Skip to main content
NCBI home page
The American Journal of Pathology logoLink to The American Journal of Pathology
. 2004 Sep;165(3):775–784. doi: 10.1016/S0002-9440(10)63340-8

Lis1 Is Necessary for Normal Non-Radial Migration of Inhibitory Interneurons

Matthew F McManus *, Ilya M Nasrallah *, MacLean M Pancoast †, Anthony Wynshaw-Boris ‡, Jeffrey A Golden *‡
PMCID: PMC2336110  PMID: 15331402

Abstract

Type I lissencephaly is a central nervous system (CNS) malformation characterized by mental retardation and epilepsy. These clinical features suggest a deficit in inhibitory neurons may, in part, underlie the pathogenesis of this disorder. Mutations in, or deletions of, LIS1 are the most commonly recognized genetic anomaly associated with type I lissencephaly. The pathogenesis of type I lissencephaly is believed to be a defect in radial neuronal migration, a process requiring LIS1. In contrast the inhibitory neurons migrate non-radially from the basal forebrain to the neocortex and hippocampus. Given that Lis1 is expressed in all neurons, we hypothesized that Lis1 also functions in non-radial migrating inhibitory neurons. To test this hypothesis we used a combination of in vivo and in vitro studies with Lis1 mutant mice and found non-radial cell migration is also affected. Our data indicate Lis1 is required for normal non-radial neural migration and that the Lis1 requirement is primarily cell autonomous, although a small cell non-autonomous effect could not be excluded. These data indicate inhibitory neuron migration is slowed but not absent, similar to that found for radial cell migration. We propose that the defect in non-radial cell migration is likely to contribute to the clinical phenotype observed in individuals with a LIS1 mutation.

Lissencephaly is a central nervous system (CNS) malformation characterized by the loss of normal gyri and sulci on the surface of the brain and a thickened and disorganized cerebral cortex. Patients with this malformation exhibit variable degrees of mental retardation and epilepsy, along with other neurological signs.1 Lissencephaly has traditionally been separated into two types, type I and type II, or classical and cobblestone, respectively. Only type I lissencephalies have been clearly linked to a deficit in cell migration, in contrast, type II lissencephalies are primarily an over-migration defect. To date, mutations in four genes have been causally associated with type I lissencephaly. The first identified was LIS1.2 Heterozygous deletions of 17p13.3, containing the LIS1 gene, or point mutations therein can lead to either isolated lissencephaly sequence (ILS) or the Miller-Dieker syndrome (MDS), MDS having a more severe phenotype than ILS and is likely the result of a contiguous gene deletion syndrome involving the linked gene 14–3-3ε.3–5 Mutations in doublecortin (DCX)6,7 and ARX,8 both X-linked genes, also lead to type I lissencephaly in males and distinct malformations in females. Finally, mutations in REELIN have been linked to AR lissencephaly.9 Data from patients and animal models indicate that each of these mutations is associated with at least a radial cell migration defect.

LIS1 mutations are the most commonly identified genetic basis for lissencephaly. The LIS1 gene product is a 45-kd, ubiquitously expressed protein that is found in a particularly high concentration in neurons.10,11 LIS1 is the non-catalytic subunit of platelet-activating factor acetyl hydrolase (PAFAH) 1B.2 Homologues of LIS1 are known to exist in fungi,12 yeast,13 Drosophila,14 and mice10,15 and function to stabilize cytoplasmic dynein and microtubules.11,16,17 This results in Lis1 or its homologues playing an important role in cytoplasmic dynamics important for cell division and movement. In yeast, the LIS1 homologous protein PAC1 (33%) is required for segregation of chromosomes during mitosis and for nuclear orientation.13 Similarly, in Aspergillus nidulans, NUDF, which is 42% identical to LIS1, is necessary for the distribution and migration of nuclei.18 Drosophila Lis1, dLis1, is 70% identical to LIS1 and is necessary for the normal development of egg chambers and for germline cell division.14 In mice, LIS1 is localized to the centrosome19 and is involved in interkinetic nuclear migration, neuroblast proliferation, and programmed cell death of cortical ventricular zone neuroblasts.20,21 Additionally, mice heterozygous for a null allele of Lis1 exhibit abnormal cerebral cortical, hippocampal, cerebellar, and olfactory bulb development, as well as impaired radial neuronal migration and hippocampal electrophysiological abnormalities.10,21,22 In vitro experiments have shown that cerebellar granule cells containing only one functional copy of Lis1 exhibit cell autonomous migration defects.10,21

Radial migration from the ventricular zone out to the surface of the developing brain, perhaps the best-characterized migratory pathway for neurons,23,24 is deficient in Lis1 heterozygous mice.10 A second pathway of migration, perpendicular to radial migration, has more recently been identified. This tangential or, more accurately, non-radial cell migration (NRCM) pathway has been described at nearly all levels of the developing nervous system,25–27 and plays a significant role in the migration of GABAergic interneurons from the ganglionic eminence to the cerebral cortex and hippocampus.28–30 Significantly, the loss or compromise of inhibitory interneuron function has been associated with human epileptogenesis.31,32

Lis1 is expressed in all neurons, and defects in inhibitory neurons are a cause of epilepsy (one of the clinical features in patients with lissencephaly). We, therefore, hypothesized that mutations in LIS1 would similarly affect NRCM. To investigate this hypothesis, we compare the characteristics of GABAergic NRCM in Lis1 +/− and wild-type mice and have studied the number of inhibitory interneurons in several patients with MDS. Our data indicate that Lis1 is required for normal NRCM in mice, and that there is both a cell autonomous and a smaller cell non-autonomous effect on NRCM. We propose that the NRCM defect contributes to the CNS anomalies and clinical manifestations in patients with lissencephaly associated with a LIS1 deletion or mutation.

Materials and Methods

Mouse Strains and Genotyping

The background of all Lis1 mice used was a mixture of 129SvEvTac and NIH Black Swiss. All mice have been backcrossed to Black Swiss at least eight generations, making this the predominant background. Timed-pregnant mice were considered to be embryonic day 0.5 (E0.5) on the morning a vaginal plug was found, in addition all embryos were morphologically staged.33 Genotyping was conducted using embryo tongues and the following primers for pcr: null and wild-type allele forward 5′-GTGTGGGATTATGAGACTGG-3′; Lis1-Neo (null) allele reverse 5′-GATCTCTCGTGGGATCATTG-3′; and Lis1-wt wild-type control reverse 5′-CCAGATGGTTTAAGTATGAGTC-3′ (positive control for the wild-type allele).

Tissue Preparation, Histology, and Immunohistochemistry

Timed-pregnant Lis1 mice were bred in our animal facility. All animal breeding, handling, and experimental procedures were approved by the institution animal care and use committee. Embryos were collected in ice-cold Hanks Balanced Salt Solution/Streptomycin/Penicillin (HanksSP). For immunohistochemistry, brains were dissected in ice cold HanksSP, and after the meninges removed, they were fixed in 4% paraformaldehyde (PFA) or 4% PFA/0.25% gluteraldehyde (for anti-GABA staining) in phosphate-buffered saline (PBS), embedded in 2% agarose (SeaKem LE) and vibratome sectioned at 50 μm.

Immunofluorescence was performed as previously described.34 Briefly, sections were blocked in 10% normal goat serum for 1 hour at room temperature (RT) with 0.1% Triton-100x. Primary antibodies used were diluted in PBS and included: anti-calretinin, 1:2000 (Swant, Bellonzona, Switzerland); and anti-GABA, 1:1500 (Sigma, St. Louis, MO). Sections were incubated in 2 to 10% normal goat serum with primary antibody and 0.01% to 0.3% Triton-100x from 1.5 days (anti-GABA) to 3 days (anti-calretinin and -calbindin) at 4°C. Secondary antibodies used (2 to 10% normal goat serum, 1 to 4 hours at RT) were biotinylated goat anti-rabbit; biotinylated goat anti-rat and biotinylated goat anti-mouse (Jackson ImmunoResearch, West Grove, PA). Biotinylated secondary antibodies were subsequently incubated with streptavidin-conjugated Cy3 (1:500, Jackson ImmunoResearch). Nuclei were counterstained with DAPI (1:1000, Molecular Probes, Eugene, OR). For immunohistochemistry, sections were incubated with biotinylated secondary antibodies and detected according to the Vector ABC detection system standard protocol (Vector Laboratories, Burlingame, CA) and counterstained with hematoxylin and eosin. Images were obtained using a Leica DMR microscope equipped with epifluorescence and either a Hamamatsu C5180 camera or a Hamamatsu ORCA-ER C4742–95 and OpenLab 3.0.8 or a Nikon Eclipse TE300 equipped with a Hamamatsu C4742–95 using Phase 3 Imaging software for image acquisition. Images were then analyzed using either Phase 3 Imaging or Adobe Photoshop 7.0 software. All distances were calculated, using Phase 3 Imaging software, from the cortico-striatal notch in a straight-line distance to the center of the immunolabeled cells’ somata. Animals from three litters were used for GABAergic and calretinin studies. A minimum of seven sections were counted and averaged for each animal.

Cell Migration Assay

For the in vitro migration studies, embryonic mouse brains were dissected as described above, embedded in 2% low-melt agarose (Fischer, Morris Plains, NJ), sectioned on a vibratome in ice-cold HanksSP at 250 μm and transferred onto 12-mm Millicell-CM cell culture inserts (Millipore, Billerica, MA) in 45%DMEM/45% F-12/10% fetal calf serum (HiClone, Logan, UT)/1X PS/6.5g/L glucose (DFS medium). For initial migration assays (no transplantation), DiI crystals (Molecular Probes) of approximately equal size were placed at the dorsal boundary of the striatum, and the slice was surrounded by matrigel (BD Bioscience, Bedford, MA). Cultures were then incubated for 2 hours in DFS before being change to DMEM/1X N2 medium supplement (Invitrogen, Carlsbad, CA)/1X PS/6.5g/L glucose (DM) and cultured for 48 to 60 hours at 37°C, 5% CO2 before fixing in 4% PFA for 2 hours at 4°C. Cortical transplantations were performed by incising from the cortical notch ventro-laterally to the piriform cortex after vibratome sectioning. Cortices and ganglionic eminences were then re-apposed randomly and cultured as above (see Figure 6 below). All migration analyses were conducted without knowledge of the animal’s genotype. For DiI implant studies without transplantation, a total of 35 animals were taken from five separate litters. For transplantation studies, a total of 36 animals from six separate litters were used. All distances were calculated, using Phase 3 Imaging software, from the dorso-lateral edge of the DiI crystal in a straight-line distance to the center of the labeled cells’ somata. Comparisons were made using two-tailed Student’s t-tests.

Figure 6.

Figure 6

Open in a new tab

Schematic representation of E14.5 transplant migration assay. Brain slices were obtained as for standard slice culture. Before DiI crystal implantation, the neocortex was excised from the ventral telencephalon along dotted line, and re-apposed to either a homo- or heterogenic ventral telencephalon (source of migratory, GE-derived cells) (blue and tan denote different genotypes, they can be interchanged such that if blue represents mutant and thus the source of GE cells in the image the figure would show mutant cells migrating on wild-type substrate. The reverse genotype, ie, tan being mutant, would give wild-type cells migrating on mutant substrate. For those slices where blue is re-apposed with blue or tan with tan, the resulting section would mutant with mutant or wild-type with wild-type giving all possible combinations; see Figure 7). DiI crystals (red sphere) were then implanted, and slices were cultured and assayed as cell migration (far right).

Results

Inhibitory Interneuron Migration Is Abnormal in Lis1 Mutant Mice

Radial cell migration is slowed but not absent in Lis1 +/− mice.35 We hypothesized a similar defect would exist for NRCM. To determine whether an in vivo defect exists in NRCM in the Lis1 +/− mice, we studied the expression of inhibitory interneuron markers at various developmental stages spanning the period during which NRCM occurs. Inhibitory interneurons were identified by the expression of several markers including GABA and calretinin. While GABA labels the total set of interneurons, calretinin only labels a subset of interneurons.36,37 We found GABA and calretinin labeling gave proportionately similar results despite calretinin only labeling a subset of the interneuron population (see Figures 1 and 2). Although calretinin represents only a subpopulation of interneurons, calretinin labeling of individual cells is much clearer than GABA labeling (Figure 1). As a result, calretinin was used in many of the subsequent studies to represent the interneuron population, although recognizably not the entire population. While GABAergic cells can be found leaving the ganglionic eminence (GE) in transit to the cortex as early as E1329 in both wild-type and mutant animals (data not shown), we focused on E14 to E14.5 animals to allow cells time to migrate sufficiently far to assess differences between mutants and wild types, yet before the leading cells migrated past the dorsal cortex (by E15.5, data not shown), which would complicate quantification and interpretation. At E14.5, inhibitory interneurons orientation non-radially with morphologies typical of migrating cells are observed from the GE into and through the neocortex in both wild-type and mutant embryos (Figure 1). We focused exclusively on non-radially migrating cells in the intermediate zone and subventricular zone, as cells migrating in layer 1 (molecular layer) are a mixed population of glutamatergic, calretinin-positive neurons (Cajal-Reitzius cells), and GABAergic interneurons.38,39

Figure 1.

Figure 1

Open in a new tab

Inhibitory interneurons migrate a shorter distance in E14.5 Lis1 +/− animals. Immunohistochemistry for calretinin (a, b, d, e) and GABA (c and f) on coronal sections of the telencephalon in E14.5 wild-type (a, b, c) and Lis1 +/− (d, e, f) embryos. GABA (c)-labeled and calretinin (a)-labeled neurons have migrated farther in wild-type embryos when compared to the mutant mice (f for GABA, d for calretinin) (see Figure 2 for quantitation). At higher power, taken from boxed areas in a and d, the leading edge cells with morphologies and orientations of non-radially migration interneurons can be seen in the dorso-lateral cortex of the wild-type (b), but only in medio-lateral cortex of the Lis1 +/− animals, indicating that interneurons have not migrated as far in the mutant animals at similar points in development.

Figure 2.

Figure 2

Open in a new tab

Total distance and percentage of distance from GE to dorsum of cortex migrated in Lis1 +/− and wild-type E14.5 embryos. The averaged distance migrated by all calretinin-positive cells past the cortico-striatal notch is significantly greater in wild-type animals by E14.5 (A). When only the leading GABAergic cells are examined, the differences between mutant and wild-type animals is greater in magnitude, as are the total distances traveled (B). Comparing the percentage of the distance from the cortico-striatal notch to the dorsum of the cortex in both cases reveals that Lis1 +/− animals also traverse a significantly smaller portion of the developing brain than do wild-type age-matched controls (C and D)

To determine whether mutant cells lagged behind wild-type cells, we quantified the average distance traveled past the cortico-striatal notch both by all calretinin-positive cells and also by the leading GABAergic cells. In all cases we assessed only those cells with a non-radial orientation and excluded those in the marginal zone. The average distance traveled by all calretinin-positive cells past the border to the GE in E14.5 animals was calculated for both the Lis1 +/− and the wild-type animals (245 μm versus 564 μm, respectively). Similarly, the distance traveled by the leading 12 cells, representing the cell migrating front, was calculated for E14 and E14.5 Lis1 +/− and Lis +/+ littermates (629 μm versus 1045 μm, respectively, in the E14.5 animals). These differences were significant at both ages (Figure 2, A, B, and data not shown). As the level of section varies slightly within an animal and differences in total brain size were possible between heterozygous and wild-type animals, the percent distance traveled from the notch to the dorsal-most aspect of the neocortex was also calculated to normalize for differences in forebrain size. Using this measure, mutant cells still progressed a significantly shorter distance than did wild-type cells (Figure 2, C and D). These findings indicate that differences in distance traveled are not due to variations in brain size between or within genotypes.

Having data from two time points in development allowed us to make an approximate calculation of the average migration speed. Unfortunately, the staging at the time points we used in mice is imprecise. We used morphological staging to separate out younger (E14) and older embryos (E14.5). Younger embryos harvested at E14 were defined as those with identifiable digits with interdigital webs whereas those harvested at E14.5 and used in these studies had no interdigital webs but a radial ray of digits on the hindlimb.33 While we recognize this is not a perfect system, it allowed us to separate embryos and calculate an approximate time difference for speed calculations. Using these criteria, we calculated the change in the average distance traveled for the leading 12 GABAergic cells and divided by the average time of 12 hours from E14 and E14.5 to generate approximate speeds of cell migration. Our data indicate wild-type cells move at approximately 43.8 μm/hr, while mutants progress an average of only 23 μm/hour. Due to the difficulty in staging these embryos, these numbers should be used as approximate values but show a clear difference between wild-type and Lis1 +/− animals.

NRCM Is Slowed in Lis1 Mutant Mice in Vitro

The in vivo data are consistent with a decrease in the rate of cell migration; alternatively, a delay in initiation of migration from the progenitor zone could account for the observed in vivo data. To determine whether non-radially migrating cells in Lis1 +/− mice migrate more slowly than wild-type cells, inhibitory interneurons were labeled in vitro and assayed for distance migrated over time. GABAergic cells have already begun migrating in both wild-type and mutant animals by E14.5 (see above), placing this time point approximately midway in the migratory process. Therefore, any observed differences in labeled cell positions should be due to varied speeds of migration and not to a delay in the start of migration.

To follow the migration of inhibitory neurons from the GE in vitro, the LGE of coronal forebrain slices were implanted with DiI crystals to label migrating cells. After 48 to 60 hours, cells are observed migrating out from the site of crystal implantation in both wild-type (Figure 3a) and mutant (Figure 3b) cultures, with morphologies typical of migrating cells (Figure 3c). As has been previously reported, wild-type and mutant cells labeled with DiI also co-label with markers of GABAergic neurons30 (Figure 3, c to e).

Figure 3.

Figure 3

Open in a new tab

Migration of DiI-labeled inhibitory interneurons is impaired in Lis1 +/− E14.5 embryonic slice culture. Representative examples of E14.5 embryonic brain slice cultures with DiI crystal implants in wild-type (a) and Lis1 +/− animals (b). Arrows in (a) and (b) indicate approximate average distance migrated dorsally from the DiI crystal edge by all labeled (arrowhead) or the leading 25 (arrow) cells (for quantification, see Figure 5). Cells labeled with DiI displaying characteristic morphologies of non-radially migrating cells (c) also label with antibody against a marker of inhibitory interneurons, calretinin (d, and merged image in e).

To compare NRCM in wild types to that in mutants, three measures were used: average distance traveled by all cells; average distance traveled by the leading cells; and the percent distance from the cortico-striatal notch to the dorsum of the cortex. The first metric examined was the average distance migrated by all cells in a particular genotype. Wild-type cells moved significantly farther on average than did mutant cells (840 μm versus 667 μm for mutants, P < 4 × 10−11; Figure 4A). Cells continue to migrate from the GE during the days in culture, and so averages of all labeled cells include the distance traveled both by those that have just begun and those that have had several days in which to migrate. The earlier labeled cells, having had more time to migrate, should also be located farthest from the DiI crystal. To examine this subpopulation of cells that should most clearly show any differences in migrational ability, the farthest 25 cells in each condition were also examined. When comparing the mutant and wild-type-labeled cells, the wild-type cells were again observed to have migrated significantly farther than the mutants (1887 μm versus 1330 μm for mutants, P < 4 × 10−14; Figure 4B). In addition to assessing the average distance migrated by the whole or a particular subset of cells, one can also bin the number of cells by distance migrated from the edge of the DiI crystal (Figure 5) to compare the distribution of distances migrated by each genotype. While there is no significant difference in number of cells found in the bins nearest the DiI crystal, a significantly higher number of wild-type cells are found at distances farther from the crystal edge, with the farthest binned distances containing no mutant cells (Figure 5).

Figure 4.

Figure 4

Open in a new tab

Average distance migrated by cells from Lis1 +/− mice in E14.5 slice cultures is significantly less than by wild-type cells. The average distance migrated by all DiI-labeled cells dorsally from the DiI crystal edge is statistically less in Lis1 +/− mice than in wild-type animals, at E14.5 (A). Similarly, when comparing the average distance traveled by the leading 25 cells, Lis1 +/− animals exhibit a migratory defect relative to wild-type animals (B)

Figure 5.

Figure 5

Open in a new tab

Distribution of distance migrated by mutant and wild-type cells in slice culture. Cells from both wild-type (white bars) and Lis1 +/− (black bars) eminences are placed into bins by distances of 100 μm migrated from the DiI crystal edge. The distributions of both populations are similarly shaped and approximately normal. There is no statistical difference in the number of cells at distances between 300 and 500 μm, inclusive. At all farther distances there is a statistically greater number of wild-type than mutant cells, and at the farthest distances only wild-type cells are found (beyond 2 mm past the DiI crystal). Bins less than 300 μm cannot be considered as the halo from the DiI crystal often made identification of individual cells impossible. Comparison within bins were made by two-tailed Student’s t-test; *, denotes P < 0.01

Lis1 Cell Autonomous and Non-Autonomous NRCM Effects

A cell autonomous radial migration defect in Lis1 mutant animals leads to a disorganized cortex, even in the presence of normal radial glial projections.10,21 Non-radially migrating inhibitory interneurons traverse the neocortex en route to their final locations in the forebrain, and some populations of these cells are known to have specific substrate requirements.40,41 We hypothesize that the NRCM defect observed in Lis1 +/− animals results from both cell autonomous and cell non-autonomous factors, the latter as a result of the disorganized cortical axonal migratory substrate.20 To test this hypothesis, GEs from wild-type and mutant animals were excised from coronal brain sections and apposed to similarly excised cortices (ie, migratory substrate) from both wild-type and mutant sections and cultured as above, giving four genetic combinations (Figure 6). In this way, cell autonomous effects are observed by comparing wild-type and mutant cell migration on genotypically identical cortical substrates. Similarly, cell non-autonomous effects were assessed by comparing migration of homogenic cells (either wild-type or mutant) on heterogenic cortical substrates (Figure 7a to d).

Figure 7.

Figure 7

Open in a new tab

Migratory defect in Lis1 +/− animals is both cell autonomous and non-autonomous. Representative examples of E14.5 embryonic brain slice transplant cultures with DiI crystal implants in cultures with either wild-type GEs (a and b) or Lis1 +/− GEs (c and d) apposed to either wild-type (a and c) or Lis1 +/− cortices (b and d). Arrows in (a–d) indicate approximate average distance migrated dorsally from the DiI crystal edge by all labeled (arrowhead) or the leading 25 (arrow) cells (for quantification, see Figure 8). (GE, ganglionic eminence; cortex, cortical substrate, see Figure 6)

The average distance migrated by all cells in each combination of genotypes demonstrated a clear cell autonomous NRCM defect. Wild-type GE cells migrate significantly farther than mutant cells, without regard to the migratory substrate’s genotype (Figure 8). No significant difference is observed between cortical substrate types within GE genotypes, indicating that no cell non-autonomous effect is seen using this metric (Figure 8). In contrast, when the farthest 25 cells, those having had the most time to migrate and therefore most clearly exhibit any migratory differences, are examined, the cell autonomous effect is again significant (Figure 8). In addition, a significant cell non-autonomous effect is also seen between wild-type and mutant cortex serving as a substrate for the wild-type GEs, with the wild-type cortex supporting farther migration than the mutant substrate (Figure 8). In summary, a cell autonomous effect is observed in all conditions, whereas the substrate-dependent cell non-autonomous effect was only observed between mutant and wild-type cortices using wild-type eminences.

Figure 8.

Figure 8

Open in a new tab

Quantification of cell autonomous and non-autonomous migratory defect in Lis1 +/− mice in E14.5 slice cultures. Migratory cells derived from wild-type GEs migrated farther than those from Lis1 GEs, regardless of the cortical migratory substrate’s genotype, as assessed by the averaged distance migrated dorsally by all (A) or the leading 25 (B) DiI-labeled cells. This indicates a cell autonomous defect in migratory ability in the Lis1 +/− animals. When comparing how well cells migrated on wild-type or mutant cortex, a cell non-autonomous effect is evidenced by the significantly greater distance traveled by the leading 25 wild-type cells in wild-type rather than Lis1 +/− cortex (B).

Discussion

Haploinsufficiency of LIS1 is known to result in type I lissencephaly in humans4 and in a cell autonomous defect in radial migration in mice.10,21 The radial migration defect has been proposed as the mechanism leading to various organizational and electrophysiological abnormalities in both mice and humans.10,21,22 Here we report on a developmental study of the non-radial migration of inhibitory interneurons from the basal forebrain into the cortex and hippocampus in Lis1 mice. Our in vivo and in vitro studies identified both a cell autonomous and cell-non-autonomous effect on NRCM, although the cell autonomous effect was greater. This NRCM defect results in inhibitory interneurons arriving late to the appropriate cortical region. A defect in inhibitory neurons could explain, at least in part, the clinical phenotype of both epilepsy and mental retardation observed in all patients with a LIS1 deletion.

Although others have used BrdU labeling of cells to visualize radial migration,10 this approach will not work for studying NRCM, as labeled cells found in the dorso-lateral and dorsal cortex that have migrated from the basal forebrain cannot be distinguished from those migrating from the cortical VZ. However, non-radially migrating cells in the dorsal forebrain are primarily inhibitory interneurons. Visualization of this population is facilitated by the availability of various immunohistochemical markers, including antibodies against GABA, GAD, and subpopulations of cells expressing markers such as calretinin, somatostatin, parvalbumin, neuro-peptide Y, and nitric oxide synthetase.42

Calretinin labels a subset of approximately 20 to 30 percent of GABAergic inhibitory interneurons while antibodies to GABA label all inhibitory neurons.36,37,43,44 Using these markers we observed cells to be significantly closer to the dorsal telencephalon in Lis1 +/+ mice when compared to Lis1 +/− littermates (littermates were used to ensure an age-matched control). An approximate speed can also be determined by comparing the distance traveled by cells at different time points, E14 and E14.5. Over the course of this half-day of development, the leading dozen wild-type cells moved at an average speed of 43.8 μm/hour, whereas the mutant cells progressed at only 23 μm/hour. The calculated rate of migration for the wild-type cells was similar to calculations made by others in different systems.45 Interestingly, the approximately 50% reduction (43.8 μm/hour to 23 μm/hour) in the cell migration rate we observed is comparable to that observed for a granular cell assay of granular cell migration in the presence of platelet activating factor, where a 46% decrease in the rate of migration was observed.46 Thus, differences in the distances traveled by inhibitory cells can be interpreted as differences in the speed of migration. Additional studies, potentially using direct imaging or time-lapse video-microscopy will be required to confirm these data.

An alternative explanation is that the Lis1 +/− animals are developmentally delayed by one-half to one full day, and the inhibitory interneurons are not slowed but rather delayed in commencing migration. While the morphological staging of embryo and immunohistochemistry for GABA at various stages of development beginning at E12.5 were similar (data not shown), a developmental delay could not be formally excluded. To address this possibility, an in vitro slice assay was used. The comparison between the average distance from the DiI source of all cells in the wild-type and the Lis1 +/− slices indicates a greatly decreased average distance of migration in the Lis1 +/− animals. A difference in the timing of the onset of migration, rather than differential migration rates, could only account for the observed difference if the average speed of cell migration increases greatly from E14 to E14.5. Preliminary data from E13.5 to E16 embryos using time-lapse video-microscopy indicate this is not the case (Nasrallah IM, Golden JA, unpublished observations).

A difference in the distribution of migratory speeds in Lis1 +/− cells could potentially generate a difference in population averages, without any difference in the maximum average cell speed. To assess whether the fastest Lis1 +/− cells were capable of migrating at the same speed as their wild-type littermate cells, the leading 25 cells in each slice were examined. The difference in average distance traveled between these two groups was highly significant, suggesting that the migration defect is seen in all Lis1 +/− cells and not in simply a subset of affected interneurons. Population averages, however, take into account all identifiable cells and can be weighted toward the slower moving cells, those closer to the DiI crystal, as these are in the greatest abundance. In cultures from Lis1 +/− animals, fewer total cells migrate out far enough to be counted. This difference in total cell count could result in the observed migratory differences in the leading 25 cells simply reflecting a greater number of wild-type cells being observed at the farthest distances, and not from a difference in overall distribution of cells. To ascertain if the distribution of distances was dissimilar, cells were placed in bins based on the distance traveled from the DiI crystal, and the number in either Lis1 +/− or wild-type slices compared. The number of cells observed in the nearest bins was not different between genotypes. However, at intermediate distances there were significantly fewer mutant cells when compared to wild-type cells, with essentially no mutant cells found at the farthest distances wild-type cells had migrated. This, as well as the farthest individual cell calculations, indicates that the decreased number of mutant cells found farthest from the DiI crystal is not due simply to an overall decreased number of migratory cells.

The observed difference in total cell number between Lis1 +/− and wild-type animals is likely an artifact resulting from the inability to count labeled cells close to the DiI crystal (less than approximately 250 μm), as it produces a bright fluorescent halo in which one cannot observe sufficient cellular detail to distinguish individual labeled cells. This is similar to data previously reported for the in vitro migration of cerebellar granule neurons from Lis1 +/− animals away from cellular re-aggregate clusters.21 It is likely that many interneurons in the Lis1 +/− slice cultures take up the DiI label but migrate less than 250 μm, a region where they can be counted due to the DiI fluorescence haze. This effectively decreases the total number of labeled cells tallied but not generated in mutant slices.

Recent in vivo and in vitro work suggests that non-radially migrating cells may use axonal processes as migratory substrates;40,41 in much the same way radial migrants use glia (reviewed in47). Lis1 +/− animals exhibit aberrant thalamo-cortical innervation20 in addition to a generally disorganized cortex.10,21 We hypothesized that this could lead to a perturbation of the migratory substrate in NRCM and consequently a cell non-autonomous migration defect. To test this hypothesis, both wild-type and mutant GEs were excised from slice cultures, apposed to either wild-type or mutant cortical explants, and implanted with DiI crystals as above. A strong cell autonomous effect was observed by comparing cultures with wild-type to those with Lis1 +/− GEs, using the average distance traveled by both the entire population and the 25 leading cells again as the migratory metric. A significant cell autonomous effect was observed, evidenced by a difference between wild-type and Lis1 +/− GE cells without regard to substrate (cortex) genotype. To determine whether a cell non-autonomous effect exists, we compared migration holding the genotype of migrating cells constant while varying the substrate’s genotype. A significant cell non-autonomous effect was observed when comparing the distance the 25 leading wild-type cells migrated on wild-type and Lis1 +/− cortices, with migration on mutant cortex proceeding less far. The leading wild-type cells are both the most competent and have presumably had the longest time in which to migrate. It is perhaps not surprising then, that this was the only condition in which a cell non-autonomous effect was observed, as population data tend to diminish small differences and Lis1 +/− cells may not be able to migrate far enough in the time course of our experiments to exhibit a difference. Nonetheless, the substrate effect, though significant, is not as dramatic as the cell autonomous defect.

A slowing of radial cell migration over distances as short as from the VZ to the CP of the developing forebrain disrupts the laminar organization of the neocortex and hippocampus.10 For cell populations that migrate over significantly longer distances, any impediment to migration could potentially affect both their ability to arrive at their target locations at the appropriate time in development and also to form appropriate connectivity with other neurons in developmentally restricted time frames. In the Lis1 +/− model of type I lissencephaly, a number of electrophysiological defects in non-locally born hippocampal cells in organotypic slice cultures have been described22 and postulated as causal for the Lis1 +/− animal’s susceptibility to lethal status-epilepticus.10,22 Indeed, the parvalbuminergic, non-pyramidal inhibitory interneurons of the hippocampus are locally derived and born at the same time as the principal neurons,48,49 having to migrate only short distances, and this has been suggested as the reason they appropriately innervate their targets.22 The slowed NRCM reported here could lead to GE-derived inhibitory interneurons missing their developmental window to arrive at and appropriately connect within the hippocampus and neocortex, contributing significantly to the pathogenesis of the epileptic phenotype in the Lis1 +/− animals. However, it is important to recognize, like RCM, the non-radially migrating interneurons do eventually reach the cortex in approximately normal numbers22 (also Pancoast MM, Nasrallah IM, Golden JA, unpublished observations).

The data presented in this report indicate the role of Lis1 in cell migration is not limited to radial cell migration along but clearly has a role in NRCM. It is possible that a NRCM defect also exists in other forms of lissencephaly, as might be predicted for at least XLAG.8 Characterization of the contribution of NRCM to the structural and clinical phenotypic features of different lissencephaly syndromes awaits further evaluation.

Footnotes

Address reprint requests to Jeffrey A. Golden, Department of Pathology, Abramson Research Center, Room 516h, Children’s Hospital of Philadelphia, 3400 Civic Center Blvd., Philadelphia, PA 19104. E-mail: goldenj@mail.med.upenn.edu.

Supported by NS39949 and HD26979 (J.A.G.) and institutional training grants GM17517 (M.F.M.), and GM07170 (I.M.N.).

M.F.M. and I.M.N. contributed equally to the work presented in this manuscript.

References

  1. Walsh CA. Genetic malformations of the human cerebral cortex. Neuron. 1999;23:19–29. doi: 10.1016/s0896-6273(00)80749-7. [DOI] [PubMed] [Google Scholar]
  2. Hattori M, Adachi H, Tsujimoto M, Arai H, Inoue K. Miller-Dieker lissencephaly gene encodes a subunit of brain platelet-activating factor acetylhydrolase. Nature. 1994;370:216–218. doi: 10.1038/370216a0. [DOI] [PubMed] [Google Scholar]
  3. Cardoso C, Leventer RJ, Ward HL, Toyo-Oka K, Chung J, Gross A, Martin CL, Allanson J, Pilz DT, Olney AH, Mutchinick OM, Hirotsune S, Wynshaw-Boris A, Dobyns WB, Ledbetter DH. Refinement of a 400-kb critical region allows genotypic differentiation between isolated lissencephaly, Miller-Dieker syndrome, and other phenotypes secondary to deletions of 17p13.3. Am J Hum Genet. 2003;72:918–930. doi: 10.1086/374320. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Dobyns WB: Lissencephaly Overview, www.geneclinics.org, 2000 [Google Scholar]
  5. Reiner O, Carrozzo R, Shen Y, Wehnert M, Faustinella F, Dobyns WB, Caskey CT, Ledbetter DH. Isolation of a Miller-Dieker lissencephaly gene containing G protein beta-subunit-like repeats. Nature. 1993;364:717–721. doi: 10.1038/364717a0. [DOI] [PubMed] [Google Scholar]
  6. des Portes V, Pinard JM, Billuart P, Vinet MC, Koulakoff A, Carrie A, Gelot A, Dupuis E, Motte J, Berwald-Netter Y, Catala M, Kahn A, Beldjord C, Chelly J. A novel CNS gene required for neuronal migration and involved in X-linked subcortical laminar heterotopia and lissencephaly syndrome. Cell. 1998;92:51–61. doi: 10.1016/s0092-8674(00)80898-3. [DOI] [PubMed] [Google Scholar]
  7. Gleeson JG, Allen KM, Fox JW, Lamperti ED, Berkovic S, Scheffer I, Cooper EC, Dobyns WB, Minnerath SR, Ross ME, Walsh CA. Doublecortin, a brain-specific gene mutated in human X-linked lissencephaly and double cortex syndrome, encodes a putative signaling protein. Cell. 1998;92:63–72. doi: 10.1016/s0092-8674(00)80899-5. [DOI] [PubMed] [Google Scholar]
  8. Kitamura K, Yanazawa M, Sugiyama N, Miura H, Iizuka-Kogo A, Kusaka M, Omichi K, Suzuki R, Kato-Fukui Y, Kamiirisa K, Matsuo M, Kamijo S, Kasahara M, Yoshioka H, Ogata T, Fukuda T, Kondo I, Kato M, Dobyns WB, Yokoyama M, Morohashi K. Mutation of ARX causes abnormal development of forebrain and testes in mice and X-linked lissencephaly with abnormal genitalia in humans. Nat Genet. 2002;32:359–369. doi: 10.1038/ng1009. [DOI] [PubMed] [Google Scholar]
  9. Hong SE, Shugart YY, Huang DT, Al Shahwan S, Grant PE, Hourihane JOB, Martin NDT, Walsh CA. Autosomal recessive lissencephaly with cerebellar hypoplasia (LCH) is associated with human reelin gene mutations. Nat Genet. 2000;26:93–96. doi: 10.1038/79246. [DOI] [PubMed] [Google Scholar]
  10. Hirotsune S, Fleck MW, Gambello MJ, Bix GJ, Chen A, Clark GD, Ledbetter DH, McBain CJ, Wynshaw-Boris A. Graded reduction of Pafah1b1 (Lis1) activity results in neuronal migration defects and early embryonic lethality. Nat Genet. 1998;19:333–339. doi: 10.1038/1221. [DOI] [PubMed] [Google Scholar]
  11. Smith DS, Niethammer M, Ayala R, Zhou Y, Gambello MJ, Wynshaw-Boris A, Tsai LH. Regulation of cytoplasmic dynein behaviour and microtubule organization by mammalian Lis1. Nat Cell Biol. 2000;2:767–775. doi: 10.1038/35041000. [DOI] [PubMed] [Google Scholar]
  12. Xiang X, Osmani AH, Osmani SA, Xin M, Morris NR. NudF, a nuclear migration gene in Aspergillus nidulans, is similar to the human LIS-1 gene required for neuronal migration. Mol Biol Cell. 1995;6:297–310. doi: 10.1091/mbc.6.3.297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Geiser JR, Schott EJ, Kingsbury TJ, Cole NB, Totis LJ, Bhattacharyya G, He L, Hoyt MA. Saccharomyces cerevisiae genes required in the absence of the CIN8-encoded spindle motor act in functionally diverse mitotic pathways. Mol Biol Cell. 1997;8:1035–1050. doi: 10.1091/mbc.8.6.1035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Liu Z, Steward R, Luo L. Drosophila Lis1 is required for neuroblast proliferation, dendritic elaboration, and axonal transport. Nat Cell Biol. 2000;2:776–783. doi: 10.1038/35041011. [DOI] [PubMed] [Google Scholar]
  15. Peterfy M, Hozier JC, Hall B, Gyuris T, Peterfy K, Takecs L. Localization of the mouse lissencephaly-1 gene to mouse chromosome 11B3, in close proximity to D11Mit65. Somat Cell Mol Genet. 1995;21:345–349. doi: 10.1007/BF02257469. [DOI] [PubMed] [Google Scholar]
  16. Faulkner NE, Dujardin DL, Tai CY, Vaughan KT, O’Connell CB, Wang Y, Vallee RB. A role for the lissencephaly gene LIS1 in mitosis and cytoplasmic dynein function. Nat Cell Biol. 2000;2:784–791. doi: 10.1038/35041020. [DOI] [PubMed] [Google Scholar]
  17. Sapir T, Elbaum M, Reiner O. Reduction of microtubule catastrophe events by LIS1, platelet-activating factor acetylhydrolase subunit. EMBO J. 1997;16:6977–6984. doi: 10.1093/emboj/16.23.6977. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Morris NR, Efimov VP, Xiang X. Nuclear migration, nucleokinesis, and lissencephaly. Trends Cell Biol. 1998;8:467–470. doi: 10.1016/s0962-8924(98)01389-0. [DOI] [PubMed] [Google Scholar]
  19. Feng Y, Olson EC, Stukenberg PT, Flanagan LA, Kirschner MW, Walsh CA. LIS1 regulates CNS lamination by interacting with mNudE, a central component of the centrosome. Neuron. 2000;28:665–679. doi: 10.1016/s0896-6273(00)00145-8. [DOI] [PubMed] [Google Scholar]
  20. Cahana A, Escamez T, Nowakowski RS, Hayes NL, Giacobini M, von Holst A, Shmueli O, Sapir T, McConnell SK, Wurst W, Martinez S, Reiner O. Targeted mutagenesis of Lis1 disrupts cortical development and LIS1 homodimerization. Proc Natl Acad Sci USA. 2001;98:6429–6434. doi: 10.1073/pnas.101122598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gambello MJ, Darling DL, Yingling J, Tanaka T, Gleeson JG, Wynshaw-Boris A. Multiple dose-dependent effects of Lis1 on cerebral cortical development. J Neurosci. 2003;23:1719–1729. doi: 10.1523/JNEUROSCI.23-05-01719.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Fleck MW, Hirotsune S, Gambello MJ, Phillips-Tansey E, Suares G, Mervis RF, Wynshaw-Boris A, McBain CJ. Hippocampal abnormalities and enhanced excitability in a murine model of human lissencephaly. J Neurosci. 2000;20:2439–2450. doi: 10.1523/JNEUROSCI.20-07-02439.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Rakic P. Principles of neural cell migration. Experientia. 1990;46:882–891. doi: 10.1007/BF01939380. [DOI] [PubMed] [Google Scholar]
  24. Cajal R. Madrid: Histologie du Systeme Nerveux de l’Homme et des Vertebres. 1952 [Google Scholar]
  25. Cepko C, Golden J, Szele F, Lin J. Cowen W, Jessell T, Zipursky S, editors. Oxford: Oxford University Press,; Lineage analysis in the vertebrate central nervous system. Neuronal Development. 1997 [Google Scholar]
  26. Marin O, Rubenstein JL. A long, remarkable journey: tangential migration in the telencephalon. Nat Rev Neurosci. 2001;2:780–790. doi: 10.1038/35097509. [DOI] [PubMed] [Google Scholar]
  27. Pilz D, Stoodley N, Golden JA. Neuronal migration, cerebral cortical development, and cerebral cortical anomalies. J Neuropathol Exp Neurol. 2002;61:1–11. doi: 10.1093/jnen/61.1.1. [DOI] [PubMed] [Google Scholar]
  28. Anderson SA, Eisenstat DD, Shi L, Rubenstein JL. Interneuron migration from basal forebrain to neocortex: dependence on Dlx genes. Science. 1997;278:474–476. doi: 10.1126/science.278.5337.474. [DOI] [PubMed] [Google Scholar]
  29. Anderson SA, Marin O, Horn C, Jennings K, Rubenstein JL. Distinct cortical migrations from the medial and lateral ganglionic eminences. Development. 2001;128:353–363. doi: 10.1242/dev.128.3.353. [DOI] [PubMed] [Google Scholar]
  30. Pleasure SJ, Anderson S, Hevner R, Bagri A, Marin O, Lowenstein DH, Rubenstein JL. Cell migration from the ganglionic eminences is required for the development of hippocampal GABAergic interneurons. Neuron. 2000;28:727–740. doi: 10.1016/s0896-6273(00)00149-5. [DOI] [PubMed] [Google Scholar]
  31. Ferrer I, Pineda M, Tallada M, Oliver B, Russi A, Oller L, Noboa R, Zujar MJ, Alcantara S. Abnormal local-circuit neurons in epilepsia partialis continua associated with focal cortical dysplasia. Acta Neuropathol (Berl) 1992;83:647–652. doi: 10.1007/BF00299415. [DOI] [PubMed] [Google Scholar]
  32. Treiman DM. GABAergic mechanisms in epilepsy. Epilepsia. 2001;42:8–12. doi: 10.1046/j.1528-1157.2001.042suppl.3008.x. [DOI] [PubMed] [Google Scholar]
  33. Theiler K. New York: Springer-Verlag,; The house mouseatlas of embryonic development. 1989:pp 178. [Google Scholar]
  34. Heffron DS, Golden JA. DM-GRASP is necessary for non-radial cell migration during chick diencephalic development. J Neurosci. 2000;20:2287–2294. doi: 10.1523/JNEUROSCI.20-06-02287.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Wynshaw-Boris A, Gambello MJ. LIS1 and dynein motor function in neuronal migration and development. Genes Dev. 2001;15:639–651. doi: 10.1101/gad.886801. [DOI] [PubMed] [Google Scholar]
  36. Xu Q, Cobos I, De La Cruz E, Rubenstein JL, Anderson SA. Origins of cortical interneuron subtypes. J Neurosci. 2004;24:2612–2622. doi: 10.1523/JNEUROSCI.5667-03.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Xu Q, de la Cruz E, Anderson SA. Cortical interneuron fate determination: diverse sources for distinct subtypes? Cereb Cortex. 2003;13:670–676. doi: 10.1093/cercor/13.6.670. [DOI] [PubMed] [Google Scholar]
  38. Hevner RF, Neogi T, Englund C, Daza RA, Fink A. Cajal-Retzius cells in the mouse: transcription factors, neurotransmitters, and birthdays suggest a pallial origin. Brain Res Dev Brain Res. 2003;141:39–53. doi: 10.1016/s0165-3806(02)00641-7. [DOI] [PubMed] [Google Scholar]
  39. Ang ES, Jr, Haydar TF, Gluncic V, Rakic P. Four-dimensional migratory coordinates of GABAergic interneurons in the developing mouse cortex. J Neurosci. 2003;23:5805–5815. doi: 10.1523/JNEUROSCI.23-13-05805.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Denaxa M, Chan CH, Schachner M, Parnavelas JG, Karagogeos D. The adhesion molecule TAG-1 mediates the migration of cortical interneurons from the ganglionic eminence along the corticofugal fiber system. Development. 2001;128:4635–4644. doi: 10.1242/dev.128.22.4635. [DOI] [PubMed] [Google Scholar]
  41. McManus MF, Nasrallah IM, Gopal PP, Back WS, Golden JA: Axon-mediated interneuron migration. J Neuropathol Exp Neurol 2004, in press [DOI] [PubMed] [Google Scholar]
  42. Marin O, Anderson SA, Rubenstein JL. Origin and molecular specification of striatal interneurons. J Neurosci. 2000;20:6063–6076. doi: 10.1523/JNEUROSCI.20-16-06063.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Gonchar Y, Burkhalter A. Three distinct families of GABAergic neurons in rat visual cortex. Cereb Cortex. 1997;7:347–358. doi: 10.1093/cercor/7.4.347. [DOI] [PubMed] [Google Scholar]
  44. del Rio JA, Martinez A, Fonseca M, Auladell C, Soriano E. Glutamate-like immunoreactivity and fate of Cajal-Retzius cells in the murine cortex as identified with calretinin antibody. Cereb Cortex. 1995;5:13–21. doi: 10.1093/cercor/5.1.13. [DOI] [PubMed] [Google Scholar]
  45. Edmondson J, Hatten M. Glial-guided granule neuron migration in vitro: a high-resolution time-lapse video-microscopic study. J Neurosci. 1987;7:1928–1934. doi: 10.1523/JNEUROSCI.07-06-01928.1987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Bix GJ, Clark GD. Platelet-activating factor receptor stimulation disrupts neuronal migration in vitro. J Neurosci. 1998;18:307–318. doi: 10.1523/JNEUROSCI.18-01-00307.1998. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Hatten ME. Central nervous system neuronal migration. Annu Rev Neurosci. 1999;22:511–539. doi: 10.1146/annurev.neuro.22.1.511. [DOI] [PubMed] [Google Scholar]
  48. Soriano E, Frotscher M. A GABAergic axo-axonic cell in the fascia dentata controls the main excitatory hippocampal pathway. Brain Res. 1989;503:170–174. doi: 10.1016/0006-8993(89)91722-8. [DOI] [PubMed] [Google Scholar]
  49. Seress L, Ribak CE. The development of GABAergic neurons in the rat hippocampal formation: an immunocytochemical study. Brain Res Dev Brain Res. 1988;44:197–209. doi: 10.1016/0165-3806(88)90218-0. [DOI] [PubMed] [Google Scholar]

Articles from The American Journal of Pathology are provided here courtesy of American Society for Investigative Pathology

RESOURCES