Tota

l In

tegr

ated

Sig

nal p

er C

ell i

n R

OI

Development of a Quantitative Immunofluorescence Imaging Assay to Assess b-catenin Translocation and Multiplex Biomarkers for Epithelial-Mesenchymal Transition (EMT) in Tumor Tissues

Tony Navas 1, Robert J. Kinders 1, Karun Mutreja 1, Scott M. Lawrence 1, Donna Butcher 2, Melinda Hollingshead 3, Ralph E. Parchment 1, Joseph E. Tomaszewski 4, and James H. Doroshow 4

1Laboratory of Human Toxicology and Pharmacology, Applied/Developmental Directorate, SAIC-Frederick, Inc., Frederick National Laboratory, Frederick MD 21702; 2Pathology/Histotechnology Laboratory, Laboratory Animal Sciences Program, SAIC-Frederick, Inc., Frederick National Laboratory, Frederick, MD 21702;

3Biological Testing Branch, Developmental Therapeutics Program, National Cancer Institute, Frederick, MD 21702; 4Division of Cancer Treatment and Diagnosis and Center for Cancer Research, National Cancer Institute, Bethesda, MD 20810

Background Wnt/β-catenin signaling and cadherin-mediated cell adhesion are two processes that promote cell proliferation and epithelial-mesenchymal transition (EMT) of carcinomas. Therapeutic inhibitors of Wnt-induced tumor activation prevent stabilization and nuclear translocation of β-catenin and putatively lead to inhibition of tumor metastasis by upregulation of E-Cadherin and downregulation of N-Cadherin and Vimentin.

Abstract # 1500

Conclusions

Target Mab Source Host,

Clonality Clone

ID Species

Reactivity Target

Specificity

Chamber Slides (ICF)

10% NBF; Triton-X

FFPE (IHF)

10% NBF HIER Citrate

E-Cadherin-AF488 BD Biosciences Ms, Mab Clone 36 Hu, Rat, Ms, Dg (Western, IFA)

N-Cadherin-AF647 BD Biosciences Ms, Mab Clone 32 Hu, Rat, Ms, Chk (Western, IFA)

b -Catenin-AF546 (CTNNB1)

Epitomics Rb, Mab E247D Hu, Rat, Ms (Western, IFA)

Vimentin-AF790 Abcam Ms, Mab V9 Hu, Dg, Rat, Ch (Western, IFA)

Pancytokeratin-AF790 Santa Cruz Ms, Mab C11 Hu, Ms, Rat (IFA)

EMT Multiplex IFA Assay Panel 0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

HCT116 Parental HCT116 Betacat (D45/-)

B-cat/GAPDH HCT116 Parental

CTNNB1 (D45/WT)

HCT116 KO

CTNNB1 (D45/-)

IFA: Western:

Validation of b-catenin antibody specificity on parental vs. CTNNB1 heterozygous KO HCT116 cell lines by confocal immunofluorescence and Western analysis

b-catenin Antibody Validation in CTNNB1 Knockout Cell Line

A431 HT29 PC3 SW620

N-Cadherin + Anti-Ms AF488

293T

A431 293T HT29 PC3 SW620

E-Cadherin-AF488

A431 293T HT29 PC3 SW620

b-catenin + Anti-Rb AF488

Summary of EMT Multiplex IFA Staining on CRC, TNBC, and Melanoma FFPE Tumor Cell Models

Multiplex IFA Staining Validation of EMT Biomarkers in Mesenchymal and Epithelial Tumor Models

Multiplex Staining Validation of EMT Biomarkers on Rat Kidney Tissue

Fitness For Purpose Assay Validation: b-catenin Nuclear Translocation In Vitro

0

20

40

60

80

100

0h 4h 8h 24h 0h 4h 8h 24h

DMSO GSK3 b inhibitor

% Cell w/ Low Expression

% Cell w/ Med Expression

% Cell w/ High Expression

% C

ells

wit

h N

ucl

ear b

-cat

enin

0h 4h 8h 24h

GSK-3bi

DMSO

Time course for b-catenin Nuclear Translocation In Vitro

EMT Multiplex IFA Assay Validation: FL-1

A multiplex quantitative immunofluorescence assay (qIFA) was developed to assess changes in levels of EMT biomarkers (E-Cadherin, N-Cadherin, Vimentin/Pancytokeratin) as well as monitor intracellular translocation of β-catenin in tumor tissues. A panel of highly specific monoclonal antibodies against EMT targets, directly conjugated to various fluorophores, was optimized for multiplex staining of FFPE tissues using colorectal (CRC) and triple-negative breast (TNBC) cancer cell lines and xenograft tissues as tumor controls for target expression. Rat kidney was used as a positive control for endogenous biomarker expression in tissues. Image acquisition was performed with confocal and epifluorescence microscopy. Image segmentation and quantitative multiplex analysis were performed using NIS Elements and Definiens Tissue Studio software to obtain total mean integrated signal per biomarker within a tissue region of interest.

Vimentin (Clone V9)

Pancytokeratin (Clone C11)

A375 HCT116 HT29 MDA-MB-231 SKMEL28

A375 HCT116 HT29 MDA-MB-231 SKMEL28

A37

5

HT2

9

HCT

116

MD

A-M

B-2

31

SKM

EL-2

8

GAPDH

Vimentin

Vimentin and Pancytokeratin Antibody Validation

Validation of monoclonal antibody specificity on epithelial or mesenchymal tumor cell models by Immunofluorescence Confocal Microscopy and Western Analysis

b -Catenin- AF546

N-Cadherin- AF647

E-Cadherin- AF488

Vimentin-AF790/ Pancytokeratin-

AF790

DAPI

MD

A-M

B-4

35

BT5

49

PC

3

DU

14

5

MC

F7

N-Cadherin

GAPDH

E-Cadherin

Actin

DU145

E-Cadherin (Clone 36)

N-Cadherin (Clone 32)

PC3 MDA-MB-435

MCF7 PC3 MDA-MB-435

PC3 DU145 MCF7 BT549

BT549

Validation of monoclonal antibody specificity on epithelial and mesenchymal tumor cell models by Immunofluorescence and Western Analysis

E-Cadherin and N-Cadherin Antibody Specificity Validation:

Staining Validation of Various EMT Targets on Control FFPE Tumor Cell Models by Confocal Microscopy

Cell Line E-Cadherin CTNNB1 N-Cadherin Vimentin Tumor Type

SW620 +

(Patchy)

+++ Membrane, Cytoplasm

Nucleus -

+ (Patchy)

CRC

Colo201 +++ +++

Membrane, Cytoplasm - - CRC

Colo205 +++ +++

Membrane, Cytoplasm - - CRC

DLD-1 ++ +++

Membrane, Cytoplasm - - CRC

HT29 ++ ++

Membrane - - CRC

HCT116 - +

Membrane - - CRC

MDA-MB-231 - + - +++ TNBC

BT549 - + +++ +++ TNBC

A375 - + ++ ++ Melanoma

MDA-MB-435 - ++

Membrane +++ ++ Melanoma

SKMEL28 + +

Membrane + ++ Melanoma

A375 BT549 Colo205 HT29 SKMEL28

Mesenchymal Epithelial

0

200

400

600

800

1000

1200

A375 BT549 Colo201 Colo205 DLD1 HCT116 HT29 MDA MB 435

SKMEL28 SW620 BT549 Iso Colo201 Iso

Betacatenin

E-Cadherin

N-Cadherin

Vimentin

To

tal In

teg

rate

d S

ign

al

pe

r C

ell i

n R

OI

Definiens Tissue Studio Analysis: N-Cadherin, E-Cadherin, Vimentin, and b-catenin

GSK-3bi

LPA +

GSK-3bi

Merge Cytoplasmic Mask Nuclear Mask

DMSO

LPA

Fitness for purpose evaluation of the biomarkers was performed in vitro to demonstrate nuclear translocation of b-catenin. A431 squamous epithelial cell lines treated with lysophosphatidic acid (LPA) induced adherens junction (AJ) dissociation of membrane E-Cadherin and b-catenin, while co-treatment with specific inhibitors to glycogen synthase kinase-3b (GSK-3b) induced a 5-fold increase in the nuclear- to-cytoplasmic mean intensity ratio of b-catenin after 24h.

Merge E-Cadherin b-catenin N-Cadherin Vimentin

Definiens Tissue Studio Quantitative Analysis:

ROI 2 ROI 1

ROI Analysis Nuclear Mask Membrane Mask

Fitness For Purpose Assay Validation: E-Cadherin to N-Cadherin Membrane Intensity Ratio Quantitation

E-Cadherin N-Cadherin Merge + DAPI

Control

24h TGF

48h TGF

72h TGF

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

No TGF 24h 72h

E:N-Cadherin Ratio in TGF-Treated NMuMG Cells

ROI 1

ROI 2

ROI 3

E:N

Ra

tio

(M

ea

n In

ten

sit

y)

TGF-b-treatment of the NMuMG mouse epithelial cell line induced a significant decrease in E- to N-Cadherin membrane intensity ratio after 24 and 72h in vitro, determined from images segmented and analyzed with NIS Elements software.

We have developed a quantitative imaging assay that integrates multiplex changes in EMT biomarkers to assay the pharmacodynamic activities of potential targeted agents currently in development at the National Cancer Institute (NCI). This assay could also be applied for retrospective analysis of clinical samples to document drug-induced changes in tumor phenotypes. Funded by NCI Contract No. HHSN261200800001E.

EMT Target Protein or Peptide Source Soluble/

Purity His Tag

removed Molecular

Weight

Molar Excess

For Blocking Total μg

N-Cadherin 42-mer Peptide NEP 95% N/A 4.7 k 200 fold 49.1 μg

b-catenin Rec. FL Protein ANL Yes/SDS Yes 87.8 kD 10 fold 37.1 μg

E-Cadherin Rec. C-terminus (a.a. 731-882)

ANL Yes Yes 17.4 kD 20 fold 18.7 μg

Vimentin Rec. FL Protein Abcam 9M Urea

Dialyzed into PBS Yes 54.0 kD 10 fold 28.8 μg

Colorectal (CRC), melanoma or triple-negative breast cancer (TNBC) tumor pellets were stained with the validated panel of EMT multiplex IFA antibodies and images were analyzed by Definiens Tissue Studio software to quantify biomarker levels of N-Cadherin (red), E-Cadherin (green), Vimentin (blue), and b-catenin (gray). Bar graphs show the total integrated signal per cell in the region of interest (ROI), which reflects the variations in EMT biomarkers evident in particular mesenchymal to epithelial phenotypes.

Rat kidney tissue was stained with the validated panel of EMT multiplex IFA or isotype control antibodies and images were analyzed by Definiens Tissue Studio software to determine relative levels of N-Cadherin (pink), E-Cadherin (green), b-catenin (red) and Vimentin (blue) biomarkers. Rat kidneys endogenously express each biomarker (tissue expression varies by location and relative levels). Rat kidney tissue could be used as an assay calibrator for reagent lots and for validation of the multiplex staining method. The upper panel shows expression of each biomarker within different epifluorescent channels. Bar graphs show the total integrated signal per cell in the ROI, corrected against the background obtained from an isotype control-stained adjacent tissue section.

Epithelial to Mesenchymal Transition (EMT):

E-Cadherin

N-Cadherin

Vimentin

Fibronectin

Nuclear translocation of β-catenin (CTNNB1)

Wnt Pathway Activation

0

200

400

600

800

1000

1200

EMT Multiplex EMT Isotype EMT minus Betacatenin EMT minus N-cadherin EMT minus vimentin vimentin N-cadherin Betacatenin

Betacatenin

N-Cadherin

Vimentin

FL -1 Single FL

Multiplex Isotype

EMT

EMT Minus

b-catenin

EMT Minus

N-Cadherin

EMT Minus

Vimentin

Single FL

Vimentin N-Cadherin b-catenin

Multiplex

EMT (FL-1)

BT549 Tissue Studio Analysis: b-catenin, N-Cadherin, and Vimentin

To

tal In

teg

rate

d S

ign

al

pe

r C

ell i

n R

OI

Fluorescence Minus One experiment compares the fluorescence signals from each channel when all 4 antibodies are mixed together (EMT multiplex) or any one of the antibodies is left out (FL-1), versus when only one antibody is used for staining tissue sections (single FL). This experiment determines whether any of the antibodies interact within the multiplex mixture during staining (antibody reactivity) as well as determines any potential fluorescence overlap or interference that may arise based on the multiplex IFA platform (signal interference).

Recombinant Protein and Peptide Blockers of EMT Antibodies

EMT Multiplex IFA Target Validation: b-catenin and N-Cadherin Target Blocking

0

100

200

300

400

500

600

700

Betacatenin + buffer Betacatenin + Betacatenin Block EMT Multiplex + buffer EMT Multiplex + Betacatenin Block

EMT Multiplex + N-Cadherin Block

EMT Isotype

Betacatenin

N-Cadherin

Vimentin

A375 Blocking Expt. Studio Analysis: β-catenin, N-Cadherin, and Vimentin

EMT Multiplex Isotype Control

+

N-Cadherin peptide

b-catenin

+

blocking buffer

+

b-catenin protein

+

b-catenin protein

+

blocking buffer

+

blocking buffer

To

tal In

teg

rate

d S

ign

al p

er

Ce

ll in

RO

I

0

50

100

150

200

ROI 1 ROI 2

Rat Nephron EMT Multiplex Bckg Corr

N-Cadherin

E-Cadherin

Betacatenin

Vimentin

Target validation was performed by incubating each antibody with molar-fold excess concentrations of the recombinant protein or peptide corresponding to each target at 4 0C overnight then staining tissue sections with the EMT multiplex antibody mixture. Complete blocking of target binding was observed when the signal from each EMT antibody was blocked by the overexpressed target either alone or in combination with the multiplex of EMT antibodies, demonstrating specificity of each Mab to its tissue target.

The Frederick National Laboratory is accredited by AAALAC International and follows the Public Health Service Policy for the Care and Use of Laboratory Animals. Animal care was provided in accordance with the procedures outlined in the "Guide for Care and Use of Laboratory Animals" (National Research Council; 1996; National Academy Press; Washington, D.C.)

TIFF Image Region of Interest

(ROI) N-Cadherin

Marker Mask DAPI

Nuclear Mask

Isotype Control

EMT Multiplex

# of nuclei = 264 Marker intensity = 48

Marker Area (um2) = 1942

8

353

b-catenin Marker Mask

Marker intensity = 91

Marker Area (um2) = 343

118

12

Composite

N-Cadherin = 345

β-catenin = 106

Definiens Tissue Studio : Relative EMT Biomarker Signal Analysis on Mouse Testis

Total integrated Signal per cell in ROI =

(Marker intensity) X (Marker area) *

(# of nuclei)

*Average marker intensity and area were thresholded based on isotype control-stained adjacent tissue section

# of nuclei = 235 Marker intensity = 38

Marker Area (um2) = 51

Marker intensity = 36

Marker Area (um2) = 82

Results

Methods