HIF-1-Dependent Stromal Adaptation to Ischemia Mediates In Vivo Tumor Radiation Resistance

Department of Radiation Oncology, The University of Texas M.D. Anderson Cancer Center, Houston, Texas, USA.
Molecular Cancer Research (Impact Factor: 4.5). 03/2011; 9(3):259-70. DOI: 10.1158/1541-7786.MCR-10-0469
Source: PubMed

ABSTRACT Hypoxia-inducible factor 1 (HIF-1) promotes cancer cell survival and tumor progression. The specific role played by HIF-1 and tumor-stromal interactions toward determining tumor resistance to radiation treatment remains undefined. We applied a multimodality preclinical imaging platform to mechanistically characterize tumor response to radiation, with a focus on HIF-1-dependent resistance pathways.
C6 glioma and HN5 human squamous carcinoma cells were stably transfected with a dual HIF-1 signaling reporter construct (dxHRE-tk/eGFP-cmvRed2XPRT). Reporter cells were serially interrogated in vitro before and after irradiation as monolayer and multicellular spheroid cultures and as subcutaneous xenografts in nu/nu mice.
In vitro, single-dose irradiation of C6 and HN5 reporter cells modestly impacted HIF-1 signaling in normoxic monolayers and inhibited HIF-1 signaling in maturing spheroids. In contrast, irradiation of C6 or HN5 reporter xenografts with 8 Gy in vivo elicited marked upregulation of HIF-1 signaling and downstream proangiogenic signaling at 48 hours which preceded recovery of tumor growth. In situ ultrasound imaging and dynamic contrast-enhanced (DCE) MRI indicated that HIF-1 signaling followed acute disruption of stromal vascular function. High-resolution positron emission tomography and dual-contrast DCE-MRI of immobilized dorsal skin window tumors confirmed postradiotherapy HIF-1 signaling to spatiotemporally coincide with impaired stromal vascular function. Targeted disruption of HIF-1 signaling established this pathway to be a determinant of tumor radioresistance.
Our results illustrate that tumor radioresistance is mediated by a capacity to compensate for stromal vascular disruption through HIF-1-dependent proangiogenic signaling and that clinically relevant vascular imaging techniques can spatially define mechanisms associated with tumor irradiation.

Download full-text


Available from: Arun K Thittai, Jun 30, 2015
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Cytokines play a significant role in cancer diagnosis, prognosis and therapy. The immune system's failure to recognize the malignant tumor cells and mount an effective response may be the result of tumor-associated cytokine deregulation. Glioblastoma Multiforme (GBM) has a characteristic cytokine expression pattern, and abnormalities in cytokine expression have been implicated in gliomagenesis. Within the heterogeneous GBM microenvironment, the tumor cells, normal brain cells, immune cells, and stem cells interact with each other through the complex cytokine network. This review summarizes the current understanding of the functions of key cytokines on GBM, and highlights potential therapeutic applications targeting these cytokines.
    Cancer letters 11/2011; 316(2):139-50. DOI:10.1016/j.canlet.2011.11.001 · 5.02 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Object Hypoxia induces an aggressive phenotype in some brain tumors in part due to hypoxia-inducible factor-1α (HIF-1α) and integrin expression. The importance of hypoxia in medulloblastoma is unclear and the interaction of HIF-1α and c-Myc in medulloblastoma has not been explored. The objective of this study was to determine if hypoxia induces an aggressive phenotype in human medulloblastoma cells that constitutively express high (D283 Med) or low (DAOY) levels of c-Myc and to determine if blocking α(v) integrins with the monoclonal antibody intetumumab inhibits hypoxia-induced cellular stress responses. Methods Cells were grown at 21% and 1% O(2) and in the presence or absence of intetumumab. Measures of malignancy evaluated included cell proliferation, cell migration, and expression of vascular endothelial growth factor (VEGF), α(v) integrins, HIF-1α, and c-Myc. Results Both cell lines robustly expressed α(v) integrins. Hypoxic DAOY cells showed significantly increased proliferation compared with normoxic controls (p < 0.05), whereas D283 Med cells did not. Both cell lines exhibited a dose-dependent decrease in proliferation when treated with intetumumab (p < 0.05). Hypoxia did not increase DAOY migration, but intetumumab significantly inhibited migration at both oxygen conditions (p < 0.05). Intetumumab significantly decreased VEGF levels in DAOY cells at both oxygen conditions (p < 0.05) and in normoxic D283 cells (p < 0.01). Neither cell line demonstrated increased HIF-1α expression in response to hypoxia. However, hypoxic D283 Med cells grown in the presence of intetumumab demonstrated significantly decreased c-Myc expression (p < 0.05). Conclusions Hypoxia did not clearly induce a more aggressive phenotype in medulloblastoma cells. Despite this result, intetumumab decreased medulloblastoma cell proliferation and migration and variably decreased VEGF and c-Myc expression in hypoxic conditions. Targeting α(v) integrins represents a promising potential adjuvant modality in the treatment of medulloblastoma, particularly subtypes that metastasize and overexpress VEGF and c-Myc.
    Journal of Neurosurgery Pediatrics 10/2012; 11(1). DOI:10.3171/2012.9.PEDS12268 · 1.37 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Object The survival of patients with high-grade gliomas remains unfavorable. Mibefradil, a T-type calcium channel inhibitor capable of synchronizing dividing cells at the G1 phase, has demonstrated potential benefit in conjunction with chemotherapeutic agents for gliomas in in vitro studies. In vivo study of mibefradil and radiosurgery is lacking. The authors used an intracranial C6 glioma model in rats to study tumor response to mibefradil and radiosurgery. Methods Two weeks after implantation of C6 cells into the animals, each rat underwent MRI every 2 weeks thereafter for 8 weeks. After tumor was confirmed on MRI, the rats were randomly assigned to one of the experimental groups. Tumor volumes were measured on MR images. Experimental Group 1 received 30 mg/kg of mibefradil intraperitoneally 3 times a day for 1 week starting on postoperative day (POD) 15; Group 2 received 8 Gy of cranial radiation via radiosurgery delivered on POD 15; Group 3 underwent radiosurgery on POD 15, followed by 1 week of mibefradil; and Group 4 received mibefradil on POD 15 for 1 week, followed by radiosurgery sometime from POD 15 to POD 22. Twenty-seven glioma-bearing rats were analyzed. Survival was compared between groups using Kaplan-Meier methodology. Results Median survival in Groups 1, 2, 3, and 4 was 35, 31, 43, and 52 days, respectively (p = 0.036, log-rank test). Two animals in Group 4 survived to POD 60, which is twice the expected survival of untreated animals in this model. Analysis of variance and a post hoc test indicated no tumor volume differences on PODs 15 and 29. However, significant volume differences were found on POD 43; mean tumor volumes for Groups 1, 2, 3, and 4 were 250, 266, 167, and 34 mm(3), respectively (p = 0.046, ANOVA). A Cox proportional hazards regression test showed survival was associated with tumor volume on POD 29 (p = 0.001) rather than on POD 15 (p = 0.162). In vitro assays demonstrated an appreciable and dose-dependent increase in apoptosis between 2- and 7-μM concentrations of mibefradil. Conclusions Mibefradil response is schedule dependent and enhances survival and reduces glioblastoma when combined with ionizing radiation.
    Journal of Neurosurgery 11/2012; 118(4). DOI:10.3171/2012.11.JNS121087 · 3.15 Impact Factor