|Home | About | Journals | Submit | Contact Us | Français|
Survivin is a member of the inhibitor of apoptosis protein family and has an essential role in mitosis. Survivin is overexpressed in a large variety of human cancers and represents an attractive target for cancer therapy. Epidermal growth factor receptor and Her/neu-transformed human tumors in particular exhibit high levels of survivin. The survivin protein forms dimers through a conserved region that is critical for subcellular localization and biological functions of the protein. We identified small molecules that target a specific cavity adjacent to the survivin dimerization surfaces. S12, a lead compound identified in the screen, can bind to the survivin protein at the intended target site. Moreover, S12 alters spindle formation, causing mitotic arrest and cell death, and inhibits tumor growth in vitro and in vivo. Cell death occurs in premetaphase stage following mitotic arrest and is not a consequence of general toxicity. Thus, the study validates a novel therapeutic target site in the survivin protein and provides a promising strategy to develop a new class of therapeutic small molecules for the treatment of human cancers.
Survivin was initially identified as a member of the inhibitor of apoptosis protein family (Li et al., 1998). Overexpression of survivin is observed in a large portion of human cancers and, in many cases, correlates with poor prognosis (Altieri, 2003). Survivin appears to have two distinct biological roles. First, survivin may exhibit anti-apoptotic activity only under certain experimental conditions although this is controversial (Li et al., 1998, 1999; Tamm et al., 1998). Elevation of survivin levels have been implicated in radioresistance in cancers (Rodel et al., 2005; Wang and Greene, 2005), whereas downregulation of survivin sensitizes cancer cells to genotoxic agents. The mechanistic role of survivin in apoptosis is currently unclear.
Second, more recent studies indicate that survivin is involved in chromosome segregation during mitosis (Skoufias et al., 2000; Uren et al., 2000). Survivin acts as an integral component of the chromosome passenger protein complex, which also includes Aurora kinase B, INCENP, TD-60 and Borealin. Survivin is localized to the kinetochore following chromosome condensation and becomes relocated to the midbody after chromosome segregation and until cytokinesis (Skoufias et al., 2000; Uren et al., 2000). Survivin has an essential role in modulating the formation of the mitotic spindle, which modulates accurate chromosome segregation during mitosis. Overexpression of survivin might corrupt the spindle checkpoint control and contribute to chromosome instability and aneuploidy.
Loss of survivin functions may affect microtubule dynamics during mitosis. siRNA depletion of survivin increased the number of microtubules nucleated at centrosomes and promoted microtubule instability (Rosa et al., 2006). Cells microinjected with a polyclonal antibody to survivin exhibit defective spindles and depletion of microtubules (Giodini et al., 2002). Clearly, the chromosomal passenger complex is required for microtubule stabilization and spindle assembly during mitosis (Sampath et al., 2004).
Our previous studies indicated that activation of the ErbB family receptor tyrosine kinases, such as epidermal growth factor receptor, leads to upregulation of survivin (Wang and Greene, 2005). Modulation of survivin levels by epidermal growth factor receptor is dependent on the PI-3 kinase pathway but not the mitogen-activated protein kinase pathway. Similarly, the signaling events initiated by p185Her2/neu and ErbB3 also cause increase of survivin levels (Asanuma et al., 2005; Xia et al., 2006). In some cells, disabling p185Her2/neu has been suggested to facilitate cell death in a survivin-dependent manner (Xia et al., 2006).
The structure of survivin has been determined crystallographically (Chantalat et al., 2000; Verdecia et al., 2000) and by nuclear magnetic resonance (NMR) (Sun et al., 2005). The N-terminal region of survivin contains a zinc-binding fold similar to the classic baculovirus inhibitor of apoptosis protein repeat (BIR) motif, which binds to phosphorylated histone and is involved in recruiting the chromosome passenger complex (CPC) proteins to the chromosome (Kelly et al., 2010; Wang et al., 2010; Yamagishi et al., 2010). The survivin BIR domain consists of a three-stranded b-sheet and four a-helices (Chantalat et al., 2000; Verdecia et al., 2000) and the survivin protein forms a dimer that resembles a ‘bow tie’. The N-terminal region contains the structural elements involved in dimerization and subcellular localization to the kinetochore and the midbody (Li and Ling, 2006). In addition, mutations of the aminoacid residues in this region affects the function of survivin (Li et al., 1998). Ubiquitination of survivin in the N-terminal region modulates localization as well as degradation (Vong et al., 2005). A number of survivin splice variants have been identified, such as survivindelta Ex3 lacking exon 3 and survivin-2B retaining a part of intron 2 as a cryptic exon. Of note, the N-terminal region, which we have focused on, is shared by the variant forms (Li and Ling., 2006; Noton et al., 2006).
More recently, the structure of a survivin-borealin-INCENP core complex has been defined (Jeyaprakash et al., 2007). Borealin and INCENP associate with the c-terminal helical domain of survivin to form a three-helical bundle of 1:1:1 stoichiometry. The interactions of the core components are essential for central spindle and midbody localization of the complex. Both survivin and borealin bind to the N-terminus of INCENP (corresponding to the amino-acid residues 1–58 of human INCENP), which is sufficient for targeting to the centromere (Ainsztein et al., 1998). Survivin is required for localization of the chromosome passenger protein complex to the centromere.
The dual role of survivin in influencing some forms of apoptosis and as a critical element in mitosis makes it an attractive target for cancer therapy. Resistance to apoptosis is seen in most human cancers. Developing molecular interventions to commit tumor cells to the cell death pathways is therefore a general strategy for pharmaceutical therapy (Nicholson, 2000). Moreover, one of the widely explored approaches to develop cancer therapeutics is to induce aberrant mitosis in tumor, which causes cell death. For example, taxanes, a class of compounds that blocks mitosis by stabilizing tubulin polymerization and interfering with the formation of the mitotic spindle, have been used to treat various forms of human cancer. However, because microtubule polymerization is involved in a large variety of physiological processes, these drugs cause profound side effects.
Survivin is selectively overexpressed in many human cancers but its levels are low in normal tissues and undetectable in non-dividing cells (Wang and Greene, 2005). Indeed, ablation of survivin function can cause mitotic arrest and death to dividing cells, including a variety of cancer cell lines. Therefore, small molecules that target and disable survivin function represent a rational targeted-therapeutic approach that interferes with both apoptosis and mitosis in cancer cells.
In this study, we use small molecules to probe a discreet survivin function. Using a recently developed algorithm that identifies cavity interacting small molecules at critical allosteric sites, we successfully used the compounds to disrupt survivin functions related to metaphase. We have performed biological studies to examine the effects of these molecules on cell viability, mitosis and in vivo tumor growth. Our results identify a new targeted therapeutic approach to affect metaphase mitotic events in human cancers.
We chose the N-terminal region of survivin protein as this region includes the structural elements important for dimerization, subcellular localization and the biological functions. The ‘cavity-induced allosteric modification’ (CIAM) approach has been used to identify small molecules that modulate protein function by inducing allosteric conformational changes (Murali et al., 2005). We have applied this algorithm to perform a highly targeted in silico screen of small molecules that can bind survivin in a cavity close to the dimeric interface, which includes an extensive hydrophobic contact formed between Leu 98 from one monomer and Leu 6, Trp 10, Phe 93, Phe 101 and Leu 102 from the neighboring monomer (Verdecia et al., 2000). We hypothesized that binding of the small molecules may produce allosteric conformational changes that disable the normal functions of the survivin dimer (Figure 1). Using the CIAM algorithm, we identified a number of small molecules having the potential to fit into the cavity and producing conformational changes that may disrupt the functions of survivin (Figure 1).
Our chemical biology and algorithm screens identified a compound, designated S12, which we have shown to bind to survivin and exhibit promising biological activities. Moreover, we have obtained three additional candidate molecules. These molecules are structurally distinct from S12 or its analogs, but showed the same potency to cause mitotic arrest or bind to the survivin protein. The biophysical and biological properties of these molecules will be reported elsewhere. Here we concentrate on the data of the S12 molecule because of its superior solubility and its favorable biological properties.
We evaluated the binding properties of the surviving-targeting molecules to survivin by isothermal titration calorimetry (ITC). S12 bound to purified recombinant survivin (Figure 2a, Ka=447±3.65 × 106/m). The ITC profile shows binding isotherm that is consistent with a competing equilibrium between individual S12 molecules (Figure 2a). Fitting of the heat profile is consistent with a 1:1 stoichiometry (Figure 2a). In addition, we found that S10 can also bind to survivin in ITC assays (Supplementary information, Figure S2). Another S12 analog, namely S2, is significantly less soluble and showed modest binding affinity.
To assess the atomic specificity of the binding site, we generated point mutations on the survivin protein at the site presumably targeted by S12. The two mutations were substitutions of phenylalanine 86 with alanine (F86A) and valine 89 with tyrosine (V89Y). As mentioned above, the targeted cavity is adjacent to the interface involved in dimerization of the survivin protein. On the basis of structural analyses, we predicted that alterations of either of these two amino-acid residues would abolish the binding of S12 without affecting the overall structure of the survivin protein. The mutant survivin protein, as well as the native form, were prepared and purified from a bacterial expression system. ITC analyses showed that, in contrast to the native survivin protein, the mutants Sur-F86A and Sur- V89Y did not bind to the survivin-targeting compound S12 at all (Figure 2a).
To rule out the possibility that the mutations abolished binding to S12 as a result of protein misfolding, we used circular dichroism spectroscopy to analyze the global structural features of the survivin mutant proteins. Consistent with our prediction, the circular dichroism spectra of the mutant proteins matched well with that of the native protein (Figure 2b), indicating that both Sur(F86A) and Sur(V89Y) mutants were properly folded and that no global structural changes occurred as a consequence of the mutations.
The candidate survivin-targeting molecules were examined for the ability to inhibit cell proliferation and mitosis. HeLa cells were treated with the compounds at various concentrations for 15 h followed by examination of the DNA content by fluorescence-activated cell sorting analysis. We found that S12 led to accumulation of cells in the G2/M stage of the cell cycle (Figure 3). In addition, we observed an increase of cells in the sub-G1 population, which is indicative of cell death. S12 affected cell proliferation and viability, consistent with the anticipated outcome caused by ablation of survivin functions. We extended these studies to test a number of molecules that are structurally similar to S12, including S2 and S10 (Supplementary information, Figure S1). As expected, these S12 analogs also exhibit the ability to arrest the cells at the G2/M stage (Supplementary information, Figure S3).
Effects of the survivin-targeting compound on cell cycle progression were examined by immunofluorescence microscopy. HeLa cells were plated on cover glass and treated with S12 (40 µm) or with the solvent dimethyl sulfoxide (DMSO) alone for 15 h. Cells were stained with anti-aurora B antibody or antisurvivin antibody. Upon S12 treatment, it was observed that cells accumulate in a stage with condensed chromosomes, but without proper alignment of chromosomes on the mitotic plate expected at the metaphase (Figure 3b). The percentage of cells that progress to metaphase or beyond was reduced by S12 treatment (Figure 3c).
In the presence of the survivin-targeting compound, multiple aster-like microtubule bundles formed in each mitotic cell (Supplementary information, Figures S3 and S4A). These microtubules are typically very short, with one end attached to the centrosome and the other end to the kinetochore of the chromosomes. S12 can cause defects in the formation of the mitotic spindle. Cells were arrested by S12 in an early stage of mitosis. However, S12 did not alter the localization of survivin or Aurora B to the kinetochore (Figure 3b). Moreover, localization of survivin to the midbody was also not affected (Supplementary information, Figure S3B). Therefore, the predominant effect is at pre-metaphase related events during mitosis.
The survivin-targeting molecules identified by the CIAM screen were tested for the ability to inhibit cell proliferation by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. AsPC1 pancreatic cancer cells were treated with DMSO or the survivin-targeting compound S12, S2 or S10 at various concentrations for 48 h. The number of viable cells was evaluated by MTT. The data show that the survivin-targeting compounds can inhibit cell proliferation and reduce cell viability in a dose-dependent manner (Figure 4a). Similarly, the survivin antagonists can inhibit proliferation of a variety of other human cancer cells, including HeLa cervical cancer cell (Figure 4b), SKBR3 breast cancer cell (Figure 4c) and DAOY medulloblastoma cell (Figure 4d). Inhibition of cell proliferation by the survivin-targeting compound is independent of p53 status. S12 is equally effective at reducing cell viability in either p53+/+ or p53−/− HCT116 cells (Figure 4e). In addition, S12 also inhibited proliferation of immortalized fibroblast cells, which is consistent with the essential role of survivin in normal proliferating cells.
We examined whether S12 disabled survivin would lead to apoptosis of cells. HeLa cells were incubated with S12 or with DMSO alone for 12 h. To compare the effect with that of other therapeutic agents, HeLa cells were treated with taxol, nocodazole or etoposide, respectively. The cell lysates were collected and subjected to western blot analysis using an anti-poly-(ADP-ribose)-polymerase antibody. As shown in Figure 5, cleavage of poly-(ADP-ribose)polymerase, which is mediated by caspase 3 has been established as a hallmark of caspase activation during apoptosis (Boulares et al., 1999; Nicholson et al., 1995; Rosenthal et al., 1997), was detected following S12 treatment (Figure 5). We also re-probed the blot with an anti-survivin antibody. S12 did not significantly affect the overall survivin levels (Figure 5a). These collective observations indicate that the survivin-targeting molecule S12 blocks mitosis and leads to apoptosis.
We wanted to determine if the survivin-targeting compounds induce cell death during mitosis but could not affect the phenotype or influence in other stages of the cell cycle. To this end, we blocked HeLa cells in the S phase by thymidine treatment and examined whether the cells were susceptible to cell death induced by S12. Our results show that when the cells were arrested in S phase, they were insensitive to S12 (Figure 5b). In contrast, unsynchronized cells underwent apoptosis in the presence of the survivin-targeting compound (Figure 5b). Moreover, S12 did not affect DNA synthesis (Supplementary information, Figure S4), which substantiates that the effect of the surviving-targeting molecule is restricted to mitosis. Thus, the current data support the notion that S12-induced death of cells is a consequence of mitotic arrest rather than general cytotoxicity and indicate a unique cell cycle specific functionality of survivin in this process.
The effect of S12 on tumor growth was examined using a mouse xenograft model. AsPC1 human pancreatic cells were injected subcutaneously into athymic mice and formed palpable tumors within 10 days. S12 was then administered at the concentration of 5 or 15 mg/kg three times per week for up to 48 days after the initial grafting. The size of the tumors was monitored. Our results indicate that S12 treatment significantly reduced the tumor volume in a dose-dependent manner (Figure 6a, left panel). In a separate set of studies, we raised the dose of S12 to 50 mg/kg and observed a modest enhancement of inhibition of tumor growth (Figure 6a, right panel). In all these studies, we observed no overtly toxic effects to the mice.
We performed immunohistochemistry studies to analyze the cellular effects of S12 treatment on the tumors. Using Ki67 as a marker for cell proliferation, we found that S12 drastically reduced proliferation of tumor cells (Figure 6b). In addition, we examined apoptosis using the TUNEL (terminal deoxynucleotidyl transferase dUTP nick end labeling) assay. Consistent with the findings of the in vitro studies, S12 treatment increased the levels of apoptosis in the tumors (Figure 6b).
We used the CIAM algorithm to identify allosteric modulators that target a survivin structural site important for cellular functions. Our studies identified a cavity that is linked to an allosteric locus in the survivin molecule. The small molecules we selected target a cavity adjacent to the structural features critical to dimerization and subcellular localization (Li and Ling, 2006). Mutations of the amino-acid residues close to in this region affect other functions associated with survivin (Li et al., 1998). Our ITC analyses confirmed that the small molecules isolated in the virtual screen bind to the survivin protein at this site. Our mutagenesis studies corroborated the specificity of these compounds. S12 can bind to the native survivin protein but not to the mutants that bear small structural changes in the site targeted by the small molecules. Because the targeted site is shared by all the variant forms (Li and Ling, 2006; Noton et al., 2006), it is likely that these small molecules target all the survivin variants. The compounds are highly selective and disable a survivin functionality important for mitotic metaphase. Our results show that the designed compound S12 and its analogs disrupt spindle formation and cause mitotic arrest. These observations indicate that survivin is involved in the regulation of microtubule dynamics in dividing cells (Giodini et al., 2002; Rosa et al., 2006). Chromosomal passenger proteins, which include survivin, are required for microtubule stabilization and spindle assembly during mitosis (Sampath et al., 2004).
Survivin-binding molecules have been identified by Wendt et al. (2007) in a high-throughput, affinity-based NMR screen. Of note, the NMR studies indicate that these molecules bind to survivin at the region that partly overlaps with our targeting site developed using the CIAM approach. However, none of the compounds identified by the NMR screen displayed biological activities (Wendt et al., 2007). According to the structural data obtained by NMR, the compounds bind to the homodimeric survivin complex in the intermolecular space between the two monomers and make molecular contacts with both interacting monomers. This binding mode suggests that instead of inhibiting dimerization, the compounds identified by NMR-based screen may actually stabilize the dimer by creating additional bonds between the monomers (Wendt et al., 2007). Importantly, it should be noted that the compounds described by Wendt et al. (2007), as well as the NMR structure of survivin, were obtained using a truncated form of the protein, which lacks 20 aminoacid residues in the C-terminus. Although the overall folding of the protein is not affected by the deletion, a close comparison of the truncated and the full-length survivin proteins revealed significant differences in conformation, especially in the binding pocket for the small molecules we developed using the CIAM method (Supplementary information, Figure S2B). On the basis of the structure of the full-length protein, we have identified at least four molecules using the CIAM algorithm and demonstrated that these compounds can cause phenotypes indicative of loss of specific survivin functions.
In cells treated with S12 or its analogs, the spindle does not form properly. Instead, multiple short, asterlike microtubule bundles can be frequently observed. These structures resemble the kinetochore fibers reported in previous studies (Goshima and Vale, 2003; Khodjakov et al., 2003, Maiato and Sunkel, 2004). It was proposed that, in normal cell division, the microtubule arrays associated with the kinetochore make physical contact with the microtubule network that arises from the MTOC and, thereby, establish the attachment of the chromosomes to the mitotic spindle (Goshima et al., 2005;Maiato et al., 2004; Rieder, 2005). Our immunofluorescence studies showed that inhibition of survivin by S12 alters the dynamics of microtubule formation at the MTOC and the kinetochore. Moreover, the chromosomes were not properly aligned to the metaphase plate. S12 appears to perturb the microtubule dynamics and activate the mitotic checkpoint.
Survivin is an essential gene for dividing cells. Hence, it is not possible to test specificity and toxicity in a complete survivin-null background. Nevertheless, we demonstrated that the anti-proliferation effects of S12 are the consequence of inhibition of mitosis, rather than general toxicity. First, cell death caused by both S12 accompanies mitotic arrest, whereas cells blocked in S phase are resistant to the treatment. The current data indicate that the pro-apoptotic effect of the surviving-targeting molecule S12 is cell cycle specific and accompanies mitotic arrest. Second, S12 does not affect other fundamental S phase processes, such as DNA synthesis. These unique compounds specifically target a mitotic checkpoint in dividing cells and identify a cell cycle specific-function of survivin in this process. Moreover, our in vitro studies indicate that S12 does not cause discernible toxic effects on non-proliferating cells and in normal resting cells.
The survivin-targeting molecules inhibit cell proliferation independently of p53 status. Ablation of survivin by siRNA also blocks cell proliferation in both p53-positive and p53-negative cells (Kappler et al., 2004). Collectively, these results imply that this class of surviving-targeting therapeutics may be useful to treat cancer regardless of p53 mutations, which are frequently associated with malignancy.
In vivo the survivin-targeting molecule S12 effectively inhibits tumor growth and appears to be well tolerated. We have observed no systematic toxicity in the animals despite prolonged treatment. Survivin is selectively overexpressed in cancer cells and to a much lesser degree in normal dividing cells. Survivin is not required for non-dividing cells. Clearly, optimization is required to increase the affinity and efficacy of these surviving-targeting molecules. Nevertheless, the current data establish a principle that targeting the survivin protein using small molecules can be used as a general strategy to target mitosis-related events in many human cancers.
The compounds have been designed using a stepwise procedure that identifies pseudo-allosteric cavities used to induce allosteric modification (CIAM) as described previously (Murali et al., 2005). The procedure involved identification of the pseudo-allosteric cavity, virtual screening using DOCK 4 (DesJarlais et al., 1988) and molecular simulations using the DISCOVER function in INSIGHTII (Accelrys Inc., San Diego, CA, USA).
The human embryonic kidney 293T cells, cervical cancer HeLa cells, osteosarcoma U2OS cells, breast cancer SKBR3 cell and medulloblasoma DAOY cells were cultured in Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal bovine serum. The following antibodies were used in this study: monoclonal anti-survivin antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA), monoclonal anti-tubulin antibody and monoclonal anti-γ-tubulin antibody (Sigma, St Louis, MO, USA), fluorescein isothiocyanate- or Cy3-conjugated anti-mouse immunoglobulin antibodies and the fluorescein isothiocyanate- or Cy3-conjugated anti-rabbit immunoglobulin antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA).
The plasmid pGEX-survivin was a kind gift of Dr Robert L Margolis (Sanford-Burnham Medical Research Institute, La Jolla, CA, USA). The point mutations were generated by using the QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA). The recombinant glutathione S-transferase-survivin fusion proteins were expressed in Escherichia coli strain BL21(DE3). Protein expression was induced with isopropyl β-D-1-thiogalactopyranoside for 3 h at 30 °C. The cell lysate was prepared by sonication in phosphate buffered saline (PBS) supplemented with the Complete Mini protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN, USA) and 1 mm Dithiothreitol. The glutathione S-transferase-survivin protein was purified with glutathione Sepharose beads. The glutathione S-transferase tag was removed by incubation with factor Xa in digestion buffer. Factor Xa was then removed by using the Xa removal resin (Qiagen, Valencia, CA, USA). The purified proteins were analyzed size exclusion by sodium dodecyl sulfate–polyacrylamide gel electrophoresis-PAGE and high-performance liquid chromatography. For ITC or circular dichroism spectroscopy, the proteins were dialyzed in 20 mm phosphate buffer containing 10 µm ZnSO4 and 1 mm 2-mercaptoethanol. ITC studies were performed using the VP-ITC Microcalorimeter (Microcal, Century City, CA, USA).
Cells seeded on cover glass were fixed in buffer 1 (3% paraformaldehyde in PBS) at room temperature and subsequently with cold methanol. Permeablization was performed using buffer 2 (0.5% Triton X-100 in 20 mm 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, pH 7.5 and 50 mm NaCl, 3 mm MgCl2). The cells were then placed in blocking buffer 3 (buffer 2 with 0.1% Triton X-100 and 2% bovine serum albumin) for 15 min, incubated with the primary antibody in buffer 3 for 30 min and washed four times with buffer 4 (20 mm 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, pH 7.5 and 50 mm NaCl, 3 mm MgCl2). After incubation with the secondary antibody and subsequent washes, the cells were mounted in VectaShield medium containing 4′,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA, USA). The slides were analyzed by fluorescence microscopy (Leica DMIRBE inverted microscope (Leica Microsystems, Buffalo Grove, IL, USA) and Openlab 3.1.4 software (PerkinElmer, Waltham, MA, USA)).
To analyze the DNA content, cells were stained in PBS containing 60 µg/ml propidium iodide (Sigma) and 10 µg/ml RNase (Roche Applied Science) for 30 min at 4 °C. Cell cycle analysis was performed using a Becton Dickinson fluorescence-activated cell analyzer and the CellQuest version 1.2 software (Becton Dickinson Immunocytometry Systems, Mansfield, MA, USA). At least 10 000 cells were analyzed for each sample. The percentage of cells in each stage of the cell cycle was calculated using the ModFit LT version 3.1 Software (Verity Software House Inc., Topsham, ME, USA).
Cell proliferation was analyzed by MTT incorporation assay. Cells were plated on 96-well plates and treated with either the compounds or the solvent DMSO alone. MTT was added to the cells for 2 h. Then cells were lysed in 20% w/v sodium dodecyl sulfate/50% N,N-Dimethylformamide, pH4.7 and maintained at 37 °C overnight. Proliferation was assessed by measuring optical density readings at 570nm using the Spectra Fluor ELISA reader (Tecan, Durham, NC, USA). Percentage viability was calculated by using DMSO-treated cells as 100%. The error bars represent s.d.
The cells were incubated medium containing 10uM BrdU for 60 min. Single-cell suspension was then prepared and the cells were fixed with cold methanol with vigorous mixing. The DNA was denatured by incubation of cells in 2N HCl for 20 min. The pH was neutralized by adding 5N NaOH. The cells were subsequently washed with PBS and incubated in blocking solution (3% bovine serum albumin in PBS). The fluorescein isothiocyanate-conjugated anti-BrdU antibody was added. Fluorescence-activated cell sorting analysis was performed after washing cells with PBS.
The University of Pennsylvania Institutional Animal Care and Use Committee has approved this protocol. NCr homozygous athymic (nude) female mice (6–8 weeks old) were purchased from the National Cancer Institute. Overall, 4 × 106 AsPC1 cells were suspended in 100 µl of PBS and injected subcutaneously into a flank of each animal. Vehicle (2.5% DMSO in PBS) or S12 (5, 15 or 50 mg/kg) was administered intraperitoneally starting on day 10 after the development of small palpable tumors and then three times a week for the duration of the experiment. Tumors were measured with calipers in three directions to calculate tumor growth at the time of each treatment. Administration of vehicle had no effect on tumor growth. Student’s t-test was used for statistical analysis.
Tumors were harvested, fixed in 10% buffered formalin overnight, dehydrated and paraffin embedded using the standard methods for histological analysis. Following rehydration of tumor sections, antigens were unmasked using Antigen Unmasking Solution (Vector Labs, Burlingame, CA, USA) and endogenous peroxidase activities were blocked with hydrogen peroxide. To examine proliferation, tumor sections were stained by standard immunohistochemical techniques using an antibody against Ki67 (1:50, Abcam, Cambridge, MA, USA) overnight at 4 °C. Detection was performed by incubation with a diluted biotinylated anti-rabbit antibody (Vector Labs) at 1:200 for 1 h at 25 °C and visualized using the avidin–biotin complex method and 3,30-diaminobenzidine (Vector Labs) according to the manufacturer’s recommendations. Sections were counterstained with Harris’s hematoxylin (Fisher Scientific, Pittsburgh, PA, USA). Negative controls (rabbit serum) did not produce a stain. To examine apoptosis, tumor sections were processed using the TACS-XL in situ Apoptosis Detection Kit (R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s recommendations.
We thank Dr Robert L Margolis (the Sanford-Burnham Medical Research Institute, La Jolla, CA, USA), and Mark I. Greene (Grant # R01 CA089481-10) for providing the pGEX-survivin plasmid. This work was supported by grants from the Abramson Family Cancer Research Institute of the University of Pennsylvania, The National Cancer Institute, the Breast Cancer Research Foundation and the Nidus Laboratories, Inc.
Conflict of interest
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)