| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Clinical Research |
Departments of 1 Hematology/Oncology, 2 Neurosurgery, 3 Radiology, and 4 Pathology and 5 Winship Cancer Institute, Emory University, School of Medicine, Atlanta, Georgia; 6 Department of Hematology/Oncology, Weill Cornell Medical College of Cornell University, New York, New York; 7 Department of Radiation Oncology, Ajou University Hospital, Suwon, Korea; and 8 EntreMed, Inc., Rockville, Maryland
Requests for reprints: Hyunsuk Shim, Winship Cancer Institute, 1701 Uppergate Drive, C5008, Atlanta, GA 30322. Phone: 404-778-4564; Fax: 404-778-5550; E-mail: hyunsuk.shim{at}emory.org.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
protein levels, and microtubule destabilization as assessed by deacetylation. Our results indicate that 2-methoxyestradiol may be a promising chemotherapeutic agent for the treatment of malignant gliomas, with significant growth inhibition. Further studies are needed to assess the effect of low or intermediate doses of 2-methoxyestradiol in combination with chemotherapeutic agents in clinical studies focused on malignant gliomas. In addition to showing tumor growth inhibition, we identified three potential surrogate biomarkers to determine the efficacy of 2-methoxyestradiol therapy: decreased HIF-1 levels, -tubulin acetylation, and degree of hypoxia as determined by pimonidazole staining. (Cancer Res 2006; 66(24): 11991-7) |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
One characteristic feature of GBM is the presence of hypoxic areas associated with increased angiogenesis. Hypoxia-inducible factor-1 (HIF-1) regulates the expression of target genes critical for the formation of new vasculature and provides a plausible explanation for the vascular hyperplasia seen in GBM (1, 2). This prosurvival transcription factor is involved in the response of both normal and tumor cells to hypoxia and consists of two subunits: HIF-1 and HIF-1ß. The HIF-1 subunit is rapidly degraded under normoxic conditions, whereas the HIF-1ß subunit is constitutively expressed. Under hypoxic conditions, HIF-1 is stabilized and dimerizes with HIF-1ß to form an active transcription factor. The dimer then binds to the DNA sequence 5'-RCGTG-3' [hypoxia-responsive element (HRE)], located in the promoter of target genes. This subsequently leads to up-regulation of factors that promote tumor growth and angiogenesis, including glycolytic enzymes and vascular endothelial growth factor (VEGF; refs. 3, 4). These data suggest that HIF-1 is a promising new target for the treatment of GBM.
HIF-1 transcriptionally up-regulates VEGF production as part of the oxygen regulation system that stimulates the production of new blood vessels to support tumor growth (5). Overexpressed VEGF leads to increased capillary leakiness and vascular permeability, which is followed by vascular hyperplasia, leading to increased vessel density. Therefore, high expression of VEGF in tumor tissues is a strong indicator of poor treatment response. Among the most promising new therapies coming to clinical trials for GBM are antiangiogenic therapies and treatments targeting HIF-1.
2-Methoxyestradiol is a naturally occurring estrogen metabolite that exhibits antiproliferative and antiangiogenic activity by inducing apoptosis as well as inhibiting HIF-1, microtubules, and tumor angiogenesis (6–9). 2-Methoxyestradiol binds directly to the colchicine binding site of the tubulin protein, thereby inhibiting polymerization (10). This disruption not only interferes with the mitotic spindle apparatus but also inhibits HIF-1 translation and its nuclear translocation (6, 11). Mabjeesh et al. (6) reported that 2-methoxyestradiol down-regulated HIF-1, downstream of microtubule disruption, by inhibiting its de novo synthesis. 2-Methoxyestradiol was also shown to preferentially kill tumor cells over normal cells by causing reactive oxygen species (ROS) accumulation in cancer cells (12). Pelicano et al. further reported that inhibition of the mitochondrial electron transport chain by arsenic trioxide, an antineoplastic medicine, enhanced ROS generation and increased its anticancer effect when combined with 2-methoxyestradiol (13).
Panzem (EntreMed, Inc., Rockville, MD), or 2-methoxyestradiol, is currently in phase I/II clinical studies to investigate its safety, efficacy, and pharmacokinetics. Studies with xenograft and metastatic cancer animal models indicate that 2-methoxyestradiol targets both the tumor and the tumor vasculature in sarcomas and melanomas (14). Because of the prominent vasculature of GBM and the dependence on angiogenesis for malignant progression, it is of great interest to investigate the potential therapeutic effects of Panzem for GBM. Here, we evaluate the efficacy and mechanisms of Panzem in the 9L rat orthotopic glioma model using a combination of noninvasive imaging methods [magnetic resonance imaging (MRI) and bioluminescence imaging (BLI)] to measure tumor volume and histology at termination.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In vitro Cell Growth Inhibition
2-Methoxyestradiol (EntreMed) was dissolved in DMSO to obtain a 30 mmol/L stock solution. 9L cells were plated in 60-mm tissue culture Petri dishes in complete medium. Cells were then treated the following day with fresh medium containing 2-methoxyestradiol at concentrations of 10 or 50 µmol/L and then incubated for 48 hours. Cell growth was assessed by counting cells using a hemacytometer. All experiments were done in quadruplicate. In the hypoxic experiment, 9L cells in 60-mm Petri dishes were transferred to the Invivo2 Hypoxia Workstation 1000 (Biotrace, Bothell, WA) immediately after seeding. The cells were treated with 2-methoxyestradiol concentrations of 0 µmol/L (control), 10 µmol/L, or 50 µmol/L 24 hours after seeding. Two humidified hypoxic tissue culture incubators were maintained with an atmospheric pressure of either 1% O2 or 5% O2 with 5% CO2 in nitrogen gas. Forty-eight hours following 2-methoxyestradiol treatment, the Petri dishes were removed from the hypoxic incubator and trypsinized for cell counting.
Implant of Tumor Cells to Rat Brain
We stereotactically injected 9L-V6R cells (50,000 in 5 µL volume) into the brains of Fischer 344 rats (average body weight = 150 g) as reported by Barker et al. (16) at stereotactic coordinates 1 mm forward of the frontal zero plane, 3 mm to the right of midline, and 4.5 mm deep.
2-Methoxyestradiol Treatment
For in vivo experiments, Panzem (nanocrystal colloidal dispersion; Elan Drug Delivery, King of Prussia, PA) was used. Rats (n = 6 per group) were treated with an i.p. injection of the vehicle (60, 200, or 600 mg/kg/d of 2-methoxyestradiol/Panzem) for nine consecutive days beginning on the 8th day after the initial tumor cell injection. The experiment was repeated a second time using three rats per group.
Imaging Studies
BLI. Seven days after the tumor cell injection, the viable hypoxic tumor was identified by noninvasive BLI (Xenogen, Alameda, CA). BLIs were obtained using a Xenogen Small Animal Imager (IVIS Imaging System) equipped with Living Image software. Eight days after the tumor cell injection and before initiation of treatment, rats were anesthetized by i.p. injection of a ketamine (80 mg/kg)/xylazine (4 mg/kg) mixture. Rats were then injected with luciferin (100 mg/kg of luciferin; Xenogen) i.p., and after 15 minutes of incubation, 1-minute image acquisition at medium binning was taken. Imaging by BLI was also done on the 9th day of treatment.
MRI. The response to 2-methoxyestradiol treatment was assessed by the measurement of tumor volume using noninvasive MRI before and after the treatment. Brain images of each animal were obtained on the first day of the treatment (4 hours after BLI to allow animals to recover) and on the 8th day of the treatment. The MRI scan was carried out using a 3T MRI scanner (Philips Intera) and a small volume coil (5-cm diameter). The animals were anesthetized by an i.p. injection of a ketamine (80 mg/kg)/xylazine (4 mg/kg) mixture and then placed in the coil. The head was secured using foam padding to minimize possible movements. Each animal received 1.0 ml/kg (0.2 mmol/L/kg) of Gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA) i.v. A set of multi-slice, T1-weighted, spin echo images were obtained in the coronal section by using a repetition time of 400 ms, an echo time of 14 ms, and an imaging matrix of 128 x 128 with a field of view of 50 x 50 mm2. To match histologic analysis, a slice thickness of 2 mm was used without a slice gap. The number of signal averages was three for the majority of the scans. Tumors shown in the MRI were measured in three orthogonal dimensions. Tumor volume (V) was calculated as: V (mm3) = 
(a x b x c) / 6, where a, b, and c represent width, height, and thickness, respectively. The mean rat brain volume was about 550 to 600 mm3, which was consistent with the size reported by Sahin et al. (17) using histologic measurements of rat brain sections. A mean of these individual values was used. Following the MRI scans, rats were grouped to obtain an even distribution of tumor sizes.
All protocols for animal studies were reviewed and approved by the Institutional Animal Care and Use Committee at Emory University.
Immunohistochemistry
Pimonidazole is used to delineate the hypoxic site of the tumors due to its specific binding to cells in a partial oxygen tension <10 mm Hg. We injected animals with pimonidazole hydrochloride (hydroxyprobe-1; Chemicon, Temecula, CA) i.p. (60 mg/kg) 90 minutes before the animals were sacrificed. Upon sacrifice of the animal, the entire brain was collected and briefly fixed in 10% neutral buffered formalin for 5 minutes. The harvested rat brain was then placed in a formalin bath with an acrylic brain matrix (Ted Pella, Inc., Redding, CA) and immobilized using two razor blades. The isolated brain was sliced into 2-mm thickness from the tip of the frontal lobe to the end of occipital lobe to match the coronal sections of the MRI. Each slice was placed with the frontal part facing down in the uni-cassettes (Tissue-Tek, Fisher Scientific, Hampton, NH) before embedding in paraffin. The intersection of the septum pellucidum and the corpus callosum was used as a focal point to match histology with MRI. Several 6-µm-thick slices were sectioned from each paraffin-embedded block, and one slice from each block was stained with H&E to match the MRI data. The selected tissue sections were subjected to immunohistochemical staining using antibodies against pimonidazole (hypoxyprobe-1; Chemicon), HIF-1 (Novus Biologicals, Littleton, CO), and acetylated -tubulin (Sigma, St. Louis, MO).
To visualize pimonidazole staining, deparaffinized sections were immunostained with a hypoxyprobe-1 antibody following manufacturer instructions (Chemicon). Images were acquired with a Nikon Eclipse E800 microscope.
The detailed procedure of HIF-1 immunostaining was described previously (18), and similar procedures were used for acetylated -tubulin immunostaining (anti–acetylated -tubulin antibody, 1 µg/mL).
For quantitative analysis of immunohistochemical results, two to four representative pictures were acquired from each slide using a Zeiss Axioskop microscope equipped with a Sony CCD video camera system. Images were collected at magnifications of x400 for HIF-1 and acetylated -tubulin and x100 for pimonidazole. The overall degree of positive staining, as indicated by the brown color, was semiquantitatively estimated with four grades (–, +, ++, and +++) by three different people (S.K., D.J.B., and H.S.). Grids were placed on the captured images, and the entire area of the picture, excluding the non-tumor area, was counted. Additionally, an immunostaining score was calculated by the multiplication of the percentage of positive tumor cells (0–100%) and the staining intensity (grades 1–4), producing a total range of 0 to 4 (19).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
synthesis through disruption of microtubule function in vitro (6), whereas 2-methoxyestradiol-induced HIF-1 inhibition in vivo has never been shown before. To study the relation between 2-methoxyestradiol and its effect on HIF-1 function in vivo, we generated a 9L rat glioma cell line (9L-V6R) stably transfected with a HIF-responsive luciferase expression vector (V6R). These cells were injected orthotopically, allowing for detection of HIF-1 activity in tumor cells by noninvasive BLI. Four groups of rats were treated with an i.p. injection of the vehicle (60, 200, or 600 mg/kg/d of 2-methoxyestradiol/Panzem) for nine consecutive days beginning on the 8th day after the initial tumor cell injection. Tumor size and localization were determined by T1-weighted contrast-enhanced (Gd-DTPA) MRI. Representative coronal MRI images from each group, before and after treatment, showed significant dose-dependent growth inhibition as a result of 2-methoxyestradiol treatment (Fig. 2A ). Changes in tumor volume were determined by comparing MRI images obtained before the start of therapy (day 8 after tumor cell implantation) with those obtained on the 9th day of treatment for each dose of 2-methoxyestradiol and the vehicle control (n = 6 for each group). There were minor variations in initial tumor volume before the treatment; however, these variations among the groups were not statistically significant. Tumor volumes at the end of treatment were 283.25 ± 49.87, 70.21 ± 36.12, 56.81 ± 26.80, and 12.61 ± 8.03 mm3 for 0, 60, 200, and 600 mg/kg/d, respectively (n = 6; P = 0.0194, P = 0.00204, P = 0.0147 for the 60 versus 200 mg/kg/d, the 60 versus 600 mg/kg/d, and the 200 versus 600 mg/kg/d groups, respectively). Tumor volumes for individual animals before and after 2-methoxyestradiol treatment are shown in Fig. 2B. Even at the lowest dose tested (60 mg/kg/d), tumor volume was significantly reduced (4-fold reduction, P = 0.00596) compared with the control animals; at the highest dose (600 mg/kg/d), there was an even greater reduction in tumor volume (23-fold reduction, P = 0.0509). Some drug related toxicity was observed at 600 mg/kg/d dose, which was most frequently seen as diarrhea and weight loss (12–15%) and the death of one animal. These MRI data showed that tumor growth was blocked by 2-methoxyestradiol treatment in a dose-dependent manner.
|
The effect of 2-methoxyestradiol on HIF-1 activity was determined using BLI. Figure 3 shows that 2-methoxyestradiol treatment, particularly at the highest dose, dramatically decreased HIF-1 transcriptional activity. At low and intermediate doses, HIF-1 transcriptional activity measured by BLI was comparable with the tumor volume reduction determined by MRI. In other words, the relative HIF-1 activity measured by BLI over tumor volume measured by MRI was not significantly reduced. Thus, it seemed that 2-methoxyestradiol treatment did not appreciably affect HIF-1 transcriptional activity at the 60 or 200 mg/kg/d doses based on BLI data alone.
|
protein and the hypoxic marker pimonidazole. The harvested rat brain was processed as described in the legend of Fig. 2C to match the coronal sections of the MRI.
Figure 4
shows immunohistochemical staining of HIF-1 (A) and pimonidazole (B) of one representative tumor tissue per group. The tumor tissue from the control group was strongly HIF-1 positive, whereas the tumor tissues from the 60, 200, and 600 mg/kg/d groups showed reduced staining for HIF-1 as summarized in Table 1
. The percentage of cells with strong HIF-1-positive staining (+++) was significantly decreased to 24.0%, 18.2%, and 12.0% for 60, 200, and 600 mg/kg/d, respectively, compared with the vehicle-treated group (35.5%). Thus, 2-methoxyestradiol treatment reduced the gross HIF-1 protein levels in a dose-dependent manner (Fig. 4A; Table 1). HIF-1 staining results at the highest dosage (600 mg/kg/d) correlated well with BLI results. However, anti-HIF1 activity of 2-methoxyestradiol determined by immunohistochemistry contradicted BLI results at intermediate doses (60 or 200 mg/kg/d).
|
|
|
status (20–23). In agreement with HIF-1 immunostaining, tumor tissues from all three treated groups exhibited weaker pimonidazole staining than the control tissues, as summarized in Table 1. The percentage of cells with strong pimonidazole-positive staining (+++) was significantly decreased in the 2-methoxyestradiol-treated group (36.0% for 60 mg/kg/d and 0% for 200 and 600 mg/kg/d) compared with the vehicle-treated group (86.5%). This may be attributed to the dramatic inhibition of tumor growth in a dose-dependent manner following 2-methoxyestradiol treatment.
Escuin et al. have shown that inhibition of HIF-1 activity by 2-methoxyestradiol requires disruption of microtubules (24). To determine if the antitumor and HIF-1 inhibition observed in vivo could be correlated with the 2-methoxyestradiol effect on the microtubules of the tumor cells, we analyzed the levels of acetylated -tubulin. Most -tubulins are known to be post-translationally modified by acetylation on the epsilon-amino group of a lysine residue in the NH2 terminus (25). Acetylation is associated with more stable and mature microtubules (26); hence, drugs that hyperstabilize microtubules, such as taxanes, enhance tubulin acetylation, whereas drugs that depolymerize microtubules, such as 2-methoxyestradiol, decrease tubulin acetylation. Therefore, we used tubulin acetylation as a surrogate marker for the effects of 2-methoxyestradiol on microtubule stability. Figure 4C shows one representative tumor tissue section per group after 2-methoxyestradiol treatment. Results of immunohistochemistry for acetylated -tubulin from all samples are summarized in Table 2
. There was a dose-dependent reduction of positive staining for acetylated -tubulin in 2-methoxyestradiol-treated rats, with highly stained areas decreasing from 57% in vehicle-treated animals to 20.0%, 7.4%, and 0% in 60, 200, and 600 mg/kg/d, respectively.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
function as a transcriptional factor in vitro in a dose-dependent manner (6). However, 2-methoxyestradiol activity had not yet been evaluated in vivo in GBM in an orthotopically implanted brain tumor model. Using noninvasive imaging techniques, we measured tumor growth and HIF-1 activity in response to 2-methoxyestradiol treatment in the 9L orthotopic animal model of GBM. Although MRI was used to monitor the tumor volume, BLI was used to monitor the HIF-1 activity using a luciferase vector, whose expression was driven by six HRE motifs. The inhibition of tumor growth in rats with implanted glioma cells occurred in a dose-dependent manner following treatment with 2-methoxyestradiol. We found that the depth of the tumor mass within the brain affected the imaging by BLI, especially in the intermediate dose groups. Components of the tumor closest to the surface were brighter than tumor cells a few millimeters deeper because of the limited penetration depth of scattered light. Thus, the BLI data can only be evaluated in conjunction with MRI to verify that the tumor location is identical between groups in an orthotopic rat brain model. Under such conditions, BLI data are a useful, qualitative measure of HIF-1 activity.
There was a significant inhibition in tumor growth even in the 60 mg/kg/d group accompanied by a dramatic reduction in pimonidazole and HIF-1 staining. Thus, the hypoxic area inside the tumor is decreased by 2-methoxyestradiol as tumor size decreases and tumor oxygenation improves. This reduction in the overall tumor size may reduce the level of hypoxia within the tumor. Alternatively, it is possible that, similar to other antiangiogenic agents (29), a vascular normalization occurs in 2-methoxyestradiol-treated tumors, which could improve blood flow and thus potentially improve results with combined therapies. We postulate that inhibition of tumor growth by the disruption of microtubule formation in actively dividing tumor cells is one of the main mechanisms of action by which 2-methoxyestradiol works in our tumor model. Tubulin acetylation, a post-translational modification that occurs at the conserved lysine residue at position 40 in the NH2 terminus of -tubulin, has been shown to be decreased by 2-methoxyestradiol in MDA-MB-231 breast cancer cells (30) and in IA9 ovarian cancer cells.9 As an in vivo measurement of 2-methoxyestradiol-mediated microtubule disruption, we compared the extent of microtubule acetylation in the tumor sections (6). Consistent with previous findings, the microtubule deacetylation effect by 2-methoxyestradiol treatment was evident in the treated groups compared with the vehicle-treated group (Table 2), which correlated well with the decrease in tumor growth.
In summary, by using noninvasive imaging techniques, we have measured the changes of tumor size in primary tumors and HIF-1 activity to evaluate the response of 2-methoxyestradiol treatment in an orthotopic animal model of gliomas. The reduction of tumor volume in rats with implanted glioma cells and decreased HIF-1 protein levels, tumor hypoxia, and acetylated tubulin occurred in a dose-dependent manner after treatment with 2-methoxyestradiol. A combination of different imaging modalities may be applicable in the comprehensive in vivo monitoring of tumor development and treatment response.
Our results suggest that 2-methoxyestradiol may be a promising chemotherapeutic agent for the treatment of gliomas. This drug is particularly well suited to GBM because of the abundance of neovasculature, and the high expression of HIF-1 is positively associated with the tumor grades in GBM. The demonstration of reduced hypoxia and reduced acetylation in rat tumors may prove useful as biomarkers in future clinical studies. Further preclinical studies are ongoing to assess the effect of low or intermediate doses of 2-methoxyestradiol in combination with chemotherapeutic agents or radiation therapy to guide further clinical studies of this agent in gliomas.
|
Acknowledgments |
|---|
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 Dr. Charlie Hao, a Pathologist, for help in evaluating the immunohistochemistry; Adam Marcus for the protocol of tubulin staining and critical review of the article; Ajou University Hospital for supporting the financial expenses for S-H. Kang to carry out this study at Emory University; and Dr. Joann Brooks and Ghazala Datoo for proofreading.
|
Footnotes |
|---|
Received 4/11/06. Revised 8/29/06. Accepted 9/22/06.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Valable, D. Eddi, J.-M. Constans, J.-S. Guillamo, M. Bernaudin, S. Roussel, and E. Petit MRI assessment of hemodynamic effects of angiopoietin-2 overexpression in a brain tumor model Neuro-oncol, January 1, 2009; 11(5): 488 - 502. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Y.E. Liaw, N. K. Salam, M. J. McKay, A. M. Cunningham, D. E. Hibbs, and M. Kavallaris Class I {beta}-tubulin mutations in 2-methoxyestradiol-resistant acute lymphoblastic leukemia cells: implications for drug-target interactions Mol. Cancer Ther., October 1, 2008; 7(10): 3150 - 3159. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Yeom, J.-K. Chung, J. H. Kang, Y. H. Jeon, K. I. Kim, Y. N. Jin, Y. M. Lee, J. M. Jeong, and D. S. Lee Visualization of Hypoxia-Inducible Factor-1 Transcriptional Activation in C6 Glioma Using Luciferase and Sodium Iodide Symporter Genes J. Nucl. Med., September 1, 2008; 49(9): 1489 - 1497. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Clines and L. A. Kingsley Novel Targets for and Mechanisms of Bone Metastasis: Meeting Report from Skeletal Complications of Malignancy V: October 25-27, 2007 in Philadelphia, Pennsylvania, USA IBMS BoneKEy, August 1, 2008; 5(8): 297 - 301. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Cicek, U. T. Iwaniec, M. J. Goblirsch, A. Vrabel, M. Ruan, D. R. Clohisy, R. R. Turner, and M. J. Oursler 2-Methoxyestradiol Suppresses Osteolytic Breast Cancer Tumor Progression In vivo Cancer Res., November 1, 2007; 67(21): 10106 - 10111. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
