Figure 1.
Transient interactions between dendritic filopodia in cultured hippocampal neurons.
(A) Triple immunostaining for F-actin, β-catenin and either Tau-1 or MAP2 in cultured hippocampal neurons. Cultured neurons on 5 DIV were triple-stained for F-actin, β-catenin and either Tau-1 or MAP2. (A1) F-actin, β-catenin and Tau-1; (A2) F-actin, β-catenin and MAP2. Upper rows, low magnification images; lower rows, high magnification images of the boxed areas in the upper rows. Bars, upper rows 10 µm; lower rows 2.5 µm. Dendrites and axons were identified by the signals for MAP2 and Tau-1. Arrows indicate axons and arrowheads indicate dendritic filopodia-filopodia contact sites, which were identified by the signal for β-catenin. (B) Time-lapse imaging of EGFP-expressing neurons. Time-lapse images of EGFP-expressing neurons on 4 DIV were acquired every 30 min. Bar, 5 µm. Arrowheads indicate dendritic filopodia-filopodia contact sites. We identified neurites to be dendrites in (B) as follows: we identified MAP2-poitive and Tau-1-negative neurites to be dendrites and MAP2-negative and Tau-1-positive neurites to be axons; in cultured neurons on 4 DIV, axons were generally longer than dendrites; and the diameters of axons, which were 10-µm apart from the cell body, became smaller than those of dendrites at the same distance. We identified neurites to be dendrites by these characteristic morphologies.
Figure 2.
Localization of ZO-1 at the tips of free dendritic filopodia and filopodia-filopodia contact sites in cultured hippocampal neurons.
Triple immunostaining for F-actin, ZO-1 and either Tau-1 or MAP2 in cultured hippocampal neurons. Cultured neurons on 5 DIV were triple-stained for F-actin, ZO-1 and either Tau-1 or MAP2. (A) F-actin, ZO-1 and Tau-1; (B) F-actin, ZO-1 and MAP2. Upper rows, low magnification images; lower rows, high magnification images of the boxed areas in the upper rows. Bars, upper rows 10 µm; lower rows 2.5 µm. Dendrites and axons were identified by the signals for MAP2 and Tau-1. Arrows indicate the growth cones of dendrites. Double arrowheads and arrowheads indicate the tips of free dendritic filopodia and dendritic filopodia-filopodia contact sites, respectively. It is noted that multiple Tau-1-positive neurites were observed in the neurons shown in (A) as described previously [34]. One of these multiple Tau-1-positive neurites is specified as an axon in the later stages of neuronal development [34].
Figure 3.
Enhancement of dendritic filopodia protrusion and changes in dendrite morphology in ZO-1-overexpressing neurons.
(A) Dendrite morphologies of the control and HA-tagged ZO-1-overexpressing neurons. Cultured neurons were transfected with EGFP and either the empty vector (MOCK) or the HA-tagged ZO-1 vector (ZO-1) on 0 DIV and triple-stained for EGFP, HA and MAP2 on 7 DIV. (Aa) low magnification images; (Ab) high magnification images of the boxed areas in (Aa). Upper rows, control neurons, lower rows, HA-tagged ZO-1-overexpressing neurons. Bars, (Aa) 10 µm; (Ab) 2.5 µm. Arrows and arrowheads indicate axons and dendritic filopodia-filopodia contact sites, respectively. (B) Time-lapse imaging of the control and HA-tagged ZO-1-overexpressing neurons. Time-lapse images of the neurons expressing EGFP and either the empty vector or the HA-tagged ZO-1 vector were acquired every 8 h from 5 to 6 DIV. (Ba) low magnification images; (Bb) high magnification images of the boxed areas in (Ba). Upper rows, neurons expressing EGFP and the empty vector; lower rows, neurons expressing EGFP and the HA-tagged ZO-1 vector. Bars, (Ba) 10 µm; (Bb) 2.5 µm. Arrowheads in (Ba) and (Bb) indicate dendrites and dendritic filopodia-filopdoia contact sites, respectively. (Bc) Quantitation of the dendritic filopodia-filopodia contact time. The data are presented as mean plus SEM (error bars) for each sample (n = 25). *, p value < 0.05 (Mann-Whitney U test). (C) Quantitation of the dendrite morphologies of HA-tagged ZO-1-overexpressing neurons. The average number of the shafts of the primary dendrites, the average length of the shafts of the primary dendrites, the average number of the shafts of the branches of the primary dendrites, and the average density of the dendritic filopodia protruding from the primary dendrites in the neurons transfected with the empty vector or the HA-tagged ZO-1 vector on 0 DIV were measured on 7 DIV. The data are presented as mean plus SEM (error bars) for each sample (n = 20 for the number of the shafts of the primary dendrites (Number of primary dendrite shafts); n = 64 for the length of the shafts of the primary dendrites (Length of primary dendrite shafts) and the number of the shafts of the branches of the primary dendrites (Number of branch shafts); n = 46 for the density of the dendritic filopodia protruding from the primary dendrites (Density of dendritic filopodia). *, p value < 0.05 (Student’s t-test)).
Figure 4.
Reduction of dendritic filopodia protrusion and changes in dendrite morphology in ZO-1-knockdown neurons.
(A) Dendrite morphologies of the control and ZO-1-knockdown (ZO-1 KD) neurons. Cultured neurons were transfected with the control siRNA or the ZO-1 siRNA on 3 DIV, then transfected with EGFP on 5 DIV, and double-immunostained for EGFP and either ZO-1 or MAP2 on 7 DIV. (A1) EGFP and ZO-1; (A2a) EGFP and MAP2; (A2b) EGFP, ZO-1 and MAP2. Upper rows, control neurons; lower rows, ZO-1-knockdown neurons. Bars, (A1) 10 µm; (A2a) 20 µm; (A2b) 2.5 µm. (B) Time-lapse imaging of the control and ZO-1-knockdown (ZO-1 KD) neurons. Time-lapse images of the neurons expressing EGFP and either the control siRNA or the ZO-1 siRNA were acquired every 8 h from 5 to 6 DIV. (Ba) low magnification images; (Bb) high magnification images of the boxed areas in (Ba). Upper rows, neurons expressing EGFP and the control siRNA; lower rows, neurons expressing EGFP and the ZO-1 siRNA. Bars, (Ba) 10 µm; (Bb) 2.5 µm. Arrowheads indicate dendrites. (C) Quantitation of the dendrite morphologies of ZO-1-knockdown neurons. The average number of the shafts of the primary dendrites, the average length of the shafts of the primary dendrites, the average number of the shafts of the branches of the primary dendrites, the average density of the dendritic filopodia protruding from the primary dendrites and the median angle of the shafts of the primary dendrites in the neurons transfected with the control siRNA or the ZO-1 siRNA on 3 DIV were measured on 7 DIV. The data are presented as mean plus SEM (error bars) for each sample (n = 21 for the number of the shafts of the primary dendrites (Number of primary dendrite shafts); n = 60 for the length of the shafts of the primary dendrites (Length of primary dendrite shafts), the number of the shafts of the branches of the primary dendrites (Number of branch shafts) and the angles of the shafts of the primary dendrites (Angle of primary dendrite shafts); n = 33 for the density of the dendritic filopodia protruding from the primary dendrites (Density of dendritic filopodia)). Statistical analyses of the dendrite morphology of the ZO-1-knockdown neurons except for that of the angles of the shafts of each dendrite were performed using Student’s t-test. Statistical analysis of the angles of the shafts of each primary dendrite was performed using the Mann-Whitney U test. *, p value < 0.05; and **, p value < 0.01 (Student’s t-test) and *, p value <0.05 (Mann-Whitney U test).
Figure 5.
Co-localization of the components of the nectin and cadherin systems with ZO-1 in cultured hippocampal neurons.
Cultured hippocampal neurons on 7 DIV were triple-stained for F-actin, ZO-1 and one of nectin-1, nectin-3, afadin, N-cadherin, β-catenin or αN-catenin. (A1) F-actin, ZO-1 and nectin-1; (A2) F-actin, ZO-1 and nectin-3; (A3) F-actin, ZO-1 and afadin; (B1) F-actin, ZO-1 and N-cadherin; (B2) F-actin, ZO-1 and β-catenin; (B3) F-actin, ZO-1 and αN-catenin. Upper rows, low magnification images; lower rows, high magnification images of the boxed areas in the upper rows. Bars, upper rows 10 µm; lower rows 2.5 µm. Arrowheads indicate dendritic filopodia-filopodia contact sites. (C) Quantitation of the co-localization of the components of the nectin and cadherin systems with ZO-1 in control neurons. In each experiment, 30 punctate immunofluorescence signals for F-actin at dendritic filopodia-filopodia contact sites were randomly chosen and the percentage of the punctate signal for ZO-1, nectin-1, nectin-3, afadin, N-cadherin, β-catenin or αN-catenin, which was co-localized with that for F-actin (% of co-localization), was counted.
Figure 6.
Increased accumulation of the components of the nectin and cadherin systems at the dendritic filopodia-filopodia contact sites in ZO-1-overexpressing neurons.
Cultured ZO-1-overexpressing hippocampal neurons on 7 DIV were prepared as described in the legend to Figure 3A and triple-stained for HA, EGFP and one of nectin-1, nectin-3, afadin, N-cadherin, β-catenin or αN-catenin. (A1) EGFP, HA and nectin-1; (A2) EGFP, HA and nectin-3; (A3) EGFP, HA and afadin; (B1) EGFP, HA and N-cadherin; (B2) EGFP, HA and β-catenin; (B3) EGFP, ZO-1 and αN-catenin. (a) low magnification images; (b) high magnification images of the boxed areas in (a). Upper rows, control neurons; lower rows, HA-tagged ZO-1-overexpressing neurons. Bars, upper rows 10 µm; lower rows 2.5 µm. Arrowheads indicate dendritic filopodia-filopodia contact sites. (C) Quantitation of the co-localization of the components of the nectin and cadherin systems with ZO-1 in ZO-1-overexpressing neurons. The data are presented as mean plus SEM (error bars) for each sample (n = 40 for nectin-1, afadin and N-cadherin; n= 50 for nectin-3 and β-catenin; n = 60 for αN-catenin). *, p value < 0.05; and **, p value < 0.01 (Student’s t-test) (AU) arbitrary unit.
Figure 7.
Reduced accumulation of the components of the nectin and cadherin systems at the dendritic filopodia-filopodia contact sites in ZO-1-knockdown neurons.
Cultured ZO-1-knockdown hippocampal neurons on 7 DIV were prepared as described in the legend to Figure 4A and double-stained for EGFP and one of nectin-1, nectin-3, afadin, N-cadherin, β-catenin or αN-catenin. (A1) EGFP and nectin-1; (A2) EGFP and nectin-3; (A3) EGFP and afadin; (B1) EGFP and N-cadherin; (B2) EGFP and β-catenin; (B3) EGFP and αN-catenin; (a) low magnification images; (b) high magnification images of the boxed areas in (a). Upper rows, control neurons; lower rows, ZO-1-knockdown neurons. Bars, upper rows 10 µm; lower rows 2.5 µm. Arrowheads indicate dendritic filopodia-filopodia contact sites. (C) Quantitation of the co-localization of the components of the nectin and cadherin systems with ZO-1 in ZO-1-knockdown neurons. The data are presented as mean plus SEM (error bars) for each sample (n = 30 for nectin-1, afadin, β-catenin and αN-catenin; n = 40 for nectin-3; n = 20 for N-cadherin) *, p value < 0.05; and **, p value < 0.01 (Student’s t-test) (AU) arbitrary unit. It is noted that the immunofluorescence signals for the components of the nectin and cadherin systems were still observed at some dendritic filopodia-filopodia contact sites in the ZO-1 knockdown neuronal cultures as shown in (A) and (B). We estimated that the transfection efficiency of the siRNA in cultured neurons was about 80%. Thus, ZO-1 might not be knocked-down in some neurons, in which these immunofluorescence signals were observed.