Skip Navigation


Carcinogenesis Advance Access originally published online on January 22, 2008
Carcinogenesis 2008 29(4):696-703; doi:10.1093/carcin/bgn019
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
29/4/696    most recent
bgn019v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Chintalapudi, M. R.
Right arrow Articles by Hsu, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chintalapudi, M. R.
Right arrow Articles by Hsu, T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2008. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oxfordjournals.org

Cyr61/CCN1 and CTGF/CCN2 mediate the proangiogenic activity of VHL-mutant renal carcinoma cells

Mastan R. Chintalapudi1, Margaret Markiewicz2, Nurgun Kose1, Vincent Dammai1,3, Kristen J. Champion1, Rana S. Hoda1,4, Maria Trojanowska2,3 and Tien Hsu1,3,*

1 Department of Pathology and Laboratory Medicine
2 Department of Medicine, Division of Rheumatology
3 Hollings Cancer Center, Medical University of South Carolina, Charleston, SC 29425, USA
4 Present address: Cytopathology Unit, University of Rochester Medical Center, 601 Elmwood Avenue, Box 626, Rochester, NY 14642, USA

* To whom correspondence should be addressed. Tel: +1 843 792 0638; Fax: +1 843 792 5002; Email: hsut{at}musc.edu


   Abstract
Top
 Abstract
Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
 Funding
 References
 
The von Hippel–Lindau (VHL) protein serves as a negative regulator of hypoxia-inducible factor (HIF)-{alpha} subunits. Since HIF regulates critical angiogenic factors such as vascular endothelial growth factor (VEGF) and lesions in VHL gene are present in a majority of the highly vascularized renal cell carcinoma (RCC), it is believed that deregulation of the VHL–HIF pathway is crucial for the proangiogenic activity of RCC. Although VEGF has been confirmed as a critical angiogenic factor upregulated in VHL-mutant cells, the efficacy of antiangiogenic therapy specifically targeting VEGF signaling remains modest. In this study, we developed a three-dimensional in vitro assay to evaluate the ability of RCC cells to promote cord formation by the primary human dermal microvascular endothelial cells (HDMECs). Compared with VHL wild-type cells, VHL-mutant RCC cells demonstrated a significantly increased proangiogenic activity, which correlated with increased secretion of cysteine-rich 61 (Cyr61)/cysteine-rich 61-connective tissue growth factor-nephroblastoma overexpressed (CCN) 1, connective tissue growth factor (CTGF)/CCN2 and VEGF in conditioned culture medium. Both CCN proteins are required for HDMEC cord formation as shown by RNA interference knockdown experiments. Importantly, the proangiogenic activities conferred by the CCN proteins and VEGF are additive, suggesting non-overlapping functions. Expression of the CCN proteins is at least partly dependent on the HIF-2{alpha} function, the dominant HIF-{alpha} isoform expressed in RCC. Finally, immunohistochemical staining of Cyr61/CCN1 and CTGF/CCN2 in RCC tissue samples showed that increased expression of these proteins correlates with the loss of VHL protein expression. These findings strengthened the notion that the hypervascularized phenotype of RCC is afforded by multiple proangiogenic factors that function in parallel pathways.

Abbreviations: CCN, cysteine-rich 61-connective tissue growth factor-nephroblastoma overexpressed; cRCC, clear cell renal cell carcinoma; CTGF, connective tissue growth factor; Cyr61, cysteine-rich 61; EGFP, enhanced green fluorescence protein; HDMEC, human dermal microvascular endothelial cell; HEK, human embryonic kidney; HIF, hypoxia-inducible factor; IHC, immunohistochemistry; PBS, phosphate buffered saline; RCC, renal cell carcinoma; shRNA, short-hairpin RNA; siRNA, short interfering RNA; TGF-β, transforming growth factor-β; VEGF, vascular endothelial growth factor; VHL, von Hippel–Lindau; WISP, Wnt-induced secreted protein; 3-D, three-dimensional


   Introduction
Top
 Abstract
 Introduction
Materials and methods
 Results
 Discussion
 Supplementary material
 Funding
 References
 
Renal cell carcinoma (RCC) is a highly vascularized cancer of the kidney. This phenotype can be partly linked to the loss of von Hippel–Lindau (VHL) tumor suppressor gene, since VHL loss of heterozygosity and/or epigenetic inactivation have been found in 70–90% of the RCC of the clear cell type. VHL protein is shown to possess E3 ubiquitin ligase activity that recognizes prolyl-hydroxylated hypoxia-inducible factor (HIF)-{alpha} subunits, leading to their ubiquitination and degradation (1). Since HIF factors are transcription activators of several genes encoding critical angiogenic factors such as vascular endothelial growth factor (VEGF), VHL mutations can result in constitutive stabilization of HIF-{alpha} subunits and lead to angiogenic induction by VHL tumors. These findings prompted the high expectation of anti-RCC therapies based on antagonists of VEGF-signaling pathway. However, in a number of clinical trials, such therapeutics showed only modest improvement on survival time (2–4). Inhibitors against VEGF and another HIF target platelet-derived growth factor increased the response rate, measured by delay of tumor progression, but still could not achieve remission (5). These promising but modest outcomes are not simply the result of suboptimal treatment design since other inhibitors against a much broader spectrum of signaling pathways could achieve much better efficacy (6). However, these multitargeted inhibitors, such as sorafenib and sunitinib, pose an inherently higher risk of side effects. As such, the possibility of additional contributing angiogenic factors secreted by the cancer cells needs to be examined.

Recently, HIF-1 has been shown to upregulate connective tissue growth factor (CTGF) in kidney cells (7). This is interesting since CTGF has been suggested to be a potent angiogenic factor (8–10). CTGF belongs to a family of proteins consisting of cysteine-rich 61 [Cyr61/cysteine-rich 61-connective tissue growth factor-nephroblastoma overexpressed (CCN) 1], (CTGF/CCN2), nephroblastoma overexpressed (NOV/CCN3) and Wnt-induced secreted proteins-1 (WISP-1/CCN4), -2 (WISP-2/CCN5) and -3 (WISP-3/CCN6) (11,12). These proteins are secreted and are associated with extracellular matrix and cell membrane. The archetypical CCN proteins contain four distinct modules: insulin-like growth factor-binding protein-like, von Willebrand factor-like, thrombospondin-like and a C-terminal cysteine knot that has been implicated in protein–protein interaction. The functions of CCN proteins are pleiotropic, but their mode of action is likely to induce cellular responses by promoting cell adhesion to the matrix and facilitating interaction between integrins and growth factor receptors (13,14). Several more recent reports have shown that both Cyr61/CCN1 and CTGF/CCN2 can promote angiogenesis in vitro and in tumor angiogenesis model (14–17). In mouse models, ~30% of Ccn1–/– embryos die by E9.5 due to failure in chorioallantoic fusion, and ~70% die at mid-gestation with placental vascular defects, hemorrhage of embryonic vessels and defects in heart partitions (18,19). In contrast, Ccn2-null mice are perinatal lethal due to skeletal malformation, resulting from defects in chondrogenesis and angiogenesis (20). We reason that if HIF can upregulate CCN genes, this new class of angiogenic factors may complement the action of VEGF in VHL-mutant RCC. In this report, we examine the role of Cyr61/CCN1 and CTGF/CCN2 in the proangiogenic activity of RCC cells.


   Materials and methods
Top
 Abstract
 Introduction
 Materials and methods
Results
 Discussion
 Supplementary material
 Funding
 References
 
Cell lines and transformants
VHL-null RCC lines 786-0 and A498 and VHL+ human embryonic kidney cells HEK293 were from American Type Culture Collection (Manassas, Virginia, USA). 786-vec and 786-VHL and A498-vect and A498-VHL cells were generated by stable transfection of the parental cells with pCMV-EGFP and pCMV-VHL (see below), respectively, and polyclonal selection by G418 (Invitrogen, Carlsbad, California, USA). Cells were maintained in Dulbecco’s modified Eagle’s medium (high glucose) supplemented with 10% dialyzed fetal bovine serum (Invitrogen, Carlsbad, California, USA) and used within eight passages. G418 was excluded in all assay conditions. Primary human dermal microvascular endothelial cells (HDMECs) (Cambrex, East Rutherford, New Jersey, USA) were cultured on bovine collagen I (from Inamed, Fremont, California, USA)-coated tissue culture plates in Endothelial Cell Medium-2 (EBM2) medium (Cambrex). The plates were coated with 50 µg/ml collagen I in EBM2 media containing 0.01 M HCl. HDMECs were maintained in culture for no more than eight passages. pCMV-EGFP and pCMV-VHL were constructed by polymerase chain reaction cloning enhanced green fluorescence protein (EGFP) and human VHL into EcoRV site of pCDNA3.1 (Invitrogen) as described previously (21). The constitutively active version of HIF-2{alpha} [HIF-2{alpha} (P-A)] is an amino acid substitution of the proline residue at position 531 that is the target of hydroxylation in normoxia (22). Both wild-type and the constitutive HIF-2{alpha}-coding sequences are cloned in the pCDNA3.0 vector and are gifts from W.Kaelin of the Harvard Medical School.

Antibodies and reagents
Antibodies used in immunostaining and western blotting are as follows: rabbit anti-CCN1 (H-78, Santa Cruz Biotechnology, Santa Cruz, California, USA), goat anti-CCN2 (L-20, Santa Cruz), mouse anti-VEGF (C-1, Santa Cruz), mouse anti-HIF-2{alpha} (Novus Biologicals, Littleton, Colorado, USA), rabbit anti-VHL (Cell Signaling Technology, Danvers, Massachusetts, USA) and mouse anti-β-actin (Sigma-Aldrich, St. Louis, Missouri, USA). Secondary antibodies used in western blotting are peroxidase-conjugated donkey anti-goat IgG (Santa Cruz), peroxidase-conjugated sheep anti-mouse IgG (Amersham Biosciences, Pittsburgh, Pennsylvania, USA) and peroxidase-conjugated sheep anti-rabbit IgG (Amersham Biosciences). Secondary antibodies used in immunohistochemistry (IHC) are peroxidase-conjugated goat anti-rabbit IgG (DAKO) and peroxidase-conjugated goat anti-mouse IgG (DAKO, Glostrup, Denmark). Colorimetric reagents for IHC are from Vector Laboratories (Burlingame, California, USA). Other reagents include growth factor-reduced Matrigel (BD Biosciences, San Jose, California, USA), collagen I (Inamed), Lipofectamine 2000 (Invitrogen), CM-Dil (Molecular Probes) and protease inhibitors mixture (Roche, Basel, Switzerland).

RNA interference
The following short interfering RNA (siRNA) oligomers are from Santa Cruz: control siRNA (sc-37007), Cyr61 siRNA (sc-39331), CTGF siRNA (sc-35124) and VEGF siRNA (sc-29520). Transfection follows the instructions from the supplier. In addition, the following Cyr61/CCN1 and CTGF/CCN2-specific oligo duplexes were used for confirmation: Cyr61, CGAGGUGGAGUUGACGAGAAA, corresponding to 772–792 bp of the Cyr61 cDNA sequence (accession number NM_001554); CTGF, GCCCAGACCCAACUAUGAUUA, corresponding to 711–731 bp of the CTGF cDNA sequence (accession number NM_001901). Plasmid vector expressing VHL-specific short-hairpin RNA (shRNA) has been used successfully and has been described (21). Plasmid vector expressing HIF-2{alpha}-specific shRNA has been used successfully in 786-0 cells (21,22) and is a gift from W.Kaelin of the Harvard Medical School.

Three-dimensional cord formation assay
Three-dimensional (3-D) cord formation assays were performed according to Velazquez et al. (23) with modifications. HDMECs were cultured as monolayer on collagen-coated 48-well plates at 1 x 105 cells per well for 24 h. The endothelial cells were labeled with CM-Dil (1 µg/ml) for 5 min in phosphate buffered saline (PBS) and washed twice with medium. Gel overlay was prepared as follows: 1 ml chilled bovine collagen I solution at 2.9 mg/ml (PureCol from Inamed) was mixed with 125 µl 10x PBS and 125 µl 0.1 N NaOH before 25 µl of 0.1 M HCl was added. The neutralized collagen I solution was then mixed with growth-factor-reduced Matrigel (BD Biosciences) on ice with a ratio of 3:1. The medium was removed from the labeled HDMECs and proper amount of gel mixture was added to the cell monolayer to achieve an average thickness of 1 mm. The gel mixture was allowed to congeal at 37°C for 30 min. RCC cells (1 x 105) were then mixed with the neutralized collagen I gel solution on ice (enough volume for a 3 mm layer) and laid over the acellular 1 mm layer, allowed to congeal at 37°C for 30 min, before EBM2 containing 1% serum (or appropriate conditioned medium) was added. After 24 h, the wells were examined directly on an inverted scope (Olympus IX70) equipped with Fluoview 300 confocal capability. To quantify the cord network, three different confocal photomicrographs (x10 objective) focused at ~100 µm from the bottom and near the center per well were taken, and the areas of network coverage within the field of view, excluding isolated cells, were marked and measured by the ImageJ software. The extent of cord coverage was expressed as the percent total area of view occupied by the cord structures.

Tumor-endothelial cell coculture transwell assays
Transwell plates (12 well), with cell culture inserts of 12 mm in diameter and 12.0 µm pore size, were used (Corning, Lowell, Massachusetts, USA). Tumor cells were seeded on the bottom wells (100 000 cells per well) of the transwell apparatus and allowed to grow for 24 h before the complete medium was replaced with minimal EBM2 medium containing 1% serum and the incubation continued for 24 h. The 50 000 HDMECs were resuspended in EBM2 containing 1% serum, plated on the collagen I-coated top well (cell culture inserts), grown for 24 h, before the upper chamber was inserted into the bottom well containing the tumor cells with conditioned medium. Twenty-four hours later, the migrated cells attached to the bottom surface of the membrane were stained with crystal violet for 30 min at room temperature and counted in 10 fields of view using bright field microscopy.

Conditioned media collection and heparin pull down
RCC cells were grown to 80% confluency in Dulbecco’s modified Eagle’s medium containing 10% serum before the medium was replaced with EBM2 medium containing 1% serum. The EBM2 medium was conditioned for 48 h and then centrifuged at 10 000g for 10 min. The supernatant was used as conditioned medium. For heparin pull down, 500 µl of heparin–agarose beads (at 750–1000 µg/ml; Sigma) were added to 10 ml of conditioned medium and incubated at 4°C for 2 h. The heparin beads were collected and washed with PBS thrice and the bound proteins were released by 2 ml of PBS containing 2 M NaCl. The eluted fraction was centrifuged at 10 000g for 10 min, dialyzed against minimal EBM2 medium and normalized to the original volume (10 ml) for 3-D cord formation assay or boiled in 2x sample buffer for sodium dodecyl sulfate–polyacrylamide gel electrophoresis and western blotting.

Tissue samples and immunohistochemical staining
Formalin-fixed and paraffin-embedded 4 µm normal kidney and RCC tissue sections were obtained from Medical University of South Carolina Tissue Bank. Patient identities were code protected and all protocols followed the guidelines of and were approved by the MUSC Institutional Review Board. Kidney tissue microarray was obtained from Cybrdi (catalog # CC07-01), which consists of 3 normal kidneys and 32 clear cell renal cell carcinoma (cRCC) samples. The tissue sections were deparaffinized by xylene, rehydrated in graded alcohols and subsequently submitted to microwave epitope retrieval (2 x 5 min at medium cycle of 600 W output in 300 ml of 0.01 M sodium citrate buffer, pH 6.0). The tissue sections were then incubated with specific antibodies against VHL, CCN1, CCN2 or VEGF proteins at 1:100 dilutions for 30 min at room temperature. The samples were washed and incubated with corresponding peroxidase- or fluorescence-conjugated secondary antibodies for 30 min at room temperature. Specificity of the antibody staining was ensured by the following negative controls: (i) using secondary antibodies alone; (ii) in place of polyclonal rabbit anti-Cyr61, rabbit anti-VHL and goat anti-CTGF antibodies, the samples were incubated with 1% of the sera from the animals of the antibody origins (rabbit or goat; from Santa Cruz), followed by standard protocol with secondary antibodies incubation and signal detection; (iii) in place of monoclonal mouse anti-VEGF antibody, equivalent mouse IgG (isotype 2a) was used in place of the primary antibody, followed by standard protocol with secondary antibodies incubation and signal detection and (iv) where available, the commercial primary antibodies, in this case anti-Cyr61, were preabsorbed with peptide antigens the respective antibodies were made against (from Santa Cruz). For tissue sections, signals were visualized with stable diaminobenzidene substrate (Vector Laboratories) and the samples were counterstained with hematoxylin. For the tissue array, Alexaphor488-conjgated secondary antibodies (Invitrogen) were used and the samples were counterstained with TO-PRO (Invitrogen). For quantification, normal tubule or cRCC cells within fields of x20 images were counted. Scoring used the following criteria: –, staining in <1% of cells; +, staining in 1–20% of cells; ++, staining in 20–50% of cells and +++, staining in >50% of cells.

Statistical analysis
Values are expressed as mean and standard deviations. Comparisons of numerical data were made by Student’s t-test using GraphPad InStat statistics software to determine statistical significance. Distributions of categorical variables, such as those obtained from IHC of kidney tissues, were analyzed using chi-square. P values ≤0.05 were considered statistically significant.


   Results
Top
 Abstract
 Introduction
 Materials and methods
 Results
Discussion
 Supplementary material
 Funding
 References
 
VHL-mutant RCC cells exhibit proangiogenic characteristics in vitro
The human 786-0 cell line isolated from RCC has been widely used as the model VHL-null mutant (for example, 21,24–27) and is in the NCI-60 collection as a representative cRCC line. This parental line was transfected with either EGFP (786-vect) or human wild-type full-length VHL sequence (786-VHL). Pools of stable transformants were collected and maintained. We have shown previously that 786-vect showed no difference in cell motility and proliferation rate as compared with the parental 786-0 (21). To avoid accumulation of potential secondary mutations in the stable transformants, the 786-VHL pool was maintained for no more than eight passages. We adopted a three-dimentional (3-D) in vitro angiogenesis assay developed previously (23). Primary HDMECs are grown on collagen I-coated well to ~80% confluency. A 1 mm layer of gel matrix composed of 25% Matrigel and 75% collagen I was placed over the HDMEC monolayer. When incubated in low-serum (1%) media, the majority of the endothelial cells remained at the bottom. In the coculture scheme mimicking the physiological endothelium–stromal interactions, a second layer of gel matrix (3 mm thick) containing supporting cells such as fibroblasts was placed over the acellular gel layer. In such conditions, the HDMECs were induced to migrate into the acellular gel and form cord-like network within 24 h (Figure 1A). Thus, this system is useful for quantifying the activity of individual ‘stromal’ (i.e. relative to the endothelium) components without complicating effects of multiple cell types encountered in in vivo environment. In our modified design, the HDMECs are marked with lipophilic live-stain CM-Dil, so that cord formation can be observed without fixing and fluorescence staining, avoiding the risk of disturbing the 3-D architecture. Since the doubling time of HDMECs in monolayer under complete medium (10% serum) is ~48 h (M.R.C. and T.H., unpublished data), proliferation is not a factor in our 3-D cord formation assay. As shown in Figure 1B, when no supporting cells were embedded in the top gel layer, only few scattered HDMECs were detected in the 1 mm gel without cord formation [Control (gel)]. The majority of the cells remained at the bottom in a monolayer [Control (bottom)]. T cells appear viable based on the lipophilic live-stain used for visualization. They also exhibit a spread-out morphology indicative of non-invading cells (inset). Note that in subsequent experiments, cells at the bottom were routinely examined (although not shown) to ensure that a lack of cord formation was not due to cell death.


Figure 1
View larger version (52K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. VHL-mutant cells induce cord formation by primary microvascular endothelial cells. (A) Schematic representation of the 3-D coculture system. Cord formation by the HDMECs is supported by stromal cells such as fibroblasts, and can be observed by confocal microscopy usually at a depth of ~100 µm (indicated as ‘cord-forming region’) from the bottom of the well. An example of fibroblast (FB)-induced cord formation is shown in the lower panel. (B) 3-D coculture assay with indicated cRCC variants (786-0 or A498) embedded in the top gel layer. vect, parental cells transfected with empty vector; VHL, parental cells transfected with vector expressing VHL. In control experiment (no supporting cells), few HDMECs migrated into the gel (up to 100 µm) while the rest remained as a monolayer at the bottom. These cells show spread-out morphology (inset). In contrast, parental and vector-transfected cells induce cord formation at a depth up to 100 µm. VHL-transfected cells showed only limited cord-inducing activity. (C) Quantification of cord formation observed in (B). The values are the percentile of total area of view occupied by cord structures. Isolated cells were not included. ***P ≤ 0.0001; **P ≤ 0.001; *P ≤ 0.01. Bars are 200 µm.

 
When 786-0 or 786-0 transfected with empty vector (786-vect) were included in the top gel layer, organized cord structures comparable with those induced by fibroblasts were observed at a depth of ~100 µm from the bottom. This also confirmed that stable transformant selection per se did not alter the proangiogenic property of the parental cells. In contrast, reexpression of VHL in 786-0 cells significantly reduced the proangiogenic activity of the cancer cells. The in vitro proangiogenic activity can be quantified by the percent of total area in the field of view occupied by the cord structures (Figure 1C). To further verify the role of VHL protein in regulating proangiogenic activity, a different VHL-null NCI-60 cRCC line A498 was examined. As shown in Figure 1C, reexpression of VHL in A498 also greatly reduced the cancer cells’ capability to induce cord formation by the endothelial cells.

Diffusible factors mediate the proangiogenic activity of RCC cells
Although the RCC cells and the HDMECs are separated by at least 1 mm of matrix in the 3-D assay, it is formally possible that the cancer cells can exert their influence on the behavior of HDMECs by physically contracting the matrix. To demonstrate that the proangiogenic activity of 786-0 cells is mediated by secreted factors, conditioned media were collected and added to the 3-D culture without the supporting cells. As shown in Figure 2A, 786-0-conditioned media could support cord formation whereas 786-VHL-conditioned media could not. Furthermore, heparin-bound fraction of the 786-0-conditioned media exhibited proangiogenic activity when added to minimal media in the 3-D culture (Figure 2A). In contrast, 786-0-conditioned media devoid of heparin-bound fraction lost the proangiogenic activity (Figure 2A). The presence of diffusible angiogenic factor is further demonstrated using a modified transwell assay (Figure 2B). In this assay, 786-0 and the variants were plated in the bottom well and the HDMECs were plated in the insert on the collagen-coated porous membrane. There was no possibility of direct contact of the two cell types or physical modification of the matrix by the cancer cells in this system. Thus, reexpression of VHL inhibited the chemotactic motility of the HDMECs toward the cancer cells. Taken together, we conclude that VHL-mutant RCC cells secrete diffusible proangiogenic factors that act over a long distance.


Figure 2
View larger version (50K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Secreted factors mediate proangiogenic activity of RCC cells. (A) 3-D cord formation assays using conditioned media (CM) collected from 786-0 or 786-VHL cultures (upper panels). 786-VHL-conditioned media lost the proangiogenic activity exhibited by 786-0-conditioned media. The 786-0-conditioned media were depleted for heparin-binding protein fraction [conditioned media no heparin-bound fraction (HBF)] and the heparin-bound fraction collected from 786-0 conditioned media was added to minimal EBM2 media (MM + heparin-bound fraction). These two reconstituted media were used in cord formation assay. Minimal media + heparin-bound fraction could support cord formation while conditioned media no heparin-bound fraction could not. Bars are 200 µm. (B) Schematic representation of the transwell coculture migration assay (left panel). The migration efficiency induced by 786-vect and 786-VHL is expressed as the percentile of the total number of HDMECs migrated in the presence of 786-0. ***P ≤ 0.0001. (C) Conditioned media collected from cultures of indicated cells were incubated with heparin-agarose beads and the heparin-bound fractions were western blotted with the indicated antibodies on the left. The apparent molecular weights of the detected proteins are indicated on the right. For loading control, the cells from which conditioned media were collected were lysed and probed for actin. (D) 786-0 cells were transfected with pCDNA3.0 expressing a shRNA of random sequence (shControl) or a HIF-2{alpha}-specific duplex (shHIF2{alpha}, as used successfully in refs 21,22). The conditioned media were analyzed for the heparin-bound proteins as in (C) and the cell layer was lysed. Each fraction was western blotted for the proteins indicated on the right. Cyr61/CCN1, CTGF/CCN2 and VEGF levels are reduced in HIF-2{alpha} knockdown cells.

 
Cyr612/CCN1 and CTGF/CCN2 are overexpressed in VHL-mutant cells
One obvious candidate of the secreted angiogenic factor is VEGF since it is a well-known target of HIF. Surprisingly, when neutralizing anti-VEGF antibody was added to the 786-0-conditioned media, at a concentration in excess of the established effective dosage in a similar assay (23), only moderate inhibition was observed (from ~40% coverage of cord structure to ~30% coverage or a 25% inhibition; supplementary Figure S1 is available at Carcinogenesis Online) as compared with the inhibition by reexpression of VHL (from ~40% coverage of cord structures to ~10% coverage or a 75% inhibition; Figure 1C). Even VEGF knockdown in RCC cells resulted in only ~40% inhibition (see below). We therefore suspected the involvement of other angiogenic factors in conferring hypervascular activity of the VHL-mutant cells.

Recently, the angiogenic factor CTGF/CCN2 has been identified as a HIF target (7). We therefore examined whether CTGF/CCN2 and its close relative Cyr61/CCN1, also an angiogenic factor, are overexpressed in VHL-mutant cells-conditioned media. Since both Cyr61/CCN1 and CTGF/CCN2 are known heparin-binding factors (28,29), heparin pull down was used to enrich the growth factors. VEGF, also a heparin-binding growth factor, was included as a control. Western blot of the heparin-binding fraction detected the mature form of the three factors and showed a significant increase of these factors in VHL mutants, in both 786-0 and A498 variants (Figure 2C). When HIF-2{alpha} was knocked down in 786-0 cells [HIF-2 is the dominant HIF factor in RCC cells such as 786-0 (30)], Cyr61/CCN1 and CTGF/CCN2 expression was significantly reduced (Figure 2D), consistent with the notion that HIF factor contributes to the overexpression of CCN factors in RCC cells.

CCN factors contribute to the proangiogenic activity of RCC cells
To examine whether CCN factors contribute to the proangiogenic activity of 786-0 cells, the expression of these factors were knocked down using specific siRNA (Figure 3A). Media conditioned by these knockdown cells were collected and used in the 3-D culture (Figure 3B and C). Individual knockdown of CCN1, CCN2 and VEGF resulted in 40–50% reduction of cord formation as compared with the control siRNA. The CCN knockdown phenotypes were confirmed by using a different set of siRNA sequences (supplementary Figure S2 is available at Carcinogenesis Online). Interestingly, double knockdown of each of the combinations of two factors resulted in further reduction (65–70%) of cord formation (Figure 3C). That is, the actions of these factors are at least partly additive, indicating non-overlapping functions. The additive functions of these factors are also demonstrated in the transwell assay (Figure 3D).


Figure 3
View larger version (59K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. CCN proteins and VEGF show non-overlapping proangiogenic activities. (A) Western blots showing specific knockdown by siRNA. Conditioned media from 786-0 cells knocked down for Cyr61/CCN1, CTGF/CCN2 and VEGF were collected and the heparin-binding fractions were isolated. The fractions and the corresponding cell lysates were western blotted using the antibodies indicated on the left. (B) 3-D cord formation assay using conditioned media collected from 786-0 cells knocked down for the indicated factors. (C) Quantification of cord formation from (B) and indicated double knockdowns. (D) Transwell coculture migration assays with 786-0 cells knocked down or double knocked down for the indicated factors. ***P ≤ 0.0001; **P ≤ 0.001. Bars are 200 µm.

 
To further confirm that VHL loss of function and HIF activity is responsible for CCN upregulation and elevated proangiogenic activity, we sought verification in VHL wild-type human embryonic kidney cell line HEK293. Wild-type HIF-2{alpha} or a mutant form of HIF-2{alpha} that is insensitive to VHL-mediated degradation was transfected into the VHL+ embryonic kidney cells HEK293. Overexpression of HIF-2{alpha}, especially the constitutively stable mutant, increased the levels of CCN1, CCN2 and VEGF (Figure 4A). In a reciprocal experiment, HEK293 cells were transfected with plasmids expressing control or VHL-specific shRNA. The VHL knockdown cells showed significant upregulation of CCN1 and CCN2 proteins and modest increase of VEGF (Figure 4A). The increased expression levels of these three factors correlate with the ability of the altered HEK293 VHL+ cells (overexpressing HIF-2{alpha} or shVHL) to support cord formation by HDMECs (Figure 4B). Furthermore, the acquired proangiogenic activity of HEK293 cells transfected with HIF-2{alpha} (P-A) could be significantly reduced by HIF-2{alpha}-specific shRNA expression (Figure 4C).


Figure 4
View larger version (57K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. VHL knockdown and HIF-2{alpha} expression in HEK293 induce CCN and VEGF expression and cord formation by HDMECs. (A) HEK293 cells were transfected with plasmids encoding green fluorescence protein (GFP), HIF-2{alpha} or HIF-2{alpha} (P-A) or with plasmids expressing control shRNA or shVHL, and the heparin-binding fractions from the conditioned media were analyzed for CCN and VEGF proteins. Actin from the corresponding cells lysates was used as loading controls. (B) 3-D cord formation assays with HEK293 variant cells described in (A). (C) HEK293 cells transfected with HIF-2{alpha} (P-A) were double transfected with plasmids expressing control or HIF-2{alpha}-specific shRNA and the corresponding media were used for cord formation assay. ***P ≤ 0.0001; **P ≤ 0.001. Bars are 200 µm.

 
Cyr61/CCN1 and CTGF/CCN2 are overexpressed in the RCC tissues
To establish whether CCN proteins are relevant in RCC samples, we first compared a battery of clinically resected RCC tissues with normal counterparts (Figure 5A). In normal kidney tissues, Cyr61/CCN1 and CTGF/CCN2 are expressed at low levels in the tubule epithelia as well as in the glomerulus. VEGF, on the other hand, is expressed strongly in the glomerula and not in the tubule cells. In contrast, all three are expressed strongly in the cancer foci, identifiable by the clear cell morphology of the cancer lesion (insets). Importantly, overexpression of these three proteins correlates inversely with the presence of VHL proteins. That is, normal kidney tubule cells express high levels of VHL whereas the cRCC samples express no or very low levels of VHL. The specificity of the antibodies was verified by using sera of the animals in which the primary antibodies were generated in place of the polyclonal anti-Cyr61 (rabbit), anti-CTGF (goat) and anti-VHL (rabbit) or by using isotype-specific mouse IgG2a in place of mouse monoclonal anti-VEGF (Figure 5A, lower panels). In addition, where available, the primary antibodies were preabsorbed with the antigenic peptides and the mixtures were used as primary antibodies. Preabsorbed anti-Cyr61 lost their ability to detect signals in the tissue sections. Standard negative controls such as secondary antibody alone also showed no staining (data not shown).


Figure 5
View larger version (94K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Expression of CCN proteins in cRCC tissues. (A) Paraffin sections from normal or cRCC samples were processed for immunostaining with indicated antibodies on the top. Hematoxylin and eosin (H&E)-stained samples are shown for visualizing the cellular morphology and for identifying tissue structures. In the normal tissue sample, glomerulus (G) is the large circular structure that is rich in blood vessels. The rest of the view shows cross section of collecting tubules with defined lumen. In the cRCC samples, the clear cell morphology is clearly seen (insets). Negative control stainings using respective sera or isotype-specific IgG in place of primary antibodies, or anti-CCN1 preabsorbed with the peptide antigen, are shown in middle panels. A total of two normal and five cRCC samples was examined. Bars are 100 µm. (B) IHC followed the protocol used in (A), except fluorescence-tagged secondary antibodies were used for detection. DNA dye TO-PRO was used for counterstain. Serial sections on separate slides were used for different antibodies. Representative stained samples are shown. White bars are 50 µm and red bars are 20 µm.

 
To further strengthen the notion that VHL-mutant cRCCs overexpress CCN proteins, we performed IHC on tissue microarrays containing 32 cRCC and 3 normal kidney tissues. Slides containing neighboring sections were incubated with antibodies against VEGF, Cyr61, CTGF or VHL. To ensure unbiased signal detection across multiple samples and different slides, we chose to use fluorescent dye so that no inconsistency of colorimetric development is involved. As shown in Figure 5B, normal kidney samples express high VHL and very low VEGF, Cyr61/CCN1 and CTGF/CCN2 as expected. The combined tissue staining results, based on the number of cells stained in fields of views (see Materials and Methods), are shown in Table I. We set the threshold of <20% stained cells as weak expression (– and + in Table I) and >20% (++ and +++) as strong expression. Using this criterion, chi-square analysis (Fisher’s exact test in this case) shows that there are no statistically significant differences between different grades of tumors. For example, the P value for CCN1 expression between grade 1 and grade 2 is >0.4. However, there are significant differences between normal (2 from MUSC Tissue Bank and 3 from tissue array) and tumors (5 from MUSC Tissue Bank and 32 from tissue array) for the overexpression of CCNs and VEGF (P = 0.007, 0.01 and 0.004 for CCN1, CCN2 and VEGF, respectively). It should be noted that since no VHL mutation analysis was available on the tumor samples shown, it is not known how many of the VHL protein-expressing tumors are in fact VHL mutants. However, if only VHL protein-null tumors (a total of 26 from 37 cRCC) are considered, there is a nearly perfect inverse correlation between loss of VHL and overexpression of CCNs. That is, 25 of 26 VHL-null tumors overexpress CCNs.


View this table:
[in this window]
[in a new window]

  		Table I. Correlation of CCN and VEGF upregulation in RCC

 

   Discussion
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
Supplementary material
 Funding
 References
 
The 3-D coculture system used in this report provides a reproducible and quantifiable in vitro assay for the angiogenic activity, in which cellular and molecular components can be tested in a defined matrix environment. This assay should be useful as a standardized method for defining proangiogenic activities of tumor cells. The use of HDMECs is also more physiologically relevant than the widely used human umbilical vein cells for assaying tumor-induced angiogenesis, since the latter are not part of the normal cancer stromal environment. While all in vitro analyses should ultimately be verified in vivo, the 3-D coculture system is a good approximation of angiogenesis, and in specific ways more advantageous, for the following reasons: (i) this is a defined assay system in which individual stromal cell types or angiogenic factors can be analyzed; (ii) the matrix into which the cords are formed consists mainly of collagen I, which is the in vivo interstitial matrix protein the endothelial cells encounter when new vessels break out from the basement membrane; (iii) in the 3-D culture, the HDMECs do not spontaneously mobilize until exogenous angiogenic factors are provided; therefore the angiogenic response can be measured accurately; (iv) some of the cords in the 3-D matrix in fact form lumen, thus mimicking true microvasculature (ref. 23 and data not shown) and (v) the in vitro assay also excludes the complicating factors such as vasculogenesis originating from circulating progenitor cells unavoidably encountered in vivo, so that angiogenic response from the endothelial cells is specifically analyzed.

In this report, we show that CCN proteins Cyr61 and CTGF are part of the proangiogenic program upregulated in the VHL-mutant RCC cells. The overexpression is at least in part dependent on the activity of HIF factor, either in RCC or in the transformed VHL+ human embryonic kidney cell line (HEK293). Importantly, double knockdown of combinations of two of the three factors examined showed additive effects. This indicates that CCN proteins and VEGF confer non-overlapping proangiogenic functions. In light of the modest effectiveness of specific anti-VEGF therapeutics in clinical trials against RCC, this finding may suggest a new therapeutic strategy. We should also emphasize that HIF may not be the only regulator of CCN proteins in RCC cells. CCN proteins, especially CTGF, are a prominent target of transforming growth factor-β (TGF-β) signaling, mediated by Jun N-terminal kinase or Extracellular signal-regulated kinase pathways (31). Specifically, CTGF has been shown to be induced by TGF-β in renal proximal tubule cells (32), although this signaling event is more directly relevant to inflammatory responses in normal kidneys. TGF-β has also been shown to induce Cyr61 expression in breast cancer cells (33). It has not yet been shown conclusively whether TGF-β signaling plays a role in RCC tumorigenesis. On the other hand, we have recently shown that in VHL-mutant RCC cells and in resected RCC tissue samples, fibroblast growth factor (FGF) receptor is overaccumulated on the cell surface, which confers oversensitized response to FGF stimulation with elevated levels of active ERK (21). In fibroblasts, CTGF is upregulated by ERK signaling via the activity of ETS1 transcription factor (34). It is therefore possible that the intrinsically elevated proangiogenic activities in RCC cells, mediated by increased level of HIF function, can be further preferentially induced by extrinsic growth factors. We have tested this possibility by blocking FGFR signaling in RCC cells and examined their proangiogenic activity under serum-induced condition. Fibriblast growth factor receptor (FGFR) knockdown cells indeed showed reduced capability to induce cord formation (data not shown). Such indirect signaling mechanism is necessarily complicated and the exact functional relationships between CCN proteins and multiple receptor kinases require further investigation.

The functions of CCN proteins are pleiotropic. They can promote proliferation or apoptosis depending on the cell types and assay conditions. In our assay systems, however, we did not observe effects of CCN knockdown on cell viability, either on the RCC cells or indirectly on the endothelial cells (Figure 1 and M.R.C. and T.H., unpublished data). The proangiogenic functions of Cyr61/CCN1 and CTGF/CCN2 have been studied extensively. It has been shown that CCN proteins bind to integrins and induce downstream signaling events or promote cell attachment (14,15,35). Engagement of integrin may also facilitate aggregation of receptor tyrosine kinases within the integrin-centered signaling complex (36–39). In addition, since CTGF can induce collagen production (11,12), it may alter the matrix environment indirectly. Although a function of matrix modification by the CCN proteins cannot be completely ruled out, our transwell migration data demonstrate that both Cyr61 and CTGF can function as chemotactic factors that promote the motility of HDMECs.

Our examination of CCN expression in RCC tissue samples suggests that Cyr61 and CTGF play a role in the progression of VHL-mutant tumors. Our in vitro data suggest that they are likely part of the proangiogenic program exhibited by the VHL-mutant RCC. However, a direct function in promoting the growth and spread of RCC is formally possible since CCN proteins have been implicated in the progression and metastasis of several cancers (40), including ovarian and breast. Future studies should determine whether Cyr61 and/or CTGF can play a similar role in RCC progression.


   Supplementary material
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
Funding
 References
 
Supplementary Figures S1 and S2 can be found at http://carcin.oxfordjournals.org/


   Funding
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
 Funding
References
 
National Cancer Institute (RO1CA109860 to T.H., PO1CA78582 to T.H. and M.E.T., and RO1CA128002 to V.D.); National Institutes of Health (R21AR050798 to M.E.T.); Scleroderma Foundation to M.M.; Abney Scholarship for Cancer Research to K.J.C.


   Acknowledgments
 
We thank Dr W.G.Kaelin for providing the plasmid clones of HIF-2{alpha} and HIF-2{alpha} (P-A) and HIF-2{alpha} siRNA.

Conflict of Interest Statement: None declared.


   References
Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Supplementary material
 Funding
 References
 

  1. Kaelin WG Jr. Molecular basis of the VHL hereditary cancer syndrome. Nat. Rev. Cancer (2002) 2:673–682.[CrossRef][Web of Science][Medline]
  2. Yang JC, et al. A randomized trial of bevacizumab, an anti-vascular endothelial growth factor antibody, for metastatic renal cancer. N. Engl. J. Med. (2003) 349:427–434.[Abstract/Free Full Text]
  3. Reuter CW, et al. Targeting vascular endothelial growth factor (VEGF)-receptor-signaling in renal cell carcinoma. World J. Urol. (2007) 25:59–72.[CrossRef][Web of Science][Medline]
  4. Srinivasan R, et al. Anti-angiogenic therapy in renal cell cancer. BJU Int. (2007) 99:1296–1300.[CrossRef][Web of Science][Medline]
  5. George DJ. Phase 2 studies of sunitinib and AG013736 in patients with cytokine-refractory renal cell carcinoma. Clin. Cancer Res. (2007) 13:753s–757s.[Abstract/Free Full Text]
  6. Speca J, et al. Improving outcomes with novel therapies for patients with newly diagnosed renal cell carcinoma. Clin. Genitourin. Cancer (2006) 5(suppl. 1):S24–S30.[CrossRef][Web of Science][Medline]
  7. Higgins DF, et al. Hypoxic induction of Ctgf is directly mediated by Hif-1. Am. J. Physiol. Renal Physiol. (2004) 287:F1223–F1232.[Abstract/Free Full Text]
  8. Shimo T, et al. Inhibition of endogenous expression of connective tissue growth factor by its antisense oligonucleotide and antisense RNA suppresses proliferation and migration of vascular endothelial cells. J. Biochem. (1998) 124:130–140.[Abstract/Free Full Text]
  9. Shimo T, et al. Connective tissue growth factor induces the proliferation, migration, and tube formation of vascular endothelial cells in vitro and angiogenesis in vivo. J. Biochem. (1999) 126:137–145.[Abstract/Free Full Text]
  10. Shimo T, et al. Connective tissue growth factor as a major angiogenic agent that is induced by hypoxia in a human breast cancer cell line. Cancer Lett. (2001) 174:57–64.[CrossRef][Web of Science][Medline]
  11. Perbal B. CCN proteins: multifunctional signalling regulators. Lancet (2004) 363:62–64.[CrossRef][Web of Science][Medline]
  12. Leask A. All in the CCN family: essential matricellular signaling modulators emerge from the bunker. J. Cell Sci. (2006) 119:4803–4810.[Abstract/Free Full Text]
  13. Lau LF, et al. The CCN family of angiogenic regulators: the integrin connection. Exp. Cell Res. (1999) 248:44–57.[CrossRef][Web of Science][Medline]
  14. Brigstock DR. Regulation of angiogenesis and endothelial cell function by connective tissue growth factor (CTGF) and cysteine-rich 61 (CYR61). Angiogenesis (2002) 3:153–165.
  15. Chen N, et al. Identification of a novel integrin alphavbeta3 binding site in CCN1 (CYR61) critical for pro-angiogenic activities in vascular endothelial cells. J. Biol. Chem. (2004) 279:44166–44176.[Abstract/Free Full Text]
  16. Kale S, et al. Microarray analysis of in vitro pericyte differentiation reveals an angiogenic program of gene expression. FASEB J. (2005) 19:270–271.[Abstract/Free Full Text]
  17. Yang F, et al. Stromal expression of connective tissue growth factor promotes angiogenesis and prostate cancer tumorigenesis. Cancer Res. (2005) 65:8887–8895.[Abstract/Free Full Text]
  18. Mo FE, et al. CCN1 (CCN1) is essential for placental development and vascular integrity. Mol. Cell. Biol. (2002) 22:8709–87020.[Abstract/Free Full Text]
  19. Mo FE, et al. The matricellular protein CCN1 is essential for cardiac development. Circ. Res. (2006) 99:961–969.[Abstract/Free Full Text]
  20. Ivkovic S, et al. Connective tissue growth factor coordinates chondrogenesis and angiogenesis during skeletal development. Development (2003) 130:2779–2791.[Abstract/Free Full Text]
  21. Hsu T, et al. Endocytic function of von Hippel-Lindau tumor suppressor protein regulates surface localization of fibroblast growth factor receptor 1 and cell motility. J. Biol. Chem. (2006) 281:12069–12080.[Abstract/Free Full Text]
  22. Kondo K, et al. Inhibition of HIF2alpha is sufficient to suppress pVHL-defective tumor growth. PLoS Biol. (2003) 1:E83.[Medline]
  23. Velazquez OC, et al. Fibroblast-dependent differentiation of human microvascular endothelial cells into capillary-like 3-dimensional networks. FASEB J. (2002) 16:1316–1318.[Abstract/Free Full Text]
  24. Ohh M, et al. The von Hippel-Lindau tumor suppressor protein is required for proper assembly of an extracellular fibronectin matrix. Mol. Cell (1998) 1:959–968.[CrossRef][Web of Science][Medline]
  25. Esteban-Barragan MA, et al. Role of the von Hippel-Lindau tumor suppressor gene in the formation of beta1-integrin fibrillar adhesions. Cancer Res. (2002) 62:2929–2936.[Abstract/Free Full Text]
  26. Jiang Y, et al. Gene expression profiling in a renal cell carcinoma cell line: dissecting VHL and hypoxia-dependent pathways. Mol. Cancer Res. (2003) 1:453–462.[Abstract/Free Full Text]
  27. Wykoff CC, et al. Gene array of VHL mutation and hypoxia shows novel hypoxia-induced genes and that cyclin D1 is a VHL target gene. Br. J. Cancer (2004) 90:1235–1243.[CrossRef][Web of Science][Medline]
  28. Brigstock DR, et al. Purification and characterization of novel heparin-binding growth factors in uterine secretory fluids. Identification as heparin-regulated Mr 10,000 forms of connective tissue growth factor. J. Biol. Chem. (1997) 272:20275–20282.[Abstract/Free Full Text]
  29. Chen CC, et al. The angiogenic factors Cyr61 and connective tissue growth factor induce adhesive signaling in primary human skin fibroblasts. J. Biol. Chem. (2001) 276:10443–10452.[Abstract/Free Full Text]
  30. Turner KJ, et al. Expression of hypoxia-inducible factors in human renal cancer: relationship to angiogenesis and to the von Hippel-Lindau gene mutation. Cancer Res. (2002) 62:2957–2961.[Abstract/Free Full Text]
  31. Leask A, et al. Connective tissue growth factor gene regulation. Requirements for its induction by transforming growth factor-beta 2 in fibroblasts. J. Biol. Chem. (2003) 278:13008–13015.[Abstract/Free Full Text]
  32. Phanish MK, et al. TGF-beta1-induced connective tissue growth factor (CCN2) expression in human renal proximal tubule epithelial cells requires Ras/MEK/ERK and Smad signalling. Nephron Exp. Nephrol. (2005) 100:e156–e165.[CrossRef][Medline]
  33. Bartholin L, et al. The human Cyr61 gene is a transcriptional target of transforming growth factor beta in cancer cells. Cancer Lett. (2007) 246:230–236.[CrossRef][Web of Science][Medline]
  34. Nakerakanti SS, et al. Fli1 and Ets1 have distinct roles in connective tissue growth factor/CCN2 gene regulation and induction of the profibrotic gene program. J. Biol. Chem. (2006) 281:25259–25269.[Abstract/Free Full Text]
  35. Leu SJ, et al. Pro-angiogenic activities of CYR61 (CCN1) mediated through integrins alphavbeta3 and alpha6beta1 in human umbilical vein endothelial cells. J. Biol. Chem. (2002) 277:46248–46255.[Abstract/Free Full Text]
  36. Clark EA, et al. Integrins and signal transduction pathways: the road taken. Science (1995) 268:233–239.[Abstract/Free Full Text]
  37. Friedlander M, et al. Definition of two angiogenic pathways by distinct {alpha}v integrins. Science (1995) 270:1500–1502.[Abstract/Free Full Text]
  38. Miyamoto S, et al. Integrins can collaborate with growth factors for phosphorylation of receptor tyrosine kinases and MAP kinase activation: roles of integrin aggregation and occupancy of receptors. J. Cell Biol. (1996) 135:1633–1642.[Abstract/Free Full Text]
  39. Giancotti FG, et al. Integrin signaling. Science (1999) 285:1028–1032.[Abstract/Free Full Text]
  40. Bleau AM, et al. CCN proteins and cancer: two to tango. Front. Biosci. (2005) 10:998–1009.[Web of Science][Medline]
Received June 29, 2007; revised December 28, 2007; accepted January 8, 2008.


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nephrol Dial TransplantHome page
S. Kroening, E. Neubauer, J. Wessel, M. Wiesener, and M. Goppelt-Struebe
Hypoxia interferes with connective tissue growth factor (CTGF) gene expression in human proximal tubular cell lines
Nephrol. Dial. Transplant., November 1, 2009; 24(11): 3319 - 3325.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
K. L. Bennewith, X. Huang, C. M. Ham, E. E. Graves, J. T. Erler, N. Kambham, J. Feazell, G. P. Yang, A. Koong, and A. J. Giaccia
The Role of Tumor Cell-Derived Connective Tissue Growth Factor (CTGF/CCN2) in Pancreatic Tumor Growth
Cancer Res., February 1, 2009; 69(3): 775 - 784.
[Abstract] [Full Text] [PDF]

Home page
 Mol Cancer ResHome page
J. Samarin, M. Rehm, B. Krueger, J. Waschke, and M. Goppelt-Struebe
Up-Regulation of Connective Tissue Growth Factor in Endothelial Cells by the Microtubule-Destabilizing Agent Combretastatin A-4
Mol. Cancer Res., February 1, 2009; 7(2): 180 - 188.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow Supplementary Data
Right arrow All Versions of this Article:
29/4/696    most recent
bgn019v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (5)
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Chintalapudi, M. R.
Right arrow Articles by Hsu, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chintalapudi, M. R.
Right arrow Articles by Hsu, T.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?