CCSer2 knockout causes a decrease in microtubule polarization during migration and retrograde-trafficking defects of integrins. (A) Illustration of dynein’s role in microtubule polarization during cell migration. Dynein anchored to the leading edge of a migrating cell facilitates the polarization of the nucleus–centrosome axis by pulling on the cortical microtubules. The Golgi apparatus is localized near the centrosome and reports on centrosome position. (B) Internalized integrins are trafficked by dynein before being shuttled to recycling endosomes or the lysosome. (C and D) Fluorescence microscopy images of fixed WT (C) or CCSer2-KO (D) cells at 0 and 4 h after wounding or 4 h after wounding with the expression of exogenous CCSer2^WT. Cells were stained with α-GM130 (green) to label Golgi apparatus, phalloidin (pink) to label actin, DAPI (blue) to label the nucleus, and α-GFP to label CCSer2^WT-transfected cells (yellow). Rose histogram plots next to each image indicate the probability of finding Golgi signal 360° around the nucleus, relative to the leading edge. Each concentric circle corresponds to the fraction of the total signal found at a given angle and 90° indicates the direction perpendicular to the angle of the wound. n = 90 cells for nontransfected samples across three biological replicates. n = 98 and 112 cells for WT and CCSer2-KOs expressing CCSer2^WT from 4 to 3 biological replicates, respectively. (E) Fluorescence microscopy images of fixed WT and CCSer2-KO cells, stained with α-paxillin (green) and α-β1-integrin (magenta) to label intracellular and transmembrane FA components, respectively, and DAPI (blue). Grayscale images are set to the same LUT for each antibody; however, LUTs of merged images have been selected for clarity of viewing. (F) Quantification of the raw integrin intensity values per cell, reporting either total cellular intensity (Total), intensity exclusively at paxillin puncta (FAs), or intensity outside paxillin puncta (Outside FAs). n = 74 cells analyzed, across three biological replicates. Error bars are the median ± interquartile range, and statistics were determined with a Kruskal–Wallis test with Dunn’s multiple comparisons. (G) Quantification of the average FA size (paxillin puncta) per cell of WT and CCSer2-KOs. n = 71 cells analyzed, across three biological replicates. Error bars are the median ± interquartile range. Statistical analysis was performed with a Mann–Whitney test. (H) Schematic of a cell in a confluent layer migrating upward to fill in a wound. The orientation of the polarized microtubule network within the dashed box establishes upward moving vesicles as anterograde (pink arrow) and downward moving vesicles as retrograde (green arrow). (I) Representative tracks (>5 μm) of integrin-containing vesicles in WT (top) or CCSer2-KO (bottom) cells. 17 tracks are shown from the WT cell, and 14 tracks are shown from the CCSer2-KO cell. (J) Directional change rate of both anterograde and retrograde individual integrin tracks averaged per cell. n = 62 and 66 track averages analyzed for 31 and 33 WT and CCSer2-KO cells, respectively, across three biological replicates. The error bars are the median ± interquartile range, and statistics were determined with a Mann–Whitney test. (K) Percentage of retrograde events in WT and CCSer2 KO cells. n = 31 and 33 cells analyzed for WT and CCSer2-KO cells, respectively, across three biological replicates. Error bars are the median ± interquartile range. Statistical analysis was performed with a Mann–Whitney test. (L) Maximum speed of the averaged retrograde and anterograde integrin tracks per cell of WT and CCSer2-KO cells. n = 31 and 33 cells analyzed for WT and CCSer2-KO cells, respectively, across three biological replicates. Error bars are the median ± interquartile range. Statistical analysis was performed with a Kruskal–Wallis test with Dunn’s multiple comparisons.