Moguntinones—New Selective Inhibitors for the Treatment of Human Colorectal Cancer

Published OnlineFirst April 17, 2014; DOI: 10.1158/1535-7163.MCT-13-0224

Molecular Cancer Small Molecule Therapeutics Therapeutics

Annett Maderer1, Stanislav Plutizki2, Jan-Peter Kramb2, Katrin Gopfert€ 1, Monika Linnig1, Katrin Khillimberger1, Christopher Ganser2, Eva Lauermann2, Gerd Dannhardt2, Peter R. Galle1, and Markus Moehler1

Abstract 3-Indolyl and 3-azaindolyl-4-aryl maleimide derivatives, called moguntinones (MOG), have been selected for their ability to inhibit protein kinases associated with angiogenesis and induce apoptosis. Here, we characterize their mode of action and their potential clinical value in human colorectal cancer in vitro and in vivo. MOG-19 and MOG-13 were characterized in vitro using kinase, viability, and apoptosis assays in different human colon cancer (HT-29, HCT-116, Caco-2, and SW480) and normal colon cell lines (CCD-18Co, FHC, and HCoEpiC) alone or in combination with topoisomerase I inhibitors. Intracellular signaling pathways were analyzed by Western blotting. To determine their potential to inhibit tumor growth in vivo, the human HT-29 tumor xenograft model was used. Moguntinones prominently inhibit several protein kinases associated with tumor growth and metastasis. Specific signaling pathways such as GSK3b and mTOR downstream targets b were inhibited with IC50 values in the nanomolar range. GSK3 signaling inhibition was independent of KRAS, BRAF, and PI3KCA mutation status. While moguntinones alone induced apoptosis only in concentrations >10 mmol/L, MOG-19 in combination with topoisomerase I inhibitors induced apoptosis synergistically at lower concentrations. Consistent with in vitro data, MOG-19 significantly reduced tumor volume and weight in combination with a topoisomerase I inhibitor in vivo. Our in vitro and in vivo data present significant proapoptotic, antiangiogenic, and antiproliferative effects of MOG-19 in different human colon cancer cells. Combination with clinically relevant topoisomerase I inhibitors in vitro and xenograft mouse model demonstrate a high potency of moguntinones to complement and improve standard chemotherapy options in human colorectal cancer. Mol Cancer Ther; 13(6); 1399–409. 2014 AACR.

Introduction ing structural features of 3 natural compounds (combre- Targeting of tyrosine kinases in angiogenesis and other tastatin, rebeccamycin, staurosporin). They showed inhibition of the protein kinases VEGFR-2/3, PDGFRb, FLT3, cellular signaling pathways has proved to be effective in b advanced or metastatic human cancer (1). Inhibition of and GSK3 , antiangiogenic efficacy, and antiproliferative VEGFR1–3 and its ligands in the angiogenic pathway by activity. In addition, they induce apoptosis in FLT3-ITD– different classes of inhibitors is an integral part of the dependent acute myeloblastic leukemia (AML) cell lines clinical armamentarium (2). VEGFs induce angiogenesis and patient blasts at low micromolar concentrations (7). by binding to its cognate receptors VEGFR1–3. Intracel- While key function of VEGF and platelet-derived growth factor (PDGF) receptors may be clearly attributed to lular signal transduction results in endothelial cell pro- b liferation, migration, and new vessel formation (3). angiogenesis, GSK3 regulates proliferation in a cell- Recently, we have developed and characterized a novel type–specific manner. One of the important functions of GSK3b is the regulation of transcription factor b-catenin. class of selective VEGFR-2/3 inhibitors (4–6). These novel b b indolyl and azaindolyl-aryl-maleimides, called mogunti- Phosphorylation of -catenin by a GSK3 protein complex nones, are synthetically designed small molecules, includ- leads to its ubiquitination and proteosomal degradation, thereby abolishing its active role in cell proliferation. GSK3b can be inactivated by protein Disheveled if the Authors' Affiliations: Departments of 1Internal Medicine I and 2Pharmacy Wnt signaling pathway is activated resulting in cell pro- and Biochemistry, Johannes Gutenberg University of Mainz, Mainz, b Germany liferation (8). However, until now, the role of GSK3 inhibition (9) is still controversial in a multiplicity of Note: Supplementary data for this article are available at Molecular Cancer Therapeutics Online (http://mct.aacrjournals.org/). indications and is under investigation (10, 11). FLT3, also known as CD135, binds FLT3 ligands and in turn activates Corresponding Author: Markus Moehler, University Medical Centre of the Johannes Gutenberg University Mainz, First Department of Internal Medicine, signal transduction molecules regulating cell survival and Langenbeckstrasse 1, Mainz 55101, Germany. Phone: 49-6131-176076; proliferation in a cell-type–specific manner. While func- Fax: 49-6131-176621; E-mail: [email protected] tion of FLT3 in colorectal cancer remains unclear, clinical doi: 10.1158/1535-7163.MCT-13-0224 interest for treatment is based on its immunologic func- 2014 American Association for Cancer Research. tion (12). Injection of FLT ligands was demonstrated to

www.aacrjournals.org 1399

Maderer et al.

A MOG-19 MOG-13

H H O N O O N O

O O

N N N O Figure 1. Characterization of MOG- H O H O O 19 and MOG-13. A, chemical structures of MOG-19 and MOG- 13. B, HET-CAM assay showed a B Control MOG-19 (82%) MOG-13 (61%) signiﬁcant inhibition of vessel growth for MOG-19 and MOG-13 in comparison with the control.

mobilize dendritic cells resulting in antitumor activity of ferent specificity and the possibility of combination with the immune system in patients with colon cancer. chemotherapy (20). We have selected moguntinones for their potential to In this article, we describe the effects of moguntinones effectively serve a dual role: inhibition of VEGFR-2/3 as on cell signaling and apoptosis in different human colon well as cell signaling downstream of GSK3b. On the basis cancer cell lines versus normal human epithelial cells, as of our preselection process of more than a hundred sub- well as in vivo significant antiangiogenic properties in stances (4–6), 2 moguntinones (MOG-13 and MOG-19) subcutaneously grown human colon tumors. For summa- were chosen for further characterization in vitro and in vivo ry of key features, see Fig. 1 and Table 1. with respect to their potential to treat human colorectal cancer. Materials and Methods Other tested small-molecule tyrosine kinase inhibitors Cell culture and authentication of cell lines (TKI) to VEGFR and PDGFR like sunitinib or sorafenib, The human colorectal cancer cell lines HT-29, HCT-116, although being effective in some indications, cause side Caco-2, and SW480 were obtained from the German effects as hand–foot syndrome and general fatigue (13). Resource Centre for Biological Material (DSMZ, Braunsch- Such side effects were also observed with, for instance, the weig, Germany, in 2006–2009). Caco-2 was cultured in novel multikinase inhibitor regorafinib (VEGFR, TIE-2) in colorectal cancer (14–16).

Especially in this indication, effective treatment of met- Table 1. IC50 values of MOG-19 and MOG-13 astatic disease has also been shown by blockage of intracellular signaling pathways promoting cell growth, for MOG-19 MOG-13 example, by inhibiting EGF receptor (EGFR), KRAS, and TargetAV (nmol/L) (nmol/L) PI3K/mTOR-dependent signaling cascades (17). Howev- er, EGFR blocking by monoclonal antibody, for example, Receptors cetuximab or panitumumab, is limited in clinical use for FLT3 18 132 KRAS wild-type patients and also by severe side effects VEGFR-2 70 37 (18). The third approved targeted therapy for colorectal Signaling cascade b > cancer is bevacizumab, a VEGF-binding antibody. Recent- GSK3 71 10.000

ly, continuation of bevacizumab after first progression NOTE: IC50 values of MOG-19 and MOG-13 were carried out plus standard second-line chemotherapy proved the clin- at Millipore (http://www.millipore.com/life_sciences/flx4/ ical efficiency of continuous angiogenesis inhibition (19). ld_kinases) by a kinase assay (in nmol/L). MOG-19 is known

With respect to the work presented here, it is of special as a potent VEGFR-2/3 inhibitor (IC50: 70/5 nmol/L) as well as interest that several studies on combination of chemo- FLT3 and GSK3b inhibitor and a moderate PDGFRb inhibitor

therapy and different TKI therapy have conﬁrmed a lack (IC50, 680 nmol/L). With the exception of GSK3b, the same is of efﬁcacy or increased toxicity. Taken together, this true for MOG-13. demonstrates the need of novel TKI inhibitors with dif-

1400 Mol Cancer Ther; 13(6) June 2014 Molecular Cancer Therapeutics

Moguntinones—New Selective Inhibitors

McCoy’s 5A medium (Invitrogen) supplemented with 20% Apoptosis detection was done by nuclear staining with heat-inactivated fetal calf serum (FCS; PAA), 100 units/mL propidium iodide (PI; ref. 22) to measure the sub-G1 cells. penicillin, and 100 mg/mL streptomycin (1%, Cambrex). Cells were incubated with MOG-19 or MOG-13 at the The other cell lines were cultured in RPMI-1640 (Invitro- required dose alone and in combination with cytostatic gen) supplemented with 10% FCS and 1% penicillin/ drugs such as Irinotecan (0.8 mg/mL) and topotecan (20 streptomycin. The normal human epithelial colon cell lines nmol/L) for 7 days. Stained cells were measured by flow FHC and HCoEpiC were obtained from American Type cytometric analysis (FACS) by a BD FACS Calibur. Each Culture Collection (ATCC; CRL-1831) and Sciencell experiment was performed in triplicate. Research Laboratories (2950). FHC was cultured in DMEM:F12 Medium (Invitrogen) supplemented with 10 Caspase assay mmol/L HEPES, 10 ng/mL cholera toxin, 0.005 mg/mL Assays were performed in white 96-well microtiter insulin, 0.005 mg/mL transferrin, 100 ng/mL hydrocorti- plates (Nunc) with the Caspase-Glo 3/7 Assay of Pro- sone, 10% FCS, and 1% penicillin/streptomycin. HCoEpiC mega, according to the manufacturer’s protocol. Cells was cultured in Colonic Epithelial Cell Medium (CECM; were treated for 6 hours with placebo, MOG-19 (5 or 8 Cat. No. 2951) and 1% penicillin/streptomycin. The mmol/L), or irinotecan (0.8 mg/mL) as well as in combi- human fibroblast cell line CCD-18Co was obtained from nation of each other. Caspase-3/7 levels were detected by ATCC (CRL-1459) and cultured in EMEM (Invitrogen), luminescence using a plate reader (Appliscan, Thermo 10% FCS and 1% penicillin/streptomycin. Fisher). Analyses were performed in triplicate. Cultures were maintained in a 5% CO2 humidified incubator at 37C. All experiments were performed with Western blot analysis exponentially growing cells. IdentiCell confirmed, by short A total of 2 106 HT-29, HCT-116, SW480, and Caco-2 tandem repeat analysis, the authenticity of the cell lines in cells were harvested after 4-hour exposure to placebo, 2011. The 4 cell lines were free of any cross-contamination. MOG-19, and MOG-13 at different concentrations Characterization of putative targets (VEGFR-2/3, between 1 and 30 mmol/L. Cells were washed twice with PDGFRb, FLT3, and GSK3b) in these cell lines was done PBS and lysed in 2 RIPA solution. The xenograft materi- by reverse transcriptase PCR (RT-PCR) as well as FACS als were unfreezed on ice, covered with 2 RIPA buffer (21), and mutational analyses were realized by high-res- supplemented with Protease Inhibitor (Roche), and lysed olution melting PCR as described earlier (22). with Precellyse ceramic beads 2.8 (Peqlab). For Western blot analysis, 60 mg of protein was loaded on 10% SDS- Inhibitors PAGE gels. After separation, the gel was transferred to 3,4-Diarylmaleimides (MOG-19 and MOG-13) were nitrocellulose transfer membranes (Schleicher & Schuell). synthesized at the Department of Pharmacy and Biochem- Proteins were detected with specific primary mouse and istry (Johannes-Gutenberg-University of Mainz, Mainz, rabbit antibodies (all Cell Signaling, used in indicated Germany; refs. 4–6). They were dissolved and diluted in concentrations, 4C, overnight) in 5% BSA. The specific dimethyl sulfoxide (DMSO; Sigma) to a stock concentra- secondary antibodies (anti-rabbit, anti-mouse from Santa tion of 10 mmol/L and stored at 80C. The chemother- Cruz) diluted 1:2,000 in 5% non-fat dried milk were apeutic agents 5-fluorouracil (5-FU), oxaliplatin, irinote- exposed for 1 hour at room temperature. The ECL Chemi- can, and topotecan were diluted in water by the Pharmacy luminescence Detection Kit (Perkin Elmer) was used for of the University Medical Centre. Vandetanib and suni- visualization. tinib were obtained from LC Laboratories. Untreated control cells were incubated with the equivalent doses of Animals DMSO in all experiments to ensure that there were no Female NOD/SCID gc / mice were purchased from possible side effects of DMSO. the central animal facility (ZVTE, University of Mainz). The mice were maintained in a laminar airflow cabinet In vitro cytotoxicity and apoptosis assay under pathogen-free conditions and used at 7 to 10 weeks The MTT assay was used to measure the effects of the of age. Mice were housed in micro-isolator cages with free test drugs on cancer tumor cell proliferation (22). The cells access to laboratory chow and tap water. All animal were plated in 96-well plates in different densities, experiments were performed in accordance with the depending on their doubling time, to enable log linear UKCCCR regulations for the Welfare of Animals (23) and growth for 72 hours. Cells were allowed to attach over- of the German Animal protection Law and approved by night before drugs were added. Moguntinones were the local responsible authorities. added in final concentrations ranging between 0.01 and 20 mmol/L. Growth inhibition was assessed as inhibitory Treatment of subcutaneous human colorectal concentration 50% (IC50) after 72-hour drug exposure to carcinoma xenografts the cells compared with vehicle-treated controls and rel- HT-29 colon carcinoma tumors were established as ative to cell growth at the time of drug addition. Each xenografts by injecting 1 107 cells, mixed in PBS medium experiment was performed in triplicate and 3 indepen- (1:1) subcutaneously into the left flank of female NOD/ dent experiments were performed. SCID gc mice. Ten days after cell injection, all mice bore a

www.aacrjournals.org Mol Cancer Ther; 13(6) June 2014 1401

Maderer et al.

tumor with a minimum diameter of 4 mm. At this time, the the 2-sample Student t test using SPSS statistical analysis mice were randomized into 4 groups of 5 animals and software. For more than 2 independent samples, the total treated with either saline, MOG-19 at 25 or 50 mg/kg 3 significance level a ¼ 0.05 was Bonferroni-adjusted for each times a week, irinotecan at 25 mg/kg twice a week, or pairwise test. Results are expressed as mean SD. P < 0.05 both in combination (25 mg/kg MOG-19 and 25 mg/kg was considered to indicate significant differences. irinotecan). All substances were administered intraperi- toneally at a volume of 0.2 mL/injection. The dose and schedule of MOG-19 and irinotecan were based on data Results and Discussion from previous studies. The health of the treated animals Of over a hundred different derivatives, MOG-13 and was checked daily. Body weight and tumor diameters MOG-19 were selected as potent inhibitors of vessel were measured 2 times weekly with a caliper. Tumor growth in HET-CAM assay as well as having subnano- V ¼ p volumes were calculated according to /6 shorter molar IC50 values for the inhibition of VEGFR-2/3, diameter2 longer diameter. At each measurement day, PDGFRb, and FLT3 in a kinase screen (see Fig. 1 the mean tumor volumes were calculated. After 36 days of and Table 1; refs. 4, 5 for details). MOG-19 in comparison treatment, mice were sacrificed by cervical dislocation, with MOG-13 is also able to inhibit the intracellular b inspected for organ changes, and tumors were dissected signaling by GSK3 with an IC50 value of 71 nmol/L. In and weighed. this investigation, the in vitro and in vivo antitumor activ- ities of MOG-13 and MOG-19 were investigated as well as RNA isolation and quantitative RT-PCR their effects on signal transduction pathways in 4 human RNA isolation was performed using the RNeasy Kit colorectal cancer cell lines: Caco-2, HT-29, HCT-116, according to the manufacturer’s recommendations (Qia- SW480, and the normal human colon cell lines FHC, gen) from the ectope HT-29 xenograft tumors (24). Tran- CCD-18Co, and HCoEpiC. scription of the housekeeping gene glycerinaldehyde-3- First, the presence of putative angiogenesis targets phosphatedehydrogenase (GAPDH), VEGF-A, CXCR4, (VEGFR-2/3, PDGFRb, FLT3) in these colorectal cancer Hif1a, and EpCAM were analyzed by a one-step RT-PCR cell lines was determined by RT-PCR (Supplementary Fig. using a LightCycler 480 system (Roche). RT-PCR was S1A). VEGFR-2 was only weak expressed in HCT-116 and performed with 0.1 mg of RNA in a 35-cycle reaction (20 detectable in HT-29 and SW480 cells. VEGFR-3 was mL total volume; QuantiTect SYBRGreen RT-PCR, Qia- expressed in all investigated cell lines. PDGFRb was gen) according to the recommendations of the manufac- expressed in all but HCT-116 cells. The translation into turer. All RT-PCR reactions were done in 2 replicates. All the respective protein was measured (data not shown) PCRs were established with an exponential phase effi- and corresponded with the RNA expression. While strong ciency of 2 to guarantee that the data were comparable. RNA expression of FLT3 was detected in Caco-2, HCT- The evaluation of the expression of the target genes was 116, and SW480, no receptor protein was verifiable. Taken performed relative to the expression of GAPDH. Control together, these selected human cell lines display a differ- and test samples of the xenograft experiments were ana- ential expression of tyrosine kinase receptors involved in DDC lyzed with the software of the LC480 based on the t angiogenesis mediation and exhibit putative targets of approach. RT-PCR primers and used programs for all moguntinones. targets could be requested by the authors. To demonstrate moguntinone action downstream of key players KRAS, BRAF, and PIK3CA in colon cancer, Immunohistochemical staining the mutational background of human KRAS (exons 12 and Frozen tissue samples of the HT-29 xenografts were cut 13), BRAF (V600E), and PIK3CA (exons 9 and 20) for the and immunohistochemically stained for Ki67 (DAKO, cell lines was examined. Caco-2 did not show hot spot 1:100), CD31 (abcam, 1:200; ref. 25), LYVE-1 (abcam, mutations in clinical relevant amino acids of any of the 1:100), and VEGF-A (Santa Cruz, 1:100). The acetone-fixed molecules, whereas we detected BRAF and PI3K hot spot tissues were peroxidase blocked in 3% H2O2 in methanol mutations in HT-29, KRAS, and PI3K mutations in HCT- for 10 minutes, blocked with 5% BSA, and subsequently 116 and a KRAS mutation in SW480 (Supplementary Fig. incubated with the respective primary antibody over- S1B; for a more detailed mutation status, see http://www. night. The secondary antibody (anti-rabbit-mouse-goat- sanger.ac.uk/genetics/CGP). The diversity of mutational antibody) was incubated for 20 minutes at room temper- hits in KRAS and PI3K signaling cascades in the colorectal ature, followed by incubation with strepavidin-POD cancer cell lines selected represents a reasonable basis for (DAKO) for 20 minutes. Antibody binding was visualized the analysis of moguntinones downstream activity. using 3,30-diaminobenzidine (DAB) solution (DAKO). The cytotoxic activity and apoptosis induction of MOG- Finally, the tissues were counterstained by hemalaun 19 and MOG-13 was assayed using MTT and PI assays (see solution (DAKO). Fig. 2 and Supplementary Fig. S2C). MOG-19 did not show cytotoxic activity in the dose range tested up to Statistical analysis 20 mmol/L after 3-day drug exposure in all cancer and 2 Significant effects between treatment groups or between normal human colon cell lines. Only the human epithelial treatment groups and control were accomplished by using cell line HCoEpiC indicated a concentration-dependent

1402 Mol Cancer Ther; 13(6) June 2014 Molecular Cancer Therapeutics

Moguntinones—New Selective Inhibitors

A 140 C 140 120 120

100 100

80 80 HT-29 60 60 HT-29 SW480 Viability (%) Viability (%) 40 SW480 40 Caco 2 Caco 2 20 20 Hct-116 Hct-116 0 0 0 0.01 0.1 1 10 20 0 0.01 0.1 1 10 20 MOG-19 (mmol/L) MOG-13 (mmol/L)

B 120 D 120

100 100

80 80

60 60 FHC FHC Viability (%)

40 Viability (%) 40 CCD-18Co CCD-18Co

20 HCoEpiC 20 HCoEpiC

0 0 0 0.1 1 10 20 0 0.1 1 10 20 MOG-19 (mmol/L) MOG-13 (mmol/L)

Figure 2. Effects of MOG-19 and MOG-13 on cell viability. A, MOG-19 did not reduce cell viability of human colon cancer cells (HT-29, SW480, HCT-116, Caco-2) after 3 days of treatment. B, in the normal human colon epithelial cell line (FHC) as well as in the human fibroblast colon cell line (CCD-18Co), MOG-19 did not influence cell viability. The second normal human epithelial colon cell line, HCoEpiC, showed a concentration-dependent reduction of cell viability. C, with the exception of Caco-2, MOG-13 decreased concentration-dependently viability in all cancer cell lines during the same incubation period. D, MOG-13 also induced a concentration-dependent decrease of viability in the normal epithelial and fibroblast colon cell line (CCD-18Co < FHC < HCoEpiC). reduction of viability starting up 1 mmol/L. In contrast, Moreover, we investigated in vitro the effect of mogun- MOG-13 showed a concentration-dependent cytotoxicity, tinones in combination with cytotoxic drugs approved for which was also cell line–dependent. SW480 was the most treatment of colorectal cancer. No additional effect was sensitive with a reduction of viable tumor cells to 40% at 1 observed when moguntinones were combined with 5-FU mmol/L MOG-13 compared with the vehicle control (Fig. and oxaliplatin. Furthermore, no supportive effect was m m 2; IC50: SW480, 0.8 mol/L; HCT-116, 14.7 mol/L; HT-29, detected when cancer and normal human cells were 28.3 mmol/L; HCoEpiC, 9 mmol/L; FHC, CCD-18Co, and treated with MOG-13 in combination with irinotecan or m Caco-2 not reached IC50 up to 30 mol/L). Interestingly, topotecan (data not shown). However, MOG-19 signifi- MOG-13 had no cytotoxic activity on the nonmutated cantly raised the fraction of cells in sub-G1 phase in the 4 KRAS, BRAF, and PIK3CA Caco-2 cell line. Moguntinones colorectal cancer cell lines in the presence of 0.8 mg/mL alone induced apoptosis only in higher concentrations irinotecan or 20 nmol/L topotecan (Fig. 3A and B, Sup- (>10 mmol/L) or after a longer incubation period of more plementary Fig. S3). The most impressive effect was than 5 days. Despite the very low IC50 values of the kinase observed in HT-29 cells. MOG-19 induced a 6- and 4-fold scan, only partial reduction in viability or induction of elevated apoptosis level in combination with topoisom- apoptosis could be detected in these colorectal cancer and erase I inhibitors topotecan or irinotecan. The induced normal human cell lines. The same moguntinones were, apoptosis was synergistically meaning the combination however, able to induce significant cytotoxicity and apo- clearly exceeded the monotherapy with the topoisomer- ptosis in nanomolar range in an FLT3-ITD–mutated AML ase I inhibitor or MOG-19 as well as the addition of these leukemia cell model and in low micromolar range in single treatments. This synergistic induction of apoptosis human AML blasts (7). was not achieved in normal human colon cells (Fig. 3C).

www.aacrjournals.org Mol Cancer Ther; 13(6) June 2014 1403

Maderer et al.

Figure 3. Apoptosis and caspase-3/7 induction by combination MOG-19 with topoisomerase inhibitors irinotecan or topotecan as well as benchmarking experiments with sunitinib and vandetanib. Cells were treated with 8 mmol/L of MOG-19 and combined with 0.8 mg/mL irinotecan (Iri) or 20 nmol/L topotecan (Topo) for 7 days. Sub-G1 phase was detected by flow cytometry. A, combination of irinotecan with MOG-19 led to a synergistic induction of apoptosis in all human colon cancer cells. B, treatment of topotecan combined with MOG-19 also intensified synergistic proapoptotic effects in all cancer cells. C, MOG-19 (8 mmol/L) induced apoptosis in 25% to 35% of normal human colon cells. The combination of irinotecan with MOG-19 led to no additional induction of apoptosis. D, combination treatment of irinotecan with vandetanib or sunitinib in comparison with treatment with MOG-19 induced less apoptotic cells. The moguntinones showed greater effectiveness when added to standard chemotherapy. E, HT-29 cells were incubated with MOG-19 (5 or 8 mmol/L) and combined with 0.8 mg/mL irinotecan for 1 day. Caspase-3/7 levels were detected by luminescence assay. Combinations of irinotecan with MOG-19 led to a significant induction of caspase-3/7. Results are displayed as mean with SDs of 3 independent experiments. Statistical significance (, P < 0.05) is shown. The synergistic effect could not be shown on HUVECs (normal tissue). MOG-13 induced no synergistic effects.

MOG-19 induced apoptosis in the human ﬁbroblast and Moguntinones showed especially during the ﬁrst days epithelial cell lines in only 25% to 35% of the cells after 7 of treatment a typical G1 arrest (Supplementary Fig. S2A), days of treatment. The combination with irinotecan again which is associated with a proliferation inhibition (4). In showed only minimal apoptosis-inducing effects in these addition to the G2 shift caused by the topoisomerase I cells. inhibitors, the cells balanced between the cell-cycle phases

1404 Mol Cancer Ther; 13(6) June 2014 Molecular Cancer Therapeutics

Moguntinones—New Selective Inhibitors

and consequently induced apoptosis. Strikingly, MOG-19 in combination with topoisomerase I did not induce apoptosis in human umbilical vein endothelial cells (HUVEC) and normal human colon cells, indicating that this combination is predominantly effective in oncogenic cells (4) and reduces the potential of adverse events. MOG-19 showed comparable suppressive effects independent of the KRAS, BRAF, and PI3KCA mutation panel of the different human colorectal cancer cells. Having in mind that these mutations may lead to the loss of important therapeutic options in the treatment of metastatic colorectal cancer, this clearly points toward a promising treatment strategy (26). In contrast to the synergistic effect of MOG-19 with the 2 topoisomerase I inhibitors, combination treatment of irinotecan with the TKIs vandetanib or sunitinib induced less apoptotic cells (Fig. 3D). These benchmarking experiments favor the treatment with MOG-19 as combination and improvement of standard chemotherapy. Using identical concentrations, we next investigated whether this effect was caspase-3/7–mediated in HT-29 cells. As shown in Fig. 3E, MOG-19 increased apoptosis via a doubling of caspase-3/7 in combination with irinotecan only. In this setting, we suspect that checkpoint kinases (Chk) could be further possible direct or indirect targets of moguntinones. Similar effects could be seen in a Figure 4. Western blot analysis of different human colon carcinoma cells breast cancer model where the combination therapy of were treated with increasing concentrations of MOG-19 and MOG-13 for irinotecan and a Chk1 inhibitor induced checkpoint 4 hours. Western blot of total cell lysates shows phosphorylated and total b b bypass and as consequence apoptosis in some cell lines. GSK3 , -catenin, phosphorylated 4EBP1, and tubulin, respectively. Inhibition of GSK3b led to an increase of b-catenin. In this connection, cells with speciﬁc mutational background respond to this combination therapies that com- bine DNA damage with Chk1 inhibition (27). of cyclin D1 expression level by regulating mRNA tran- As moguntinones used in this work were selected for scription and protein degradation (30). In lung cancer, the their ability to act downstream of key players in cell modulation of GSK3b/b-catenin pathway inhibited signaling known to be highly mutated in colorectal cancer, angiogenesis and proliferation (32). Also in HUVECs, the we next examined the intracellular signaling pathways inhibition of Akt/GSK3b and mTOR signaling led to a using Western blot analysis. Both moguntinones reduce suppression of angiogenesis in vitro and in vivo (33). phosphorylated GSK3b in all human colon cancer cell Furthermore, in silencing experiments, Zhou and collea- lines in a concentration-dependent manner (Fig. 4). In line gues demonstrated that the elimination of GSK3b inhib- with the fact that phosphorylated GSK3b is inactive and ited tumor growth and angiogenesis in a pancreatic cancer unable to target b-catenin to proteasomal degradation, an model (34). Concordantly, we demonstrated that both increase of b-catenin was observed in HT-29, HCT-116, as moguntinones reduce the mTOR signaling in HT-29 (Fig. well as Caco-2 cells. MOG-13 only weak reduced the 4). This cell line reacted most sensitive to the treatment phosphorylation of GSK3b in SW480. Overall, this clearly with moguntinones in all in vitro analyses. Therefore, we indicates targeting of GSK3b pathway by moguntinones. used it in HT-29 tumor xenografts. Thereby, b-catenin is only one possible indirect target of To check the suitability for clinical applications in vivo, the moguntinones by GSK3b. To target the Wnt pathway we analyzed the effect of moguntinones on vessel forma- is one-therapy objective not only for colon cancer (28). tion, tumor volume, and tumor weight of female NOD/ Further functions of this complex modulator could be SCID gc / mice. Mice were treated with saline, MOG-19, found in regulation of inﬂammation (29) and cyclins for irinotecan, or both in combination (Fig. 5). Development cell cycle (30). Their activation resulted in the observed G1 of HT-29 tumors with respect to tumor volume measured arrest and in the combination treatment the induced was reduced to 74% (524 mm3) using irinotecan after 36 apoptosis. Other GSK3b inhibitors, for example, luteolin days of treatment in comparison with the control group leads also to de-phosphorylation and activation of GSK3b (725 mm3). However, further inhibition could be obtained (31). The activated protein then targets cyclin D1 and by adding MOG-19 resulting in additional 30% (312 mm3) b induces G1 arrest. GSK3 , as a multiplayer in cell metab- reduction of tumor volume (Fig. 5A). After 36 days of olism, represents a promising therapeutic target. It has treatment, the weight of isolated tumors was compared. been shown to play an important role in the determination Treatment with MOG-19 (1.3 g) alone caused tumor

www.aacrjournals.org Mol Cancer Ther; 13(6) June 2014 1405

Maderer et al.

A 1,250 B 1.50 Irinotecan + MOG-19 1,000 )

3 Irinotecan

750 MOG-19 1.00 Control

500 0.50 Tumor weight (g) Tumor Tumor volume (mm volume Tumor 250

0 0.00 0 2 7 11 14 16 21 24 28 31 36 Days after treatment start Control MOG-19 Irinotecan MOG-19 Irinotecan +

C 1,750

1,500 Control ) 3 1,250 MOG-19 25 mg/kg MOG-19 50 mg/kg 1,000

750

500 Tumor volume (mm volume Tumor 250

0 0 2 4 7 9 11 14 16 18 21 23 25 28 Days after treatment start

Figure 5. In vivo effects of MOG-19 against subcutaneously grown human colon tumors. HT-29 tumor–bearing animals were treated with either MOG-19 alone or in combination with irinotecan. After sacriﬁce of the animals, tumors were removed and analyzed as described. A, antitumor activity of MOG-19, irinotecan, and their combination in HT-29. The maximum shrinkage of volume could be detected in the combination of MOG-19 and irinotecan compared with monotherapy. B, tumor weight was measured at the end of the experiment. The lowest mean weight was found in the combination arm. C, application of MOG-19 in 2 different concentrations. About 25 mg/kg showed only marginal tumor volume changes in comparison with the control group, instead of 50 mg/kg, which induced shrinkage up to one third. Statistical signiﬁcance (, P < 0.05) is shown.

masses comparable to control (1.25 g), whereas treatment the first multikinase inhibitor with survival benefits for with irinotecan resulted in a reduction to 64% (0.8 g). The metastatic colorectal cancer that will be approved by the combination of irinotecan and MOG-19 resulted in drastic U.S. Food and Drug Administration (FDA; ref. 37) after a reduction of the tumor mass to 15% (0.18 g) compared positive monotherapy phase III study (CORRECT; ref. 16). with controls (Fig. 5B). None of the treated animals The used moguntinone concentration is similar to other showed pathologic changes to internal organs such as the TKIs such as regorafenib in human colon cancer (38). On lung, liver, or spleen. MOG-19 was well-tolerated in mice, the contrary, MOG-19 may be used in a more nontoxic no significant weight loss, diarrhea, or abnormal deaths concentration of 25 mg/kg to support the chemotherapy were observed. MOG-19 alone showed no effect in this and make it more effective. concentration, but after doubling to 50 mg/kg (Fig. 5C), To further characterize moguntinone properties on the the tumor volume was reduced to a third of the control tumor environment in vivo, RNA and protein levels of group. In this setting, MOG-19 is similar to the new oral proliferation and angiogenesis-regulating proteins were multikinase inhibitor regorafenib (35). This small mole- measured. The mRNA levels of angiogenesis-associated cule exhibits comparable moguntinones targets (36) and is proteins were significantly reduced in the combination

1406 Mol Cancer Ther; 13(6) June 2014 Molecular Cancer Therapeutics

Moguntinones—New Selective Inhibitors

Figure 6. RNA and protein analyses of angiogenic markers in HT-29 xenograft mice. A, quantitative Hif1a, CXCR4, VEGF-A, and EpCAM mRNA expression analysis of HT-29 xenografts. After expression levels were correlated with GAPDH, a reduction is seen particularly by the combination of Irinotecan and MOG-19. B, Western blot analysis of b-catenin, pGSK3b, GSK3b, Hif1a, CXCR4, VEGF-A, EpCAM, and tubulin of subcutaneously grown HT-29 tumor xenografts. In the combination of MOG-19 and irinotecan, a decrease in protein levels of pGSK3b, Hif1a, CXCR4, VEGF-A, and EpCAM was detected. C, analyses of vessel formation (CD31, LYVE-1), angiogenic ligand (VEGF-A), and proliferation (Ki67) were done by immunohistochemistry (IHC) with subcutaneously grown tumor. IHC stainings were done on control (top line), irinotecan-treated (second line), MOG-19–treated animals (third line), and on the combination of irinotecan and MOG-19 group (bottom line). Magniﬁcation, 200. The combined treatment with irinotecan and MOG-19 showed a clear decrease in all presented markers compared with the controls and single agents.

group of irinotecan and MOG-19 (Fig. 6A). Hif1a, which is Again, expression of EpCAM, just like CXCR4, is linked to activated under hypoxic conditions, was halved in the the frequency of metastasis and tumor progression (40). MOG-19 group compared with the control and irinotecan- Thus, the reduction of these characterized pathways clear- treated animals. The combination of a cytotoxic agent and ly demonstrates the high activity of combinations of the MOG-19 led to further reduction in Hif1a mRNA moguntinones with chemotherapeutic agents against levels. VEGF-A, the ligand of VEGFR-1, also inducing tumor growth and progression, particularly in human vessel angiogenesis, was reduced for about 40% only in colorectal cancer. The Western blot analyses of the same the combination whereas the other control groups were targets are in accordance with the mRNA data (Fig. 6B). unaffected. mRNA of the cancer-associated antigen MOG-19 as GSK3b inhibitor in vitro again reduced the EpCAM and the chemokine receptor CXCR4 showed a phosphorylation of GSK3b in vivo, especially in combina- similar reduction in the double-treated animals in com- tion with irinotecan. As a consequence, an associated parison with the other control groups. The additional accumulation of b-catenin was conﬁrmed in the same inhibition of the chemokine receptor CXCR4 support the tumors. Double treatment of irinotecan and MOG-19 antitumoral and antiangiogenic properties of mogunti- resulted in protein level reductions of Hif1a, VEGF-A, nones, as CXCR4 and CXCR7 regulate angiogenesis (39). EpCAM, and CXCR4. Hif1a, induced by irinotecan, was

www.aacrjournals.org Mol Cancer Ther; 13(6) June 2014 1407

Maderer et al.

considerably reduced by MOG-19. Concordantly, this various tissues, organs, and for metastases (45–47). Com- kind of Hif1a reduction was also demonstrated under pared with combretastatin, the moguntinones lost the treatment with irinotecan and rapamycin in colon cancer effect of tubulin depolymerization (6). (41). Even more, the reduction of VEGF-A levels again Thus, these in vitro and in vivo experiments argue for a reflects the antiangiogenic potential of moguntinones, high potency of this new class of compounds comple- which was also seen in mRNA analyses and in immuno- menting the current therapeutic options in human colo- histochemical stainings. rectal cancer. Currently, our focus is to further character- As b-catenin was shown to be an inducer of EpCAM in ize the pharmacodynamic and pharmacokinetic aspects in colon (42) and hepatocellular carcinoma (43), EpCAM larger animal species to clearly establish the scientific role activation may have been promoted in the MOG-19 group of moguntinones for an upcoming clinical phase I trial. by increasing b-catenin levels induced by moguntinones alone. This effect was abrogated by the addition of irino- Disclosure of Potential Conflicts of Interest tecan combined with MOG-19. No potential conflicts of interest were disclosed. Comparison of vessel formation of isolated tumors by CD31 and LYVE-1 staining showed a visible decrease in Authors' Contributions large vessel formation in the double-treated group com- Conception and design: A. Maderer, S. Plutizki, G. Dannhardt, P.R. Galle, pared with MOG-19 alone and, to a lesser extent, with M. Moehler € irinotecan alone (Fig. 6C). Again, a reduced smaller vessel Development of methodology: A. Maderer, S. Plutizki, K. Gopfert, M. Linnig, M. Moehler density in the combination group was observed. Further- Acquisition of data (provided animals, acquired and managed patients, more, in this group, Ki67, a well-established proliferation provided facilities, etc.): A. Maderer, S. Plutizki, J.-P. Kramb, K. Gopfert,€ M. Linnig, K. Khillimberger, C. Ganser, E. Lauermann, P.R. Galle, marker, was reduced and not found in the tumor cell M. Moehler nuclei. In contrast, the active form of Ki67 could be Analysis and interpretation of data (e.g., statistical analysis, biostatis- measured in a higher intensity in the tumor cell nuclei tics, computational analysis): A. Maderer, S. Plutizki, J.-P. Kramb, C. Ganser, E. Lauermann, G. Dannhardt, P.R. Galle, M. Moehler in the control and monotherapy groups. In addition, Writing, review, and/or revision of the manuscript: A. Maderer, VEGF-A mRNA levels were reduced and clearly weakly C. Ganser, E. Lauermann, G. Dannhardt, P.R. Galle, M. Moehler expressed in MOG-19/irinotecan animals than in the Administrative, technical, or material support (i.e., reporting or orga- nizing data, constructing databases): A. Maderer, S. Plutizki, K. Gopfert,€ single controls. E. Lauermann, G. Dannhardt, P.R. Galle, M. Moehler Taken together, in vivo experiments demonstrate that Study supervision: A. Maderer, P.R. Galle, M. Moehler MOG-19 significantly reduced tumor volume and weight as well as vessel density (CD31, LYVE-1) in combination Acknowledgments with irinotecan. Concordantly, angiogenesis-associated The authors thank Julia Altmaier from the FACS Core Facility, Mainz, a Germany, for assistance with the FACS analysis and H. Hendriks for co- proteins like VEGF-A or Hif1 were clearly reduced in editing the article. Special thanks go to Stiftung Rheinland-Pfalz fur€ the combination group. These changes in vessel formation Innovation for their financial support. are most likely attributed to the composition of the moguntinones, as moguntinones include structural fea- Grant Support tures of 3 natural antiangiogenic and antiproliferative M. Moehler and G. Dannhardt were supported by a grant from the compounds. One backbone of the moguntinones is com- Stiftung Rheinland-Pfalz fur€ Innovation, Germany (grant number 15202- 386261). bretastatin, which binds the colchicine site of tubulin and The costs of publication of this article were defrayed in part by the prevents the formation of microtubules. It has also been payment of page charges. This article must therefore be hereby marked shown to induce vascular disruption in tumors (44). advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Furthermore, combretastatin A-4 and other derivatives indicated typical antitumor effects by irreversible stop- Received March 25, 2013; revised March 4, 2014; accepted March 18, page of tumor tissue blood flow for tumors growing in 2014; published OnlineFirst April 17, 2014.

References 1. Linkous AG, Yazlovitskaya EM. Novel therapeutic approaches for (3,4,5-trimethoxyphenyl)-1 H-pyrrole-2,5-dione (1) as a highly selec- targeting tumor angiogenesis. Anticancer Res 2012;32:1–12. tive VEGF-R2/3 inhibitor. J Med Chem 2006;49:7549–53. 2. Ellis LM, Hicklin DJ. VEGF-targeted therapy: mechanisms of anti- 6. Peifer C, Stoiber T, Unger E, Totzke F, Schachtele C, Marme D, tumour activity. Nat Rev Cancer 2008;8:579–91. et al. Design, synthesis, and biological evaluation of 3,4-diarylma- 3. Takahashi Y, Nishioka K. Therapeutic approaches targeting tumor leimides as angiogenesis inhibitors. J Med Chem 2006;49: vasculature in gastrointestinal cancers. Front Biosci (Elite Ed) 2011; 1271–81. 3:541–8. 7. Heidel FH, Mack TS, Razumovskaya E, Blum MC, Lipka DB, Ballaschk 4. Ganser C, Lauermann E, Maderer A, Stauder T, Kramb JP, Plutizki S, A, et al. 3,4-Diarylmaleimides-a novel class of kinase inhibitors-effec- et al. Novel 3-Azaindolyl-4-arylmaleimides exhibiting potent antian- tively induce apoptosis in FLT3-ITD-dependent cells. Ann Hematol giogenic efﬁcacy, protein kinase inhibition, and antiproliferative activ- 2012;91:331–44. ity. J Med Chem 2012;55:9531–40. 8. Manoukian AS, Woodgett JR. Role of glycogen synthase kinase-3 in 5. Peifer C, Krasowski A, Hammerle N, Kohlbacher O, Dannhardt G, cancer: regulation by Wnts and other signaling pathways. Adv Cancer Totzke F, et al. Proﬁle and molecular modeling of 3-(indole-3-yl)-4- Res 2002;84:203–29.

1408 Mol Cancer Ther; 13(6) June 2014 Molecular Cancer Therapeutics

Moguntinones—New Selective Inhibitors

9. Jope RS, Johnson GV. The glamour and gloom of glycogen synthase 30. Takahashi-Yanaga F, Sasaguri T. GSK-3beta regulates cyclin D1 kinase-3. Trends Biochem Sci 2004;29:95–102. expression: a new target for chemotherapy. Cell Signal 2008;20: 10. Mishra R. Glycogen synthase kinase 3 beta: can it be a target for oral 581–9. cancer. Mol Cancer 2010;9:144. 31. Ong CS, Zhou J, Ong CN, Shen HM. Luteolin induces G1 arrest in 11. Phukan S, Babu VS, Kannoji A, Hariharan R, Balaji VN. GSK3beta: role human nasopharyngeal carcinoma cells via the Akt-GSK-3beta-Cyclin in therapeutic landscape and development of modulators. Br J Phar- D1 pathway. Cancer Lett 2010;298:167–75. macol 2010;160:1–19. 32. Fei HR, Cui LY, Zhang ZR, Zhao Y, Wang FZ. Caudatin inhibits 12. Spagnoli GC, Kloth J, Terracciano L, Trutmann M, Chklovskaia E, carcinomic human alveolar basal epithelial cell growth and angiogen- Remmel E, et al. FLT3 ligand gene expression and protein production in esis through modulating GSK3beta/beta-catenin pathway. J Cell Bio- human colorectal cancer cell lines and clinical tumor specimens. Int J chem 2012;113:3403–10. Cancer 2000;86:238–43. 33. Wang CY, Tsai AC, Peng CY, Chang YL, Lee KH, Teng CM, et al. 13. Ibrahim N, Yu Y, Walsh WR, Yang JL. Molecular targeted therapies for Dehydrocostuslactone suppresses angiogenesis in vitro and in vivo cancer: sorafenib mono-therapy and its combination with other ther- through inhibition of Akt/GSK-3beta and mTOR signaling pathways. apies (review). Oncol Rep 2012;27:1303–11. PLoS One 2012;7:e31195. 14. Strumberg D, Schultheis B. Regorafenib for cancer. Expert Opin 34. Zhou W, Wang L, Gou SM, Wang TL, Zhang M, Liu T, et al. ShRNA Investig Drugs 2012;21:879–89. silencing glycogen synthase kinase-3 beta inhibits tumor growth and 15. Waddell T, Cunningham D. Evaluation of regorafenib in colorectal angiogenesis in pancreatic cancer. Cancer Lett 2012;316:178–86. cancer and GIST. Lancet 2013;381:273–5. 35. Wilhelm SM, Dumas J, Adnane L, Lynch M, Carter CA, Schutz G, 16. Grothey A, Van Cutsem E, Sobrero A, Siena S, Falcone A, Ychou M, et al. Regorafenib (BAY 73-4506): a new oral multikinase inhibitor et al. Regorafenib monotherapy for previously treated metastatic of angiogenic, stromal and oncogenic receptor tyrosine kinases colorectal cancer (CORRECT): an international, multicentre, rando- with potent preclinical antitumor activity. Int J Cancer 2010;129: mised, placebo-controlled, phase 3 trial. Lancet 2013;381:303–12. 245–55. 17. Kohne CH, Lenz HJ. Chemotherapy with targeted agents for the 36. Aprile G, Macerelli M, Giuliani F. Regorafenib for gastrointestinal treatment of metastatic colorectal cancer. Oncologist 2009;14:478–88. malignancies: from preclinical data to clinical results of a novel 18. Lynch TJ Jr, Kim ES, Eaby B, Garey J, West DP, Lacouture ME. multi-target inhibitor. BioDrugs 2013;27:213–24. Epidermal growth factor receptor inhibitor-associated cutaneous toxi- 37. FDA approves regorafenib (Stivarga) for metastatic colorectal cancer. cities: an evolving paradigm in clinical management. Oncologist Oncology (Williston Park) 2012;26:896. 2007;12:610–21. 38. Abou-Elkacem L, Arns S, Brix G, Gremse F, Zopf D, Kiessling F, et al. 19. Bennouna J, Sastre J, Arnold D, Osterlund P, Greil R, Van Cutsem E, Regorafenib inhibits growth, angiogenesis, and metastasis in a highly et al. Continuation of bevacizumab after first progression in metastatic aggressive, orthotopic colon cancer model. Mol Cancer Ther colorectal cancer (ML18147): a randomised phase 3 trial. Lancet Oncol 2013;12:1322–31. 2013;14:29–37. 39. Kollmar O, Rupertus K, Scheuer C, Nickels RM, Haberl GC, Tilton B, 20. Chu E. An update on the current and emerging targeted agents in et al. CXCR4 and CXCR7 regulate angiogenesis and CT26.WT tumor metastatic colorectal cancer. Clin Colorectal Cancer 2012;11:1–13. growth independent from SDF-1. Int J Cancer 2010;126:1302–15. 21. Lyros O, Mueller A, Heidel F, Schimanski CC, Gockel I, Galle PR, et al. 40. Lin CW, Liao MY, Lin WW, Wang YP, Lu TY, Wu HC. Epithelial cell Analysis of anti-proliferative and chemosensitizing effects of sunitinib adhesion molecule regulates tumor initiation and tumorigenesis via on human esophagogastric cancer cells: synergistic interaction with activating reprogramming factors and epithelial-mesenchymal tran- vandetanib via inhibition of multi-receptor tyrosine kinase pathways. sition gene expression in colon cancer. J Biol Chem 2012;287: Int J Cancer 2010;127:1197–208. 39449–59. 22. Mueller A, Bachmann E, Linnig M, Khillimberger K, Schimanski CC, 41. Pencreach E, Guerin E, Nicolet C, Lelong-Rebel I, Voegeli AC, Galle PR, et al. Selective PI3K inhibition by BKM120 and BEZ235 alone Oudet P, et al. Marked activity of irinotecan and rapamycin com- or in combination with chemotherapy in wild-type and mutated human bination toward colon cancer cells in vivo and in vitro is mediated gastrointestinal cancer cell lines. Cancer Chemother Pharmacol 2012; through cooperative modulation of the mammalian target of rapa- 69:1601–15. mycin/hypoxia-inducible factor-1alpha axis. Clin Cancer Res 2009; 23. United Kingdom Co-ordinating Committee on Cancer Research 15:1297–307. (UKCCCR). Guidelines for the Welfare of Animals in Experimental 42. Liu D, Sun J, Zhu J, Zhou H, Zhang Y. The expression and clinical Neoplasia (Second Edition). Br J Cancer 1998;77:1–10. significance of b-catenin and colorectal cancer stem cells marker 24. Moehler M, Frings C, Mueller A, Gockel I, Schimanski CC, Biesterfeld EpCAMhigh/CD44þ in colorectal cancer. Chin-Ger J Clin Oncol S, et al. VEGF-D expression correlates with colorectal cancer aggres- 2013;12:581. siveness and is downregulated by cetuximab. World J Gastroenterol 43. Yamashita T, Budhu A, Forgues M, Wang XW. Activation of hepatic 2008;14:4156–67. stem cell marker EpCAM by Wnt-beta-catenin signaling in hepatocel- 25. Kroger C, Vijayaraj P, Reuter U, Windoffer R, Simmons D, Heukamp L, lular carcinoma. Cancer Res 2007;67:10831–9. et al. Placental vasculogenesis is regulated by keratin-mediated hyper- 44. Lippert JW III. Vascular disrupting agents. Bioorg Med Chem 2007; oxia in murine decidual tissues. Am J Pathol 2011;178:1578–90. 15:605–15. 26. Lievre A, Blons H, Laurent-Puig P. Oncogenic mutations as predictive 45. Hori K. Starvation tactics for solid tumors: tumor blood flow interrup- factors in colorectal cancer. Oncogene 2010;29:3033–43. tion via a combretastatin derivative (Cderiv), and its microcirculation 27. Ma CX, Cai S, Li S, Ryan CE, Guo Z, Schaiff WT, et al. Targeting Chk1 in mechanism. Cancer Metastasis Rev 2011;31:109–22. p53-deficient triple-negative breast cancer is therapeutically beneficial 46. Hori K, Saito S, Kubota K. A novel combretastatin A-4 derivative, in human-in-mouse tumor models. J Clin Invest 2012;122:1541–52. AC7700, strongly stanches tumour blood flow and inhibits growth of 28. Fu Y, Zheng S, An N, Athanasopoulos T, Popplewell L, Liang A, et al. tumours developing in various tissues and organs. Br J Cancer beta-catenin as a potential key target for tumor suppression. Int J 2002;86:1604–14. Cancer 2011;129:1541–51. 47. Hori K, Saito S, Nihei Y, Suzuki M, Sato Y. Antitumor effects due to 29. Klamer G, Song E, Ko KH, O'Brien TA, Dolnikov A. Using small irreversible stoppage of tumor tissue blood flow: evaluation of a novel molecule GSK3beta inhibitors to treat inflammation. Curr Med Chem combretastatin A-4 derivative, AC7700. Jpn J Cancer Res 1999;90: 2010;17:2873–81. 1026–38.

www.aacrjournals.org Mol Cancer Ther; 13(6) June 2014 1409

Moguntinones−−New Selective Inhibitors for the Treatment of Human Colorectal Cancer

Annett Maderer, Stanislav Plutizki, Jan-Peter Kramb, et al.

Mol Cancer Ther 2014;13:1399-1409. Published OnlineFirst April 17, 2014.

Updated version Access the most recent version of this article at: doi:10.1158/1535-7163.MCT-13-0224

Supplementary Access the most recent supplemental material at: Material http://mct.aacrjournals.org/content/suppl/2014/04/17/1535-7163.MCT-13-0224.DC1

Cited articles This article cites 47 articles, 7 of which you can access for free at: http://mct.aacrjournals.org/content/13/6/1399.full#ref-list-1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/13/6/1399. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.