A MusD retrotransposon insertion in the mouse Slc6a5 gene causes alterations in neuromuscular junction maturation and behavioral phenotypes

Abstract

Glycine is the major inhibitory neurotransmitter in the spinal cord and some brain regions. The presynaptic glycine transporter, GlyT2, is required for sustained glycinergic transmission through presynaptic reuptake and recycling of glycine. Mutations in SLC6A5, encoding GlyT2, cause hereditary hyperekplexia in humans, and similar phenotypes in knock-out mice, and variants are associated with schizophrenia. We identified a spontaneous mutation in mouse Slc6a5, caused by a MusD retrotransposon insertion. The GlyT2 protein is undetectable in homozygous mutants, indicating a null allele. Homozygous mutant mice are normal at birth, but develop handling-induced spasms at five days of age, and only survive for two weeks, but allow the study of early activity-regulated developmental processes. At the neuromuscular junction, synapse elimination and the switch from embryonic to adult acetylcholine receptor subunits are hastened, consistent with a presumed increase in motor neuron activity, and transcription of acetylcholine receptors is elevated. Heterozygous mice, which show no reduction in lifespan but nonetheless have reduced levels of GlyT2, have a normal thermal sensitivity with the hot-plate test, but differences in repetitive grooming and decreased sleep time with home-cage monitoring. Open-field and elevated plus-maze tests did not detect anxiety-like behaviors; however, the latter showed a hyperactivity phenotype. Importantly, grooming and hyperactivity are observed in mouse schizophrenia models. Thus, mutations in Slc6a5 show changes in neuromuscular junction development as homozygotes, and behavioral phenotypes as heterozygotes, indicating their usefulness for studies related to glycinergic dysfunction.

Figure 1. Overt phenotype and mapping of the new mutation.

A. P15 mutant (left, white) and unaffected littermate (right, brown) in the NOD x DBA/2 mapping cross. Picture was taken at the time of a generalized tremor of the mutant. B. Growth curve for wild-type, heterozygous and homozygous mutants, completed once the genotyping assay was established. C. Haplotype of affected (left) and unaffected (right) mice at three loci of chromosome 7 flanking the Slc6a5 (Slc6a5) gene. Simple sequence length polymorphism (Mit markers) different in NOD and DBA/2 and their positions in megabases (Mb) are indicated. The number of mice of each genotype is indicated below the diagram. D. Schematic representation of chromosome 7 with centromere at the top, location of MIT markers used and Slc6a5. Positions are given in megabases (Mb) on the left and centimorgans (cM) on the right.

Figure 2. Characterization of the new Slc6a5 mutation as a null-allele.

A. Schematic of the Slc6a5 gene organization, with boxes and lines representing exons and introns respectively, and boundaries of PCRs (1 to 4) run on reverse-transcribed cDNAs. B. Agarose gel of the PCR amplification products from cDNA. Note the extra band in PCR 2 on mutant brain cDNAs (arrow). C. Schematic representation of intron 5, and location of the insertion and splice donor and acceptor sites in the retrotransposon sequence. The 10 base pairs immediately 5′ of the insertion site, 1833 bp into intron 5, are shown, as well as the first base-pairs of the LTR element of the retrotransposon. The 183 bp sequence of the transposon that is spliced-in Slc6a5 mRNA is indicated with its 10 first and last base-pairs and the inferred acceptor and donor splice sites. The primers used to localize the insertion (int5F and MusR), and to amplify the transposon from mutant genomic DNA (int5F and int5R) are indicated by arrows. D. Agarose gel of the products of the PCR on genomic DNA with primers flanking the insertion site (int5F/R of C.). A 3 kb product was amplified from wild-type gDNA, but only a 9 kb product was amplified from mutant gDNA. E. Anti-GLYT2 Western blot on spinal cord cell membrane preparations, with beta-dystroglycan used as a loading control. Genotypes are indicated above the blot. F. Anti-GLYT2 immunocytochemistry (DAB) on brain sagittal sections and spinal cord cross sections. Nuclei were counterstained with hematoxylin before mounting. The lateral brain stem is shown in the top panels with the ventral lobules of the cerebellum visible in the top right corners and the ventral horn of the spinal cord in shown in the bottom panels.

Figure 3. Neuromuscular junction phenotypes.

A. Representative pictures of P7 NMJs in WT (left) and mutant (right) triangularis sterni muscles. The axons are labeled by the Thy1-YFP16 transgene (green) and the acetylcholine receptor plaques by fluorescent alpha-bungarotoxin (red). NMJs innervated by more than one axon terminal (arrowheads) are more abundant in the WT than in the mutant. B. Time course of the synapse elimination. The curves represent the average percentage of polyinnervated NMJs in wild-type (gray) and mutant (black) triangularis sterni muscles at ages P3 to P12. a: Wilcoxon signed-rank paired test, p = 0.09. The bars represent the average percentage of polyinnervated NMJs in the P3-P5 and P6-P9 age groups. * : Mann-Whitney test, p<0.05. Error bars represent the SEM. C. Schematic representation of the molecular switch from gamma to epsilon AChR subunits in the muscle pentameric acetylcholine receptor, also indicating the constitutive subunits alpha, beta and delta. D. Quantification of the fold-change of AChR subunits alpha, gamma and epsilon, and myogenin, in P5 and two week old synaptic regions of the mutant diaphragm, expressed as a ratio to age-matched WT values ( = 1.0). Biological replicates: 5 for P5, 7 for two week. Error bars represent the confidence interval. * : Mann-Whitney test, p<0.05.

Figure 4. Normal thermal nociception in heterozygotes.

Results of the hot-plate assay. Reaction time before flickering/retraction of one paw after mice have been placed on a 50 deg hot plate. Individual values are shown by circles, mean +/- SEM are shown by the chart and error bars. The number of mice tested is indicated.

Figure 5. Nocturnal behavioral phenotypes in the home cage monitoring system.

A. Table of the major behaviors that did not show any change between genotypes, in % of the total nocturnal time (mean and SEM). B. Behaviors with a significant change. a : p = 0.08 for the amount of sleep and b : p = 0.06 for the amount of grooming in the 7 months old mutants, * : p<0.05, ** : p<0.01, Mann-Whitney test. See “Methods” section for more detailed description. 6 female mice per age group and genotype were observed.

Figure 6. Anxiety-related behaviors.

A, B : Openfield test, C, D : elevated plus-maze test. Wild-type and heterozygotes values are shown for two age groups, 6 weeks and 7 months old. A. Average time spent per zone of the openfield, in sec., zone 1 being the most peripheral, zone 5 the central zone. B. Average track length per zone (in inches) during the 10 min observation. C. Average time spent in the open or closed arm of the elevated plus-maze. D. Average velocity in the open and closed arms. 6 mice per age group and genotype were observed. * : p<0.05, ** : p<0.01, Mann-Whitney test.