INTRODUCTIONAngiogenesis, the process of forming new blood vessels from existing vasculature, plays a central role in the development and metastasis of many solid tumors. Tumors are unable to grow beyond a few millimeters in diameter unless new blood vessels are created to supply actively proliferating tumor cells with nutrients and oxygen. Tumor cells release multiple angiogenic factors to stimulate vascular growth. Consequently, inhibiting tumor angiogenesis has been investigated as a mechanism for preventing cancer progression.

Angiogenesis inhibitors are being paired with traditional chemotherapy drugs to treat multiple metastatic cancers including breast cancer. The mechanisms utilized to block tumor-associated angiogenesis include antibodies that bind and sequester circulating angiogenic factors, antibodies that bind and block activation of cell-surface receptors, and small molecule tyrosine kinase inhibitors that disrupt angiogenic signal transduction pathways (1). However, phase III trials testing the use of antiangiogenic drugs in treating metastatic breast cancer have yet to report an overall survival benefit with these agents (1,2).

The heterogeneity of many cancers presents a challenge in the development of effective antiangiogenic agents. The heterogeneous

nature of breast cancer is reflected in the diversity of breast cancer cell lines, which exhibit a variety of malignant gene expression patterns including DNA polymorphisms in genes known to be important in the angiogenesis pathway (3,4). Different breast cancer cell lines secrete various combinations of angiogenic factors, which can confound preclinical experiments that utilize single analyte assays to investigate a potential antiangiogenic agent. The use of multianalyte profiling techniques can provide a more detailed analysis of the full range of angiogenic factors secreted by cancer cells and screen potential antiangiogenic agents for efficacy or off-target activity in a cost-effective and timely manner.

THE PROBLEM• Quickly identifying multiple angiogenic factors secreted by various

breast cancer cell lines

• Performing pharmacological screens of antiangiogenic compounds in a timely manner

• Identifying off-target pharmacological effects

David Finkel, Amy James, Greta Wegner

Using Antibody Arrays for Profiling Angiogenic Factors, RTK Phosphorylation, and Off-Target Responses of Potential Angiogenesis Inhibitors

WHITE PAPER

THE PROBLEMIt is difficult to quickly screen the effects of potential antiangiogenic agents on multiple secreted angiogenic factors and phosphorylated receptor tyrosine kinases (RTKs) when using time-consuming traditional techniques such as Western blot.

THE SOLUTIONProteome Profiler™ Antibody Arrays can rapidly detect multiple secreted factors and screen pharmacological responses of angiogenesis inhibitors. There are no gels to run and no proteins to transfer. If you can collect data from an immunoblot, you have the equipment necessary to run an array experiment today. For details please visit RnDSystems.com/ProteomeProfiler.

Molecules and Phosphorylation Sites Detected

Membrane-based Array

Many Analytes, Select Samples

Array Membrane

Chemiluminescent Substrate Light

Plate-based Array

Select Analytes, Many Samples

Capture Antibody

Target AnalyteDetection Antibody

HRP

Arrays are composed of capture and control antibodies spotted in duplicate on either nitrocellulose membranes or on the bottoms of transparent 96-well polystyrene plates. For both array formats, cell supernatants or extracts were diluted, mixed with a cocktail of biotinylated detection antibodies, and incubated overnight with either the Proteome Profiler Human Angiogenesis Antibody Array (Catalog # ARY007), the Proteome Profiler Human Phospho-RTK Antibody Array (Catalog # ARY001B), the Proteome Profiler 96 Human Phospho-RTK Array1 (Catalog # ARZ001), or the Proteome Profiler 96 Human Phospho-RTK Array 3 (Catalog # ARZ003). Streptavidin-HRP and chemiluminescent detection reagents were applied to the arrays, and the signal produced at each capture spot corresponded to the amount of protein bound.

Proteome Profiler Human Phospho-RTK Array(Catalog # ARY001B)

ALK/CD246 EphA10 FGF R4 ROR1

Axl EphB1 Flt-3/Flk-2 ROR2

DDR1 EphB2 HGF R/c-MET Ryk

DDR2 EphB3 IGF-I R SCF R/c-kit

Dtk EphB4 Insulin R/CD220 Tie-1

EGF R EphB6 M-CSF R Tie-2

EphA1 ErbB2 Mer TrkA

EphA2 ErbB3 MSP R/Ron TrkB

EphA3 ErbB4 MuSK TrkC

EphA4 FGF R1 PDGF Rα VEGF R1/Flt-1

EphA5 FGF R2α PDGF Rβ VEGF R2/KDR

EphA6 FGF R3 c-Ret VEGF R3/Flt-4

EphA7

Proteome Profiler 96 Human Phospho-RTK Array 1(Catalog # ARZ001)

Proteome Profiler 96 Human Phospho-RTK Array 3(Catalog # ARZ003)

RS

Erb-B2

Erb-B4

Erb-B3

INSR

M-CSF R

SCFR

RSPDGFRβ

PDGFRα

VEGF R2

VEGFR3

EGFR

IGF-IR

MSPR

HGFR

Tie-2 VEGF R1

EphB4

Tie-1

VEGFR2

VEGFR3

RS

Tie-2 VEGFR1

PDGFRα

PDGFRβ

Proteome Profiler Human Angiogenesis Array(Catalog # ARY007)

Activin A Endostatin/Collagen XVIII CXCL8/IL-8 PlGF

ADAMTS-1 Endothelin-1 LAP (TGF-β1) Prolactin

Angiogenin FGF acidic Leptin Serpin B5/Maspin

Angiopoietin-1 FGF basic CCL2/MCP-1 Serpin E1/PAI-1

Angiopoietin-2 FGF-4 CCL3/MIP-1α Serpin F1/PEDF

Angiostatin/Plasminogen FGF-7/KGF MMP-8 TIMP-1

Amphiregulin GDNF MMP-9 TIMP-4

Artemin GM-CSF NRG1-β1 Thrombospondin-1

Tissue Factor/Factor III HB-EGF Pentraxin 3 Thrombospondin-2

CXCL16 HGF PD-ECGF uPA

DPPIV/CD26 IGFBP-1 PDGF-AA Vasohibin

EGF IGFBP-2 PDGF-AB/PDGF-BB VEGF

EG-VEGF IGFBP-3 Persephin VEGF-C

Endoglin/CD105 IL-1β CXCL4/PF4

THE SOLUTIONArrays that can simultaneously detect multiple proteins increase the amount of information collected from a single experiment. They can reveal changes in protein expression or phosphorylation in context, providing a more complete understanding of the overall intracellular response to treatment. Proteome Profiler Antibody Arrays are designed to simultaneously measure the expression of multiple proteins in a single sample. These multianalyte assays are available in either a membrane-based or a microplate-based format. The membrane-based assays are ideal for profiling a wide range of proteins within a small number of samples and require no specialized equipment beyond what is used for data acquisition when employing Western blot. The microplate-based assays require a small sample size and are amenable to high-throughput analysis of a specific set of cellular proteins. Both assay formats can be performed in approximately 3.5 hours.

To highlight the arrays’ utility, we profiled the relative expression of 55 angiogenesis-related proteins in supernatants from several breast cancer cell lines using the Proteome Profiler Human Angiogenesis Antibody Array. We also screened a panel of VEGF R inhibitors to assess their effects on RTK phosphorylation using the Proteome Profiler Human Phospho-RTK Antibody Array and Proteome Profiler 96 Phospho-RTK Antibody Arrays. Our results showed variations in angiogenic factors secreted by the various breast cancer cell lines and revealed that inhibitors varied in their effectiveness of blocking RTK phosphorylation in each cell line.

MATERIALS AND METHODSProteome Profiler Array Assay Principle

Additional Kits and ReagentsELISA kits from R&D Systems were used to confirm protein or phosphorylated RTK levels: Human Phospho-VEGF R2/KDR DuoSet® IC ELISA Development System (Catalog # DYC1766), Human Angiogenin Quantikine® ELISA Kit (Catalog # DAN00), Human VEGF Quantikine ELISA Kit (Catalog # DVE00), and Human TIMP-1 Quantikine ELISA Kit (Catalog # DTM100). Recombinant RTK ligands from R&D Systems used included: Human VEGF165 (Catalog # 293-VE), PDGF-BB (Catalog # 220-BB), EGF (Catalog # 236-EG) and NRG1-b1/HRG1-b1 (Catalog # 396-HB). Tocris Bioscience small molecule inhibitors included: Axitinib (Catalog # 4350), DMH4 (Catalog #4471), (E)-FeCP-oxindole (Catalog # 3882), (Z)-FeCP-oxindole (Catalog # 3883), Ki 8751 (Catalog # 2542), and ZM 323881 hydrochloride (Catalog # 2475).

RESULTS

Profiling Secreted Angiogenic Factors in Supernatants from Breast Cancer Cell LinesProfile analysis of 55 angiogenic factors in the supernatants from the T47D, MCF-7, and MDA-MB-453 human breast cancer cell lines demonstrated that all three cell lines secreted multiple angiogenesis-related proteins (Figure 1A). Comparison of the relative expression levels of VEGF, Angiogenin, and TIMP-1 in the supernatants from the three breast cancer cell lines revealed that the highest levels of VEGF were present in MDA-MB-453 supernatant while the TIMP-1 was highest in T47D and MCF-7 supernatants (Figure 1B). Concentrations of VEGF, Angiogenin, and TIMP-1 in the breast cancer cell supernatants were also determined using Quantikine ELISAs. The semi-quantitative results obtained from the Proteome Profiler membrane array were comparable to protein concentrations obtained from the ELISAs (Figure 1B).

Conditioned Media from Breast Cancer Cells Phosphorylates VEGF R2Conditioned media from various breast cancer cells lines were tested for their ability to induce RTK phosphorylation. HUVEC human umbilical vascular endothelial cells were treated with conditioned media from MCF-7, T47D, MDA-MB-468, and MDA-MB-453 human breast cancer cell lines, and cell extracts were then analyzed for RTK phosphorylation. Conditioned media from all four breast cancer cells lines induced phosphorylation of VEGF R2; however, the highest levels of phospho-VEGF R2 were found in the extracts of HUVEC cells treated with conditioned media from MDA-MB-453 cells (Figure 2A). VEGF R2 phosphorylation in HUVEC cell extracts was also analyzed using a DuoSet IC ELISA Development System. Results obtained from the two detection methods were comparable (Figure 2B).

T47D CellsAngiogenin

TIMP-1 VEGF

MCF-7 Cells Angiogenin

TIMP-1 VEGF

MDA-MB-453 CellsAngiogenin

TIMP-1 VEGF

T47D MCF-7 MDA-MB-453M

ean

Pix

el D

ensi

ty VE

GF (ng/m

L)

7000

14000

21000

28000

0

3

6

9

12

15

0

35000Membrane-based ArrayELISA

VEGF

T47D MCF-7 MDA-MB-453

Mea

n P

ixel

Den

sity

Angiogenin (ng/m

L)

7000

14000

21000

28000

0

0.25

0.5

0.75

1.0

0

35000Membrane-based ArrayELISA

Angiogenin

T47D MCF-7 MDA-MB-453

Mea

n P

ixel

Den

sity TIM

P-1 (ng/m

L)

10000

20000

30000

40000

0

20

40

60

80

0

50000

TIMP-1

Membrane-based ArrayELISA

A

B

Figure 1. Profiling Angiogenic Proteins in Breast Cancer Cell Line Supernatants. A) Images of Proteome Profiler Human Angiogenesis Antibody Array and the corresponding histogram profiles are shown. The Proteome Profiler membrane array was used to detect secreted angiogenic proteins in the supernatant from T47D, MCF-7, and MDA-MB-453 human breast cancer cell lines. B) Pixel densities obtained from the Proteome Profiler membrane array (blue bars) and protein concentrations obtained from Quantikine ELISAs (gray bars) are shown for selected analytes. Results obtained using these detection methods are comparable.

Figure 1.

learn more | RnDSystems.com/ProteomeProfiler

Measuring Efficacy and Off-Target Responses of VEGF R InhibitorsHUVECs were used to generate a VEGF-induced phosphorylation profile and to assess the effects of several small molecule VEGF R inhibitors (Figure 3). The results indicate the varied levels of efficacy of the different inhibitors. Axitinib and Ki 8751 were the most potent inhibitors of VEGF R1 and VEGF R2 phosphorylation. However, DMH4, (E)-FeCP-oxindole, and (Z)-FeCP-oxindole displayed little or no effect on VEGF-induced phosphorylation of VEGF R1 and VEGF R2. Axitinib also inhibited VEGF R2 phosphorylation induced by conditioned media from the MDA-MB-453 human breast cancer cell line (Figure 4). The effects of Axitinib appear to be specific to VEGF R1 and VEGF R2. Axitinib did not block the phosphorylation of PDGF-Ra or PDGF-Rb induced by PDGF-BB, EGF-induced EGF R phosphorylation, or the phosphorylation of ErbB2, ErbB3, or ErbB4 induced by NRG1-b1/HRG1-b1 (Figure 5).

Untreated (E)-FeCP-oxindole/VEGF

VEGF (Z)-FeCP-oxindole/VEGF

(Z)-FeCP-oxindole/VEGF

Axitinib/VEGF Ki 8751/VEGF

DMH4/VEGF ZM 323881 hydrochloride/VEGF

VEGF R2VEGF R1

MDA-MB-453 CM PDGF-BB EGF NRG1-β1/HRG1-β1

VEGF R2 PDGF-Rα PDGF-Rβ EGF RErbB2

ErbB3ErbB4

AxitinibTreated

Untreated

A) HUVECs B) CCD-1070Sk Cells C) A431 Cells D) MDA-MB-453 Cells

Reference Spot

0.10 30.3 1

Mea

n P

ixel

Den

sity

7000

14000

21000

0

28000

VEGF R2

Axitinib (nM)+ MDA-MB-453 CM

Reference Spot

Untreated

Array Wells

EphB4

Figure 3. Induction and Inhibition of RTK Phosphorylation in HUVECs. HUVEC human umbilical vascular endothelial cells were untreated or treated with 100 nM of the VEGF R inhibitors Axitinib, DMH4, (E)-FeCP-oxindole, (Z)-FeCP-oxindole, Ki 8751 or ZM 323881 hydrochloride for 45 hours. Cells were subsequently treated with 50 ng/mL Recombinant Human VEGF165 for 5 minutes or remained untreated. Phosphorylated VEGF R1 and VEGF R2 in cell extracts were detected using the Proteome Profiler Human Phospho-RTK Antibody Array. Images of Proteome Profiler membrane arrays are shown.

Figure 4. Phosphorylation of VEGF R2 by Media from Cultured MDA-MB-453 Cells. HUVEC human umbilical vascular endothelial cells were untreated or treated with increasing concentrations of the VEGF R inhibitor Axitinib for 45 minutes. Cells were subsequently treated with conditioned media (CM) from the MDA-MB-453 human breast cancer cell line for 5 minutes or remained untreated. Phosphorylation of VEGF R2 was detected using the Proteome Profiler 96 Human Phospho-RTK Array 3. Well images of each treatment condition are shown along with corresponding histograms.

Figure 5. Screening for Off-Target RTK Dephosphorylation. Cell lines were untreated or treated with 10 nM of the VEGF R inhibitor Axitinib for 45 minutes and then treated with conditioned media or a RTK-specific ligand for 5 minutes. The different experimental conditions included: A) HUVEC human umbilical vascular endothelial cells treated with conditioned media (CM) from the MDA-MB-453 human breast cancer cell line; B) the CCD-1070Sk human foreskin fibroblast cell line treated with 100 ng/mL Recombinant Human PDGF-BB; C) the A431 human epithelial carcinoma cell line treated with 100 ng/mL Recombinant Human EGF; and D) the MDA-MB-453 human breast cancer cell line treated with 100 ng/mL NRG1-b1/HRG1-b1. Phosphorylation of 16 different RTKs can be detected using the Proteome Profiler 96 Human Phospho-RTK Array 1. Well images for each treatment condition are shown.

Figure 2. Phosphorylation of VEGF R2 by Media from Cultured Breast Cancer Cells. HUVEC human umbilical vascular endothelial cells were treated for 5 minutes with conditioned media from the MCF-7, T47D, MDA-MB-468, and MDA-MB-453 human breast cancer cell lines. Phosphorylated VEGF R2 in HUVEC cell extracts was detected using the Proteome Profiler Human Phospho-RTK Antibody Array and the Human Phospho-VEGF R2/KDR DuoSet IC ELISA Development System. Spot images of Proteome Profiler membrane arrays and the corresponding histogram profiles (blue bars) are shown. The optical densities obtained from the DuoSet IC ELISA (gray bars) are also shown. Results obtained using these detection methods are comparable

T47DMCF-7 MDA-MB-453MDA-MB-468

T47DMCF-7A

B

MDA-MB-453MDA-MB-468

Mea

n P

ixel

Den

sity O

ptical Density

5000

10000

15000

0

0.2

0.4

0.6

0.8

0

20000

Membrane-based ArrayELISA

Figure 2.

Figure 3.

Figure 4.

Figure 5.

WP_2.14_AntibodyArrays

CONCLUSIONSThe Proteome Profiler Human Angiogenesis Array and the Proteome Profiler and Proteome Profiler 96 Human Phospho-RTK Arrays are economical alternatives to traditional methods for screening differences in soluble protein expression and phosphorylation. The membrane-based and plate-based array assays are easy to perform. Both can be completed in approximately 3.5 hours, making them far more time-effective than performing multiple IP-Western blots. In addition, these assays employ a chemiluminescence detection method, thus, no specialized equipment beyond what is typically used to collect Western blot data is required. These arrays are also sensitive enough to compare changes in the phosphorylation of RTKs caused by both ligand and inhibitor treatment, making them extremely useful for performing screens to assess the efficacy and/or off-target effects of potential pharmacological agents.

REFERENCES1. Rugo, H.S. (2012) J. Clin. Oncol. 30:898.2. Kerbel, R.S. (2011) Breast 20:S56.3. Wehland, M. et al. (2012) Curr. Pharm. Des. 18:4244.4. Schneider, B.P. & K.D. Miller (2005) J. Clin. Oncol. 23:1782.

learn more | RnDSystems.com/ProteomeProfiler