Autophagy initiation triggers p150Glued-AP-2β interaction on the lysosomes and facilitates their transport

Abstract

The endocytic adaptor protein 2 (AP-2) complex binds dynactin as part of its noncanonical function, which is necessary for dynein-driven autophagosome transport along microtubules in neuronal axons. The absence of this AP-2-dependent transport causes neuronal morphology simplification and neurodegeneration. The mechanisms that lead to formation of the AP-2-dynactin complex have not been studied to date. However, the inhibition of mammalian/mechanistic target of rapamycin complex 1 (mTORC1) enhances the transport of newly formed autophagosomes by influencing the biogenesis and protein interactions of Rab-interacting lysosomal protein (RILP), another dynein cargo adaptor. We tested effects of mTORC1 inhibition on interactions between the AP-2 and dynactin complexes, with a focus on their two essential subunits, AP-2β and p150^Glued. We found that the mTORC1 inhibitor rapamycin enhanced p150^Glued-AP-2β complex formation in both neurons and non-neuronal cells. Additional analysis revealed that the p150^Glued-AP-2β interaction was indirect and required integrity of the dynactin complex. In non-neuronal cells rapamycin-driven enhancement of the p150^Glued-AP-2β interaction also required the presence of cytoplasmic linker protein 170 (CLIP-170), the activation of autophagy, and an undisturbed endolysosomal system. The rapamycin-dependent p150^Glued-AP-2β interaction occurred on lysosomal-associated membrane protein 1 (Lamp-1)-positive organelles but without the need for autolysosome formation. Rapamycin treatment also increased the acidification and number of acidic organelles and increased speed of the long-distance retrograde movement of Lamp-1-positive organelles. Altogether, our results indicate that autophagy regulates the p150^Glued-AP-2β interaction, possibly to coordinate sufficient motor-adaptor complex availability for effective lysosome transport.

Keywords: AP-2 adaptor complex; Autophagy; Dynactin; Lysosomes; mTORC1; p150Glued.

Fig. 1

mTORC1 inhibition increases p150^Glued–AP-2β interaction. A Western blot analysis of levels of endogenous p150^Glued, AP-2β, and P-S6 (Ser235/236) and co-immunoprecipitation of endogenous AP-2β with p150^Glued from brain lysates from rats that were treated with rapamycin (RAPA +) or vehicle (RAPA-). INPUT, 10% of lysate used for immunoprecipitation. Shown is a representative example from N = 3 independent experiments. B–E Dynamics of p150^Glued-tdTomato and AP-2β-GFP co-transport in axons of neurons that were treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA). B Representative snapshots of 60 μm segment of axon and C corresponding kymographs of p150^Glued-tdTomato- and AP-2β-GFP-positive objects. See also Supplementary Fig. S1 and Movies 1–2. Scale bar = 10 μm. D Number of all p150^Glued-tdTomato/AP-2β-GFP objects in axons per 60 μm. The data are expressed as the mean ± SEM. N = 4 independent experiments. n = 15 cells per variant. *p < 0.05 (Mann–Whitney test). E Number of mobile p150^Glued-tdTomato/AP-2β-GFP-positive objects in axons per 60 μm. The data are expressed as the mean ± SEM. N = 4 independent experiments. n = 15 cells per variant. ns nonsignificant (Mann–Whitney test). F Western blot analysis of levels of endogenous p150^Glued, AP-2β, and P-S6 (Ser235/236) and co-immunoprecipitation of endogenous p150^Glued with AP-2β from HEK293T cells that were treated for 2 h with 0.1% DMSO (RAPA-) or 100 nM rapamycin (RAPA +). Input, 10% of lysate used for immunoprecipitation. Shown is a representative example from N = 5 independent experiments. G Representative images of Rat2 fibroblasts that were treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA) with p150^Glued–AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. H Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in G. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 5 independent experiments. n = 201 cells for each experimental variant. ***p < 0.001 (Student’s t-test). (I) Representative images of Rat2 fibroblasts that were treated for 2 h with 0.1% DMSO or 100 nM AZD-8055 with p150^Glued-AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. J Quantification of the number of p150^Glued-AP-2β PLA puncta in cells that were treated as in I. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 5 independent experiments. n = 211 cells (DMSO), 188 cells (AZD-8055). *p < 0.05 (Student’s t-test)

Fig. 2

AP-2β interaction with p150^Glued is indirect and requires intact dynactin-dynein complex. A Diagram of the full p150^Glued and N, C, and C2 fragments that were used in the study. B Western blot analysis of E. coli-produced His-AP-2β-ear binding to in vivo biotinylated AviHA-tagged p150^Glued N or C fragments or AviGFP-tagged p150^Glued or AviGFP-tagged β-galactosidase that was pulled down from HEK293T cells using Avi-tag pull down. Input, 10% of lysate added to the assay. Shown is a representative example from N = 3 independent experiments. C Western blot analysis of E. coli-produced His-AP-2β-ear binding to in vivo biotinylated AviHA-tagged p150^Glued C or C2 fragments or AviGFP-tagged p150^Glued or AviGFP-tagged β-galactosidase that was pulled down from HEK293T cells using Avi-tag pull down. Input, 10% of lysate added to the assay. Shown is a representative example from N = 2 independent experiments. D, E Western blot analysis of pull down of E. coli-purified recombinant GST- p150^Glued C, GST- p150^Glued C2, GST-Eps15, or GST (negative control) with recombinant His-AP-2β ear. Shown is a representative example from N = 2 independent experiments. F Western blot analysis of endogenous p62 and Arp1 binding to in vivo biotinylated AviHA-tagged p150^Glued N or C2 fragments or full AviGFP-tagged p150^Glued or AviGFP-tagged β-galactosidase that was pulled down from HEK293T cells using Avi-tag pull down. Shown is a representative example from N = 3 independent experiments. G Western blot analysis of the co-immunoprecipitation of endogenous p150^Glued with other subunits of dynactin-dynein complex (DIC1/2, p62, and Arp1) from HEK293T cells that were transfected with pEGFPC1 or pEGFPC1-p50 plasmids. Shown is a representative example from N = 3 independent experiments. H Representative images of Rat2 fibroblasts that were transfected with pEGFPC1 or pEGFPC1-p50 (green) and treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA) with p150^Glued–AP-2β PLA signals (magenta) and DAPI-stained nuclei (blue). Scale bar = 10 μm. I Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in H. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (GFP + DMSO) ± SEM. N = 4 independent experiments. n = 96 cells (GFP + DMSO), 81 cells (GFP + RAPA), 91 cells (GFP-p50 + DMSO), 92 cells (GFP-p50 + RAPA). **p < 0.01, ***p < 0.001, ns, nonsignificant (two-way ANOVA followed by Tukey’s multiple-comparison post hoc test)

Fig. 3

CLIP-170 but not microtubule dynamics is needed for p150^Glued–AP-2β interaction. A Western blot analysis of endogenous CLIP-170, p150^Glued, AP-2β, S6, and P-S6 (Ser235/236) levels and co-immunoprecipitation of endogenous AP-2β with p150^Glued in HEK293T cells that were transfected with control siRNA (siCtrl) or siRNA against human CLIP-170 (siCLIP-170) and treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA). Input, 10% of lysate added for immunoprecipitation. Shown is a representative example from N = 3 independent experiments. B Western blot analysis of endogenous CLIP-170 and CLIP-115 levels in non-transfected (NT) Rat2 cells or Rat2 cells that were transfected with siCtrl or rat siCLIP-170. C Representative images of Rat2 cells that were transfected with siCtrl or rat siCLIP-170, with immunofluorescently labeled endogenous CLIP-170 (green) and nucleus stained with Hoechst 33258 (blue). Scale bar = 10 µm. D Quantitative analysis of CLIP-170 immunofluorescence in Rat2 cells that were treated as in C. The results are presented as the average intensity of CLIP-170 immunofluorescence in the cell ± SEM. N = 5 independent experiments. **p < 0.01 (Student’s t-test). E Representative images of Rat2 fibroblasts that were transfected with siCtrl or rat siCLIP-170, and treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA), with p150^Glued–AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. F Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in E. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (siCtrl + DMSO) ± SEM. N = 4 independent experiments. n = 149 cells (siCtrl + DMSO), 130 cells (siCtrl + RAPA), 161 cells (siCLIP-170 + DMSO), 142 cells (siCLIP-170 + RAPA). **p < 0.01, ns, nonsignificant (two-way ANOVA followed by Tukey’s multiple-comparison post hoc test). G Representative images of Rat2 fibroblasts with p150^Glued–AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Cells were treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA) or treated for 2 h 15 min with 100 nM nocodazole alone or in combination with 100 nM rapamycin that was added 15 min after nocodazole (PRIOR nocodazole + RAPA). Scale bar = 10 μm. H Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in G. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 3 independent experiments. n = 126 cells (DMSO), 125 cells (RAPA), 133 cells (nocodazole), 139 cells (PRIOR nocodazole + RAPA). *p < 0.05, **p < 0.01, ns, nonsignificant (one-way ANOVA followed by Bonferroni multiple-comparison post hoc test). I Representative images of Rat2 fibroblasts with p150^Glued–AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Cells were treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA) or treated for 1 h with 100 nM nocodazole alone or added in the middle of 2 h of 100 nM rapamycin incubation (RAPA + POST nocodazole). Scale bar = 10 μm. J Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in I. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 3 independent experiments. n = 128 cells (DMSO), 118 cells (RAPA), 136 cells (nocodazole), 121 cells (RAPA + POST nocodazole). **p < 0.01, ***p < 0.001, ns, nonsignificant (one-way ANOVA followed by Bonferroni multiple-comparison post hoc test)

Fig. 4

Autophagy induction upon mTORC1 inhibition is needed for p150^Glued–AP-2β interaction. A Western blot analysis of endogenous p150^Glued, AP-2β, S6, and P-S6 (Ser235/236) levels and co-immunoprecipitation of endogenous AP-2β with p150^Glued in HEK293T cells that were treated with 0.1% DMSO for 2 h, 100 nM rapamycin (RAPA) for 2 h, 25 μM SBI-0206965 for 2 h 30 min, or 25 μM SBI-0206965 for 30 min and 100 nM rapamycin for 2 h (RAPA + SBI-0206965). Input, 10% of lysate used for immunoprecipitation. Shown is a representative example from N = 2 independent experiments. B Results of quantitative analysis of co-immunoprecipitation of experiments performed is as in A. C Representative images of Rat2 fibroblasts with p150^Glued–AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Cells were treated with 0.1% DMSO for 2 h, 100 nM rapamycin (RAPA) for 2 h, 25 μM SBI-0206965 for 2 h 30 min, or 25 μM SBI-0206965 for 30 min and 100 nM rapamycin for 2 h (RAPA + SBI-0206965). Scale bar = 10 μm. D Quantification of the number of p150^Glued-AP-2β PLA puncta in cells that were treated as in B. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 5 independent experiments. n = 184 cells (DMSO), 189 cells (RAPA), 169 cells (SBI-0206965), 174 cells (RAPA + SBI-0206965). *p < 0.05, ns, nonsignificant (one-way ANOVA followed by Bonferroni multiple-comparison post hoc test). E Representative images of Rat2 fibroblasts that were transfected with siCtrl or rat siAtg5 for 72 h and then treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA), with PLA p150^Glued–AP-2β signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. F Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in D. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (siCtrl + DMSO) ± SEM. N = 4 independent experiments. n = 199 cells (siCtrl + DMSO), 178 cells (siCtrl + RAPA), 195 cells (siAtg5 + DMSO), 211 cells (siAtg5 + RAPA). ***p < 0.001, **p < 0.01, ns, nonsignificant (two-way ANOVA followed by Tukey’s multiple-comparison post hoc test)

Fig. 5

p150^Glued–AP-2β interaction is induced by autophagy even when mTOR activity is preserved. A Western blot analysis of endogenous actin, LC3B I, LC3B II, P-S6 (Ser235/236), and S6 levels in Rat2 fibroblasts that were treated for 3 h with 0.1% DMSO or 100 μM L-690330. Shown is a representative example from N = 3 independent experiments. B Densitometry analysis of normalized LC3B II/LCB I ratio in Rat2 cells that were treated as in A. The data are presented as mean of the normalized ratio of LC3B II to LC3B I levels ± SEM. N = 3 independent experiments. *p < 0.05 (one-tailed Mann–Whitney test). C Representative images of Rat2 cells that were treated for 3 h with 0.1% DMSO or 100 μM L-690330 with immunofluorescently labeled endogenous LC3B (green) and nuclei stained with Hoechst 33,258 (blue). Scale bar = 10 µm. D Representative images of Rat2 fibroblasts that were treated for 3 h with 0.1% DMSO or 100 μM L-690330 with p150^Glued–AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. E Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in D. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 5 independent experiments. n = 200 cells (DMSO), 181 cells (L-690330). **p < 0.01 (Student’s t-test). F Western blot analysis of endogenous CLIP-170, p150^Glued, and AP-2β levels and co-immunoprecipitation of endogenous AP-2β with p150^Glued in HEK293T cells that were transfected with control siRNA (siCtrl) or siRNA against human CLIP-170 (siCLIP-170) and treated for 3 h with 0.1% DMSO or 100 μM L-69330. Input, 10% of lysate used for immunoprecipitation. Shown is a representative example from N = 2 independent experiments. G Representative images of Rat2 fibroblasts that were transfected with siCtrl or rat siCLIP-170 and treated for 3 h with 0.1% DMSO or 100 μM L-690330 with PLA p150^Glued–AP-2β signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. H Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in G. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (siCtrl + DMSO) ± SEM. N = 4 independent experiments. n = 167 cells (siCtrl + DMSO), 163 cells (siCtrl + L-690330), 146 cells (siCLIP-170 + DMSO), 164 cells (siCLIP-170 + L-690330). **p < 0.01, ns, nonsignificant (two-way ANOVA followed by Tukey’s multiple-comparison post hoc test). I Western blot analysis of endogenous CLIP-170, P-S6 (Ser235/236), S6, tubulin, LC3B I, and LC3B II levels in Rat2 fibroblasts that were transfected with siCtrl or rat siCLIP-170 and treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA) or treated for 3 h with 100 μM L-690330. Shown is a representative example from N = 6 independent experiments. J Densitometry analysis of normalized LC3B II/LCB I ratio in Rat2 cells that were treated as in I. The data are presented as mean of the normalized ratio of LC3B II to LC3B I levels ± SEM. N = 6 independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ns, nonsignificant (two-way ANOVA followed by Tukey’s multiple-comparison post hoc test). K Representative images of Rat2 cells that were transfected with siCtrl or rat siCLIP-170 and treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA) or treated for 3 h with 100 μM L-690330 with immunofluorescently labeled endogenous LC3B (green) and nuclei stained with Hoechst 33258 (blue). Scale bar = 10 µm

Fig. 6

ABMA and chloroquine prevent rapamycin-induced p150^Glued–AP-2β interaction. A Representative images of Rat2 fibroblasts that were treated for 2 h with 0.1% DMSO, 100 nM rapamycin (RAPA), 60 μM ABMA (ABMA), or 60 μM ABMA and 100 nM rapamycin (ABMA + RAPA) with p150^Glued–AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue Scale bar = 10 μm. B Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in A. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 4 independent experiments. n = 140 cells (DMSO), 144 cells (RAPA), 148 cells (ABMA), 134 cells (ABMA + RAPA). *p < 0.05, ns nonsignificant (one-way ANOVA followed by Bonferroni multiple-comparison post hoc test). C Representative images of Rat2 fibroblasts that were treated for 2 h with 0.1% DMSO, 100 nM rapamycin (RAPA), 50 μM chloroquine (CQ), or 50 μM chloroquine and 100 nM rapamycin for 2 h (CQ + RAPA) with PLA p150^Glued–AP-2β signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. D Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in C. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 3 independent experiments. n = 142 cells (DMSO), 143 cells (RAPA), 135 cells (CQ), 154 cells (CQ + RAPA). **p < 0.01, ns nonsignificant (one-way ANOVA followed by Bonferroni multiple-comparison post hoc test). E Representative images of Rat2 fibroblasts that were transfected with siCtrl or rat siSnap29 for 72 h and then treated for 2 h with 0.1% DMSO or 100 nM rapamycin (RAPA), with p150^Glued–AP-2β PLA signals (magenta), immunofluorescently labeled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. F Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in E. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (siCtrl + DMSO) ± SEM. N = 4 independent experiments. n = 172 cells (siCtrl + DMSO), 161 cells (siCtrl + RAPA), 172 cells (siSnap29 + DMSO), 179 cells (siSnap + RAPA). *p < 0.05, ns nonsignificant (two-way ANOVA followed by Tukey’s multiple-comparison post hoc test). G Representative images of Rat2 fibroblasts that were treated with 0.1% DMSO for 2 h, 100 nM rapamycin (RAPA) for 2 h, 20 mM NH₄Cl (NH₄Cl) for 3 h, or pretreated with 20 mM NH₄Cl for 1 h and treated with 100 nM rapamycin for 2 h (NH₄Cl + RAPA) with p150^Glued-AP-2β PLA signals (magenta), immunofluorescently labelled tubulin (green), and DAPI-stained nuclei (blue). Scale bar = 10 μm. H Quantification of the number of p150^Glued–AP-2β PLA puncta in cells that were treated as in G. The data are expressed as the mean number of PLA puncta per cell, normalized to the control variant (DMSO) ± SEM. N = 3 independent experiments, n = 122 cells (DMSO), 109 cells (RAPA), 125 cells (NH₄Cl), 114 cells (NH₄Cl + RAPA). *p < 0.05, ns nonsignificant (one-way ANOVA followed by Bonferroni multiple-comparison post hoc test)

Fig. 7

Rapamycin induces p150^Glued-AP-2β interaction on lysosomes and regulates lysosomal mobility. A Representative electron microscopy images of Rat2 fibroblasts after 2 h treatment with 0.1% DMSO or 100 nM rapamycin (RAPA) and PLA analysis of p150^Glued–AP-2β with and without (negative control) primary antibodies. White arrows point to PLA signals that co-occurred with organelles that resembled lysosomes or autophagosomes. Black boxes indicate the regions shown in higher magnification. n = 30 cells per variant. N = 3 independent experiments. Scale bar = 500 nm. B Representative images of Rat2 cells treated with 100 nM rapamycin (RAPA) with p150^Glued–AP-2β PLA signals (magenta) immunofluorescently labeled endogenous Lamp1 (green) and DAPI-stained nuclei (blue). Images were acquired using the AiryScan module. Scale bar = 10 μm. (Upper panel) Representative photograph of a single cell. (Lower panel) Close-up of a site with a high intensity of PLA-Lamp1 co-localization. C Representative images of Rat2 cells transfected with Lamp1-GFP (green) for 24 h and then treated with 100 nM rapamycin (RAPA) with p150^Glued–AP-2β PLA signals (magenta), and DAPI-stained nuclei (blue). Images were acquired using the AiryScan module. Scale bar = 10 μm. (Top) Representative photograph of a single cell. (Bottom) Close-up of a site with a high intensity of PLA-Lamp1-GFP co-localization. D Representative images of cells that expressed Lamp1-GFP. The upper row shows the first frames from the time-lapse movies. The area inside the golden circle is considered the “center” compartment, and all movements outside this area are “peripheries.” The lower row shows trajectories (tracks) that were identified by the ImageJ “TrackMate” plugin that were longer than 6.8 mm (100 pixels). Trajectories were color-coded based on their directions, which were established using Pearson correlation coefficient (PCC) calculated by change in the distance from the cell center in time. If the distance was increasing with consecutive frames (PCC: 0.5 to 1) tracks were considered to move to the cell membrane. If distance was decreasing (PCC: − 1 to − 0.5), direction was described as moving to the center. Values in between were marked as oscillating. E Difference in speed of Lamp1-GFP vesicles' movements between rapamycin-treated (RAPA) and control (DMSO) cells. The values are mean trajectories that were identified by the ImageJ “TrackMate” plugin that were at least 6.8 µm (100 pixels) long, divided according to their initial location (center or periphery) and direction (center or cell membrane) as indicated above the graphs. The single dot represents the mean value from one measured cell. N = 4 independent experiments. n = 25 cells for both RAPA and DMSO. *p < 0.05, ns, nonsignificant (Mann–Whitney test)

Fig. 8

Postulated mechanism of autophagy initiation-induced recruitment of p150^Glued and AP-2β to lysosomes. Administration of rapamycin or L-69030 initiates autophagosome formation (AP), which depends on the presence of CLIP-170. Simultaneously, autophagy inducers lead to a decrease in lysosomal pH (LY pH) and recruitment of the dynatin-AP2 complex to the lysosome (LY), presumably leading to intensification of LY retrograde transport to the perinuclear region and subsequent LY fusion with the mature AP. knd knockdown, SBI-0206965 - Ulk inhibitor