| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cell and Tumor Biology |
and CXCR4 Expression in Hemangioblastoma and Clear Cell-Renal Cell Carcinoma: von Hippel-Lindau Loss-of-Function Induces Expression of a Ligand and Its Receptor
1 Microvascular and Molecular Neuro-oncology Laboratory, 2 Department of Pathology, 3 Division of Neuropathology, 4 Department of Neurosurgery, and 5 New York University Cancer Institute of New York University School of Medicine, New York, New York and 6 Vascular Program, Institute for Cell Engineering; Departments of Pediatrics, Medicine, Oncology, and Radiation Oncology; and Institute of Genetic Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland
Requests for reprints: David Zagzag, Department of Pathology, Division of Neuropathology, New York University Medical Center, 550 First Avenue, New York, NY 10016. Phone: 212-263-6449; Fax: 212-263-8994; E-mail: dz4{at}nyu.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
(HIF-1). In hemangioblastomas and CC-RCCs, HIF-1 is constitutively overexpressed leading to increased transcription of HIF-1–regulated genes, including vascular endothelial growth factor (VEGF). Because loss of VHL function is associated with increased expression of the chemokine receptor CXCR4 in CC-RCCs, we investigated the expression of HIF-1, CXCR4, and its ligand stromal cell–derived factor-1 (SDF-1) in hemangioblastomas and CC-RCCs. Immunohistochemistry revealed overexpression of both CXCR4 and SDF-1 within tumor cells and endothelial cells of hemangioblastomas and CC-RCCs. HIF-1 was detected in tumor cell nuclei of both hemangioblastomas and CC-RCCs. A specific ELISA showed that hemangioblastomas and CC-RCCs expressed SDF-1 protein at levels that were significantly higher than those found in normal tissue. Analysis of the VHL-null RCC line 786-0 revealed that SDF-1 mRNA levels were 100-fold higher than in a subclone transfected with the wild-type VHL gene. Expression of CXCR4 and SDF-1 mRNA was significantly decreased in HIF-1-null compared with wild-type mouse embryo fibroblasts (MEFs). ELISA and Western blot studies for SDF-1 and CXCR4 protein expression confirmed the RNA findings in RCC lines and MEFs. These results suggest that loss-of-function of a single tumor suppressor gene can up-regulate the expression of both a ligand and its receptor, which may establish an autocrine signaling pathway with important roles in the pathogenesis of hemangioblastoma and CC-RCC. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CC-RCCs are the leading cause of morbidity and mortality in VHL patients (3). CC-RCCs represent 
75% of kidney carcinomas (4). Although the vast majority of CC-RCCs are sporadic or noninherited, there is a small proportion (4%) that is hereditary, including those associated with the VHL syndrome (4).
Hemangioblastomas are usually well circumscribed and slowly growing. They are composed of stromal cells, endothelial cells, pericytes, and mast cells. The endothelial and pericytic components are obviously vascular, but many studies have failed to resolve the issue of the origin of the stromal cells, believed to be the true neoplastic cells of hemangioblastomas, and the histogenesis of the hemangioblastomas remains unclear (5). Sporadic hemangioblastomas are typically single, often cystic neoplasms with one or multiple mural nodules, located in the cerebellum, and appearing at an average age greater than 40 years. In contrast, hemangioblastomas of the CNS in the context of VHL syndrome tend to be multiple and present 10 years earlier (6).
Hypoxia-inducible factor-1 (HIF-1) is up-regulated in stromal cells of hemangioblastomas (7) and tumor cells of CC-RCCs (8, 9). HIF-1, a ubiquitously expressed and highly conserved heterodimeric basic-helix-loop-helix-PAS transcription factor composed of HIF-1 and HIF-1ß subunits, plays an essential role in oxygen homeostasis (10). As cellular oxygen concentration decreases, HIF-1 levels increase, thus determining the level of HIF-1 activity (10). HIF-1 activates a large battery of genes whose protein products function either to increase oxygen availability or to allow metabolic adaptation to oxygen deprivation. HIF-1 target genes include those encoding erythropoietin, vascular endothelial growth factor (VEGF), glucose transporters, and glycolytic enzymes. Such genes share the presence of hypoxia response elements, which contain binding sites for HIF-1 (10).
VHL binds to the HIF-1 subunit and targets it for ubiquitination and degradation (11). In the presence of oxygen, HIF-1 is hydroxylated at proline residues 402 and/or 564, which is necessary and sufficient for binding of VHL (12). Under hypoxic conditions, HIF-1 is not prolyl hydroxylated and escapes recognition by VHL. Loss-of-function resulting from mutations or silencing of the VHL gene by hypermethylation has been reported as a frequent occurrence in hemangioblastomas and CC-RCCs (13). HIF-2 is a related hypoxia-inducible protein that, like HIF-1, dimerizes with HIF-1ß and is overexpressed in VHL-null cells.
Stromal cell–derived factor-1 (SDF-1), also known as CXCL12, is the only ligand of the chemokine receptor CXCR4, which is also known as fusin (14). CXCR4 and SDF-1 are required for normal embryonic development of the nervous, hematopoietic, and cardiovascular systems. Knockout mice die in utero with defects in neuronal migration in the cerebellum, ventricular septation in the heart, vascularization of the gastrointestinal tract, and B-cell lymphopoiesis and myelopoiesis (15). SDF-1 and CXCR4 are involved in the trafficking of hematopoietic progenitor and stem cells. For example, SDF-1 is chemotactic for CD34+ progenitor cells, induces proliferation of B-cell progenitors, and regulates leukocyte and endothelial precursor migration (16). Angiogenic effects of SDF-1 have been shown both ex vivo and in vitro (17–19).
Because CXCR4 gene expression has been reported to be positively regulated by HIF-1 and negatively regulated by VHL (20), we examined SDF-1, CXCR4, and HIF-1 protein expression in tissues from 22 hemangioblastomas and 15 CC-RCCs by immunohistochemistry and ELISA. In addition, CXCR4 and SDF-1 mRNA and protein expression was analyzed in VHL-null versus VHL-rescued CC-RCC lines and HIF-1-null versus wild-type mouse embryo fibroblasts (MEF). These studies revealed that the expression of both CXCR4 and SDF-1 is associated with loss of VHL function. Our data suggest that loss-of-function of a single tumor suppressor gene (VHL) up-regulates the expression of a ligand (SDF-1) and its receptor (CXCR4), thus identifying an autocrine signaling pathway that may play important roles in the pathogenesis of VHL-null neoplasms.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Immunohistochemistry for stromal cell–derived factor-1 and CXCR4. Formalin-fixed, paraffin-embedded tissue was prepared from surgical excisions using conventional histologic methods. Serial sections (6 µm) were cut from each paraffin block and one was stained with H&E. For immunohistochemistry, tissue sections were deparaffinized in xylene (three changes), rehydrated through graded alcohol (three changes of 100% ethanol and three changes of 95% ethanol), and rinsed in distilled water. Serial sections were stained for SDF-1 and CXCR4 using one or two antibodies, respectively. Heat-induced epitope retrieval was done by boiling in 10 mmol/L citrate buffer (pH 6.0) for 20 minutes for SDF-1 and 10 minutes for CXCR4. The sections were allowed to cool to room temperature for 30 minutes, rinsed in distilled water, washed in 0.05 mol/L Tris-HCl (pH 7.6) containing 0.3 mol/L NaCl and 0.1% Tween 20, and blocked for endogenous peroxidase with 3% H2O2. SDF-1 was detected with mouse anti-SDF-1 monoclonal antibody (mAb; clone 7801B, IgG1 class, R&D Systems, Minneapolis, MN) diluted 1:50. CXCR4 was detected first with goat polyclonal anti-CXCR4 antibody directed against the amino terminus of the molecule (sc-6279, Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1:50. The cases were also stained with a rabbit polyclonal anti-CXCR4 diluted 1:10 (IMG-537, Imgenex, San Diego, CA) that recognizes amino acids 328 to 338 of human CXCR4. All primary antibodies were incubated overnight at room temperature. The detection of SDF-1 and CXCR4 (when done with the Imgenex antibody) used prediluted biotinylated goat anti-mouse or anti-rabbit antibodies, respectively (Ventana Medical Systems, Tucson, AZ). When CXCR4 was detected with the Santa Cruz Biotechnology antibody, a biotinylated horse anti-goat antibody (Vector Laboratories Burlingame, CA) diluted 1:200 was used. In each case, the secondary antibody was applied for 32 minutes followed by horseradish peroxidase (HRP)–conjugated streptavidin with 3,3'-diaminobenzidine as the chromogen. Detection was carried out at 37°C on a NexES instrument (Ventana Medical Systems). Slides were washed in distilled water. Nuclei were lightly counterstained with hematoxylin, dehydrated, and mounted with permanent medium. Control procedures included isotype-matched murine mAb and rabbit polyclonal of irrelevant specificity. In addition, to rule out nonspecific avidin-biotin–induced immunoreactivity, we did serial dilution experiments for both primary and secondary antibodies. Each series included a single (no primary antibody) and double (no primary or secondary antibody) negative controls.
Immunohistochemistry for hypoxia-inducible factor-1. Flanking sections (6 µm) were stained for HIF-1 using the Catalyzed Signal Amplification System (DAKO, Carpinteria, CA) as described (7, 8). After deparaffinization and rehydration, slides were treated with target retrieval solution (DAKO) at 97°C for 45 minutes; then, the manufacturer's instructions were followed. Nuclei were counterstained with hematoxylin. Negative controls were done with nonimmune serum or PBS used instead of primary antibody. The mAb against HIF-1 (1 mg/mL, H167, Novus Biologicals, Littleton, CO) was diluted to 1:1,000. Automated analysis was done on a Tech-Mate 100 automated stainer (Ventana-BioTek Solutions, Inc., Tucson, AZ).
Histologic assessment. Three investigators (D.Z., H.Y., and J.M.) independently evaluated the immunohistochemistry. When the evaluations differed, the final decision was made by consensus. The immunohistochemical analysis of SDF-1, CXCR4, and HIF-1 was scored as follows: –, no staining; +, staining of <1% of cells; ++, staining of 1% to 10% of cells; +++, staining of 10% to 50% of cells; and ++++, staining of > 50% of cells. Vascular and tumor cells were assessed separately. Immunoreactivity within adjacent normal tissue was also examined.
Cell culture. VHL-null RCC lines RCC4 and 786-0 and subclones stably transfected with an expression vector encoding VHL and neomycin resistance [i.e., RCC4T-3-13 (gift of Celeste Simon, University of Pennsylvania, Philadelphia, PA) and 786-0+VHL (provided by William Kaelin, Dana-Farber Cancer Institute, Boston, MA)] were maintained in DMEM containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin (22, 23). G418 (1 mg/mL) was added to the medium of RCC4T-3-13 and 786-0+VHL cells. Wild-type and Hif1a–/– MEFs were cultured in high-glucose DMEM with 15% FBS and 1% each of penicillin/streptomycin, sodium pyruvate, and nonessential amino acids (Invitrogen, Carlsbad, CA; ref. 24).
Immunoblot assays. MEFs, 786-0 cells, and RCC4 cells were exposed to 20% O2 (95% air/5% CO2) or 1% O2 (in a modular incubator flushed with a gas mixture containing 1% O2, 5% CO2, and 94% N2) for 4 hours and lysed in radioimmunoprecipitation assay buffer. Lysate (100 µg) was fractionated by 7% SDS-PAGE and transferred to a nitrocellulose membrane. Immunoblot assays were done using mouse mAb H167 against HIF-1 (8) or rabbit polyclonal antibodies against HIF-2 (NB-100-122, Novus Biologicals) at 1:500 dilution for 2 hours at room temperature. HRP-conjugated anti-mouse and anti-rabbit secondary antibodies were used to detect the signals. The blots were stripped and reprobed with mAb H1ß234 against HIF-1ß (7) at 1:1,000 dilution. For CXCR4 detection, cells were seeded at low density and HEPES (pH 7.4) was added to 50 mmol/L before exposing the cells to 20% or 1% O2 for 48 hours. Lysate (20 µg) was fractionated by 10% SDS-PAGE and transferred to nitrocellulose membrane. The membrane was probed with rabbit polyclonal anti-CXCR4 (IMG-537) also at 1:500 dilution overnight at 4°C. The same antibody was also used for immunohistochemistry. Goat polyclonal anti-ß-actin antibody (sc-1616, Santa Cruz Biotechnology) was used as a loading control at 1:3,000 dilution with detection using HRP-conjugated donkey anti-goat secondary antibody.
Preparation of nuclear extracts. RCC cells were grown in DMEM with 10% FCS, 6.6 mmol/L L-glutamine, and 50 µg/mL gentamicin, exposed to 1% O2 for 4 hours, trypsinized, and washed twice with ice-cold PBS. Nuclear extracts were prepared as described (25), except that 1 µg/mL aprotinin, 1 µg/mL leupeptin, 100 mmol/L NaF, and 10 mmol/L tetrasodium pyrophosphate were added to buffers.
Quantitative real-time reverse transcription-PCR. Cells (1.5 x 106) were seeded onto a 15-cm dish and exposed to 20% or 1% O2 for 24 hours. RNA was isolated using QIAshredder and RNeasy Mini kits (Qiagen, Valencia, CA). Primers were designed using Beacon Designer 2.1 software and cDNA was prepared using the iScript cDNA Synthesis kit (Bio-Rad, Hercules, CA). cDNA samples were diluted 1:15 and real-time PCR was done using iQ SYBR Green Supermix and the iCycler Real-time PCR Detection System (Bio-Rad). For each primer pair (sequences available on request), annealing temperature was optimized by gradient PCR. The fold change in expression of each target mRNA relative to 18S rRNA was calculated based on the threshold cycle (Ct) as 2–
(Ct), where Ct = Cttarget – Ct18S and (Ct) = Ct1% – Ct20%.
Stromal cell–derived factor-1 ELISA. The quantity of human SDF-1 present in tissue homogenates in hemangioblastomas, CC-RCCs, and normal brain and kidney was determined by specific ELISA. Briefly, tissue sample (0.4 mg) was powdered in liquid nitrogen and homogenized in 2 mL T-PER Tissue Protein Extraction Reagent (Pierce, Rockford, IL) in the presence of protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN). Extracts were centrifuged at 10,000 rpm for 1 hour, supernatant was collected, and protein concentration was measured by BCA assay kit (Pierce). Extracts were analyzed using the human SDF-1 Quantikine kit (R&D Systems). SDF-1 protein expression was also examined in conditioned medium from RCC cell lines.
Statistical analysis. Statistical differences between the means for the different groups were evaluated by one-way ANOVA using StatView 4.5 (Abacus Concepts, Berkeley, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Immunohistochemistry for stromal cell–derived factor-1 and CXCR4 in hemangioblastomas. SDF-1 expression was noted in 21 of 22 cases (Table 1). Stromal cells showed staining in all 21 cases. In 8 cases, SDF-1 was localized to the cytoplasm only, whereas in 13 cases both cytoplasmic and nuclear localization was detected (Fig. 2A and B). SDF-1 immunoreactivity highlighted cellular contours in a pattern consistent with membranous staining in one case with 4+ nuclear and cytoplasmic staining. Of the cases showing only cytoplasmic staining, 5 were 4+, 2 were 3+, and 1 was 2+. Of the cases showing nuclear and cytoplasmic staining, 12 were 4+. SDF-1 was also detected in the cytoplasm of vascular cells (Fig. 2B).
|
|
(7), high levels of CXCR4 and SDF-1 were detected in stromal cells of hemangioblastomas.
The paucity of adjacent normal CNS tissue precluded a definitive analysis of the expression of SDF-1 and CXCR4. However, in 11 cases where adjacent attenuated and compressed cerebellar or spinal cord tissue was present on the sections, there was no immunoreactivity for SDF-1, except for rare vascular channels that showed weak SDF-1 and CXCR4 immunoreactivity as described previously (27). Staining was not observed with isotype-matched murine mAb controls or in the absence of primary antibody.
Immunohistochemistry for stromal cell–derived factor-1 and CXCR4 in clear cell-renal cell carcinomas. The CC-RCCs had 4+ SDF-1 expression with nuclear staining in all cases (Fig. 3A; Table 2). Cytoplasmic staining was also seen in 4 of 15 cases. Membranous staining for SDF-1 was observed in 2 cases (Fig. 3A). Most blood vessels had 4+ staining for SDF-1 (Fig. 3A). Only 2 of 15 cases had both nuclear and cytoplasmic staining, whereas 7 showed only cytoplasmic and 6 nuclear staining only. The border between tumor and adjacent kidney was demarcated by immunohistochemistry for SDF-1 and CXCR4 (data not shown). SDF-1 immunoreactivity was observed in normal kidney (Fig. 3B; Table 3) where glomerular capillary endothelial cells, visceral and parietal cells of the Bowman's capsule, and mesangial cells had variable nuclear and cytoplasmic expression of SDF-1 (Fig. 3B). Proximal and distal convoluted tubules had nuclear and cytoplasmic expression. There was greater staining intensity for SDF-1 in the distal, compared with the proximal, convoluted tubules (Fig. 3B). SDF-1 was also seen in the arcuate and lobular vascular channels. No staining was observed with isotype-matched murine mAb controls or in the absence of primary antibody.
|
|
|
and CXCR4 (data not shown).
The kidney is rich in biotin-containing enzymes that can cause false-positive labeling (28). To determine whether the staining of the normal kidney was the result of the endogenous biotin, we did serial dilution experiments for both primary and secondary antibodies. Each series of experiments included single (no primary antibody) and double (no primary and secondary antibody) negative controls. Sections stained with CXCR4 showed immunoreactivity in the tubular epithelium, whereas sections processed without primary antibody failed to show staining (data not shown). Staining was also absent when both primary and secondary antibodies were omitted (data not shown). The staining for SDF-1 was in distal tubules, whereas the highest biotin concentration is in proximal tubules (29) in which staining was absent. We therefore conclude that staining of normal tubular epithelium adjacent to the CC-RCCs for SDF-1 and CXCR4 is specific and not the result of endogenous biotin activity.
Expression of stromal cell–derived factor-1 in tissue homogenates of hemangioblastomas and clear cell-renal cell carcinomas by ELISA. Although CXCR4 expression in CC-RCCs and its regulation by VHL was already described (20), the expression of SDF-1 in hemangioblastomas and its regulation by VHL is a novel finding. To further explore this result, we examined SDF-1 expression in tissue homogenates of surgically obtained hemangioblastomas and CC-RCCs. We also studied normal brain and kidney tissue adjacent to the CC-RCCs. We performed ELISA for SDF-1 on tumor and normal tissues (30). All of the hemangioblastomas expressed SDF-1 protein at levels that were significantly higher than those found in normal brain (Fig. 1C). CC-RCCs also had levels of SDF-1 that were significantly higher than adjacent normal kidney (Fig. 1D). The results in normal kidney we report here are similar to those reported previously (30). The low levels of SDF-1 in normal brain tissue by ELISA are consistent with immunohistochemical data (17, 27).
Immunohistochemistry for hypoxia-inducible factor-1 in hemangioblastomas and clear cell-renal cell carcinomas. In all hemangioblastomas, stromal cells, including those adjacent to vascular channels, showed intense HIF-1 nuclear staining (Fig. 4A) as described previously (7). Tumor cells lining microcysts expressed HIF-1. No staining was observed in the absence of primary antibody (data not shown). Ten hemangioblastomas were graded as 4+, 4 were 3+, and 8 were 2+. In CC-RCCs, in accordance with published immunohistochemical observations (7), very little expression of HIF-1 was found in normal kidney tissue. In 13 of 15 cases, the staining pattern for HIF-1 was homogeneous, with virtually all tumor cell nuclei staining positive (Fig. 4A). In 2 cases, expression of HIF-1 was focal with either single cells or clusters of cells accumulating the protein in their nuclei. Thirteen CC-RCCs were graded as 4+ and 2 were 2+.
|
Expression of vascular endothelial growth factor, stromal cell–derived factor-1, and CXCR4 in clear cell-renal cell carcinoma lines. VHL-null 786-0 cells expressed high levels of HIF-2 under nonhypoxic and hypoxic conditions (Fig. 5A). When VHL expression was rescued by transfection, HIF-2 reverted to the normal oxygen-regulated pattern of expression. HIF-1 expression was not detected in 786-0 or 786-0+VHL cells. HIF-1ß expression was detected but unaffected by VHL status or oxygen concentration. In RCC4 and RCC4T-3-13 cells, a subclone stably transfected with VHL, the expression of both HIF-1 and HIF-2 was negatively regulated by VHL and oxygen. To determine whether increased expression of CXCR4 and SDF-1 was due to loss of VHL activity, we analyzed expression of the mRNAs encoding these proteins in VHL-null RCC lines 786-0 and RCC4 compared with the subclones 786-0+VHL and RCC4T-3-13. As described previously (22, 23), VEGF mRNA expression was increased in the VHL-null compared with the respective VHL-expressing subclones (Fig. 5B).
|
protein levels as determined by ELISA in 786-0 cells but not in RCC4 cells (Fig. 5C), whereas CXCR4 protein expression was VHL- and oxygen-regulated in RCC4 cells (Fig. 5D) but was not detected in 786-0 cells.
Expression of vascular endothelial growth factor, stromal cell–derived factor-1, and CXCR4 in mouse embryo fibroblasts. VHL negatively regulates VEGF mRNA expression by binding to HIF-1 and targeting it for ubiquitination and proteasomal degradation. We analyzed expression of CXCR4 and SDF-1 in MEFs that either were wild-type and expressed HIF-1 under hypoxic conditions or were homozygous for a knockout allele at the Hif1a locus and lacked HIF-1 expression (Fig. 6A). HIF-2 was not expressed in wild-type or HIF-1-null MEFs. Similar to VEGF, CXCR4 and SDF-1 mRNA expression was significantly reduced in HIF-1-null MEFs (Fig. 6B). Immunoblot assay showed that CXCR4 protein expression was hypoxia inducible in a HIF-1-dependent manner (Fig. 6C).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
are overexpressed in stromal cells of hemangioblastomas and in tumor cells of CC-RCCs. Vascular cells in both hemangioblastomas and CC-RCCs also expressed CXCR4 and SDF-1. Our results provide evidence that the VHL-HIF-1 pathway regulates the expression of CXCR4, a receptor,and SDF-1, its specific ligand, in hemangioblastomas and CC-RCCs. Within these tumors, at least three interconnected pathways, involving hypoxia-induced expression mediated by HIF-1, VHL loss-of-function, and VEGF cross-talk, may contribute to the overexpression of CXCR4 and SDF-1. In addition, our studies show cytosolic and nuclear localization of both CXCR4 and SDF-1 in hemangioblastomas and CC-RCCs.
Regulation of CXCR4 and stromal cell–derived factor-1 expression by hypoxia and hypoxia-inducible factor-1. Hypoxia has been shown to increase CXCR4 expression through HIF-1 activation in human monocytes, macrophages, endothelial cells, and cancer cells (20, 32). Induction of CXCR4 mRNA and protein expression in response to hypoxia is markedly reduced in HIF-1-null MEFs (Fig. 6). A cis-acting hypoxia response element has been identified in the CXCR4 gene, thus implicating HIF-1 in the direct regulation of CXCR4 (20). Hypoxic regulation of SDF-1 has been reported in cultured human synovial cells (33) and human umbilical vein endothelial cells (34). HIF-1 and SDF-1 expression colocalized in ischemic murine skin and were not detected in nonischemic tissue (34). HIF-1 binding to the SDF-1 promoter in hypoxic cells was shown (34). We observed coexpression of HIF-1, SDF-1, and CXCR4 in tumors. Taken together with previous studies, our data show that both CXCR4 and SDF-1 are regulated in an oxygen-dependent manner by HIF-1.
Regulation of CXCR4 and stromal cell–derived factor-1 expression by the von Hippel-Lindau protein. CXCR4 and SDF-1 were expressed throughout the highly vascularized and nonnecrotic tumors, including cells contiguous to vascular channels, suggesting that SDF-1 and CXCR4 expression was not driven by hypoxia, which occurs in cells >100 µm away from a functional blood supply (35), a condition that is rarely met in hemangioblastomas and CC-RCCs. Among the genes most strongly suppressed by the reintroduction of functional VHL into CC-RCC cells is CXCR4 (20). Taken together, these data provide evidence that expression of CXCR4 and SDF-1 is negatively regulated by VHL. The expression of either SDF-1 or CXCR4, but not both, in RCC lines stands in contrast to the expression of both ligand and receptor in most hemangioblastomas and CC-RCCs. Although it is not possible to draw definitive conclusions from two cell lines, these findings suggest that there may be a selection against coexpression in tissue culture. Thus, the impaired degradation of HIF-1 and HIF-2 in hemangioblastomas (7) and CC-RCCs (8, 36) leading to CXCR4 and SDF-1 expression is driven by a genetic alteration rather than a physiologic stimulus.
Regulation of CXCR4 and stromal cell–derived factor-1 expression by vascular endothelial growth factor. Hemangioblastomas and CC-RCCs express high levels of VEGF (1, 37, 38). Resting endothelial cells have low levels of CXCR4 mRNA that increase in response to VEGF (19), which stimulates migration toward SDF-1. VEGF also induces SDF-1 expression in human brain microvascular endothelial cells (39). The pattern of CXCR4 and SDF-1 expression observed in hemangioblastomas and CC-RCCs is similar to the pattern of HIF-1 and VEGF expression described previously (1, 37, 38). Experimental data suggest that angiogenesis mediated by SDF-1/CXCR4 is regulated at the receptor level by VEGF (32). Thus, SDF-1 or CXCR4 expression may be induced, at least in part, indirectly by VEGF. However, the identification of a hypoxia response element in the SDF-1 gene promoter (34) and our demonstration of decreased SDF-1 mRNA in HIF-1-null cells indicate that HIF-1 directly regulates expression of SDF-1.
Subcellular localization of CXCR4 and stromal cell–derived factor-1. The intracellular localization of these proteins in tumor cells of hemangioblastomas and CC-RCCs has been reported in other human cancers, including nuclear localization of CXCR4 in hepatocellular (40), breast (41), and non–small cell lung (42) carcinomas using several different antibodies. Nuclear and cytoplasmic staining for SDF-1 has been detected in prostate cancers (43). Intracellular expression of both CXCR4 and SDF-1 has been shown in acute myelogenous leukemia cells (44). These results suggest the involvement of CXCR4 and SDF-1 in autocrine and/or intracrine signaling pathways. Further studies are required to establish the functional significance of these observations.
Stromal cell–derived factor-1/CXCR4 as a novel angiogenic pathway in hemangioblastomas and clear cell-renal cell carcinomas. A hallmark of hemangioblastomas and CC-RCCs is their high degree of vascularization, which has been linked to the overexpression of VEGF (1, 37, 38), a critical hypoxia-inducible angiogenic factor (45). VHL loss-of-function is responsible for abnormal accumulation of VEGF in hemangioblastomas and CC-RCCs (1, 37, 38, 46). Our findings, in combination with previous studies, suggest that in addition to the VEGF/VEGF receptor axis the SDF-1/CXCR4 pathway might also be important in inducing the exuberant vascularization present in hemangioblastomas and CC-RCCs. CXCR4 is required for vascularization of the gastrointestinal tract (15). In addition, endothelial cells and their progenitors express CXCR4 (47). SDF-1 acts as a chemoattractant for endothelial cells and their progenitors, induces endothelial cell proliferation in vitro, and induces microvessel formation and capillary sprouting in vivo (19, 48). An angiogenic role of SDF-1 has been shown at sites of inflammation (48), in gliomas (17), and in ischemic tissue (34).
SDF-1 increases VEGF expression and enhances VEGF-induced proliferation of endothelial cells (48, 49). In addition, both VEGF and SDF-1 up-regulate EGR-1, a transcription factor that up-regulates expression of VEGF-receptor Flt-1 (50). Taken together, these studies suggest that SDF-1 and VEGF may collaborate to induce angiogenesis in hemangioblastomas and CC-RCCs. The chemotactic influence of SDF-1 (18, 19) combined with the proliferative effects of VEGF may represent a powerful angiogenic signal in these tumors.
In summary, we have shown that CXCR4 and SDF-1 are overexpressed in tumor and vascular cells of hemangioblastomas and CC-RCCs. Our results indicate that VHL negatively regulates both CXCR4 and SDF-1 expression. To the best of our knowledge, this is the first report suggesting that loss-of-function for a tumor suppressor (VHL) results in increased expression of both a receptor (CXCR4) and its ligand (SDF-1), thus establishing autocrine, intracrine, and/or paracrine signaling pathways that may play important roles in the pathogenesis of hemangioblastoma and CC-RCC. Additional studies are required to elucidate the consequences of CXCR4 and SDF-1 expression and to determine whether small-molecule antagonists of CXCR4 may have therapeutic effects in patients with hemangioblastoma or CC-RCC.
|
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 are grateful to Karen Padgett of Novus Biologicals Inc. (Littleton, CO) for providing anti-HIF 2 antibodies. We thank Bill Kaelin and Celeste Simon for providing cell lines and Eugene Lukyanov for his help in preparing the article.
Received 12/14/04. Revised 4/20/05. Accepted 4/28/05.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in brain tumors: association with angiogenesis, invasion, and progression. Cancer 2000;88:2606–18.[CrossRef][Medline]
in common human cancers and their metastases. Cancer Res 1999;59:5830–5. targeted for VHL-mediated destruction by proline hydroxylation: implications for O2 sensing. Science 2001;292:464–8. (HIF-1) and HIF-2 in hypoxic gene regulation. Mol Cell Biol 2003;23:9361–74. and insulin-like growth factor 2. Cancer Res 1999;59:3915–8. in clear cell renal carcinomas. Cancer Res 2001;61:5215–22. expression in renal cell carcinomas. J Urol 2003;170:588–92.[CrossRef][Medline]
induced gene-expression in human endothelial cells. Mol Cell Probes 2003;17:245–52.[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  P. L. Wagner, E. Hyjek, M. F. Vazquez, D. Meherally, Y. F. Liu, P. A. Chadwick, T. Rengifo, G. L. Sica, J. L. Port, P. C. Lee, et al. CXCL12 and CXCR4 in adenocarcinoma of the lung: association with metastasis and survival. J. Thorac. Cardiovasc. Surg., March 1, 2009; 137(3): 615 - 621. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Pencreach, E. Guerin, C. Nicolet, I. Lelong-Rebel, A.-C. Voegeli, P. Oudet, A. K. Larsen, M.-P. Gaub, and D. Guenot Marked Activity of Irinotecan and Rapamycin Combination toward Colon Cancer Cells In vivo and In vitro Is Mediated through Cooperative Modulation of the Mammalian Target of Rapamycin/Hypoxia-Inducible Factor-1{alpha} Axis Clin. Cancer Res., February 15, 2009; 15(4): 1297 - 1307. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Zhao, J. Najbauer, E. Garcia, M. Z. Metz, M. Gutova, C. A. Glackin, S. U. Kim, and K. S. Aboody Neural Stem Cell Tropism to Glioma: Critical Role of Tumor Hypoxia Mol. Cancer Res., December 1, 2008; 6(12): 1819 - 1829. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. L. Wagner, T.-A. Moo, N. Arora, Y.-F. Liu, R. Zarnegar, T. Scognamiglio, and T. J. Fahey III The Chemokine Receptors CXCR4 and CCR7 are Associated with Tumor Size and Pathologic Indicators of Tumor Aggressiveness in Papillary Thyroid Carcinoma Ann. Surg. Oncol., October 1, 2008; 15(10): 2833 - 2841. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Marlow, P. Strickland, J. S. Lee, X. Wu, M. PeBenito, M. Binnewies, E. K. Le, A. Moran, H. Macias, R. D. Cardiff, et al. SLITs Suppress Tumor Growth In vivo by Silencing Sdf1/Cxcr4 within Breast Epithelium Cancer Res., October 1, 2008; 68(19): 7819 - 7827. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Zagzag, M. Esencay, O. Mendez, H. Yee, I. Smirnova, Y. Huang, L. Chiriboga, E. Lukyanov, M. Liu, and E. W. Newcomb Hypoxia- and Vascular Endothelial Growth Factor-Induced Stromal Cell-Derived Factor-1{alpha}/CXCR4 Expression in Glioblastomas: One Plausible Explanation of Scherer's Structures Am. J. Pathol., August 1, 2008; 173(2): 545 - 560. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Li, L. Zhang, X. Zhang, Q. Yan, Y. A. Minamishima, A. F. Olumi, M. Mao, S. Bartz, and W. G. Kaelin Jr. Hypoxia-Inducible Factor Linked to Differential Kidney Cancer Risk Seen with Type 2A and Type 2B VHL Mutations Mol. Cell. Biol., August 1, 2007; 27(15): 5381 - 5392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. P. Rastaldi Necrotizing crescentic glomerulonephritis--a conditional knockout model discloses new therapeutic challenges Nephrol. Dial. Transplant., April 1, 2007; 22(4): 1020 - 1022. [Full Text] [PDF]
|
|
|
|
 |
|
 I. Kryczek, S. Wei, E. Keller, R. Liu, and W. Zou Stroma-derived factor (SDF-1/CXCL12) and human tumor pathogenesis Am J Physiol Cell Physiol, March 1, 2007; 292(3): C987 - C995. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Maroni, P. Bendinelli, E. Matteucci, and M. A. Desiderio HGF induces CXCR4 and CXCL12-mediated tumor invasion through Ets1 and NF-{kappa}B Carcinogenesis, February 1, 2007; 28(2): 267 - 279. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. G. Kaelin Jr. The von Hippel-Lindau Tumor Suppressor Protein and Clear Cell Renal Carcinoma Clin. Cancer Res., January 15, 2007; 13(2): 680s - 684s. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. C. Spoo, M. Lubbert, W. G. Wierda, and J. A. Burger CXCR4 is a prognostic marker in acute myelogenous leukemia Blood, January 15, 2007; 109(2): 786 - 791. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Evans, R. C. Russell, O. Roche, T. N. Burry, J. E. Fish, V. W. K. Chow, W. Y. Kim, A. Saravanan, M. A. Maynard, M. L. Gervais, et al. VHL Promotes E2 Box-Dependent E-Cadherin Transcription by HIF-Mediated Regulation of SIP1 and Snail Mol. Cell. Biol., January 1, 2007; 27(1): 157 - 169. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K S Kimbro and J W Simons Hypoxia-inducible factor-1 in human breast and prostate cancer. Endocr. Relat. Cancer, September 1, 2006; 13(3): 739 - 749. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Melillo and G. L. Semenza Meeting Report: Exploiting the Tumor Microenvironment for Therapeutics. Cancer Res., May 1, 2006; 66(9): 4558 - 4560. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Glasker, J. Li, J. B. Xia, H. Okamoto, W. Zeng, R. R. Lonser, Z. Zhuang, E. H. Oldfield, and A. O. Vortmeyer Hemangioblastomas share protein expression with embryonal hemangioblast progenitor cell. Cancer Res., April 15, 2006; 66(8): 4167 - 4172. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Burger and T. J. Kipps CXCR4: a key receptor in the crosstalk between tumor cells and their microenvironment Blood, March 1, 2006; 107(5): 1761 - 1767. [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 |
