Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.

CAN-09-3722

Therapeutics, Targets, and Chemical Biology

Cancer
Research
Sunitinib Acts Primarily on Tumor Endothelium rather than
Tumor Cells to Inhibit the Growth of Renal Cell Carcinoma
Dan Huang1, Yan Ding1, Yan Li1, Wang-Mei Luo1, Zhong-Fa Zhang1, John Snider1,
Kristin VandenBeldt2, Chao-Nan Qian1,3, and Bin Tean Teh1

Abstract
Sunitinib is a broad-spectrum small-molecule inhibitor of receptor tyrosine kinases (RTK) that serves as the
present standard of care for first-line therapy of advanced clear cell renal cell carcinoma (ccRCC). A full under-
standing of the targets and mechanism of action of sunitinib in ccRCC treatment remains incomplete. In this
study, we evaluated several tumor cell and endothelial targets of sunitinib and investigated which RTK(s) may
specifically contribute to its therapeutic effects. Microarray expression profiling and Western blot analysis re-
vealed that among known sunitinib targets, only platelet-derived growth factor receptor-β and vascular endo-
thelial growth factor receptor-2 (VEGFR-2) were overexpressed in ccRCCs relative to normal tissues. Sunitinib
was unable to inhibit survival or proliferation of ccRCC cells at pharmacologically relevant concentrations
(∼0.1 μmol/L) that inhibit RTK targets. In contrast, sunitinib inhibited endothelial cell proliferation and motility
at the same concentrations by suppressing VEGFR-2 signaling. Moreover, whereas sunitinib inhibited the growth
of ccRCC xenograft tumors and decreased tumor microvessel density as soon as 12 hours after treatment,
sunitinib showed no significant effects on tumor cell proliferation or apoptosis up to 72 hours after treatment.
Our findings indicate that sunitinib inhibits ccRCC growth primarily through an antiangiogenic mechanism
and not through direct targeting of ccRCC tumor cells. Cancer Res; 70(3); 1053–62. ©2010 AACR.

Introduction perivascular cells (pericytes) and play critical roles in angio-

genesis. Sunitinib may also have direct antiproliferative and/
Sunitinib is a multitargeted inhibitor of receptor tyrosine or apoptotic effects on tumor cells, which express target ki-
kinases (RTK) and has shown clinical efficacy in the treat- nases. Sunitinib has been shown to directly inhibit the sur-
ment of human cancers. It is currently Food and Drug Admin- vival and proliferation of a variety of cancer cells, including
istration approved for the treatment of advanced clear cell small cell lung carcinoma, gastrointestinal stromal tumors,
renal cell carcinoma (ccRCC) and gastrointestinal stromal tu- acute myelogenous leukemia, and chronic myelogenous
mors. It is also under intense investigation as a treatment for monocytic leukemia, which harbored either activating muta-
several other cancers, including breast cancer, colorectal can- tions or translocations of the target RTKs or their ligands (1,
cer, and non–small cell lung cancer (1–4). Sunitinib has been 3, 4, 6–8). However, direct antitumor effects of sunitinib in
shown to inhibit the tyrosine kinase activity of vascular endo- the treatment of ccRCC have not been clearly established.
thelial growth factor receptors (VEGFR-1, VEGFR-2, and To date, no somatic mutations in target RTK genes have
VEGFR-3), platelet-derived growth factor receptors (PDGFR- been found in human ccRCC (9). Although one recent study
α and PDGFR-β), fms-related tyrosine kinase 3 (FLT3), stem reported apoptotic effects of sunitinib on ccRCC cells in vitro,
cell growth factor receptor KIT, and RET (5). the doses observed to inhibit cell survival were far higher
The ability of sunitinib to suppress tumor angiogenesis has than pharmacologically relevant doses (10). Thus, despite
been well established. Two of the major targets of sunitinib, the efficacy of sunitinib in the treatment of ccRCC, its targets
VEGFR and PDGFR, are expressed on endothelial cells and and complete mechanism of action against ccRCC remain
unclear. In particular, it is unclear whether sunitinib acts pri-
marily through an antiangiogenic mechanism or whether it
Authors' Affiliations: 1Laboratory of Cancer Genetics and 2Laboratory of
Analytical, Cellular, and Molecular Microscopy, Van Andel Research
may also act to directly target ccRCC tumor cells.
Institute, Grand Rapids, Michigan and 3The State Key Laboratory of Sunitinib is currently considered the standard of care for
Oncology in South China, Sun Yat-sen University Cancer Center, first-line therapy of metastatic ccRCC, a disease that is resis-
Guangzhou, People's Republic of China
tant to traditional chemotherapy and radiotherapy and that
Note: Supplementary data for this article are available at Cancer has long had a very poor patient survival rate. Compared
Research Online (http://cancerres.aacrjournals.org/).
with traditional IFN treatment, sunitinib more than doubles
Corresponding Author: Bin Tean Teh, Laboratory of Cancer Genetics,
Van Andel Research Institute, 333 Bostwick Avenue North East, Grand progression-free survival of metastatic ccRCC (11, 12). How-
Rapids, MI 49503. Phone: 616-234-5296; Fax: 616-234-5297; E-mail: ever, not all patients respond to sunitinib, and the vast ma-
[email protected]. jority eventually develop resistance to sunitinib therapy. A
doi: 10.1158/0008-5472.CAN-09-3722 complete understanding of the mechanisms of sunitinib action
©2010 American Association for Cancer Research. against ccRCC is thus critical in understanding and improving

www.aacrjournals.org 1053

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Huang et al.

treatment of this disease. In this study, we evaluated the pri- total PDGFR-β and VEGFR-2 levels, membranes were re-
mary targets of sunitinib in both ccRCC tumor cells and endo- probed with the same anti–PDGFR-β and anti–VEGFR-2 an-
thelial cells. We determined that, at pharmacologically tibody that was used for the immunoprecipitation. Western
relevant concentrations, the primary action of sunitinib in blot analysis of phospho–extracellular signal-regulated ki-
ccRCC is through an antiangiogenic mechanism and not nase (ERK) 1/2 (Cell Signaling), ERK1/2 (Cell Signaling),
through direct targeting of ccRCC cells. phospho-AKT (Cell Signaling), and AKT (Cell Signaling)
was performed on whole-cell lysates (30 μg). β-Actin (Ab-
Materials and Methods cam) was used as a loading control.

Reagents Ligand-Dependent Cell Proliferation Assay

Sunitinib was provided by Pfizer Global Pharmaceuticals Cells were plated in 96-well plates at a density of 3,000 per
and was prepared as a 20 mmol/L stock solution in DMSO well and starved in low serum overnight in a medium con-
(Sigma) for in vitro studies. taining 0.1% FBS. Cells were treated with sunitinib and stim-
ulated with 50 ng/mL recombinant human PDGF-BB or
Expression of RTKs in Clear Cell RCC PDGF-AA or VEGF165 (R&D Systems) and cultured continu-
The mRNA expression of RTK-related genes was measured ously in the presence of sunitinib for 72 h. Cell viability was
in 174 clear cell renal tumors and 15 normal kidney samples assessed using the MTS assay (Promega). MTS assays were
using Affymetrix HGU133 Plus 2.0 microarrays, as described also performed on RCC cells growing in complete medium
elsewhere (13). (10% FBS) treated with sunitinib. Results were expressed as
a percentage of viable cells relative to cells treated with
Cells and Cell Culture DMSO. Experiments were performed in triplicate and repeat-
ACHN, A-498, and 786-O ccRCC cell lines were obtained ed at least three times.
from the American Type Culture Collection. SN12C cells were
kindly provided by Dr. George Vande Woude (Van Andel Re- Small Interfering RNA Transfection
search Institute, originally from National Cancer Institute). RCC cells were seeded in 96-well or 6-well plates and,
The cells were maintained in DMEM or RPMI 1640 (Invitro- 24 h later, transfected with individual small interfering
gen) supplemented with 10% fetal bovine serum (FBS; Invi- RNA (siRNA; Invitrogen) targeting PDGFR-α, PDGFR-β, and
trogen), 100 IU/mL penicillin, and 100 μg/mL streptomycin VEGFR-1, respectively, using Oligofectamine (Invitrogen) ac-
(Invitrogen) in a humidified incubator containing 5% CO2 cording to the manufacturer's protocol. At the same time,
at 37°C. Human umbilical vascular endothelial cells (HUVEC) nontargeting siRNA (Invitrogen) was transfected as a nega-
and human lymphatic microvascular endothelial cells tive control (Neg-L or Neg-H means with low or high GC con-
(HLMVEC) were obtained from Clonetics and maintained tent). Twenty-four, 48, and 72 h after transfection, cell
in Clonetics EBM-2 medium supplemented with EGM-2 Sin- viability was determined using MTS assay, and the knock-
gleQuots or EGM-MV SingleQuots (Lonza). All the cell lines down of individual gene was confirmed by Western blotting
were used within 20 passages. 72 h after transfection.

Immunoprecipitation and Immunoblotting Soft Agar Colony Formation Assay

Cells were starved in low serum (0.1% FBS) overnight be- Soft agar colony formation assay was performed as de-
fore treatment with sunitinib for 2 h. After treatment, cells scribed elsewhere (14). Cells were treated with either suniti-
were stimulated with 50 ng/mL recombinant human nib (0.1, 1, and 5 μmol/L) or 0.1% DMSO (control).
PDGF-BB or VEGF165 (R&D Systems) for 10 min. Cells were Experiments were performed in triplicate and repeated three
then lysed with cold HNTG lysis buffer (50 mmol/L HEPES, times.
150 mmol/L NaCl, 10% glycerol, 1% Triton X-100, 1.5 mmol/L
MgCl2, and 1 mmol/L EGTA) containing protease and phos- Cell Cycle Analysis
phatase inhibitors [10 mmol/L sodium fluoride, 1 mmol/L Cells were incubated with either 0.1 μmol/L sunitinib or
sodium orthovanadate, and protease inhibitor cocktail DMSO (control) for 72 h. Cells were collected and analyzed
(Roche Diagnostics)], and protein concentration was deter- using a cellular DNA flow cytometric analysis kit (Roche Di-
mined using the detergent-compatible protein assay accord- agnostics). Cell cycle profiles were determined by flow cyto-
ing to the manufacturer's instructions (Bio-Rad). Protein metric analysis. Daunorubicin (0.1 μmol/L) was used as a
(1 mg) from each sample was immunoprecipitated overnight positive control for cell cycle arrest at G2-M phase.
at 4°C with an anti–PDGFR-β (Santa Cruz Biotechnology) or
anti–VEGFR-2 (Cell Signaling) antibody and protein G/A aga- Endothelial Cell Invasion Assay
rose beads (Pierce). Immune complexes were washed with Endothelial cell invasion was analyzed using the BD Bio-
cold HNTG lysis buffer containing inhibitors. Proteins were Coat endothelial cell invasion system according to the in-
eluted by boiling in SDS sample buffer, separated by SDS- structions of the manufacturer (BD Biosciences). Briefly,
PAGE, and transferred to nitrocellulose membranes. Mem- HUVEC cells (2 × 104) were plated onto the upper well of
branes were probed with an anti-phosphotyrosine antibody the chamber with medium (serum-free) alone with or with-
(Upstate) and then stripped with stripping buffer. To detect out sunitinib; the lower chamber contained 50 ng/mL VEGF

1054 Cancer Res; 70(3) February 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Mechanism of Action for Sunitinib in ccRCC

(R&D Systems) as a chemoattractant. All experimental sam- proliferation index, and apoptosis index was also performed
ples were set up in duplicate. Cells were incubated at 37°C in as described elsewhere (14).
CO2 for 16 to 24 h, and then residual cells on the upper sur-
face of the chamber membranes were removed using a cot- Statistical Analysis
ton swab. Invaded cells on the lower membrane surface were All values are expressed as mean ± SD. Values were compared
visualized by Diff-Quik staining kit (Allegiance) according to using Student's t test. P < 0.05 was considered significant.
the manufacturer's protocol and examined under the micro-
scope. Pictures of invaded cells were taken for three fields per Results
well under ×100 magnification, and the number of invaded
cells was counted by ImageJ v1.37v software. PDGFR-β and VEGFR-2 are highly expressed in human
ccRCC. To identify potential RTK(s) that would likely be the
Xenograft Models targets of sunitinib in ccRCC, we examined the expression
Tumor implantation and growth. All animal studies were levels of several RTKs, including VEGFR, PDGFR, FLT3, c-
in compliance with Van Andel Research Institute Institution- KIT, and RET, in 174 human ccRCC samples as determined
al Animal Care and Use Committee policies. Six-week-old fe- by gene expression profiling. We found that PDGFR-β and
male BALB/c nu/nu nude mice (Charles River) were given s.c. VEGFR-2, but not others, were highly expressed in human
injections of 3 × 106 A-498, ACHN, SN12C, or 786-O cells in ccRCC when compared with normal controls (P < 0.001;
the right flank. Tumor size was measured twice per week us- Fig. 1A). We also examined the expression and activation
ing digital calipers (Mitutoyo), and tumor volume was calcu- of these RTKs in ccRCC cell lines and endothelial cell lines
lated as length × width × height × 0.5. Tumor growth ratio is by Western blotting. PDGFR-β was expressed by 4 of 12
presented as mean + SD and normalized to the initial volume RCC cell lines, PDGFR-α was expressed by most RCC cell
when sunitinib treatment began. When tumors had grown to lines, VEGFR-1 was expressed in almost all RCC and endo-
an average volume of 200 to 300 mm3, tumor-bearing mice thelial cell lines, whereas VEGFR-2 was expressed only in
were separated into four groups of 10 animals (two groups the endothelial cell lines. However, no significant basal acti-
for A-498 and 786-O xenografts). Three treatment groups re- vation of these RTKs was detected either in the ccRCC cells
ceived an oral gavage of sunitinib as a citrate-buffered (pH or in the endothelial cells (Fig. 1B; data not shown). FLT3, c-
3.5) solution once daily at the dosages of 20, 40, and 80 KIT, and RET expression were undetectable in all cell lines.
mg/kg (40 mg/kg for A-498 and 786-O xenografts), respec- Based on the combination of these data, we focused our
tively. At the same time, one vehicle control group received study on PDGFR and VEGFR signaling in these respective
citrate buffer (pH 3.5) only. Mice were euthanized at the end cells.
of the treatment period. Tumors were removed and fixed in Sunitinib inhibits RTK signaling pathways in ccRCC
4% paraformaldehyde and paraffin embedded, and then 4- cells in vitro. Pharmacokinetic and pharmacodynamic
μm-thick sections were prepared. Some sections were studies have revealed that the pharmacologically relevant
stained with H&E and the others were used for subsequent concentration for sunitinib is 50 to 100 ng/mL (0.125–
immunohistochemical analysis. 0.25 μmol/L; refs. 4, 15). To examine the effect of sunitinib
Target modulation study. To determine the mechanism at these concentrations on the activation of PDGFR-β in
of action of sunitinib in these ccRCC xenograft models, a ccRCC cells, two PDGFR-β–expressing cell lines, SN12C
separate study was designed in which sunitinib treatment and ACHN, were chosen for evaluation in vitro. Sunitinib
was initiated when tumors grew untreated to a size of 400 decreased PDGF-BB–stimulated phosphorylation of
to 500 mm3. Mice were then administered sunitinib or ve- PDGFR-β at 0.01 μmol/L in both cell lines and completely
hicle control at the indicated concentration once daily for inhibited the phosphorylation of PDGFR-β at concentra-
1 to 4 d. At the indicated time after dosing, individual tions ≥0.1 μmol/L (Fig. 1C).
mice were sacrificed and tumors were resected and fixed RTKs transduce signals of proliferation, migration, and
in 4% paraformaldehyde. Microvessel density (MVD) was survival from the extracellular environment through a series
determined by CD34 staining in the control- or sunitinib- of downstream signal transduction pathways. The ERK and
treated tumor sections. Tumor cell proliferation and apo- AKT kinases are two major downstream effectors of PDGFR
ptosis were also determined by proliferating cell nuclear signaling (16, 17). Western blot analysis showed that, in
antigen (PCNA) and terminal deoxynucleotidyl transferase– comparison with DMSO controls, sunitinib inhibited PDGF-
mediated dUTP nick end labeling (TUNEL) staining as BB–stimulated activation of ERK and AKT at ∼0.1 μmol/L
described below. (Fig. 1D), the same concentration shown to completely
inhibit activation of PDGFR-β. Because sunitinib shows
Immunohistochemistry similar inhibitory potency (Ki value) on both PDGFR and
Immunohistochemical staining for PCNA, TUNEL, and VEGFR in biochemical assays (5), we speculate that sunitinib
CD34 was performed as described elsewhere (14). PCNA an- could inhibit the activation of these RTKs simultaneously at
tibody (Abcam) was used at a dilution of 1:500. TUNEL stain- the above concentrations.
ing was performed according to the protocol of the Proliferation of ccRCC cells is not affected by sunitinib
manufacturer (Promega). CD34 (MEC 14.7, Abcam) antibody at concentrations that inhibit RTK signaling. To further ex-
was used at a dilution of 1:50. The quantification of MVD, plore the role of PDGFR and VEGFR signaling pathways in

www.aacrjournals.org Cancer Res; 70(3) February 1, 2010 1055

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Huang et al.

ccRCC cells in vitro, cancer cells were serum starved and then attributable to the direct targeting of ccRCC cells. To test this
PDGFR and VEGFR signaling pathways were activated with hypothesis, we further examined the effect of knockdown of
PDGF and VEGF. The IC50 for sunitinib to inhibit PDGF- or PDGFR and VEGFR on the proliferation of ccRCC cells using
VEGF-dependent proliferation of ccRCC cells was 4 to 10 siRNA in vitro. Knockdown of individual PDGFR or VEGFR
μmol/L, which is much higher than the concentration of su- isoforms did not affect the proliferation of ccRCC cells com-
nitinib sufficient to inhibit RTK phosphorylation (0.01–0.1 pared with the nontargeting siRNA controls (negative; Fig.
μmol/L). Sunitinib showed similar effects on proliferation 2C; Supplementary Fig. S3).
of ccRCC cells in PDGF- or VEGF-only medium and in com- Sunitinib inhibits anchorage-independent growth of
plete medium (Fig. 2A; Supplementary Fig. S1A). This indi- ccRCC cells at higher concentrations. Because anchorage-
cates that inhibition of the PDGFR and VEGFR signaling independent growth is a characteristic of transformed cells,
pathways is not sufficient to inhibit the proliferation of we next used a soft agar colony formation assay to evaluate
ccRCC cells in vitro. To our surprise, PDGF or VEGF alone whether sunitinib could reverse malignant properties of
did not induce the proliferation of ccRCC cells after serum ccRCC cells at pharmacologically relevant concentrations. Al-
starvation (Fig. 2B; Supplementary Fig. S1B), although PDGF though sunitinib significantly inhibited the anchorage-inde-
has been reported to stimulate the proliferation of PDGFR- pendent colony formation of ACHN and A-498 cells at 5
transfected NIH-3T3 cells (4). Cell cycle analysis also showed μmol/L (P < 0.01), it showed weak effects at 1 μmol/L and
no change in cell cycle progression after sunitinib treatment no effect at 0.1 μmol/L (Fig. 3; Supplementary Fig. S4).
compared with DMSO controls (Supplementary Fig. S2). Sunitinib inhibits VEGFR-2–mediated endothelial cell
Knockdown of PDGFR and VEGFR does not affect prolif- growth and invasion in vitro. Because VEGFR has been
eration of ccRCC cells. Based on the above findings, we hy- found important for tumor angiogenesis and is expressed
pothesized that the effect of sunitinib on ccRCC is not in tumor-associated endothelial cells, we next investigated

Figure 1. Expression profile of RTKs and the effect of sunitinib on RTK signaling in ccRCC cells. A, overexpression of PDGFR-β and VEGFR-2 in 174
human clear cell RCC samples (CC) and 15 normal controls (NO) is shown by Affymetrix microarray analysis. Columns, average gene expression level
values of each group; bars, SD. B, expression of PDGFR and VEGFR in ccRCC cell lines and endothelial cell lines is shown by Western blots.
C, phosphorylation of PDGFR-β was inhibited by sunitinib (SU) at pharmacologically relevant concentrations in SN12C and ACHN cells. D, sunitinib
inhibited PDGF-BB–stimulated activation of ERK and AKT signaling in comparison with DMSO controls at about 0.01 to 0.1 μmol/L, the same
concentrations that inhibited activation of PDGFR-β.

1056 Cancer Res; 70(3) February 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Mechanism of Action for Sunitinib in ccRCC

Figure 2. Proliferation of ccRCC cells is not affected by sunitinib

at concentrations that completely inhibit RTK signaling. A,
viability (%) of ACHN and A-498 cells treated with sunitinib under
ligand-dependent and normal conditions. Cell viability was
determined by MTS assay. Results were normalized to DMSO
controls. B, PDGF or VEGF alone failed to stimulate the proliferation
of ccRCC cells in vitro. Results were normalized to starved controls.
S, starve. **, P < 0.01. C, knockdown of individual PDGFR or
VEGFR isoform with siRNA did not affect proliferation of ACHN and
A-498 cells in vitro. Cell viability was determined by MTS assay
and shown as absorbance (A) values detected at different time
points after transfection. D, knockdown of individual RTK by siRNA in
ACHN cells was shown by Western blots.

www.aacrjournals.org Cancer Res; 70(3) February 1, 2010 1057

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Huang et al.

the effect of sunitinib on endothelial cell lines in vitro. lines were treated with sunitinib. Four ccRCC cell lines
First, we examined the expression of the RTKs in five hu- were chosen for xenograft study: A-498 and 786-O cells
man endothelial cell lines. VEGFR-1 and VEGFR-2 were contain loss of heterozygosity VHL, whereas ACHN and
highly expressed in those cell lines; in contrast, PDGFR ex- SN12C cells harbor wild-type VHL. Sunitinib treatment
pression was weak or undetectable (Fig. 1B). Analogous to showed either growth inhibition (40 mg/kg) or stasis (80
its effect on PDGFR activation in ccRCC cell lines, suniti- mg/kg) effects on SN12C xenografts (data not shown),
nib could inhibit the phosphorylation of VEGFR in whereas sunitinib treatment caused stasis effects at 40
HLMVEC cells at concentrations between 0.01 and 0.1 mg/kg on A-498 and 786-O xenografts and induced growth
μmol/L (Fig. 4A). In addition, ERK signaling downstream inhibition at 20 mg/kg and regression at 40 and 80 mg/kg
of VEGFR was also inhibited by sunitinib at this concen- for ACHN xenografts (Fig. 5A; Supplementary Fig. S5A).
tration [VEGFR-mediated activation of AKT (Ser473) was Quantification of tumor MVD at the end of sunitinib treat-
undetectable in our system]. Compared with ccRCC tumor ment showed that MVD was significantly decreased in
cells, sunitinib showed more robust inhibitory effect on the sunitinib-treated tumors compared with vehicle controls
proliferation of HUVEC and HLMVEC endothelial cells. (Fig. 5B; Supplementary Fig. S5B).
IC50 values on endothelial cell lines under VEGF-dependent Sunitinib primarily inhibits ccRCC tumor angiogenesis
conditions were ∼0.01 μmol/L, which was similar to the but not ccRCC cells in vivo. To further elucidate the mech-
concentrations required to inhibit activation of VEGFR-2 anism of sunitinib-mediated growth inhibition or stasis effect
(Fig. 4B). We further found that after serum starvation, in vivo, we collected tumor samples treated with sunitinib for
VEGF alone could induce the proliferation of endothelial 24 to 96 hours and examined the effect of sunitinib on tumor
cells (Fig. 4C). At pharmacologically relevant concentration, vasculature as well as ccRCC cell proliferation and apoptosis.
sunitinib also inhibited VEGF-induced endothelial cell inva- In A-498 xenografts, tumor MVD was significantly inhibited
sion (Fig. 4D). These results show that VEGFR signaling is by sunitinib as soon as 12 hours after treatment compared
required for proliferation and invasion of endothelial cells with vehicle-treated controls (Fig. 6A; also see results for
and that inhibition of VEGFR signaling with sunitinib sup- SN12C xenografts in Supplementary Fig. S6A). The percent-
presses the function of endothelial cells. Thus, all the age of PCNA-positive cells (shown as proliferation index) and
above results indicate that at pharmacologically relevant the percentage of TUNEL-positive cells (shown as apoptosis
concentrations, sunitinib may primarily target tumor endo- index) in the viable tumor section did not significantly alter
thelium rather than ccRCC cells in vivo. up to 72 hours after sunitinib treatment when compared
Sunitinib inhibits RCC xenograft tumor growth and tu- with vehicle controls (Fig. 6B; also see results for SN12C xe-
mor angiogenesis in vivo. To investigate the effect of su- nografts in Supplementary Fig. S6B). This result suggests that
nitinib in vivo and the effect of VHL status on response to the inhibitory effect of sunitinib on xenograft tumor growth
sunitinib, mice bearing tumor xenografts of ccRCC cell in vivo was not caused by direct targeting of ccRCC cells. The

Figure 3. Sunitinib does not affect

the anchorage-independent
growth of ccRCC cells at
pharmacologically relevant
concentrations. Cells were treated
with either sunitinib (0.1, 1, and
5 μmol/L) or 0.1% DMSO
(control). Although sunitinib
significantly inhibited the
anchorage-independent colony
formation of ACHN and A-498 cells
at 5 μmol/L, it showed weak effects
at 1 μmol/L and no effect at
0.1 μmol/L. *, P < 0.05; **, P < 0.01.

1058 Cancer Res; 70(3) February 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Mechanism of Action for Sunitinib in ccRCC

Figure 4. Sunitinib inhibits phosphorylation of VEGFR-2 and VEGF-dependent proliferation and invasion of endothelial cells at pharmacologically relevant
concentrations. A, sunitinib inhibited the phosphorylation of VEGFR-2 and downstream ERK signaling in HLMVEC endothelial cells at pharmacologically
relevant concentrations (0.01–0.1 μmol/L). B, sunitinib inhibited VEGF-dependent proliferation of endothelial cells as analyzed by MTS assay. Results
were normalized to DMSO controls. C, VEGF induced the proliferation of endothelial cells in vitro. **, P < 0.01. D, sunitinib inhibited VEGF-dependent
invasion of endothelial cells at pharmacologically relevant concentrations. VEGF was used as a chemoattractant for control (con)–treated, DMSO-treated,
or sunitinib (0.1 μmol/L)–treated samples; negative (neg) means no chemoattractant added. **, P < 0.01.

inhibition of endothelial cell function by sunitinib in vitro tion of sunitinib required to inhibit RTK signaling in our cell
and the reduction of MVD in the sunitinib-treated tumors lines (0.01–0.1 μmol/L; see Fig. 1). Other groups have also re-
in vivo indicate that the antitumor effect of sunitinib is main- ported that sunitinib concentrations much lower than 5
ly mediated by inhibition of tumor angiogenesis. μmol/L are sufficient to inhibit RTK signaling. In cellular as-
says, Mendel and colleagues (4) found that sunitinib inhib-
Discussion ited ligand-dependent phosphorylation of VEGFR-2 and
PDGFR-β with IC50 values of ∼0.01 μmol/L, consistent with
Sunitinib is currently considered first-line therapy for the our own findings. Finally, both clinical and preclinical phar-
treatment of advanced ccRCC. However, the relationship be- macodynamic and pharmacokinetic studies indicate that
tween clinical efficacy of sunitinib on ccRCC and its mecha- pharmacologically and clinically relevant plasma concentra-
nism of action has not been established. To study this tions of sunitinib are in the range of 50 to 100 ng/mL or 0.1 to
problem, we directly examined the effect and action of suni- 0.2 μmol/L (4, 15). Thus, we believe that the high concentra-
tinib on ccRCC cells and endothelial cells in vitro and in vivo tions of sunitinib (∼5 μmol/L), which Xin and colleagues
at pharmacologically relevant concentrations. Our studies in- observed to inhibit proliferation and viability of ccRCC cells,
dicate that sunitinib inhibits ccRCC growth primarily are likely not pharmacologically or clinically relevant.
through an antiangiogenic mechanism and not through di- In our study, we observed that pharmacologically relevant
rect targeting of ccRCC tumor cells. concentrations of sunitinib (∼0.1 μmol/L) did not affect the
A recent study by Xin and colleagues (10) reported that viability or proliferation of ccRCC cell lines in vitro. In con-
sunitinib induces apoptosis in ccRCC cells through inhibition trast, sunitinib did inhibit endothelial cell proliferation and
of signal transducer and activator of transcription 3. Howev- invasion and did so at concentrations that also inhibited
er, Xin and colleagues found that the IC50 values required to VEGFR signaling in these cells. Moreover, studies of ccRCC
inhibit ccRCC cell viability and proliferation in vitro were in xenografts treated with pharmacologically relevant concen-
the range of 5 μmol/L and above, >50 times the concentra- trations of sunitinib showed that sunitinib rapidly inhibits

www.aacrjournals.org Cancer Res; 70(3) February 1, 2010 1059

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Huang et al.

tumor angiogenesis in vivo but did not affect the proliferation sion) was not sufficient to inhibit the proliferation of ccRCC
or apoptosis of ccRCC tumor cells. Together, these data indi- cells. Knockdown of VEGFR-1 expression in ccRCC cells also
cate that sunitinib acts primarily to target tumor vasculature failed to inhibit proliferation. Thus, PDGFR and VEGFR-1 sig-
rather than tumor cells in the treatment of ccRCC. naling in ccRCC cells are likely redundant growth signaling
Our analysis of RTK expression in ccRCC suggested that pathways. In contrast, we found that VEGFR signaling is re-
PDGFR and VEGFR were likely relevant targets of sunitinib quired for the proliferation of endothelial cells. Sunitinib in-
in the treatment of ccRCC. PDGFR-β and VEGFR-2 were hibited phosphorylation of VEGFR- and VEGF-dependent
both overexpressed in human ccRCC tumor samples com- proliferation of endothelial cells at similar concentrations
pared with normal tissue. Analysis of ccRCC and endothelial (0.01 μmol/L), indicating the important role of VEGFR signal-
cell lines found that, of the known sunitinib-sensitive RTKs, ing for proliferation of endothelial cells.
only PDGFR-α, PDGFR-β, and VEGFR-1 were expressed by Although PDGFR signaling may not be required for growth
ccRCC cells, whereas VEGFR-1 and VEGFR-2 were highly ex- of ccRCC cells, it has been shown as a requirement for the sur-
pressed by endothelial cells. Overexpression of c-KIT has vival of pericytes, perivascular cells surrounding newly formed
been shown in chromophobe RCC and renal oncocytoma blood vessels, in several cancer types (22–24). In the neovascu-
but not ccRCC (18, 19). Although higher PDGFR-α expression larization process, endothelial cells secrete growth factors that
has been correlated with RCC progression (20, 21), our in vitro recruit pericytes to the newly formed blood vessel to stabilize
studies showed that inhibition of PDGFR signaling (by either and mature the vascular network. PDGF-B is expressed by
sunitinib treatment or siRNA knockdown of PDGFR expres- sprouting capillary endothelial cells, whereas its receptor

Figure 5. Sunitinib inhibits ccRCC xenograft tumor growth and tumor angiogenesis in vivo. A, growth of ACHN and A-498 xenograft tumors was
inhibited by sunitinib. **, P < 0.01. B, decreased MVD was found in sunitinib-treated tumor sections as evaluated by CD34 staining. Scale bar, 0.20 mm.
CON, control. *, P < 0.05; **, P < 0.01.

1060 Cancer Res; 70(3) February 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Mechanism of Action for Sunitinib in ccRCC

Figure 6. Target modulation study shows that sunitinib primarily acts on tumor endothelium rather than via direct targeting of ccRCC cells in vivo.
A, sunitinib inhibited A-498 xenograft tumor angiogenesis as soon as 12 h after treatment. *, P < 0.05; **, P < 0.01. B, sunitinib showed minimal effect on
A-498 tumor cell proliferation and did not induce apoptosis of A-498 cells up to 72 h after treatment in vivo.

PDGFR-β is localized on pericytes, suggesting a paracrine sig- giogenic effect and also not through direct targeting of ccRCC
naling circuit between the two cell types (25–27). Although su- cells. This understanding of sunitinib action should have im-
nitinib does not affect the proliferation of ccRCC cells, our portant implications for the improved treatment of ccRCC.
preliminary results indicate that it might inhibit pericytes sur-
vival through PDGFR suppression (Supplementary Fig. S7) and Disclosure of Potential Conflicts of Interest
contribute to the inhibition of tumor angiogenesis. Further
studies are needed to validate this hypothesis. B.T. Teh: commercial research grant, Pfizer Global Pharmaceuticals. The
Mutations in the VHL gene are associated with the major- other authors disclosed no potential conflicts of interest.
ity of ccRCC tumors (28). We did not see a correlation be-
tween VHL mutational status and response to sunitinib. Acknowledgments
We assessed the effect of sunitinib on four ccRCC xenografts
(two with VHL−/− and two with VHL+/+). The most sensitive We thank Eric Kort (Laboratory of Cancer Genetics) for the generous
xenograft was the VHL+/+ ACHN (40 mg/kg sunitinib causes sharing of his image analysis software; Lisa DeCamp (Vivarium Operations,
regression). The mid-sensitive xenografts were A-498 and Van Andel Research Institute) for her help with the animal experiments;
Kristin Buzzitta (Laboratory of Cancer Genetics, Van Andel Research
786-O (VHL−/−; 40 mg/kg sunitinib causes stasis effect). The Institute) and Bree Berghuis, Eric Hudson, and J.C. Goolsby (Laboratory of
least sensitive xenograft was SN12C (VHL+/+; 40 mg/kg suni- Analytical, Cellular, and Molecular Microscopy, Van Andel Research Institute)
tinib causes growth inhibition). The lack of correlation be- for their help in immunohistochemical staining; Rich West (Laboratory of Cell
Structure and Signal Integration) for his help in fluorescence-activated cell
tween VHL mutational status and response to sunitinib sorting analysis; Vanessa Fogg and David Nadziejka (Van Andel Research
therapy is consistent with clinical observations. In a retro- Institute) for scientific and technical editing of the manuscript; and Sabrina
Noyes (Van Andel Research Institute) for assisting in preparation and sub-
spective analysis of ccRCC patients, Choueiri and colleagues mission of the manuscript.
(29) found that VHL mutation status had little effect on
patient response to sunitinib therapy; response rates of
VHL-mutated and VHL wild-type ccRCC patients to sunitinib Grant Support
therapy were 56% and 52%, respectively.
Pfizer Global Pharmaceuticals.
In summary, we found that, under pharmacologically rele- The costs of publication of this article were defrayed in part by the payment
vant concentrations, sunitinib targets the ccRCC tumor endo- of page charges. This article must therefore be hereby marked advertisement in
thelium and not ccRCC cells directly. To our knowledge, this accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
is the first study to clearly show that the clinical efficacy of Received 10/7/09; revised 11/26/09; accepted 12/1/09; published OnlineFirst
sunitinib on ccRCC is mediated primarily through an antian- 1/26/10.

www.aacrjournals.org Cancer Res; 70(3) February 1, 2010 1061

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Huang et al.

References
1. Abrams TJ, Lee LB, Murray LJ, Pryer NK, Cherrington JM. SU11248 pathways suppressed renal cell carcinoma growth and angiogenesis
inhibits KIT and platelet-derived growth factor receptor β in preclin- in vivo. Cancer Res 2008;68:81–8.
ical models of human small cell lung cancer. Mol Cancer Ther 2003; 15. Faivre S, Delbaldo C, Vera K, et al. Safety, pharmacokinetic, and anti-
2:471–8. tumor activity of SU11248, a novel oral multitarget tyrosine kinase
2. Abrams TJ, Murray LJ, Pesenti E, et al. Preclinical evaluation of the inhibitor, in patients with cancer. J Clin Oncol 2006;24:25–35.
tyrosine kinase inhibitor SU11248 as a single agent and in combina- 16. Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell
tion with “standard of care” therapeutic agents for the treatment of 2000;103:211–25.
breast cancer. Mol Cancer Ther 2003;2:1011–21. 17. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell 2000;100:
3. Ikezoe T, Nishioka C, Tasaka T, et al. The antitumor effects of suni- 57–70.
tinib (formerly SU11248) against a variety of human hematologic ma- 18. Pan CC, Chen PC, Chiang H. Overexpression of KIT (CD117) in chro-
lignancies: enhancement of growth inhibition via inhibition of mophobe renal cell carcinoma and renal oncocytoma. Am J Clin
mammalian target of rapamycin signaling. Mol Cancer Ther 2006;5: Pathol 2004;121:878–83.
2522–30. 19. Yamazaki K, Sakamoto M, Ohta T, Kanai Y, Ohki M, Hirohashi S.
4. Mendel DB, Laird AD, Xin X, et al. In vivo antitumor activity of Overexpression of KIT in chromophobe renal cell carcinoma. Onco-
SU11248, a novel tyrosine kinase inhibitor targeting vascular endo- gene 2003;22:847–52.
thelial growth factor and platelet-derived growth factor receptors: 20. Sulzbacher I, Birner P, Traxler M, Marberger M, Haitel A. Expression
determination of a pharmacokinetic/pharmacodynamic relationship. of platelet-derived growth factor-αα receptor is associated with tu-
Clin Cancer Res 2003;9:327–37. mor progression in clear cell renal cell carcinoma. Am J Clin Pathol
5. Faivre S, Demetri G, Sargent W, Raymond E. Molecular basis for su- 2003;120:107–12.
nitinib efficacy and future clinical development. Nat Rev Drug Discov 21. Tawfik OW, Kramer B, Shideler B, Danley M, Kimler BF, Holzbeierlein
2007;6:734–45. J. Prognostic significance of CD44, platelet-derived growth factor re-
6. Ikezoe T, Yang Y, Nishioka C, et al. Effect of SU11248 on gastro- ceptor α, and cyclooxygenase 2 expression in renal cell carcinoma.
intestinal stromal tumor-T1 cells: enhancement of growth inhibition Arch Pathol Lab Med 2007;131:261–7.
via inhibition of 3-kinase/Akt/mammalian target of rapamycin signal- 22. Bergers G, Song S, Meyer-Morse N, Bergsland E, Hanahan D.
ing. Cancer Sci 2006;97:945–51. Benefits of targeting both pericytes and endothelial cells in the tu-
7. Seandel M, Shia J, Linkov I, Maki RG, Antonescu CR, Dupont J. The mor vasculature with kinase inhibitors. J Clin Invest 2003;111:
activity of sunitinib against gastrointestinal stromal tumor seems to 1287–95.
be distinct from its antiangiogenic effects. Clin Cancer Res 2006;12: 23. Sennino B, Falcon BL, McCauley D, et al. Sequential loss of tu-
6203–4. mor vessel pericytes and endothelial cells after inhibition of platelet-
8. Prenen H, Cools J, Mentens N, et al. Efficacy of the kinase inhibitor derived growth factor B by selective aptamer AX102. Cancer Res
SU11248 against gastrointestinal stromal tumor mutants refractory 2007;67:7358–67.
to imatinib mesylate. Clin Cancer Res 2006;12:2622–7. 24. Shaheen RM, Tseng WW, Davis DW, et al. Tyrosine kinase inhibition
9. Greenman C, Stephens P, Smith R, et al. Patterns of somatic muta- of multiple angiogenic growth factor receptors improves survival in
tion in human cancer genomes. Nature 2007;446:153–8. mice bearing colon cancer liver metastases by inhibition of endothe-
10. Xin H, Zhang C, Herrmann A, Du Y, Figlin R, Yu H. Sunitinib inhibition lial cell survival mechanisms. Cancer Res 2001;61:1464–8.
of Stat3 induces renal cell carcinoma tumor cell apoptosis and re- 25. Bergers G, Benjamin LE. Tumorigenesis and the angiogenic switch.
duces immunosuppressive cells. Cancer Res 2009;69:2506–13. Nat Rev 2003;3:401–10.
11. Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon 26. Gerhardt H, Betsholtz C. Endothelial-pericyte interactions in angio-
alfa in metastatic renal-cell carcinoma. N Engl J Med 2007;356: genesis. Cell Tissue Res 2003;314:15–23.
115–24. 27. Holash J, Thurston G, Rudge JS, et al. Inhibitors of growth factor
12. Motzer RJ, Michaelson MD, Redman BG, et al. Activity of SU11248, receptors, signaling pathways and angiogenesis as therapeutic mo-
a multitargeted inhibitor of vascular endothelial growth factor recep- lecular agents. Cancer Metastasis Rev 2006;25:243–52.
tor and platelet-derived growth factor receptor, in patients with me- 28. Nickerson ML, Jaeger E, Shi Y, et al. Improved identification of von
tastatic renal cell carcinoma. J Clin Oncol 2006;24:16–24. Hippel-Lindau gene alterations in clear cell renal tumors. Clin Cancer
13. Furge KA, Chen J, Koeman J, et al. Detection of DNA copy number Res 2008;14:4726–34.
changes and oncogenic signaling abnormalities from gene expres- 29. Choueiri TK, Vaziri SA, Jaeger E, et al. von Hippel-Lindau gene status
sion data reveals MYC activation in high-grade papillary renal cell and response to vascular endothelial growth factor targeted therapy
carcinoma. Cancer Res 2007;67:3171–6. for metastatic clear cell renal cell carcinoma. J Urol 2008;180:860–5,
14. Huang D, Ding Y, Luo WM, et al. Inhibition of MAPK kinase signaling discussion 865–6.

1062 Cancer Res; 70(3) February 1, 2010 Cancer Research

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.
Published OnlineFirst January 26, 2010; DOI: 10.1158/0008-5472.CAN-09-3722

Sunitinib Acts Primarily on Tumor Endothelium rather than

Tumor Cells to Inhibit the Growth of Renal Cell Carcinoma
Dan Huang, Yan Ding, Yan Li, et al.

Cancer Res 2010;70:1053-1062. Published OnlineFirst January 26, 2010.

Updated version Access the most recent version of this article at:
doi:10.1158/0008-5472.CAN-09-3722

Supplementary Access the most recent supplemental material at:

Material http://cancerres.aacrjournals.org/content/suppl/2010/01/26/0008-5472.CAN-09-3722.DC1.html

Cited Articles This article cites by 29 articles, 16 of which you can access for free at:
http://cancerres.aacrjournals.org/content/70/3/1053.full.html#ref-list-1

Citing articles This article has been cited by 20 HighWire-hosted articles. Access the articles at:
http://cancerres.aacrjournals.org/content/70/3/1053.full.html#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].

Downloaded from cancerres.aacrjournals.org on May 13, 2015. © 2010 American Association for Cancer Research.