Periventricular heterotopia may result from radial glial fiber disruption

A transient window of hypothyroidism alters neural progenitor cells and results in abnormal brain development

Cortical heterotopias are clusters of ectopic neurons in the brain and are associated with neurodevelopmental disorders like epilepsy and learning disabilities. We have previously characterized the robust penetrance of a heterotopia in a rat model, induced by thyroid hormone (TH) disruption during gestation. However, the specific mechanism by which maternal TH insufficiency results in this birth defect remains unknown. Here we first determined the developmental window susceptible to endocrine disruption and describe a cellular mechanism responsible for heterotopia formation. We show that five days of maternal goitrogen treatment (10 ppm propylthiouracil) during the perinatal period (GD19-PN2) induces a periventricular heterotopia in 100% of the offspring. Beginning in the early postnatal brain, neurons begin to aggregate near the ventricles of treated animals. In parallel, transcriptional and architectural changes of this region were observed including decreased Sonic hedgehog (Shh) expression, abnormal cell adhesion, and altered radial glia morphology. As the ventricular epithelium is juxtaposed to two sources of brain THs, the cerebrospinal fluid and vasculature, this progenitor niche may be especially susceptible to TH disruption. This work highlights the spatiotemporal vulnerabilities of the developing brain and demonstrates that a transient period of TH perturbation is sufficient to induce a congenital abnormality.

Pathophysiological analyses of periventricular nodular heterotopia using gyrencephalic mammals

Although periventricular nodular heterotopia (PNH) is often found in the cerebral cortex of people with thanatophoric dysplasia (TD), the pathophysiology of PNH in TD is largely unknown. This is mainly because of difficulties in obtaining brain samples of TD patients and a lack of appropriate animal models for analyzing the pathophysiology of PNH in TD. Here we investigate the pathophysiological mechanisms of PNH in the cerebral cortex of TD by utilizing a ferret TD model which we recently developed. To make TD ferrets, we electroporated fibroblast growth factor 8 (FGF8) into the cerebral cortex of ferrets. Our immunohistochemical analyses showed that PNH nodules in the cerebral cortex of TD ferrets were mostly composed of cortical neurons, including upper layer neurons and GABAergic neurons. We also found disorganizations of radial glial fibers and of the ventricular lining in the TD ferret cortex, indicating that PNH may result from defects in radial migration of cortical neurons along radial glial fibers during development. Our findings provide novel mechanistic insights into the pathogenesis of PNH in TD.

The value of early and comprehensive diagnoses in a human fetus with hydrocephalus and progressive obliteration of the aqueduct of Sylvius: Case Report

Background

Mutant rodent models have highlighted the importance of the ventricular ependymal cells and the subcommissural organ (a brain gland secreting glycoproteins into the cerebrospinal fluid) in the development of fetal onset hydrocephalus. Evidence indicates that communicating and non-communicating hydrocephalus can be two sequential phases of a single pathological phenomenon triggered by ependymal disruption and/or abnormal function of the subcommissural organ. We have hypothesized that a similar phenomenon may occur in human cases with fetal onset hydrocephalus.

Case presentation

We report here on a case of human fetal communicating hydrocephalus with no central nervous system abnormalities other than stenosis of the aqueduct of Sylvius (SA) that became non-communicating hydrocephalus during the first postnatal week due to obliteration of the cerebral aqueduct. The case was followed closely by a team of basic and clinic investigators allowing an early diagnosis and prediction of the evolving pathophysiology. This information prompted neurosurgeons to perform a third ventriculostomy at postnatal day 14. The fetus was monitored by ultrasound, computerized axial tomography and magnetic resonance imaging (MRI). After birth, the follow up was by MRI, electroencephalography and neurological and neurocognitive assessments. Cerebrospinal fluid (CSF) collected at surgery showed abnormalities in the subcommissural organ proteins and the membrane proteins L1-neural cell adhesion molecule and aquaporin-4. The neurological and neurocognitive assessments at 3 and 6 years of age showed neurological impairments (epilepsy and cognitive deficits).

Conclusions

(1) In a hydrocephalic fetus, a stenosed SA can become obliterated at perinatal stages. (2) In the case reported, a close follow up of a communicating hydrocephalus detected in utero allowed a prompt postnatal surgery aiming to avoid as much brain damage as possible. (3) The clinical and pathological evolution of this patient supports the possibility that the progressive stenosis of the SA initiated during the embryonic period may have resulted from ependymal disruption of the cerebral aqueduct and dysfunction of the subcommissural organ. The analysis of subcommissural organ glycoproteins present in the CSF may be a valuable diagnostic tool for the pathogenesis of congenital hydrocephalus.

Neural stem cells: are they the hope of a better life for patients with fetal-onset hydrocephalus?

I was honored to be awarded the Casey Holter Essay Prize in 2013 by the Society for Research into Hydrocephalus and Spina Bifida. The purpose of the prize is to encourage original thinking in a way to improve the care of individuals with spina bifida and hydrocephalus. Having kept this purpose in mind, I have chosen the title: Neural stem cells, are they the hope of a better life for patients with fetal-onset hydrocephalus? The aim is to review and discuss some of the most recent and relevant findings regarding mechanisms leading to both hydrocephalus and abnormal neuro/gliogenesis. By looking at these outcome studies, it is hoped that we will recognize the potential use of neural stem cells in the treatment of hydrocephalus, and so prevent the disease or diminish/repair the associated brain damage.

Magnetic resonance diffusion tensor imaging metrics in perilesional white matter among children with periventricular nodular gray matter heterotopia

BACKGROUND

Despite pharmacological and surgical interventions, some children with periventricular nodular heterotopia (PNH) remain refractory to treatment, which suggests more diffuse pathology potentially involving perilesional white matter.

OBJECTIVE

The purpose of this study was to evaluate MR diffusion tensor imaging (MRDTI) metrics within perilesional white matter in children with PNH.

MATERIALS AND METHODS

Six children with PNH (four boys; average age 3.2 years, range 2 months to 6 years) were studied with MRDTI at 3 T. Fractional anisotropy (FA), mean diffusivity (MD), radial diffusivity (RD), and axial diffusivity (AD) were quantified within perilesional white matter at distances of 5 mm, 10 mm, 15 mm, and 20 mm from focal areas of PNH and compared to location-matched ROIs in six healthy control patients (two boys, average age 3.3 years, range 2-6 years). Statistical significance was set at an overall level of α = 0.05, corrected for multiple comparisons.

RESULTS

Perilesional white matter showed significantly decreased fractional anisotropy and elevated mean and radial diffusivity at all evaluated distances. No significant differences in axial diffusivity were detected at any distance.

CONCLUSION

PNH is associated with microstructural white matter abnormalities as indicated by abnormal perilesional MRDTI metrics detectable at least 20 mm from visible nodular lesions.

Neuropathology of brain and spinal malformations in a case of monosomy 1p36

Monosomy 1p36 is the most common subtelomeric chromosomal deletion linked to mental retardation and seizures. Neuroimaging studies suggest that monosomy 1p36 is associated with brain malformations including polymicrogyria and nodular heterotopia, but the histopathology of these lesions is unknown. Here we present postmortem neuropathological findings from a 10 year-old girl with monosomy 1p36, who died of respiratory complications. The findings included micrencephaly, periventricular nodular heterotopia in occipitotemporal lobes, cortical dysgenesis resembling polymicrogyria in dorsolateral frontal lobes, hippocampal malrotation, callosal hypoplasia, superiorly rotated cerebellum with small vermis, and lumbosacral hydromyelia. The abnormal cortex exhibited “festooned” (undulating) supragranular layers, but no significant fusion of the molecular layer. Deletion mapping demonstrated single copy loss of a contiguous 1p36 terminal region encompassing many important neurodevelopmental genes, among them four HES genes implicated in regulating neural stem cell differentiation, and TP73, a monoallelically expressed gene. Our results suggest that brain and spinal malformations in monosomy 1p36 may be more extensive than previously recognized, and may depend on the parental origin of deleted genes. More broadly, our results suggest that specific genetic disorders may cause distinct forms of cortical dysgenesis.

A glial origin for periventricular nodular heterotopia caused by impaired expression of Filamin-A

Periventricular nodular heterotopia (PH) is a human brain malformation caused by defective neuronal migration that results in ectopic neuronal nodules lining the lateral ventricles beneath a normal appearing cortex. Most affected patients have seizures and their cognitive level varies from normal to severely impaired. Mutations in the Filamin-A (or FLNA) gene are the main cause of PH, but the underlying pathological mechanism remains unknown. Although two FlnA knockout mouse strains have been generated, none of them showed the presence of ectopic nodules. To recapitulate the loss of FlnA function in the developing rat brain, we used an in utero RNA interference-mediated knockdown approach and successfully reproduced a PH phenotype in rats comparable with that observed in human patients. In FlnA-knockdown rats, we report that PH results from a disruption of the polarized radial glial scaffold in the ventricular zone altering progression of neural progenitors through the cell cycle and impairing migration of neurons into the cortical plate. Similar alterations of radial glia are observed in human PH brains of a 35-week fetus and a 3-month-old child, harboring distinct FLNA mutations not previously reported. Finally, juvenile FlnA-knockdown rats are highly susceptible to seizures, confirming the reliability of this novel animal model of PH. Our findings suggest that the disorganization of radial glia is the leading cause of PH pathogenesis associated with FLNA mutations. Rattus norvegicus FlnA mRNA (GenBank accession number FJ416060).

Monosomy1p36.3 and trisomy 19p13.3 in a child with periventricular nodular heterotopia

Monosomy 1p36 is a clinically recognizable syndrome that is considered to be the most common terminal deletion syndrome. It has characteristic clinical features that include craniofacial dysmorphism, congenital anomalies, hearing deficits, developmental delay, mental retardation, hypotonia, seizures, and brain anomalies. Brain anomalies in patients with 1p36 deletion are frequent but inconsistent. To date, 2 cases with monosomy 1p36 associated with periventricular nodular heterotopia (PNH) have been reported. We report a 2-month-old boy with multiple congenital anomalies; brain magnetic resonance imaging revealed PNH. The first 2 described cases were pure terminal deletions, whereas our patient carried unbalanced translocation due to an adjacent 1 segregation of a balanced maternal translocation, resulting in monosomy 1p36.3 and trisomy 19p13.3 identified by whole-genome array comparative genomic hybridization analysis. Our patient, with a smaller deletion that the 2 previously reported cases, can help narrow the critical region for PNH in association with the 1p36 deletion. Several potential candidate genes are discussed.

Polymicrogyria includes fusion of the molecular layer and decreased neuronal populations but normal cortical laminar organization

Malformations of cortical development are frequently identified in surgical resections for intractable epilepsy. Among the more frequently identified are cortical dysplasia, pachygyria, and polymicrogyria. The pathogenesis of these common developmental anomalies remains uncertain. Polymicrogyria is particularly vexing because there are multiple described forms (2, 4, and 6 layers) that have been attributed to multiple etiologies (e.g. ischemic, genetic, infectious, and toxic). We reviewed the pathology in 19 cases and performed cortical laminar analysis in 10 of these cases. Our data indicate that a defining feature of polymicrogyriais fusion of the molecular layer and that most often there is a well-defined gray matter-white matter junction. Unexpectedly, the cortical laminae were normally positioned, but there were reduced neuronal populations within these laminae, particularly in the subgranular layers. On the basis of these data, we propose that the categorization of polymicrogyria according to the number of lamina is artificial and should be abandoned, and polymicrogyria should be defined according to the presence or absence of coexisting neuropathological features. Furthermore, our data indicate that polymicrogyria is not a cell migration disorder, rather it should be considered a postmigration malformation of cortical development.

A simple PCR-based genotyping method for M105I mutation of alpha-SNAP enhances the study of early pathological changes in hyh phenotype

alpha-SNAP is an essential component of the protein machinery responsible for membrane fusion events in different cell types. The hyh (hydrocephalus with hop gait) mouse carries a missense mutation in Napa gene that results in a point mutation (M105I) in alpha-SNAP protein. Homozygous animals for the mutant allele have been identified by the clinical and/or neuropathological phenotype, or by direct sequencing of PCR products. The aims of the present study were (i) to develop a high-throughput technique to genotype hyh mice, (ii) to correlate genotype-phenotype, and (iii) to analyze the earliest pathological changes of hyh mutant mice. As no restriction sites are affected by the hyh mutation, we resolved this problem by creating a BspHI restriction site with a modified (mismatch) polymerase chain reaction (PCR) primer in wild-type allele. This artificially created restriction site (ACRS)-PCR technique is a simple, rapid and reliable method to genotype hyh mice in a day-work procedure. Biochemical and histological analysis of genotyped hyh embryos at different developmental stages allowed us to identify and characterize the earliest brain pathological changes of the hyh phenotype, including the first signs of neuroepithelial disruption and neuronal ectopia. In addition, genotype-phenotype analysis of 327 animals confirmed that (i) hyh is a single-gene autosomal recessive disorder, and (ii) the disorder has 100% penetrance (i.e., the mutation was only present in affected mice). The genotyping method described here enhances the potentiality of hyh mouse as a unique in vivo model to study the role of membrane trafficking in different developmental and physiological processes.

Disruption of neural progenitors along the ventricular and subventricular zones in periventricular heterotopia

Periventricular heterotopia (PH) is a disorder characterized by neuronal nodules, ectopically positioned along the lateral ventricles of the cerebral cortex. Mutations in either of two human genes, Filamin A (FLNA) or ADP-ribosylation factor guanine exchange factor 2 (ARFGEF2), cause PH (Fox et al. in 'Mutations in filamin 1 prevent migration of cerebral cortical neurons in human periventricular heterotopia'. Neuron, 21, 1315-1325, 1998; Sheen et al. in 'Mutations in ARFGEF2 implicate vesicle trafficking in neural progenitor proliferation and migration in the human cerebral cortex'. Nat. Genet., 36, 69-76, 2004). Recent studies have shown that mutations in mitogen-activated protein kinase kinase kinase-4 (Mekk4), an indirect interactor with FlnA, also lead to periventricular nodule formation in mice (Sarkisian et al. in 'MEKK4 signaling regulates filamin expression and neuronal migration'. Neuron, 52, 789-801, 2006). Here we show that neurons in post-mortem human PH brains migrated appropriately into the cortex, that periventricular nodules were primarily composed of later-born neurons, and that the neuroependyma was disrupted in all PH cases. As studied in the mouse, loss of FlnA or Big2 function in neural precursors impaired neuronal migration from the germinal zone, disrupted cell adhesion and compromised neuroepithelial integrity. Finally, the hydrocephalus with hop gait (hyh) mouse, which harbors a mutation in Napa [encoding N-ethylmaleimide-sensitive factor attachment protein alpha (alpha-SNAP)], also develops a progressive denudation of the neuroepithelium, leading to periventricular nodule formation. Previous studies have shown that Arfgef2 and Napa direct vesicle trafficking and fusion, whereas FlnA associates dynamically with the Golgi membranes during budding and trafficking of transport vesicles. Our current findings suggest that PH formation arises from a final common pathway involving disruption of vesicle trafficking, leading to impaired cell adhesion and loss of neuroependymal integrity.

Trouble making the first move: interpreting arrested neuronal migration in the cerebral cortex

Postmitotic cortical neurons that fail to initiate migration can remain near their site of origin and form persistent periventricular nodular heterotopia (PH). In human telencephalon, this malformation is most commonly associated with Filamin-A (FLNa) mutations. The lack of genetic animal models that reliably produce PH has delayed our understanding of the underlying molecular mechanisms. This review examines PH pathogenesis using a new mouse model. Although PH have not been observed in Flna-deficient mice generated thus far, the loss of MEKK4, a regulator of Flna, produces striking PH in mice and offers insight into the mechanisms involved in neuronal migration initiation. Elucidating the basic functions of FLNa and associated molecules is crucial for understanding the causes of PH and for developing prevention for at-risk patients.

Neurodevelopmental disorders as a cause of seizures: neuropathologic, genetic, and mechanistic considerations

This review will consider patterns of developmental neuropathologic abnormalities-malformations of cortical development (MCD)--encountered in infants (often with infantile spasms), children, and adults with intractable epilepsy. Treatment of epilepsy associated with some MCD, such as focal cortical dysplasia and tubers of tuberous sclerosis, may include cortical resection performed to remove the "dysplastic" region of cortex. In extreme situations (eg, hemimegalencephaly), hemispherectomy may be carried out on selected patients. Neuropathologic (including immunohistochemical) findings within these lesions will be considered. Other conditions that cause intractable epilepsy and often mental retardation, yet are not necessarily amenable to surgical treatment (eg, lissencephaly, periventricular nodular heterotopia, double cortex syndrome) will be discussed. Over the past 10 years there has been an explosion of information on the genetics of MCD. The genes responsible for many MCD (eg, TSC1, TSC2, LIS-1, DCX, FLN1) have been cloned and permit important mechanistic studies to be carried out with the purpose of understanding how mutations within these genes result in abnormal cortical cytoarchitecture and anomalous neuroglial differentiation. Finally, novel techniques allowing for analysis of patterns of gene expression within single cells, including neurons, is likely to provide answers to the most vexing and important question about these lesions: Why are they epileptogenic?

The cerebral cortex malformation in thanatophoric dysplasia: neuropathology and pathogenesis

Thanatophoric dysplasia (TD) is a relatively common, fatal form of chondrodysplastic dwarfism in which the cerebral cortex displays a unique and complex malformation. The malformation is characterized by a combination of abnormalities, which affect the temporal lobe most severely. Salient features include temporal lobe enlargement, deep transverse sulci across the inferomedial temporal surface, and hippocampal dysplasia. TD is caused by mutations of the fibroblast growth factor (FGF) receptor 3 gene (FGFR3), which result in constitutive activation of the FGFR3 tyrosine kinase. However, the link between constitutive FGFR3 activation and malformation of the cortex has been difficult to elucidate. In this review, I describe the neuropathological features of human TD, especially the cortical malformation, ascertained by examination of 45 published cases and 5 new cases, spanning gestational ages from 18 to 42 weeks. The cortical malformation is interpreted with regard to developmental mechanisms, and observations from a mouse model of TD. The evidence suggests that FGFR3 activation perturbs three key processes in cortical development: areal patterning, progenitor proliferation, and apoptosis. Defective patterning accounts for hippocampal dysplasia, while increased proliferation and decreased apoptosis account for temporal lobe hyperplasia and premature development of aberrant sulci. Disturbances in these processes may also contribute to other cortical malformations.

Malformations of cortical development: molecular pathogenesis and experimental strategies

Malformations of cortical development (MCD) are developmental brain lesions characterized by abnormal formation of the cerebral cortex and a high clinical association with epilepsy in infants, children, and adults. Despite multiple anti-epileptic drugs (AEDs), treatment of epilepsy associated with MCD may require cortical resection performed to remove the cytoarchitecturally abnormal region of cortex. Single genes responsible for distinct MCD including lissencephaly, subcortical band heterotopia, and tuberous sclerosis, have been identified and permit important mechanistic insights into how gene mutations result in abnormal cortical cytoarchitecture. The pathogenesis of MCD such as focal cortical dysplasia, hemimegalencephaly, and polymicrogyria, remains unknown. A variety of new techniques including cDNA array analysis now allow for analysis of gene expression within MCD.

Neurobeachin is essential for neuromuscular synaptic transmission

We report a random disruption in the mouse genome that resulted in lethal paralysis in homozygous newborns. The disruption blocked expression of neurobeachin, a protein containing a BEACH (beige and Chediak-Higashi) domain implicated in synaptic vesicle trafficking and an AKAP (A-kinase anchor protein) domain linked to localization of cAMP-dependent protein kinase activity. nbea-null mice demonstrated a complete block of evoked synaptic transmission at neuromuscular junctions, whereas nerve conduction, synaptic structure, and spontaneous synaptic vesicle release were completely normal. These findings support an essential role for neurobeachin in evoked neurotransmitter release at neuromuscular junctions and suggest that it plays an important role in synaptic transmission.

The Bimodal Role of Filamin in Controlling the Architecture and Mechanics of F-actin Networks

Reconstituted actin filament networks have been used extensively to understand the mechanics of the actin cortex and decipher the role of actin cross-linking proteins in the maintenance and deformation of cell shape. However, studies of the mechanical role of the F-actin cross-linking protein filamin have led to seemingly contradictory conclusions, in part due to the use of ill-defined mechanical assays. Using quantitative rheological methods that avoid the pitfalls of previous studies, we systematically tested the complex mechanical response of reconstituted actin filament networks containing a wide range of filamin concentrations and compared the mechanical function of filamin with that of the cross-linking/bundling proteins alpha-actinin and fascin. At steady state and within a well defined linear regime of small non-destructive deformations, F-actin solutions behave as highly dynamic networks (actin polymers are still sufficiently mobile to relax the stress) below the cross-linking-to-bundling threshold filamin concentration, and they behave as covalently cross-linked gels above that threshold. Under large deformations, F-actin networks soften at low filamin concentrations and strain-harden at high filamin concentrations. Filamin cross-links F-actin into networks that are more resilient, stiffer, more solid-like, and less dynamic than alpha-actinin and fascin. These results resolve the controversy by showing that F-actin/filamin networks can adopt diametrically opposed rheological behaviors depending on the concentration in cross-linking proteins.

Cortical dysplasia and epilepsy: animal models

Cortical dysplasia syndromes--those conditions of abnormal brain structure/organization that arise during aberrant brain development--frequently involve epileptic seizures. Neuropathological and neuroradiological analyses have provided descriptions and categorizations based on gross anatomical and cellular histological features (e.g., lissencephaly, heterotopia, giant cells), as well as on the developmental mechanisms likely to be involved in the abnormality (e.g., cell proliferation, migration). Recently, the genes responsible for several cortical dysplastic conditions have been identified and the underlying molecular processes investigated. However, it is still unclear how the various structural abnormalities associated with cortical dysplasia are related to (i.e., "cause") chronic seizures. To elucidate these relationships, a number of animal models of cortical dysplasia have been developed in rats and mice. Some models are based on laboratory manipulations that injure the brain (e.g., freeze, undercut, irradiation, teratogen exposure) of immature animals; others are based on spontaneous genetic mutations or on gene manipulations (knockouts/transgenics) that give rise to abnormal cortical structures. Such models of cortical dysplasia provide a means by which investigators can not only study the developmental mechanisms that give rise to these brain lesions, but also examine the cause-effect relationships between structural abnormalities and epileptogenesis.

Disruption of postnatal progenitor migration and consequent abnormal pattern of glial distribution in the cerebrum following administration of methylmercury

Transplacental administration of methylmercury (MeHg) induces disruption of neuronal migration in the developing cerebral cortex. However, the effects of MeHg on glial progenitor migration remain unclear. To understand this, we performed double administration of MeHg and 5-bromo-2-deoxyuridine (BrdU) to neonatal rat pups on postnatal day 2 (P2), when glial cells are generated from progenitors in the subventricular zone (SVZ). Histopathological examination of a proportion of the MeHg-treated rats on P28 revealed no apparent abnormalities of cytoarchitecture or neuron count in either the primary motor or primary somatosensory cortex of the cerebrum. BrdU immunohistochemistry revealed abnormal accumulation of the labeled cells in the deeper layers of the cortices and underlying white matter of both areas, where an excessive number of astrocytes (glial fibrillary acidic protein- or S-100beta-immunolabeled cells) and oligodendrocytes (2',3'-cyclic-nucleotide 3'-phosphohydrolase-labeled cells) were located. Next, to investigate the migration of individual progenitors from the forebrain SVZ of P2 neonates, we labeled them in vivo with a retrovirus encoding green fluorescent protein (GFP), following administration of MeHg, and then examined the distribution pattern of the GFP-labeled cells in the P28 cerebrum. We found that the labeled cells developed into astrocytes and oligodendrocytes and were accumulated abnormally in the lateral white matter as well as in the adjacent deeper layer of the lateral cortex and lateral side of the striatum. Thus, exposure to MeHg in the gliogenic period induced irregular distribution of glia as a consequence of abnormal migration of the postnatal progenitors.

Neuronal migration, cerebral cortical development, and cerebral cortical anomalies

Cerebral cortical malformations are relatively common anomalies identified by neuroimaging and pathologically in patients with epilepsy and mental retardation. A disruption in neuronal migration during central nervous system development has been postulated as the pathogenesis for many of these disorders. Recently, the cell migration hypothesis has been proven accurate for lissencephaly, subcortical band heterotopia, and periventricular nodular heterotopia. Furthermore, advances in cellular and molecular biology have begun elucidating the fundamental mechanisms underlying these migration disorders. These data have resulted in redefining and recategorizing specific malformations based on their molecular genetic abnormality. In this review we shall discuss the current understanding of neuronal migration in the developing cerebral cortex, the evaluation of these patients, and attempt to describe the pathogenesis for several well-characterized human disorders of cell migration.

Neuropathologic research strategies in holoprosencephaly

Hypotheses are presented to explain the pathogenesis of several clinical features of holoprosencephaly, and neuropathologic approaches to testing these hypotheses are suggested. The traditional morphologic classification of holoprosencephaly into alobar, semilobar, and lobar forms is grades of severity, and each occurs in all of the genetic mutations known. Of the four defective genes identified as primary in human holoprosencephaly, three exhibit a ventrodorsal gradient of expression (SHH, SIX3, and TGIF) and one a dorsoventral gradient (ZIC2). But, in addition to the vertical axis, genes expressed in the neural tube also may have rostrocaudal and mediolateral gradients in the other axes. These other gradients may be equally as important as the vertical. If the rostrocaudal gradient extends as far as the mesencephalic neuromere, it may interfere with the formation, migration, or apoptosis of the mesencephalic neural crest, which forms membranous bones of the face, orbits, nose, and parts of the eyes, and may explain the midfacial hypoplasia seen in many, but not all, children with holoprosencephaly. This rostrocaudal gradient also causes noncleavage of the caudate nucleus, thalamus, and hypothalamus and contributes to the formation of the dorsal cyst of holoprosencephaly, which is probably derived from an expanded suprapineal recess of the 3rd ventricle with secondary dilation of the telencephalic monoventricle and at times may produce a unique transfontanellar encephalocele. The extent of the mediolateral gradient may explain the severe disorganization of cerebral cortical architecture in medial parts of the forebrain and normal cortex in lateral parts, including the radial glial fibers. This preserved lateral cortex may explain why some children with holoprosencephaly have better intellectual function than expected and may also be important in the pathogenesis of epilepsy, by contrast with malformations such as lissencephaly, in which the entire cerebral cortex is involved. Epilepsy in some, but not all, cases also may be related to the sequential maturation of axonal terminals in relation to the neurons they innervate. Diabetes insipidus is a complication in a majority of patients; other neuroendocrinopathies occur less frequently. Secondary down-regulation of the OTP gene or of downstream genes such as BRN2 or SIM1 may result in failure of terminal differentiation of magnocellular neurons of the supraoptic and paraventricular hypothalamic nuclei. Disoriented radial glial fibers or abnormal ependyma may allow aberrant migration of neuroepithelial cells into the ventricle. A new classification of holoprosencephaly is needed to integrate morphologic and genetic criteria.