Lissencephaly 1

A number sign (#) is used with this entry because lissencephaly-1 (LIS1) and subcortical band heterotopia (SBH) are both be caused by heterozygous mutation in the PAFAH1B1 gene (601545) on chromosome 17p13.

Description

Lissencephaly (LIS), literally meaning smooth brain, is characterized by smooth or nearly smooth cerebral surface and a paucity of gyral and sulcal development, encompassing a spectrum of brain surface malformations ranging from complete agyria to subcortical band heterotopia (SBH). Classic lissencephaly is associated with an abnormally thick cortex, reduced or abnormal lamination, and diffuse neuronal heterotopia. SBH consists of circumferential bands of heterotopic neurons located just beneath the cortex and separated from it by a thin band of white matter. SBH represents the less severe end of the lissencephaly spectrum of malformations (Pilz et al., 1999, summary by Kato and Dobyns, 2003). Agyria, i.e., brain without convolutions or gyri, was considered a rare malformation until recent progress in neuroradiology (Bordarier et al., 1986). With this technical advantage, a number of lissencephaly syndromes have been distinguished.

Classic lissencephaly (formerly type I) is a brain malformation caused by abnormal neuronal migration at 9 to 13 weeks' gestation, resulting in a spectrum of agyria, mixed agyria/pachygyria, and pachygyria. It is characterized by an abnormally thick and poorly organized cortex with 4 primitive layers, diffuse neuronal heterotopia, enlarged and dysmorphic ventricles, and often hypoplasia of the corpus callosum. (Lo Nigro et al., 1997).

Kato and Dobyns (2003) presented a classification system for neuronal migration disorders based on brain imaging findings and molecular analysis. The authors also reviewed the contributions and interactions of the 5 genes then known to cause human lissencephaly: LIS1 or PAFAH1B1, 14-3-3-epsilon (YWHAE), DCX, RELN, and ARX.

Genetic Heterogeneity of Lissencephaly

Lissencephaly is a genetically heterogeneous disorder. See also LIS2 (257320), caused by mutation in the RELN gene (600514) on chromosome 7q22; LIS3 (611603), caused by mutation in the TUBA1A gene (602529) on chromosome 12q13; LIS4 (614019), caused by mutation in the NDE1 gene (609449) on chromosome 16p13; LIS5 (615191), caused by mutation in the LAMB1 gene (150240) on chromosome 7q31; LIS6 (616212), caused by mutation in the KATNB1 gene (602703) on chromosome 16q21; LIS7 (616342), caused by mutation in the CDK5 gene (123831) on chromosome 7q36; LIS8 (617255), caused by mutation in the TMTC3 gene (617218) on chromosome 12q21; and LIS9 (618325), caused by mutation in the MACF1 gene (608271) on chromosome 1p34.

X-linked forms include LISX1 (300067), caused by mutation in the DCX gene (300121) on chromosome Xq23, and LISX2 (300215), caused by mutation in the ARX gene (300382) on chromosome Xp21.

See also Miller-Dieker lissencephaly syndrome (MDLS; 247200), a contiguous gene microdeletion syndrome involving chromosome 17p13 and including the PAFAH1B1 and YWHAE (605066) genes. Lissencephaly caused by mutations in the PAFAH1B1 gene is also called 'isolated' lissencephaly to distinguish it from the accompanying features of MDLS.

Clinical Features

Chong et al. (1996) reported a patient with isolated lissencephaly who had a mutation in the LIS1 gene (601545.0001; see MOLECULAR GENETICS). Leventer et al. (2001) described the patient reported by Chong et al. (1996) in greater detail. From infancy, the patient showed developmental delay, myoclonic jerks and spasms, seizures, generalized hypotonia, microcephaly, and dysmorphic facies. Brain MRI revealed moderate agyria in the occipital lobes transitioning to pachygyria anteriorly as well as flattening of the corpus callosum and mild dilation of the posterior horns of the lateral ventricles. The patient developed progressive spasticity and died of sepsis at age 4 years.

Leventer et al. (2001) reported a patient with generalized hypotonia and poor visual and social interaction who later developed complex partial seizures. MRI revealed moderate pachygyria, consistent with isolated lissencephaly sequence, that was most severe in the parietooccipital regions, hypoplasia of the rostral corpus callosum, and mild dilation of the posterior horns of the lateral ventricles. At age 4 years, the patient could feed himself and understand simple commands.

Leventer et al. (2001) reported a girl with isolated lissencephaly sequence who had global developmental delay and hypotonia and later developed myoclonic jerks, absence seizures, and febrile seizures. Brain MRI showed moderate generalized pachygyria that was most severe in the occipitoparietal regions, hypoplasia of the cerebellar vermis, hypoplasia of the rostral corpus callosum, and mild dilation of the lateral ventricles. At age 12 years, she walked with assistance, was toilet-trained, and had limited communication skills.

Leventer et al. (2001) reported a boy with speech and walking delay and strabismus who later developed complex partial seizures. Brain MRI showed moderate pachygyria restricted to the occipital and posterior parietal lobes, consistent with isolated lissencephaly sequence. At age 6 years, the boy attended a developmental preschool, played sports, and was found to have an IQ of 100.

Saillour et al. (2009) found that 40 of 63 patients with posterior predominant lissencephaly had a LIS1 mutation or deletion, including 1 patient with somatic mosaicism for a nonsense mutation. Most patients with LIS1 mutations had posterior agyria and anterior pachygyria (55.3%). Diffuse agyria was observed in 9 (23.7%) patients, and posterior predominant pachygyria was seen in 6 (15.8%). Twenty-two (64.7%) of 34 patients had corpus callosum abnormalities, with either thinning or abnormal thickening. Prominent perivascular spaces were seen in 23 (67.4%) cases and enlarged ventricles in 28 (73.7%). The degree of neuromotor impairment was in accordance with the severity of lissencephaly, with a high incidence of tetraplegia (61.1%). However, the severity of epilepsy could not show the same reliability, because 82.9% had early onset of seizures, and 48.7% had seizures more often than daily. Mutation type and location did not predict the severity of LIS1-related lissencephaly. In comparison, patients without LIS1 mutation tended to have less severe lissencephaly and no additional brain abnormalities.

Subcortical Laminar Heterotopia

Pilz et al. (1999) reported a boy with subcortical band heterotopia who had a mutation in the LIS1 gene (601545.0004, see MOLECULAR GENETICS). Leventer et al. (2001) described the boy reported by Pilz et al. (1999) in greater detail. As a child, he had mild global developmental delay and complex partial seizures. MRI showed posterior subcortical band heterotopia and mild dilation of the posterior horns of the lateral ventricles. At age 23 years, he worked as an unskilled manual laborer and enjoyed normal activities, although seizures remained a problem. Leventer et al. (2001) suggested that the milder phenotype may be due to somatic mosaicism.

In 2 male patients with subcortical band heterotopia, Sicca et al. (2003) identified somatic mosaicism for mutations in the LIS1 gene: arg241 to pro (R241P; 601545.0008) and arg8 to ter (R8X; 601545.0009), respectively. The mutant alleles were present in 18% and 24% of lymphocyte DNA and 21% and 31% of hair root DNA, respectively. The patients had mental retardation, seizures, and posterior SBH on brain MRI, but the phenotype was not as severe as full-blown lissencephaly. In a male patient with lissencephaly, Sicca et al. (2003) identified the R8X mutation. This third patient did not show somatic mosaicism and had a very severe phenotype. The authors noted that these examples suggested that somatic mosaicism results in a less severe phenotype.

Molecular Genetics

The majority of patients with classic lissencephaly have deletions in the LIS1 gene. Cardoso et al. (2002) found that 65 of 98 patients with isolated lissencephaly or MLDS had large deletions of the LIS1 gene. Among 41 intragenic LIS1 mutations, 36 (88%) resulted in a truncated or internally deletion protein. Only 5 (12%) of 41 were missense mutations. mutations were found in only 12% (5 of 41). Mutations occurred throughout the gene except for exon 7.

In 3 patients with isolated lissencephaly sequence in whom no deletions of 17p were detectable by FISH, Chong et al. (1996) identified 3 mutations in the PAFAH1B1 gene (601545.0001-601545.0003). See also Lo Nigro et al. (1997). Leventer et al. (2001) reported 3 novel mutations in the PAFAH1B1 gene in patients with ILS (601545.0005-601545.0007). In a patient with subcortical laminar heterotopia, Pilz et al. (1999) identified a mutation in the PAFAH1B1 gene (601545.0004).

Cardoso et al. (2003) completed a physical and transcriptional map of the 17p13.3 region from LIS1 to the telomere. Using FISH, they mapped the deletion size in 19 children with ILS, 11 children with MDS, and 4 children with 17p13.3 deletions not involving LIS1. They showed that the critical region that differentiates ILS from MDS at the molecular level can be reduced to 400 kb. Using somatic cell hybrids from selected patients, the authors identified 8 genes that are consistently deleted in patients classified as having MDS. These genes include ABR (600365), 14-3-3-epsilon (605066), CRK (164762), MYO1C (606538), SKIP (INPP5K; 607875), PITPNA (600174), SCARF1, RILP, PRP8 (607300), and SERPINF1 (172860). In addition, deletion of the genes CRK and 14-3-3-epsilon delineates patients with the most severe lissencephaly grade. On the basis of recent functional data and the creation of a mouse model suggesting a role for 14-3-3-epsilon in cortical development, Cardoso et al. (2003) suggested that deletion of 1 or both of these genes in combination with deletion of LIS1 may contribute to the more severe form of lissencephaly seen only in patients with Miller-Dieker syndrome.

Mei et al. (2008) identified mutations in the LIS1 gene in 20 (44%) of 45 patients with isolated lissencephaly showing a posterior to anterior gradient. In 19 (76%) of 25 patients in whom FISH and direct sequencing had failed to detect mutations, MLPA analysis identified 18 small genomic deletions and 1 duplication. Overall, small genomic deletions/duplications represented 49% of all LIS1 alterations identified, and LIS1 involvement was demonstrated in 39 (87%) of 45 patients. Breakpoint characterization in 5 patients suggested that Alu-mediated recombination is a major molecular mechanism underlying LIS1 deletions. Mei et al. (2008) noted the high diagnostic yield with MLPA.

Among 63 patients with posterior predominant lissencephaly, Saillour et al. (2009) identified 40 with LIS1 gene defects. There were 8 small deletions and 31 heterozygous LIS1 mutations, including 12 nonsense, 8 frameshift, 6 missense, and 5 splicing defects. The mutations were found scattered throughout the gene, except in exons 3 and 9, and all were confirmed to be de novo. One patient had a somatic truncating mutation present in 30% of the blood, but other tissues were not available for testing.

Using multiplex ligation-dependent probe amplification (MLPA) analysis, Haverfield et al. (2009) identified 12 deletions and 6 duplications involving the LIS1 gene in 18 (35%) of 52 patients with an anterior-to-posterior lissencephaly gradient in whom no molecular defect had previously been identified. The majority of patients with LIS1 deletions or duplications had grade 3 lissencephaly. Most deletions and duplications were scattered within the gene, but several deletions included genes flanking LIS1, such as HIC1 (603825), or only included noncoding putative upstream regulatory elements of LIS1. Haverfield et al. (2009) suggested that genetic testing for isolated lissencephaly should include both mutation and deletion/duplication analysis of the LIS1 gene.

Genotype/Phenotype Correlations

By direct DNA sequencing of the LIS1 and DCX genes in 25 children with sporadic lissencephaly and no deletion of the LIS1 gene by FISH analysis, Pilz et al. (1998) identified LIS1 mutations in 8 (32%) patients and DCX mutations in 5 (20%). All the LIS1 mutations were de novo: 6 were truncating, and 2 were splice site mutations. Two additional patients were found to have de novo LIS1 rearrangements by Southern blot analysis. Phenotypic studies showed that those with LIS1 mutation had more severe lissencephaly over the parietal and occipital brain regions, whereas those with DCX mutations had the reverse gradient, with more severe lissencephaly over the frontal regions. All DCX mutation carriers also had mild hypoplasia and upward rotation of the cerebellar vermis, but these changes were only seen in about 20% of patients with LIS1 mutations. Overall, mutations of LIS1 or DCX were found in 60% of patients in this study. Combined with the previously observed frequency of LIS1 mutations detected by FISH, Pilz et al. (1998) concluded that these 2 genes account for about 76% of sporadic lissencephaly.

Dobyns et al. (1999) compared the phenotype of 48 children with lissencephaly, including 12 with MDLS with large deletions including LIS1, 24 with isolated lissencephaly sequence caused by smaller LIS1 deletions or mutations, and 12 with DCX mutations. There were consistent differences in the gyral patterns, with LIS1 mutations associated with more severe malformations posteriorly, and DCX mutations associated with more severe malformations anteriorly. In addition, hypoplasia of the cerebellar vermis was more common in those with DCX mutations.

Fogli et al. (1999) reported 7 patients with lissencephaly-1 and a heterozygous mutation in the LIS1 gene, 6 with a truncating mutation and 1 with a splice site mutation resulting in the skipping of exon 4. Western blot analysis on lymphoblastoid cells of 2 patients with truncating mutations showed that the mutated allele did not produce a detectable amount of the LIS1 protein, whereas analysis of fibroblasts from the patient with the splice site mutation showed partial protein synthesis. Patients with the truncating mutations had severe developmental delay with early-onset seizures, hypotonia, and spastic quadriparesis; the patient with the splice site mutation had a less severe clinical course. Fogli et al. (1999) noted that intracellular dosage of the LIS1 protein is important to the neuronal migration process.

The mutations in the LIS1 and DCX genes causing classic lissencephaly (formerly type I) are thought to occur during corticogenesis and operate on radial migratory pathways. Viot et al. (2004) noted that heterozygous mutations in the LIS1 gene and hemizygous mutations in the DCX gene had been thought to produce a similar histologic pattern. They reported detailed neuropathologic studies in 2 unrelated fetuses, 1 with a mutation in the LIS1 gene and the other with a mutation in the DCX gene. In the fetus with the LIS1 mutation, the cortical ribbon displayed a characteristic inverted organization, also called '4-layered cortex,' whereas in the fetus with the DCX mutation, the cortex displayed a roughly ordered '6-layered' lamination. Viot et al. (2004) hypothesized that mutations in these 2 genes may not affect the same neuronal arrangement in the neocortex.

Forman et al. (2005) proposed a classification of lissencephaly based on the neuropathologic findings of 16 patients. Six had LIS1 deletions, 2 had DCX mutations, 2 had ARX mutations, and 6 had no defined genetic defect, One of the patients had SBH consistent with a DCX mutation. The cortex was thickened in all cases. Those with LIS1 and DCX mutations had 4-layer involvement, with more posterior and anterior involvement, respectively. Brains with ARX mutations showed 3-layer cortical involvement. Two of 5 patients with no known genetic defect showed a fourth type of histopathology characterized by a 2-layered cortex; these brains also had profound brainstem and cerebellar abnormalities. Forman et al. (2005) proposed that LIS1- and DCX-related lissencephaly be termed 'classic' lissencephaly and that ARX-related and the other entities with hindbrain involvement be termed 'variant' lissencephaly.

Uyanik et al. (2007) identified 14 novel and 7 previously described LIS1 mutations in 21 unrelated patients, including 18 with lissencephaly-1, 1 with subcortical band heterotopia, and 2 with lissencephaly with cerebellar hypoplasia. There were 9 truncating mutations, 6 splice site mutations, 5 missense mutations, and an in-frame deletion. Somatic mosaicism was assumed in 3 patients with partial subcortical band heterotopia or mild pachygyria. Uyanik et al. (2007) concluded that the severity of the phenotype is independent of the type of mutation and its site within the coding region of the LIS1 gene.

In a retrospective review of MRI scans from 111 patients with lissencephaly, Jissendi-Tchofo et al. (2009) found a correlation between the extent of cerebral lissencephaly and midbrain-hindbrain involvement. However, most patients with LIS1 had normal midbrain-hindbrain findings, and those with midbrain-hindbrain involvement tended to have 'variant' lissencephaly, as defined by Forman et al. (2005).

Animal Model

Hirotsune et al. (1998) produced 3 different mutant alleles in the mouse Pafah1b1 gene. Homozygous-null mice died early in embryogenesis soon after implantation. Mice with 1 inactive allele displayed cortical, hippocampal, and olfactory bulb disorganization resulting from delayed neuronal migration by a cell-autonomous neuronal pathway. Mice with further reduction of Pafah1b1 activity displayed more severe brain disorganization as well as cerebellar defects. The results demonstrated an essential, dosage-sensitive neuronal-specific role for Pafah1b1 in neuronal migration throughout the brain, and an essential role in early embryonic development. The phenotypes observed were distinct from those of other mouse mutants with neuronal migration defects, suggesting that Pafah1b1 participates in a novel pathway for neuronal migration.

Cahana et al. (2001) deleted the first coding exon from the mouse Lis1 gene. The deletion resulted in a shorter protein that initiated from the second methionine, a unique situation because most LIS1 mutations result in a null allele. This mutation mimicked a mutation described in 1 lissencephaly patient with a milder phenotype (Fogli et al., 1999). Homozygotes were early lethal, although heterozygotes were viable and fertile. The morphology of cortical neurons and radial glia was aberrant in the developing cortex, and the neurons migrated more slowly. This was the first demonstration of a cellular abnormality in the migrating neurons after Lis1 mutation. Moreover, cortical plate splitting and thalamocortical innervation were also abnormal. Biochemically, the mutant protein was not capable of dimerization, and enzymatic activity was elevated in the embryos, thus a demonstration of the in vivo role of LIS1 as a subunit of platelet-activating factor acetylhydrolase.

Yamada et al. (2009) demonstrated that inhibition or knockdown of calpains (see, e.g., CAPN1; 114220) protected the Lis1 protein from proteolysis in Lis1 +/- mouse embryonic fibroblasts. Increased protein levels rescued the aberrant distribution of cytoplasmic dynein and mitochondria observed in Lis1 +/- cells, consistent with an improvement in function. Calpain inhibitors also improved neuronal migration of Lis1 +/- cerebellar granular neurons. Intraperitoneal injection of the calpain inhibitor to pregnant Lis1 +/- dams rescued apoptotic neuronal cell death and partially rescued neuronal migration defects in Lis1 +/- offspring. Furthermore, in utero knockdown of calpain by short hairpin RNA rescued defective cortical layering in Lis1 +/- mice. Yamada et al. (2009) suggested that LIS1 is specifically degraded by calpain and that calpain inhibition could be a potential therapeutic intervention for lissencephaly due to haploinsufficiency of LIS1.