LMTK3 deficiency causes pronounced locomotor hyperactivity and impairs endocytic trafficking

Abstract

LMTK3 belongs to the LMTK family of protein kinases that are predominantly expressed in the brain. Physiological functions of LMTK3 and other members of the LMTK family in the CNS remain unknown. In this study, we performed a battery of behavioral analyses using Lmtk3(-/-) mice and showed that these mice exhibit abnormal behaviors, including pronounced locomotor hyperactivity, reduced anxiety behavior, and decreased depression-like behavior. Concurrently, the dopamine metabolite levels and dopamine turnover rate are increased in the striata of Lmtk3(-/-) mice compared with wild-type controls. In addition, using cultured primary neurons from Lmtk3(-/-) mice, we found that LMTK3 is involved in the endocytic trafficking of N-methyl-d-aspartate receptors, a type of ionotropic glutamate receptor. Altered membrane traffic of the receptor in Lmtk3(-/-) neurons may underlie behavioral abnormalities in the mutant animals. Together, our data suggest that LMTK3 plays an important role in regulating locomotor behavior in mice.

Keywords: LMTK; endocytosis; hyperactivity; locomotor; membrane trafficking.

Figure 1.

Expression profile of LMTK3 and generation of Lmtk3^−/− mice. A, Schematic diagram of LMTK3, WT allele, a targeting vector, and a targeted allele after homologous recombination. Exons 1–3 encoding 1–120 aa of Lmtk3, which include N-terminal hydrophobic membrane targeting sequences (filled boxes) and a part of the Ser/Thr kinase domain (gray box) were replaced with lacZ and Neo^r. B, Tail genomic DNA [WT (+/+), heterozygous mutant (+/−), and homozygous mutant (−/−)] was digested with BglII and hybridized with a 5′ probe (indicated in A) for Southern blot analysis. C, Genotyping PCR of tail genomic DNA [WT (+/+), heterozygous mutant (+/−), and homozygous mutant (−/−)] using primers Pr1, Pr2, and Pr3 (indicated in A). D, Western blot analysis of WT (+/+), heterozygous (+/−), and homozygous (−/−) mutant mouse brain. Newborn brain lysates were probed with anti-LMTK3, anti-CYTH2, anti-β-galactosidase, and anti-α-tubulin antibodies. Asterisks indicate nonspecific signals. E–H, X-Gal staining of sagittal (E) and coronal (H) sections of the P1 Lmtk3^+/− brain. F, G, High-magnification images of E to show strong expression in the cornu ammonis (CA) and dentate gyrus (DG) in the hippocampus (F) and prefrontal cortex (G). CTX, Cerebral cortex; HC, hippocampus; TH, thalamus. Scale bars: E, H, 0.5 mm.

Figure 2.

Hyperactive and reduced anxiety-like behaviors in Lmtk3^−/− mice. A–E, The open-field test. A, Increased horizontal activity of KO mice in a novel environment. WT, n = 10; Het, n = 10; KO, n = 9. F_(2,26) = 97.61, p < 0.0001 for genotype, two-way repeated-measures ANOVA. WT versus KO: p < 0.0001. B, Typical examples of locomotor patterns of WT (top) and KO (bottom) mice during 15 min recordings. C, Increased vertical activity in KO mice in a novel environment. n = 7 for each genotype. F_(1,12) = 5.34. *p = 0.0394 for WT versus KO in a novel environment, one-way ANOVA. D, KO mice spent less time immobile in both novel and familiar environments. WT, n = 10; Het, n = 10; KO, n = 9. F_(1,17) = 119.59. ***p < 0.0001 for WT versus KO in a novel environment, F_(1,17) = 44.21. ***p < 0.0001 for WT versus KO in a familiar environment, one-way ANOVA. E, Increased horizontal activity of KO mice in a familiar environment. WT, n = 10; KO, n = 9. F_(1,17) = 29.72, p < 0.0001 for genotype, two-way repeated-measures ANOVA. F–I, The elevated plus maze test. F, The increased total distance traveled by mutants. WT, n = 13; KO, n = 11, F_(1,22) = 117.8. ***p < 0.0001, one-way ANOVA. G, The increased total number of entries by mutants. WT, n = 13; KO, n = 11, F_(1,22) = 170.1. ***p < 0.0001, one-way ANOVA. H, Increased time spent in the open arms by the mutants. WT, n = 13; KO, n = 11, F_(1,22) = 70.3. ***p < 0.0001, one-way ANOVA. I, The increased number of entries into the open arms by mutants. WT, n = 13; KO, n = 11, F_(1,22) = 19.3. ***p < 0.001, one-way ANOVA. J, The increased number of transitions between the light and dark compartments by mutants during the light/dark transition test. WT, n = 13; KO, n = 11, F_(1,22) = 36.8. ***p < 0.0001, one-way ANOVA. K–M, Contextual and auditory fear conditioning tests. n = 15 for each genotype. K, Freezing responses on the conditioning day. A tone was presented for 10 s (solid line), and mice were given a footshock for 2 s (arrow). F_(1,28) = 5.354, p = 0.0282 for genotype, two-way repeated-measures ANOVA. L, Freezing responses in the contextual fear conditioning test 24 h after conditioning. F_(1,28) = 4.896, p = 0.0352 for genotype, two-way repeated-measures ANOVA. M, Freezing responses in the auditory fear conditioning test 48 h after conditioning. Three minutes after putting mice into a test chamber with novel contexts, the same tone was presented for 3 min (solid line). F_(1,28) = 4.796, p = 0.037 for genotype, two-way repeated-measures ANOVA. Data are mean ± SEM.

Figure 3.

Antidepression-like behaviors in Lmtk3^−/− mice. A, B, The tail suspension test. A, The decreased percentage immobility time by mutants. WT, n = 12; KO, n = 11, F_(1,21) = 30.81, p < 0.0001 for genotype, two-way repeated-measures ANOVA. B, A summary of the immobility time during the 6 min test. ***p < 0.001, one-way ANOVA. C, D, The forced swim test. C, The decreased percentage immobility time of mutants. WT, n = 13; KO, n = 11, F_(1,22) = 22.33, p < 0.0001 for genotype, two-way repeated-measures ANOVA. D, A summary of the immobility time during the 10 min test. ***p < 0.001, one-way ANOVA. Data are mean ± SEM.

Figure 4.

Behavioral analyses of Lmtk3^−/− mice by acoustic startle response and prepulse inhibition tests. A, Mean amplitudes of startle responses to acoustic stimuli. WT, n = 13; KO, n = 11, two-way ANOVA. F_(4,110) = 1.853, p = 0.1239 for startle × genotype interaction; F_(1,110) = 11.23, p = 0.0011 for genotype, followed by Bonferroni's test. **p = 0.0013 for 100 dB. B, Percentage prepulse inhibition. Prepulse sound at 75 or 80 dB was presented before the 120 dB startle stimuli. WT, n = 13; KO, n = 11, two-way ANOVA. F_(1,44) = 0.5071, p = 0.4801 for genotype.

Figure 5.

Behavioral analyses of Lmtk3^−/− mice by wire hang, tail flick, and accelerating rotarod tests. A, Wire hang test. Average latency to fall from the wire-mesh of two trials. n = 15 for each genotype. F_(1,28) = 2.95, p = 0.0967, one-way ANOVA. B, Tail flick test. Average tail flick latency in response to heat stimulus over three trials. WT, n = 13; KO, n = 12, F_(1,23) = 0.137, p = 0.704, one-way ANOVA. C, Accelerating rotarod test. Mice were trained for eight trials, four trials per day for two consecutive days. n = 15 for each genotype. F_(1,28) = 0.130, p = 0.718, two-way repeated-measures ANOVA.

Figure 6.

Effects of methylphenidate on locomotor activity. Horizontal activity was measured in the open-field test for 45 min after the injection of methylphenidate at the doses indicated (n = 12 [WT], n = 8 [KO] for vehicle (saline); n = 4 [WT], n = 10 [KO] for 0.05 mg/kg; n = 5 [WT], n = 4 [KO] for 0.3 mg/kg; n = 10 [WT], n = 6 [KO] for 3 mg/kg; n = 11 [WT], n = 8 [KO] for 10 mg/kg). F_(4,68) = 1.895, p = 0.1214 for dose × genotype interaction; F_(1,68) = 15.34, p = 0.0002 for genotype, two-way repeated-measures ANOVA, followed by Bonferroni's test. Data are mean ± SEM.

Figure 7.

Subcellular distribution of LMTK3. A, Cultured cortical neurons (DIV7) were immunostained with anti-LMTK3 (red) and anti-GM130 (green) antibodies. Neuronal somas were outlined in the merged image for clarification of the cell shape. Scale bar, 10 μm. B, Cultured cortical neurons (DIV14) were labeled with Alexa546-Tf (red) and immunostained with anti-LMTK3 antibody (green). Arrowheads indicate the colocalization of LMTK3 with Tf-positive endosomal vesicles in the neurites. Scale bar, 2 μm. C, Cultured cortical neurons (DIV14) were immunostained with anti-LMTK3 (green), anti-VGLUT1 (red), and anti-PSD-95 (blue) antibodies. Arrowheads indicate the presence of LMTK3 in the close vicinity of VGLUT1/PSD-95-positive excitatory synapses. Scale bar, 2 μm. D, Cultured cortical neurons (DIV14) from Lmtk3^+/+ (left) and Lmtk3^−/− (right) embryos were transfected with EGFP expression plasmid. Two days after transfection, neurons were immunostained with anti-LMTK3 (red) and anti-GFP (green) antibodies. LMTK3 immunofluorescence was confirmed to be rarely detectable in Lmtk3^−/− neurons. Neuronal morphology was assessed by GFP immunofluorescence. Spine densities were also quantified (n > 50 neurites in EGFP-expressing neurons of each genotype; p = 0.56). Scale bar, 10 μm. E, Biochemical fractionation of mouse brain lysate. An equal amount of protein from each fraction was probed with anti-LMTK3, anti-synaptophysin, and anti-PSD-95 antibodies. F, Biochemical preparation of the Golgi membrane fraction from mouse brain lysate by discontinuous sucrose density gradient. Equal amount proteins of each fraction were probed with anti-LMTK3, anti-GM130, anti-Rab8, and anti-EEA1 antibodies. G, Lysates from Lmtk3^+/+ or Lmtk3^−/− mouse brains were immunoprecipitated with an anti-LMTK3 antibody or a control rabbit IgG and analyzed using anti-α-adaptin antibody. Data are mean ± SEM; Student's t test.

Figure 8.

Impaired NMDA receptor trafficking in Lmtk3^−/− neurons. A, Biotinylation of surface proteins and endocytosis were performed as described in Materials and Methods. Total lysates, surface proteins, and endocytosed proteins of WT and KO neurons (DIV14) were analyzed by immunoblotting for GluN1, GluN2B, GluN2A, GluA1, TfR, GABA_ARβ2, and α-tubulin. Negligible α-tubulin immunoreactivity in “surface” and “endocytosed” lanes indicates successful biotin labeling and quenching of the surface proteins. B, Quantification of the immunoblot of A. n = 5 for GluN2B, GluN2A, and TfR; n = 3 for GluN1, GluA1, GluA2, and GABA_ARβ2. Data are mean ± SEM. *p < 0.05, Student's t test. ***p < 0.001, Student's t test. C, Cultured cortical neurons (DIV14) were immunostained with anti-LMTK3 (green) and anti-GluN1 (red) antibodies. Arrowheads indicate partial colocalization of LMTK3 with the NMDA receptors. Scale bar, 2 μm.