indeed they may be ambiguous in some cases, particularly in fetal MRI or in neonates. Abnormalities of blood vessels, large or small, chronic ischemia, and microinfarcts denoting earlier ischemia or parenchymal hemorrhage are excluded from type 1 but are identified as primary associated lesions in type 3. Chronic or acute gliosis is not a criterion. Small white matter microhemorrhages in the fetal brain do not usually leave permanent hemosiderin deposits and gliosis to mark their occurrence, as might occur in the mature brain, and are not as constant with fetal microinfarcts. Nevertheless, the columnar pattern of cortical architectural disturbance is as distinctive as other types and often is discovered because it is the site of an epileptic focus and eventually is treated by surgical excision, enabling neuropathologic examination.
Figure 4. A 9-year-old girl with progressive refractory partial epilepsy arising in foci within the margins of a left frontotemporal porencephalic cyst that resulted from middle cerebral arterial infarction in fetal life. (A, B) Computed tomography scan at levels of corpus striatum and thalamus, show left porencephaly. Resection was performed of the rostral and lateral margins of the cyst, the epileptogenic zones indicated by EEG monitoring. Preserved but atrophic cortical gray matter is superficial to the cyst. (C, D) The cyst margin shows radial microcolumnar architecture as the predominant pattern of focal dysplasia, corresponding to ILAE type 3d. No balloon cells or megalocytic neurons are evident. This pattern probably is focal maturational arrest in fetal life caused by ischemia. (C, D) Neuronal nuclear antigen (NeuN), (C) original magnification 100; (D) original magnification 250.
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
An alternative speculative explanation of the delay in maturation of the cortical architecture in DiGeorge syndrome is that the congenital heart lesions that often are part of this syndrome may result in chronic ischemia or hypoperfusion of the cortical plate during both early and late gestation but did not result from postnatal complications. The infant we here illustrate had tetralogy of Fallot and polycystic lungs. Death ensued at age 49 days after parental decision to discontinue life support because of the poor prognosis. Other chromosomopathies
In other diverse chromosomal duplications, translocations, deletions, and microdeletions, radial microcolumnar architecture also occurs within the cortex, usually multifocally; we have observed it in trisomies 13, 18, and 21, but a complete survey of cortical architectural aberrations is beyond the scope of this review. Hemimegalencephaly
This unique malformation involves one cerebral hemisphere, with or without asymmetry of development of subcortical structures such as the thalamus, cerebellum, and brainstem. It is primarily a disorder of cellular lineage and cytologic growth and differentiation; disturbances of cellular migration within the brain are secondary to both the abnormal neuroblasts and abnormal radial glial cells that guide them [52]. In one of our patients, radial microcolumnar architecture was demonstrated focally in the
print & web 4C=FPO
tetralogy of Fallot has abnormal cytoarchitecture of the cerebral cortex and caudate nucleus, but the distribution of calretinin-reactive cortical interneurons was normally preserved [50], as it was in the case here presented. Generalized radial columnar architecture is not previously recognized, but reported postmortem examinations in infants are rare. The patient here illustrated, a premature infant with a conceptional age of 35 weeks, is the youngest case yet reported; despite generalized minicolumnar histologic arrangement in all regions of the cerebral cortex, the synaptic architecture varied from the expected horizontal lamination in some areas [51] and radial synaptic layering in others (Fig 5). Immunoreactivity of neuronal nuclear antigen (NeuN), a nuclear protein expressed in late stages of neuronal maturation, shows reactive neurons in the middle and deep portions of the cortex but not in the most superficial part that corresponds to layer 2 (Fig 5B), which indicates primordial horizontal lamination of the vertically arranged microcolumns of neurons. The GABAergic interneurons arriving by tangential migration are normal in number and arrangement within the cortical plate (Fig 5C). Synaptophysin shows radial columns of synapses in most cortical regions (Fig 5D and 5F) but early horizontal layering of synapses in some regions (Fig 5G). Whether this pattern of microcolumnar architecture is a maturational delay that would have developed into horizontal lamination if this infant had survived or is a maturational arrest that would not have changed cannot be known from our case, but this developmental question merits focused observation in other cases of DiGeorge syndrome in later infancy and childhood.
265
Figure 5. (A-D) Sections of temporal neocortex at autopsy in a 6-week-old boy, born at 29 weeks’ gestation (conceptional age 35 weeks’ gestation) with DiGeorge syndrome, genetically confirmed as a 21q11.2 microdeletion. He had pharmacologically controlled generalized epilepsy and tetralogy of Fallot and polycystic lungs. MRI showed that the anterior commissure failed to form, but the corpus callosum was preserved, and gyration of the cortex was appropriate for age. Predominant radial microcolumnar architecture was generalized and equally involved all lobes and the insula. Radial microcolumns are well demonstrated by (A) Hematoxylin and eosin staining (H&E) and (B) neuronal nuclear antigen (NeuN) show lack of reactivity in the most superficial neurons that correspond to layer 2, thus indicating a primordial horizontal laminar cortical architecture. (C) Calretinin demonstrates a normal number and distribution of GABAergic inhibitory interneurons that arrive by tangential, rather than radial, migration. (D) Synaptophysin reveals strong synaptic vesicle reactivity, but the orientation of the synaptic layers are more vertical than horizontal. (E, F) Frontal neocortex shows a similar pattern with both (E) H&E and (F) synaptophysin. (G) Synaptophysin reactivity in the parietal cortex exhibits similar microcolumnar architecture, but the synaptic pattern is more mature with more prominent horizontal lamination of synapses. (A-G) Original magnification 250.
266
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
contralateral “normal” hemisphere but was limited to the cingulate gyrus (Fig 6); this gyrus also exhibited other evidence of delayed neuronal maturation not seen in other gyri of this hemisphere, which is more fully discussed in another publication [53]. Minor focal defects of cortical architecture occur frequently in the contralateral “normal” hemisphere in holoprosencephaly, as shown by autopsy, but usually are microscopic findings that would not be evident by neuroimaging; hence, this feature is not as well recognized. Tuberous sclerosis
The architecture of the cortex in cortical tubers is dysplastic, apart from the presence of balloon cells and atypical dysplastic neurons and glial cells. The dysplastic pattern in cortical tubers often includes focal areas of loose radial columnar architecture, but this is a minor feature (Fig 7). It is not necessarily related to the presence or absence of balloon cells within the tuber at the site of the microcolumns. Other genetic cerebral cortical malformations
In holoprosencephaly, the most severely organized cortex, often nodular or totally lacking in recognizable architecture, is the sagittal and parasagittal cortex. Transitional zone to normal hexalaminar cortex in the more lateral regions follow a mediolateral gradient of genetic expression [54]. This transitional zone often exhibits radial columnar architecture. In the lissencephalies, particularly type 2 associated with Walker-Warburg syndrome, focal regions of cortex include excessive columnar architecture among other abnormal patterns of dysgenesis. We also have observed this pattern in Meckel-Grüber syndrome, as well as in various other
cortical dysgeneses, many of which do not correspond to easily classified syndromes or specific genetic mutations. Fetal metabolic encephalopathies
We have observed maturational delay in general, and micro-columnar cortical architecture in particular, in several fetuses and neonates with a variety of inborn errors of metabolism expressed in fetal life. An example is shown in Fig 8, of a term neonate who died with seizures and hyperammonemia caused by organic (methylmalonic) aciduria. Aminoacidurias, such as phenylketonuria, that can be compensated in intrauterine life by placental exchange and only become manifest postnatally after feeding is initiated, would not be expected to interfere with prenatal brain development; hence, cortical lamination is normal at term. A systematic survey of neocortical architecture in fetal metabolic encephalopathies is in progress in our laboratory. Effect of postmortem autolysis on the microscopic appearance of cortical architecture in fetal and neonatal postmortem examinations
Autolysis causes microscopic swelling of the cortical neuropil that separates neurons more than in the physiological state during life. Such postmortem changes increase with time between death and fixation of the tissue, so that postmortem examinations delayed more than 24 hours after death tend to show more pronounced autolytic swelling. The subtle residual radial columnar pattern detected in promptly fixed cortical tissue, whether surgical or autopsy, becomes exaggerated by this artifactual cell separation (Fig 9). The diagnosis of focal cortical dysplasia 1a in particular thus becomes problematical in brain tissue exhibiting extensive autolytic changes. Discussion
print & web 4C=FPO
Nature of radial microcolumnar cortical architecture
Figure 6. Biopsy of right cingulate gyrus during corpus callosotomy for refractory epilepsy in a 3-week-old term neonate with severe left hemimegalencephaly (HME). The histologic pattern of this cortex is radial columnar more than laminar. The infant died after surgery at age 2½ months, after undergoing left partial hemispherectomy. At postmortem examination, the non-HME hemisphere exhibited microcolumnar architecture only in the cingulate gyrus; gyri in all other cortical regions showed the expected 6-layer cortex with horizontal lamination. Neuronal nuclear antigen (NeuN); original magnification 250.
The pathogenesis of all malformations of the central nervous system has a basis as disturbances in normal developmental processes, regardless of whether the cause is known. Causes include not only genetic mutations, but epigenetic and environmental factors such as exposure to teratogenic drugs and toxins, x-irradiation, trauma, ischemia, and infarction during fetal life. In this context, persistence after midgestation of the normal pattern of radial columnar architecture of early fetal life may be regarded as a maturational delay or arrest, the latter distinction depending on the potential for further maturation with time. Columnar architecture in focal cortical dysplasias thus is pathologic only because of age and failure of expected maturation. It nevertheless qualifies as a true malformation postnatally in the same way that heterotopic neurons in the deep subcortical white matter are pathologic because of migratory arrest before reaching their intended destination. Disturbances of the 10% of cortical neurons that migrate radially from 16 weeks to the late third trimester of gestation also may be attributed to maturational delay [55]. Another feature of delayed maturation of corticogenesis is persistence of the subplate zone [56], which normally is
267
print & web 4C=FPO
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
Figure 7. Surgical resection of cortical tuber at site of highly epileptic focus in right temporal neocortex of a 2-year-old boy with tuberous sclerosis. (A, B) This region of disorganized cortex exhibits poor lamination and loosely organized radial microcolumns of neurons. Balloon cells are not seen in this field but were focally numerous in the subcortical white matter and cortical gray matter at other sites (not shown). Neuronal nuclear antigen (NeuN), (A) original magnification 40; (B) original magnification 100.
This ischemia resulted in arrest of further maturation of the cortical plate. This columnar architecture is classified as a focal cortical dysplasia type 1a. Gliosis is not a prominent feature of radial columnar architecture, regardless of cause, although it may coexist if chronic ischemia was present, as in the margins of porencephaly. Astrocytic proliferation and excessive satellitosis occur in many focal cortical dysplasias, but diagnostically is nonspecific. The purpose of the columns in the early fetal cortical plate is not only related to the radial migratory pattern to reach the cortical plate, but to facilitate synaptic chains or a series or synapses between the future layers of neurons of the same type. Such synaptic chains are important for neuronal sequence generation as a substrate for complex neural functions [60]. Before 22 weeks’ gestation, there is little synaptophysin activity within the cortical plate to indicate synaptogenesis with synaptic vesicle formation, except in the molecular and subplate zones, and maturational delay in the normal temporal sequence of neocortical synaptogenesis after midgestation may be evident postnatally in focal cortical dysplasias [51]. Our study serves to document the neuropathologic findings of maturational delay in cortical plate architecture, but it was not designed to demonstrate pathophysiological mechanisms, which would require experimental studies in
print & web 4C=FPO
transitory and incorporated into the deepest part of the maturing cortical plate (not yet a laminar layer 6) by about 14 weeks. An accompanying feature of type 1 focal cortical dysplasia that is not really a transitory normal finding is an excessive number of scattered neurons in the molecular zone. These neurons are not Cajal-Retzius neurons because they exhibit nuclear reactivity with neuronal nuclear antigen (NeuN) as do all neurons of the cerebral cortex at maturation, except the Cajal-Retzius neurons, which are never marked by this antibody [57,58]. More recently, it is recognized that the cortex is capable of plasticity in reorganization after acquired lesions of the formed cortex in the neonate [59]. It is of interest that this anomaly of radial columnar architecture occurs in two different malformations involving predominantly midline structures of the cerebrum: DiGeorge syndrome (Fig 5) and holoprosencephaly (Fig 6). The significance of this finding in the contralateral hemisphere in an infant with hemimegalencephaly is not evident, particularly why it is so focal in the cingulate gyrus, on the medial side. Our example of the epileptogenic margin of a porencephalic cyst that formed subsequent to a middle cerebral artery infarct at or shortly before midgestation (Fig 4) likely had its origin of the columnar architecture as a maturation arrest of an ischemic zone adjacent to the zone of coagulation necrosis that subsequently became cystic.
Figure 8. Inferior temporal gyrus of a 3-day-old male infant, born at 39 weeks’ gestation, who died of hyperammonemia confirmed as methylmalonic acidemia. (A) Microcolumns of neurons are well demonstrated by NeuN and are oriented radially. (B) Synaptic layers also are predominantly perpendicular to the pial surface of the cortex. A normal number and distribution of GABAergic interneurons were demonstrated by calretinin (not shown), indicating a disorder of radial, but not tangential, neuroblast migration in this inborn error of metabolism. (A) Neuronal nuclear antigen (NeuN); (B) synaptophysin; original magnification 100. Courtesy of Drs. David George and Alfredo Pinto-Rojas, Alberta Children’s Hospital.
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
print & web 4C=FPO
268
Figure 9. Insular cortex of a spontaneous 21-week fetal loss without cerebral malformations shows extensive postmortem autolysis; the fetus had macerated skin and autolysis in multiple organs. The autolytic swelling of the cerebral tissue produces an exaggerated appearance of columnar architecture that is partly artifactual. Although not a true cortical dysplasia, in a term neonate or postnatal infant or child, it would fulfill the criteria of focal cortical dysplasia type Ia. Hematoxylin & eosin staining; original magnification 250.
animals. We have speculated that ischemia may be a factor in the margins of porencephalic cysts and in infants with congenital heart disease, such as DiGeorge syndrome. In genetic diseases, including many chromosomopathies, faulty programming of brain development affects not only morphogenesis, but also the timing of onset of developmental processes. Neurotoxic products of metabolic encephalopathies, such as hyperammonemia in our patient with methylmalonic acidemia, may superimpose additional epigenetic factors that further contribute to delayed maturation. Is persistent columnar architecture epileptogenic?
Whether the radial columnar pattern of cortical architecture is intrinsically epileptogenic is an incompletely resolved issue. At midgestation, when it is the expected architecture of the neocortex, there are so few synapses within the cortical plate that the generation of seizures would not be physiologically possible [51]. Postnatally, focal cortical dysplasias with radial columnar architecture often are highly epileptogenic and refractory to pharmacologic treatment. Porencephalic cyst rims are frequently epileptogenic (Fig 4). DiGeorge syndrome, by contrast, in which this fetal columnar pattern predominates in all regions of the cortex as a component of faulty genetic programming (Fig 5), is not usually characterized by intractable epilepsy, although such patients do have a higher than expected incidence of seizures [45]. In a familiar form of 22q12 mutation, variable epileptic foci within a family are mapped at the classical DiGeorge syndrome locus [61]. In other genetic disorders resulting in abnormal cortical architecture, the frequency of epilepsy is variable: in lissencephaly 1 and 2, nearly all patients have severe epilepsy. In holoprosencephaly, by contrast, only 40%-60% of patients have epilepsy postnatally or in childhood, despite the severity of the malformation both macroscopically and
microscopically with abnormal patterns of synaptogenesis [62,63]. Focal cortical dysplasias in otherwise well-formed brains also are often epileptogenic, and this symptom is what brings attention to the malformation in the first place by initiation of EEG and MRI examinations. A correlation may also be relevant that persistent columnar architecture occurs in a mild form in the depths of normal sulci and that this location is the most frequent site of origin of focal dysplasias and epileptic activity [64]. Microcolumnar architecture in focal cortical dysplasia type 1a is present at both the crowns of gyri and in the depths of sulci but more prominent in the crown (Fig 3). In addition to a minor persistence of this fetal architecture in normal mature brains, the originally radial synaptic layers may accompany the cellular architecture and, if too prominent in abnormal cortices, may prevent synaptic remodeling with maturation for normal mature intrinsic synaptic circuitry. Potential demonstration by fetal neuroimaging
It would be advantageous for prenatal diagnosis if the histologic patterns here described could be correlated in the living fetus by ultrasonography, MRI, diffusion tensor, or other neuroimaging techniques. Rapid advances are being made in resolution and interpretation of images by both transabdominal and transvaginal sonography and transabdominal MRI. Horizontal lamination of the germinal matrix, intermediate zone of white matter, subplate zone, cortical plate, and molecular zone are now identified in fetuses at midgestation or less [65-67]. Is radial columnar architecture of the neocortex a maturational arrest or a cortical dysgenesis?
Our response to the query posed in the title of this review is that microcolumnar architecture is both; these two interpretations are neither in conflict with each other nor superfluous. One may conclude that persistent fetal microcolumnar architecture, whether generalized or focal, is a form of maturational arrest in cortical development and, as such, is analogous to disorders of neuroblast migration but in the last stage with synaptic reorganization of the cortical plate. This pattern is incorporated in the new ILAE classification scheme of cerebral cortical dysgeneses. It differs from neuroblast migratory disorders, such as periventricular nodular heterotopia or subcortical laminar or band heterotopia because it indicates arrest in the intrinsic reorganization of the cortical plate at a later stage than as neuroblasts not proceeding further from intermediate sites in the white matter during the course of migration. It is emphasized that traces of radial microcolumnar architecture persist even in fully mature cerebral cortex and should not be misinterpreted as pathologic. Microcolumnar architecture thus begins as a physiological process and becomes pathologic by virtue of advancing gestational and postnatal age. We are grateful to Dr. Ingmar Blümcke, Erlangen, Germany, for his helpful comments on this work. We thank Gaston Guenette and Patricia McGinnis of the Histopathology Laboratory at Alberta Children’s Hospital, and Vivian King and her staff in the Immunopathology Laboratory of Calgary Laboratory Services, for their meticulous technical preparation of tissues. General (non-neuropathologic) portions of the postmortem examinations of cases here described were performed by Drs. C.L.
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270 Trevenen, A. Pinto-Rojas and W. Yu; surgical resections of cases here described were performed by Dr. W. Hader, all at Alberta Children’s Hospital.
References [1] Rakic P. Guidance of neurons migrating to the fetal monkey neocortex. Brain Res 1971;33:471e6. [2] Rakic P. Radial versus tangential migration of neuronal clones in the developing cerebral cortex. Proc Natl Acad Sci USA 1995;92: 11323e7. [3] Sarnat HB. Cerebral dysgenesis. Embryology and clinical expression. New York: Oxford University Press; 1992. [4] Sarnat HB. Regional differentiation of the human fetal ependyma: Immunocytochemical markers. J Neuropathol Exp Neurol 1992;51: 58e75. [5] Sarnat HB, Flores-Sarnat L. Cajal-Retzius and subplate neurons. Role in cerebral cortical development. Eur J Paediatr Neurol 2002; 6:91e7. [6] Jiménez D, López-Mascaraque LM, Valverde F, De Carlos JA. Tangential migration in neocortical development. Dev Biol 2002; 244:155e69. [7] Ramón y Cajal S. L’histologie du système nerveux de l’homme et des vértébrés. Paris: Masson 1909-1911. In: Reprinted in English translation, Histology of the Nervous System of Man and Vertebrates, in 2 volumes. New York: Oxford University Press; 1995. [8] Sanes JR, Yamagata M. Formation of lamina-specific synaptic connections. Curr Opin Neurobiol 1999;9:79e87. [9] Hevner RF. Layer-specific markers as probes for neuron type identity in human neocortex and malformations of cortical development. J Neuropathol Exp Neurol 2007;66:101e9. [10] Ferland RJ, Batiz L, Neal J, et al. Disruption of neural progenitors along the ventricular and subventricular zones in periventricular heterotopia. Hum Mol Genet 2009;18:497e516. [11] Kostovi c I, Judas M. Correlation between the sequential ingrowth of afferents and transitent patterns of cortical lamination in preterm infants. Anat Rec 2002;267:1e6. [12] Takahashi E, Folkerth RD, Galaburda AM, Grant PE. Emerging cerebral connectivity in the human fetal brain: An MRI tractography study. Cereb Cortex 2012;22:455e64. [13] Ben-Ari Y, Khalilov I, Represa A, Gozlan H. Interneurons set the tune of developing networks. Trends Neurosci 2004;27:422e7. [14] Akerman CJ, Cline HT. Refining the roles of GABAergic signaling during neural circuit formation. Trends Neurosci 2007;30:382e9. [15] Marcorelles P, Laquerrière A, Adde-Michel C, et al. Evidence for tangential migration disturbances in human lissencephaly resulting from a defect in LIS1, DCX and ARX genes. Acta Neuropathol 2010;120:503e15. [16] Sarnat HB, Flores-Sarnat L. Precocious and delayed neocortical synaptogenesis in fetal holoprosencephaly [e-pub ahead of print]. Clin Neurol doi: 10.5414/NP300588. [17] Takano T, Sawai C, Takeuchi Y. Radial and tangential neuronal migration disorder in Ibotenate-induced cortical lesions in hamsters: immunohistochemical study of reelin, vimentin, and calretinin. J Child Neurol 2004;19:107e15. [18] Li S, Jin Z, Koirala S, et al. GPR56 regulates pial basement membrane integrity and cortical lamination. J Neurosci 2008;28:5817e26. [19] Bahi-Buisson N, Poirier K, Boddaert N, et al. GPR56-related bilateral frontoparietal polymicrogyria: further evidence for an overlap with the cobblestone complex. Brain 2010;133:3194e209. [20] Taylor DC, Falconer MA, Bruton CJ, Corsellis JA. Focal dysplasia of the cerebral cortex in epilepsy. J Neurol Neurosurg Psychiatry 1971;34:369e87. [21] Palmini A, Najm I, Avanzini G, et al. Terminology and classification of the cortical dysplasias. Neurology 2004;62:S2e8. [22] Blümcke I, Spreafico R. An international consensus classification for focalcorticaldysplasias. Lancet Neurol 2011;10:26e7. [23] Blümcke I, Thom M, Aronica E, et al. The clinicopathologic spectrum of focal cortical dysplasias: A consensus classification proposed by an ad hoc task force of the ILAE Diagnostic Methods Commission. Epilepsia 2011;52:158e74. [24] Coras R, de Boer O, Armstrong D, et al. Good interobserver and intraobserver agreement in the evaluation of the new ILAE classification of focal cortical dysplasias. Epilepsia 2012;53:1341e8.
269
[25] Orlova KA, Tsai V, Baybis M, et al. Early progenitor cell marker expression distinguishes type II from type I focal cortical dysplasias. J Neuropathol Exp Neurol 2010;69:850e63. [26] Wieser HG, Blume WT, Fish D, et al. ILAE Commission Report. Proposal for a new classification of outcome with respect to epileptic seizures following epilepsy surgery. Epilepsia 2001;42: 282e6. [27] Hardiman O, Burke T, Philips J, et al. Microdysgenesis in resected temporal lobe neocortex: incidence and clinical significanace in focal epilepsy. Neurology 1988;38:1041e7. [28] Farrell MA, DeRosa MJ, Curran JG, et al. Neuropathologic findings in cortical resections (including hemispherectomies) performed for the treatment of intractable childhood epilepsy. Acta Neuropathol 1992;83:246e59. [29] Mischel PS, Nguyen L, Vinters HV. Cerebral cortical dysplasia associated with pediatric epilepsy. Review of neuropathological features and proposal for a grading system. J Neuropathol Exp Neurol 1995;54:137e53. [30] Blümcke I, Beck H, Suter B, et al. An increase of hippocampal calretinin-immunoreactive neurons correlates with early febrile seizures in temporal lobe epilepsy. Acta Neuropathol 1999;97:31e9. [31] Thom M, Sisodiya S, Harkeness W, Scaravilli F. Microdysgenesis in temporal lobe epilepsy. A quantitative and immunohistochemical study of white matter neurons. Brain 2001;124:2299e309. [32] Lahl R, Villagrán R, Teixeira W. Neuropathology of focal epilepsies: An atlas. Eastleigh, U.K: John Libbey; 2003. [33] Bocti C, Robitaille Y, Diadori P, et al. The pathological basis of temporal lobe epilepsy in childhood. Neurology 2003;60:191e5. [34] Hildebrand M, Pieper T, Winkler P, Kolodziejczk D, Holthausen H, Blümke I. Neuropathological spectrum of cortical dysplasia in children with severe focal epilepsies. Acta Neuropathol 2005;110: 1e11. [35] Eriksson SH, Free SL, Thom M, et al. Correlation of quantitative MRI and neuropathology in epilepsy surgical resection sectionsdT2 correlates with neuronal tissue in grey matter. Neuroimage 2007; 37:48e55. [36] Blümcke I, Pauli E, Clusmann H, et al. A new clincopathological classification system for mesial temporal sclerosis. Acta Neuropathol 2007;113:235e44. [37] Sarnat HB, Flores-Sarnat L. Neuropathology of pediatric cerebral resections for epilepsy. In: Wheless JW, Willmore LJ, Brumback RA, editors. Pediatric epilepsy. Shelton, Connecticut: People’s Medical Publishing House; 2009. p. 77e91. [38] Blümke I, Vinters HV, Armstrong D, Aronica E, Thom M, Spreafico R. Malformations of cortical development and epilepsies: neuropathological findings with emphasis on focal cortical dysplasia. Epil Disord 2009;11:181e93. [39] Krsek P, Pieper T, Karlmeier A, et al. Differential presurgical characteristics and seizure outcomes in children with focal cortical dysplasia type I or II. Epilepsia 2009;50:125e37. [40] Spreafico R, Blümcke I. Focal cortical dysplasias: Clinical implication of neuropathological classification systems. Acta Neuropathol 2010;120:359e67. [41] Aronica E, Becker AJ, Spreafico R. Malformations of cortical development. Brain Pathol 2012;22:380e401. [42] Govaert P. Prenatal stroke. Semin Fetal Neonatal Med 2009;14: 250e66. [43] Kiehl TR, Chow EWC, Mikulis DJ, George SR, Bassett AS. Neuropathologic features in adults with 22q11.2 deletion syndrome. Cereb Cortex 2009;19:153e64. [44] Meechan DW, Maynard TM, Tucker ES, LaMantia A-S. Three phases of DiGeorge/22q11 deletion syndrome pathogenesis during brain development: patterning, proliferation, and mitochondrial functions of 22q11 genes. Int J Dev Sci 2011;29:283e94. [45] Bassett AS, Chow EWC, Husted J, et al. Clinical features of 78 adults with 22q11 deletion syndrome. Am J Med Genet A 2005;138:307e13. [46] Chow EWC, Mikulis DJ, Zipursky RB, Scott LE, Weksberg R, Bassett AS. Qualitative MRI findings in adults with 22q deletion syndrome and schizophrenia. Biol Psychiatry 1999;46:1436e42. [47] Shashi V, Muddasani S, Santos CC, et al. Abnormalities of the corpus callosum in nonpsychotic children with chromosome 22q11 deletion syndrome. Neuroimage 2004;21:1399e406. [48] Schaer M, Glaser B, Cuadra MB, Debbane M, Thiran JP, Eliez S. Congenital heart disease affects local gyrification in 22q11.2 deletion syndrome. Dev Med Child Neurol 2009;51:746e53.
270
H.B. Sarnat, L. Flores-Sarnat / Pediatric Neurology 48 (2013) 259e270
[49] Schaer M, Debbané M, Bach Curadra M, et al. Deviant trajectories of cortical maturation in 22q11.2 deletion syndrome (22q11DS): A cross-sectional and longitudinal study. Schizophrenia Res 2009; 115:182e90. [50] McFadden K, Wu P, Murdoch G. Neuropathology of 22q11 deletion syndrome. J Neuropathol Exp Neurol 2012;71:559 [abstract]. [51] Sarnat HB, Flores-Sarnat L, Trevenen CL. Synaptophysin immunoreactivity in the human hippocampus and neocortex from 6 to 41 weeks of gestation. J Neuropathol Exp Neurol 2010;69:234e45. [52] Flores-Sarnat L, Sarnat HB, Dávila-Gutiérrez G, Álvarez A. Hemimegalencephaly, Part 2. Neuropathology suggests a disorder of cellular lineage. J Child Neurol 2003;18:776e85. [53] Sarnat HB, Flores-Sarnat L, Crino PB, et al. Hemimegalencephaly: foetal tauopathy, mTOR activation and neuronal lipidosis. Folia Neuropathol 2012;50:330e45. [54] Sarnat HB, Flores-Sarnat L. Neuropathological research strategies in holoprosencephaly. J Child Neurol 2001;16:918e31. [55] Sarnat HB. Disturbances of late neuronal migration in the perinatal period. Am J Dis Child 1987;141:969e80. [56] Torres-Reveron J, Friedlander MJ. Properties of persistent postnatal cortical subplate neurons. J Neurosci 2007;27:9962e74. [57] Wolf HK, Buslei R, Schmidt-Kastner R, et al. NeuN: A useful neuronal marker for diagnostic histopathology. J Histochem Cytochem 1996;44:1167e71. [58] Sarnat HB, Nochlin D, Born DE. Neuronal nuclear antigen (NeuN) as a marker of neuronal maturation in the early human fetal nervous system. Brain Dev 1998;20:88e94. [59] Marín-Padilla M, Parisi JE, Armstrong DL, Sargent SK, Kaplan JA. Shaken infant syndrome: developmental neuropathology, progressive
[60] [61]
[62]
[63]
[64]
[65]
[66]
[67]
cortical dysplasia and epilepsy. Acta Neuropathol 2002;103: 321e32. Long MA, Jin DZ, Fee MS. Support for a synaptic chain model of neuronal sequence generation. Nature 2010;468:394e9. Klein KM, O’Brien TJ, Praveen K, et al. Familial focal epilepsy with variable foci mapped to chromsome 22q12: expansion of the pheotypic spectrum [e-pub ahead of print]. Epilepsia doi: 10.1111/ j.1528e1167.2012.03585.x. Stashinko EE, Clegg NJ, Kammann HA, et al. A retrospective survey of perinatal risk factors of 104 living children with holoprosencephaly. Am J Med Genet A 2004;128:114e9. Yang MT, Lee WT, Peng SS, et al. The roles of electroencephalography and neuroimaging in children with holoprosencephaly. Epileptic Disord 2004;6:173e80. Besson P, Andermann F, Dubeau F, Bernasconi A. Small focal cortical dysplasia lesions are located at the bottom of a deep sulcus. Brain 2008;131:3246e55. Widjala E, Geibprasert S, Mahmoodabadi SZ, et al. Alteration of human fetal subplate layer and intermediate zone during normal development on MR and diffusion tensor imaging. Am J Neuroradiol 2010;31:1091e9. Pugash D, Hendson G, Dunham CP, et al. Sonographic assessment of normal and abnormal patterns of fetal cerebral lamination [e-pub ahead of print]. Ultrasound Obstet Gynecol doi: 10.1002/ uog.11164. Pugash D, Dewar K, Hendson G, et al. Endovaginal sonography of normal transient patterns of fetal cerebral lamination [e-pub ahead of print]. Ultrasound Obstet Gynecol 2012;40(S1):23 (abstract); September 2012; doi: 10.1002/uog.11297.
