A perturbation of multimodal GABA functions underlying the formation of focal cortical malformations: Assessments by using animal models

Authors


Correspondence: Atsuo Fukuda, MD, PhD, Department of Neurophysiology, Hamamatsu University School of Medicine, 20-1 Handayama 1-chome, Higashi-ku, Hamamatsu, Shizuoka 431-3192, Japan. Email: axfukuda@hama-med.ac.jp

Abstract

In order to study how the formation of focal cortical malformations is attributed to perturbation of developmentally multimodal γ-aminobutyric acid (GABA) functions, we made a focal cortical freeze-lesion on a rodent cerebral cortex at P0 (postnatal day 0). The microgyrus was formed at P7. GABA neurons were accumulated in the region surrounding necrosis at P4. Cortical plate cells born at E17.5 gathered, surrounding the GABA neurons, forming the cell dense portions in layer 2 of the microgyrus. Ambient GABA level was increased in the area corresponding to populated GABA neurons at P4. A KCC2 expression was downregulated, whereas an NKCC1 expression was upregulated in both the gathering GABA and cortical plate neurons, suggesting these cells had high intracellular Cl concentration rendering GABA action depolarizing. GABAA receptor activation was involved in Ca2+ oscillation in these gathering cells. In vivo blockade of GABAA receptor prevented the above characteristic pattern of cell accumulation and hence microgyrus formation. Thus, neonatal freeze-lesion causes characteristic accumulation of differential populations of neurons preceded by characteristic release of GABA at an early stage, which induces GABAA receptor-mediated depolarization and Ca2+ oscillation. This paracrine/autocrine GABA may underlie the formation of neocortical malformations such as polymicrogyria.

Introduction

The major inhibitory neurotransmitter GABA necessarily evokes depolarization in the immature brain, in contrast to hyperpolarization in the adult brain. Because the GABAA receptor is a chloride-negative (Cl) channel, this developmental switch of GABA actions from depolarization (Cl efflux) to hyperpolarization (Cl influx) is induced by changes in the transmembranous Cl gradient, which is regulated by Cl transporters (NKCC1 (sodium, potassium, chloride cotransporter), uptake; KCC2, extrusion). In addition, the release mechanism of GABA changes from non-vesicular and non-synaptic to vesicular and synaptic ones, so that GABA actions are generally tonic in immature neurons and are converted to phasic with development. As these tonic depolarizing (and excitatory on occasion) GABA actions are necessary for neurogenesis, differentiation, migration and synaptogenesis, we term the above-mentioned GABA actions as “developmentally multimodal GABA functions” (Fig. 1).[1, 2]

Figure 1.

Schematic drawings of modal shift of GABAA receptor-mediated actions during corticogenesis. In the ventricular zone and subventricular zone, ambient GABA affects neurogenesis. When post-mitotic neurons migrate to the cortical plate, they are tonically depolarized by ambient GABA which is non-synaptically and non-vesicularly released from tangentially migrating GABA neurons prior to forming synapses, and change migration tempo. GABA also acts in autocrine fashion to accelerate tangential migration. Vesicular release of GABA, which remains depolarized, could contribute to synapse formation. Following establishment of GABAergic synapses and prior to hyperpolarization, GABA acts as an excitatory neurotransmitter. Up to this stage, chloride negtive (Cl) levels are high due to predominant sodium/potassium/chloride cotransporter 1 (NKCC1) and subsidiary KCC2. Following upregulation of KCC2 and downregulation of NKCC1, GABA acts as an inhibitory neurotransmitter (not shown in the figure). Ambient GABA is indicated as a green-shaded circle.

On the other hand, neocortical malformations such as polymicrogyria show the presence of cortical cells in heterotopic positions; this is often caused by disruption in neuronal migration.[3, 4] Therefore, perturbation of the mulitimodal GABA function during development may underlie etiology of cortical malformations. Indeed, GABAA receptor-mediated responses are anomalously depolarizing in human focal cortical dysplasia tissue.[5, 6] However, how mulitimodality of GABA functions is disturbed is yet to be elucidated and it requires experiments using animal models.

The focal freeze-lesion (FFL) in the cerebral cortex of newborn (P0) rats produces a microgyrus, a focal cortical malformation with a small sulcus and a three- or four-layered microgyric cortex, which resembles human four-layered polymicrogyria,[7-10] a clinical condition that results from abnormal neuronal migration. Here, by using FFL models of rodents, we have addressed abnormalities in developmentally multimodal GABA functions. A part of this original data has been published and is available on-line (http://cercor.oxfordjournals.org/content/early/2012/12/16/cercor.bhs375.full?keytype=ref&ijkey=i04VZWybp7NFEDX).[11]

Abnormal Modal Shifts of Cl Homeostasis Underlying Neocortical Malformation

Altered GABAergic functions that occur during early brain development are induced by changes in Cl homeostasis and play important roles in neocortical development by modulating events, such as laminar organization and synaptogenesis (Fig. 1). A relationship between polymicrogyria pathogenesis and Cl homeostasis ontogeny in the developing parietal cortex were observed following FFL on both rats[12] and mice[11] at P0. In the FFL area, NKCC1 mRNA expression levels are greater in the lesion center-surrounding areas at P4 compared with adjacent exofocal areas. In contrast, although KCC2 mRNA re-emerges in the surrounding structure, expression levels are weaker than in the exofocal cortex. After P7, KCC2 mRNA expression increases in the microgyric region to levels similar to the intact surrounding cortex. Subsequently, NKCC1 mRNA expression in the microgyric region decreases to levels similar to exofocal cortices at P14–21.

In addition to the rat model,[12] we have established a FFL in glutamic acid decarboxylase-green fluorescent protein (GAD67-GFP) knock-in (KI) mice[13] maternally injected with 5-bromo-2′-deoxyuridine (BrdU) to distinguish GABAergic and glutamatergic neurons, with identified birthdates, responsible for the formation of a FFL-induced microgyrus.[11] This model indicated that the above expressional changes in Cl transporters were exclusively induced in the microgyrus-forming GABAergic and E17.5-born glutamatergic neurons.

Distinct Accumulation of Gabaergic and Late-born Glutamatergic Neurons in the Developing Microgyrus

Because GABAergic and glutamatergic neurons are generated, migrate and settle differently in the cortical region at distinct stages, we next addressed how the microgyrus is formed by migration of these distinct types of neurons. In combination with classical birth date analysis to label late-born neurons using BrdU, we determined GABAergic neurons and E14.5- or E17.5-born cells during formation of the microgyrus at P4[11].

In the lesioned cortex, few E14.5-born BrdU-positive cells reached the area surrounding the necrotic center (not shown). In contrast, many E17.5-born BrdU-positive neurons accumulated in the area surrounding the necrotic center. Interestingly, GFP-positive GABAergic cells accumulated around the lesioned area including in the superficial area (Fig. 2).[11] Quantitative analysis of the distribution of GFP-positive cells revealed that they also tended to occupy the inner rim within 50 μm of the FFL border rather than the outer rim 50–100 μm from the FFL border (P < 0.05). In contrast to GFP-positive cells, E17.5-born BrdU-positive cells were almost absent in the superficial region of the lesioned cortex (P < 0.001). In addition, in contrast to GFP-positive cells, E17.5-born BrdU-positive cells tended to occupy the outer rim rather than the inner rim (P < 0.05).

Figure 2.

Distribution of E17.5-born cells and GABAergic cells in the FFL cortex at post-natal day 4 (P4). Images of coronal sections of a P4 focal freeze-lesion (FFL) mouse that received a maternal injection of 5-bromo-2′-deoxyuridine (BrdU) at E17.5, immunostained for BrdU (red, A) and green fluorescent protein (GFP) (green, B). An accumulation of GFP-positive cells can be detected in the superficial part of the lesioned area (B). (C) Merged images of (A) and (B) show an accumulation of E17.5-born BrdU-positive cells surrounding the GFP-positive cells at the boundary of the lesioned area. (D) Higher magnification of the superficial part of the lesion center as marked in (C). (E) Higher magnification of the boundary of the lesion center as marked in (C). (F,G) Distribution of GFP-positive cells in the superficial region (F) and in the inner and the outer rim of the surrounding region within 100 μm of the lesion border (G). (H,I) Distribution of E17.5-born BrdU-positive cells in the superficial region (H) and in the inner and the outer rim of the surrounding region (I). Note that there is a significant difference in the cell density of E17.5-born BrdU-positive cells between the superficial and the exofocal regions (H, independent two-tailed t-test, ***P < 0.001; three preparations). Note that there are significant differences in the cell densities of GFP-positive cells (G) and E17.5-born BrdU-positive cells (I) between the inner rim and the outer rim (independent two-tailed t-test, *P < 0.05; three preparations each). The positions of the cell-count square frames (50 μm × 50 μm) are shown in (A). (J,K) Images of double-staining for the layer-specific markers Cux1 (J) or Tbr1 (K) (green) and BrdU (red). A Cux1-positive cell-dense band can be observed in the area surrounding the necrotic tissue (J). Higher magnification shows that the majority of E17.5-born BrdU-positive cells co-localize with Cux1 (inset in J). Tbr1-positive cells are absent from the dysplastic cortex at P4 (K). (L) There is a significant difference in the cell density of E17.5-born BrdU/Cux1 double-positive cells and E17.5-born BrdU/Tbr1 double-positive cells in the surrounding region within 100 μm of the lesion border (two-tailed t-test, ***P < 0.001; three preparations each). Three cell-count square frames (50 μm × 50 μm) were placed randomly in the surrounding (medial, deeper and lateral) layer within 100 μm of the lesion border (J). Scale bars: (A–C) 200 μm; (D,E) 50 μm; (J,K) 100 μm; inset of (J) 25 μm. Adapted from Wang et al.,[11] available at http://cercor.oxfordjournals.org/content/early/2012/12/16/cercor.bhs375.full?keytype=ref&ijkey=i04VZWybp7NFEDX.

In the area surrounding the FFL, a Cux1-positive, cell-dense band was detected (Fig. 2), whereas Tbr1-positive cells were absent in the area surrounding the FFL (Fig. 2) at P4. In the FFL area, E17.5-born BrdU-positive cells accumulated around the necrotic tissue and most co-localized with Cux1 (Fig. 2, inset), but not with Tbr1 (Fig. 2). These results suggest that the majority of E17.5-born cells that accumulated in the area surrounding the FFL were Cux1-positive glutamatergic cortical plate cells.

Temporal Increases in Ambient GABA in the Developing Microgyrus

Previous studies have suggested that GABA released from GABAergic neurons has a paracrine action on immature neurons,[14, 15] which may affect radial[16-21] and tangential[22-24] migration during neural development. Therefore, we next examined the spatio-temporal changes in extracellular GABA levels in the FFL cortex at different stages by means of GABA imaging recently developed.[25]

The extracellular GABA concentration in the FFL center was about 43 μmol/L, and in the area < 100 μm from the FFL border was about 26 μmol/L; both values were significantly higher than those in the more distant areas (>100 μm: 9–11 μmol/L; Fig. 3).[11] We also examined glutamate, using the modified method in which glutamate is catabolized by α-ketoglutarate by L-glutamic dehydrogenase (GDH), resulting in the production of a fluorescent reduced form of β-nicotinamide adenine dinucleotide (NADH). As shown in Figure 3, at P4, the ambient glutamate levels in the areas nearer to the FFL (within lesion: 2 μmol/L; <100 μm: 9 μmol/L) were much lower than those in the areas further from the FFL border (>100 μm: 25–26 μmol/L; Fig. 3).[11]

Figure 3.

Temporal increase in ambient GABA in focal freeze-lesion (FFL). (A) A representative image showing the distribution of GABAergic neurons in and around the dysplastic cortex at post-natal day 4 (P4). (B) An image of extracellular GABA in a P4 acute coronal slice using an enzyme-linked assay system. The dotted line indicates the border of the necrotic tissue. (C) GABA concentration in the dysplastic cortex at P4. The GABA concentration in the lesion center and in the area within 100 μm of the lesion border were significantly higher than that in the area more than 100 μm from the lesion border (post-hoc Tukey test; *P < 0.05, **P < 0.01, ***P < 0.001, n = 10 slices from six animals). As a sample for analysis of variance (ANOVA), fluorescence intensity was measured and averaged for each zone as indicated in A, from nine randomly placed square frames (50 μm × 50 μm). (D) Image of extracellular glutamate in the same slice as shown in (B). (E) Statistical analysis of glutamate concentration as measured similarly to GABA. The glutamate concentrations in the lesion center and in the area within 100 μm of the lesion site were significantly lower than that in the area more than 100 μm from the lesion site (post-hoc Tukey test; **P < 0.01, ***P < 0.001, n.s., not significant; n = three slices from three animals). Scale bars: 100 μm. Adapted from Wang et al.,[11] available at http://cercor.oxfordjournals.org/content/early/2012/12/16/cercor.bhs375.full?keytype=ref&ijkey=i04VZWybp7NFEDX.

Since this characteristic GABA accumulation was not observed in P2 or P7 FFL cortex, we speculated that the temporally increased extracellular GABA released from accumulated GABAergic neurons in the lesioned cortex might play a role in abnormal neuronal migration during formation of the microgyrus.

Intracellular Ca2+ Transients Mediated by GABAA Receptor Activation in Microgyrus-forming Cells

Given the transient increase in ambient GABA concentration and the potentially elevated intracellular Cl concentration due to reciprocal expressional changes in Cl transporters, NKCC1 and KCC2, in the vicinity of the FFL at P4, GABAA receptor-mediated depolarizing action may be responsible for the spontaneous Ca2+ oscillation that is important for regulation of neuronal migration.[26, 27]

The frequency of spontaneous Ca2+ transients was significantly higher in the cells < 100 μm from the FFL border than in those at the further site (P < 0.01; Fig. 4).[11] When the GABAA receptor antagonist bicuculline methiodide (BMI; 20 μmol/L) was bath-applied, the number of spontaneous Ca2+ transients was significantly reduced only within 100 μm of the FFL border (P < 0.01; Fig. 4). These results suggest that tonic GABAA receptor-mediated depolarizing activity produces Ca2+ transients in migrating cells that are forming the microgyrus.

Figure 4.

Spontaneous intracellular Ca2+ transients near the lesion site were mediated by tonic GABAA receptor activation. (A) Typical example of a post-natal day 4 (P4). neocortical cell in the area within 100 μm of the lesion border showing spontaneous Ca2+ transients in its soma. (B) Traces of spontaneous Ca2+ transients before and after bath application of 20 μmol/L bicuculline methiodide (BMI). (C) Numbers of spontaneous Ca2+ transients in cells in each area. The frequency of Ca2+ transients in the cells within 100 μm of the lesion border was significantly higher than that in the cells in both the upper and the lower parts > 100 μm from the lesion border (post-hoc Tukey test; **P < 0.01). (D) Number of spontaneous Ca2+ transients before and after bath application of BMI in each area. The frequency of Ca2+ transients was significantly decreased by BMI only in the area within 100 μm of the lesion border (two-tailed paired t-test, **P < 0.01). (E) Number of spontaneous Ca2+ transients before and after bath application of D-AP5 (D(–)-2-amino-5-phosphonopentanoic acid) and CNQX (6-cyano-7-nitroquinoxaline-2,3-dione) in each area. The frequency of Ca2+ transients was significantly decreased by D-AP5 + CNQX only in the areas > 100 μm from the lesion border (paired two-tailed t-test, *P < 0.05). Scale bar: 20 μm. ACSF, artificial cerebrospinal fluid control. Adapted from Wang et al.,[11] available at http://cercor.oxfordjournals.org/content/early/2012/12/16/cercor.bhs375.full?keytype=ref&ijkey=i04VZWybp7NFEDX. image, ACSF; image, AP5+CNQX.

Chronic in vivo Blockade of GABAA Receptors Perturbs the Microgyrus Formation

Next, we blocked GABAA receptors continuously in vivo after FFL to study whether the accumulation of GABAergic neurons, causing a focal ambient GABA gradient, underlies microgyrus formation. BMI was injected intraperitoneally from P2 to P4 after FFL.[11] At P7, although a sulcus-like formation was observed on the surface of the brain, the microgyric appearance of three- or four-layered cortex was obscured (Fig. 5).[11]

Figure 5.

Distribution of late-born glutamatergic neurons and GABAergic cells in the focal freeze-lesion (FFL) cortex at post-natal day 4 (P4) after intraperitoneal injection of bicuculline. (A,B) Thionin-stained coronal section from a P4 (A) and P7 (B) GAD67-GFP (glutamate decarboxylase 67 kDa – green fluorescent protein) knock-in mouse that received FFL at P0 and intraperitoneal injection of bicuculline methiodide (BMI) from P2 to P4. (C–E) Images of a coronal section of a P4 FFL mouse that received intraperitoneal injection of BMI, followed by immunostaining for Cux1 (red, C) and GFP (green, D). (E) Merged images of (C) and (D) show disturbances in the distinct distribution pattern of Cux1-positive and GFP-positive cells at the boundary of the lesioned area after in vivo application of BMI. (F) Effect of BMI on the distribution of GFP-positive cells in four different regions (superficial, exofocal control, inner rim, and outer rim). The positions of cell-count square frames (50 μm × 50 μm) are shown in (C). Two-way analysis of variance (ANOVA) (region × drug) revealed a significant region effect (P < 0.001), a drug effect (P < 0.01), and a significant region × drug interaction (P < 0.001). Note that the significant difference between the inner rim and the outer rim was abolished by the application of BMI (post-hoc Tukey test, **P < 0.01, ***P < 0.001; saline (black symbols), five preparations; BMI (gray symbols), nine preparations). Significant differences in drug effects were region-specific (independent two-tailed t-test, #P < 0.05, ###P < 0.001). (G) Effect of BMI on the distribution of Cux1-positive cells in four different regions. Two-way ANOVA (region × drug) revealed a significant region effect (P < 0.001). Note that the significant difference between the inner rim and the outer rim was abolished by the application of BMI (post-hoc Tukey test, *P < 0.05, **P < 0.01, ***P < 0.001, and independent two-tailed t-test, P > 0.05, saline (black symbols); BMI (gray symbols). Scale bars: (A,B), 100 μm; (E) 200 μm. Adapted from Wang et al.,[11] available at http://cercor.oxfordjournals.org/content/early/2012/12/16/cercor.bhs375.full?keytype=ref&ijkey=i04VZWybp7NFEDX. image, saline; image, BMI.

Quantitative analysis by cell counting revealed that the preferential accumulation of GFP-positive GABAergic cells in the inner rim versus the outer rim was abolished by intraperitoneal BMI injection (Fig. 5F, see also Fig. 2G). In addition, cell density in the superficial region and the inner rim was significantly decreased by BMI treatment, while that in the outer rim was increased (Fig. 5F). Likewise, the preferential accumulation of Cux1-positive cortical plate cells in the outer rim versus the inner rim was abolished by intraperitoneal BMI injection (Fig. 5G).

Conclusion

The temporally increased ambient GABA causes tonic GABAA receptor activation, possibly depolarization, of NKCC1-upregulated and KCC2-downregulated neurons, resulting in modulation of Ca2+ oscillations. These Ca2+ signals could differentially affect the migratory status of GABAergic (“go”) and E17.5-born Cux1-positive cortical plate cells (“stop”). Therefore, immature cortical plate neurons and GABAergic neurons anomalously migrate with distinct patterns to form the microgyrus. Thus, abnormally induced mulitimodal GABA functions could be involved in migration disorders that ultimately result in cortical malformations.

Acknowledgments

Original studies contributed to this review were supported by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science; and by Grants-in-Aid for Scientific Research on Priority Area and on Innovative Area from the Ministry of Education, Culture, Sports, Science and Technology, Japan. The authors thank all the contributors of original papers: C. Shimizu-Okabe, T. Kumada, T. Morishima, S. Iwata, T. Kaneko, Y. Yanagawa, S. Yoshida, A. Okabe, W. Kilb, K. Sato and H.J. Luhmann, and also thank T. Furukawa for excellent technical assistance.

Ancillary