Defective function of GABA-containing synaptic vesicles in mice lacking the AP-3B clathrin adaptor

Abstract

AP-3 is a member of the adaptor protein (AP) complex family that regulates the vesicular transport of cargo proteins in the secretory and endocytic pathways. There are two isoforms of AP-3: the ubiquitously expressed AP-3A and the neuron-specific AP-3B. Although the physiological role of AP-3A has recently been elucidated, that of AP-3B remains unsolved. To address this question, we generated mice lacking mu3B, a subunit of AP-3B. mu3B-/- mice suffered from spontaneous epileptic seizures. Morphological abnormalities were observed at synapses in these mice. Biochemical studies demonstrated the impairment of gamma-aminobutyric acid (GABA) release because of, at least in part, the reduction of vesicular GABA transporter in mu3B-/- mice. This facilitated the induction of long-term potentiation in the hippocampus and the abnormal propagation of neuronal excitability via the temporoammonic pathway. Thus, AP-3B plays a critical role in the normal formation and function of a subset of synaptic vesicles. This work adds a new aspect to the pathogenesis of epilepsy.

Figure 1.

Targeted disruption of μ3B gene by homologous recombination. (A) Structure of mouse μ3B gene including the exon encoding ATG start codon (wild type); the targeting vector containing 5′ (3.0 kb) and 3′ (1.5 kb) homologous regions, the EGFP gene, the Neo gene flanked with two loxP sites, and the herpes simplex virus-thymidine kinase (HSV-tk) gene for positive and negative selections (targeting vector); the resultant mutant allele generated by homologous recombination (mutant); and the mutant allele lacking the Neo gene after crossing with Cre-transgenic mice (ΔNeo). The following restriction enzyme sites are indicated: H, HindIII; Sc, SacI; Se, SpeI; and Sh, SphI. An additional HindIII site, shown by (H)*, exists only in the genome from C57BL/6 mice and not in that from 129 mice. (B–D) Southern blotting using 5′ (probe L) and 3′ (probe S) probes as depicted in A. Digestion of genome DNA with HindIII and SpeI yielded a 3.6-kb wild-type band and a 2.5-kb band generated by homologous recombination using probe S (B). Similarly, digestion of genome DNA with SacI and SphI yielded a 6.4-kb wild-type band and a 5.5-kb band generated by homologous recombination using probe L (C). The mutant allele digested with HindIII and SacI became 1.2 kb shorter (from 6.5 to 5.3 kb) using probe L after deletion of the Neo gene by crossing with Cre-transgenic mice (D, compare first and second lanes). Data shown in D were obtained using mice with C57BL/6 background. Note that the additional HindIII site ((H)*, A) in C57BL/6 genome gave a 2-kb band corresponding the C57BL/6 wild-type allele (third lane). (E) RT-PCR analysis using total RNA from brain (left) or spinal cord (right) as a template for PCR. (F) Whole brain lysates from wild-type, μ3B^+/−ΔNeo and μ3B^−/−ΔNeo mice were subjected to immunoblotting with anti-μ3 (top left) and anti-β3B (top right) antibodies. After stripping off the antibodies, both membranes were reblotted with anti–α adaptin (bottom) as an internal control for the amount of protein in each lane.

Figure 2.

Increased seizure susceptibility of μ3B ^−/− ΔNeo mice. (A) Representative electrocorticogram recorded from μ3B^−/−ΔNeo mice (n = 7) during the interictal period. Interictal spikes are underlined. (B) Seizure was induced by intravenous administration of PTZ, a GABA_A receptor antagonist, in 4-wk-old (left) and 8-wk-old (right) wild-type (closed circle; 4 wk old, n = 6; 8 wk old, n = 5) and μ3B^−/−ΔNeo (closed square; 4 wk old, n = 4; 8 wk old, n = 5) mice. Seizure stages were judged as described previously (see Materials and methods). Results are expressed as means ± SEM (* indicates P < 0.05; ** indicates P < 0.01). (C) Seizure development by amygdala kindling. Kindling experiments were performed as described previously (see Materials and methods), and the development of seizure classes in individual animals is plotted against the number of kindling stimulations. (D) Typical afterdischarge of wild-type (top) or μ3B^−/−ΔNeo mice (bottom) induced by the first kindling stimulation.

Figure 3.

Ultrastructural analysis of synaptic terminals. (A–D) Electron micrographs of asymmetric (excitatory) terminals (A and B) and symmetric (inhibitory) terminals attaching to perikarya of CA1 pyramidal neurons (C and D) in the CA1 in wild-type (A and C) and μ3B^−/−ΔNeo (B and D) mice at the age of 8 wk. Bar, 200 nm. (E–H) Developmental changes in the number of synaptic vesicles per unit area (E and F) and the diameter of synaptic vesicles (G and H) of excitatory (E and G) and inhibitory terminals (F and H) in μ3B^−/−ΔNeo mice (open square) and wild-type mice (solid square). 30–50 areas (E and F) and >500 synaptic vesicles (G and H) from both genotypes (n = 3 each) were counted, respectively. Values are expressed as means ± SEM Student's or Welch's t test was used to determine the differences between μ3B^−/−ΔNeo and wild-type mice (* indicates P < 0.05; ** indicates P < 0.01).

Figure 4.

Impairment of GABA release due to reduction of VGAT in μ3B ^−/− ΔNeo mice. (A and B) Measurement of glutamate (A) and GABA (B) release was performed as described in Materials and methods. Basal (circles) and K⁺-evoked (squares) release of wild-type (closed symbols) and μ3B^−/−ΔNeo mice (open symbols) is shown (n = 3 each). Results are expressed as means ± SEM (* indicates P < 0.05; ** indicates P < 0.01). (C) Western blotting of VGAT, VGLUT1, VGLUT2, synaptophysin, synaptotagmin, rabphilin-3A, Rab3A, and VAMP2 in total (lanes 1, 2, 5, and 6) and synaptosomal (lanes 3 and 4) or LP2 (lanes 7 and 8) lysates from hippocampus (left) and whole brain (right) of wild-type (lanes 1, 3, 5, and 7) and μ3B^−/−ΔNeo (lanes 2, 4, 6, and 8) mice. Shown are the representatives of four independent experiments. (D) Quantitative analysis of the amount of VGAT protein in total (left) and synaptosomal (right) hippocampal lysates of wild-type and μ3B^−/−ΔNeo mice (n = 4 each) as shown in C. Results are expressed as means ± SD. (* indicates P < 0.05). AU, arbitrary unit.

Figure 5.

Enhanced synaptic potentiation through reduced GABAergic synaptic inhibition in μ3B ^−/− ΔNeo mice. (A–D) The average time course of the slope of synaptic responses in μ3B^−/−ΔNeo mice and their littermate wild-type mice. Initial EPSP slopes were normalized in each experiment to the average slope value of the baseline (−30–0 min). The potentiation ratio was calculated by dividing the average slope value from 50 to 60 min by that of the baseline. Afferent fibers were tetanized at time 0 at: (A) 100 Hz for 1 s in the presence of PTX in wild-type (open circles; n = 18) and μ3B^−/−ΔNeo (closed circles; n = 19) mice; (B) 100 Hz for 1 s in the absence of PTX in wild-type (open circles; n = 16) and μ3B^−/−ΔNeo (closed circles; n = 13) mice; (C) 100 Hz for 200 ms in the presence of PTX in wild-type (open circles; n = 10) and μ3B^−/−ΔNeo (closed circles; n = 10) mice; and (D) 100 Hz for 200 ms in the absence of PTX in wild-type (open circles; n = 14) and μ3B^−/−ΔNeo (closed circles; n = 15) mice.

Figure 6.

Abnormal propagation of neuronal excitability via TA pathway in μ3B ^−/− ΔNeo mice. (A) Schematic of neuronal circuits in EC–hippocampus pathway. Black arrows indicate the trisynaptic pathway that includes the perforant, mossy fiber and Schaffer collateral pathways. Red and blue arrows indicate excitatory and inhibitory projections in the TA pathway, respectively. (B) Change in fluorescence signal associated with propagation of neuronal excitabilities in EC–hippocampal formation of 4-wk-old (4w) and 8-wk-old (8w) wild-type (+/+) and μ3B^−/−ΔNeo (−/−) mice after the electrical stimulation of EC layers II, III, and IV. Pseudo colors indicate >10³-fold (change in fluorescence intensity/initial fluorescence intensity = ΔF/F) elevation in the amplitude of the fluorescence signal. (C) Mean values of optical signals in dentate gyrus and CA1 regions after the electrical stimulation of EC layers II, III, and IV in 4-wk-old wild-type (closed circle, n = 5), 4-wk-old μ3B^−/−ΔNeo (open circle, n = 5), 8-wk-old wild-type (closed square, n = 5), and 8-wk-old μ3B^−/−ΔNeo (open square, n = 4) mice. Ordinates indicate the change in fluorescence intensity/initial fluorescence intensity (ΔF/F), and abscissas indicate time after stimulation (msec).