## Results: 8

FIGURE 2:. From: A mechanosensory system governs myosin II accumulation in dividing cells.

*iqgap1*,

*iqgap2*,

*cortII*) strains, only IQGAP2 is recruited to the micropipette in dividing cells. The IQGAP2 response magnitude is significantly higher than that of rac1A, IQGAP1, and cortexillin II (ANOVA: p = 0.001). (B) IQGAP2 does not accumulate at the micropipette in

*myoII*- or

*cortI*-null cells (white arrows; ANOVA: p = 0.008). Note that

*cortI*-mutant cells, which form a lot of blebs, showed strong accumulation of IQGAP2 at the blebs (arrowhead). Scale bars,10 μm.

FIGURE 5:. From: A mechanosensory system governs myosin II accumulation in dividing cells.

*iqgap2*-null cells. (A) In cells grown on surfaces (without perturbation), GFP-myosin II accumulated at the cleavage furrow at comparable levels in WT and

*kif12*-null cells. (B) GFP-myosin II levels at the cleavage furrow were similar in

*incenp*-rescued cells and

*incenp*-null cells. (C) Images of GFP-myosin II in WT,

*cortI*,

*iqgap1*,

*iqgap2*, and

*iqgap1/2*cells grown on surfaces. By pairwise analysis of each mutant as compared with the WT control, only the

*iqgap2*null is significantly lower (Student's

*t*test: p = 0.008). Scale bars, 10 μm. The bar graphs show the mean furrow-to-pole intensity ratio (I

_{f}/I

_{p}± SEM), and the sample sizes are listed on the bars.

FIGURE 3:. From: A mechanosensory system governs myosin II accumulation in dividing cells.

*rac1A/C*double mutants,

*cortII*single mutants, and

*iqgap1/2*double mutants (Student's

*t*test: p = 0.9, 0.1, and 0.07, respectively) but not

*iqgap2*and

*cortI*single mutants (Student's

*t*test: p < 0.0001 for both cases). GFP-myosin II responsiveness is higher in

*iqgap1*null cells than in WT (Student's

*t*test: p = 0.01). Although GFP-myosin II recruitment in

*cortI/II*is not statistically different from WT (Student's

*t*test: p = 0.4), the recruitment behavior is not WT like (see Supplemental Figure S2). (B) Rescue of

*iqgap2*cells with IQGAP2 restored the myosin II mechanosensitive accumulation, whereas IQGAP1 overexpression in WT cells inhibited myosin II mechanosensitive accumulation (Student's

*t*test: p = 0.01). (C) Cortexillin I showed a similar dependence as myosin II on IQGAP function. Cortexillin I accumulated in

*iqgap1*and

*iqgap1/2*and

*cortI/II*mutants but not in

*iqgap2*single-mutant cells (ANOVA: p = 0.0003). Scale bars (A, C), 10 μm. (D) The effective cortical tension of WT,

*iqgap1*,

*iqgap2*,

*iqgap1/2*,

*cortI*,

*cortII*, and

*cortI/II*mutant cells. Each strain is significantly different from WT (and each other; Student's

*t*test: p < 0.0001 for all except

*cortII*, p = 0.006). Sample sizes are listed on the bar graph.

FIGURE 6:. From: A mechanosensory system governs myosin II accumulation in dividing cells.

*kif12*-rescued cells (as compared with unperturbed cells; Figure 5A). This amplification was lost in

*kif12-*null cells (Student's

*t*test: p = 0.0002). (B) INCENP was required for stress-induced amplification of myosin II accumulation at the cleavage furrow cortex. Under agarose overlay, the rescued

*incenp*cell line showed threefold higher myosin II accumulation than the

*incenp*-null cells had (Student's

*t*test: p = 0.001). (C) WT cells treated with the DMSO carrier and WT cells treated with 10 μM nocodazole showed comparable accumulation of myosin II in response to agarose overlay. The spindle was confirmed to be disrupted by tracking RFP-tubulin. (D) Except for

*iqgap1*- and

*iqgap2*-rescued cells, all mutants showed significant differences in the GFP-myosin II I

_{f}/I

_{p}ratios as compared with WT cells. Statistical analysis was performed using an ANOVA with a Student Neuman–Keuls post hoc test. The p values are indicated in the inset, and level of significance is indicated with asterisks. Scale bars, 10 μm. The bar graphs show the mean furrow-to-pole intensity ratio (I

_{f}/I

_{p}± SEM), and the sample sizes are listed on the bars.

FIGURE 4:. From: A mechanosensory system governs myosin II accumulation in dividing cells.

*iqgap2*- and

*iqgap1/2*-null cells. WT control data were redrawn from Figure 1A. The difference between the WT control and the

*iqgap*single and double mutants was significant (ANOVA: p < 0.0001). (B) WT and

*iqgap1*-null cells had normal spindle morphology and cleavage furrow localization of GFP-kif12 (white arrows). The

*iqgap2*- and

*iqgap1/2*-null cells had disrupted spindle morphology, and GFP-kif12 furrow localization was reduced or absent. Scale bars (A, B), 10 μm. (C) The quantification of the intensity ratio of GFP-kif12 at the furrow to polar cytoplasm (I

_{f}/I

_{p cyto}). The I

_{f}/I

_{p cyto}values in WT and

*iqgap1*cells were indistinguishable (Student's

*t*test: p = 0.44). The intensity ratio was significantly lower in

*iqgap2*(Student's

*t*test: p = 0.03) and

*iqgap1/2*(Student's

*t*test: p = 0.02) cells as compared with WT cells. (D) The percentage of

*iqgap2*and

*iqgap1/2*null cells with normal spindles was much lower than that in WT and

*iqgap1*(comparison of proportions: p ≤ 0.0006). (E) The

*iqgap2*and

*iqgap1/2*mutants were deficient in cytokinesis, as the percentage of cells completing cytokinesis was much lower than for WT and

*iqgap1-*null cells (comparison of proportions: p ≤ 0.006). The sample sizes for C–E are listed on the bar graphs. (F) Cartoon summarizes data from this figure and Figures 1–3. Mechanical stress detected by the myosin II/cortexillin I mechanosensor is transduced through IQGAP2 to kif12 and INCENP. IQGAP2 is also important for maintaining normal spindle morphology and is required to suppress IQGAP1, which inhibits mechanosensing.

FIGURE 7:. From: A mechanosensory system governs myosin II accumulation in dividing cells.

*myoII-*null cells) accumulated at the cleavage furrow under agarose overlay, but ΔBLCBS did not integrate as tightly at the lateral edges of the cleavage furrow cortex. Line scans at the line in the fluorescence images show the distribution of myosin II in the cortex vs. the central region of these furrows. Scale bar, 10 μm. A comparison by confocal microscopy is presented in Supplemental Figure S5A. (B) The bar graph shows the quantification of the mean furrow-to-pole intensity ratio of (I

_{f}/I

_{p}) of CIT-WT and CIT-ΔBLCBS myosin II (Student's

*t*test: p = 0.77). (C) The bar graph shows the quantification of the mean ratio of lateral furrow cortex to furrow center (I

_{f}/I

_{f center}) of CIT-WT and ΔBLCBS myosin II (Student's

*t*test: p = 0.0004). Sample sizes are shown on the bar graphs in B and C. (D, E) The graphs depict the furrow-thinning dynamics of

*myoII*-null cells (black line) and

*myoII*-null cells complemented with a CIT-labeled WT myosin II (gray line), long-lever-arm 2xELC myosin II (light gray line), and the short-lever-arm ΔBLCBS myosin II. (D) The WT and 2xELC myosin II, which had indistinguishable furrow ingression dynamics, whereas

*myoII*cells showed accelerated furrow thinning. (E) Individual ΔBLCBS furrow ingression dynamics were similar to those of

*myoII*-null cells, but they failed to collapse onto a single universal curve. Unaveraged

*myoII*-null, WT, and 2xELC curves are given in Supplemental Figure S5, B–D.

FIGURE 8:. From: A mechanosensory system governs myosin II accumulation in dividing cells.

FIGURE 1:. From: A mechanosensory system governs myosin II accumulation in dividing cells.

_{p}/I

_{o}], where I

_{p}is the intensity in the pipette and I

_{o}is the intensity of the opposite cortex) in WT and rescued cells. The distributions of GFP-kif12 and GFP-INCENP response intensity ratio in WT and rescued cells are significantly higher than in WT cells expressing either GFP or mCherry alone (Student's

*t*test: p < 0.0001) and cells expressing GFP-aurora (Student's

*t*test: p = 0.003). The distribution of GFP-aurora is not different from that of GFP/mCherry (Student's

*t*test: p = 0.14). Each dot represents an aspirated cell. The numbers in the differential interference contrast images represent the start of the movie (time 0 s) and the time of the response after the pressure was applied. (B) GFP-kif12 accumulated at the micropipette independent of microtubules but dependent on myosin II. Cells were treated with 10 μM nocodazole (NOC), and the RFP-tubulin was monitored to confirm that the mitotic spindle was disrupted. The response distribution of GFP-kif12 in

*myoII*-null cells is significantly lower than in the untreated (reproduced from A) and 10 μM nocodazole-treated cells (ANOVA: p < 0.0001). Note that the bright spherical structures shown in GFP-kif12 images are the centrosomes. (C) GFP-myosin II mechanosensitive accumulation occurred similarly in WT vs.

*kif12*-null cells and

*incenp*-rescued vs.

*incenp*-null cells. The distributions are statistically indistinguishable (WT and

*kif12*null, Student's

*t*test: p = 0.22;

*incenp*rescued and

*incenp*null, Student's

*t*test: p = 0.53). Scale bars (A–C), 10 μm. (D) The effective cortical tension (T

_{eff}) of

*kif12*and

*incenp*null cells was only slightly reduced compared with WT and GFP-INCENP–rescued controls (Student's

*t*test: p = 0.08 and 0.67, respectively). Sample sizes are listed on the bar graph. (E) Cartoon summarizes the data in this figure. Mechanical stress directs the recruitment of myosin II and cortexillin I, which in turn are required to direct mechanosensitive accumulation (mechanotransduction) of kif12 and INCENP.