Clinical Cancer Research CTRC-AACR San Antonio Breast Cancer Symposium Infection and Cancer: Biology, Therapeutics, and Prevention
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
Clinical Cancer Research 13, 2441-2448, April 15, 2007. doi: 10.1158/1078-0432.CCR-06-2692
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Correction (v14,p326)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gillespie, D. L.
Right arrow Articles by Jensen, R. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gillespie, D. L.
Right arrow Articles by Jensen, R. L.

Cancer Therapy: Preclinical

Silencing of Hypoxia Inducible Factor-1{alpha} by RNA Interference Attenuates Human Glioma Cell Growth In vivo

David L. Gillespie1, Kum Whang1, Brian T. Ragel1, Jeannette R. Flynn2, David A. Kelly1 and Randy L. Jensen1

Authors' Affiliations: 1 Department of Neurosurgery and 2 Center for Children, Huntsman Cancer Institute in the Division of Pediatric Hematology/Oncology, University of Utah School of Medicine, Salt Lake City, Utah

Requests for reprints: Randy L. Jensen, Department of Neurosurgery, University of Utah, 175 N. Medical Drive East, Salt Lake City, UT 84132. Phone: 801-581-6908; Fax: 801-581-4138; E-mail: randy.jensen{at}hsc.utah.edu.


   Abstract
Top
 Abstract
Materials and Methods
 Results
 Discussion
 References
 
Purpose: Higher-grade gliomas are distinguished by increased vascular endothelial cell proliferation and peritumoral edema. These are thought to be instigated by vascular endothelial growth factor, which, in turn, is regulated by cellular oxygen tension. Hypoxia inducible factor-1{alpha} (HIF-1{alpha}) is a main responder to intracellular hypoxia and is overexpressed in many human cancers, including gliomas.

Experimental Design: We investigated the role of HIF-1{alpha} in glioma growth in vivo using RNA interference (RNAi) in U251, U87, and U373 glioma cells.

Results: We found that RNAi can be used to significantly attenuate glioma growth by reducing HIF-1{alpha} levels constitutively using short hairpin RNAs and transiently using small interfering RNAs (siRNA). HIF-1{alpha} levels on average were reduced 55% in normoxia and 71% in hypoxia. Vascular endothelial growth factor and GLUT-1 levels were reduced 81% and 71%, respectively, in the stable HIF-1{alpha}–reduced clones. These clones showed significant growth attenuation (up to 73%) compared with negative controls when grown in vivo in mouse flanks. Cellular proliferation was also reduced significantly, as determined by MIB-1 staining. Treating gliomas grown in mouse flank transiently with siRNA against HIF-1{alpha} by intratumoral injection resulted in a significant reduction of HIF-1{alpha} activity. This activity was followed using a hypoxia-responsive luciferase construct that enabled hypoxia imaging in vivo. Tumor volume in these siRNA injection experiments was reduced by 50% over the negative controls.

Conclusions: These results indicate that transient RNAi directed against HIF-1{alpha} can effectively curb glioma growth in vivo.

Among the features that distinguish higher-grade gliomas such as glioblastoma multiforme (GBM) from low-grade gliomas are vascular endothelial cell proliferation and peritumoral edema (1). Vascular endothelial growth factor (VEGF) is thought to be the major mediator of these properties and has been implicated in glioma growth and angiogenesis (2, 3). VEGF expression is regulated by cellular hypoxia mediated by hypoxia-inducible factor-1{alpha} (HIF-1{alpha}) under normal physiologic conditions (4) and possibly under pathologic situations such as tumorigenesis. Overexpression of HIF-1{alpha} has been described in common human cancers and their metastases, among them a limited number of human brain tumors, including gliomas (5, 6). The role of HIF-1 in solid tumor growth is still not entirely clear, but previous work suggests that this transcription factor is necessary for growth and angiogenesis of these tumors (7, 8). Although very little is known of the role of HIF-1{alpha} in glioma growth and angiogenesis, it is logical that similar processes take place in these tumors. A direct correlation between tumor grade and HIF-1 expression in GBMs has been shown (9). We support the proposition that HIF-1{alpha} expression represents an "angiogenic switch" that facilitates the progression of a low-grade astrocytoma to a GBM and promotes cell survival in hypoxic conditions by elevating glycolysis and angiogenesis (6). We hypothesize that reducing HIF-1{alpha} expression will lead to a reduction of glioma growth. Although researchers in several laboratories have shown that this is in fact the case (8, 10, 11), there have been contrary reports of HIF-1{alpha} acting as a tumor suppressor (12, 13). To further address this controversy, we investigated the role of HIF-1{alpha} in glioma cell growth by using RNA interference (RNAi) to silence HIF-1{alpha} in vitro and in vivo.

The transcription factor HIF-1 is composed of two heterodimeric subunits, HIF-1{alpha} and HIF-1ß (known as aryl hydrocarbon receptor nuclear translocator). At the mRNA level, HIF-1{alpha} and HIF-1ß are both constitutively expressed and do not seem to be significantly modified by hypoxia (14). Whereas HIF-1ß protein is found in normoxic cells, HIF-1{alpha} is rapidly degraded by proteasomal degradation. During low oxygen tension conditions (1-2% O2), this degradation is inhibited, leading to increased HIF-1 (15). HIF-1 binds to DNA hypoxia response elements (HRE) and induces the transcription of a number of well-characterized genes that help cells survive low oxygen conditions (16). In addition to VEGF, these genes include erythropoietin, transferrin, and its receptor GLUT-1, and almost every gene in the glycolytic pathway (4). Stabilization of HIF-1{alpha} promotes cell survival via an adaptive modification of cellular metabolism that increases these glycolytic enzymes and hence the glycolysis rate. This adaptation of cancer cells through increased glycolysis was first proposed by Warburg (17) as a necessary step toward an aggressive phenotype. The results of recent studies of HIF-1 have indicated a possible link between HIF-1 and the Warburg effect in various cell types (18, 19), and prompted the proposal that aerobic glycolysis could be controlled by dysregulation of HIF-1{alpha} (20).

The discovery of RNAi in Caenorhabditis elegans using injected double-stranded RNA (dsRNA) launched the current wave of RNAi discoveries in mammals (21). Initially, these experiments failed because of the activation of the IFN response pathway by long dsRNA, which caused global inhibition of protein synthesis. These problems were overcome by Elbashir et al. (22), who discovered that small RNAs of 19 to 23 bases, called short interfering RNA (siRNA), avoided this response. These siRNAs can be synthesized and introduced directly into cells. Unlike the extended silencing effect seen in worms and flies, the effect of siRNA in mammals is transitory, usually lasting up to 72 h in cultured cells (23). The use of plasmids to express short hairpin RNA (shRNA) allows for extended gene silencing. Here, we describe how we have successfully used both shRNA and siRNA to significantly reduce HIF-1{alpha} expression in vitro and in vivo and subsequent glioma growth in vivo.


   Materials and Methods
Top
 Abstract
 Materials and Methods
Results
 Discussion
 References
 
Hypoxia chamber cell growth studies. Cells were plated on 100-mm tissue culture dishes until they reached 80% confluence. The culture medium was changed immediately before the dishes were placed in a GasPak Plus anaerobic culture chamber containing hydrogen and a palladium catalyst (GasPak Plus Hydrogen-CO2 generators; Becton Dickinson, Sparks, MD) to reduce the oxygen concentration below 2%. Oxygen indicator strips (Becton Dickinson) were used to verify the oxygen concentration. Cells were treated for 4 to 24 h. Once the seal of the hypoxia chamber was broken, the plates were placed in an ice-water bath to minimize the degradation of HIF-1{alpha} and washed with ice-cold PBS and phosphatase inhibitor buffer (250 mmol/L NaF, 500 mmol/L ß-glycerophosphate, and 50 mmol/L Na3VO4). Cells were scraped from the culture dishes and cellular protein was isolated as described for Western blot analysis.

Protein isolation from tumors and cell lines. Tumor tissue was snap frozen in the operating room and stored in liquid nitrogen. Tumor samples were chopped and 0.3 g of tissue was homogenized with a Polytron homogenizer in 3 mL of digest buffer [10 mmol/L HEPES (pH 7.6), 2 mmol/L DTT, 1 mmol/L Na2VO4, 100 mmol/L NaF, 0.4 mmol/L phenylmethylsulfonyl fluoride, 0.1 mmol/L EGTA, 10 mmol/L Na4P2O7, 1x protease inhibitor cocktail (Sigma, St. Louis, MO)]. The homogenate was centrifuged at 580 relative centrifugal force (RCF) for 5 min at 4°C. The supernatant was transferred to a new tube, glycerol was added to a final concentration of 5%, and the solution was vortexed and centrifuged at 15,000 RCF for 15 min. The pellet was resuspended in 300 µL of lysis buffer [400 mmol/L NaCl, 20 mmol/L HEPES (pH 7.5), 10 mmol/L NaF, 10 mmol/L PNPP, 1 mmol/L Na2VO4, 0.1 mmol/L EDTA, 10 µmol/L Na2MoO4, 10 mmol/L ß-glycerophosphate, 20% glycerol, 1 mmol/L DTT, and 1x protease inhibitor cocktail] and shaken gently at 4°C for 30 min, then centrifuged at 30,000 RCF at 4°C for 30 min and stored at –70°C until ready for Western analysis.

Isolation of nuclear protein from malignant glioma cell lines grown in 60-mm dishes was done on cells treated in the hypoxia chamber as described above. After growth medium from these cells was decanted, the monolayer was washed twice with ice-cold PBS and phosphatase inhibitor buffer. The plates were kept in an ice-water bath while the cells were dislodged into the second rinse using a cell lifter and transferred to a 15-mL centrifuge tube. The cells were pelleted at 165 RCF for 5 min at 4°C, then resuspended in 600 µL of hypotonic buffer [20 mmol/L HEPES (pH 7.5), 5 mmol/L NaF, 0.1 mmol/L EDTA, 10 µmol/L Na2MoO4] and held in ice for 15 min. Nuclei were isolated by adding NP40 (0.5% final), vortexing for 10 s, and spinning at 165 RCF for 4 min. An aliquot of the supernatant was used for cytoplasmic protein analysis. The nuclei were resuspended in lysis buffer and shaken for 30 min, then clarified by centrifuging at 30,000 RCF for 20 min. The extracts were stored at –80°C until use. All steps were done on ice or at 4°C.

ELISA for HIF-1{alpha} and VEGF. HIF-1{alpha} protein from nuclear extracts was quantitated using the HIF-1{alpha} Trans-AM kit from Active Motif (Carlsbad, CA). VEGF concentrations in the growth medium and cell cytoplasm were measured using a commercially available ELISA (R&D Systems, Minneapolis, MN). Absorbance at 450 nm was measured and corrected using the 540 nm reading on a Benchmark microplate reader (Bio-Rad Laboratories, Hercules, CA). Data analysis was done using Microplate Manager III software (Bio-Rad Laboratories).

Western blot. Total protein concentrations of the nuclear extracts described above were determined spectrophotometrically in 96-well plates using the DC Total Protein Assay (Bio-Rad Laboratories) and read using a Benchmark microplate reader. Equal amounts of protein (25-50 µg) were resolved using a SDS-PAGE slab gel (4-12% continuous gradient) and transferred to polyvinylidene difluoride Hybond-p membrane (Amersham Pharmacia, Piscataway, NJ). The membrane was blocked and probed in TBS with 5% nonfat dry milk, 0.1% Tween 20 using a mouse IgG anti-human HIF-1{alpha} monoclonal antibody (nb100-296: 1:1,800; Novus Biologicals, Littleton, CO), and a mouse IgG anti-actin monoclonal antibody (MAB 1501; 1:5,000; Chemicon International, Temecula, CA). The Western blots were visualized using ECL chemiluminescent reagents (Amersham Pharmacia) and X-OMAT film (Eastman Kodak, Rochester, NY). A positive control for HIF-1{alpha} was provided with the antibody (HIF-1{alpha} protein isolated from cobalt chloride-treated COS-7 cells).

RNAi design. SiRNAs for HIF-1{alpha} were designed by searching the coding sequence of HIF-1{alpha} for two adenines followed by 19 nucleotides that had a GC content below 45% and did not contain more than three thymines or adenines in a row. These sequences were tested for possible homology to other human and mouse genes with BLAST.3 Four potential siRNA sequences were selected and prepared using Ambion (Austin, TX) Silencer siRNA construction kit. After screening, siRNA 1589 (UUCAAGUUGGAAUUGGUAG) was selected for further use. Two previously published siRNAs (1124 and 1659; ref. 24) were used for validation. A negative control was designed by randomizing the sequence of siRNA1589 (AAUUAGCGUAGAUGUAAUGUG) and checking for nonhomology to any human or mouse gene by BLAST. SiRNAs are named based on the nucleotide position in the HIF-1{alpha} mRNA sequence.

Cell transfection and imaging. U251 cells were transfected with LipofectAMINE 2000 (Invitrogen, Carlsbad, CA) following manufacturer's protocol, using a final concentration of 10 nmol/L siRNA. Some siRNAs were labeled with FAM fluorescent dye using the Silencer siRNA Labeling kit (Ambion). Dead cells were visualized using the Block-iT Transfection Optimization kit Dead Cell reagent (Invitrogen).

Creation of shRNA stable transfections. The pSilencer 2.1 U6-hygro plasmid from Ambion was used to express an shRNA controlled via the U6 promoter, following the manufacturer's protocol. The siRNA 1589 sequence was used to create pSil2.1_hygro-1589, along with the negative control, pSil2.1_hygro-Neg. U-251 cells were transfected with Fugene transfection reagent (Roche, Alameda, CA) and transfectants were selected with 300 µg/mL hygromycin in DMEM.

Xenograft tumor model. Transfected glioma cell lines or wild-type U251-HRE, U251, U87, or U373 cells were injected s.c. on the flanks of CD-1 (nu/nu genotype) athymic nude mice. Approximately 1 x 106 cells were placed in ice-cold Matrigel (BD Biosciences, Franklin Lakes, NJ). Tumors were allowed to establish for 2 weeks and then their size was measured with calipers twice weekly for up to 60 days. Tumor volume was calculated as length x width x height and expressed as cubic millimeters. Tumor volume measurements were made by a single blinded observer to prevent observer bias and to avoid interpersonal differences in caliper tumor measurement.

Real-time PCR. Cytoplasmic RNA was isolated from 60% to 80% confluent cells using the Qiagen RNeasy Mini spin column kit. The Invitrogen Superscript III first strand synthesis for reverse transcription-PCR kit was then used with oligo(dT) primers for cDNA synthesis. The cDNA was quantified by spectrophotometer and diluted to 500 ng/µL. Real-time PCR was done using the Lightcycler FastStart DNA Master plus SYBR Green 1 kit (Roche). The PCR reaction was analyzed with a Roche LightCycler. The primer sequences were 10-11f-CCTGAGCCTAATAGTCCC; 11-12r-TACTCAGGACACAGATTTAGAC. Actin was used for positive control.

SiRNA mouse treatment. Mice were injected s.c. with 1 x 106 U251-HRE, U251, U87, or U373 cells, and tumors were allowed to establish for 2 weeks. The HRE cells were a gift from Dr. G. Melillo (National Cancer Institute, Frederick, MD). They have a luciferase gene under control of a modified thymidine kinase promoter with three HREs and consistently express luciferase in a hypoxia- and HIF-1–dependent manner. Mice were then injected intratumorally with 600 pmol of siRNA complexed to 4x JetPEI (Q-Biogene, Irvine, CA) or PBS and 4x JetPEI thrice a week for up to 60 days. The siRNA was mixed with the 4x JetPEI following the manufacturer's instructions, using a nitrogen to phosphate ratio of 8. The siRNAs were synthesized at the University of Utah School of Medicine core facility.

Luciferase mouse imaging. Mice were injected weekly i.p. with 50 mg/kg firefly D-Luciferin (Xenogen Corp., Alameda, CA), 100 mg/kg ketamine, and 10 mg/kg xylazine hydrochloride (both from Sigma) in 1x PBS. Mice were photographed 15 min after injection with an IVIS 100 imaging system (Xenogen) with an exposure time of 1 min, medium binning. Images were quantified using LivingImage software (Xenogen).

MIB-1 and GLUT-1 staining and counting. Slides were cut at 4 µm, stained, and analyzed as described previously (25). GLUT-1 staining was done as previously published (25), then scored on a scale of 0 to 4 by two blinded researchers, and the duplicate scores were averaged together.

Statistical analysis. Graphpad Prism 4 (Graphpad software, San Diego, CA) was used to visualize and perform data analysis by t test, one-way and repeated measures ANOVA, Tukey's multiple comparison test, Mann-Whitney U test, and Wilcoxon signed rank test.


   Results
Top
 Abstract
 Materials and Methods
 Results
Discussion
 References
 
Knockdown of HIF-1{alpha} by siRNA in U251 cells. To test our hypothesis that HIF-1{alpha} plays a role in glioma growth, we designed and synthesized several siRNAs (called siRNA-Neg, 2048, 1124, 1589, and 1659) against HIF-1{alpha}. Two of the siRNAs (1124 and 1659) have been published previously (24) and were used for validation. These were transiently transfected into the glioma cell line U251, which produces HIF-1{alpha}, GLUT-1, and VEGF at low levels in normoxic conditions but reliably increases that expression during hypoxia. To visualize transfection efficiency, some siRNAs were labeled with FAM fluorescent dye. The cells were transfected efficiently, with relatively low cytotoxicity (data not shown), and analyzed by Western blot (Fig. 1A ). The 1124, 1659, and 1589 siRNAs successfully knocked down HIF-1{alpha} expression, whereas the 2048 and Neg siRNAs did not.


Figure 1
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 1. A, Western blot of protein isolated from transfected cells. The cells were allowed to recover for 30 h posttransfection and then placed in hypoxia (<2% O2) for 18 h. The protein was isolated 48 h posttransfection. Cells were transfected with siRNA 1124 (lane 1), siRNA 1659 (lane 2), siRNA 1589 (lane 3), siRNA 2048 (lane 4), and siRNA-Neg (lane 5). B, Western blot of nuclear extract protein from subcloned U251 glioma cells with stably integrated pSil2.1_h-Neg or pSil2.1_h-1589 plasmids after 18-h hypoxia exposure. Lanes 1 to 3, three different clones expressing shRNA-Neg. Lanes 4 to 6, clones expressing shRNA-1589. Lane 7, wild-type U251 control. C, average Hif-1{alpha} ELISA of different U251_shRNA clones grown in normoxic and hypoxic conditions. HIF-1{alpha} in the shRNA-1589 clones is reduced 71% in hypoxia and 55% in normoxia compared with wild-type (*P = 0.0017, n = 4 per clone, **P = 0.0022, n = 4 per clone, one-way ANOVA).

 
Stable RNAi of HIF-1{alpha} using shRNA in U251 cells. To produce U251 cells with a stable knockdown of HIF-1{alpha}, we designed and made DNA oligonucleotides for the siRNAs 1589 and Neg and ligated them into the pSilencer 2.1_hygro plasmid (Ambion), referred to hereafter as pSil2.1_h-1589 and pSil2.1_h-Neg, respectively. After cloning in bacteria and sequencing, the plasmids were transfected into U251 cells, and individual clones were selected by hygromycin. The clones were then screened for HIF-1{alpha} protein expression by Western blot and ELISA (Fig. 1B). All of the shRNA-1589–based clones showed significant decreases in HIF-1{alpha} when compared with the shRNA-Neg and wild-type U251. ELISA for HIF-1{alpha} with nuclear protein extracts from normoxic and hypoxic clones gave results consistent with the Western blot observations, showing a 55% reduction in HIF-1{alpha} in normoxia and a 71% reduction in hypoxia (Fig. 1C). Clones U251_Neg-8, U251_Neg-9, U251_1589-4, U251_1589-7, and U251_1589-11 (Fig. 1B, lanes 2, 3, 4, 5, and 6) were selected for further analysis. Real-time PCR was used to evaluate the levels of HIF-1{alpha} mRNA in the different clones (Fig. 2A ). The mRNA levels in all the 1589-based clones were significantly lower than those in the Neg and wild-type U251 cells. These mRNA copy numbers reflected the relative HIF-1{alpha} protein levels seen in the Western blot and ELISA. Because HIF-1{alpha} is mostly regulated posttranslationally (14), the decrease in mRNA levels indicates the shRNAs are functioning correctly at the pretranslational level.


Figure 2
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 2. A, real-time PCR data from cytoplasmic RNA collected from hypoxic shRNA-producing clones. The HIF-1{alpha} mRNA levels are significantly reduced by 84% (clone 7) and 90% (clone 11) in the 1589-based clones (*P < 0.0001, one-way ANOVA, n = 4 per clone). Data were normalized to wild-type U251 HIF-1{alpha} copy number. B, VEGF protein levels in stable transfectants after 18-h hypoxia exposure, divided by total cellular protein. VEGF levels were determined by ELISA and are significantly lower by 81% in the cells expressing shRNA-1589 (*P = 0.0015, one-way ANOVA, n = 3 per clone).

 
Reduction of HIF-1{alpha} by shRNA results in decreased VEGF and GLUT-1 expression. Because HIF-1 is thought to be a major transcriptional activator of VEGF in gliomas, we analyzed the level of this protein by ELISA in the various stable transfectants, although only the U251_Neg-8 and U251_1589-7 are shown here (Fig. 2B). These VEGF levels are reduced by 81%, which is significantly different, verifying that the HIF-1 protein levels are functionally reduced in the cells expressing shRNA-1589. This also verifies that HIF-1 is a major VEGF inducer during hypoxia in U251 cells. Because GLUT-1 has been shown to be another main target of HIF-1 and a main facilitator of cells switching to glycolysis for energy production, we stained tumor sections taken from U251_shRNA clones grown in mouse flanks. Figure 3 shows that GLUT-1 is significantly reduced by 71% in the U251_shRNA 1589 tumors compared with the shRNA-Neg tumors, which have identical staining to wild-type (data not shown).


Figure 3
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 3. A, representative positive (shRNA-Neg) and negative (shRNA-1589) GLUT-1 stains of U251_shRNA Neg and 1589 tumors, grown in mouse flanks, taken at x400. Dark stain, GLUT-1. B, analysis of GLUT-1 score as described in Materials and Methods. GLUT-1 protein levels in the shRNA-1589 (n = 20) cells is significantly decreased (71%) when compared with the shRNA Neg (n = 18) cells (*P = 0.001, Mann-Whitney U test).

 
Stable knockdown of HIF-1{alpha} significantly reduces tumor volume in vivo. To test the effect of decreased HIF-1{alpha} levels in glioma cells in vivo, we injected the flanks of nude mice with U251_1589-4, U251_1589-7, or U251_1589-11 cells on one side and U251_Neg-8 or U251_Neg-9 on the other. The mice were monitored for 60 days (Fig. 4A ) before they were sacrificed, and the resulting tumors were analyzed by Western blot (Fig. 4B) and immunohistochemistry. As can be seen in Fig. 4A, the difference in tumor volume is significant (62-73%), indicating that HIF-1{alpha} is a major factor contributing to glioma growth in this model. Further analysis of protein extracts from the tumors shows that HIF-1{alpha} is almost nonexistent in tumors expressing shRNA-1589 (Fig. 4B). To address concerns that the pSil2.1_hygro plasmid may be lost with no antibiotic selection in vivo, PCR was done to verify the presence of pSil2.1_h-1589 and pSil2.1_h-Neg in the tumor tissue. PCR primers flanking the shRNA insert were used to determine the presence of the plasmid and insert size simultaneously. The presence of the plasmid and correct insert size in the harvested tumor tissue were detectable after 120 days (data not shown). Immunohistochemical staining of the tumors for MIB-1, a measure of cell proliferation, revealed that the U251_1589 cells have significantly fewer cell division events, indicating they are not dividing as rapidly as the U251_Neg-8 tumors (Fig. 4C and D).


Figure 4
View larger version (45K):
[in this window]
[in a new window]

  		Fig. 4. A, graph of stable transfectant glioma (U251) tumor volume as measured by digital calipers in a nude mouse flank tumor model. Tumor volumes of cells transfected with all pSil2.1_h-1589 clones (clones 4, 7, and 11) are significantly smaller (62-73%) than those transfected with pSil2.1_h-Neg (clones 8 and 9) or wild-type (*P < 0.0001, one-way ANOVA, n = 10 mice per group). Tukey's multiple comparison test verifies this (P < 0.01). B, Western blot of nuclear protein extracted from flank tumors. Positive control is hypoxic U251 nuclear extract. Lanes 2, 4, and 6, tumors expressing shRNA-1589; lanes 3, 5, and 7, tumors expressing shRNA-Neg. shRNA-1589 and shRNA-Neg cells grown in the same mouse are grouped together. Multiple bands are common in tumor-extracted protein because of rapid HIF-1{alpha} degradation during extraction procedure. C, proliferation index by MIB-1 immunohistochemical staining demonstrating significant differences between negative control (U251_Neg-8, n = 68) and U251_1589-7 (n = 52) cells (*P < 0.0001, t test). D, representative slides taken at x400 showing tumor staining and morphology. Arrows, MIB-1 staining.

 
Transient siRNA treatment to knock down HIF-1{alpha} reduces tumor volume in vivo in U251, U87, and U373 cells. To test the ability of transient HIF-1{alpha} siRNA treatment to affect tumor growth, we injected 1 x 106 U251-HRE cells in mouse flanks. These cells have a luciferase gene under control of a modified thymidine kinase promoter with three HREs and consistently express luciferase in a hypoxia- and HIF-1–dependent manner (26). The cells were allowed to establish tumors for 2 weeks postinjection and then were treated thrice per week with intratumoral siRNA injections, as done previously (27). The mice were imaged weekly and measured twice a week. Before injections were begun, the mice were imaged to ensure the tumors were expressing luciferase and to establish the optimal time for luminescent imaging (data not shown). Figure 5A shows a representative image of the treated and control groups. Quantization of these images taken over the course of treatment shows that the siRNA 1589 has a significant effect on the expression of luciferase and, by inference, HIF-1{alpha}, reducing it by 77% over tumors injected with the negative control siRNA by day 15 (Fig. 5B). Because of concern that the HRE construct may be interfering with in vivo growth, the experiment was repeated using U251 wild-type cells. The measured tumor volume is also significantly reduced (50% by day 57) in these mice, although there is no definite separation in the groups until day 37, a full week after the shRNA group showed significant separation (Fig. 6A ). This is expected, given that the U251_1589 tumors have a constitutive HIF-1{alpha} reduction, whereas the U251 and U251-HRE tumors were allowed to grow for 2 weeks before treatment with siRNA. To test the efficacy of the 1124 and 1659 siRNAs in vivo, additional injection experiments were done (Fig. 6A). These siRNAs seem to function identically to siRNA 1589 in reducing tumor growth. To address concerns that this response may be specific to U251 cells, siRNA flank injections were repeated in U87 and U373 cells. These cells responded in like manner to siRNA injections knocking out HIF-1{alpha} in U251 cells (Fig. 6D). They also had a similar reduction in VEGF, GLUT-1, and HIF-1 protein levels (data not shown). It is surprising that all of these tumors also show similar proliferation rates as determined by MIB-1 staining to those of the shRNA clones (Fig. 6B and C; U251 data representative of other cell lines, not shown).


Figure 5
View larger version (36K):
[in this window]
[in a new window]

  		Fig. 5. A, image of luciferase activity in vivo on day 15 of thrice weekly intratumoral injection time course. Mice were injected on Monday, Wednesday, and Friday of each week and imaged 20 h after the Friday injection. Mice on the left were injected with siRNA 1589, whereas those on the right were injected with the scrambled siRNA control. B, graph quantifying the average radiance (normalized) of the glioma (U251-HRE) tumor cells imaged during the siRNA injection time course. Day 0 is the baseline luciferase activity before siRNA injection. The luciferase activity is significantly less in the group injected with siRNA 1589 by day 15 (*P = 0.037, t test, n = 5 mice per group).

 

Figure 6
View larger version (57K):
[in this window]
[in a new window]

  		Fig. 6. A, graphs showing mouse flank (U251 cells) tumor volume as measured by digital calipers. The volume of tumors injected with siRNAs 1589 (left) or 1124 or 1659 (right) are significantly smaller (50-62%) than that of those injected with PBS + jetPEI or siRNA-Neg (*P < 0.0001, n = 10 per group; **P = 0.0008, n = 5 per group; both are verified by repeated-measures ANOVA and Tukey's multiple comparison test, P < 0.01). B, graph showing proliferation index by MIB-1 immunohistochemical staining demonstrating significant differences between siRNA-Neg (n = 27) and siRNA 1589 (n = 37) injected tumors (*P < 0.001). C, representative slides taken at x400 showing tumor staining and morphology. Arrow, MIB-1 staining. D, graphs showing tumor volume of U87 (left) and U373 (right) cells grown in mouse flanks and injected with siRNA (Neg, 1589) or PBS + jetPEI. U87 cells and U373 cells both responded significantly to siRNA-1589 injections when compared with siRNA-Neg or carrier (*P = 0.0002, n = 5 per group; **P = 0.0001, n = 5 per group; both are verified by repeated-measures ANOVA and Tukey's multiple comparison test, P < 0.01).

 

   Discussion
Top
 Abstract
 Materials and Methods
 Results
 Discussion
References
 
The results of other studies have shown that reducing HIF-1{alpha} constitutively or through an inducible shRNA construct can negatively affect tumor growth. However, such constitutive expression or inducible systems are not directly applicable in a clinical setting. This is the first report using direct siRNA injections in vivo to reduce glioma growth, supporting the hypothesis that HIF-1{alpha} is a major activator of VEGF production in hypoxic glioma cells in vitro, and GLUT-1 in glioma cells grown in the mouse flank. It is unclear whether the inhibition of growth is caused by lower levels of HIF-1{alpha} directly or if the subsequent reduction of VEGF and/or GLUT-1 is the most important component. Previous studies have shown that a reduction of VEGF levels results in reduced tumor growth but does not seem to be as effective at this reduction as our findings indicate HIF-1{alpha} is (28). The authors of recent studies have revived Warburg's hypothesis and targeting glycolysis as a possible anticancer therapy. In several successful studies, the authors have found that targeting glycolysis can attenuate solid tumors (19, 29, 30). It is possible that the reduction of GLUT-1 has a major effect on glioma growth. Our results in U251, U87, and U373 cells confirm what has been found previously in D45MG glioma cells (10). Although we and others (25) have shown that most GBM tumors have elevated GLUT-1 levels, the effect of specifically targeting GLUT-1 in GBM is unknown. These results would indicate this as a potential target for future study.

Our results using RNAi to knock down HIF-1{alpha} indicate that the protein is a required factor for sustained GBM tumor growth. The replication of this with three different siRNAs indicates it is not produced by an off-target effect. This is consistent with previous studies in other cancer types and supports the hypothesis that HIF-1{alpha} is a growth requirement for many, if not all, solid tumors (10, 11, 31–33). Our results using siRNA injections also indicate that the reduction of HIF-1{alpha} does not need to be constitutive to affect tumor growth significantly (Fig. 6).

The use of a luciferase reporter system allows for noninvasive, real-time imaging of intratumoral hypoxia in vivo. This was used to test the effectiveness of siRNA silencing of HIF-1{alpha}. We show that an injected siRNA, complexed to JetPEI, is an effective silencer of HIF-1{alpha} expression and inhibitor of glioma growth in vivo. This mode of delivery has direct clinical application because the treatment is transient and does not involve the use of viral vectors or other permanent DNA-modifying procedures. Although the reduction of HIF-1{alpha} and associated glioma growth was very substantial using shRNA, the use of these shRNA-expressing vectors in humans is problematic and potentially dangerous, as it involves permanent changes to the cell DNA. The long-term effects of silencing HIF-1{alpha} constitutively in brain tissue are also unknown.

HIF-1{alpha} is not normally highly expressed in the adult nervous system (34), making it a possible target for angiogenic therapy. Because of its function in activating glycolytic proteins, it is a potential target for cancer cells that have switched to glycolysis for energy production. In a recent report, however, Acker et al. (12) suggested that inactivating HIF-1{alpha} or HIF-2{alpha} can actually increase tumor growth. Although the effects of HIF-2{alpha} reduction are beyond the scope of this article, it could be that this reversal of results from HIF-1{alpha} reduction are cell type specific. Indeed, when comparing the response of U251, U87, and U373 cells to that of siRNA 1589, it is apparent that U373 cells did not respond to treatment to the same extent that the U87 and U251 cells did (Fig. 6A and D). It is possible that the rat GS9L cell line used by Acker et al. responds in a different manner than the human glioma cell lines used by us and others. Another possibility is that our siRNA was much more effective at knocking down HIF-1{alpha} in hypoxic conditions. The dominant negative construct they used interfered with both HIF-1{alpha} and HIF-2{alpha}, so the specific question of what happens when only HIF-1{alpha} is removed in hypoxia was never addressed.

Blouw et al. (35) found that, whereas inactivation of HIF-1{alpha} in the mouse flank results in attenuated growth of astrocytomas, it has the reverse effect when they are grown intracranially. This study used a HIF-1{alpha} knockout astrocytoma cell model that may not accurately reproduce the formation of GBM in vivo. Although astrocytoma cells commonly transform into GBM, the implanted knockout cells have no initial HIF-1{alpha} activity. Because of this, the tumor cells are almost completely dependent on co-opted blood vessels during their initial growth phase. This requires the tumor to become much more invasive, as Blouw et al. observed. It is not known whether this same effect will occur in tumors that are initially established with normal or elevated HIF-1{alpha} levels (as found in all patients presenting with GBM) and then subjected to HIF-1{alpha} reduction. We have begun to address this question by using siRNA, shRNA, and an inducible shRNA system in an intracranial model of GBM; however, preliminary results are not available to report. This finding of differential tumor growth raises the question of the validity of the tumor flank model because of its relatively unvascularized environment when compared with the brain. This model is very popular because of its ease of use, but it does have limitations when used to model a tumor of the brain, such as GBM.

In summary, our results show that using RNAi to reduce HIF-1{alpha} levels in several glioma cell lines significantly reduces the production of VEGF and GLUT-1, cellular proliferation, and tumor growth in vivo in three different glioma cell lines. We investigated the possibility of direct intratumoral injections and show that this approach may be a useful clinical treatment regimen when transient siRNA is used.


   Acknowledgments
 
We thank Dr. G. Melillo for providing U251-HRE cells, Sheryl Tripp for the superb immunohistochemistry stains, Drs. Stephen Lessnick and Eric Huang for useful comments, Kristin Kraus for her excellent editorial guidance, and the anonymous reviewers for their helpful suggestions regarding the manuscript.


   Footnotes
 
Grant support: Huntsman Cancer Institute funding award (R.L. Jensen).

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.

Note: Current address for K. Whang: Wonju College of Medicine, Yonsei University, Wonju, Korea.

3 http://www.ncbi.nlm.nih.gov/BLAST Back

Received 11/ 9/06; revised 1/16/07; accepted 1/31/07.


   References
Top
 Abstract
 Materials and Methods
 Results
 Discussion
 References
 

  1. Kleihues P, Louis DN, Scheithauer BW, et al. The WHO classification of tumors of the nervous system. J Neuropathol Exp Neurol 2002;61:215–25; discussion 26–9.[Medline]
  2. Berkman RA, Merrill MJ, Reinhold WC, et al. Expression of the vascular permeability factor/vascular endothelial growth factor gene in central nervous system neoplasms. J Clin Invest 1993;91:153–9.[Medline]
  3. Ke LD, Shi YX, Im SA, Chen X, Yung WK. The relevance of cell proliferation, vascular endothelial growth factor, and basic fibroblast growth factor production to angiogenesis and tumorigenicity in human glioma cell lines. Clin Cancer Res 2000;6:2562–72.[Abstract/Free Full Text]
  4. Forsythe JA, Jiang BH, Iyer NV, et al. Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol 1996;16:4604–13.[Abstract]
  5. Zhong H, De Marzo AM, Laughner E, et al. Overexpression of hypoxia-inducible factor 1{alpha} in common human cancers and their metastases. Cancer Res 1999;59:5830–5.[Abstract/Free Full Text]
  6. Kaur B, Khwaja FW, Severson EA, Matheny SL, Brat DJ, Van Meir EG. Hypoxia and the hypoxia-inducible-factor pathway in glioma growth and angiogenesis. Neuro-oncol 2005;7:134–53.[Abstract]
  7. Maxwell PH, Dachs GU, Gleadle JM, et al. Hypoxia-inducible factor-1 modulates gene expression in solid tumors and influences both angiogenesis and tumor growth. Proc Natl Acad Sci U S A 1997;94:8104–9.[Abstract/Free Full Text]
  8. Ryan HE, Poloni M, McNulty W, et al. Hypoxia-inducible factor-1{alpha} is a positive factor in solid tumor growth. Cancer Res 2000;60:4010–5.[Abstract/Free Full Text]
  9. Zagzag D, Zhong H, Scalzitti JM, Laughner E, Simons JW, Semenza GL. Expression of hypoxia-inducible factor 1{alpha} in brain tumors: association with angiogenesis, invasion, and progression. Cancer 2000;88:2606–18.[CrossRef][Medline]
  10. Li L, Lin X, Staver M, et al. Evaluating hypoxia-inducible factor-1{alpha} as a cancer therapeutic target via inducible RNA interference in vivo. Cancer Res 2005;65:7249–58.[Abstract/Free Full Text]
  11. Rapisarda A, Zalek J, Hollingshead M, et al. Schedule-dependent inhibition of hypoxia-inducible factor-1{alpha} protein accumulation, angiogenesis, and tumor growth by topotecan in U251-HRE glioblastoma xenografts. Cancer Res 2004;64:6845–8.[Abstract/Free Full Text]
  12. Acker T, Diez-Juan A, Aragones J, et al. Genetic evidence for a tumor suppressor role of HIF-2{alpha}. Cancer Cell 2005;8:131–41.[CrossRef][Medline]
  13. Carmeliet P, Dor Y, Herbert JM, et al. Role of HIF-1{alpha} in hypoxia-mediated apoptosis, cell proliferation and tumour angiogenesis. Nature 1998;394:485–90.[CrossRef][Medline]
  14. Semenza GL. Regulation of mammalian O2 homeostasis by hypoxia-inducible factor 1. Annu Rev Cell Dev Biol 1999;15:551–78.[CrossRef][Medline]
  15. Richard DE, Berra E, Pouyssegur J. Angiogenesis: how a tumor adapts to hypoxia. Biochem Biophys Res Commun 1999;266:718–22.[CrossRef][Medline]
  16. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci U S A 1995;92:5510–4.[Abstract/Free Full Text]
  17. Warburg O. On the origin of cancer cells. Science 1956;123:309–14.[Free Full Text]
  18. Robey IF, Lien AD, Welsh SJ, Baggett BK, Gillies RJ. Hypoxia-inducible factor-1{alpha} and the glycolytic phenotype in tumors. Neoplasia 2005;7:324–30.[CrossRef][Medline]
  19. Xu RH, Pelicano H, Zhou Y, et al. Inhibition of glycolysis in cancer cells: a novel strategy to overcome drug resistance associated with mitochondrial respiratory defect and hypoxia. Cancer Res 2005;65:613–21.[Abstract/Free Full Text]
  20. Lu H, Forbes RA, Verma A. Hypoxia-inducible factor 1 activation by aerobic glycolysis implicates the Warburg effect in carcinogenesis. J Biol Chem 2002;277:23111–5.[Abstract/Free Full Text]
  21. Fire A, Xu S, Montgomery MK, Kostas SA, Driver SE, Mello CC. Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature 1998;391:806–11.[CrossRef][Medline]
  22. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells. Nature 2001;411:494–8.[CrossRef][Medline]
  23. Duxbury MS, Whang EE. RNA interference: a practical approach. J Surg Res 2004;117:339–44.[CrossRef][Medline]
  24. Yeung HY, Lai KP, Chan HY, Mak NK, Wagner GF, Wong CK. Hypoxia-inducible factor-1-mediated activation of stanniocalcin-1 in human cancer cells. Endocrinology 2005;146:4951–60.[Abstract/Free Full Text]
  25. Jensen RL, Ragel BT, Whang K, Gillespie D. Inhibition of hypoxia inducible factor-1{alpha} (HIF-1{alpha}) decreases vascular endothelial growth factor (VEGF) secretion and tumor growth in malignant gliomas. J Neurooncol 2006;78:233–47.[CrossRef][Medline]
  26. Rapisarda A, Uranchimeg B, Scudiero DA, et al. Identification of small molecule inhibitors of hypoxia-inducible factor 1 transcriptional activation pathway. Cancer Res 2002;62:4316–24.[Abstract/Free Full Text]
  27. Urban-Klein B, Werth S, Abuharbeid S, Czubayko F, Aigner A. RNAi-mediated gene-targeting through systemic application of polyethylenimine (PEI)-complexed siRNA in vivo. Gene Ther 2005;12:461–6.[CrossRef][Medline]
  28. Detwiller KY, Fernando NT, Segal NH, Ryeom SW, D'Amore PA, Yoon SS. Analysis of hypoxia-related gene expression in sarcomas and effect of hypoxia on RNA interference of vascular endothelial cell growth factor A. Cancer Res 2005;65:5881–9.[Abstract/Free Full Text]
  29. Pelicano H, Martin DS, Xu RH, Huang P. Glycolysis inhibition for anticancer treatment. Oncogene 2006;25:4633–46.[CrossRef][Medline]
  30. Fantin VR, St-Pierre J, Leder P. Attenuation of LDH-A expression uncovers a link between glycolysis, mitochondrial physiology, and tumor maintenance. Cancer Cell 2006;9:425–34.[CrossRef][Medline]
  31. Hanze J, Eul BG, Savai R, et al. RNA interference for HIF-1{alpha} inhibits its downstream signalling and affects cellular proliferation. Biochem Biophys Res Commun 2003;312:571–7.[CrossRef][Medline]
  32. Kung AL, Wang S, Klco JM, Kaelin WG, Livingston DM. Suppression of tumor growth through disruption of hypoxia-inducible transcription. Nat Med 2000;6:1335–40.[CrossRef][Medline]
  33. Zhang Q, Zhang ZF, Rao JY, et al. Treatment with siRNA and antisense oligonucleotides targeted to HIF-1{alpha} induced apoptosis in human tongue squamous cell carcinomas. Int J Cancer 2004;111:849–57.[CrossRef][Medline]
  34. Talks KL, Turley H, Gatter KC, et al. The expression and distribution of the hypoxia-inducible factors HIF-1{alpha} and HIF-2{alpha} in normal human tissues, cancers, and tumor-associated macrophages. Am J Pathol 2000;157:411–21.[Abstract/Free Full Text]
  35. Blouw B, Song H, Tihan T, et al. The hypoxic response of tumors is dependent on their microenvironment. Cancer Cell 2003;4:133–46.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Correction (v14,p326)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gillespie, D. L.
Right arrow Articles by Jensen, R. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gillespie, D. L.
Right arrow Articles by Jensen, R. L.

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