LGI1-associated epilepsy through altered ADAM23-dependent neuronal morphology

Abstract

Most epilepsy genes encode ion channels, but the LGI1 gene responsible for autosomal dominant partial epilepsy with auditory features produces a secreted protein. LGI1 is suggested to regulate PSD-95 via ADAM22. However, no unbiased screen of LGI1 action has been conducted. Here, we searched for brain genes supporting high affinity LGI-1 binding. ADAM23 was the only LGI1 interactor identified. The related proteins, ADAM22 and ADAM11, but not ADAM12, bind LGI1. Neither ADAM23 nor ADAM11, nor some forms of ADAM22, contain PDZ-interacting sequences, suggesting PSD-95-independent mechanisms in ADPEAF. Because ADAMs modulate integrins, we examined LGI1 effect on neurite outgrowth. LGI1 increases outgrowth from wild-type but not ADAM23-/- neurons. Furthermore, CA1 pyramidal neurons of ADAM23-/- hippocampi have reduced dendritic arborization. ADAM23-/- mice exhibit spontaneous seizures, while ADAM23+/- mice have decreased seizure thresholds. Thus, LGI1 binding to ADAM23 is necessary to correctly pattern neuronal morphology and altered anatomical patterning contributes to ADPEAF.

Figure 1. The secreted ADPEAF protein LGI1 binds to neurons

(A) The LGI1-AP fusion protein (WT) is secreted from transfected HEK293T cells. The AP-LGI1-F318C mutant (MU) is detectable in the cell lysate but is not secreted into the medium. Anti-AP immunoblot. (B) LGI1-AP fusion protein, but not AP, binds to mouse E18 DRG and hippocampal neurons. Bound AP is detected by NBT/BCIP reaction.

Figure 2. Expression cloning identifies ADAM23 as a high affinity LGI-1 binding site

(A) COS-7 cells were transfected with empty vector or ADAM23 expression vector and then exposed to 30 nM AP or LGI1-AP for 1 hr. Bound exogenous heat-stable AP was visualized as a dark enzyme reaction product after washing, fixing and heat inactivating endogenous AP. Scale bar is 100 μm. (B) LGI1-AP binding to COS-7 cells transfected with ADAM23, ADAM22, ADAM11, ADAM12 and vector. Scale bar is 100 μm. (C) The amount of LGI1-AP bound to ADAM23, ADAM22, ADAM11, ADAM12 and vector-expressing COS-7 cells is plotted as a function of ligand concentration. Data are mean ± sem from 3 independent determinations. (D) The data from C are replotted by the Scatchard method. The apparent ADAM23 Kd is 10 nM.

Figure 3. Extracellular LGI1 stimulates neurite outgrowth

(A) Exposure of chick E7 DRG or rat E18 hippocampal neurons to 100 nM LGI1-AP increases neurite outgrowth. (B) Quantitation of hippocampal neurite outgrowth as a function of AP or LGI1-AP concentration. Data are mean ± sem from 3 independent experiments. *, P<0.05 versus AP, Student's two-tailed t test.

Figure 4. Distribution of ADAM23 expression in the CNS

(A-D) β-Galactosidase staining of coronal brain sections from adult ADAM23 +/- mice reveals the distribution of cells expressing ADAM23. Note the widespread expression in pyramidal cells of both cerebral cortex (A-C) and hippocampus (B-C), in Purkinjie cell of the cerebellum (D) and in numerous diencephalic and brain stem nuclei (A-D). There is an absence of ADAM23 expression in the dentate gyrus of the hippocampus (B-C). (E) Parasagittal section of adult ADAM23 +/- mice stained for β-galactosidase activity. (F) There is no β-galactosidase reaction product in WT mouse brain sections.

Figure 5. Position of cells expressing ADAM23 does not require the protein

(A-D) Coronal sections of P12 mouse brain from an ADAM23 +/- (A, C) or ADAM23 -/- (B, D) mouse were processed for β-galactosidase. There is no evidence for cell migration abnormalities in the cerebral cortex or hippocampus in mice lacking ADAM23.

Figure 6. Reduced ability of LGI1 to stimulate hippocampal and cortical outgrowth from ADAM23-/- neurons

Cortical (A) and hippocampal (C) cultures from ADAM23+/+, ADAM23+/- and ADAM23-/- P1 mice exposed to purified AP or LGI1-AP protein at 100 nM ligand for 24 hours before fixation and staining of neuronal processes with anti- βIII-tubulin (dark in this inverted fluorescence image). Images were obtained at 10× magnification. Average outgrowth from P1 mouse cortical (B) and hippocampal (D) neurons was measured in μm/cell for ADAM23+/+ (n=6 separate cultures, each from a different mouse), ADAM23+/- (n=6) and ADAM23-/- (n=7) after culture in buffer, 100 nM AP or 100 nM LGI1. Data are mean ± sem. ANOVA: * p<0.05, ** p<0.01. Cell number varied by less than 15% between different conditions in any one experiment.

Figure 7. CA1 dendritic arborization and outgrowth in ADAM23+/+, AADAM23 +/- and ADAM23-/- mice

(A) At left, micrographs illustrate Golgi strains of CA1 pyramidal from mice of the indicated genotypes. At right, camera Lucida drawings of collapsed Z-stack of all basal dendrites in coronal 40 μm sections of CA1 pyramidal neurons from ADAM23+/+, ADAM23+/- and ADAM23 -/- P10 mice. Scale bar is 50 μm. (B) Average area enclosed by the basal dendritic arborization of ADAM23+/+ (n=43 cells from 6 mice), ADAM23+/- (n=33 from 6 mice), ADAM23-/- (n=47 from 6 mice) mice. Data are mean ± sem. Student's two-tailed t-test: * p<0.05, ** p<0.01. (C) Scholl analysis of the average number of basal dendrites crossing concentric circles at the specified distances from the cell soma. For ADAM23+/+, n=43 from 6 mice; ADAM23+/-, n=33 from 6 mice; and ADAM23-/-, n=47 from 6 mice. Data are mean ± sem. Repeated measures ANOVA: * p<0.05, ** p<0.01.

Figure 8. Reduced CA1 dendritic arbor in postnatal day 2 ADAM23-/- mice

(A) CA1 pyramidal from P2 mice of the indicated genotypes were impregnated by the Golgi method and basal dendrites of CA1 pyramidal neurons were assessed. Average area enclosed by the basal dendritic arborization of ADAM23+/+ (n=15 cells from 2 mice), ADAM23-/- (n=42 from 3 mice) mice. Data are mean ± sem. Student's two-tailed t-test: * p<0.05. (C) Scholl analysis of the average number of basal dendrites crossing concentric circles at the specified distances from the cell soma. For ADAM23+/+, n=15 from 2 mice and ADAM23-/-, n=42 from 3 mice. Data are mean ± sem. Repeated measures ANOVA: * p<0.05, ** p<0.01.

Figure 9. Epileptic events in ADAM23-/- mice and reduced seizure threshold in ADAM23+/- mice

(A) Head position plotted as a function of time for P10 ADAM23-/- mouse. Top trace (--) represents repetitive clonic movements of a seizure and bottom trace (--) represents irregular ataxic movements in the same mouse. (B) The frequency of convulsive episodes illustrated in (A) reported for a 60 min observation period (horizontal lines) for ADAM23 +/+, ADAM23+/- and ADAM23-/- P10 mice. Each occurrence is represented as a vertical line. (C) EEG recording from a P10 Adam23-/- mouse illustrates an example of tactile stimulation leading to tonic seizure. (D) Response of 6 month old ADAM23+/+ and +/- mice to single sc injection of PTZ (mg/kg). Data are mean ± sem, student t-test * p<0.05, ** p<0.01. ADAM23+/+ 20mg/kg n=3, 25mg/kg n=14, 30mg/kg n=18, 50mg/kg n=3. ADAM23+/- 20mg/kg n=3, 25mg/kg n=19, 30mg/kg n=27, 50mg/kg n=6. (E) Short term kindling of 6 month old ADAM23+/+ and ADAM23+/- mice to repeated (every 30min.) sub convulsive (15mg/kg) sc injection of PTZ (mg/kg). Data are mean ± sem, student t-test * p<0.05, ** p<0.01, ADAM23+/+ n=40, ADAM23+/- n=50.