JIP1 and JIP3 cooperate to mediate TrkB anterograde axonal transport by activating kinesin-1

Abstract

Long-range anterograde axonal transport of TrkB is important for neurons to exert appropriate BDNF responses. TrkB anterograde axonal delivery is mediated by kinesin-1, which associates with TrkB via the adaptor protein JIP3 or the Slp1/Rab27B/CRMP-2 protein complex. However, little is known about the activation mechanisms of TrkB-loaded kinesin-1. Here, we show that JIP1 mediates TrkB anterograde axonal transport using JIP1 knockout mice, sciatic nerve ligation analysis and live imaging. Next, we proved that JIP1 and JIP3 cooperate to mediate TrkB anterograde axonal transport. Finally, microtubule-binding and microfluidic chamber assays revealed that JIP1 and JIP3 cooperate to relieve kinesin-1 autoinhibition, which depends on the binding of JIP1 to kinesin-1 heavy chain (KHC) and light chain (KLC) and the binding of JIP3 to KLC and is essential for TrkB anterograde axonal transport and BDNF-induced TrkB retrograde signal. These findings could deepen our understanding of the regulation mechanism underlying TrkB anterograde axonal transport and provide a novel kinesin-1 autoinhibition-relieving model.

Keywords: APP; Autoinhibition; Co-immunoprecipitation; FEZ-1; Kif5C.

Fig. 1

JIP1 interacts with TrkB in vitro and in vivo. a, b Exogenous co-immunoprecipitation (CO-IP) of JIP1 and TrkB. Lysates from HEK293 cells transfected with HA–JIP1 and/or FLAG–TrkB constructs were immunoprecipitated with mouse anti-FLAG or rabbit anti-HA antibody. Next, western blot was employed to detect the immunoprecipitated proteins. c, d Endogenous CO-IP of JIP1 and TrkB. Rat brain lysates were immunoprecipitated with mouse anti-JIP1 or rabbit anti-TrkB antibody. Next, western blot was employed to detect the immunoprecipitated proteins. e Co-localization of JIP1 and TrkB in cultured hippocampal neurons. The cultured primary hippocampal neurons (DIV 5) of rat were immunostained with mouse anti-JIP1 (red) and goat anti-TrkB (green) antibodies. The lower panels display enlarged images of the framed regions and the z-stack images displays the images of the cell body (upper) and axon segment (lower) from the z-stack perspective at the position of drawing the blue lines in the merge image. The arrows point to the puncta that were immunostained by both anti-JIP1 and TrkB antibodies. The scale bar represents 20 μm

Fig. 2

JIP1 mediates TrkB anterograde axonal transport. a Schematic representation of the method of the sciatic nerve ligation analysis. b The sciatic nerves of JIP1−/− mice and littermate wild-type (WT) mice were ligated for 1 day at the mid-thigh level. The right sciatic nerves were ligated (Lig.), and the left sciatic nerves underwent sham operations and served as the control group (Ctr.). Next, the sciatic nerves of the different groups were isolated at the same location and probed with the indicated antibodies in the immunohistochemistry assays. The scale bar represents 50 μm. c Western blot analyses of the accumulated levels of TrkB, TrkA, KLC and JIP1 at the proximal (P) and distal (D) sides of the ligation of the sciatic nerves of the WT and JIP1−/− mice. Tubulin was used as a loading control. d Quantification of the proximal and distal sciatic nerve fractions. The data are presented as the mean ± SEM (n = 7, for WT group; n = 8, for JIP1−/− group; ampersand significant treatment effect, Psi significant genotype effect, Omega significant interaction effect, vs the WT + Ctr. group, two-way ANOVA). e Lysates from DRG tissues or cultured DRG neurons segregated from the WT and JIP1−/− mice were detected with the indicated antibodies. f Kymographs of the axons of the cultured neurons from the WT, JIP3−/− and JIP1−/− mice. Images were captured every 1 s for 60 s. The scale bar represents 10 μm. g Quantitative analyses of the relative frequencies of the immobile (Immobile), bidirectional (Bidir), anterograde (Antero) and retrograde (Retro) TrkB–mRFP-containing vesicles in the cultured hippocampal neurons from the WT, JIP3−/− and JIP1−/− mice. At least 80 vesicles of 20 axons or 40 dendrites were counted in each condition (n = 3; *p < 0.05, vs the WT group; one-way ANOVA). h Quantitative analyses of the mean velocity of TrkB–mRFP-containing vesicles in the cultured hippocampal neurons from the WT, JIP3−/− and JIP1−/− mice. The numbers of cells analyzed for each group were >20. The data are displayed as the mean ± SEM (n = 3; *p < 0.05, vs the WT group; one-way ANOVA). i Endogenous CO-IP of TrkB. Brain lysates of WT and JIP1−/− mice were immunoprecipitated with rabbit anti-TrkB antibody. Next, western blot was employed to detect the immunoprecipitated proteins

Fig. 3

JIP1 interacts with TrkB via JIP3. a Schematic representation of the TrkB mutants. Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with rabbit anti-HA antibody. Next, western blot was performed to detect the immunoprecipitated proteins. b Schematic representation of the chimeric IL2R–TrkB–JM mutants. Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with mouse anti-FLAG antibody. Next, western blot was performed to detect the immunoprecipitated proteins (n = 3, *p < 0.05, vs the HA–JIP1 + FLAG–IL2R–JM1 group; one-way ANOVA). c–e Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with mouse anti-FLAG antibody. Next, western blot was performed to detect the immunoprecipitated proteins. The data in d and e are displayed as the mean ± SEM from three independent experiments [n = 3, in d, *p < 0.05, vs the HA–JIP1 + FLAG–TrkB group (control group); n = 3, in e, *p < 0.05, vs the HA–JIP3 + FLAG–TrkB group (control group); one-way ANOVA]

Fig. 4

JIP1 and JIP3 cooperate to mediate TrkB anterograde axonal transport. a Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with mouse anti-FLAG antibody. Next, western blot was performed to detect the immunoprecipitated proteins. The data are displayed as the mean ± SEM [n = 3, *p < 0.05, vs the Myc–KLC + FLAG–TrkB group (control group); NS no significance; one-way ANOVA]. b Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with mouse anti-FLAG antibody. Next, western blot was performed to detect the immunoprecipitated proteins. The data are presented as the mean ± SEM [n = 3, *p < 0.05, vs the Myc–KLC + FLAG–TrkB group (control group); ^# p < 0.05; NS no significance; one-way ANOVA). c Representative images of cultured primary hippocampal neurons (DIV5) of rat transfected with the indicated constructs and immunostained with goat anti-TrkB (red) and/or rabbit anti-HA (green) and/or mouse anti-Myc (green or blue) antibodies. The neurons in the control group were transfected with the pcDNA3.1-EGFP construct to identify the transfection-positive neurons. The lower panels display enlarged images of the framed regions, which represent the axon tips. The scale bar represents 20 μm. d Quantification of the localization of the endogenous TrkB at the distal axon and dendrite in c. The data are presented as the mean ± SEM and >30 cells were examined in each experiment (n = 3, *p < 0.05, vs the control group; one-way ANOVA). e Kymographs of the axons of the cultured WT (pcDNA3.1-transfected), JIP3−/− (pcDNA3.1–JIP1-transfected) and JIP1−/− (pcDNA3.1–JIP3-transfected) neurons. Images were captured every 1 s for 60 s. The scale bar represents 20 μm. f Quantitative analysis of the relative frequencies of the immobile (Immobile), bidirectional (Bidir), anterograde (Antero) and retrograde (Retro) TrkB–mRFP-containing vesicles in the cultured WT + pcDNA3.1, JIP3−/− + JIP1 and JIP1−/− + JIP3 neurons. At least 80 vesicles of 20 axons or 40 dendrites were counted in each condition (n = 3; *p < 0.05, vs the WT group; one-way ANOVA). g Quantitative analysis of the mean velocity of TrkB–mRFP-containing vesicles in the cultured WT + pcDNA3.1, JIP3−/− + JIP1 and JIP1−/− + JIP3 neurons. The numbers of cells analyzed for each group were >20. The data are displayed as the mean ± SEM (n = 3; *p < 0.05, vs the WT group; one-way ANOVA)

Fig. 5

The binding of JIP1 to KHC and to KLC is necessary for relieving kinesin-1 autoinhibition. a, f Lysates from COS-7 cells transfected with the indicated constructs were examined in a microtubule-binding assay. Taxol-stabilized microtubules were added (+) or not (−), and either ATP (T) or AMPPNP (N) was used. The microtubules were centrifuged as pellets. Next, the microtubule-binding proteins in the pellets were immunostained with the indicated antibodies (P pellet, S supernatant). b Schematics of the JIP1 constructs and summaries of their reported binding domains with KLC and the KHC-tail. The mutants represent the reported JIP1 mutants that cannot bind to KLC or the KHC-tail. c–e Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with rabbit anti-HA antibody. Next, western blot was performed to detect the immunoprecipitated proteins

Fig. 6

The binding of JIP3 to KLC is necessary for relieving kinesin-1 autoinhibition. a Schematics of the JIP3 constructs and summaries of their reported binding domains with KLC and the KHC-tail. The mutants represent the reported JIP3 mutants that cannot bind to KLC or the KHC-tail. b–d Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with rabbit anti-HA antibody. Next, western blot was performed to detect the immunoprecipitated proteins. e, g Lysates from COS-7 cells transfected with the indicated constructs were examined in the microtubule-binding assay. Taxol-stabilized microtubules were added (+) or not (−), and either ATP (T) or AMPPNP (N) was used. The microtubules were centrifuged as pellets. Next, the microtubule-binding proteins in the pellets were immunostained with the indicated antibodies. f Lysates from COS-7 cells transfected with pSuper-siFEZ-1 or empty vector were examined by western blot to analyze the knock-down efficiency of the FEZ-1 siRNA. Tubulin was used as a loading control

Fig. 7

JIP1 and JIP3 mediate TrkB anterograde axonal transport by relieving kinesin-1 autoinhibition. a Schematic illustration of the interacting regions of JIP3, KLC, JIP1 and KHC that are necessary for kinesin-1 autoinhibition relief. b Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with mouse anti-FLAG antibody. Next, western blot was performed to detect the immunoprecipitated proteins. The data are presented as the mean ± SEM [n = 3, *p < 0.05, vs the Myc–KLC + FLAG–TrkB group (control group); ^# p < 0.05, vs the Myc–KLC + FLAG–TrkB + HA–JIP1 group; one-way ANOVA]. c Lysates from HEK293 cells transfected with the indicated constructs were immunoprecipitated with mouse anti-FLAG antibody. Next, western blot was performed to detect the immunoprecipitated proteins. The data are displayed as the mean ± SEM [n = 3, *p < 0.05, vs the Myc–KLC + FLAG–TrkB group (control group); one-way ANOVA]. d Representative images of cultured primary hippocampal neurons (DIV5) of rat transfected with the indicated construct and immunostained with goat anti-TrkB and/or rabbit anti-HA antibodies. Neurons in the control group were transfected with the pcDNA3.1–EGFP construct to display the transfection-positive neurons. The lower panels present enlarged images of the framed regions, which represent the axon tips. The scale bar represents 20 μm. e Quantification of the localizations of endogenous TrkB at the distal axon or dendrite in d. The data are presented as the mean ± SEM and >30 cells were examined in each experiment (n = 3, *p < 0.05, vs the control group; one-way ANOVA)

Fig. 8

JIP1 and JIP3 mediate BDNF-induced TrkB retrograde signal by mediating the binding of TrkB to KLC and relieving kinesin-1 autoinhibition. a, d Schematics of the neuronal culture and dosing in the microfluidic chamber. Primary hippocampal neurons of rat transfected with both pcDNA3.1–EGFP and the indicated JIP1 or JIP3 mutant construct were cultured in the microfluidic chambers for 7 days. BDNF (50 ng/mL) was applied to the cell body compartment a or the axon compartment d for 30 min, and then the cell bodies of the neurons were stained with pErk5. b, e Representative images of the cell bodies of the hippocampal neurons cultured in the microfluidic chambers shown in a, d. The neurons were immunostained with mouse anti-GFP (green) and rabbit anti-pErk5 (red) antibodies. The scale bar represents 10 μm. c, f Quantification of the fluorescence intensities of pErk5 in the cell bodies of the neurons in b, e. The data are presented as the mean ± SEM and >30 cells were examined in each experiment (n = 3, *p < 0.05, vs the vehicle group; ^# p < 0.05, vs the vehicle + BDNF group; one-way ANOVA)