CMPD1 induced severe mitotic arrest in multiple cancer cell lines

(A) Chemical structure of CMPD1. (B) FACS analysis of MDA-MB-231 cells treated with DMSO or 1, 5, or 10 µM CMPD1 for 24 hours. (C) Representative time-lapse images of MDA-MB-231, CAL-51, and T-47D cells treated with either DMSO or 5 μΜ CMPD1. Time is indicated in minutes post-NEBD. (D) Quantification of mitotic duration of MDA-MB-231, CAL-51, and T-47D treated with DMSO, 1, 5, or 10 μΜ CMPD1. n = 30-60 cells pooled from two independent replicates. Error bars are mean ± s.d. Asterisk above each dot plot indicates the significant difference (p < 0.0001, Tukey’s multiple comparison test) between the indicated drug-treated condition and the control.

CMPD1 treatment specifically attenuates mitotic fidelity of cancer cells

(A) Representative time-lapse images (interval: 3 min) of RPE1, MDA-MB-231, and CAL-51 cells treated with DMSO or low dose of CMPD1 (10 and 50 nM). All of these three cell lines express H2B-GFP. Time is indicated in minutes post-NEBD. (B) Quantification of mitotic duration of cells shown in (A). n = 70, 60, 60, 60, 60, 65, 50, 84, 101 cells from left to right. (C) The enlarged plot of a red box region shown in (B). (D) Representative images of mitotic defects. Note that “misaligned” and “no metaphase plate” were annotated only when the cells exhibited this phenotype upon anaphase onset. (E) Normalized percentage of mitotic cells exhibiting accurate chromosome segregation in each condition. The data derived from CMPD1-treated cells were normalized to the data derived from DMSO-treated cells for each cell line. (F) Left: representative time-lapse images (interval: 3 min) of untreated RPE1 cells, and RPE1, MDA-MB-231, and CAL-51 cells treated with 10 nM PTX. Right: The quantification of the fraction of mitotic cells showing mitotic errors. The p-value was calculated by Tukey’s multiple comparisons test. Time is indicated in minutes post-washout. (G) Representative images of RPE1, MDA-MB-231, and CAL-51 cells in CMPD1 washout experiments. Briefly, cells were treated with 2 μΜ CMPD1 for 4 hours, followed by a wash with complete media and live-cell imaging. Time is indicated in minutes post-washout. (H) The mitotic duration of mitotic cells that were arrested by CMPD1 in the beginning of the imaging. The mitotic duration was defined as the time from the start of the imaging to anaphase onset or mitotic exit. The mean value was shown at the top right of each condition. n = 54, 60, and 37 cells for RPE1, MDA-MB-231, and CAL-51 cells, respectively. The p-value was calculated by Tukey’s multiple comparisons test. (I) Quantification of the percentage of mitotic cells showing normal chromosome segregation. The mean value was shown at the top right of each condition. n = 50 cells in each condition pooled from two independent experiments. The p-value was calculated by Tukey’s multiple comparisons test. Error bars are mean ± s.d.

CMPD1 inhibits both anchorage independent growth and tumor growth in mice

(A) Representative images of anchorage independent growth assay using MDA-MB-231 cells treated with DMSO, CMPD1 at different conditions (0.01, 0.05, 0.1, 0.25, 0.5, and 1 µM), or 10 μΜ PTX. (B) The normalized number of cell colonies formed in each condition. Error bars are mean ± s.d. Three independent experiments were performed for each condition. (C) Schematic diagram of the mouse xenograft experiment and the drug treatment schedule. (D) The image of tumors in each condition at the time of necropsy. The tumors were arranged in order based on their size. (E) The quantification of gross weight of tumors at the time of necropsy. Error bars are mean ± s.d.

CMPD1 induces microtubule depolymerization

(A) Representative time-lapse images of CAL-51 cells expressing α-Tubulin-mNeonGreen and H2B-mScalret upon the treatment with DMSO, 2 μΜ CMPD1, or 10 μΜ ΜG132 along with 2 μΜ CMPD1. CMPD1 was added into the cell culture media immediately after images at the first time point were acquired. Time is indicated in minutes. (B) The quantification of the signal levels of mitotic spindles over time in each condition shown in (A). (C) Representative time-lapse images (interval: 2 min) of a G2 phase CAL-51 cell expressing α-Tubulin-mNeonGreen and H2B-mScarlet in the presence of 2 μM CMPD1. Note that the two bright dots in the Tubulin channel indicate the two clustered centrosomes before NEBD. The arrow in the DNA channel indicates the time of NEBD. Time is indicated in minutes. (D) Representative kymographs depicting microtubule plus and minus end dynamics before and after the addition of polymerization mix supplement with 15 µM tubulin alone, or 15 µM tubulin supplemented with either 20 µM CMPD1 or 5 µM vinblastine, as indicated (GMPCPP-stabilized microtubule seeds, magenta; microtubules polymerized from seeds, green; see Methods). Plus ends are positioned to the right, and minus ends are positioned to the left of the seeds in all kymographs. Arrows indicate the time of addition of tubulin alone, or tubulin plus drug. (E) Plot depicting fraction of microtubules with detectable plus or minus ends 1 minute after addition of drug, or tubulin alone (n = 27, 29, and 30 microtubules from left to right). Error bars indicate mean ± s.d. (F) Representative images from a time-lapse sequence showing microtubule plus and minus end lengths 10 seconds prior to, and 5 and 15 seconds after addition of polymerization mix supplemented with 20 µM CMPD1. (G) Plots depicting plus end catastrophe frequencies (n = 51 and 41 microtubules from left to right), plus end growth rates (n = 172 and 91 events from left to right), and normalized maximum length (n = 55 and 41 microtubules from left to right) achieved over the imaging period. (H) Plots depicting minus end catastrophe frequencies (n = 51 and 41 microtubules from left to right), minus end growth rates (n = 129 and 63 events from left to right), and normalized maximum length (n = 55 and 40 microtubules from left to right) achieved over the imaging period. Error bars indicate mean ± s.d. Data were pooled from at least 2-4 independent replicates (G-H).

CMPD1 inhibits cancer cell migration and invasion

(A) Representative images of H2B-mScarlet CAL-51 cells treated with DMSO, or CMPD1 at different concentrations (0, 0.1, 0.5, 1, 10 μΜ) at three different time points (0, 24, and 48 hr post-treatment). A wound (cell-free zone) was created using a tip, followed by the addition of indicated drugs and live-cell imaging (interval: 30 min). (B) Quantification of cell migration speed when cells were treated with indicated drugs. Each condition was normalized to the speed of DMSO-treated cells. The distance between the edge of the wound was measured using imageJ macro. Results are the mean ± s.d. N = 3 biological replicates. (C) Representative images of trans-well invasion assay using MDA-MB-231 cells treated with DMSO, CMPD1 (0.1, 1, 2, 5 μΜ), or 10 μΜ PTX. (D) The quantification of the average number of invaded cells in each condition shown in (C). (E) Left: example images of blood vessels in xenograft tumors derived from mice treated with DMSO or 1.0 mg/kg CMPD1 as shown in Fig. 3C. Two sets of example images acquired from different tumors were shown. Right: the quantification of the percentage of blood vessels infiltrated with cancer cells. Error bars are mean ± s.d.

MK2 inhibition enhances the efficacy of microtubule inhibitors in cancer cells

(A) Left: representative time-lapse images of mitotic CAL-51 cells treated with indicated combinations of drugs (10 μΜ MK2i, 1 nM vinblastine, 5 nM vinblastine). Right: quantification of the mitotic duration, mitotic error rate, mitotic slippage rate, and the frequency of death in mitosis. The fate of mitotic cells was color-coded. N = 100 cells pooled from two biological replicates for each condition. Time is indicated in minutes post-NEBD. (B) Volcano plot displaying changes in gene expression following CMPD1 treatment in MDA-MB-231 cells. RNA-seq was conducted with three biological replicates. Differentially expressed genes (DEGs) are highlighted in red (up-regulated) and blue (down-regulated). (C) Pathway enrichment analysis of differentially expressed genes using the Gene Ontology (GO) Biological Processes (BP). Enrichment analysis was performed with the DAVID online tool. (D) Comparison of GO Biological Process enrichment analysis of DEGs unique to MDA-MB-231 cells relative to RPE1 cells. Genes uniquely up- or down-regulated in MDA-MB-231 cells, but not in RPE1 cells, were subjected to GO BP enrichment analysis.

The model of CMPD1-mediated cytotoxicity effects on cancer cells.

CMPD1 exhibits its tumor-specific cytotoxicity likely via two pathways. First, CMPD1 acts as a kinase inhibitor which prevents p38 MAPK-dependent phosphorylation of MK2, leading to the disruption of proper actin remodeling and spindle formation during mitosis. Second, CMPD1 serves as a MTA which can specifically induce depolymerization at the plus ends of microtubules. Both impaired actin reorganization and attenuated microtubule filament formation inhibit cancer cell migration and invasion, thereby preventing metastasis. On the other hand, failure of spindle assembly during mitosis causes the extended prometaphase arrest and the diminished mitotic fidelity, resulting in apoptosis of tumor cells and tumor shrinkage.