Radiosensitizer Targeting Tumor Angiogenesis

α-Sulfoquinovosylmonoacylglycerol Is a Novel Potent Radiosensitizer Targeting Tumor Angiogenesis Ippei Sakimoto, Keisuke Ohta, Takayuki Yamazaki, et al. Cancer Res 2006;66:2287-2295.

Updated version

Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/66/4/2287

Cited Articles

This article cites by 28 articles, 13 of which you can access for free at: http://cancerres.aacrjournals.org/content/66/4/2287.full.html#ref-list-1

Citing articles

This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/66/4/2287.full.html#related-urls

E-mail alerts Reprints and Subscriptions Permissions

Sign up to receive free email-alerts related to this article or journal. To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at [email protected]. To request permission to re-use all or part of this article, contact the AACR Publications Department at [email protected].

Downloaded from cancerres.aacrjournals.org on June 6, 2013. © 2006 American Association for Cancer Research.

Research Article

A-Sulfoquinovosylmonoacylglycerol Is a Novel Potent Radiosensitizer Targeting Tumor Angiogenesis 1

2

2

3

Ippei Sakimoto, Keisuke Ohta, Takayuki Yamazaki, Seiji Ohtani, Hiroeki Sahara, 2 2 1 Fumio Sugawara, Kengo Sakaguchi, and Masahiko Miura

4

1 Oral Radiation Oncology, Department of Oral Restitution, Graduate School, Tokyo Medical and Dental University, Yushima, Bunkyo-ku, Tokyo; 2Department of Applied Biological Science, Frontier Research Center for Genomic Drug Discovery, Tokyo University of Science, Yamazaki, Noda, Chiba; 3Department of Biomedical Engineering, Sapporo Medical University School of Medicine, Chuo-ku, Sapporo; and 4Marine Biomedical Institute, Sapporo Medical University School of Medicine, Oshidomari, Rishirifuji, Hokkaido, Japan

Abstract Angiogenesis is a promising target for the treatment of cancer, and varying types of antiangiogenic agents have been developed. However, limitations and problems associated with antiangiogenic therapy have recently arisen. Although radiotherapy can be combined with antiangiogenic compounds to overcome these difficulties, almost all previously described angiogenesis inhibitors could still cause side effects at effective doses, and only additive effects are seen in current combination therapy. In this study, we identified a member of the sulfoquinovosylacylglycerols, A-sulfoquinovosylmonoacylglycerol (A-SQMG), originally derived from sea urchins, as a potent radiosensitizer. The agent synergistically inhibits angiogenesis at low doses when combined with ionizing radiation. Combined treatment with A-SQMG and radiation seems to promote the adoption of a senescence-like phenotype by vascular endothelial cells. Finally, the agent remarkably enhances the radioresponse of human tumors transplanted into nude mice, accompanied by a significant reduction in the vascularity of the tumors. Collectively, A-SQMG may be a novel potent radiosensitizer targeting angiogenesis. (Cancer Res 2006; 66(4): 2287-95)

Introduction

tional chemotherapeutic agents that directly target tumor cells. However, even after the prolonged administration of antiangiogenic agents, complete eradication of tumor cells is not achieved (7). In addition, angiogenesis is essential for adipose tissue growth (8), and a VEGFR inhibitor, SU5416, was associated with thrombotic or hemorrhagic complications and hypertension after prolonged use (9). It is now clear that antiangiogenic agents as monotherapy are not the anticancer panacea that they were once thought to be (7). Given these potential shortcomings, we wished to combine novel antiangiogenic agents with radiotherapy to develop more effective tumor-specific treatments. Although effective combination therapies have been reported, most use agents such as angiostatin and anti-VEGF antibodies (10, 11) and require high doses of the antiangiogenic compound causing the negative effects discussed above. In contrast, we wished to identify agents with little effect when used alone but that acted synergistically with radiotherapy. Sulfoquinovosylacylglycerols (SQAG) are sulfoglycolipids originally derived from sea urchins (12), and they are also naturally found in higher plants (13) or sea algae (14). They fall into two groups according to the number of fatty acids, monoacyl forms (SQMG) and diacyl forms (SQDG). Two isomers are structurally possible, but natural products include only the a-isoforms. We can now chemically synthesize any of these compounds, including h-isomers (15, 16). Sahara et al. reported that SQMGs inhibit the growth of the human lung adenocarcinoma, A549, s.c. transplanted in nude mice. Interestingly, this activity was apparent under in vivo conditions, but not in vitro cell culture (15). Although SQMGs could act as DNA polymerase inhibitors (14), this does not fully explain the differences in activity seen in vitro and in vivo. We hypothesized that in vivo, SQMGs may be adversely affecting angiogenesis. In this study, we found that synthetic a-SQMG (Mw = f600) was antiangiogenic at relatively high doses, however, subtherapeutic doses of a-SQMG effectively inhibited tumor growth when combined with ionizing radiation. Thus, a-SQMG may be a novel antiangiogenic agent that specifically targets cancerous tissue when combined with ionizing radiation.

A growing body of evidence indicates that angiogenesis is essential for solid tumor progression. Antiangiogenic treatments hold great promise for the treatment of cancer (1, 2). Among prospective antitumor agents, angiostatin and endostatin are naturally occurring endogenous antiangiogenic protein factors originating from plasminogen and collagen XVIII, respectively (3, 4). Additionally, antivascular endothelial growth factor (VEGF) antibodies and low molecular weight non–protein agents that inhibit the VEGF receptor (VEGFR) tyrosine kinase have been developed. These agents can effectively inhibit tumor growth in vivo in animal models (3–6). Because vascular endothelial cells, the target of these agents, are genetically stable, the development of resistance to antiangiogenic therapies is thought to be unlikely. Moreover, widespread angiogenesis does not occur in adults under normal physiologic conditions, and antiangiogenic therapies should be relatively tumor-specific (1, 2). These characteristics render antiangiogenic therapies theoretically superior to conven-

Materials and Methods

Requests for reprints: Masahiko Miura, Oral Radiation Oncology, Department of Oral Restitution, Graduate School, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8549, Japan. Phone: 81-3-5803-5545; Fax: 81-3-58030205; E-mail: [email protected]. I2006 American Association for Cancer Research. doi:10.1158/0008-5472.CAN-05-2209

Cell lines and culture conditions. SAS, TE-8, HeLa, and A549 cells were obtained from Health Science Research Resources Bank (Sendai, Japan). Bovine aortic endothelial cells (BAEC) were a generous gift from Dr. Ikuo Morita (Tokyo Medical and Dental University). Human umbilical vein endothelial cells (HUVEC) were purchased from Cambrex (Walkersville, MD). SAS, TE-8, HeLa, and A549 cells were maintained in RPMI 1640 containing 1 mmol/L pyruvate, 100 units/mL penicillin, and 100 Ag/mL

www.aacrjournals.org

2287

Cancer Res 2006; 66: (4). February 15, 2006


Cancer Research streptomycin, supplemented with 10% fetal bovine serum. BAECs were maintained in Eagle’s MEM supplemented with 5% fetal bovine serum. HUVECs were maintained according to the manufacturer’s instructions and cells in passage numbers three to five were used for the study. All the cell lines were maintained at 37jC in a humidified atmosphere containing 5% CO2. Preparation of A-SQMG. Synthesized a-SQMGs (15, 16) were dissolved in DMSO (10 mmol/L for in vitro study and 100 mmol/L for in vivo study) and sonicated at 50jC for 5 minutes. The stock solution was appropriately diluted with phosphate-buffered saline (PBS). The final concentrations of DMSO were 50 cells were counted. Multiplicity was not corrected because the a-SQMG used in this study did not affect cell growth; no variation occurred.


Senescence-associated B-galactosidase assay. Senescence-associatedh-galactosidase (h-Gal) assay was done using a h-gal staining kit (Invitrogen, Carlsbad, CA) according to the manufacturer’s instructions, except that staining was done at pH 6.0. Five days after irradiation, cells were fixed in 1 fixative solution (Invitrogen) for 10 minutes and washed in PBS twice. Cells were then stained in staining solution (pH 6.0; Invitrogen) for 10 hours and the reaction was stopped with 70% glycerol. Apoptosis assay. For morphologic analysis of BAECs following different cell treatments, cells grown on collagen I-coated chamber slides (Iwaki, Tokyo, Japan) were fixed in 2% paraformaldehyde at room temperature for 15 minutes and stained with 300 nmol/L 4V,6-diamidino-2-phenylindole (Invitrogen). Fluorescence images were obtained with a fluorescence microscopy (Olympus, Tokyo, Japan). For determining the sub-G1 populations as a marker of apoptosis treatment, BAECs were fixed in cold ethanol at 20jC overnight and washed with PBS. Cells were then treated with 0.5% RNase (Sigma-Aldrich, St. Louis, MO) for 30 minutes and treated with 30 Ag/mL propidium iodide (BioLegend, San Diego, CA). Sub-G1 populations were determined with a FACScan (BD Bioscience). Western blotting. Poly-ADP ribose polymerase (PARP) and h-actin were detected using Western blotting as described previously (17). Briefly, cells were lysed in lysis buffer [20 mmol/L Tris-HCl (pH 8.0), 100 mmol/L NaCl, 1 mmol/L EDTA, 0.5% Triton X-100, 0.1% SDS, 500 Amol/L DTT, 0.1% sodium deoxycholate, 5% glycerol, 1 mmol/L phenylmethylsulfonyl fluoride, 5 Ag/mL aprotinin, 1 mmol/L Na3VO4, and 100 mmol/L NaF] and equal amounts of cell lysate were separated using SDS-PAGE. The proteins were transferred to nitrocellulose membranes, and membranes were blocked in 5% nonfat milk. Proteins were visualized with specific primary antibodies (PARP, BD Bioscience; h-actin, Chemicon International, Temecula, CA) and secondary antibodies conjugated with horseradish peroxidase (Santa Cruz Biotechnology, Santa Cruz, CA) using the enhanced chemiluminescence system (Amersham Pharmacia Biotech). Immunofluorescence staining. BAECs and HUVECs grown on collagen I-coated chamber slides were treated with a-SQMG for 24 hours and washed in PBS. Cells were then irradiated at 4 Gy and fixed 3 days after irradiation. Cells were fixed in 2% paraformaldehyde at room temperature for 15 minutes and incubated with an anti-p53 or p21 antibody (Santa Cruz Biotechnology) together with an antiactin antibody. After extensive washing in PBS, cells were incubated with antimouse IgG conjugated with FITC (BD Bioscience). After washing, the cells were incubated with antirabbit IgG conjugated with Alexa594 (BD Bioscience). After washing in PBS again, the cells were incubated with DAPI. Confocal scanning images were obtained using a laser scanning microscope (Carl Zeiss, Jena, Germany). For histology sections, 7 days after the start of treatment, tumors were excised from mice and fixed in 10% formalin. After embedding in paraffin, 4-Am sections were cut and tissue sections were mounted. After being deparaffinized and rehydrated, H&E staining was done. For evaluation of vascular density, nonspecific binding sites were blocked and slides were simultaneously incubated with an anti–von Willebrand factor (vWF) antibody (Sigma-Aldrich) and an anti-VEGFR2 antibody (Santa Cruz Biotechnology). Each protein was visualized with secondary antibodies conjugated with FITC or AlexaFluor 594. In vivo studies using mice. In vivo studies were carried out in accordance with the Guidelines for Animal Experimentation of Tokyo Medical and Dental University. For all in vivo studies, a-SQMG was given i.p. to mice. Male ob/ob mice (5 weeks old) were fed ad libitum and the weight of mice treated with vehicle or a-SQMG were monitored daily. SAS, TE-8, or A549 cells (1 106 cells) were injected s.c. into the right thigh of male KSN nude mice (8 weeks old). After reaching 50 mm3, tumors were treated and tumor growth was monitored by palpation. The size of palpable tumors was measured with calipers every 2 to 3 days. The tumor volume (V, mm3) was estimated by V = length (mm) [(width)2] / 2. The following definitions were used in the xenograft experiments: tumor growth delay (TGD) was defined as the time in days for tumors in mice treated with radiation to reach a size of 400 mm3 minus the time for untreated control tumor to reach the same size; normalized growth delay (NGD) was defined as the time in days for tumors in mice treated with the combination of a-SQMG and radiation to reach 400 mm3 minus the time in days for

2288



a-SQMG is a Novel Potent Radiosensitizer tumors in mice treated with a-SQMG alone to reach the same size; enhancement ratio (ER) was defined as the ratio of NGD to TGD. Irradiation. For in vitro experiments, 60Co g-ray therapeutic machine (Toshiba, Tokyo, Japan) was used. For in vivo experiments, only the thighs of KSN nude mice with tumors were irradiated with X-ray therapeutic machines (250 or 225 kVp, 15 mA, 1.0 mm Cu filtration; Stabillipan-2: Siemens, Erlangen, Germany; HS-225: Shimadzu, Kyoto, Japan) by shielding the body with lead. Statistical analysis. Statistical comparisons of mean values were done by one-way ANOVA. Differences with P < 0.05 were considered statistically significant.

Results Antiangiogenic activities of A-SQMGs. The composition and structure of a-SQMGs (C14:0 and C18:0) are shown in Fig. 1A. Although originally derived from natural compounds, a-SQMGs can be readily synthesized in vitro (15, 16), and, interestingly, they exhibit greater antitumor activity in vivo than in vitro (12, 15). Given these properties of a-SQMGs, we examined the ability of a-SQMGs to inhibit angiogenesis in vitro. a-SQMG (C18:0) at 40 Amol/L inhibited capillary formation by BAECs grown on Matrigels during a 22-hour incubation, a marker of

angiogenesis in vitro (Fig. 1B). Because a-SQMGs can inhibit DNA polymerase-a (14), we examined the ability of a-SQMG to inhibit DNA replication. Neither DNA synthesis nor cell proliferation were inhibited when cells were incubated with up to 50 Amol/L a-SQMG, whereas incubation with aphidicolin (Aph), a potent DNA polymerase-a inhibitor, completely blocked both events (Fig. 1C and D). These results are consistent with data demonstrating a 100-fold decreased inhibition of DNA polymerase by a-SQMG in vivo due to the cell membrane impermeability (18). When cocultured with human fibroblasts, HUVECs form tube-like structures and adopt characteristics of newly formed blood vessels. When a-SQMGs were added to HUVEC cocultures, tube formation was inhibited by both C14:0 and C18:0 a-SQMG in a dose-dependent manner up to 40 Amol/L (Fig. 1E). Further activity was not seen at 50 Amol/L. a-SQMG is thus intrinsically antiangiogenic at relatively high doses, and this activity is likely not associated with its inhibition of DNA polymerase. Combination of A-SQMG and ionizing radiation synergistically inhibits angiogenesis in vitro . We next examined whether the combination of a-SQMG (C18:0) with ionizing radiation would synergistically inhibit angiogenesis. Because

Figure 1. a-SQMG inhibits angiogenesis in vitro. A, schematic chemical structures of a-SQMGs. C14 or C18, the number of carbons in fatty acids; 0, represents saturated fatty acids. B, antiangiogenic activity of a-SQMG in BAECs in Matrigel cultures. BAECs plated on Matrigel were treated with 40 Amol/L a-SQMG (C18:0) and capillary formation was observed as described in Materials and Methods. C, effect of a-SQMG on DNA synthesis in BAECs. Incorporation of [3H]thymidine was measured as described in Materials and Methods, and the relative DNA synthesis is for increasing concentrations of a-SQMG (C18:0; 0, 5, 30, and 50 Amol/L) or aphidicolin (Aph; 1 and 5 Ag/mL). Columns , mean of four wells; bars , F SD; **, P < 0.0001 versus untreated control. D, effect of a-SQMG on BAEC proliferation. BAECs were treated with increasing concentrations of a-SQMG (C18:0; 0, 10, 30, and 50 Amol/L) or Aph (1 Ag/mL) and absorbance values at 450 nm obtained by MTT assay were plotted as a function of incubation times. Points , mean of triplicate determinants; bars , F SD; **, P < 0.0001 versus untreated control. E, antiangiogenic activity of a-SQMGs in HUVEC-fibroblast cocultures. The relative tube formation of HUVECs was measured after treatment with increasing concentrations of the agents (0, 10, 20, 30, 40, and 50 Amol/L) compared with untreated HUVECs as described in Materials and Methods. Columns , mean of three to four different areas; bars , F SD; *, P < 0.02; **, P < 0.0001 versus untreated control.


2289



Cancer Research

Figure 2. Combined treatment with radiation and a-SQMG synergistically inhibits angiogenesis in vitro. A, capillary formation of HUVECs on Matrigel following combined treatment with radiation and a-SQMG. Cells on Matrigels were untreated (a) or treated with 1 Amol/L (b), 5 Amol/L (c) a-SQMG, 2 Gy of irradiation (d), 1 Amol/L a-SQMG plus 2 Gy (e), and 5 Amol/L a-SQMG plus 2 Gy (f). Photographs were taken 22 hours after the initiation of treatment as described in Materials and Methods. B, invasion of HUVECs following combined treatment. Cells which passed through the Matrigels were counted following combined treatment with a-SQMG and radiation. *, P < 0.01; **, P < 0.0001 versus radiation alone. Points , mean of triplicate determinants; bars , FSD. C, tube formation of HUVECs in HUVEC-fibroblast cocultures following combined treatment. Cells were treated with each concentration of a-SQMG for 24 hours, unwashed or washed with PBS, and irradiated at 0, 2, or 4 Gy. Unwashed cells were further incubated in a-SQMG during the experimental period with replacement of fresh agents every 3 days. Eleven days after the start of the a-SQMG treatment, cells were immunostained using an anti-CD31 antibody and tube formation was quantitated by Image++. *, P < 0.003; **, P < 0.0001 versus radiation alone.

systemic effects arise when angiogenesis is inhibited, we wished to limit the concentration of a-SQMG used. Consequently, we examined capillary network formation by HUVECs on Matrigels over the course of 24 hours in the presence or absence of 3-4 days) after combined treatment, however, cell growth was decreased compared with cells undergoing irradiation alone (Fig. 3D). Senescence-like phenotype is enhanced in BAECs following combined treatment. While examining cells treated as above, we observed varying colony morphologies (Fig. 4). In the flasks of SAS cells, there were no morphologic differences between the colonies receiving radiation alone or the combined treatment. In the flasks of BAECs receiving irradiation alone, many large colonies were observed, as well as a small number of tiny colonies were seen. In contrast, in the flasks receiving the combined treatment, a large number of only small colonies, most consisting of f15 cells were observed. Colonies of this size represent cells that ceased to grow

2290



a-SQMG is a Novel Potent Radiosensitizer

after several mitoses, assuming a doubling time of 24 hours. The kinetics of this response are consistent with the results shown in Fig. 3D. Furthermore, close inspection revealed that the cells comprising the small colonies were extended and large (Fig. 4B), a phenotype reminiscent of senescent cells (21). To test this possibility, we stained cells by senescence-associated h-Gal assay, a representative marker of senescence (21). As expected, only the extremely large cells exhibited staining consistent with senescence. To confirm the senescence-like phenotype, we examined the induction of p21 by immunofluorescence (Fig. 4C). In addition to p21, actin was visualized to determine cell size and nuclei were visualized with DAPI. Three days after combined treatment with radiation and a-SQMG, both cell size and p21 expression were increased in both BAECs and HUVECs compared with untreated cells, consistent with cellular senescence (22). Interestingly, many p21-expressing cells were binucleated or multinucleated, with accompanying micronuclei (Fig. 4C). The senescent cells also upregulated p53 expression (data not shown), an additional senescence marker (22). Taken together, these results indicate that a-SQMG does not directly enhance radiation-induced inhibition of endothelial cell growth at early time points, but it likely acts at later times by promoting the adoption of senescence-like growth arrest. Absence of apoptosis in BAECs following combined treatment. The combination of radiation and a-SQMG might inhibit angiogenesis in BAECs through the induction of apoptosis, and the observed decrease in absorbance in MTT assay is consistent with this hypothesis. However, we did not observe any significant induction of morphologic apoptotic changes following treatment with 10 Amol/L a-SQMG treatment, 4 Gy of irradiation, or their combination (Fig. 5A). A high frequency of binucleated cells with micronuclei were observed in the combined treatment group, but

these cells were easily distinguished from apoptotic cells based on their size. Additionally, PARP cleavage, an event mediated by activated caspase-3 or caspase-7 and a marker of apoptosis, did not occur during the 1 to 5 day period following combined treatment (Fig. 5B). Finally, flow cytometric analysis showed that only 10 injections of 10 mg/kg a-SQMG. Food intake was not significantly affected by either treatment (data not shown). Therefore, we selected a dose of 1 mg/kg, a much safer dose, for further testing of tumor growth inhibition in vivo. We next used a tumor growth delay assay to examine the antitumor effects of a-SQMG in vivo (Fig. 6). SAS cells were s.c.

Figure 3. Combined treatment with radiation and a-SQMG decreases endothelial cell but not tumor cell survival in vitro. The fractions of surviving cells were plotted against a-SQMG concentrations at varying doses of radiation in BAECs (A) and SAS cells (B). Cells grown in plastic flasks were treated with varying concentrations of a-SQMG for 24 hours and washed with PBS. Cells incubated in fresh medium without the agent were irradiated and incubated for an additional 10 to 12 days. Colonies consisting of >50 cells were counted and surviving fractions were calculated. C, short-term effect on DNA synthesis following combined treatment of BAECs. BAECs were treated with a-SQMG (10 or 20 Amol/L) and irradiated as described in Materials and Methods. Thirty minutes after irradiation, [3H]thymidine was incorporated for 1 hour. The relative incorporated activities were determined compared with untreated cells. D, long-term effect on cell growth following combined treatment of BAECs. Cell proliferation was evaluated by MTT assay. Points , mean of triplicate determinants; bars , F SD; **, P < 0.0001 versus radiation alone.


2291



Cancer Research

caused a synergistic delay in tumor growth with the NGD of 12 and the ER of 2.0. Interestingly, when 8 Gy was administered on days 0 and 3, tumor growth was inhibited to a greater extent than five cycles of 2 Gy, and, when combined with a-SQMG treatment, the NGD was 23 and the ER was 2.9, demonstrating dramatic inhibition of tumor growth. We applied this sequence to the human esophageal carcinoma cell line TE-8 and a similarly profound antitumor response was seen (ER, 3.2). We also treated mice implanted with the slowly growing lung adenocarcinoma A549 with two cycles of 4 Gy (on days 0 and 3) and a-SQMG. Although the effect of radiation alone was not apparent, remarkably, no significant tumor regrowth was seen in animals receiving the combined therapy over the observation period. A-SQMG synergistically enhances the radiation-induced reduction of tumor vascular density in vivo. Histologic analysis of TE-8 tumors 7 days after initiation of combined therapy revealed extensive necrosis compared with irradiation alone in H&E staining (Fig. 7A). Vehicle-treated and drug-treated animals exhibited only narrow areas of necrosis (data not shown). We next tried to detect endothelial cells in the implanted tumors by staining for the endothelial cell–specific proteins of vWF and VEGFR2 to evaluate the antiangiogenic effects of combined treatment. Tumors from control or drug-treated animals were positive for both proteins, and the endothelial markers almost completely colocalized. Two cycles of 8 Gy irradiation significantly reduced the endothelial staining, and, in mice undergoing

Figure 4. Combined treatment with radiation and a-SQMG induces cellular senescence in endothelial cells. A, images of representative colonies of BAECs and SAS cells. a, 2 Gy alone; b , 5 Amol/L a-SQMG plus 2 Gy. B, magnified morphology of BAECs. a, a large colony following 2 Gy of irradiation alone; b and c, a small colony and a single cells following 5 Amol/L a-SQMG and 2 Gy of irradiation, respectively. Senescence-associated h-Gal staining in BAECs. d, e, and f , corresponding to (a ), (b), and (c ), respectively, from (B). Cells were stained 5 days after irradiation. C, immunofluorescence staining of p21 in BAECs and HUVECs. Actin and p21 were double-immunostained using specific antibodies as described in Materials and Methods. The nuclei were stained with DAPI.

transplanted into the thighs of nude mice, and animals were vehicle-treated, treated with drug alone, radiation alone, or with a combination of drug and radiation. Treatment with a-SQMG alone via i.p. injection (1 mg/kg on days 0-4) did not inhibit tumor growth, but no apparent side effects were seen. When mice were irradiated with 2 Gy daily for 5 days, tumor growth was delayed and the TGD was 6. The addition of daily a-SQMG injections


Figure 5. Apoptotic activity in BAECs following combined treatment. A, morphologic observation. BAECs were treated with 10 Amol/L a-SQMG for 24 hours (a), 4 Gy of irradiation (b ), or in combination (c ), and stained with DAPI 72 hours after irradiation. Typical apoptotic bodies are shown in (d) following 1.5 mmol/L H2O2 treatment. B, Western blotting for PARP. Cells treated as in (A ) were prepared for Western blotting and equal mounts of cell lysates were separated by SDS-PAGE. PARP was visualized as described in Materials and Methods. Day 1, cells that were lysed 1 day after irradiation.

2292



a-SQMG is a Novel Potent Radiosensitizer

Figure 6. Combined therapy delays tumor growth in vivo. A, five fractions of 2 Gy irradiation (on days 0-4) plus five injections of a-SQMG (1 mg/kg/injection, i.p.) in SAS-induced tumors. B, two fractions of 8 Gy irradiation (on days 0 and 3) plus five injections of a-SQMG (1 mg/kg/ injection, i.p.) in SAS-induced tumors. C, two fractions of 8 Gy irradiation (on days 0 and 3) plus five injections of a-SQMG (1 mg/kg/injection, i.p.) in TE-8-induced tumors. D, two fractions of 4 Gy irradiation (on days 0 and 3) plus five injections of a-SQMG (1 mg/kg/injection, i.p.) in A549-induced tumors. After the tumor volume reached 50 mm3, the volume was plotted against the time after the initiation of the treatment. Points , means of three to four mice; bars , FSD; open arrows , radiation; closed arrows , a-SQMG injection.

combined treatment, vWF and VEGFR2 staining was barely detected even in nonnecrotic areas (Fig. 7A). When these results were quantitated, endothelial protein expression was significantly reduced by the combination of radiation and a-SQMG compared with all other treatment modalities (Fig. 7B). These results strongly suggest that the antiangiogenic properties observed and characterized in vitro likely occur in vivo, as well.

Discussion In the present study, we characterized a promising candidate radiosensitizer, a-SQMG, a member of the monoacyl forms of sulfoglycolipids SQAGs. This agent exhibited antiangiogenic activity in vitro at relatively high doses, but, interestingly, significantly inhibited in vitro blood vessel formation when combined with ionizing radiation. Decreased cell survival was seen only in endothelial cells following combination therapy, but tumor cells were not affected. Additionally, the combination of a-SQMG and radiation caused endothelial cells to adopt a senescence phenotype, but apoptosis was not significantly induced. Finally, the radioresponse of several different tumors was dramatically enhanced by the coadministration of a-SQMG at doses well below those causing any systemic toxicity. The enhanced tumor response was likely due to decreased angiogenesis because tumors from mice undergoing combined therapy exhibited significantly decreased vascularity. Although a-SQMG was initially characterized as a DNA polymerase inhibitor (13, 14), the antiangiogenic activity observed at 30 to 50 Amol/L is likely independent of this function. Notably, C14:0 and C18:0 derivatives of a-SQMGs inhibited angiogenesis to the same extent up to 50 Amol/L. Inhibition of DNA polymerase


strongly correlates with the length of the fatty acid carbon chains (23). Indeed, neither cell proliferation nor DNA synthesis were affected in BAECs by a-SQMG (C18:0) up to 50 Amol/L (Fig. 1C and D). Furthermore, clonogenic cell survival was not affected in either BAECs or tumor cells up to 25 Amol/L (Fig. 3); these data are consistent with a previous report that identified the effective dose for cell growth inhibition as >100 Amol/L (16). These compounds likely are relatively cell-impermeable because the IC50 for DNA polymerase-a is 2.2 Amol/L in a cell-free system (16). This suggests that a-SQMG may also act on an extracellular protein to inhibit angiogenesis. We recently found that a-SQMG binds the extracellular domains of Tie-1 and Tie-2, which are receptors specifically expressed on vascular endothelial cells (24) and play important roles in angiogenesis together with VEGFRs, using phage display screening and surface plasmon resonance analysis.5 The relationship between the antiangiogenic activity of a-SQMG and binding with such molecules is currently under investigation. These compounds may also represent a novel class of angiogenesis inhibitors; compounds previously identified including angiostatin, endostatin, or VEGF-related inhibitors typically directly inhibit endothelial cell growth. Thus, endothelial cell growth and angiogenesis may be uncoupled by a-SQMG. When endothelial cells were treated with a combination of low doses of a-SQMG and ionizing radiation, notable effects were seen. Capillary network formation and endothelial invasion were synergistically inhibited at relatively early times after treatment. Furthermore, growth inhibition resulted from p53 and p21

2293

5

Ohta et al., unpublished data.



Cancer Research

Figure 7. Combined treatment leads to decreased tumor vascularity in vivo. A, HE and immunofluorescence staining of vWF and VEGFR2 in TE-8 tumors receiving radiation alone and a-SQMG plus radiation. Tumors were excised 7 days after the initiation of treatment as shown in Fig. 6C. Histologic preparations were made followed by immunofluorescence staining as described in Materials and Methods. B, quantitative analysis of vWF and VEGFR2 expressions in each treatment group. Columns , mean of three to four different areas; bars , F SD; *, P < 0.02 versus control; **, P < 0.002 versus radiation alone; ***, P < 0.0001 versus control.

induction, leading to senescence accompanied by abnormal nuclear events including binucleation and micronuclei formation (Fig. 4). Interestingly, the induction of a senescence-like phenotype was not observed in the human tumor cell lines examined. The combination of a-SQMG with radiation inhibits angiogenesis at two distinct stages; early inhibition of capillary formation and endothelial cell invasion, and the specific induction of a senescence-like phenotype in endothelial cells at later times, with

subsequent growth arrest. We speculate that the pronounced inhibition of tumor growth observed in vivo results from a novel coordinated mechanism affecting vascular endothelial cells, which efficiently inhibits many key angiogenic processes. To our knowledge, this is the first description of such a multistep inhibition of angiogenesis including the induction of the senescence-like phenotype. A variety of cellular stressors, including ionizing radiation, could induce a senescence-like phenotype, which is unlikely to be related to shortening of telomeres (25). The precise molecular mechanism and biological significance of this phenotype in vivo remain unclear, but p53 and p21 clearly play crucial roles in this process (21). It was recently reported that the accumulation of double-strand breaks ultimately determines the induction of senescence (26). Thus, it is also possible that low doses of a-SQMG might affect the repair of radiationinduced double-strand breaks, which are specifically regulated in endothelial cells. Kolesnick and coworkers reported that tumor vascular endothelial cell apoptosis determines the clonogenic dysfunction of tumor cells and consequently tumor radioresponse (27). The present results may provide additional insight into the potential function of senescence in endothelial cells in relation to radioresponse of tumors. Further modifications of our protocol by varying the radiation dose/fractionation regimen may provide for enhanced efficacy. In general, optimal radiotherapy would directly kill the maximum number of cancer cells whereas angiogenesis inhibitors would kill cells indirectly by depriving tumors of nutrients and oxygen. However, reduced oxygen levels could render cells radioresistant, which makes tumor cells resistant in a fractionation regimen of irradiation. It was recently proposed that some degree of vascular normalization occurs in tumors with time, and the decreased hypoxia promotes radiosensitization (28). Because vascular density is dramatically reduced after combined treatment in our study (Fig. 6), examination of the physiologic changes within tumor tissue such as oxygen tension or interstitial fluid pressure during treatment may reveal the best timing to improve overall outcomes. When further optimized, treatment protocols using a-SQMG or similar compounds in conjunction with radiotherapy may hold great promise for the treatment of solid tumors.

Acknowledgments Received 6/27/2005; revised 10/29/2005; accepted 12/6/2005. Grant support: Grant-in-aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan and a research grant from Toyo Suisan, Co. Ltd. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. We thank Drs. I. Morita, T. Suganami, and Y. Ogawa for their useful discussion; and K. Igarashi for her help in FACS analysis.

1. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995;1:27–31. 2. Folkman J, Kalluri R. Cancer without disease. Nature 2004;427:787. 3. O’Reilly MS, Boehm T, Shing Y, et al. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell 1997;88:277–85. 4. O’Reilly MS, Holmgren L, Shing Y, et al. Angiostatin: a

novel angiogenesis inhibitor that mediates the suppression of metastases by a Lewis lung carcinoma. Cell 1994; 79:315–28. 5. Gerber HP, Ferrara N. Pharmacology and pharmacodynamics of bevacizumab as monotherapy or in combination with cytotoxic therapy in preclinical studies. Cancer Res 2005;65:671–80. 6. Fong TA, Shawver LK, Sun L, et al. SU5416 is a potent and selective inhibitor of the vascular endothelial growth factor receptor (Flk-1/KDR) that inhibits


2294

References

tyrosine kinase catalysis, tumor vascularization, and growth of multiple tumor types. Cancer Res 1999;59: 99–106. 7. O’Reilly MS. Targeting multiple biological pathways as a strategy to improve the treatment of cancer. Clin Cancer Res 2002;8:3309–10. 8. Rupnick MA, Panigrahy D, Zhang CY, et al. Adipose tissue mass can be regulated through the vasculature. Proc Natl Acad Sci U S A 2002;99:10730–5. 9. Kuenen BC, Levi M, Meijers JC, et al. Potential role of



a-SQMG is a Novel Potent Radiosensitizer platelets in endothelial damage observed during treatment with cisplatin, gemcitabine, and the angiogenesis inhibitor SU5416. J Clin Oncol 2003;21:2192–8. 10. Mauceri HJ, Hanna NN, Beckett MA, et al. Combined effects of angiostatin and ionizing radiation in antitumour therapy. Nature 1998;394:287–91. 11. Lee CG, Heijn M, di Tomaso E, et al. Anti-vascular endothelial growth factor treatment augments tumor radiation response under normoxic or hypoxic conditions. Cancer Res 2000;60:5565–70. 12. Sahara H, Ishikawa M, Takahashi N, et al. In vivo antitumour effect of 3V-sulphonoquinovosyl 1V-monoacylglyceride isolated from sea urchin (Strongylocentrotus intermedius) intestine. Br J Cancer 1997;75:324–32. 13. Mizushina Y, Watanabe I, Ohta K, et al. Studies on inhibitors of mammalian DNA polymerase a and h: sulfolipids from a pteridophyte, Athyrium niponicum . Biochem Pharmacol 1998;55:537–41. 14. Ohta K, Mizushina Y, Hirata N, et al. Sulfoquinovosyldiacylglycerol, KM043, a new potent inhibitor of eukaryotic DNA polymerases and HIV-reverse transcriptase type 1 from a marine red alga, Gigartina tenella. Chem Pharm Bull (Tokyo) 1998;46:684–6. 15. Sahara H, Hanashima S, Yamazaki T, et al. Antitumor effect of chemically synthesized sulfolipids based on sea urchin’s natural sulfonoquinovosylmonoacylglycerols. Jpn J Cancer Res 2002;93:85–92.


16. Murakami C, Yamazaki T, Hanashima S, et al. A novel DNA polymerase inhibitor and a potent apoptosis inducer: 2-mono-O -acyl-3-O -(a- D-sulfoquinovosyl)glyceride with stearic acid. Biochim Biophys Acta 2003; 1645:72–80. 17. Yu D, Watanabe H, Shibuya H, Miura M. Redundancy of radioresistant signaling pathways originating from insulin-like growth factor I receptor. J Biol Chem 2003; 278:6702–9. 18. Ohta K, Hanashima S, Mizushina Y, et al. Studies on a novel DNA polymerase inhibitor group, synthetic sulfoquinovosylacylglycerols: inhibitory action on cell proliferation. Mutat Res 2000;467:139–52. 19. Thompson EW, Nakamura S, Shima TB, et al. Supernatants of acquired immunodeficiency syndrome-related Kaposi’s sarcoma cells induce endothelial cell chemotaxis and invasiveness. Cancer Res 1991;51:2670–6. 20. Saito M, Hamasaki M, Shibuya M. Induction of tube formation by angiopoietin-1 in endothelial cell/fibroblast co-culture is dependent on endogenous VEGF. Cancer Sci 2003;94:782–90. 21. Dimri GP, Lee X, Basile G, et al. A biomarker that identifies senescent human cells in culture and in aging skin in vivo . Proc Natl Acad Sci U S A 1995;92: 9363–7. 22. Chang BD, Swift ME, Shen M, Fang J, Broude EV, Roninson IB. Molecular determinants of terminal

growth arrest induced in tumor cells by a chemotherapeutic agent. Proc Natl Acad Sci U S A 2002;99: 389–94. 23. Hanashima S, Mizushina Y, Ohta K, Yamazaki T, Sugawara F, Sakaguchi K. Structure-activity relationship of a novel group of mammalian DNA polymerase inhibitors, synthetic sulfoquinovosylacylglycerols. Jpn J Cancer Res 2000;91:1073–83. 24. Sato TN, Qin Y, Kozak CA, Audus KL. Tie-1 and tie-2 define another class of putative receptor tyrosine kinase genes expressed in early embryonic vascular system. Proc Natl Acad Sci U S A 1993;90:9355–8. 25. Suzuki K, Mori I, Nakayama Y, Miyakoda M, Kodama S, Watanabe M. Radiation-induced senescence-like growth arrest requires TP53 function but not telomere shortening. Radiat Res 2001;155:248–53. 26. Sedelnikova OA, Horikawa I, Zimonjic DB, Popescu NC, Bonner WM, Barrett JC. Senescing human cells and ageing mice accumulate DNA lesions with unrepairable double-strand breaks. Nat Cell Biol 2004; 6:168–70. 27. Garcia-Barros M, Paris F, Cordon-Cardo C, et al. Tumor response to radiotherapy regulated by endothelial cell apoptosis. Science 2003;300:1155–9. 28. Jain RK. Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 2005;307:58–62.

2295

