Fig 1.
Expression of angiomotin proteins in neurons.
(A) Western blot analysis of Amot expression in different regions of the mouse brain. “OB” = olfactory bulbs; “Mb” = midbrain; “Hipp” = hippocampus; “Th” = thalamus; “Cx” = cortex; “Cb” = cerebellum. (B, C) Amot localization in neurons in cryostat sections of the P30 mouse brain. (B) CA1 region of the hippocampus stained for Amot (green), Syn (blue), and Map2 (red). (C) Purkinje cell dendritic trees in the cerebellum stained for Amot (green), Syn (red), and calbindin (blue). (D) Model of hippocampal neuron development in vitro. Development of the somatodendritic compartment (black) and axon (blue) is correlated with DIVs on the axis at the bottom. (E) Amot expression increases at later stages of neuronal development. Neuronal extracts that were collected on the indicated DIVs were analyzed by western blot. (F, G) Localization of Amot protein in cultured neurons. (F) Rat hippocampal neurons (DIV7) were immunolabeled for Amot (green) and Map2 (red). (G) Amot localization in DIV21 cultured rat hippocampal neurons, showing Amot (green) and Syn (red). The lower panels in B, C, and G show higher magnifications of the boxed areas in the upper panels. Scale bar = 10 μm in B, C, F, and G (upper panel) and 5 μm in G (lower panel). Amot, angiomotin; DIV, day in vitro; Map2, microtubule-associated protein 2; P, postnatal day; Syn, synaptophysin.
Fig 2.
Amot depletion in cultured hippocampal neurons affects dendritic tree organization.
(A, B) Validation of Amot-knockdown efficiency in neurons by qRT-PCR (n = 3/group; p = 0.0041) (A) and western blot (B). (C) Rat hippocampal neurons were transfected on DIV7 with an empty pLKO plasmid (control), a plasmid with Amot shRNA, and a plasmid with Amot shRNA and a plasmid that expressed mouse Amot–GFP (Amot shRNA + Amot; rescue experiment). (D, E) Quantification of TDL (D) and Sholl analysis (E) of hippocampal neurons that were transfected with plasmids as in (C) or transfected with Amot–GFP plasmid. Control: n = 60; Amot: n = 38; Amot shRNA: n = 60; Amot shRNA + Amot: n = 58. To Control; p = 0.2678, p < 0.0001; to Amot shRNA; p < 0.0001. (F) Reduction of dendritic tree complexity in hippocampal neurons from Amot fl/fl mice that were transfected with a Cre-expressing plasmid. (G, H) Quantification of TDL (G) and Sholl analysis (H) of dendritic trees of neurons from (F). Control vector: n = 33; Cre plasmid: n = 40. p < 0.0001. (I) Cre expression in Amot fl/fl hippocampal neurons led to the down-regulation of Amot expression, determined by western blot. (J) Domain architecture of Amot protein. The Amot ID domain contains LPXY and PPXY motifs (red) that bind Yap1 and the AB (blue). (K) Quantification of TDL of Amot-knockdown rat hippocampal neurons that were transfected with the indicated Amot constructs. Control cells were transfected with an empty pLKO vector. Control: n = 107; Amot shRNA: n = 80; Amot shRNA + Amot: n = 100; Amot shRNA + Amot ID domain: n = 37; Amot shRNA + Amot ΔID domain: n = 45; Amot shRNA + Amot ΔAB: n = 59. To control; p < 0.0001; to Amot shRNA; p < 0.0001, p = 0.0064, p = 0.0015, p < 0.0001; to Amot shRNA + Amot; p = 0.0047, p = 0.0033, p = 0.09695. Images were obtained from at least three independent cultures. Statistical significance was analyzed using two-tailed and unpaired t tests (A, G), one-way analysis of variance followed by Tukey’s post hoc test (D, K), and two-way analysis of variance followed by Bonferroni’s test (E, H). Numerical values that underlie the graphs are shown in S1 Data. See S1 Table for detailed statistics for E and H. **p < 0.01, ****p < 0.0001. Bars represent the mean ± SEM. Scale bars = 50 μm. AB, actin-binding motif; Amot, angiomotin; CC, coiled-coil domain; DIV, day in vitro; GFP, green fluorescent protein; ID, intrinsic disordered; ns, not significant; PR, proline-rich region; qRT-PCR, quantitative real-time polymerase chain reaction; SEM, standard error of the mean; shRNA, short-hairpin RNA; TDL, total dendrite length; Yap1, Yes-associated protein 1.
Fig 3.
Amot interaction with Yap1 is required for proper growth of dendritic arbors.
(A) Schematic illustration of Amot protein with mutated Yap1-binding sites where PPXY motifs were modified into PPXA. (B) Rat hippocampal neurons that were transfected on DIV7 with the indicated constructs. (C) Quantification of TDL of Amot-knockdown rat hippocampal neurons that were cotransfected with plasmids as in B. Control cells were transfected with an empty pLKO vector. Control: n = 107; Amot shRNA: n = 80; Amot shRNA + Amot: n = 100; Amot shRNA + Amot ΔYap1: n = 59. To control; p < 0.0001; to Amot shRNA p < 0.0001; p = 0.3853. Images were obtained from at least from three independent cultures. (D) Yap1 phosphorylation at serine-127 was visibly reduced in Amot-depleted neurons, whereas the total cellular level of Yap1 was unchanged. (E) Yap1 coprecipitated with Amot from mouse brain homogenates. Uncoated magnetic beads were used as the control. (F) Western blot analysis of Yap1 expression in different regions of the mouse brain. “OB” = olfactory bulbs; “Mb” = midbrain; “Hipp” = hippocampus; “Th” = thalamus; “Cx” = cortex; “Cb” = cerebellum. (G, H) Yap1 localization in the brain, showing cryostat sections of the P30 mouse hippocampus (G) and cerebellum (H) that were stained for Yap1 (green) and Syn (red). Overlay images additionally showed nuclei, visualized with DAPI (blue in G), and Purkinje cell dendritic processes, visualized with anti-calbindin antibody (blue in H). (I) Yap1 expression increased at later stages of neuronal development. Neuronal extracts that were collected on the indicated DIVs were analyzed by western blot. (J) Yap1 localized to dendritic processes in young cultured neurons. Rat hippocampal neurons (DIV8) were immunolabeled for Yap1 (green) and Map2 (red). (K) Yap1 localized to synapses in mature cultured neuronal cells. Rat hippocampal neurons (DIV21) were immunolabeled for Yap1 (green) and Syn (red). The lower panels in G, H, and K show higher magnifications of the boxed areas in the upper panels. Scale bars = 50 μm in B; 10 μm in G, H, J, and K (upper panel); and 5 μm in K (lower panel). Numerical values that underlie the graphs are shown in S1 Data. Statistical significance was analyzed using one-way analysis of variance followed by Tukey’s post hoc test. ****p < 0.0001. Bars represent the mean ± SEM. AB, actin-binding motif; CC, coiled-coil domain; DIV, day in vitro; ID, intrinsic disordered; IP, immunoprecipitation; Map2, microtubule-associated protein 2; ns, not significant; P, postnatal day; PR, proline-rich region; SEM, standard error of the mean; shRNA, short-hairpin RNA; Syn, synaptophysin; TDL, total dendrite length; Yap1, Yes-associated protein 1.
Fig 4.
Yap1 regulates dendritic tree complexity.
(A) Efficacy of Yap1 shRNA in neurons, analyzed by western blot. (B) Rat hippocampal neurons were transfected on DIV7 with the indicated constructs. An empty pLKO plasmid was used as the control. (C, D) Quantification of TDL (C) and Sholl analysis (D) of hippocampal neurons that were transfected with plasmids as in B or cotransfected with Yap1 shRNA and human Yap1-cDNA with a deleted TEAD-binding domain (Yap1 shRNA + Yap1 del60-89). Control: n = 40; Yap1 shRNA: n = 60; Yap1 shRNA + Yap1: n = 47; Yap1 shRNA + Yap1 del60-89: n = 51. To control; p < 0.0001; to Yap1 shRNA; p = 0.0011, p = 0.0264; to Yap1 shRNA + Yap1; p = 0.7244. (E) Reduction of dendritic tree complexity in hippocampal neurons from Yap1 fl/fl mice that were transfected with a Cre-expressing plasmid. (F) Cre expression in Yap1 fl/fl hippocampal neurons led to the down-regulation of Yap1 expression, analyzed by western blot. (G, H) Quantification of TDL (G) and Sholl analysis (H) of dendritic trees of hippocampal Yap1 fl/fl neurons that were transfected with either a control vector (n = 53) or a Cre-expressing plasmid (n = 53). p < 0.0001. Images were obtained from at least from three independent cultures. Statistical significance was analyzed using two-tailed unpaired t test (G), one-way analysis of variance followed by Tukey’s post hoc test (C), and two-way analysis of variance followed by Bonferroni’s post hoc test (D, H). Numerical values that underlie the graphs are shown in S1 Data. See S1 Table for detailed statistics for D and H. *p < 0.05, **p < 0.01, ****p < 0.0001. Bars represent the mean ± SEM. Scale bar = 50 μm. DIV, day in vitro; ns, not significant; SEM, standard error of the mean; shRNA, short-hairpin RNA; TDL, total dendrite length; TEAD, TEA domain; Yap1, Yes-associated protein 1.
Fig 5.
Neuronal deletion of Amot affects cerebellar morphology.
(A) Strategy for the generation of Amot conditional knockout mice, showing the Amot flox allele (upper diagram) and recombined allele after Cre-mediated excision (lower diagram). (B) Lower Amot expression in the cerebellum of Amot fl/fl;Syn-Cre P30 mice, analyzed by western blot. (C) Amot fl/fl;Syn-Cre mice exhibited a smaller cerebellum compared with control Amot fl/fl littermates. (D) Quantitative analysis of cerebellar weight in Amot fl/fl (n = 9) and sex-matched Amot fl/fl;Syn-Cre (n = 6) mice on P30. p = 0.0411. The measurements were normalized to the whole-brain weight. (E) Sagittal cerebellar sections of Amot fl/fl and Amot fl/fl;Syn-Cre P30 mice that were immunolabeled for calbindin (green) to visualize Purkinje cells and the molecular layer and NeuN (red) to visualize the granule cell layer. Numbers indicate individual lobes. (F) Quantification of Purkinje cell number in sagittal sections of Amot fl/fl mice (n = 3 mice and at least three sections per mouse) and Amot fl/fl;Syn-Cre mice (n = 4 mice and at least three sections per mouse). p = 0.5405. (G) Decrease in molecular layer thickness in Amot fl/fl;Syn-Cre mice, visualized by anti-calbindin antibody; lobe III down the preculminate fissure. (H-J) Thickness of the molecular (H), granular (I), and Purkinje (J) cell layers in the cerebellum of Amot fl/fl mice (n = 3 mice and at least three sections per mouse) and Amot fl/fl;Syn-Cre mice (n = 5 mice and at least three sections per mouse) on P30 measured in lobe III down the preculminate fissure. p = 0.0135, p = 0.4169, and p = 0.4615. (K, L) Granule cell distribution (K) and quantification (L) in the molecular layer of the cerebellum in P30 Amot fl/fl mice (n = 3 mice and at least three sections per mouse) and Amot fl/fl;Syn-Cre mice (n = 5 mice and at least three sections per mouse), shown as a percentage of NeuN-positive cells per 20,000 μm2. p = 0.7333. Numerical values that underlie the graphs are shown in S1 Data. Statistical significance was analyzed using two-tailed unpaired t test. *p < 0.05. Bars represent the mean ± SD. Scale bars = 5 mm in C, 1,000 μm in E and 20 μm in G and K. Amot, angiomotin; ns, not significant; P, postnatal day; SD, standard deviation.
Fig 6.
Amot deletion in neurons impairs dendritic tree morphology of Purkinje cells.
(A) Tomato expression in single Purkinje cells in STOP-Tom reporter mice as a result of Cre activity upon virus injection. The upper panel shows a schematic illustration of the experimental procedure. P0 STOP-Tom pups were injected with AAV8-Syn-Cre virus and humanely killed on P21. (B) Images of the dendritic tree morphology of Purkinje cells in STOP-Tom and Amot fl/fl;STOP-Tom mice that were infected with AAV8-Syn-Cre. (C-G) Quantitative analysis of dendritic tree area (C), height (D), and width (E); primary and secondary dendrite lengths (F); and number of dendritic branching points (G) of Purkinje cells from STOP-Tom (n ≥ 45 cells and 4 mice in C–F; n = 35 cells and 4 mice in G) and Amot fl/fl;STOP-Tom (n ≥ 43 cells and 4 mice in C–F; n = 39 cells and 4 mice in G) brains that were infected with AAV8-Syn-Cre. p < 0.0001, p < 0.0001, p = 0.0012, p < 0.0001, and p = 0.0007. (H-M) Distribution of inhibitory and excitatory synapses in the cerebellum in Amot fl/fl and Amot fl/fl;Syn-Cre P30 mice. Cerebellar sagittal sections were immunostained for VGAT (H), VGLUT1 (L), and VGLUT2 (M). Anti-calbindin antibody was used to visualize Purkinje cells. (I-K) Quantification of VGAT (I), VGLUT1 (J), and VGLUT2 (K) puncta per 1,000 μm2 in Amot fl/fl mice (n = 3 mice and at least three sections per mouse) and Amot fl/fl;Syn-Cre mice (n = 4 mice and at least three sections per mouse). p = 0.4891, p = 0.9017, and p = 0.0970. Numerical values that underlie the graphs are shown in S1 Data. Statistical significance was analyzed using two-tailed unpaired t tests. **p < 0.01, ***p < 0.001, ****p < 0.0001. Bars represent the mean ± SD. Scale bars = 200 μm in A, 50 μm in B and 20 μm in H, L, M. Amot, angiomotin; ns, not significant; P, postnatal day; SD, standard deviation; STOP-Tom, STOP-tdTomato; VGAT, vesicular γ-aminobutyric acid transporter; VGLUT, vesicular glutamate transporter.
Fig 7.
Yap1 deletion in neurons affects cerebellar morphology.
(A) Strategy for generating Yap1 conditional knockout mice, showing the Yap1 flox allele (upper diagram) and recombined allele after Cre-mediated excision (lower diagram). (B) Lower Yap1 expression in the cerebellum of Yap1 fl/fl;Syn-Cre P30 mice analyzed by western blot. (C) Yap1 fl/fl;Syn-Cre mice exhibited a smaller cerebellum compared with control sex-matched littermates (Yap1 fl/fl). (D) Quantitative analysis of cerebellar weight in Yap1 fl/fl (n = 10) and Yap1 fl/fl;Syn-Cre (n = 4) P30 mice. p = 0.0178. Measurements were normalized to whole-brain weight. (E) Sagittal cerebellar sections of Yap1 fl/fl and Yap1 fl/fl;Syn-Cre P30 mice immunolabeled for calbindin (green) to visualize Purkinje cells and the molecular layer and NeuN (red) to visualize the granule cell layer. Numbers indicate individual lobes. (F) Quantification of Purkinje cells in sagittal sections of Yap1 fl/fl mice (n = 4 mice and at least three sections per mouse) and Yap1 fl/fl;Syn-Cre mice (n = 4 mice and at least three sections per mouse). p = 0.1782. (G) Decrease in cerebellar molecular layer thickness in Yap1 fl/fl;Syn-Cre mice visualized with calbindin; lobe III down the preculminate fissure at P30. (H-J) Thickness of the molecular (H), granular (I), and Purkinje (J) cell layers in the cerebellum of Yap1 fl/fl (n = 4 mice and at least three sections per mouse) and Yap1 fl/fl;Syn-Cre (n = 4 mice and at least three sections per mouse) P30 mice measured in lobe III down the preculminate fissure. p = 0.0128, p = 0.4679, and p = 0.1885. (K, L) Granule cell distribution (K) and quantification (L) in the molecular layer in P30 Yap1 fl/fl mice (n = 4 mice and at least three sections per mouse) and Yap1 fl/fl;Syn-Cre mice (n = 4 mice and at least three sections per mouse) in the cerebellum, shown as a percentage of NeuN-positive cells per 20,000 μm2. p = 0.0010. Numerical values that underlie the graphs are shown in S1 Data. Statistical significance was analyzed using two-tailed unpaired t test. *p < 0.05, **p < 0.01. Bars represent the mean ± SD. Scale bar = 5 mm in C, 1,000 μm in E, and 20 μm in G and K. ns, not significant; P, postnatal day; SD, standard deviation; Yap1, Yes-associated protein 1.
Fig 8.
Impairments in dendritic tree morphology of Purkinje cells in Yap1 mutant mice.
(A) Dendritic trees of Purkinje cells from STOP-Tom (control) and Yap1 fl/fl;STOP-Tom mice that were infected with AAV8-Syn-Cre. (B-F) Quantitative analysis of dendritic tree area (B) and width (C), primary and secondary dendrite length (D) and height (E), and the number of dendritic branching points (F) of Purkinje cells from STOP-Tom (n ≥ 28 cells and 5 mice in B-E; n = 33 cells and 5 mice in F) and Yap1 fl/fl;STOP-Tom (n ≥ 68 cells and 8 mice in B-E; n = 48 cells and 6 mice in F) brains infected with AAV8-Syn-Cre. p < 0.0001, p = 0.0003, p < 0.0001, p = 0.2194, p < 0.0001. (G-L) Distribution of inhibitory and excitatory synapses in the cerebellum in Yap1 fl/fl and Yap1 fl/fl;Syn-Cre P30 mice. Cerebellar sagittal sections were immunostained for VGAT (G), VGLUT1 (K), and VGLUT2 (L). Anti-calbindin antibody was used to visualize Purkinje cells. (H-J) Quantification of VGAT (H), VGLUT1 (I), and VGLUT2 (J) puncta per 1,000 μm2 in Yap1 fl/fl mice (n = 4 mice and at least three sections per mouse) and Yap1 fl/fl;Syn-Cre mice (n = 4 mice and at least three sections per mouse). p = 0.0727, p = 0.6267, and p = 0.3417. Numerical values that underlie the graphs are shown in S1 Data. Statistical significance was analyzed using two-tailed unpaired t tests. ***p < 0.001, ****p < 0.0001. Bars represent the mean ± SD. Scale bars = 50 μm in A and 50 μm in G, K, and L. ns, not significant; P, postnatal day; SD, standard deviation; STOP-Tom, STOP-tdTomato; VGAT, vesicular γ-aminobutyric acid transporter; VGLUT, vesicular glutamate transporter; YAP1, Yes-associated protein 1.
Fig 9.
Impairment in locomotor coordination in Amot fl/fl;Syn-Cre and Yap1 fl/fl;Syn-Cre mice.
(A, B) General mobility in Amot fl/fl;Syn-Cre mice (n = 9) and control age- and sex-matched Amot fl/fl mice (n = 9) in the open field, showing the (A) time spent immobile (p = 0.6602) and (B) total distance traveled (p = 0.2787). (C, D) General mobility in Yap1 fl/fl;Syn-Cre mice (n = 6) and control age- and sex-matched Yap1 fl/fl mice (n = 8) in the open field, showing the (C) time spent immobile (p = 0.5806) and (D) total distance traveled (p = 0.2298). (E) Image from the rotarod experiment. (F) Locomotor coordination in Amot fl/fl mice (n = 10) and Amot fl/fl;Syn-Cre mice (n = 6) on P150, showing the latency to fall from the rotarod. p = 0.0092, p = 0.0022, p = 0.0051, p = 0.0422, and p = 0.0483. (G) Locomotor coordination in Yap1 fl/fl mice (n = 4) and Yap1 fl/fl;Syn-Cre mice (n = 6) on P150, showing the latency to fall from the rotarod. p = 0.3018, p = 0.1009, p = 0.0150, p = 0.0005, and p = 0.0416. (H) CatWalk gait analysis. The footprints of the right paws are shown in red, and the footprints of the left paws are shown in green. (I) Duty cycle in Amot fl/fl;Syn-Cre mice (n = 5) and Amot fl/fl (n = 11) control littermates. p = 0.9463 and p = 0.0401. (J) Duty cycle in Yap1 fl/fl;Syn-Cre (n = 6) and Yap1 fl/fl (n = 4) mice. p = 0.0538 and p = 0.0024. (K) Image from the foot-fault experiment. (L, M) Foot-fault analysis of Amot fl/fl (n = 7) and Amot fl/fl;Syn-Cre (n = 5) mice. (L) Time traveled. p = 0.0441. (M) Number of foot slips. p < 0.0001. (N, O) Foot-fault analysis of Yap1 fl/fl (n = 4) and Yap1 fl/fl;Syn-Cre (n = 6) mice. (N) Time traveled. p = 0.0271. (O) Number of foot slips. p = 0.0430. Numerical values that underlie the graphs are shown in S1 Data. Statistical significance was analyzed using two-tailed unpaired t test. *p < 0.05, **p < 0.01, ***p < 0.001. Bars represent the mean ± SD in A, B, C, D, I, J, K, M, N, and O and mean ± SEM in F and G. Amot, angiomotin; BOS, base of support; FP, forepaws; HP, hind paws; ns, not significant; P, postnatal day; SD, standard deviation; SEM, standard error of the mean; Yap1, Yes-associated protein 1.
Fig 10.
Mechanism underlying Amot- and Yap1-mediated dendritogenesis.
(A, B) Reduction of phosphorylation of S6 ribosomal protein in cerebellar homogenates from Amot fl/fl;Syn-Cre (A) and Yap1 fl/fl;Syn-Cre (B) P30 mice, analyzed by western blot (see also S11A and S11B Fig for additional blots and quantification). (C, D) Reduction of phosphorylation of S6 ribosomal protein levels in lysates from Amot-knockdown neurons (C) and Yap1-knockdown neurons (D), analyzed by western blot (see also S11C Fig for additional blots and quantification). (E) Decrease in phosphorylation of S6 ribosomal protein levels in cultured rat hippocampal neurons on DIV7 transfected with constructs that encoded Amot shRNA or Yap1 shRNA, analyzed by immunocytochemistry (see also S11D for quantification). The neurons were cotransfected with a plasmid that expressed mRFP (red) to visualize transfected cells (arrows in the left panel) and immunostained with anti-phospho-S6 antibody (green). Scale bar = 10 μm. (F, G) Reduction of phosphorylation of p70 S6K at Thr389 in cerebellar homogenates from Amot fl/fl;Syn-Cre (F) and Yap fl/fl;Syn-Cre (G) P30 mice, analyzed by western blot. (H-J) The ectopic expression of S6K rescued morphological phenotypes in Amot- or Yap1-knockdown neurons. (H) Quantification of TDL of Amot-knockdown rat hippocampal neurons that were transfected with the indicated constructs. Control: n = 66; Amot shRNA: n = 80; Amot shRNA + Amot: n = 48; Amot shRNA + S6K C.A.: n = 78; Amot shRNA + S6K K.D.: n = 38). To control p < 0.0001; to Amot shRNA p < 0.0001, p < 0.0001, p = 0.9995. (I) Quantification of TDL of Yap1-knockdown rat hippocampal neurons that were transfected with the indicated constructs. Control: n = 66; Yap1 shRNA: n = 92; Yap1 shRNA + Yap1: n = 46; Yap1 shRNA + S6K C.A.: n = 75; Yap1 shRNA + S6K K.D.: n = 46. To control p < 0.0001; to Yap1 shRNA p < 0.0001, p < 0.0001, p = 0.9933. (J) Quantification of TDL of rat hippocampal neurons that were transfected with the indicated plasmids. Control: n = 66; S6K C.A.: n = 37; S6K K.D.: n = 38. p = 0.5395, p = 0.9516. The analysis was performed for neurons from at least three independent cultures. (K) Model of a pathway in which Amot interacts with Yap1 and regulates dendritic tree morphology through the regulation of S6 phosphorylation by S6K. Numerical values that underlie the graphs are shown in S1 Data. Statistical significance was analyzed using one-way analysis of variance followed by Tukey’s post hoc test. ****p < 0.0001. Bars represent the mean ± SEM. Amot, angiomotin; DIV, day in vitro; mRFP, monomeric RFP; ns, not significant; P, postnatal day; RFP, red fluorescent protein; S6K, S6 kinase; S6K C.A., constitutively active mutant of S6K; S6K K.D., kinase-dead mutant of S6K; SEM, standard error of the mean; shRNA, short-hairpin RNA; TDL, total dendrite length; Thr, threonine; Yap1, Yes-associated protein 1.