Cantharidin-induced mitotic arrest is associated with the formation of aberrant mitotic spindles and lagging chromosomes resulting, in part, from the suppression of PP2Aalpha

Abstract

Cantharidin, a natural vesicant, inhibits the activity of several PPP family phosphatases, displays antitumor activity, and induces apoptosis in many types of tumor cells. However, the molecular mechanisms underlying the antitumor activity of cantharidin are not clear. Here, dose-response studies confirm a strong correlation between the suppression of phosphatase activity and cell death. Flow cytometry analysis indicates that before apoptosis, cantharidin delays cell cycle progression following DNA replication with no apparent effect on G(1)-S or S-G(2) phase progression. In contrast, studies with double thymidine-synchronized populations of cells indicate that cantharidin can rapidly arrest growth when added during G(2) or early M phase. Immunostaining indicates that cell cycle arrest occurs before the completion of mitosis and is associated with the appearance of aberrant mitotic spindles. Live cell imaging with time-lapse microscopy shows that cantharidin disrupts the metaphase alignment of chromosomes and produces a prolonged mitotic arrest, with the onset of apoptosis occurring before the onset of anaphase. To explore the contribution of individual phosphatases, antisense oligonucleotides and small interfering RNA were developed to suppress the expression of cantharidin-sensitive phosphatases. The suppression of PP2Aalpha, but not PP2Abeta, is sufficient to induce metaphase arrest, during which time lagging chromosomes are observed moving between the spindle poles and the metaphase plate. Immunostaining revealed slightly abnormal, yet predominately bipolar, mitotic spindles. Nonetheless, after a 10- to 15-hour delay, the cells enter anaphase, suggesting that an additional cantharidin-sensitive phosphatase is involved in the progression from metaphase into anaphase or to prevent the onset of apoptosis in cells arrested during mitosis.

Figure 1. Cantharidin inhibits PPP-family phosphatases and cell proliferation

A) Effect of cantharidin (insert) on the divalent cation-independent phosphatase activity contained in a crude homogenate of A549 cells. Cells from a single ~70% confluent 100 mm dish were rinsed with 4°C phosphate buffered saline, scraped and sonicated in 1 ml of Tris buffer (20 mM Tris-HCL, 1mM EDTA, and 2 MM dithiothreitol, pH 7.4). The homogenate was then diluted (~65 μg protein/ml) and assayed for phosphohistone phosphatase activity in the presence of the indicated amount of cantharidin as described in methods. B) Effect of cantharidin on cell viability. A549 cells were plated in 96 well plates (5,000 cells/well) and allowed to grow. After 24 hours, the indicated concentration of cantharidin was added and the cells were incubated for the time indicated at 37°C. Cell viability was then measured using a Cell Titer 96 Aqueous One Solution Cell proliferation Assay according to the methods of the manufacturer (Promega). The data shown represents the mean +/− SD of three independent experiments each conducted in triplicate.

Figure 2. DNA content flow cytometric histograms of propidium iodide stained A549 cells

A) Cantharidin dose-response. Logarithmically growing A549 cells were treated with cantharidin at the concentrations indicated for 24 hours. Controls were treated with solvent alone. At the time points indicated, the cells were harvested and DNA content flow cytometry analysis was performed as described in methods. DNA content is shown on the X-axis as fluorescence, and populations of cells containing 2N and 4N DNA are shaded with dark gray (2N and 4N populations are indicated by arrow heads below the graph). S-phase cells are shaded with diagonal lines. The sub G-1 populations (dead and apoptotic cells) are shaded as light gray. B & C) FACS analysis of A549 cell cultures following release from double thymidine induced G1-growth arrest. Double thymidine treatments were used to arrest A549 cell cultures in late G1- phase of the cell cycle. The cells were then treated with solvent (B) or cantharidin (C) and released from growth suppression. At the times indicated, the cells were fixed and processed for DNA content FACS analysis as described above. D) FACS analysis of A549 cells treated with cantharidin at timed intervals after release from growth suppression. A549 cells were arrested with double-thymidine treatment and then released from growth suppression as above. Cantharidin was added to the culture at the times indicated. The cells were then fixed 16 hours after they were released from double thymidine growth arrest. The results shown are representative of at least three independent experiments each conducted in triplicate.

Figure 3. Abnormal spindle formation in cells treated with cantharidin

Representative A549 cells treated with solvent alone (A) or cantharidin (B-E). A549 cells were treated with 2 µM cantharidin and fixed on cover slips 24 hrs later. Microtubules (green) were then visualized by immunofluorescence following treatment with anti-α-tubulin (primary) and Alexa Fluor 488-labeled (secondary) antibodies. DNA (red) was visualized by staining with PI. E) Serial z-sections obtained by confocal microscopy of cantharidin treated cells stained for anti-α tubulin antibodies as above. F) Effect of cantharidin on spindle formation. A549 cells were treated with the indicated amount of cantharidin. After 24 hours the cells were fixed. Microtubules and DNA were then visualized as described above. Metaphase cells (300 for each concentration of cantharidin) were then scored as containing normal or aberrant spindles. The data is plotted as percentage containing normal bipolar spindles (white bar) or aberrant spindles (filled bar). For A–F similar results were obtained with ≥ 4 independent experiments.

Figure 3. Abnormal spindle formation in cells treated with cantharidin

Representative A549 cells treated with solvent alone (A) or cantharidin (B-E). A549 cells were treated with 2 µM cantharidin and fixed on cover slips 24 hrs later. Microtubules (green) were then visualized by immunofluorescence following treatment with anti-α-tubulin (primary) and Alexa Fluor 488-labeled (secondary) antibodies. DNA (red) was visualized by staining with PI. E) Serial z-sections obtained by confocal microscopy of cantharidin treated cells stained for anti-α tubulin antibodies as above. F) Effect of cantharidin on spindle formation. A549 cells were treated with the indicated amount of cantharidin. After 24 hours the cells were fixed. Microtubules and DNA were then visualized as described above. Metaphase cells (300 for each concentration of cantharidin) were then scored as containing normal or aberrant spindles. The data is plotted as percentage containing normal bipolar spindles (white bar) or aberrant spindles (filled bar). For A–F similar results were obtained with ≥ 4 independent experiments.

Figure 4. Inhibition of PP2A mRNA and protein expression by treatment with antisense oligondeoxynucleotides

A) Relative positioning of the predicted hybridization sites within the human PP2Aα or PP2Aβ mRNA of 43 antisense oligodeoxynucleotides that were evaluated for their ability to inhibit PP2A expression in cultured A549 cells. Antisense oligodeoxynucleotides that inhibit the expression of PP2Aα and PP2Aβ mRNA were identified by treating A549 cells with the indicated oligodeoxynucleotides at a concentration of 300 nM. mRNA was prepared 24 hours later, and analyzed for PP2Aα, PP2Aβ and glyceraldehyde-3-phosphodehydrogenase (G3PDH) mRNA levels by northern blot analysis. B) Inhibition of PP2Aβ mRNA levels by ISIS 110179, ISIS 110181, ISIS 110186 or ISIS 110189. C) Inhibition of PP2Aα by ISIS 110159 or ISIS 110163. D) Specificity of oligos targeting PP2Aα (ISIS 110159) and PP2Aβ (ISIS 110181). A549 cells were treated with increasing concentrations (0–300 nM) of the indicated antisense oligodeoxynucleotides or corresponding mismatched control analogues (indicated as MM) that contain 13 base changes (mismatches) within the sequence of the indicated target. Total mRNA was prepared 24 h later and analyzed for PP2Aα PP2Aβ and G3PDH mRNA levels by northern blot analysis. IC₅₀ values were estimated from plots produced by quantification of PP2A mRNA levels after normalization to G3PDH mRNA in A549 cells following treatment with increasing concentrations of oligodeoxynucleotides (illustrated for ISIS 110159). E) Western blots of PP2A protein levels in A549 cells. Cells were treated with the mismatch control oligodeoxynucleotides, ISIS 110159 (targeting PP2Aα) or ISIS 110181 (targeting PP2Aβ), and protein extracts were prepared 48 or 76 hours later. Each lane contained 40 μg protein. F) Target specific inhibition of ISIS 110159 and ISIS 110181. A549 cells were treated with 200 or 400 nM antisense oligodeoxynucleotides targeting the indicated isoform of PP2A. Total protein was prepared 48 or 76 hours later and analyzed for PP4 and PP6 by western blot analysis.

Figure 4. Inhibition of PP2A mRNA and protein expression by treatment with antisense oligondeoxynucleotides

A) Relative positioning of the predicted hybridization sites within the human PP2Aα or PP2Aβ mRNA of 43 antisense oligodeoxynucleotides that were evaluated for their ability to inhibit PP2A expression in cultured A549 cells. Antisense oligodeoxynucleotides that inhibit the expression of PP2Aα and PP2Aβ mRNA were identified by treating A549 cells with the indicated oligodeoxynucleotides at a concentration of 300 nM. mRNA was prepared 24 hours later, and analyzed for PP2Aα, PP2Aβ and glyceraldehyde-3-phosphodehydrogenase (G3PDH) mRNA levels by northern blot analysis. B) Inhibition of PP2Aβ mRNA levels by ISIS 110179, ISIS 110181, ISIS 110186 or ISIS 110189. C) Inhibition of PP2Aα by ISIS 110159 or ISIS 110163. D) Specificity of oligos targeting PP2Aα (ISIS 110159) and PP2Aβ (ISIS 110181). A549 cells were treated with increasing concentrations (0–300 nM) of the indicated antisense oligodeoxynucleotides or corresponding mismatched control analogues (indicated as MM) that contain 13 base changes (mismatches) within the sequence of the indicated target. Total mRNA was prepared 24 h later and analyzed for PP2Aα PP2Aβ and G3PDH mRNA levels by northern blot analysis. IC₅₀ values were estimated from plots produced by quantification of PP2A mRNA levels after normalization to G3PDH mRNA in A549 cells following treatment with increasing concentrations of oligodeoxynucleotides (illustrated for ISIS 110159). E) Western blots of PP2A protein levels in A549 cells. Cells were treated with the mismatch control oligodeoxynucleotides, ISIS 110159 (targeting PP2Aα) or ISIS 110181 (targeting PP2Aβ), and protein extracts were prepared 48 or 76 hours later. Each lane contained 40 μg protein. F) Target specific inhibition of ISIS 110159 and ISIS 110181. A549 cells were treated with 200 or 400 nM antisense oligodeoxynucleotides targeting the indicated isoform of PP2A. Total protein was prepared 48 or 76 hours later and analyzed for PP4 and PP6 by western blot analysis.

Figure 4. Inhibition of PP2A mRNA and protein expression by treatment with antisense oligondeoxynucleotides

A) Relative positioning of the predicted hybridization sites within the human PP2Aα or PP2Aβ mRNA of 43 antisense oligodeoxynucleotides that were evaluated for their ability to inhibit PP2A expression in cultured A549 cells. Antisense oligodeoxynucleotides that inhibit the expression of PP2Aα and PP2Aβ mRNA were identified by treating A549 cells with the indicated oligodeoxynucleotides at a concentration of 300 nM. mRNA was prepared 24 hours later, and analyzed for PP2Aα, PP2Aβ and glyceraldehyde-3-phosphodehydrogenase (G3PDH) mRNA levels by northern blot analysis. B) Inhibition of PP2Aβ mRNA levels by ISIS 110179, ISIS 110181, ISIS 110186 or ISIS 110189. C) Inhibition of PP2Aα by ISIS 110159 or ISIS 110163. D) Specificity of oligos targeting PP2Aα (ISIS 110159) and PP2Aβ (ISIS 110181). A549 cells were treated with increasing concentrations (0–300 nM) of the indicated antisense oligodeoxynucleotides or corresponding mismatched control analogues (indicated as MM) that contain 13 base changes (mismatches) within the sequence of the indicated target. Total mRNA was prepared 24 h later and analyzed for PP2Aα PP2Aβ and G3PDH mRNA levels by northern blot analysis. IC₅₀ values were estimated from plots produced by quantification of PP2A mRNA levels after normalization to G3PDH mRNA in A549 cells following treatment with increasing concentrations of oligodeoxynucleotides (illustrated for ISIS 110159). E) Western blots of PP2A protein levels in A549 cells. Cells were treated with the mismatch control oligodeoxynucleotides, ISIS 110159 (targeting PP2Aα) or ISIS 110181 (targeting PP2Aβ), and protein extracts were prepared 48 or 76 hours later. Each lane contained 40 μg protein. F) Target specific inhibition of ISIS 110159 and ISIS 110181. A549 cells were treated with 200 or 400 nM antisense oligodeoxynucleotides targeting the indicated isoform of PP2A. Total protein was prepared 48 or 76 hours later and analyzed for PP4 and PP6 by western blot analysis.

Figure 5. Mitotic abnormalities associated with the suppression of PP2Aα

Representative confocal microscopic images of A549 cells treated with ISIS 110159. A549 cells were treated with 300 nM ISIS 110159 and fixed on cover slips 24 hrs later. A) Microtubules (green) were then visualized by immunofluorescence following treatment with anti-α-tubulin and Alexa Fluor 488-labeled antibodies as described in Figure 3. B) Centrosomes were visualized by treatment with pericentrin. C) DNA (blue) was visualized by staining with Draq5. D) Computer assisted merge of images A, B and C. E) Representative images of serial z-sections of ISIS 110519 treated mitotic cells stained for anti-α tubulin. Similar results were obtained with ≥ 5 independent experiments.

Figure 6. Localization of H2B-GFP protein in HeLa cells

Representative fluorescent and corresponding differential interference contrast (DIC) microscopic images of live HeLa cells expressing H2B-GFP in various phases of the cell cycle. A) Representative images selected from time-lapse studies of control cells illustrating prophase (0 min), early metaphase (20 min), late metaphase (70 min), anaphase (80 min) and telophase (86 min). B) Cantharidin treatment of prophase cells prevents the alignment of chromosomes at the metaphase plate. H2B-GFP expressing HeLa cells were treated at time 0 with 4 μM cantharidin and time-lapse live cell imaging studies were conducted for 24 hours. Representative images taken at the times indicated are shown. C) Cantharidin treatment of metaphase cells delays anaphase and disrupts the metaphase alignment of condensed chromosomes. H2B-GFP-expressing HeLa cells were treated with cantharidin and time-lapse live cell imaging was conducted for 24 hours. Representative images taken at the times indicated are shown. D) Representative fluorescent and DIC images from time-lapse studies monitoring cell cycle progression and H2B-GFP-expression in cells treated with ISIS 110159. H2B-GFP expressing HeLa cells were treated with ISIS 110159 and 30 hours later time-lapse imaging studies were conducted as described above. E) Enlarged images highlighting differences in control, cantharidin and ISIS 110159 treated cells. Arrows indicate the fluorescence of lagging sister chromatids.

Figure 6. Localization of H2B-GFP protein in HeLa cells

Representative fluorescent and corresponding differential interference contrast (DIC) microscopic images of live HeLa cells expressing H2B-GFP in various phases of the cell cycle. A) Representative images selected from time-lapse studies of control cells illustrating prophase (0 min), early metaphase (20 min), late metaphase (70 min), anaphase (80 min) and telophase (86 min). B) Cantharidin treatment of prophase cells prevents the alignment of chromosomes at the metaphase plate. H2B-GFP expressing HeLa cells were treated at time 0 with 4 μM cantharidin and time-lapse live cell imaging studies were conducted for 24 hours. Representative images taken at the times indicated are shown. C) Cantharidin treatment of metaphase cells delays anaphase and disrupts the metaphase alignment of condensed chromosomes. H2B-GFP-expressing HeLa cells were treated with cantharidin and time-lapse live cell imaging was conducted for 24 hours. Representative images taken at the times indicated are shown. D) Representative fluorescent and DIC images from time-lapse studies monitoring cell cycle progression and H2B-GFP-expression in cells treated with ISIS 110159. H2B-GFP expressing HeLa cells were treated with ISIS 110159 and 30 hours later time-lapse imaging studies were conducted as described above. E) Enlarged images highlighting differences in control, cantharidin and ISIS 110159 treated cells. Arrows indicate the fluorescence of lagging sister chromatids.