Download Targeting Cancer Initiating Cells in Pancreatic Cancer

Targeting Cancer Initiating Cells in Pancreatic Cancer
Matteo Ligorio1, Francesco Sabbatino2,Yangyang Wang2, Elvira Favoino2, Ling Yu2, Soldano Ferrone2, Xinhui Wang2, Keith D.
Lillemoe1, Cristina R. Ferrone1
1
Department of Surgery, Massachusetts General Hospital, Harvard Medical School
2 Department of Immunology, University of Pittsburgh
Introduction!
Pancreatic ductal adenocarcinoma (PDAC) is the most lethal tumor of the
gastrointestinal track1.
The American Cancer Society predicts 37,390 deaths and 43,920 new cases of
PDACs in 20121. Currently, surgical resection is the only therapy which can provide a
5 year survival and even with a complete surgical resection, and an adjuvant therapy,
the actual 5 year survival is only 12%. The majority of patients succumbs to the
disease within 20-24 months from their diagnosis2.
The cancer stem cell theory predicts the presence of a subpopulation of cancer
cells, the cancer initiating cells (CICs), which has the potential to initiate and
sustain tumor progression3. In patients, these cells are thought to be responsible for
local recurrence and distant metastases and they have to be eradicated in order to
“cure” a malignant disease.
Recent studies have demonstrated that a subpopulation of human PDAC cells
which expresses high levels of aldehyde-dehydrogenase1A1, ALDHbright cells,
can be identified as a pancreatic cancer initiating population4.
These findings underscore the urgent need of novel therapeutic strategies which have
the potential to target differentiated cancer cells as well as CICs.
To address this need, we have recently focused our attention on a new cancerspecific antigen, the glucose-regulated protein 94 kDa (Grp94). Grp94 is a
member of the heat shock protein 90 kDa (HSP90) family and is selectively expressed
on the cell surface of cancer cells, but not on normal cells5.
To target this antigen we have generated a fully human monoclonal antibody
(mAb), W9, which recognizes an extracellular epitope of Grp94.
Hypothesis!
Results!
Generation of mAb W9 and epitope recognition
Human mAb W9
scFv W9 light chain variable
region
secrete
Phage Display
Library
Cancer cells
ABCB5 shRNA
100$KD$
VH
Myeloma cell line
P3-X63-Ag8.653 cells
R.ELISNASDLLDK.I$
K.GVVDSDDLPLNVSR.$E$
scFv W9 heavy chain
variable region
Not$transduced$
Transduced$
Figure 3. Proof that mAb W9 uniquely binds
Grp94. mAb W9 does not react with the human
melanoma FO-1 cells in which Grp94 mRNA was
knocked down by transduction with lentivirus
carrying Grp94 siRNA. (ABCB5) siRNA was used as
a control.
Figure 2. Experimental verification that Grp94 is the
specific antigen recognized by mAb W9. Chromatographytandem mass spectrometry analysis of the 100KDa molecule
specifically precipitated by mAb W9 from the human melanoma
WM1158 cell lysate reveals two triptic peptides uniquely
derived from Grp94. The unrelated Ab 119 was used as control.
Figure 1. Generation of fully human antibody W9. scFv W9 variable light (VL) and
heavy (VH) regions were excised from the phagemid vector and cloned into pFUSE2CLIg-hk and pFUSE-CHIg-hG1, respectively. The two plasmids were then co-transfected
in NS0 mouse myeloma cells to allow the expression of a completely recombinant IgG1
antibody, the mAb W9.
AIM.2 Grp94 is expressed in PDAC cell lines
AIM.1 mAb W9 stains human PDAC lesions but not human tissues
B
A
Breast
mAb W9 specifically targets pancreatic cancer cells as well as cancer initiating
cells and inhibits their growth.
Normal Cells
Grp94 shRNA
Grp94
VL
Cancer Initiating Cells
Testis
Medulla
oblongata
Cerebrum
Thyroid gland
Ovary
Colon
Esophagus
Kidney
Lung
Pancreas
Placenta
N=12
= Grp94 - Endoplasmic Reticulum Expression
Prostate
= Grp94 - Cell Surface Expression
Skeletal
muscle
Skin
Small intestine
Human mAb W9
Specific Aims!
Figure 5. Expression of the extracellular Grp94 epitope in several
human pancreatic adenocarcinoma cell lines. Tumor cells were
incubated on ice with mAb W9 for 30 min. Binding was detected using
RPE-labeled F(ab’)2 fragments of goat anti-human IgG Fcγ Ab. Stained
cells were analyzed by flow cytomety. Percentage of stained cells and
Mean Fluorescence Intensity (MFI) are indicated. Human Ig (HIg) was used
as control.
Figure 4. Staining of human pancreatic adenocarcinomas and human normal tissues with mAb W9. Frozen slides of pancreatic
adenocarcinoma from 12 patients (A) and a microarray of normal human tissues derived from many different organs (B) were stained
with mAb W9. 10 out of 12 human PDAC lesions were positive, while all normal pancreatic tissues from the 12 patients were negative
for mAb W9 immunostaining. A representative image of PDAC lesions and normal pancreas tissues is shown in panel A. The panel B
shows a broad range of normal tissues negatively stained for mAb W9. Frozen sections derived from cell lines MV3 (Grp94+) and Raji
(Grp94neg) were simultaneously stained as positive and negative controls (data not shown).
1.  Testing whether mAb W9 selectively stains human PDAC lesions, but not
normal human tissues.
2.  Determining if Grp94 is expressed in PDAC cell lines.
3.  Evaluating the ability of mAb W9, alone or in combination with other chemotherapeutic agents, to inhibit the in vitro proliferation of cancer cells and a
subpopulation of pancreatic cancer initiating cells, ALDHbright cells.
AIM.3. mAb W9 inhibits the in vitro proliferation of cancer cells and cancer initiating cells, ALDHbright cells
A
mAb W9+ 5-FU
1.  mAb W9 selectively stains human PDAC lesions, but not normal human
tissues.
MIA PACA-2
PANC3.27
PANC2.03
Methods!
Optical Density
ALIVE CELLS
2.  Grp94 is overexpressed on the cell surface membrane of several human
PDAC cell lines.
3.  mAb W9 significantly inhibits the in vitro proliferation of differentiated cancer
cells as well as a subpopualtion of CICs, ALDHbright cells, alone or in
combination with other chemotherapeutic agents, such as 5-FU,
radiotherapy and cyclopamine (a sonic hedgehog pathway inhibitor).
**
**
**
**
***
***
mAb W9 + RadioTherapy + Cyclopamine
mAb W9+ 5-FU
**
**
***
***
MIA PACA-2
B ALDEFLUOR + DEAB
ALDEFLUOR
ALDHBright Cells (%)
Conclusions!
**
**
**
**
***
***
*
***
! Generation of mAb W9: scFv W9 was isolated from a phage display scFv library with human Grp94+
melanoma cells WM1158; the variable light (VL) and heavy (VH) antibody regions of scFv W9 were cloned for
the expression of a completely recombinant antibody W9 (Fig. 1).
! Immunoprecipitation and mass spectrometry: This approach was used to proof that Grp94 is the specific
antigen recognized by mAb W9 (Fig. 2).
! siRNA knock-down Grp94 expression: A siRNA targeting Grp94 was used to verify the unique specificity of
mAb W9 (Fig. 3).
* p value < 0.05; ** p value < 0.01; *** p value < 0.001
** p value < 0.01; *** p value < 0.001
TREATMENTS
TREATMENTS
References!
1.  http://www.cancer.org/Cancer/PancreaticCancer/DetailedGuide/pancreatic-cancer-key-statistics
2.  Oettle, et Al. Drugs, 2007. Adjuvant therapy in pancreatic cancer: a critical appraisal.
3.  Visvader, et Al. Nat Rev Cancer, 2008. Cancer stem cells in solid tumours: accumulating evidence and
unresolved questions.
4.  Visus, et Al. Clin Cancer Res, 2011. Targeting ALDH(bright) human carcinoma-initiating cells with ALDH1A1specific CD8 T cells.
5.  Altmeyer, et Al. Int J Cancer, 1996. Tumor-specific cell surface expression of the-KDEL containing,
endoplasmic reticular heat shock protein gp96.
Figure 6. Growth inhibition of two human PDAC cell lines, PANC2.03 AND PANC3.27,
using mAb W9 and 5-FU. Cells (4x105/ml) were starved for 3 hours then incubated with
mAb W9 (2 µg/ml) and/or 5-FU (10 µM) in RPMI 1640 medium containing 1.5% FCS. Cell
viability was then tested by MTT assay. HIg was used as a negative control. Results were
shown as optical density and error bars represent the standard deviation. The mAb W9 and
5-FU were able to inhibit the proliferation of both cancer cell lines (p<0.01). The
combination of mAb W9 and 5-FU seems to have an additive effects in inhibiting the
proliferation of cancer cells in both cell lines (p<0.001). !
Figure 6. ADLHbright have a higher in vivo
tumorigenicity and a high expression of Grp94.
(A) 500 sorted ALDHbright and ALDHnegative MIA
PaCa-2 cells were injected into 3 NOD/SCID mice.
Tumor growth was observed after 4 months. (B)
MIA PaCa-2 cells were incubated with
ALDEFLUOR and its inhibitor, DEAB, and with
mAb W9. All the ALDHpositive cells express high
level of Grp94 on their surface membrane.!
Figure 9. The decrease of ALDHbright cells, the cancer initiating population, in a human PDAC cell
line after the treatment with mAb W9, cyclopamine and radiation therapy. MIA PaCa-2 cells were
incubated with mAb W9 (2µg/ml) and cyclopamine (20 µM) for 48 hrs at 37°C, and/or irradiated with 20
Gys. After incubation, cells were harvested and stained with ALDEFLUOR, with or without its inhibitor,
the DEAB. Cells were then analyzed by flow cytometry to obtain the number of ALDHbright cells. HIg was
used as a negative control. Results are shown as the percentage of cancer initiating cells after
treatment. mAb W9, cyclopamine and radiation therapy were able to significantly reduce the number of
ALDHbright cells in monotherapy; furthermore their combination was able to eradicate them.!