LAMP5 Fine-Tunes GABAergic Synaptic Transmission in Defined Circuits of the Mouse Brain

Abstract

LAMP5 is member of the LAMP family of membrane proteins. In contrast to the canonical members of this protein family, LAMP1 and LAMP2, which show widespread expression in many tissues, LAMP 5 is brain specific in mice. In C. elegans, the LAMP5 ortholog UNC-46 has been suggested to act a trafficking chaperone, essential for the correct targeting of the nematode vesicular GABA-transporter UNC-47. We show here that in the mouse brain LAMP5 is expressed in subpopulations of GABAergic forebrain neurons in the striato-nigral system and the olfactory bulb. The protein was present at synaptic terminals, overlapping with the mammalian vesicular GABA-transporter VGAT. In LAMP5-deficient mice localization of the transporter was unaffected arguing against a conserved role in VGAT trafficking. Electrophysiological analyses in mutants showed alterations in short term synaptic plasticity suggesting that LAMP5 is involved in controlling the dynamics of evoked GABAergic transmission. At the behavioral level, LAMP5 mutant mice showed decreased anxiety and deficits in olfactory discrimination. Altogether, this work implicates LAMP5 function in GABAergic neurotransmission in defined neuronal subpopulations.

Fig 1. Differential expression of LAMP5 mRNA and protein in the brain.

(A,C) In situ hybridization and (B,D) immunohistochemistry for LAMP5 on sagittal brain sections (A,B) and on coronal olfactory bulb (OB) sections (C,D). Strongest expression of LAMP5 mRNA is found in the neocortex (CX), piriform cortex (Pir), hippocampus (Hp), striatum (ST) and the granular cell layer (GCL) of the OB. LAMP5 protein is strongly present in the Globus Pallidus/Ventral Pallidum complex (GP/VP), the Substantia Nigra pars reticulata (SNr) and the entopeduncular nucleus (EP), that are the main output structures of the striatal GABAergic projection neurons, and in the external plexiform layer (EPL), in which granule cells positioned in the granule cell layer (GCL) form GABAergic synapses. (E-F) Schematic representation of LAMP5 mRNA (light grey) and protein (dark grey) expression in the rodent forebrain (E) and OB (F). (G) qRT-PCR analysis of LAMP5 expression in different brain tissue samples. Coherent with immunohistochemical stainings, strongest expression of LAMP5 mRNA is detected in the cortex, striatum and the OB. (H) Western blotting and its quantification (I) demonstrates that LAMP5 protein is strongly expressed in GP and SNr while striatal tissue (ST) and cortex show only weak signal. Th: thalamus; Scale bar: 1 mm in A,B; 0.5 mm in C,D.

Fig 2. LAMP5 deficient mice validate mAb 34.2 anti-LAMP5 antibody specificity.

(A) Targeting strategy to generate a LAMP5 deficient mouse line. The targeting vector was designed to remove exons 3–5 after CRE-induced recombination, leading to a null allele. LAMP5 null animals are viable and therefore used for most analyses. (B) Western blot of KO (-/-), heterozygote (+/-) and wild type (+/+) mouse brains demonstrates absence of protein in homozygous KOs. (C) Immunohistochemistry with mAb 34.2 anti-LAMP5 antibody on LAMP5^+/+ and LAMP5^-/- tissue sections of GP and OB validates the absence of LAMP5 protein in KOs. Cortex was always negative. Scale bars: C, 200 μm for GP and OB; 100 μm for cortex.

Fig 3. LAMP5 is specifically expressed in GABAergic synapses.

(A) Immunofluorescence labeling for LAMP5 and GAD65 proteins demonstrates co-labelling in GABAergic axon terminals in the GP (arrowheads). (B) LAMP5 immunoreactivity never overlaps with vGLUT1 (arrowheads), thus it is absent from glutamatergic synapses. (C, C') In the OB, LAMP5 is localized on the synapses (arrowheads) of newly generated granule neurons, labeled with GFP by in vivo brain electroporation. (D) Electron microscopy immunogold labeling validates synaptic localization of LAMP5. Electron dense gold particles are always associated with symmetric synaptic densities (arrow) typical of GABAergic synapses formed by granule neurons onto mitral cell dendrites. (E) Like in the GP, glutamatergic synapses in the OB are devoid of LAMP5 staining (arrowheads). Scale bars: 5 μm in A,B,C,E, F; 0,1 μm in D.

Fig 4. LAMP5 is present on VGAT positive synaptic vesicles.

(A) Histological sections of the GP and the OB immunostained for LAMP5 and VGAT. LAMP5 labeling largely overlaps with VGAT immunoreactivity in both structures (arrowheads) (B) Quantitative evaluation of synaptic puncta shows that the vast majority of LAMP5 positive synapses co-express VGAT (n = 4 photomicrographs for GP and for OB). (C) Immunoisolation of vesicles using a VGAT antibody followed by western blotting. LAMP5 is specifically found in the VGAT positive fraction. Scale bar: 5 μm.

Fig 5. LAMP5 knockout mice show normal VGAT distribution.

(A, B) Immunostaining of LAMP5 and VGAT on GP (A) and OB (B) histological sections show no change in tissue and subcellular distribution in the absence of LAMP5. (C) Signal intensity of VGAT staining in the GP and in the OB, assessed by ImageJ analysis, is also unchanged. Mean intensity was calculated over 3 photomicrographs. Scale bar: A, 100 μm (left), 5 μm (right); C, 50 μm (left), 5 μm (right).

Fig 6. Absence of LAMP5 does not alter brain structure nor spine density in the OB.

(A) Cresyl violet staining of forebrain and OB sections shows no obvious structural changes in the brain in the absence of LAMP5. (B) Dendrites of control (LAMP5 flox/+) and mutant (LAMP5 flox/flox) newly generated granule neurons observed in the EPL of the OB 28 days after their electroporation (28dpe) with GFP and Cre recombinase expression plasmids. (C) Quantification of spine density on the dendrites of control (CTL) and KO newly integrated neurons. Dendrite morphology and spine density are unchanged in LAMP5 KO. Dendrites n = 26 (flox/+) and 27 (flox/flox). Statistics: Wilcoxon test p-value = 0.9645. Scale bars: A, 1mm (0.5 mm for the OB); B, 20 μm (left), 5 μm (right). ST: Striatum; Th: Thalamus.

Fig 7. Short-term plasticity of the striatopallidal synapse is altered in LAMP5 deficient mice.

(A) Top: Representative traces of mIPSCs from WT and KO mice. Bottom: Cumulative inter-event interval (left) and amplitude (right) distributions of mIPSCs obtained in WT and KO mice (n = 400 events per cell). The frequency of mIPSCs is significantly increased in LAMP5 KO compared to WT mice (p < 0.001; WT amplitude n = 3 mice, n = 4 slices, n = 7 cells / KO amplitude n = 2 mice, n = 4 slices, n = 5 cells; WT frequency n = 3 mice, n = 4 slices, n = 7 cells / KO frequency n = 2 mice, n = 4 slices, n = 6 cells). (B) Parasagittal section of the mouse brain showing the recording and stimulation sites. Evoked striatopallidal GABAergic IPSCs are blocked by the GABA_A receptor antagonist, picrotoxin (ptx). St: striatum, GP: globus pallidus. (C) Mean PPR values from WT and KO mice at different interstimulus intervals. Sample traces of PPR are shown above the graph (traces were scaled to first IPSCs). * p < 0.05, ** p < 0.01 vs. WT mice. (D) Synaptic depression during repeated stimulation (10 pulses at 20 and 50 Hz) in WT mice was replaced by facilitation in KO mice. Representative traces to 20 and 50 Hz trains are shown in WT and KO mice. * p < 0.05, ** p < 0.01 vs. WT mice. Error bars represent SEM.

Fig 8. Absence of LAMP5 disrupts olfaction and leads to anxiety-like behavior.

(A) Spontaneous activity of naive wild type (open circles) and KO mice (dark circles). Horizontal locomotor activity is recorded every 10 min for 2 hours. The exploration is similar over time between WT (n = 11) and KO (n = 16). (B) Motor coordination is assessed in a rotarod essay where mice are placed on a cylinder that rolls at increasing speed with time. The latency before the mice fall is measured. No difference is found between WT (n = 8) and KO (n = 7) mice. Kruskal-Wallis test p-value = 0.4875. (C) Catalepsy test. WT and KO mice are injected with haloperidol (i.p. 1mg/kg). The length of time the animals stay in a constrained position is recorded to a maximum of 300 seconds every 20 min. (WT n = 7; KO n = 7). ANOVA testing shows no genotype effect (F_1,13 = 0.213, p-value = 0.652) and no interaction between time and genotype (F_7,91 = 0.558; p = 0.788). (D) Elevated Plus Maze test. Left: Total number of entries in closed and open arms, an index of activity. Middle: percentage of time spent into the open arms. Right: percentage of open arm entries. The mutant mice are less anxious as they spend significantly more time exploring the open arms (Kruskal-Wallis test p-value = 0.01221) and enter preferentially into the open arms (Kruskal-Wallis test p-value = 0.03415; WT n = 11, KO n = 16). (E) Odor discrimination test. A first odorant (Oh: habituation odor) is presented for four consecutive trials followed by the presentation of a test odor (Ot). The time spent by the mice investigating the respective odorant is recorded. Habituation occurs similarly for both WT and KO mice as shown by a significant decrease in investigation time between the first (Oh1) and last (Oh4) trials (*, p<0.05; **, p<0.01; ***, p<0.001). Discrimination is partially impaired in KO mice as they do not discriminate between isoamyl butyrate and heptanol or between hexanol and limonene^- (Ot vs Oh4: $, p<0.05; $ $, p<0.01; ns: not significant). Statistics: Wilcoxon test. Number of animals: isoamyl butyrate/ heptanol WT n = 14, KO n = 27; hexanol/limonene WT n = 10, KO n = 13; isoamyl actetate/carvon WT n = 7, KO n = 7. Error bars represent SEM.