Depletion of CHC17 clathrin amplifies mitotic centrosome number and induces multipolar spindle formation. (A) HeLa-GFP–α-tubulin cells were treated (72 h) with siRNA-targeting CHC17, CLCs LCa and LCb (siCLC), CHC22, both CHC17 and CLC, or with nonsilencing siRNA (siNonsilencing). Cell lysates were analyzed for protein depletion by immunoblotting for proteins at the left. β-actin serves as a loading control. Molecular mass marker position is shown right. (B and C) HeLa-GFP–α-tubulin cells were treated with siRNA, as in A, and processed for IF to detect γ-tubulin and CLC or CHC22. GFP–α-tubulin was visualized in green. Red-green merge is yellow. Bars, 7.5 µm. (insets) Boxes in each frame are magnified threefold. Bars, 2.0 µm. All images are 2D. (D) The percentage of HeLa-GFP–α-tubulin cells with more than two centrosomes (shaded bars) and the percentage of cells with more than two spindle poles (open bars) were quantified for ≥200 mitotic cells in prometaphase or metaphase for each siRNA treatment. Centrosomes were identified by γ-tubulin staining at spindle poles. Occasional spindle poles without obvious centrosomes were observed, so these were scored separately. Data represent the mean ± SD of four to five independent experiments. Statistically significant results are shown in comparison with nonsilencing siRNA conditions (*, P < 0.1 based on a one-way analysis of variance Friedman’s matched pairs test with Dunn’s multiple comparison post-hoc test).

Centrosome fragmentation is observed in early mitotic cells depleted for CHC17 and CLCs. HeLa-GFP–α-tubulin cells treated (72 h) with siRNA, as indicated (left), were processed for IF to detect PCNT2 and clathrin (CHC17). GFP–α-tubulin was visualized in green. Red-green merge is yellow. Centrosome fragments, defined by PCNT2 staining, that are not localized to a spindle pole are indicated (arrowheads). Bar, 7.5 µm. (insets) Boxed regions are magnified threefold and highlight more diffuse PCNT2 staining at spindle poles in clathrin-depleted samples compared with the nonsilencing siRNA (siNonsilencing) treatment. Bar, 2 µm. All images are 2D.

Interphase centriole centrin-2 puncta are marginally increased by CHC17 and CLC depletion. (A) HeLa–centrin-2–GFP cells were treated (72 h) with siRNA, as indicated (top). Depletion of proteins (left) was detected by immunoblotting siRNA-treated cell lysate. Molecular mass marker position is shown on the right. (B) HeLa–centrin-2–GFP cells were treated (72 h) with siRNA, as indicated (left), and then processed for IF to detect PCNT2 and CHC17. Centrin-2–GFP was visualized in green. Red-green merge is yellow. Bars, 10 µm. (insets) Boxes in each frame are magnified threefold. 2D confocal images are shown with the corresponding brightfield image. Bars, 2 µm. (C) The percentage of HeLa–centrin-2–GFP cells with more than four centrin-2 puncta per cell in interphase (representing G1, S, and G2 phases) was quantified for ≥50 cells for each siRNA treatment indicated. Centrin-2 puncta were identified by centrin-2–GFP that colocalized with PCNT2 detected by IF. Examples of interphase cells with more than two centrin-2 puncta are seen in B for the indicated siRNA treatments. Data represent the mean ± SEM for four independent experiments.

Depletion of either CHC17 or CHC22 increases multinucleation. (A) HeLa-GFP–α-tubulin cells were treated (48 h) with siRNA, as indicated (top). Lysates of treated cells were immunoblotted for the proteins indicated (left). Molecular mass marker position is shown on the right. (B) The percentage of binuclear and multinuclear cells was assessed at 48 or 72 h after each siRNA treatment indicated by nuclear staining with DAPI (>100 cells per treatment per experiment were scored). Targeted siRNA effects were compared with nonsilencing siRNA (siNonsilencing) effects for statistical significance. The mean ± SD of four to five independent experiments is shown (*, P < 0.1).

Clathrin localizes at centrosomes of interphase and metaphase cells. (A) HeLa cells were methanol fixed and processed for IF to detect CHC (X22 mAb) or CLC (rabbit polyclonal) and centrosomal markers PCNT2 (goat polyclonal) and γ-tubulin (mAb) at indicated cell cycle phases. Colocalization of green and red is shown as yellow. Bars, 10 µm. (B) HeLa-SNAP-uLCa clone 3.3 cells were methanol fixed and processed for IF to detect LCa with mAb X16 (CLC), PCM1 (rabbit polyclonal), and PCNT2 (goat polyclonal). Merged colors showing colocalization between labeled proteins are shown as yellow (red and green) and white (red, green, and yellow or cyan and red). Bars: (top) 2 µm; (bottom) 10 µm. (C) HeLa-SNAP-uLCa clone 3.3 cells were methanol fixed and processed for IF to detect TfR (mAb) or EEA1 (mAb), CLC with polyclonal rabbit antibody, and PCNT2 (goat polyclonal). Merged colors showing colocalization between CLC and PCNT2 are shown as pink. Bar, 10 µm. (A–C, insets) Boxes in each frame are magnified threefold. Bars, 2 µm. Images shown are 3D maximum projections.

Covalent cross-linking of SNAP-uLCa disrupts CHC17-dependent pathways of transferrin uptake and recycling. (A) HeLa-SNAP-uLCa clone 3.3 cells were treated with SNAP-tag homodimerizer (BG-GLA-BG, 10 µM) or mock treated (0 µM) with solvent (DMSO) equivalent to the amount used for the BG-GLA-BG treatment. Cell lysates were immunoblotted for dimerized SNAP-uLCa (114** kD), monomeric SNAP-uLCa (57* kD), and endogenous uLCa (32 kD) using antibodies to the SNAP-tag or LCa (mAb X16), as indicated (left). β-actin was detected as a loading control. Protein mass based on migration relative to marker proteins is shown on the right. (B) Clone 3.3 cells were treated (2 h) with 5 µM BG-GLA-BG or equivalent DMSO (mock). Internalization of prebound (4°C) Alexa Fluor 488–transferrin (Tf) was visualized after 10 min or 1 h at 37°C after IF labeling with antibodies to the SNAP-tag or CHC17. Red-blue colocalization in merge is pink. Three-color colocalization in merge is white. Bars, 10 µm. (C) Clone 3.3 cells internalized prebound Alexa Fluor 488–transferrin for 1 h and were labeled by IF with antibodies to the SNAP-tag and EEA1. Red-green colocalization in merge is yellow. Three-color colocalization in merge is white. Bars, 10 µm. (B and C, insets) Boxes are magnified threefold. Bars, 1 µm. Images are 2D. (D) Clone 3.3 cells were treated with BG-GLA-BG or DMSO (mock), as in B, labeled with NHS-S-S-biotin, and then kept on ice (0* and 0 min) or incubated at 37°C for 5, 10, or 15 min. After reduction (stripping) of all samples except 0*, cells were lysed and analyzed by immunoblotting for total TfR and GAPDH (lysate), or biotinylated proteins were SA bound and immunoblotted to detect internalized TfR or control GAPDH. Protein mass based on migration relative to marker proteins is shown on the right. (E) TfR uptake (signal at internalization time point minus signal at time 0) was quantified by densitometry and plotted relative to total biotinylated TfR (0* minus signal at time 0) for four independent experiments, performed as in D for clone 3.3 cells treated with BG-GLA-BG (shaded squares) or DMSO (mock; open circles). Internalized TfR is shown in arbitrary units. Data represent the mean ± SEM of three separate experiments.

Acute inactivation of clathrin during S phase causes centrosome fragmentation in early mitosis. (A) To inactivate clathrin at specific times during the cell cycle, cells expressing the SNAP-uLCa were enriched in S phase by T/T and treated (2 h) with BG-GLA-BG immediately (T₀) upon thymidine washout (0–2 h after T/T) or at 7 h (T₇; 7–9 h after T/T). The timing of treatment relative to cell cycle phase is shown, as empirically determined for the transfected HeLa clones in this study. Corresponding cellular events are illustrated with the nucleus (N), microtubule networks (MT), Golgi (G), centrosomes (C), and focal adhesion structures (FA) indicated. (B) Clone 3.3 cells were enriched to S phase and then treated (2 h) with 5 µM BG-GLA-BG or mock treated, starting at T₀ or T₇ after T/T. Lysates prepared 2, 4, 6, 8, and 12 h after T/T for the T₀ treatment and at 12 h after T/T (12*) for the T₇ treatment were immunoblotted with antibodies to the SNAP-tag or β-actin (loading control). Migration positions of the SNAP-uLCa dimer (**), monomer (*), and molecular mass marker proteins are shown on the right. (C) Clone 3.3 cells were enriched to S phase by T/T block, and, at T₀, cells were treated with BG-GLA-BG (10 µM for 2 h), MLN8054 (500 nM for 4 h), or DMSO (mock; 4 h). 4 h after T₀, cells were methanol fixed and processed for IF to detect CLC (X16), PCM1, and PCNT2. Red-blue localization is pink in merge, and three-color localization is white. Bar, 10 µm. (D) Clone 3.3 cells were enriched to S phase by T/T block, and then, at T₀, cells were treated with BG-GLA-BG (10 µM for 2 h), MLN8054 (500 nM for 8 h), or DMSO (mock; 8 h). As cells entered mitosis (8 h after T₀), they were fixed with PFA and processed for IF to detect the SNAP-tag, γ-tubulin, and α-tubulin. Red-blue localization is pink in merge, and green-blue localization is cyan. Bar, 10 µm. (C and D, insets) Boxed regions are magnified threefold. Bars, 2 µm. Images are 3D maximum projections. (E) Clone 3.3 cells were enriched to S phase by T/T block, and, at T₀, cells were treated with BG-GLA-BG (10 µM for 2 h), MLN8054 (500 nM for 4 or 8 h), or DMSO (mock; 4 or 8 h, as in C and D). Treated cells were processed for IF to detect γ-tubulin and α-tubulin and were scored for centrosome fragmentation (γ-tubulin fragments shown by arrowheads in D). Data represent the mean ± SEM of two separate experiments (***, P < 0.001). 70 cells were scored per experiment. (F) From the experiment in D, 3D maximum projections of images of prometaphase and metaphase cells were analyzed for the distribution of γ-tubulin at the spindle pole–associated centrosome using the Radial Profile Extended plug-in in ImageJ. The normalized, integrated fluorescence intensity (y axis) is compared with the distance (pixels) of detected γ-tubulin at and surrounding the centrosome for cells treated with BG-GLA-BG, MLN8054, or DMSO (mock). The inset graph shows a magnified portion of the main graph and highlights the γ-tubulin fragments detected further out from the spindle pole–associated centrosome. Data represent the mean ± SEM of two to five prometaphase or metaphase cells (***, P < 0.001).

Clathrin colocalizes with TACC3 and ch-TOG at interphase centrosomes, and ch-TOG levels are reduced by clathrin inactivation at S phase. (A) Asynchronous clone 3.3 cells were treated (2 h) with 5 µM BG-GLA-BG or DMSO (mock) and then methanol fixed and processed for IF to detect LCa (mAb X16), PCNT2, and ch-TOG. Three-color colocalization is white in the merged image, and red-blue colocalization is pink. Bar, 10 µm. (B) Asynchronous clone 3.3 cells were treated as in A, methanol fixed, and then processed for IF to detect LCa (rabbit polyclonal), PCNT2, and TACC3. Three-color colocalization is white in the merged image. Bar, 10 µm. (A and B, insets) Boxed regions are magnified threefold. Bars, 1 µm. Images are 3D maximum projections. (C) Clone 3.3 cells were enriched to S phase by T/T block and treated (2 h) at T₀ with BG-GLA-BG or mock treated, as in A. Lysates prepared 4, 6, and 8 h after T/T were immunoblotted to detect ch-TOG, α-tubulin, and GAPDH, as indicated. Molecular mass marker position is shown on the right. (D) Clone 3.3 cells were enriched to S phase and treated with BG-GLA-BG or DMSO, as in C. Lysates prepared 2, 4, 6, 8, and 12 h after T/T were immunoblotted to detect TACC3 and GAPDH, as indicated. The immunoblot shown here comes from the same transfer membrane used to generate the immunoblot in Fig. S3 A. The transfer membrane was cut into horizontal strips of different molecular mass zones to detect all proteins shown here and in Fig. S3 A. Molecular mass marker position is shown on the right. (E) The densitometric quantification of immunoblots from multiple experiments performed as in C and D for clone 3.3 cells treated with BG-GLA-BG (shaded squares) or mock treated (open circles) relative to the loading control GAPDH. Data represent the mean ± SD of three to four separate experiments.

Clathrin inactivation or inhibition of Aurora A kinase reduces clathrin and ch-TOG at centrosomes. (A) Clone 3.3 cells were synchronized by T/T block and treated at T₀ with BG-GLA-BG for 2 h or the Aurora A kinase inhibitor MLN8054 for 8 h. 8 h after T₀, cells were fixed in methanol and processed for IF to visualize clathrin (mAb X16), PCNT2, and ch-TOG. Three-color colocalization in merge is white, and red-blue colocalization in merge is pink. Bars, 5 µm. (insets) Boxed regions are magnified. Bars, 2 µm. (B) Quantification by Pearson’s coefficient of overlap between the centrosomal marker PCNT2 and clathrin or ch-TOG for each treatment in A. Data represent the mean ± SEM from two independent experiments (**, P < 0.01; ***, P < 0.001).

Clathrin and pericentrin traffic along a shared route toward late-interphase centrosomes. (A) Asynchronous clone 3.3 cells were transiently transfected with GFP-pericentrin 48 h before labeling SNAP-uLCa with SNAP-TMR-Star. Cells were analyzed over 2 h by time-lapse confocal microscopy, taking 2D images every 10 min (37°C in 5% CO₂ at 95% humidity). Colocalization (yellow in merged images) between SNAP-TMR-Star–labeled SNAP-uLCa and GFP-pericentrin occurs at centrosomes (arrowheads) and adjacent submicrometer vesicles (arrows). Top right corners of merged images show time interval in minutes. Bar, 10 µm. Images are 3D maximum projections. (B) Asynchronous clone 3.3 cells transiently transfected with GFP-pericentrin and labeled with SNAP-TMR-Star were imaged by confocal microscopy. Colocalization of pericentrin and SNAP-uLCa is shown (yellow in merged image) at two separated centrosomes of a late-interphase cell and overlayed with the brightfield image. Bar, 10 µm. Images are 2D. (C) The cell shown in B was magnified twofold and analyzed by time-lapse confocal microscopy for 2 min with 10-s imaging intervals. SNAP-uLCa localizes with GFP-pericentrin at the centrosome, and pericentriolar SNAP-uLCa vesicles move toward the centrosome (arrowheads). Top right corners of merged images show the time interval. Bars, 5 µm. (insets) Boxed regions are magnified threefold. Bars, 2 µm. Images are 3D maximal projections.