MISP regulates the IQGAP1/Cdc42 complex to collectively orchestrate spindle orientation and mitotic progression

Precise mitotic spindle orientation is essential for both cell fate and tissue organization while defects in this process are associated with tumorigenesis and other diseases. In most animal cell types, the dynein motor complex is anchored at the cell cortex and exerts pulling forces on astral microtubules to position the spindle. The actin-binding protein MISP controls spindle orientation and mitotic progression in human cells. However, the exact underlying mechanism remains to be elucidated. Here we report that MISP interacts with the multidomain scaffolding protein IQGAP1. We further show that MISP binds to the active form of Cdc42 through IQGAP1. Depletion of MISP promotes increased accumulation of IQGAP1 at the cell cortex and a decrease in its Cdc42-binding capacity leading to reduced active Cdc42 levels. Interestingly, overexpression of IQGAP1 can rescue mitotic defects caused by MISP downregulation including spindle misorientation, loss of astral microtubules and prolonged mitosis and also restores active Cdc42 levels. Importantly, we find that IQGAP1 acts downsteam of MISP in regulating astral microtubule dynamics and the localization of the dynactin subunit p150glued that is crucial for proper spindle positioning. We propose that MISP regulates IQGAP1 and Cdc42 to ensure proper mitotic progression and correct spindle orientation.


Immunoprecipitations (IPs)
For IPs with overexpressed proteins, cells from one 15-cm dish were used.
FLAG IPs were performed using FLAG M2 affinity beads (Sigma). For GFP IPs, home-made GFP trap beads were used. Briefly, GFP-binding protein (#49172, Addgene) was purified from E.coli, and after size-exclusion chromatography, covalently coupled to Sepharose beads 4 . Cells for mitotic interactions were pretreated with 250 ng/ml nocodazole (AppliChem) or 100 nM taxol (Sigma) for 17 h. For endogenous IPs, lysates from 10 cm dishes were precleared with sepharose CL-4B beads (Pharmacia Biotech) on a rotating wheel at 4°C for 30 min. Precleared cell lysates were first incubated with 2 µg antibodies against the protein of interest (or normal mouse or rabbit IgG as control, Santa Cruz, sc-2025/sc-2027) on a rotating wheel at 4°C. After one hour, 20 µl protein G-(for mouse antibodies) or protein A-(for rabbit antibodies) coupled sepharose beads were added to the reactions and incubated for another hour. Protein complexes were collected by centrifugation and washed three times with lysis buffer. Immunoprecipitated proteins were eluted from the beads by 5-min boiling in 2x Laemmli buffer.

GTPase activation assay
To determine the activation status of the Rho family GTPases, the RhoA/Rac1/Cdc42 G-LISA Activation Assay (BK135) from Cytoskeleton was used according to the kit manual. Briefly, cells were scraped and lysed on ice with lysis buffer and protease inhibitors provided with the kit, supplied with phosphatase inhibitor cocktail (Roche, 04906845001). After clearing the lysate by centrifugation (10,000 g), protein concentrations were measured and equalized. GTPase activity measurements were carried out according to the manufacturer's protocol.

Cdc42 GEF assay
To find out if MISP is a guanine-nucleotide exchange factor for Cdc42, the RhoGEF exchange assay (BK100) from Cytoskeleton was used according to the manufacturer's instructions. Briefly, purified Cdc42 was incubated with mant-(N-methylanthraniloyl)-GTP, a fluorophore-labeled GTP analog, whose intensity increases dramatically when bound to the GTPase. Fluorescence intensity is measured over time after the addition of the purified potential GEF (MISP) in different concentrations or the positive control Dbs (a proven Cdc42 GEF) with a plate reader at 440 nm (excitation 360 nm).

Quantification of cortical IQGAP1 signals
For quantification of cortical IQGAP1 intensity, cells were fixed with PFA and stained with the rabbit IQGAP1 antibody (Abcam). In case of mitotic cells, the equatorial z-section was used to calculate the relative intensity of the cortical and overall IQGAP1 signal. A circle was drawn just around the mitotic cell (a) and another one right below the cortex (b). Results were obtained with the integrated densities of the circles using the following formula: (a-b)/a.

Quantification of astral microtubule intensity
Quantifications of astral MT intensities were done on methanol fixed samples stained for α-tubulin or EB1. In sum projection of z-stack images, integrated densities were measured using the oval function of ImageJ of the whole mitotic cell (a) and the mitotic spindle itself (b). Astral/spindle MT intensity was calculated with the following formula: (a-b)/b.

Spindle orientation experiments
Cells were seeded onto glass coverslips freshly pre-incubated with 10 µg/ml fibronectin solution for 30 min at RT. After treatment and staining with pericentrin antibody, z-stack confocal images (0.5 µm/stack) were acquired of metaphase cells with the Zeiss LSM-700 or LSM-710 system. After Z-projection of a line going through the two centrosomes in x-y dimension, the angle of the line connecting the two centrosomes in x-z dimension was measured with ImageJ.

Live cell imaging
The length of mitosis was studied in HeLa cells with a Zeiss Observer Z1 Carbamidomethylation of cysteine was set as fixed modification. Variable modifications included oxidation of methionine and deamidation of asparagine and glutamine. One missed cleavage site in case of incomplete trypsin hydrolysis was allowed. A hit was considered significant if the protein score was above the MASCOT identity threshold. Threshold means that an event would be expected to occur at random with a frequency of less than 1% (p<0.01). Proteins were considered as identified if at least one unique peptide 14 had an individual ion score exceeding the identity threshold. The protein score was ranking criteria (the protein with the highest protein score was ranked first), calculated as the sum of all individual peptide scores. Every peptide got its individual ion score based on probability based scoring, which were summed up to give the protein score. If a peptide was fragmented more than once only the highest peptide score was taken for the protein score. For acquiring pre-bleach intensities, five consecutive images were taken at 10% laser power. Then a 4 x 2 µm rectangle at the cell cortex was bleached with 5 laser pulses of 4 lasers (405/458/476/488) at 100% power, each lasting for 1.3 seconds. For recovery measurements 20 single section images were collected at 3s intervals with 10% laser power. Intensities were normalized to pre-bleach data and plots were generated with the FRAP wizard of the Leica LAS AF software. Data points were exported to excel for better visualization.