Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Stathmin/oncoprotein 18, a microtubule regulatory protein, is required for survival of both normal and cancer cell lines lacking the tumor suppressor, p53

Stathmin/oncoprotein 18, a microtubule regulatory protein, is required for survival of both... ReseaRCh papeR ReseaRCh p apeR Cancer Biology & Therapy 9:9, 699-709; May 1, 2010; © 2010 Landes Bioscience Stathmin/oncoprotein 18, a microtubule regulatory protein, is required for survival of both normal and cancer cell lines lacking the tumor suppressor, p53 Bruce K. Carney and Lynne Cassimeris* Department of Biological s ciences; Lehigh University; Bethlehem, pa Usa Key words: stathmin, p53, apoptosis, cell cycle, microtubule stathmin, a microtubule regulatory protein, is overexpressed in many cancers and required for survival of several cancer lines. In a study of breast cancer cell lines, proposed that stathmin is required for survival of cells lacking p53, but this hypothesis was not tested directly. h ere we tested their hypothesis by examining cell survival in cells depleted of stathmin, p53 or both proteins. Comparing h CT116 colon cancer cell lines differing in T p53 genotype, stathmin depletion resulted in significant death only in cells lacking p53. a s a second experimental system, we compared the effects of stathmin depletion from h eLa cells, which normally lack detectable levels of p53 due to expression of the hp V e6 protein. s tathmin depletion caused a large percentage of h eLa cells to die. Restoring p53, by depletion of hpV e6, p53-/- rescued h eLa cells from stathmin-depletion induced death. Cleaved pa Rp was detected in h CT116 cells depleted of stathmin and cell death in stathmin-depleted h eLa cells was blocked by the caspase inhibitor Z-Va D-FMK, consistent with apoptotic death. The stathmin-dependent survival of cells lacking p53 was not confined to cancerous cells because both proteins were required for survival of normal human fibroblasts. In h CT116 and h eLa cells, depletion of both stathmin and p53 leads to a cell cycle delay through G . Our results demonstrate that stathmin is required for cell survival in cells lacking p53, suggesting that stathmin depletion could be used therapeutically to induce apoptosis in tumors without functional p53. generally poor prognosis, supporting the use of stathmin expres- Introduction 7,14 sion as a biomarker for cancer stage progression. Microtubules are dynamic polymers of α/β tubulin dimer sub- While stathmin level may serve as a biomarker in many cancers, units that contribute to a number of cell processes including several studies have demonstrated that stathmin depletion may chromosome movement and mitosis. Several successful chemo- also have therapeutic value; in a number of cancer cell lines stath- therapies, including paclitaxel and vinblastine, are thought to act min depletion results in cell cycle arrest and apoptosis. Mistry et by targeting microtubules or their tubulin subunits, disrupting al. found that expression of anti-stathmin ribozymes depleted mitotic spindle function, activating the spindle assembly check- stathmin and resulted in apoptosis in the androgen-independent 1-3 point and inducing cell death. While microtubules have been prostate cancer cell line LNCaP. Depletion of stathmin by siR NA 1 21 22 a successful target for chemotherapy development, accessory or shRNA in HeLa cells, osteosarcoma cell lines and several proteins that regulate microtubule assembly and disassembly are breast cancer cell lines also demonstrated reduced cell prolifera- 4,5 also targets for novel chemotherapeutics. Of these regulatory tion and apoptosis after stathmin reduction. In contrast to these proteins, stathmin/oncoprotein 18, a microtubule destabilizing cancer cell lines, stathmin knockout mice are viable. These protein, has received attention recently as a potential target for results have raised interest in stathmin reduction as a potential 6-8 cancer therapy. cancer therapy, possibly active only in cancerous cells. Stathmin is overexpressed in many cancers including It is currently not known why even partial stathmin depletion 9 10 6 11 leukemia, lymphoma, oral squamous-cell, adenoid cystic, is sufficient to induce apoptosis in several cancer cell lines. Most 12 13 14,15 lung and hepatocellular carcinomas, adenocarcinoma, pros- studies have indicated that stathmin depletion results in a G /M 7 16 17 18,19 7,21,25 tate cancer, sarcomas, glioma, breast cancer and ovarian block, but normally stathmin is inactivated by phosphory- 20 26-28 cancer. Cancers where stathmin is overexpressed have been lation at entry into mitosis and does not have a detectable shown to have a high malignancy phenotype, which leads to microtubule destabilizing activity during mitosis. These latter *Correspondence to: Lynne Cassimeris; Email: [email protected] Submitted: 09/03/09; Revised: 01/08/10; Accepted: 02/08/10 Previously published online: www.landesbioscience.com/journals/cbt/article/11430 www.landesbioscience.com Cancer Biology & Therapy 699 p53W T p53-/- Figure 1. s tathmin depletion from h CT116 and h CT116 cell lines. Matched lines of h CT116 cells differing in p53 genotype were treated with p53-/- siRNa targeting stathmin. ( a) s tathmin depletion from h CT116 cells was clearly detectable 24 hrs after transfection and stayed low for at least 5 d W Tp53 post transfection. (B) siRNa depletion of stathmin from h CT116 cells. The sample shown at 5 days included a second siRNa transfection, 72 h after the initial siRNa transfection, which was necessary to keep stathmin level low. s amples at 1 and 3 d are from a single siRNa transfection. In both ( a and B), the lysates from non-targeting siRNa treated cells were loaded in a dilution series to estimate the extent of depletion by stathmin siRN a . Blots were re-probed for actin as an indicator of equal loading between control (100% load) and stathmin-depleted samples. (C) Low magnification phase contrast images of living cells acquired at the times given after transfection with control or stathmin siRNa . The same field is shown within an experi - p53-/- mental condition over 4–5 d. h CT116 cells depleted of stathmin remain relatively sparse over the time course while cells grow to near confluency in the other experimental conditions. Data shown are representative of three separate experiments. data strongly suggest that stathmin depletion does not lead to Results apoptosis by inducing a mitotic block. An alternative mecha- nism was suggested by Alli et al. who noted that those cell lines Depletion of stathmin from HCT116 colon cancer cell lines. requiring stathmin for survival also expressed mutant p53. This To test directly whether p53 is required for survival of cells intriguing hypothesis was not tested directly because Alli et al. depleted of stathmin, we used HCT116 colon cancer cell lines p53-/- were unable to deplete stathmin from cell lines expressing wild- differing in TP53 genotype. The HCT116 line was devel- type p53. Further complicating analysis, some of the breast cancer oped by knocking out both copies of the TP53 gene from the 23 W Tp53 32 cell lines used by Alli et al. have additional mutations targeting parental HCT116 line (HCT116 ; ). We first examined 30,31 other tumor suppressor genes, including Rb and PTEN. the extent and time course of stathmin depletion after siRNA Here we used several experimental systems to test whether transfection, compared to cells transfected with siGlo, a non- p53-/- stathmin is specifically required for survival of those cells lacking targeting control siRNA (Dharmacon). For HCT116 cells, p53. First, we compared cell proliferation and death after stath- stathmin was depleted by approximately 75% within 24 h after min depletion from matched colon cancer cell lines (HCT116) transfection and remained low for 5 d (Fig. 1A). We found that W Tp53 differing in p53 genotype. As an alternate approach, we restored stathmin depletion was more difficult in HCT116 , consis- p53 to HeLa cells by depletion of the human papillomavirus tent with a previous study of breast cancer cell lines. By 72 h (HPV) E6 protein, which normally binds p53 and marks it for after siRNA transfection, stathmin protein level was reduced by W Tp53 destruction. Finally, we depleted stathmin and/or p53 from nor- approximately 50% in HCT116 (Fig. 1B). Stathmin level mal human foreskin fibroblasts (HFF cells) to address whether often returned at later time points (not shown), but a second stathmin is specifically required for cancer cell survival. We siRNA transfection at 72 h kept the stathmin level low at day found that stathmin is required for the survival of both cancerous 5 (time after the first transfection, Fig. 1B ) and up to 7 d after and non-cancerous derived cell lines, but only in the absence of the initial transfection (not shown). These results demonstrate p53. In cells lacking both stathmin and p53, cells arrest or delay that stathmin can be successfully knocked down in HCT116 at G of the cell cycle. matched cell lines. 700 Cancer Biology & Therapy Volume 9 Issue 9 p53-/- Figure 2. s tathmin is required for surival of h CT116 cells. h CT116 cell lines were treated with non-targeting control or stathmin RNa i. a t the indicated time points, cells were trypsinized, incubated in Trypan Blue to identify dead cells, and numbers of living and dead cells counted. Cell proliferation rates (a) and the percentage of dead cells (B) represent the means of three independent experiments ± s D. Stathmin is required for cell survival in HCT116 cells each HCT116 cell line at the 5 d time point. The difference in p53-/- lacking p53. The HCT116 cell lines were depleted of stathmin the mean percentage of dead cells in HCT116 control and using the conditions described above and were observed for up stathmin-depleted cells was highly significant (p < 0.001), while W Tp53 to 5 d post-siRNA transfection to observe the effect of stath- the means were not significantly different in HCT116 cells W Tp53 min depletion on cell proliferation and death. First, we followed and HCT116 cells depleted of stathmin. Taken together, living cells by imaging the same regions of coverslips over a 4–5 these results provide strong evidence that stathmin is required W Tp53 d time course. For HCT116 cells, either expressing stathmin for cell survival only in cells lacking p53, but the differences in or depleted of stathmin, cell density increased over time, reaching time course of stathmin knockdown in cells differing in TP53 near conu fl ency by day 4–5 ( Fig. 1C). The same growth charac- genotype complicates analysis. p53-/- teristics were observed for the HCT116 cell line transfected Restoration of p53 in HeLa cells rescues cells from stath- with a control siRNA. In contrast, depletion of stathmin from min-depletion induced death. To further test whether stathmin p53-/- the HCT116 line significantly reduced the increase in cell is required for survival of cells lacking p53, we used HeLa cells, number over time (Fig. 1C). which are wildtype for the TP53 gene, but fail to maintain p53 Following living cells over time suggested that stathmin protein level because they stably express the human papilloma- depletion slowed cell proliferation and/or led to cell death only virus (HPV) E6 protein. The HPV E6 protein binds p53 and p53-/- 33 in the HCT116 cell line. To examine cell growth and death marks it for destruction. The ability to deplete the HPV E6 more directly, we measured both cell proliferation, by count- protein by RNAi provides a mechanism to restore p53 in HeLa ing the number of living cells, and cell death, by counting the cells, allowing comparison of cells depleted of stathmin in the percentage of trypan blue positive cells over a 5 d time course. presence and absence of p53. HCT116 cells expressing wildtype p53 and depleted of stathmin We first confirmed the results of Koivusalo et al. that E6 proliferated at a slower rate than cells transfected with a control depletion restores p53 in HeLa cells. Where noted, cells were RNA (solid lines, Fig. 2A). Although the stathmin depleted treated with the DNA-damaging agent doxorubicin for the W Tp53 HCT116 cells grew more slowly, they remained viable, as 24 h prior to gel sample preparation to stabilize p53 and enhance measured by trypan blue exclusion (Fig. 2B). Stathmin depletion its detection. As shown in Figure 3A, HeLa cells have a very low p53-/- from HCT116 cells showed very slow proliferation over a 5 d level of p53, even after doxorubicin treatment. Depletion of E6 time course (Figs. 2A, see also 1C) and these cells died at a much by siRNA results in a significant increase in p53 level, which W Tp53 greater rate than HCT116 cells depleted of stathmin, or is also detectable in cells depleted of both stathmin and E6 either of the HCT116 cell lines transfected with a control RNA (Fig. 3A). (Fig. 2B). There was a high variability in the percentage of dead Stathmin was also readily depleted from HeLa cells, as shown p53-/- cells in HCT116 cells depleted of stathmin, as seen by the in Figure 3B. At 72 h after siRNA transfection, stathmin was large standard deviations. At this time we do not know the reason reduced by at least 75%, independent of whether cells were also for the inter-experiment variability. To confirm that the differ - depleted of E6 (p53 restored). Stathmin level was also reduced, ence in means was significant, we compared the mean percent - although to a lesser extent, in cells depleted only of E6. In this ages of dead cells for control and stathmin-depleted cells within latter case, reduced stathmin level may reflect the ability of p53 www.landesbioscience.com Cancer Biology & Therapy 701 Figure 3. Restoring p53 in h eLa cells. (a) h eLa cells were treated with control or siRNa targeting hp V e6 for 72 h and exposed to 0.5 µM doxorubicin for the final 24 h to induce DN a damage and stabilize p53, facilitating its detection by immunoblot. In control-treated h eLa cells, p53 is present at a very low level, even in the presence of doxorubicin. Depletion of hp V e6 restored p53. (B) h eLa cells were treated with siGlo control siRNa (or empty vector controls) or siRNa s targeting stathmin, hp V e6 or both mRNa s. a nti-stathmin immunoblot of cell lysates were collected 72 h after siRNa transfection. s tathmin was depleted by at least 75% after treatment with siRNa targeting stathmin or stathmin and hp V e6. Depletion of hp V e6 also 34,35 caused a reduction in stathmin level, possibly by p53-mediated repression of stathmin expression. For both (a and B), blots were re-probed for actin as a marker of equal loading between control (100%) and siRNa samples. (C) Low magnification phase contrast images of living cells acquired at the times given after transfection with control or siRNa s targeting stathmin, hp V e6 or both mRNa s. The same field is shown for each experimental condition over 4–5 d. h eLa cells depleted of e6 or stathmin and e6 grow slightly slower than control siRNa treated cells, while stathmin depleted cells show minimal growth. Data shown are representative of three separate experiments. to negatively regulate stathmin expression when p53 level is showed proliferation rates and percentages of dead cells compara- 34,35 transiently increased. ble to control-treated or HPV E6 siRNA-treated cells. Finally, we We next examined cell density over time, similar to the confirmed the siRNA results by transfecting cells with a plasmid experiments performed with HCT116 cell lines. Stathmin deple- encoding an shRNA directed against a different sequence in the tion from HeLa cells reduced the increase in cell number over stathmin mRNA (see methods). As shown in Figure 4C for cells time observed in control-siR NA treated cells (Fig. 3C). Restoring examined 72 h after transfection, shRNA depletion of stathmin p53 (E6 depletion), in the absence or presence of stathmin deple- significantly increased the percentage of dead cells, compared to tion, resulted in slower cell growth compared to control cells, but cells transfected with a control plasmid. Together with the results not nearly as significant an inhibition as that observed in stath - from HCT116 cell lines, these results provide strong evidence min depleted HeLa cells (Fig. 3C). We then measured both cell that stathmin is required for cell survival only in the absence of proliferation (live cell number over time) and the percent dead p53. cells (trypan blue positive) for these experimental conditions. In the above experiments, we used trypan blue exclusion as Figure 4 shows that stathmin depletion from HeLa cells (lack- a simple assay to differientiate living and dead cells. It is likely ing p53) resulted in both reduced cell proliferation and a large that stathmin depletion results in death by apoptosis, as others 7,21,23,25 increase in the percentage of dead cells. Comparison of mean have demonstrated previously. To confirm that cells die via percentages of dead cells at day 5 showed that stathmin depletion apoptosis we examined whether stathmin-depletion induced cell signicantly increased the percentage of dead cells over that mea- death was inhibited by the capsase inhibitor Z-VAD-FMK. As sured in control siRNA treated cells (p < 0.001). Treatment of expected, Z-VAD-FMK significantly blocked the death of HeLa cells with siRNA to deplete both stathmin and HPV E6 (restor- cells depleted of stathmin (Fig. 5A and B). Stathmin depleted ing p53) reversed the effects of stathmin depletion alone; cells cells treated with Z-VAD-FMK also showed cell proliferation 702 Cancer Biology & Therapy Volume 9 Issue 9 Figure 4. Restoring p53 rescues h eLa cells from stathmin-depletion induced death. Cell proliferation rates and the percentage of dead cells were mea- sured as described in Figure 2. (a and B) s tathmin depletion slows cell proliferation and increases the percentage of dead (trypan blue positive) cells. Depletion of hp V e6 (restoring p53) alone or in combination with stathmin depletion does not lead to significant cell death. (C) Transfection of cells with an shRNa targeting stathmin resulted in stathmin depletion (top) and a significant increase in the percentage of dead cells at 72 h post- transfection. Controls were transfected with the negative control plasmid provided by the manufacturer. *denotes p < 0.05. a ll data represent the means of three independent experiments ± s D. rates intermediate between control-treated and stathmin-siRNA achieved using RNAi and shRNA respectively (Fig. 6A and B). treated cells. In additional experiments, we found increased lev- A 50% knockdown of stathmin was seen as early as day 3 and els of cleaved poly-ADP ribose polymerase (PARP) in stathmin- continued to approximately 75% knockdown by day 5. By day p53-/- depleted HCT116 cells (Fig. 5C and D), but not in stath- 2, p53 showed a significant knockdown as well. Note that the W Tp53 min-depleted HCT116 cells. These results confirm previous only cells treated with doxorubucin were those used to prepare 7,21,23,25 results of others that stathmin depletion initiates apop- gel samples for p53 detection (as noted in Fig. 6B). Depletion of totic death in cells lacking p53. stathmin or p53 alone did not inhibit cell proliferation rate and Depletion of p53 and stathmin from non-cancerous cells did not increase cell death compared to cells transfected with leads to cell death. The experiments above, along with previ- control siRNA or the empty vector used for shRNA delivery (Fig. 7,21,23 ous results of others, demonstrated that stathmin depletion 6C and D). Interestingly, however, depleting both stathmin and from cancer cell lines results in cell death by apoptosis. We next p53 together resulted in significant cell death (p < 0.001 at day 5) asked whether non-cancerous cells also require stathmin for sur- and a lack of cell proliferation (Fig. 6C and D). These data dem- vival using human foreskin fibroblasts (HFF) as a model non- onstrate that stathmin is required for cell survival in both human cancerous cell line. To deplete p53 from these cells, we trans- cancerous and non-cancerous cell lines lacking p53. fected cells with a plasmid for expression of an shRNA targeting Cells arrest in G of the cell cycle prior to apoptosis. Both 36 34 37 p53. A successful knockdown of both stathmin and p53 was the overexpression of wildtype p53, and the absence of p53, www.landesbioscience.com Cancer Biology & Therapy 703 Figure 5. s tathmin-depletion induced cell death likely occurs via apoptosis. (a) The caspase inhibitor Z-V a D-FMK blocks stathmin-depletion induced cell death in h eLa cells. h eLa cells were treated with 10 µm Z-Va D-FMK for 24 h prior to transfection with either siGlo control siRNa (or empty vector controls) or RNa i against stathmin. plots of cell proliferation rates (a) and the percentage of dead cells (B) represent the means of three independent experiments ± sD. (C) h CT116 matched cell lines were transfected with control or stathmin RNa i, fixed 72 h later and stained for cleaved pa Rp. The p53-/- percentage of cells positive for cleaved pa Rp are plotted. Only h CT116 cells depleted of stathmin show a signficant increase in cells positive for cleaved pa Rp. Data shown are the means ± sD of three independent experiments and a total of ∼ 200 cells per treatment. (D) Representative image p53-/- from h CT116 cells depleted of stathmin and staining positive for cleaved pa Rp in the nucleus (arrow). s everal weakly stained cells (counted as negative) are also included. s cale bar = 10 µm. have been shown to cause cell cycle arrest, at least under some phosphosphorylated on Tyr15 (Phospho-CDK1(Y15); inhibi- experimental conditions. To address whether stathmin deple- tory phosphorylation on CDK1 present during G and removed tion leads to a unique cell cycle arrest that is dependent upon at entry into mitosis ). Cells were co-stained with antibodies p53 status, we examined cell cycle distributions in HCT116 to tubulin (Figs. 7 and 8), allowing visual classification of cells matched cells depleted of stathmin and HeLa cells depleted in M phase. Using these staining conditions and counting all p53-/- of stathmin, HPV E6 (restoring p53) or both proteins. Others cells within multiple images, we found that HCT116 cells have previously followed DNA levels and identified a G /M depleted of stathmin showed a significant (p < 0.05) increase in cell cycle arrest after stathmin depletion in cancer-derived cell non-mitotic cells staining positively for TPX 2 (Fig. 7A and B) or 7,23 lines, but these studies measured DNA content and cannot Cyclin B (Fig. 7C and D) 48 h post transfection. This result was differentiate between G and M phases. Therefore, we used an most pronounced in TPX2 stained samples. The mitotic index alternative protocol, by staining fixed cells with antibodies to was approximately equal among the two HCT116 lines whether TPX2 (present in the nuclei of S and G cells and bound to stathmin was depleted or not. the mitotic spindle microtubules in M phase ), Cyclin B (pres- We then extended these staining protocols to HeLa cells. ent in the cytoplasm of G cells and associated with the mitotic Depletion of stathmin from HeLa cells also increased the per- spindle until anaphase onset during M phase ), and CDK1 centage of non-mitotic TPX2 positive cells (Fig. 8A and B) and 704 Cancer Biology & Therapy Volume 9 Issue 9 Figure 6. Depleting both p53 and stathmin from normal human fibroblast ( h FFs) cells leads to cell death. h FF cells were transfected with Transs ilent empty vector, p53 shRNa plasmid and/or siRN a targeting stathmin as indicated. ( a) a nti-stathmin immunoblot from cell lysates were isolated 3 d after transfection. a knockdown of >75% was observed for both stathmin siRN a alone or when used in combination with shRN a against p53. (B) a nti-p53 immunoblot from cell lysates isolated at the times indicated. When present, doxorubucin was included for the final 24 h of incubation. The level of p53 is detectably reduced 1 d after shRNa transfection, but is more significantly depleted by 2–3 d after transfection. (C and D) p lots of h FF cell proliferation rates (C) and the percentage of dead cells (D) represent the means of three independent experiments ± s D. concomitantly significantly decreased the percentage of cells not prior to entry into mitosis. Depletion of either stathmin or p53 recognized as TPX2 positive. Untreated HeLa cells, cells treated alone is not sufficient to delay progression through the cell cycle. with a non-targeting siRNA or depleted or E6 to restore p53 showed a much smaller percentage of non-mitotic cells stain- Discussion ing positive for TPX2. As a second marker for cells in G , we also stained HeLa cells with antibodies recognizing phospho- In the results presented here we demonstrated that stathmin is CDK1(Y15), one of the two inhibitory phosphorylations hold- required for the survival of cell lines lacking p53, whether those ing CDK1 inactive until entry into M phase. Consistent with lines are derived from cancerous or normal tissues. Stathmin p53-/- results from TPX2 staining, HeLa cells depleted of stathmin depletion from HCT116 and HeLa cells showed significantly showed a significant increase (p < 0.05) in non-mitotic cells reduced proliferation rates and an increased percentage of dead stained positive for phospho-CDK1(Y15) (Fig. 8C and D). Cells cells, while depletion of stathmin had little effect on the viabil- W Tp53 depleted of E6, or of E6 and stathmin, did not show significant ity of HCT116 cells or of HeLa cells expressing p53. These changes in the percentage of non-mitotic, phospho-CDK1(Y15) observations were not confined to cancer-derived cell lines since positive cells. We confirmed these results by transfecting HeLa normal human fibroblasts (HFFs) also required stathmin for pro - cells with a control plasmid or a plasmid expressing an shRNA liferation and survival, but only in the absence of p53. For both W Tp53 directed against stathmin and again found an increase in non- HCT116 and HPV E6-depleted HeLa cells, stathmin deple- mitotic phospho-CDK1(Y15) positive cells in stathmin-depleted tion slowed cell proliferation somewhat, but did not result in cell cells (Fig. 8E). From these data, it is evident that stathmin deple- death and did not impose a cell cycle block. Our results provide tion disrupts cell cycle progression, leading to an increase in cells direct support for the model proposed by Alli et al. demonstrat- positive for G markers, but only in the absence of p53. Given ing that stathmin is required for survival in cells lacking p53. that others have reported a G /M block in stathmin depleted Several groups have previously reported slowing of cell prolif- cells, based on analysis of DNA content, and our results indicate eration and cell death after stathmin depletion, but these stud- a G block, it most likely that cells are blocked or delayed in G , ies did not address a stathmin requirement for cell survival in 2 2 www.landesbioscience.com Cancer Biology & Therapy 705 p53-/- Figure 7. s tathmin depletion from h CT116 cells delays G of the cell cycle. Matched h CT116 cell lines were treated with either siGlo control siRNa or siRNa targeting stathmin mRN a s. (a and B) h CT116 cells were fixed 48 h after siRN a transfection and stained with antibodies against α -tubulin (green in merged images) and TpX2 (red in merged images). TpX2 is present in s /G and M phases of the cell cycle. The numbers of TpX2 positive, negative and mitotic cells were counted, and the percentage of cells in each group are shown in (a). Images shown in (B) are from a representative field of cells transfected with stathmin siRNa . Negative cells were identified by their array of interphase microtubules and their lack of T pX2 staining. Cells in inter- phase staining positively for TpX2 were scored as positive. Mitotic cells were identified by the presence of a mitotic spindle (tubulin staining). s cale bar = 25 µm. (C and D) h CT116 cell lines were fixed 48 h after siRN a transfections and stained with antibodies against α -tubulin (green in merged images) and cyclin B (red in merged images). Cytoplasmic cyclin B is a marker of G cells. The percentage of non-mitotic cyclin B positive and negative cells are shown in (C). Images shown in (D) are from representative cells fixed 48 h after stathmin siRN a transfection. s cale bar = 25 µm. The data represent the means of three (C) or four (a) independent experiments, each including ∼ 1,000 cells ± sD for each treatment group. *denotes p < 0.05. non-cancerous cells. Zhang et al. developed a method for expres- proliferation and increased cell death contrasts with the reported sion of stathmin shRNA driven by the survivin promoter, limit- viability of mice with both copies of the stathmin and p53 genes ing shRNA expression to those cells expressing survivin. Since knocked out. It is possible that these mice compensate through normal differentiated cells do not express survivin, it is unclear increased or decreased expression of other proteins allowing nor- whether the survivin promoter would be active, and thus express- mal development and survival in these mice. ing stathmin shRNA, in the endothelial cells used as controls. Stathmin and p53 are required for cell cycle progression. To Alli et al. were unable to deplete stathmin from breast cancer address why stathmin is required for cell survival only in cells lines expressing p53, making it unclear whether stathmin was lacking p53, we examined cell cycle distributions in HCT116 cell required for survival of these cells. Wang et al. and Mistry et lines depleted of stathmin and in HeLa cells expressing stathmin al. examined stathmin depletion in cancer cell lines and did not and p53, either protein alone, or depleted of both proteins. We test stathmin’s requirement for survival of non-cancerous cells. found that cells depleted of both stathmin and p53 show a delay in Our finding that stathmin and p53 depletions from nor - G (based on staining for TPX2, cyclin B, phospho-CDK1(Y15) mal human fibroblasts results in significantly reduced cell and tubulin) while others have reported a G /M delay in cancer 706 Cancer Biology & Therapy Volume 9 Issue 9 Figure 8. s tathmin depletion from h eLa cells delays G of the cell cycle. h eLa cells were transfected with either siGlo control siRNa or siRN a targeting stathmin, hp V e6 or both mRNa s. (a and B) h eLa cells were fixed and stained with antibodies against α -tubulin (green in merged images) and TpX2 (red in merged images). Cells were then counted as described in Figure 7a , and the percent cells in each group are shown in (a). Images shown (B) are from representative cells fixed 48 h after siRN a transfections. a rrows in the merged image denote several TpX2 positive cells. s cale bar = 50 µm. (C and D) h eLa cells were fixed and stained with antibodies against α -tubulin (red in merged images) and phospho-CDK1(Y15) (green in merged images; this inhibitory phosphorylation must be removed for entry into M phase). The percentage of non-mitotic cells staining positive for phospho-CDK1(Y15) is shown in (C). Representative images from phospho-CDK1(Y15) stained cells are shown in (D). a rrows in the merged image denote phospho-CDK1(Y15) positive cells. s cale bar = 50 µm. (e) Depletion of stathmin by shRNa also increased the percentage of phospho-CDK1(Y15) positive non-mitotic cells at 48 h after transfection. each plot represents the mean of three independent experiments ± sD including ∼1,000 cells per treatment. *denotes p < 0.05. 7,21,23,25 cells depleted of stathmin. As discussed above, we think mitosis and is not required for spindle assembly, it is unlikely that it is likely that cells depleted of stathmin and p53 are blocked stathmin depletion would prevent entry into mitosis by disrupt- in G , prior to entry into mitosis. It is interesting that induced ing microtubule assembly and turnover. expression of p53 both represses stathmin expression and induces Activation of p53, typically in response to DNA damage, a G /M block. Thus, it is likely that the levels of stathmin and primarily induces a G block, but a G block is also possible 2 1 2 p53, rather than simply their presence or absence, contribute to under some conditions. It is intriguing that many proteins G /M cell cycle progression. functioning in the G /M checkpoint, including p53, are associ- 2 2 It is not yet clear why both stathmin and p53 are required ated with the centrosome and we recently found that stathmin for cell cycle progression through G or for entry into M phase. regulates microtubule nucleation from the centrosome during Stathmin is phosphorylated during mitosis, resulting in loss of interphase. These data suggest that depletion of both stath- 28,41 its microtubule destabilizing activity. Stathmin inactivation is min and p53 may influence either centrosome function or the 28,41 required for proper assembly of the mitotic spindle and stath- function(s) of G /M checkpoint proteins at the centrosome min depletion does not impact microtubule formation in the to cause a cell cycle delay during G . An alternative model is spindle. Because stathmin is normally turned off at entry into also possible, where loss of p53 and stathmin impact separate www.landesbioscience.com Cancer Biology & Therapy 707 pathways that together function synergistically to slow cell Cell growth and death measurements. Cells were grown cycle progression. for 1–5 d post-transfection with siRNA or shRNA, trypsinized, A role for p53, or loss of p53, in regulating G /M progression stained with Trypan Blue (0.4%) and counted using a hemo- was also identified recently by comparing transcriptome differ - cytometer. Cell viability was assessed by Trypan Blue exclusion. ences between matched cell lines differing in p53 status, includ- Live and dead cells were counted, averaged and pooled for three ing the HCT116 colon cancer cells used here. Cells lacking p53 separate experiments for each cell line and treatment. Data shown upregulated expression of genes functioning in G /M and were are means ± standard deviations. sensitive to treatment with a Plk (polo-like kinase) inhibitor (Plk Groups of cells were also followed over time by plating cells on functions in G –M progression). In terms of cell proliferation, grid-etched coverslips attached to the bottoms of 35 mm dishes stathmin depletion acts similarly to Plk inhibition, since each (MatTek Corporation). Cells were plated and incubated for treatment slows proliferation of cells lacking p53. Stathmin is 24 h before transfection of siRNAs. Cells were allowed to grow one target of Plk, but this phosphorylation inhibits stathmin’s for an additional 24 h and then observed using a 20 X objective on microtubule regulatory activity. Thus inhibiting Plk will keep an inverted microscope (Nikon TE2000E) equipped with phase stathmin in an active form, which could contribute to a G /M contrast optics and MetaVu image acquisition software. Ten grid 26,28 block by preventing proper spindle assembly. It is not clear squares were chosen randomly and images were acquired from why stathmin depletion acts similarly to Plk inhibition to slow these squares each day for up to 5 d. Three separate imaging time G /M progression; possibly a balance of stathmin and p53 func- course experiments were performed and representative images are tions are necessary to pass through G . shown in Figures 1 and 3. Statistical analysis of cell counts, including those determined Methods after immunofluorescent staining (below), were performed using Paired t-tests in Microsoft Excel or GraphPad Software Cell culture. Cells were grown at 37°C in a humidified atmo - (www.graphpad.com/quickcalcs/ttest1.cfm). sphere of 5% CO . HCT116 and HFF cells were grown in Indirect immunou fl orescence and confocal microscopy. Cells DMEM (GIBCO) supplemented with 3.7 g/L sodium bicar- were fixed and imaged as described previously. Primary antibod- bonate, x1 antibiotic/antimycotic (Sigma), 1% sodium pyru- ies used were mouse monoclonal α-tubulin (B512 ; Sigma-A ldrich), vate, and 10% fetal bovine serum (FBS) (GIBCO-Invitrogen). rabbit anti-TPX2, (gift from Duane Compton, Dartmouth HeLa cells were grown in MEM (GIBCO) supplemented with Medical School), rabbit anti-cyclin B (Sigma-Aldrich), rabbit 2.2 g/L sodium bicarbonate, x1 antibiotic/antimycotic and anti-phospho-CDK1 (Tyr 15) (Cell Signaling Technology) and 10% FBS. mouse anti-cleaved PARP (ASP214) (Cell Signaling Technology). Drugs. Doxorubicin was added to cells to induce DNA dam- Goat anti-mouse or rabbit Alexa Fluor 488 or 563 (Invitrogen) age and stabilize p53, facilitating its detection by western blot. were used as the secondary antibodies in these experiments. When used, doxorubicin (0.5 µM) was added to cells for the Confocal microscopy was used to image stained cells as described last 24 h of an experimental treatment. Doxorubicin was never previously. Images were acquired using a 40 X/1.3NA objective. used in experiments to measure cell proliferation or cell death. Image stacks were converted to maximum intensity projections, Some cells were treated with the caspase inhibitor, Z-VAD-FMK exported as TIFF files and assembled using Photoshop. (10 µm; Sigma-Aldrich) or DMSO as a vehicle control. Z-VAD- Protein isolation and western blotting. Soluble cell extracts FMK was added to cells 24 h prior to any other treatment and were prepared for SDS-polyacrylamide gel electrophoresis as remained for the duration of the experiment. described previously. Protein concentrations were measured by RNA interference and shRNA transfections. RNA interfer- Bradford assay. Membranes were probed and imaged as previ- ence (RNAi) was achieved using GeneSilencer reagents following ously described using primary antibodies mouse anti-p53 (Vision the manufacturer’s protocol. Cells were grown on 35 mm dishes Bio Systems), rabbit anti-stathmin (Sigma-Aldrich), or rabbit anti- for 1–2 d before the addition of siRNA. Cells were serum starved actin (Sigma-Aldrich) followed by goat anti-mouse or rabbit horse- 30 min pre-transfection and 4 h post-transfection to improve radish peroxidase-linked IgG (Sigma-Aldrich). Protein depletions transfection efc fi iency. RNAi oligonucleotides (Dharmacon) used were estimated by comparison of western blot signals to those gen- included: SMTN1 (Op18-443), 5'-CGU UUG CGA GAG AAG erated by serial dilution of the control-treated cell lysate. GAU Adtdt-3', and HPV E6 (18E6-385), 5'-CUA ACA CUG GGU UAU ACA Adtdt-3'. SiGlo Risc-Free siRNA (Dharmacon) Conclusions was used as a control siRNA sequence for these experiments. A stathmin shRNA plasmid and control plasmid were obtained Stathmin depletion is required for cell proliferation and sur- from Superarray Bioscience Corporation and used to confirm vival of those cells lacking p53. Given that p53 is mutated in at results from siRNA. Cells were grown similarly to those treated least 50% of human cancers, mechanisms to specifically target with RNAi except that Fugene 6 was used to transfect cells with these cells for death hold great promise in treatment of a wide shRNA plasmids. The shRNA (manufacturer’s #4) used recog- range of cancers. Although the cellular mechanism responsible nizes 607–627 of exon 4 of the stathmin gene. Some cells were is unknown, our data add stathmin depletion to the handful of also transfected with TransSilent empty vector or TransSilent p53 strategies available to block proliferation of those cells lacking a 36 47,48 shRNA plasmids (Panomics; ) using Fugene 6. functional p53. Given that nitrosoureas appear to target stathmin 708 Cancer Biology & Therapy Volume 9 Issue 9 45 and control glioma cell migration, these compounds may be for a generous gift of HCT116 cell lines. Thanks also to Bob useful agents to induce apoptosis in a mutant p53 background. Skibbens, Danielle Ringhoff and Victoria Caruso for their advice during the course of these studies. Supported by grants from NIH Acknowledgements (GM058025) and the Pennsylvania Department of Health to L.C. We are indebted to Dr. Maureen Murphy, Fox Chase Cancer The Pennsylvania Department of Health specicfi ally disclaims Center, for many helpful discussions and to Dr. B. Vogelstein responsibility for any analyses, interpretations or conclusions. 18. Brattsand G. Correlation of oncoprotein 18/stathmin 34. Johnsen JI, Aurelio ON, Kwaja Z, Jorgensen GE, References expression in human breast cancer with established Pellegata NS, Plattner R, et al. p53-mediated negative 1. Jordan MA, Kamath K. How do microtubule-targeted prognostic factors. Br J Cancer 2000; 83:311-8. regulation of stathmin/Op18 expression is associ- drugs work? An overview. Curr Cancer Drug Targets ated with G(2)/M cell cycle arrest. Int J Cancer 2000; 19. Bieche I, Lachkar S, Becette V, Cifuentes-Diaz C, Sobel 2007; 7:730-42. 88:685-91. A, Lidereau R, et al. Overexpression of the stathmin 2. Rieder CL, Maiato H. Stuck in division or passing gene in a subset of human breast cancer. Br J Cancer 35. Ahn J, Murphy M, Kratowicz S, Wang A, Levine through: what happens when cells cannot satisfy the 1998; 78:701-9. AJ, George DL. Downregulation of the stathmin/ spindle assembly checkpoint. Dev Cell 2004; 7:637- Op18 and FKBP25 genes following p53 induction. 20. Price DK, Ball JR, Bahrani-Mostafavi Z, Vachris JC, Oncogene 1999; 18:5954-8. Kaufman JS, Naumann RW, et al. The phosphoprotein 3. Weaver BA, Cleveland DW. Decoding the links Op18/stathmin is differentially expressed in ovarian 36. Giono LE, Manfredi JJ. Mdm2 is required for inhibi- between mitosis, cancer and chemotherapy: the mitotic cancer. Cancer Invest 2000; 18:722-30. tion of Cdk2 activity by p21, thereby contributing to checkpoint, adaptation and cell death. Cancer Cell p53-dependent cell cycle arrest. Mol Cell Biol 2007; 21. Zhang HZ, Wang Y, Gao P, Lin F, Liu L, Yu B, et al. 2005; 8:7-12. 27:4166-78. Silencing stathmin gene expression by survivin promot- 4. Bhat KM, Setaluri V. Microtubule-associated proteins er-driven siRNA vector to reverse malignant phenotype 37. Sur S, Pagliarini R, Bunz F, Rago C, Diaz LA Jr, Kinzler as targets in cancer chemotherapy. Clin Cancer Res of tumor cells. Cancer Biol Ther 2006; 5:1457-61. KW, et al. A panel of isogenic human cancer cells sug- 2007; 13:2849-54. gests a therapeutic approach for cancers with inactivat- 22. Wang R, Dong K, Lin F, Wang X, Gao P, Wei SH, et al. 5. Chin GM, Herbst R. Induction of apoptosis by ed p53. Proc Natl Acad Sci USA 2009; 106:3964-9. Inhibiting proliferation and enhancing chemosensitiv- monastrol, an inhibitor of the mitotic kinesin Eg5, is ity to taxanes in osteosarcoma cells by RNA interfer- 38. Brito DA, Rieder CL. Mitotic checkpoint slippage in independent of the spindle checkpoint. Mol Cancer ence-mediated downregulation of stathmin expression. humans occurs via cyclin B destruction in the presence Ther 2006; 5:2580-91. Mol Med 2007; 13:567-75. of an active checkpoint. Curr Biol 2006; 16:1194- 6. Kouzu Y, Uzawa K, Koike H, Saito K, Nakashima 23. Alli E, Yang JM, Hait WN. Silencing of stathmin D, Higo M, et al. Overexpression of stathmin in oral induces tumor-suppressor function in breast cancer cell 39. Clute P, Pines J. Temporal and spatial control of cyclin squamous-cell carcinoma: correlation with tumour lines harboring mutant p53. Oncogene 2007; 26:1003- B1 destruction in metaphase. Nat Cell Biol 1999; progression and poor prognosis. Br J Cancer 2006; 12. 1:82-7. 94:717-23. 24. Schubart UK, Yu J, Amat JA, Wang Z, Hoffmann MK, 40. Borgne A, Meijer L. Sequential dephosphorylation of 7. Mistry SJ, Bank A, Atweh GF. Targeting stathmin in Edelmann W. Normal development of mice lacking p34(cdc2) on Thr-14 and Tyr-15 at the prophase/meta- prostate cancer. Mol Cancer Ther 2005; 4:1821-9. metablastin (P19), a phosphoprotein implicated in cell phase transition. J Biol Chem 1996; 271:27847-54. 8. Rana S, Maples PB, Senzer N, Nemunaitis J. Stathmin cycle regulation. J Biol Chem 1996; 271:14062-6. 41. Holmfeldt P, Brannstrom K, Stenmark S, Gullberg M. 1: a novel therapeutic target for anticancer activity. 25. Wang Y, Ji P, Liu J, Broaddus RR, Xue F, Zhang Aneugenic activity of Op18/stathmin is potentiated by Expert Rev Anticancer Ther 2008; 8:1461-70. W. Centrosome-associated regulators of the G(2)/M the somatic Q18→e mutation in leukemic cells. Mol 9. Melhem R, Hailat N, Kuick R, Hanash SM. checkpoint as targets for cancer therapy. Mol Cancer Biol Cell 2006; 17:2921-30. Quantitative analysis of Op18 phosphorylation in 2009; 8:8. 42. O’Connell MJ, Cimprich KA. G damage checkpoints: childhood acute leukemia. Leukemia 1997; 11:1690- 26. Larsson N, Melander H, Marklund U, Osterman O, what is the turn-on? J Cell Sci 2005; 118:1-6. Gullberg M. G /M transition requires multisite phos- 43. Budde PP, Kumagai A, Dunphy WG, Heald R. 10. Nylander K, Marklund U, Brattsand G, Gullberg M, phorylation of oncoprotein 18 by two distinct protein Regulation of Op18 during spindle assembly in Roos G. Immunohistochemical detection of oncopro- kinase systems. J Biol Chem 1995; 270:14175-83. Xenopus egg extracts. J Cell Biol 2001; 153:149-58. tein 18 (Op18) in malignant lymphomas. Histochem J 27. Marklund U, Larsson N, Gradin HM, Brattsand G, 44. Wiman KG. Restoration of wild-type p53 function in 1995; 27:155-60. Gullberg M. Oncoprotein 18 is a phosphorylation- human tumors: strategies for efficient cancer therapy. 11. Nakashima D, Uzawa K, Kasamatsu A, Koike H, Endo responsive regulator of microtubule dynamics. EMBO Adv Cancer Res 2007; 97:321-38. Y, Saito K, et al. Protein expression profiling identi- J 1996; 15:5290-8. 45. Liang XJ, Choi Y, Sackett DL, Park JK. Nitrosoureas fies maspin and stathmin as potential biomarkers of 28. Larsson N, Marklund U, Gradin HM, Brattsand G, inhibit the stathmin-mediated migration and invasion adenoid cystic carcinoma of the salivary glands. Int J Gullberg M. Control of microtubule dynamics by of malignant glioma cells. Cancer Res 2008; 68:5267- Cancer 2006; 118:704-13. oncoprotein 18: dissection of the regulatory role of 12. Nishio K, Nakamura T, Koh Y, Kanzawa F, Tamura T, multisite phosphorylation during mitosis. Mol Cell 46. Holmfeldt P, Stenmark S, Gullberg M. Interphase- Saijo N. Oncoprotein 18 overexpression increases the Biol 1997; 17:5530-9. specific phosphorylation-mediated regulation of tubu- sensitivity to vindesine in the human lung carcinoma 29. Ringhoff DN, Cassimeris L. Stathmin regulates cen- lin dimer partitioning in human cells. Mol Biol Cell cells. Cancer 2001; 91:1494-9. trosomal nucleation of microtubules and tubulin 2007; 18:1909-17. 13. Yuan RH, Jeng YM, Chen HL, Lai PL, Pan HW, Hsieh dimer/polymer partitioning. Mol Biol Cell 2009; 47. Piehl M, Cassimeris L. Organization and dynamics of FJ, et al. Stathmin overexpression cooperates with p53 20:3451-8. growing microtubule plus ends during early mitosis. mutation and osteopontin overexpression, and is asso- 30. DeGraffenried LA, Fulcher L, Friedrichs WE, Grunwald Mol Biol Cell 2003; 14:916-25. ciated with tumour progression, early recurrence and V, Ray RB, Hidalgo M. Reduced PTEN expression in 48. Warren JC, Rutkowski A, Cassimeris L. Infection with poor prognosis in hepatocellular carcinoma. J Pathol breast cancer cells confers susceptibility to inhibitors replication-deficient adenovirus induces changes in the 2006; 209:549-58. of the PI3 kinase/Akt pathway. Ann Oncol 2004; dynamic instability of host cell microtubules. Mol Biol 14. Chen G, Wang H, Gharib TG, Huang CC, Thomas 15:1510-6. Cell 2006; 17:3557-68. DG, Shedden KA, et al. Overexpression of oncoprotein 31. Wang NP, To H, Lee WH, Lee EY. Tumor suppressor 49. Garrett S, Auer K, Compton DA, Kapoor TM. hTPX2 18 correlates with poor differentiation in lung adeno- activity of RB and p53 genes in human breast carci- is required for normal spindle morphology and cen- carcinomas. Mol Cell Proteomics 2003; 2:107-16. noma cells. Oncogene 1993; 8:279-88. trosome integrity during vertebrate cell division. Curr 15. Friedrich B, Gronberg H, Landstrom M, Gullberg M, 32. Bunz F, Dutriaux A, Lengauer C, Waldman T, Zhou Biol 2002; 12:2055-9. Bergh A. Differentiation-stage specific expression of S, Brown JP, et al. Requirement for p53 and p21 to 50. Bradford MM. A rapid and sensitive method for the oncoprotein 18 in human and rat prostatic adenocarci- sustain G arrest after DNA damage. Science 1998; 2 quantitation of microgram quantities of protein utiliz- noma. Prostate 1995; 27:102-9. 282:1497-501. ing the principle of protein-dye binding. Anal Biochem 16. Belletti B, Nicoloso MS, Schiappacassi M, Berton S, 33. Koivusalo R, Krausz E, Helenius H, Hietanen S. 1976; 72:248-54. Lovat F, Wolf K, et al. Stathmin activity influences Chemotherapy compounds in cervical cancer cells sarcoma cell shape, motility and metastatic potential. primed by reconstitution of p53 function after short Mol Biol Cell 2008; 19:2003-13. interfering RNA-mediated degradation of human pap- 17. Ngo TT, Peng T, Liang XJ, Akeju O, Pastorino S, illomavirus 18 E6 mRNA: opposite effect of siRNA Zhang W, et al. The 1p-encoded protein stathmin and in combination with different drugs. Mol Pharmacol resistance of malignant gliomas to nitrosoureas. J Natl 2005; 68:372-82. Cancer Inst 2007; 99:639-52. www.landesbioscience.com Cancer Biology & Therapy 709 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cancer Biology & Therapy Taylor & Francis

Stathmin/oncoprotein 18, a microtubule regulatory protein, is required for survival of both normal and cancer cell lines lacking the tumor suppressor, p53

Carney, Bruce K.; Cassimeris, Lynne
Cancer Biology & Therapy , Volume 9 (9): 11 – May 1, 2010
Download PDF
11 pages

Loading next page...
 
/lp/taylor-francis/stathmin-oncoprotein-18-a-microtubule-regulatory-protein-is-required-ijjoH210mc

References (50)

Publisher
Taylor & Francis
Copyright
Copyright © 2010 Landes Bioscience
ISSN
1555-8576
eISSN
1538-4047
DOI
10.4161/cbt.9.9.11430
Publisher site
See Article on Publisher Site

Abstract

ReseaRCh papeR ReseaRCh p apeR Cancer Biology & Therapy 9:9, 699-709; May 1, 2010; © 2010 Landes Bioscience Stathmin/oncoprotein 18, a microtubule regulatory protein, is required for survival of both normal and cancer cell lines lacking the tumor suppressor, p53 Bruce K. Carney and Lynne Cassimeris* Department of Biological s ciences; Lehigh University; Bethlehem, pa Usa Key words: stathmin, p53, apoptosis, cell cycle, microtubule stathmin, a microtubule regulatory protein, is overexpressed in many cancers and required for survival of several cancer lines. In a study of breast cancer cell lines, proposed that stathmin is required for survival of cells lacking p53, but this hypothesis was not tested directly. h ere we tested their hypothesis by examining cell survival in cells depleted of stathmin, p53 or both proteins. Comparing h CT116 colon cancer cell lines differing in T p53 genotype, stathmin depletion resulted in significant death only in cells lacking p53. a s a second experimental system, we compared the effects of stathmin depletion from h eLa cells, which normally lack detectable levels of p53 due to expression of the hp V e6 protein. s tathmin depletion caused a large percentage of h eLa cells to die. Restoring p53, by depletion of hpV e6, p53-/- rescued h eLa cells from stathmin-depletion induced death. Cleaved pa Rp was detected in h CT116 cells depleted of stathmin and cell death in stathmin-depleted h eLa cells was blocked by the caspase inhibitor Z-Va D-FMK, consistent with apoptotic death. The stathmin-dependent survival of cells lacking p53 was not confined to cancerous cells because both proteins were required for survival of normal human fibroblasts. In h CT116 and h eLa cells, depletion of both stathmin and p53 leads to a cell cycle delay through G . Our results demonstrate that stathmin is required for cell survival in cells lacking p53, suggesting that stathmin depletion could be used therapeutically to induce apoptosis in tumors without functional p53. generally poor prognosis, supporting the use of stathmin expres- Introduction 7,14 sion as a biomarker for cancer stage progression. Microtubules are dynamic polymers of α/β tubulin dimer sub- While stathmin level may serve as a biomarker in many cancers, units that contribute to a number of cell processes including several studies have demonstrated that stathmin depletion may chromosome movement and mitosis. Several successful chemo- also have therapeutic value; in a number of cancer cell lines stath- therapies, including paclitaxel and vinblastine, are thought to act min depletion results in cell cycle arrest and apoptosis. Mistry et by targeting microtubules or their tubulin subunits, disrupting al. found that expression of anti-stathmin ribozymes depleted mitotic spindle function, activating the spindle assembly check- stathmin and resulted in apoptosis in the androgen-independent 1-3 point and inducing cell death. While microtubules have been prostate cancer cell line LNCaP. Depletion of stathmin by siR NA 1 21 22 a successful target for chemotherapy development, accessory or shRNA in HeLa cells, osteosarcoma cell lines and several proteins that regulate microtubule assembly and disassembly are breast cancer cell lines also demonstrated reduced cell prolifera- 4,5 also targets for novel chemotherapeutics. Of these regulatory tion and apoptosis after stathmin reduction. In contrast to these proteins, stathmin/oncoprotein 18, a microtubule destabilizing cancer cell lines, stathmin knockout mice are viable. These protein, has received attention recently as a potential target for results have raised interest in stathmin reduction as a potential 6-8 cancer therapy. cancer therapy, possibly active only in cancerous cells. Stathmin is overexpressed in many cancers including It is currently not known why even partial stathmin depletion 9 10 6 11 leukemia, lymphoma, oral squamous-cell, adenoid cystic, is sufficient to induce apoptosis in several cancer cell lines. Most 12 13 14,15 lung and hepatocellular carcinomas, adenocarcinoma, pros- studies have indicated that stathmin depletion results in a G /M 7 16 17 18,19 7,21,25 tate cancer, sarcomas, glioma, breast cancer and ovarian block, but normally stathmin is inactivated by phosphory- 20 26-28 cancer. Cancers where stathmin is overexpressed have been lation at entry into mitosis and does not have a detectable shown to have a high malignancy phenotype, which leads to microtubule destabilizing activity during mitosis. These latter *Correspondence to: Lynne Cassimeris; Email: [email protected] Submitted: 09/03/09; Revised: 01/08/10; Accepted: 02/08/10 Previously published online: www.landesbioscience.com/journals/cbt/article/11430 www.landesbioscience.com Cancer Biology & Therapy 699 p53W T p53-/- Figure 1. s tathmin depletion from h CT116 and h CT116 cell lines. Matched lines of h CT116 cells differing in p53 genotype were treated with p53-/- siRNa targeting stathmin. ( a) s tathmin depletion from h CT116 cells was clearly detectable 24 hrs after transfection and stayed low for at least 5 d W Tp53 post transfection. (B) siRNa depletion of stathmin from h CT116 cells. The sample shown at 5 days included a second siRNa transfection, 72 h after the initial siRNa transfection, which was necessary to keep stathmin level low. s amples at 1 and 3 d are from a single siRNa transfection. In both ( a and B), the lysates from non-targeting siRNa treated cells were loaded in a dilution series to estimate the extent of depletion by stathmin siRN a . Blots were re-probed for actin as an indicator of equal loading between control (100% load) and stathmin-depleted samples. (C) Low magnification phase contrast images of living cells acquired at the times given after transfection with control or stathmin siRNa . The same field is shown within an experi - p53-/- mental condition over 4–5 d. h CT116 cells depleted of stathmin remain relatively sparse over the time course while cells grow to near confluency in the other experimental conditions. Data shown are representative of three separate experiments. data strongly suggest that stathmin depletion does not lead to Results apoptosis by inducing a mitotic block. An alternative mecha- nism was suggested by Alli et al. who noted that those cell lines Depletion of stathmin from HCT116 colon cancer cell lines. requiring stathmin for survival also expressed mutant p53. This To test directly whether p53 is required for survival of cells intriguing hypothesis was not tested directly because Alli et al. depleted of stathmin, we used HCT116 colon cancer cell lines p53-/- were unable to deplete stathmin from cell lines expressing wild- differing in TP53 genotype. The HCT116 line was devel- type p53. Further complicating analysis, some of the breast cancer oped by knocking out both copies of the TP53 gene from the 23 W Tp53 32 cell lines used by Alli et al. have additional mutations targeting parental HCT116 line (HCT116 ; ). We first examined 30,31 other tumor suppressor genes, including Rb and PTEN. the extent and time course of stathmin depletion after siRNA Here we used several experimental systems to test whether transfection, compared to cells transfected with siGlo, a non- p53-/- stathmin is specifically required for survival of those cells lacking targeting control siRNA (Dharmacon). For HCT116 cells, p53. First, we compared cell proliferation and death after stath- stathmin was depleted by approximately 75% within 24 h after min depletion from matched colon cancer cell lines (HCT116) transfection and remained low for 5 d (Fig. 1A). We found that W Tp53 differing in p53 genotype. As an alternate approach, we restored stathmin depletion was more difficult in HCT116 , consis- p53 to HeLa cells by depletion of the human papillomavirus tent with a previous study of breast cancer cell lines. By 72 h (HPV) E6 protein, which normally binds p53 and marks it for after siRNA transfection, stathmin protein level was reduced by W Tp53 destruction. Finally, we depleted stathmin and/or p53 from nor- approximately 50% in HCT116 (Fig. 1B). Stathmin level mal human foreskin fibroblasts (HFF cells) to address whether often returned at later time points (not shown), but a second stathmin is specifically required for cancer cell survival. We siRNA transfection at 72 h kept the stathmin level low at day found that stathmin is required for the survival of both cancerous 5 (time after the first transfection, Fig. 1B ) and up to 7 d after and non-cancerous derived cell lines, but only in the absence of the initial transfection (not shown). These results demonstrate p53. In cells lacking both stathmin and p53, cells arrest or delay that stathmin can be successfully knocked down in HCT116 at G of the cell cycle. matched cell lines. 700 Cancer Biology & Therapy Volume 9 Issue 9 p53-/- Figure 2. s tathmin is required for surival of h CT116 cells. h CT116 cell lines were treated with non-targeting control or stathmin RNa i. a t the indicated time points, cells were trypsinized, incubated in Trypan Blue to identify dead cells, and numbers of living and dead cells counted. Cell proliferation rates (a) and the percentage of dead cells (B) represent the means of three independent experiments ± s D. Stathmin is required for cell survival in HCT116 cells each HCT116 cell line at the 5 d time point. The difference in p53-/- lacking p53. The HCT116 cell lines were depleted of stathmin the mean percentage of dead cells in HCT116 control and using the conditions described above and were observed for up stathmin-depleted cells was highly significant (p < 0.001), while W Tp53 to 5 d post-siRNA transfection to observe the effect of stath- the means were not significantly different in HCT116 cells W Tp53 min depletion on cell proliferation and death. First, we followed and HCT116 cells depleted of stathmin. Taken together, living cells by imaging the same regions of coverslips over a 4–5 these results provide strong evidence that stathmin is required W Tp53 d time course. For HCT116 cells, either expressing stathmin for cell survival only in cells lacking p53, but the differences in or depleted of stathmin, cell density increased over time, reaching time course of stathmin knockdown in cells differing in TP53 near conu fl ency by day 4–5 ( Fig. 1C). The same growth charac- genotype complicates analysis. p53-/- teristics were observed for the HCT116 cell line transfected Restoration of p53 in HeLa cells rescues cells from stath- with a control siRNA. In contrast, depletion of stathmin from min-depletion induced death. To further test whether stathmin p53-/- the HCT116 line significantly reduced the increase in cell is required for survival of cells lacking p53, we used HeLa cells, number over time (Fig. 1C). which are wildtype for the TP53 gene, but fail to maintain p53 Following living cells over time suggested that stathmin protein level because they stably express the human papilloma- depletion slowed cell proliferation and/or led to cell death only virus (HPV) E6 protein. The HPV E6 protein binds p53 and p53-/- 33 in the HCT116 cell line. To examine cell growth and death marks it for destruction. The ability to deplete the HPV E6 more directly, we measured both cell proliferation, by count- protein by RNAi provides a mechanism to restore p53 in HeLa ing the number of living cells, and cell death, by counting the cells, allowing comparison of cells depleted of stathmin in the percentage of trypan blue positive cells over a 5 d time course. presence and absence of p53. HCT116 cells expressing wildtype p53 and depleted of stathmin We first confirmed the results of Koivusalo et al. that E6 proliferated at a slower rate than cells transfected with a control depletion restores p53 in HeLa cells. Where noted, cells were RNA (solid lines, Fig. 2A). Although the stathmin depleted treated with the DNA-damaging agent doxorubicin for the W Tp53 HCT116 cells grew more slowly, they remained viable, as 24 h prior to gel sample preparation to stabilize p53 and enhance measured by trypan blue exclusion (Fig. 2B). Stathmin depletion its detection. As shown in Figure 3A, HeLa cells have a very low p53-/- from HCT116 cells showed very slow proliferation over a 5 d level of p53, even after doxorubicin treatment. Depletion of E6 time course (Figs. 2A, see also 1C) and these cells died at a much by siRNA results in a significant increase in p53 level, which W Tp53 greater rate than HCT116 cells depleted of stathmin, or is also detectable in cells depleted of both stathmin and E6 either of the HCT116 cell lines transfected with a control RNA (Fig. 3A). (Fig. 2B). There was a high variability in the percentage of dead Stathmin was also readily depleted from HeLa cells, as shown p53-/- cells in HCT116 cells depleted of stathmin, as seen by the in Figure 3B. At 72 h after siRNA transfection, stathmin was large standard deviations. At this time we do not know the reason reduced by at least 75%, independent of whether cells were also for the inter-experiment variability. To confirm that the differ - depleted of E6 (p53 restored). Stathmin level was also reduced, ence in means was significant, we compared the mean percent - although to a lesser extent, in cells depleted only of E6. In this ages of dead cells for control and stathmin-depleted cells within latter case, reduced stathmin level may reflect the ability of p53 www.landesbioscience.com Cancer Biology & Therapy 701 Figure 3. Restoring p53 in h eLa cells. (a) h eLa cells were treated with control or siRNa targeting hp V e6 for 72 h and exposed to 0.5 µM doxorubicin for the final 24 h to induce DN a damage and stabilize p53, facilitating its detection by immunoblot. In control-treated h eLa cells, p53 is present at a very low level, even in the presence of doxorubicin. Depletion of hp V e6 restored p53. (B) h eLa cells were treated with siGlo control siRNa (or empty vector controls) or siRNa s targeting stathmin, hp V e6 or both mRNa s. a nti-stathmin immunoblot of cell lysates were collected 72 h after siRNa transfection. s tathmin was depleted by at least 75% after treatment with siRNa targeting stathmin or stathmin and hp V e6. Depletion of hp V e6 also 34,35 caused a reduction in stathmin level, possibly by p53-mediated repression of stathmin expression. For both (a and B), blots were re-probed for actin as a marker of equal loading between control (100%) and siRNa samples. (C) Low magnification phase contrast images of living cells acquired at the times given after transfection with control or siRNa s targeting stathmin, hp V e6 or both mRNa s. The same field is shown for each experimental condition over 4–5 d. h eLa cells depleted of e6 or stathmin and e6 grow slightly slower than control siRNa treated cells, while stathmin depleted cells show minimal growth. Data shown are representative of three separate experiments. to negatively regulate stathmin expression when p53 level is showed proliferation rates and percentages of dead cells compara- 34,35 transiently increased. ble to control-treated or HPV E6 siRNA-treated cells. Finally, we We next examined cell density over time, similar to the confirmed the siRNA results by transfecting cells with a plasmid experiments performed with HCT116 cell lines. Stathmin deple- encoding an shRNA directed against a different sequence in the tion from HeLa cells reduced the increase in cell number over stathmin mRNA (see methods). As shown in Figure 4C for cells time observed in control-siR NA treated cells (Fig. 3C). Restoring examined 72 h after transfection, shRNA depletion of stathmin p53 (E6 depletion), in the absence or presence of stathmin deple- significantly increased the percentage of dead cells, compared to tion, resulted in slower cell growth compared to control cells, but cells transfected with a control plasmid. Together with the results not nearly as significant an inhibition as that observed in stath - from HCT116 cell lines, these results provide strong evidence min depleted HeLa cells (Fig. 3C). We then measured both cell that stathmin is required for cell survival only in the absence of proliferation (live cell number over time) and the percent dead p53. cells (trypan blue positive) for these experimental conditions. In the above experiments, we used trypan blue exclusion as Figure 4 shows that stathmin depletion from HeLa cells (lack- a simple assay to differientiate living and dead cells. It is likely ing p53) resulted in both reduced cell proliferation and a large that stathmin depletion results in death by apoptosis, as others 7,21,23,25 increase in the percentage of dead cells. Comparison of mean have demonstrated previously. To confirm that cells die via percentages of dead cells at day 5 showed that stathmin depletion apoptosis we examined whether stathmin-depletion induced cell signicantly increased the percentage of dead cells over that mea- death was inhibited by the capsase inhibitor Z-VAD-FMK. As sured in control siRNA treated cells (p < 0.001). Treatment of expected, Z-VAD-FMK significantly blocked the death of HeLa cells with siRNA to deplete both stathmin and HPV E6 (restor- cells depleted of stathmin (Fig. 5A and B). Stathmin depleted ing p53) reversed the effects of stathmin depletion alone; cells cells treated with Z-VAD-FMK also showed cell proliferation 702 Cancer Biology & Therapy Volume 9 Issue 9 Figure 4. Restoring p53 rescues h eLa cells from stathmin-depletion induced death. Cell proliferation rates and the percentage of dead cells were mea- sured as described in Figure 2. (a and B) s tathmin depletion slows cell proliferation and increases the percentage of dead (trypan blue positive) cells. Depletion of hp V e6 (restoring p53) alone or in combination with stathmin depletion does not lead to significant cell death. (C) Transfection of cells with an shRNa targeting stathmin resulted in stathmin depletion (top) and a significant increase in the percentage of dead cells at 72 h post- transfection. Controls were transfected with the negative control plasmid provided by the manufacturer. *denotes p < 0.05. a ll data represent the means of three independent experiments ± s D. rates intermediate between control-treated and stathmin-siRNA achieved using RNAi and shRNA respectively (Fig. 6A and B). treated cells. In additional experiments, we found increased lev- A 50% knockdown of stathmin was seen as early as day 3 and els of cleaved poly-ADP ribose polymerase (PARP) in stathmin- continued to approximately 75% knockdown by day 5. By day p53-/- depleted HCT116 cells (Fig. 5C and D), but not in stath- 2, p53 showed a significant knockdown as well. Note that the W Tp53 min-depleted HCT116 cells. These results confirm previous only cells treated with doxorubucin were those used to prepare 7,21,23,25 results of others that stathmin depletion initiates apop- gel samples for p53 detection (as noted in Fig. 6B). Depletion of totic death in cells lacking p53. stathmin or p53 alone did not inhibit cell proliferation rate and Depletion of p53 and stathmin from non-cancerous cells did not increase cell death compared to cells transfected with leads to cell death. The experiments above, along with previ- control siRNA or the empty vector used for shRNA delivery (Fig. 7,21,23 ous results of others, demonstrated that stathmin depletion 6C and D). Interestingly, however, depleting both stathmin and from cancer cell lines results in cell death by apoptosis. We next p53 together resulted in significant cell death (p < 0.001 at day 5) asked whether non-cancerous cells also require stathmin for sur- and a lack of cell proliferation (Fig. 6C and D). These data dem- vival using human foreskin fibroblasts (HFF) as a model non- onstrate that stathmin is required for cell survival in both human cancerous cell line. To deplete p53 from these cells, we trans- cancerous and non-cancerous cell lines lacking p53. fected cells with a plasmid for expression of an shRNA targeting Cells arrest in G of the cell cycle prior to apoptosis. Both 36 34 37 p53. A successful knockdown of both stathmin and p53 was the overexpression of wildtype p53, and the absence of p53, www.landesbioscience.com Cancer Biology & Therapy 703 Figure 5. s tathmin-depletion induced cell death likely occurs via apoptosis. (a) The caspase inhibitor Z-V a D-FMK blocks stathmin-depletion induced cell death in h eLa cells. h eLa cells were treated with 10 µm Z-Va D-FMK for 24 h prior to transfection with either siGlo control siRNa (or empty vector controls) or RNa i against stathmin. plots of cell proliferation rates (a) and the percentage of dead cells (B) represent the means of three independent experiments ± sD. (C) h CT116 matched cell lines were transfected with control or stathmin RNa i, fixed 72 h later and stained for cleaved pa Rp. The p53-/- percentage of cells positive for cleaved pa Rp are plotted. Only h CT116 cells depleted of stathmin show a signficant increase in cells positive for cleaved pa Rp. Data shown are the means ± sD of three independent experiments and a total of ∼ 200 cells per treatment. (D) Representative image p53-/- from h CT116 cells depleted of stathmin and staining positive for cleaved pa Rp in the nucleus (arrow). s everal weakly stained cells (counted as negative) are also included. s cale bar = 10 µm. have been shown to cause cell cycle arrest, at least under some phosphosphorylated on Tyr15 (Phospho-CDK1(Y15); inhibi- experimental conditions. To address whether stathmin deple- tory phosphorylation on CDK1 present during G and removed tion leads to a unique cell cycle arrest that is dependent upon at entry into mitosis ). Cells were co-stained with antibodies p53 status, we examined cell cycle distributions in HCT116 to tubulin (Figs. 7 and 8), allowing visual classification of cells matched cells depleted of stathmin and HeLa cells depleted in M phase. Using these staining conditions and counting all p53-/- of stathmin, HPV E6 (restoring p53) or both proteins. Others cells within multiple images, we found that HCT116 cells have previously followed DNA levels and identified a G /M depleted of stathmin showed a significant (p < 0.05) increase in cell cycle arrest after stathmin depletion in cancer-derived cell non-mitotic cells staining positively for TPX 2 (Fig. 7A and B) or 7,23 lines, but these studies measured DNA content and cannot Cyclin B (Fig. 7C and D) 48 h post transfection. This result was differentiate between G and M phases. Therefore, we used an most pronounced in TPX2 stained samples. The mitotic index alternative protocol, by staining fixed cells with antibodies to was approximately equal among the two HCT116 lines whether TPX2 (present in the nuclei of S and G cells and bound to stathmin was depleted or not. the mitotic spindle microtubules in M phase ), Cyclin B (pres- We then extended these staining protocols to HeLa cells. ent in the cytoplasm of G cells and associated with the mitotic Depletion of stathmin from HeLa cells also increased the per- spindle until anaphase onset during M phase ), and CDK1 centage of non-mitotic TPX2 positive cells (Fig. 8A and B) and 704 Cancer Biology & Therapy Volume 9 Issue 9 Figure 6. Depleting both p53 and stathmin from normal human fibroblast ( h FFs) cells leads to cell death. h FF cells were transfected with Transs ilent empty vector, p53 shRNa plasmid and/or siRN a targeting stathmin as indicated. ( a) a nti-stathmin immunoblot from cell lysates were isolated 3 d after transfection. a knockdown of >75% was observed for both stathmin siRN a alone or when used in combination with shRN a against p53. (B) a nti-p53 immunoblot from cell lysates isolated at the times indicated. When present, doxorubucin was included for the final 24 h of incubation. The level of p53 is detectably reduced 1 d after shRNa transfection, but is more significantly depleted by 2–3 d after transfection. (C and D) p lots of h FF cell proliferation rates (C) and the percentage of dead cells (D) represent the means of three independent experiments ± s D. concomitantly significantly decreased the percentage of cells not prior to entry into mitosis. Depletion of either stathmin or p53 recognized as TPX2 positive. Untreated HeLa cells, cells treated alone is not sufficient to delay progression through the cell cycle. with a non-targeting siRNA or depleted or E6 to restore p53 showed a much smaller percentage of non-mitotic cells stain- Discussion ing positive for TPX2. As a second marker for cells in G , we also stained HeLa cells with antibodies recognizing phospho- In the results presented here we demonstrated that stathmin is CDK1(Y15), one of the two inhibitory phosphorylations hold- required for the survival of cell lines lacking p53, whether those ing CDK1 inactive until entry into M phase. Consistent with lines are derived from cancerous or normal tissues. Stathmin p53-/- results from TPX2 staining, HeLa cells depleted of stathmin depletion from HCT116 and HeLa cells showed significantly showed a significant increase (p < 0.05) in non-mitotic cells reduced proliferation rates and an increased percentage of dead stained positive for phospho-CDK1(Y15) (Fig. 8C and D). Cells cells, while depletion of stathmin had little effect on the viabil- W Tp53 depleted of E6, or of E6 and stathmin, did not show significant ity of HCT116 cells or of HeLa cells expressing p53. These changes in the percentage of non-mitotic, phospho-CDK1(Y15) observations were not confined to cancer-derived cell lines since positive cells. We confirmed these results by transfecting HeLa normal human fibroblasts (HFFs) also required stathmin for pro - cells with a control plasmid or a plasmid expressing an shRNA liferation and survival, but only in the absence of p53. For both W Tp53 directed against stathmin and again found an increase in non- HCT116 and HPV E6-depleted HeLa cells, stathmin deple- mitotic phospho-CDK1(Y15) positive cells in stathmin-depleted tion slowed cell proliferation somewhat, but did not result in cell cells (Fig. 8E). From these data, it is evident that stathmin deple- death and did not impose a cell cycle block. Our results provide tion disrupts cell cycle progression, leading to an increase in cells direct support for the model proposed by Alli et al. demonstrat- positive for G markers, but only in the absence of p53. Given ing that stathmin is required for survival in cells lacking p53. that others have reported a G /M block in stathmin depleted Several groups have previously reported slowing of cell prolif- cells, based on analysis of DNA content, and our results indicate eration and cell death after stathmin depletion, but these stud- a G block, it most likely that cells are blocked or delayed in G , ies did not address a stathmin requirement for cell survival in 2 2 www.landesbioscience.com Cancer Biology & Therapy 705 p53-/- Figure 7. s tathmin depletion from h CT116 cells delays G of the cell cycle. Matched h CT116 cell lines were treated with either siGlo control siRNa or siRNa targeting stathmin mRN a s. (a and B) h CT116 cells were fixed 48 h after siRN a transfection and stained with antibodies against α -tubulin (green in merged images) and TpX2 (red in merged images). TpX2 is present in s /G and M phases of the cell cycle. The numbers of TpX2 positive, negative and mitotic cells were counted, and the percentage of cells in each group are shown in (a). Images shown in (B) are from a representative field of cells transfected with stathmin siRNa . Negative cells were identified by their array of interphase microtubules and their lack of T pX2 staining. Cells in inter- phase staining positively for TpX2 were scored as positive. Mitotic cells were identified by the presence of a mitotic spindle (tubulin staining). s cale bar = 25 µm. (C and D) h CT116 cell lines were fixed 48 h after siRN a transfections and stained with antibodies against α -tubulin (green in merged images) and cyclin B (red in merged images). Cytoplasmic cyclin B is a marker of G cells. The percentage of non-mitotic cyclin B positive and negative cells are shown in (C). Images shown in (D) are from representative cells fixed 48 h after stathmin siRN a transfection. s cale bar = 25 µm. The data represent the means of three (C) or four (a) independent experiments, each including ∼ 1,000 cells ± sD for each treatment group. *denotes p < 0.05. non-cancerous cells. Zhang et al. developed a method for expres- proliferation and increased cell death contrasts with the reported sion of stathmin shRNA driven by the survivin promoter, limit- viability of mice with both copies of the stathmin and p53 genes ing shRNA expression to those cells expressing survivin. Since knocked out. It is possible that these mice compensate through normal differentiated cells do not express survivin, it is unclear increased or decreased expression of other proteins allowing nor- whether the survivin promoter would be active, and thus express- mal development and survival in these mice. ing stathmin shRNA, in the endothelial cells used as controls. Stathmin and p53 are required for cell cycle progression. To Alli et al. were unable to deplete stathmin from breast cancer address why stathmin is required for cell survival only in cells lines expressing p53, making it unclear whether stathmin was lacking p53, we examined cell cycle distributions in HCT116 cell required for survival of these cells. Wang et al. and Mistry et lines depleted of stathmin and in HeLa cells expressing stathmin al. examined stathmin depletion in cancer cell lines and did not and p53, either protein alone, or depleted of both proteins. We test stathmin’s requirement for survival of non-cancerous cells. found that cells depleted of both stathmin and p53 show a delay in Our finding that stathmin and p53 depletions from nor - G (based on staining for TPX2, cyclin B, phospho-CDK1(Y15) mal human fibroblasts results in significantly reduced cell and tubulin) while others have reported a G /M delay in cancer 706 Cancer Biology & Therapy Volume 9 Issue 9 Figure 8. s tathmin depletion from h eLa cells delays G of the cell cycle. h eLa cells were transfected with either siGlo control siRNa or siRN a targeting stathmin, hp V e6 or both mRNa s. (a and B) h eLa cells were fixed and stained with antibodies against α -tubulin (green in merged images) and TpX2 (red in merged images). Cells were then counted as described in Figure 7a , and the percent cells in each group are shown in (a). Images shown (B) are from representative cells fixed 48 h after siRN a transfections. a rrows in the merged image denote several TpX2 positive cells. s cale bar = 50 µm. (C and D) h eLa cells were fixed and stained with antibodies against α -tubulin (red in merged images) and phospho-CDK1(Y15) (green in merged images; this inhibitory phosphorylation must be removed for entry into M phase). The percentage of non-mitotic cells staining positive for phospho-CDK1(Y15) is shown in (C). Representative images from phospho-CDK1(Y15) stained cells are shown in (D). a rrows in the merged image denote phospho-CDK1(Y15) positive cells. s cale bar = 50 µm. (e) Depletion of stathmin by shRNa also increased the percentage of phospho-CDK1(Y15) positive non-mitotic cells at 48 h after transfection. each plot represents the mean of three independent experiments ± sD including ∼1,000 cells per treatment. *denotes p < 0.05. 7,21,23,25 cells depleted of stathmin. As discussed above, we think mitosis and is not required for spindle assembly, it is unlikely that it is likely that cells depleted of stathmin and p53 are blocked stathmin depletion would prevent entry into mitosis by disrupt- in G , prior to entry into mitosis. It is interesting that induced ing microtubule assembly and turnover. expression of p53 both represses stathmin expression and induces Activation of p53, typically in response to DNA damage, a G /M block. Thus, it is likely that the levels of stathmin and primarily induces a G block, but a G block is also possible 2 1 2 p53, rather than simply their presence or absence, contribute to under some conditions. It is intriguing that many proteins G /M cell cycle progression. functioning in the G /M checkpoint, including p53, are associ- 2 2 It is not yet clear why both stathmin and p53 are required ated with the centrosome and we recently found that stathmin for cell cycle progression through G or for entry into M phase. regulates microtubule nucleation from the centrosome during Stathmin is phosphorylated during mitosis, resulting in loss of interphase. These data suggest that depletion of both stath- 28,41 its microtubule destabilizing activity. Stathmin inactivation is min and p53 may influence either centrosome function or the 28,41 required for proper assembly of the mitotic spindle and stath- function(s) of G /M checkpoint proteins at the centrosome min depletion does not impact microtubule formation in the to cause a cell cycle delay during G . An alternative model is spindle. Because stathmin is normally turned off at entry into also possible, where loss of p53 and stathmin impact separate www.landesbioscience.com Cancer Biology & Therapy 707 pathways that together function synergistically to slow cell Cell growth and death measurements. Cells were grown cycle progression. for 1–5 d post-transfection with siRNA or shRNA, trypsinized, A role for p53, or loss of p53, in regulating G /M progression stained with Trypan Blue (0.4%) and counted using a hemo- was also identified recently by comparing transcriptome differ - cytometer. Cell viability was assessed by Trypan Blue exclusion. ences between matched cell lines differing in p53 status, includ- Live and dead cells were counted, averaged and pooled for three ing the HCT116 colon cancer cells used here. Cells lacking p53 separate experiments for each cell line and treatment. Data shown upregulated expression of genes functioning in G /M and were are means ± standard deviations. sensitive to treatment with a Plk (polo-like kinase) inhibitor (Plk Groups of cells were also followed over time by plating cells on functions in G –M progression). In terms of cell proliferation, grid-etched coverslips attached to the bottoms of 35 mm dishes stathmin depletion acts similarly to Plk inhibition, since each (MatTek Corporation). Cells were plated and incubated for treatment slows proliferation of cells lacking p53. Stathmin is 24 h before transfection of siRNAs. Cells were allowed to grow one target of Plk, but this phosphorylation inhibits stathmin’s for an additional 24 h and then observed using a 20 X objective on microtubule regulatory activity. Thus inhibiting Plk will keep an inverted microscope (Nikon TE2000E) equipped with phase stathmin in an active form, which could contribute to a G /M contrast optics and MetaVu image acquisition software. Ten grid 26,28 block by preventing proper spindle assembly. It is not clear squares were chosen randomly and images were acquired from why stathmin depletion acts similarly to Plk inhibition to slow these squares each day for up to 5 d. Three separate imaging time G /M progression; possibly a balance of stathmin and p53 func- course experiments were performed and representative images are tions are necessary to pass through G . shown in Figures 1 and 3. Statistical analysis of cell counts, including those determined Methods after immunofluorescent staining (below), were performed using Paired t-tests in Microsoft Excel or GraphPad Software Cell culture. Cells were grown at 37°C in a humidified atmo - (www.graphpad.com/quickcalcs/ttest1.cfm). sphere of 5% CO . HCT116 and HFF cells were grown in Indirect immunou fl orescence and confocal microscopy. Cells DMEM (GIBCO) supplemented with 3.7 g/L sodium bicar- were fixed and imaged as described previously. Primary antibod- bonate, x1 antibiotic/antimycotic (Sigma), 1% sodium pyru- ies used were mouse monoclonal α-tubulin (B512 ; Sigma-A ldrich), vate, and 10% fetal bovine serum (FBS) (GIBCO-Invitrogen). rabbit anti-TPX2, (gift from Duane Compton, Dartmouth HeLa cells were grown in MEM (GIBCO) supplemented with Medical School), rabbit anti-cyclin B (Sigma-Aldrich), rabbit 2.2 g/L sodium bicarbonate, x1 antibiotic/antimycotic and anti-phospho-CDK1 (Tyr 15) (Cell Signaling Technology) and 10% FBS. mouse anti-cleaved PARP (ASP214) (Cell Signaling Technology). Drugs. Doxorubicin was added to cells to induce DNA dam- Goat anti-mouse or rabbit Alexa Fluor 488 or 563 (Invitrogen) age and stabilize p53, facilitating its detection by western blot. were used as the secondary antibodies in these experiments. When used, doxorubicin (0.5 µM) was added to cells for the Confocal microscopy was used to image stained cells as described last 24 h of an experimental treatment. Doxorubicin was never previously. Images were acquired using a 40 X/1.3NA objective. used in experiments to measure cell proliferation or cell death. Image stacks were converted to maximum intensity projections, Some cells were treated with the caspase inhibitor, Z-VAD-FMK exported as TIFF files and assembled using Photoshop. (10 µm; Sigma-Aldrich) or DMSO as a vehicle control. Z-VAD- Protein isolation and western blotting. Soluble cell extracts FMK was added to cells 24 h prior to any other treatment and were prepared for SDS-polyacrylamide gel electrophoresis as remained for the duration of the experiment. described previously. Protein concentrations were measured by RNA interference and shRNA transfections. RNA interfer- Bradford assay. Membranes were probed and imaged as previ- ence (RNAi) was achieved using GeneSilencer reagents following ously described using primary antibodies mouse anti-p53 (Vision the manufacturer’s protocol. Cells were grown on 35 mm dishes Bio Systems), rabbit anti-stathmin (Sigma-Aldrich), or rabbit anti- for 1–2 d before the addition of siRNA. Cells were serum starved actin (Sigma-Aldrich) followed by goat anti-mouse or rabbit horse- 30 min pre-transfection and 4 h post-transfection to improve radish peroxidase-linked IgG (Sigma-Aldrich). Protein depletions transfection efc fi iency. RNAi oligonucleotides (Dharmacon) used were estimated by comparison of western blot signals to those gen- included: SMTN1 (Op18-443), 5'-CGU UUG CGA GAG AAG erated by serial dilution of the control-treated cell lysate. GAU Adtdt-3', and HPV E6 (18E6-385), 5'-CUA ACA CUG GGU UAU ACA Adtdt-3'. SiGlo Risc-Free siRNA (Dharmacon) Conclusions was used as a control siRNA sequence for these experiments. A stathmin shRNA plasmid and control plasmid were obtained Stathmin depletion is required for cell proliferation and sur- from Superarray Bioscience Corporation and used to confirm vival of those cells lacking p53. Given that p53 is mutated in at results from siRNA. Cells were grown similarly to those treated least 50% of human cancers, mechanisms to specifically target with RNAi except that Fugene 6 was used to transfect cells with these cells for death hold great promise in treatment of a wide shRNA plasmids. The shRNA (manufacturer’s #4) used recog- range of cancers. Although the cellular mechanism responsible nizes 607–627 of exon 4 of the stathmin gene. Some cells were is unknown, our data add stathmin depletion to the handful of also transfected with TransSilent empty vector or TransSilent p53 strategies available to block proliferation of those cells lacking a 36 47,48 shRNA plasmids (Panomics; ) using Fugene 6. functional p53. Given that nitrosoureas appear to target stathmin 708 Cancer Biology & Therapy Volume 9 Issue 9 45 and control glioma cell migration, these compounds may be for a generous gift of HCT116 cell lines. Thanks also to Bob useful agents to induce apoptosis in a mutant p53 background. Skibbens, Danielle Ringhoff and Victoria Caruso for their advice during the course of these studies. Supported by grants from NIH Acknowledgements (GM058025) and the Pennsylvania Department of Health to L.C. We are indebted to Dr. Maureen Murphy, Fox Chase Cancer The Pennsylvania Department of Health specicfi ally disclaims Center, for many helpful discussions and to Dr. B. Vogelstein responsibility for any analyses, interpretations or conclusions. 18. Brattsand G. Correlation of oncoprotein 18/stathmin 34. Johnsen JI, Aurelio ON, Kwaja Z, Jorgensen GE, References expression in human breast cancer with established Pellegata NS, Plattner R, et al. p53-mediated negative 1. Jordan MA, Kamath K. How do microtubule-targeted prognostic factors. Br J Cancer 2000; 83:311-8. regulation of stathmin/Op18 expression is associ- drugs work? An overview. Curr Cancer Drug Targets ated with G(2)/M cell cycle arrest. Int J Cancer 2000; 19. Bieche I, Lachkar S, Becette V, Cifuentes-Diaz C, Sobel 2007; 7:730-42. 88:685-91. A, Lidereau R, et al. Overexpression of the stathmin 2. Rieder CL, Maiato H. Stuck in division or passing gene in a subset of human breast cancer. Br J Cancer 35. Ahn J, Murphy M, Kratowicz S, Wang A, Levine through: what happens when cells cannot satisfy the 1998; 78:701-9. AJ, George DL. Downregulation of the stathmin/ spindle assembly checkpoint. Dev Cell 2004; 7:637- Op18 and FKBP25 genes following p53 induction. 20. Price DK, Ball JR, Bahrani-Mostafavi Z, Vachris JC, Oncogene 1999; 18:5954-8. Kaufman JS, Naumann RW, et al. The phosphoprotein 3. Weaver BA, Cleveland DW. Decoding the links Op18/stathmin is differentially expressed in ovarian 36. Giono LE, Manfredi JJ. Mdm2 is required for inhibi- between mitosis, cancer and chemotherapy: the mitotic cancer. Cancer Invest 2000; 18:722-30. tion of Cdk2 activity by p21, thereby contributing to checkpoint, adaptation and cell death. Cancer Cell p53-dependent cell cycle arrest. Mol Cell Biol 2007; 21. Zhang HZ, Wang Y, Gao P, Lin F, Liu L, Yu B, et al. 2005; 8:7-12. 27:4166-78. Silencing stathmin gene expression by survivin promot- 4. Bhat KM, Setaluri V. Microtubule-associated proteins er-driven siRNA vector to reverse malignant phenotype 37. Sur S, Pagliarini R, Bunz F, Rago C, Diaz LA Jr, Kinzler as targets in cancer chemotherapy. Clin Cancer Res of tumor cells. Cancer Biol Ther 2006; 5:1457-61. KW, et al. A panel of isogenic human cancer cells sug- 2007; 13:2849-54. gests a therapeutic approach for cancers with inactivat- 22. Wang R, Dong K, Lin F, Wang X, Gao P, Wei SH, et al. 5. Chin GM, Herbst R. Induction of apoptosis by ed p53. Proc Natl Acad Sci USA 2009; 106:3964-9. Inhibiting proliferation and enhancing chemosensitiv- monastrol, an inhibitor of the mitotic kinesin Eg5, is ity to taxanes in osteosarcoma cells by RNA interfer- 38. Brito DA, Rieder CL. Mitotic checkpoint slippage in independent of the spindle checkpoint. Mol Cancer ence-mediated downregulation of stathmin expression. humans occurs via cyclin B destruction in the presence Ther 2006; 5:2580-91. Mol Med 2007; 13:567-75. of an active checkpoint. Curr Biol 2006; 16:1194- 6. Kouzu Y, Uzawa K, Koike H, Saito K, Nakashima 23. Alli E, Yang JM, Hait WN. Silencing of stathmin D, Higo M, et al. Overexpression of stathmin in oral induces tumor-suppressor function in breast cancer cell 39. Clute P, Pines J. Temporal and spatial control of cyclin squamous-cell carcinoma: correlation with tumour lines harboring mutant p53. Oncogene 2007; 26:1003- B1 destruction in metaphase. Nat Cell Biol 1999; progression and poor prognosis. Br J Cancer 2006; 12. 1:82-7. 94:717-23. 24. Schubart UK, Yu J, Amat JA, Wang Z, Hoffmann MK, 40. Borgne A, Meijer L. Sequential dephosphorylation of 7. Mistry SJ, Bank A, Atweh GF. Targeting stathmin in Edelmann W. Normal development of mice lacking p34(cdc2) on Thr-14 and Tyr-15 at the prophase/meta- prostate cancer. Mol Cancer Ther 2005; 4:1821-9. metablastin (P19), a phosphoprotein implicated in cell phase transition. J Biol Chem 1996; 271:27847-54. 8. Rana S, Maples PB, Senzer N, Nemunaitis J. Stathmin cycle regulation. J Biol Chem 1996; 271:14062-6. 41. Holmfeldt P, Brannstrom K, Stenmark S, Gullberg M. 1: a novel therapeutic target for anticancer activity. 25. Wang Y, Ji P, Liu J, Broaddus RR, Xue F, Zhang Aneugenic activity of Op18/stathmin is potentiated by Expert Rev Anticancer Ther 2008; 8:1461-70. W. Centrosome-associated regulators of the G(2)/M the somatic Q18→e mutation in leukemic cells. Mol 9. Melhem R, Hailat N, Kuick R, Hanash SM. checkpoint as targets for cancer therapy. Mol Cancer Biol Cell 2006; 17:2921-30. Quantitative analysis of Op18 phosphorylation in 2009; 8:8. 42. O’Connell MJ, Cimprich KA. G damage checkpoints: childhood acute leukemia. Leukemia 1997; 11:1690- 26. Larsson N, Melander H, Marklund U, Osterman O, what is the turn-on? J Cell Sci 2005; 118:1-6. Gullberg M. G /M transition requires multisite phos- 43. Budde PP, Kumagai A, Dunphy WG, Heald R. 10. Nylander K, Marklund U, Brattsand G, Gullberg M, phorylation of oncoprotein 18 by two distinct protein Regulation of Op18 during spindle assembly in Roos G. Immunohistochemical detection of oncopro- kinase systems. J Biol Chem 1995; 270:14175-83. Xenopus egg extracts. J Cell Biol 2001; 153:149-58. tein 18 (Op18) in malignant lymphomas. Histochem J 27. Marklund U, Larsson N, Gradin HM, Brattsand G, 44. Wiman KG. Restoration of wild-type p53 function in 1995; 27:155-60. Gullberg M. Oncoprotein 18 is a phosphorylation- human tumors: strategies for efficient cancer therapy. 11. Nakashima D, Uzawa K, Kasamatsu A, Koike H, Endo responsive regulator of microtubule dynamics. EMBO Adv Cancer Res 2007; 97:321-38. Y, Saito K, et al. Protein expression profiling identi- J 1996; 15:5290-8. 45. Liang XJ, Choi Y, Sackett DL, Park JK. Nitrosoureas fies maspin and stathmin as potential biomarkers of 28. Larsson N, Marklund U, Gradin HM, Brattsand G, inhibit the stathmin-mediated migration and invasion adenoid cystic carcinoma of the salivary glands. Int J Gullberg M. Control of microtubule dynamics by of malignant glioma cells. Cancer Res 2008; 68:5267- Cancer 2006; 118:704-13. oncoprotein 18: dissection of the regulatory role of 12. Nishio K, Nakamura T, Koh Y, Kanzawa F, Tamura T, multisite phosphorylation during mitosis. Mol Cell 46. Holmfeldt P, Stenmark S, Gullberg M. Interphase- Saijo N. Oncoprotein 18 overexpression increases the Biol 1997; 17:5530-9. specific phosphorylation-mediated regulation of tubu- sensitivity to vindesine in the human lung carcinoma 29. Ringhoff DN, Cassimeris L. Stathmin regulates cen- lin dimer partitioning in human cells. Mol Biol Cell cells. Cancer 2001; 91:1494-9. trosomal nucleation of microtubules and tubulin 2007; 18:1909-17. 13. Yuan RH, Jeng YM, Chen HL, Lai PL, Pan HW, Hsieh dimer/polymer partitioning. Mol Biol Cell 2009; 47. Piehl M, Cassimeris L. Organization and dynamics of FJ, et al. Stathmin overexpression cooperates with p53 20:3451-8. growing microtubule plus ends during early mitosis. mutation and osteopontin overexpression, and is asso- 30. DeGraffenried LA, Fulcher L, Friedrichs WE, Grunwald Mol Biol Cell 2003; 14:916-25. ciated with tumour progression, early recurrence and V, Ray RB, Hidalgo M. Reduced PTEN expression in 48. Warren JC, Rutkowski A, Cassimeris L. Infection with poor prognosis in hepatocellular carcinoma. J Pathol breast cancer cells confers susceptibility to inhibitors replication-deficient adenovirus induces changes in the 2006; 209:549-58. of the PI3 kinase/Akt pathway. Ann Oncol 2004; dynamic instability of host cell microtubules. Mol Biol 14. Chen G, Wang H, Gharib TG, Huang CC, Thomas 15:1510-6. Cell 2006; 17:3557-68. DG, Shedden KA, et al. Overexpression of oncoprotein 31. Wang NP, To H, Lee WH, Lee EY. Tumor suppressor 49. Garrett S, Auer K, Compton DA, Kapoor TM. hTPX2 18 correlates with poor differentiation in lung adeno- activity of RB and p53 genes in human breast carci- is required for normal spindle morphology and cen- carcinomas. Mol Cell Proteomics 2003; 2:107-16. noma cells. Oncogene 1993; 8:279-88. trosome integrity during vertebrate cell division. Curr 15. Friedrich B, Gronberg H, Landstrom M, Gullberg M, 32. Bunz F, Dutriaux A, Lengauer C, Waldman T, Zhou Biol 2002; 12:2055-9. Bergh A. Differentiation-stage specific expression of S, Brown JP, et al. Requirement for p53 and p21 to 50. Bradford MM. A rapid and sensitive method for the oncoprotein 18 in human and rat prostatic adenocarci- sustain G arrest after DNA damage. Science 1998; 2 quantitation of microgram quantities of protein utiliz- noma. Prostate 1995; 27:102-9. 282:1497-501. ing the principle of protein-dye binding. Anal Biochem 16. Belletti B, Nicoloso MS, Schiappacassi M, Berton S, 33. Koivusalo R, Krausz E, Helenius H, Hietanen S. 1976; 72:248-54. Lovat F, Wolf K, et al. Stathmin activity influences Chemotherapy compounds in cervical cancer cells sarcoma cell shape, motility and metastatic potential. primed by reconstitution of p53 function after short Mol Biol Cell 2008; 19:2003-13. interfering RNA-mediated degradation of human pap- 17. Ngo TT, Peng T, Liang XJ, Akeju O, Pastorino S, illomavirus 18 E6 mRNA: opposite effect of siRNA Zhang W, et al. The 1p-encoded protein stathmin and in combination with different drugs. Mol Pharmacol resistance of malignant gliomas to nitrosoureas. J Natl 2005; 68:372-82. Cancer Inst 2007; 99:639-52. www.landesbioscience.com Cancer Biology & Therapy 709

Journal

Cancer Biology & Therapy – Taylor & Francis

Published: May 1, 2010

There are no references for this article.