Fig 2: CCL19 gradient induces asymmetric shootin1b activation to drive chemotaxis. A,B) A gradient of CCL19 was applied to dendritic cells cultured in a mixture of collagen gel and Matrigel (left panel, A). One hour after gradient application, time‐lapse phase‐contrast/fluorescence images of WT (A) and shootin1b KO (B) dendritic cells expressing EGFP were obtained. Nuclei were also visualized by Hoechst to accurately trace the trajectories of cell migrations (shown in right panels and Video S8, Supporting Information). The pictures show representative images from the time‐lapse series taken every 1 min for 90 min. The tracing lines (magenta) indicate dendritic cell migration for 90 min. Scale bar: 50 µm. The right panels depict trajectories of dendritic cell migrations. The initial cell positions are normalized at x = 0 µm and y = 0 µm. C) Scheme of chemotaxis index. Chemotaxis index was calculated as the ratio of the straight distance toward the CCL19 source (S) to the total distance (T) by tracing the migration trajectories in (A, B). D,E) Analyses of migration speed (D) and chemotaxis index (E) of WT and shootin1b KO dendritic cells expressing flag‐GST (control flag‐tagged protein), and KO cells expressing flag‐shootin1b‐WT (H), flag‐shootin1b‐AA (I) and flag‐shootin1b‐DD (J) under the CCL19 gradient. For multiple comparison, one‐way ANOVA with Turkey's post hoc test was performed. WT, n = 72 cells; KO, n = 65 cells; KO + flag‐shootin1b‐WT, n = 73 cells; KO + flag‐shootin1b‐DD, n = 72 cells; KO + flag‐shootin1b‐AA, n = 64 cells. See Video S8 (Supporting Information). F) Dendritic cells were transfected with myc‐shootin1b to visualize shootin1b. After the stimulation by CCL19 gradients for 30 min, they were fixed and immunolabeled with anti‐myc and anti‐pSer249 shootin1 antibodies. Fluorescence images show the detected phosphorylated shootin1b and shootin1b in a dendritic cell. White dashed lines indicate the boundary of a dendritic cell and the center line that separates the high side (CCL19 source side) and low side. Scale bar: 10 µm. G) Quantitative data for shootin1b activation (phospho‐shootin1b/shootin1b) in the CCL19 source side (high side) and low side of dendritic cells. Two‐tailed Mann–Whitney U‐test for phospho‐shootin1b/total shootin1b between the high side and low side (n = 30 cells). H–J) Migration trajectories of shootin1b KO cells expressing flag‐shootin1b‐WT (H), flag‐shootin1b‐AA (I), and flag‐shootin1b‐DD (J) stimulated by the CCL19 gradient. Data represent means ± SEM; *, p < 0.05; **p < 0.02; ***p < 0.01; ns, not significant.

Fig 3: Shootin1b mediates generation of weak forces for dendritic cell migration. A) Schema of traction force microscopy in a semi‐3D condition. Dendritic cells were cultured on laminin‐coated polyacrylamide gels embedded with 200‐nm fluorescent beads in a mixture of collagen gel and Matrigel. Traction force under the cells was monitored by visualizing force‐induced deformation of the gel, which is reflected by the bead movement (red arrows). B) Overlayed differential interface contrast (DIC) and fluorescence images showing a dendritic cell migrating under the semi‐3D condition in (A) in the presence of 200 ng mL−1 CCL19. See Video S1, Supporting Information. The pictures show representative images from the time‐lapse series taken every 3 s for 270 s. The original and displaced positions of the beads in the gel are indicated by green and red colors, respectively. The cells were visualized by CMFDA staining (blue color); dashed lines indicate the boundaries of the cells. The kymographs (panel below) along the axis of bead displacement (white dashed arrows) at indicated areas 1 and 2 show movement of beads recorded by every 3 s. The bead in area 2 is a reference bead. Scale bar: 5 µm (in the inset, 1 µm). C) Analyses of the magnitude of the traction force under WT and shootin1b KO dendritic cells stimulated by 20 and 200 ng mL−1 CCL19. For multiple comparisons, one‐way ANOVA with Turkey's post hoc test was performed. WT + 20 ng mL−1 CCL19, n = 12 cells; WT + 200 ng mL−1 CCL19, n = 12 cells; KO + 20 ng mL−1 CCL19, n = 12 cells; KO + 200 ng mL−1 CCL19, n = 12 cells. D) Analyses of migration speed of WT and shootin1b KO dendritic cells cultured in a mixture of collagen gel and Matrigel in the presence of 20 and 200 ng mL−1 CCL19. For multiple comparisons, one‐way ANOVA with Turkey's post hoc test was performed. WT + 20 ng mL−1 CCL19, n = 111 cells; WT + 200 ng mL−1 CCL19, n = 137 cells; KO + 20 ng mL−1 CCL19, n = 107 cells; KO + 200 ng mL−1 CCL19, n = 110 cells. See Figures S1 and Video S2, Supporting Information. E) Scheme showing the angles of traction force (θ1) and dendritic cell migration (θ2) (left panel). The angles were calculated from the data of sequential 30 images of migrating dendritic cells in the presence of 200 ng mL−1 CCL19 (B). Right panel shows the correlation analysis between the traction force angle and migration angle of dendritic cells. F) Fluorescence images of a dendritic cell co‐stained with anti‐shootin1b antibody and phalloidin‐Alexa 555 for F‐actin. An enlarged view of the rectangular region is shown to the right. Arrowheads indicate shootin1b co‐localization with F‐actins in filopodia. The images were obtained by STED microscopy. Scale bar: 10 µm (in the inset, 2 µm). G) Overlayed DIC and fluorescence images showing a shootin1b KO dendritic cell migrating under the semi‐3D condition in (A) in the presence of 200 ng mL−1 CCL19. For detailed explanations, see (B). See Video S1 (Supporting Information). H, I) Effects of shootin1b KO on the traction force (H) and migration speed (I) increased by 200 ng mL−1 CCL19. Two‐tailed unpaired Student′s t‐test was performed (n = 3 independent experiments). Data represent means ± SEM; **p < 0.02; ***p < 0.01; ns, not significant.

Fig 4: Sema3A affects DC cytoskeletal changes and migration. A Transwell chambers analysis of rSema3A-triggered DC migration, n=3/group. B, C qRT-PCR and western blot analysis the expression of Sema3A under CCL19 treatment. n=3/group. D Flow cytometry analysis of CCR7 expression under rSema3A treatment. n=3/group, NS, not significant. E, F immunofluorescence staining revealed the rSema3A-triggered cytoskeletal remodel (E), and Glu, si-Sema3A-triggered cytoskeletal remodel (F). The right side of F is a local enlarged image. DNA (blue) was stained with DAPI, F-actin (green). Scale bars, 10 μm. G Confocal imaging revealed DC-T cell interactions and synapses in WT/CKO-derived DCs following Glu and si-Sema3A treatment, DNA (blue) was stained with DAPI, F-actin (green). Scale bars, 10 μm. Data are presented as mean ± SD. Statistical analysis was performed using two-tailed Student's t-test (B), One-Way ANOVA (A, D). *P < 0.05, ***P < 0.001.

Fig 5: (A) Volcano plot of phosphoproteomic profiling of FACS-sorted naive CD8+ T cells, showing differential phosphosite abundance between experimental groups. Log2 fold change is plotted on the x-axis, significantly regulated phosphosites are highlighted in red (n=3). (B) STRING functional association network of TAOK3, generated using STRING v12 for Mus musculus. (C) Polarization index of FACS-sorted naive CD8+ T cells following CCL19 stimulation over time, normalised to baseline (mean of 9 wells derived from 3 biological replicates, each measured in technical triplate). Cell polarization was assessed by immunostaining for CD11a (leading edge) and CD43 (uropod). A representative snapshot at 3 min post-stimulation is shown. (D) F-actin content assessed by Acti-stain Phalloidin staining in naive CD8+ T cells. Splenocytes were stimulated with CCL19 over time and gated on viable CD45+CD3+TCRβ+CD8+CD62L+CD44low cells (n=2). (E) Intracellular ROS levels in naive CD4+ and CD8+ T cells were quantified by DHR123 and MitoSOX staining using flow cytometry. Cells were gated on CD45+CD3+TCRβ+CD62L+CD44low cells. (F) Mitochondrial polarization of naive CD8+ T cells measured by flow cytometry as the ratio of TMRM to MitoTracker Green staining. (G) Representative flow cytometry plots of TMRM and MitoTracker Green corresponding to panel (F, H). Mitochondrial membrane polarisation in FACS-sorted naive CD8+ T cells following 1,5h ex vivo treatment with multiple doses of NADPH oxidase inhibitor VAS2870. Mitochondrial polarization was assessed by TMRM staining using flow cytometry (n=2). (I) Representative flow cytometry histograms of TMRM fluorescence intensity corresponding to panel (H, J). Seahorse extracellular flux analysis of oxygen consumption rate (OCR) in purified naive CD8+ T cells. OCR was measured under basal conditions and following sequential treatment with oligomycin (Oligo), FCCP, and rotenone plus antimycin A (Rot/AA), as indicated. Graphs represent the mean ± SEM of four biological replicates, with each data point being calculated from averaged triplicate (TN) or quadruplet (TM) readings across 5 sequential time points right before oligo addition (basal OCR), or 3 sequential time points following oligo or FCCP treatment (maximal OCR)(n=4) (K) TEM analysis of naive CD8+ T-cell ultrastructure with representative high-magnification views of the Golgi apparatus, mitochondria, and autolysosomal compartments. Taok3-deficient cells display Golgi swelling, altered mitochondrial morphology, and increased autophagolysosomal structures. (L) Mitochondrial morphology was scored in a blinded manner and classified as normal, mildly defective, or severely defective based on shape, outer membrane integrity, cristae organisation, and evidence of mitophagy. 207 Taok3 KD+/+; 334 Taok3 KD KI/KI mitochondria, derived from Taok3 KD+/+ and 72 Taok3 KD KI/KI images of randomly selected cells. Statistical analysis: (C) Two-way ANOVA. (E, F) Welch’s T test. (L) χ² test of independence. Data are representative of at least two independent experiments. E,F = each dot represents an individual mouse. For all other panels, this is indicated in the corresponding figure legends. *P <.05, **P <.01, ***P <.001 and ****P<.0001.