Comprehensive screening for antigens overexpressed on carcinomas via isolation of human mAbs that may be therapeutic

Article (PDF Available)inProceedings of the National Academy of Sciences 105(20):7287-92 · June 2008with31 Reads
DOI: 10.1073/pnas.0712202105 · Source: PubMed
Abstract
Although several murine mAbs that have been humanized became useful therapeutic agents against a few malignancies, therapeutic Abs are not yet available for the majority of the human cancers because of our lack of knowledge of which antigens (Ags) can become useful targets. In the present study we established a procedure for comprehensive identification of such Ags through the extensive isolation of human mAbs that may become therapeutic. Using the phage-display Ab library we isolated a large number of human mAbs that bind to the surface of tumor cells. They were individually screened by immunostaining, and clones that preferentially and strongly stained the malignant cells were chosen. The Ags recognized by those clones were isolated by immunoprecipitation and identified by MS. We isolated 2,114 mAbs with unique sequences and identified 21 distinct Ags highly expressed on several carcinomas. Of those 2,114 mAbs 356 bound specifically to one of the 21 Ags. After preparing complete IgG1 Abs the in vitro assay for Ab-dependent cell-mediated cytotoxicity (ADCC) and the in vivo assay in cancer-bearing athymic mice were performed to examine antitumor activity. The mAbs converted to IgG1 revealed effective ADCC as well as antitumor activity in vivo. Because half of the 21 Ags showed distinct tumor-specific expression pattern and the mAbs isolated showed various characteristics with strong affinity to the Ag, it is likely that some of the Ags detected will become useful targets for the corresponding carcinoma therapy and that several mAbs will become therapeutic agents. • phage Ab library • therapeutic Ab • tumor-associated antigen

Figures

Figure
Figure
Figure

Full-text (PDF)

Available from: Nobuhiro Hayashi
Comprehensive screening for antigens overexpressed
on carcinomas via isolation of human mAbs that
may be therapeutic
Gene Kurosawa
a
, Yasushi Akahori
a
, Miwa Morita
a
, Mariko Sumitomo
b
, Noriko Sato
c
, Chiho Muramatsu
b
, Keiko Eguchi
b
,
Kazuki Matsuda
b
, Akihiko Takasaki
d
, Miho Tanaka
a
, Yoshitaka Iba
a
, Susumu Hamada-Tsutsumi
a
, Yoshinori Ukai
e
,
Mamoru Shiraishi
e
, Kazuhiro Suzuki
e
, Maiko Kurosawa
a
, Sally Fujiyama
f
, Nobuhiro Takahashi
f
, Ryoichi Kato
g
,
Yoshikazu Mizoguchi
h
, Mikihiro Shamoto
i
, Hiroyuki Tsuda
j
, Mototaka Sugiura
k
, Yoshinobu Hattori
l
, Shuichi Miyakawa
l
,
Ryoichi Shiroki
c
, Kiyotaka Hoshinaga
c
, Nobuhiro Hayashi
d
, Atsushi Sugioka
l
, and Yoshikazu Kurosawa
a,m
a
Division of Antibody Project and
d
Department of Biomedical Polymer Science, Institute for Comprehensive Medical Science,
b
21st Century Center of
Excellence Research Center, Departments of
c
Urology,
g
Radiology,
h
Pathology,
k
Internal Medicine, and
l
Surgery, School of Medicine, Fujita Health
University, Toyoake, Aichi 470-1192, Japan;
e
Institute for Antibodies, Ltd., Toyoake, Aichi 470-1192, Japan;
f
Department of Biotechnology,
United Graduate School of Agriculture, Tokyo University of Agriculture and Technology, Saiwai-cho, Fuchu-shi, Tokyo 183-8509,
Japan;
i
Division of Clinical Laboratory, Yachiyo Hospital, Anjo, Aichi 446-8510, Japan; and
j
Department of Toxicology,
Graduate School of Medical Sciences, Nagoya City University , Kawazumi, Mizuho, Nagoya 467-8601, Japan
Edited by Frederick W. Alt, Harvard Medical School, Boston, MA, and approved March 14, 2008 (received for review December 29, 2007)
Although several murine mAbs that have been humanized became
useful therapeutic agents against a few malignancies, therapeutic
Abs are not yet available for the majority of the human cancers
because of our lack of knowledge of which antigens (Ags) can
become useful targets. In the present study we established a
procedure for comprehensive identification of such Ags through
the extensive isolation of human mAbs that may become thera-
peutic. Using the phage-display Ab library we isolated a large
number of human mAbs that bind to the surface of tumor cells.
They were individually screened by immunostaining, and clones
that preferentially and strongly stained the malignant cells were
chosen. The Ags recognized by those clones were isolated by
immunoprecipitation and identified by MS. We isolated 2,114
mAbs with unique sequences and identified 21 distinct Ags highly
expressed on several carcinomas. Of those 2,114 mAbs 356 bound
specifically to one of the 21 Ags. After preparing complete IgG
1
Abs
the in vitro assay for Ab-dependent cell-mediated cytotoxicity
(ADCC) and the in vivo assay in cancer-bearing athymic mice were
performed to examine antitumor activity. The mAbs converted to
IgG
1
revealed effective ADCC as well as antitumor activity in vivo.
Because half of the 21 Ags showed distinct tumor-specific expres-
sion pattern and the mAbs isolated showed various characteristics
with strong affinity to the Ag, it is likely that some of the Ags
detected will become useful targets for the corresponding carci-
noma therapy and that several mAbs will become therapeutic
agents.
phage Ab library therapeutic Ab tumor-associated antigen
S
ince the discovery of a method to produce mAbs numerous
scientists have been trying to identify and produce mAbs that
could be used for immunotherapy against various malignancies.
The success for example of alemtuzumab against CD52, trastu-
zumab against HER2, and rituximab against CD20 for treatment
of chronic lymphocytic leukemia, breast cancer, and non-
Hodgkins lymphoma, respectively (1–3), suggests that mAbs are
likely to become very important therapeutic agents also against
a wider range of cancers. However, for the majority of the human
cancers useful therapeutic Abs are not yet available because of
our lack of knowledge of which antigens (Ags) are likely to
become useful targets (4). Therefore, several groups of investi-
gators have been trying to identify other potential Ags as targets
for immunotherapy using microarray technology (5, 6). Al-
though many differences in transcripts have been revealed
between malignant cells and the normal counterpart cells, it will
take more time and laborious work to examine which Ags could
be targets and to prepare therapeutic Abs against them. Fur-
thermore, the presence of a large amount of transcripts does not
always indicate expression of a large amount of the proteins.
Our experimental approach was designed in the opposite way
to the strategy with the microarray technology mentioned above
and was based on the phage-display technology (7). First we
isolated a large number of mAbs that bind to the surface of
cancer cells using a huge phage Ab library and many kinds of
cancer-derived cell lines. Then using fresh tumor tissues we
selected clones that gave significant staining of malignant cells
but were negative or very weakly positive on the normal cells in
the histological sections. At the third step the Ags recognized by
the respective clones were isolated by immunoprecipitation and
identified by MS analysis. Finally mAbs were converted to
complete human IgG
1
and the antitumor activity was examined.
Thus, the procedure adopted in our study enabled us to succeed
in comprehensive identification of tumor-associated Ags
(TAAs) and simultaneous isolation of mAbs against them.
Recently several groups of investigators have been using the
phage-display method to screen for tumor-specific Ags accord-
ing to a procedure similar to ours (8, 9), but the number of TAAs
identified by them was limited, and to the best of our knowledge
none converted their clones to complete Abs, which are essential
for further studies to try to evaluate their potential therapeutic
effects.
Results
Isolation of mAbs That Differentially Bound to Cancer Cells. Using 33
different tumor cell lines from seven carcinomas, hepatocarci-
noma, renal carcinoma, pancreatic carcinoma, lung carcinoma,
colonic carcinoma, gastric carcinoma, and ovarian carcinoma,
the phage Ab library was screened 51 times for isolation of mAbs
that bound to molecules present on the cell surface. The number
Author contributions: Y.K. designed research; G.K., Y.A., M.M., M. Sumitomo, N.S., C.M.,
K.E., K.M., M.T., Y.I., S.H.-T., Y.U., M. Shiraishi, K.S., and M.K. performed research; A.T., S.F.,
and N.T. contributed new reagents/analytic tools; G.K., Y.A., M.M., A.T., S.F., N.T., R.K.,
Y.M., M. Shamoto, H.T., M. Sugiura, Y.H., S.M., R.S., K.H., N.H., and A.S. analyzed data;
and Y.K. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Freely available online through the PNAS open access option.
m
To whom correspondence should be addressed at: Institute for Comprehensive Medical
Science, Fujita Health University, 1-98, Dengakugakubo, Kutsukake-cho, Toyoake, Aichi
470-1192, Japan. E-mail: kurosawa@fujita-hu.ac.jp.
© 2008 by The National Academy of Sciences of the USA
www.pnas.orgcgidoi10.1073pnas.0712202105 PNAS
May 20, 2008
vol. 105
no. 20
7287–7292
MEDICAL SCIENCES
Advertisement:
of clones that were picked up in each screening is indicated in the
column ‘‘Isolated’’ in Table 1. A total of 9,395 clones were picked
up. Those clones were then screened by ELISA using anti-cp3 Ab
to examine expression of the intact single-chain Fv (scFv)
molecules on the phage because scFv fused with a truncated cp3
was expressed in our system. The number of clones that were
Table 1. Summary of screenings
Cancer type Screening no. Cell Isolated Intact Kinds Select Super Select
Hepatocarcinoma 035 HepG2 240 162 91
040 Nuk-1 286 254 100
041 OCTH-18 239 197 86
042 HepG2 96 21 20
044 Hep3B 190 120 112
045 HepG2 428 270 189
046 Clinical sample 0722* 190 156 14
047 Clinical sample 0722* 142 116 18
048 Clinical sample 0722* 142 137 8
049 Clinical sample 0722* 190 160 36
050 Clinical sample 0722* 190 138 51
051 HepG2 168 68 49
052 HepG2 208 187 94
053 HepG2 208 149 71
063 HLF 190 141 45
3172 Clinical sample 0317
111
054 RBE 250 204 106
Total 17 3,358 2,481 1,091 967
Renal carcinoma 057 Caki-1 190 168 90
059 CCF-RC1 190 148 80
061 Caki-1 190 146 53
062 CCF-RC1 190 140 111
060 ACHN 190 160 97
Total 5 950 762 431 341
Pancreatic carcinoma 055 PANC-1 286 181 62
058 MIA PaCa-2 190 159 50
085 BxPC-3 190 145 61
087 Capan-1 190 44 27
Total 4 856 529 200 180
Lung carcinoma 064 A549 189 172 56
065 PC-14 379 349 60
066 NCI-H441 190 167 71
068 Calu-3 48 34 22
067 EBC-1 285 210 107
079 RERF-LC-AI 190 172 73
080 LK-2 190 158 86
086 VMRC-LCP 190 177 69
Total 8 1,661 1,439 544 437
Colonic cancer 028 Caco-2 190 170 102
029 CW-2 190 153 92
082 SW480 190 175 46
084 HT-29 190 177 70
Total 4 760 675 310 279
Gastric cancer 031 MKN-45 190 159 90
075 NCI-N87 190 145 50
077 SNU-5 190 143 65
081 KATO III 190 162 79
Total 4 760 609 284 240
Ovarian cancer 015 SKOv3 240 183 81
021 SKOv3 48 10 9
022 SKOv3 48 15 10
025 SKOv3 48 10 6
026 SKOv3 48 20 8
039 SKOv3 48 43 36
074 KF28 190 143 60
076 RMG-1 190 177 76
078 RMG-2 190 176 79
Total 9 1,050 777 365 287
Total 51 9,395 7,272 3,225 2,731 2,114
*Clinical sample 0722 was derived from a male patient with hepato cell carcinoma HCV () stage II.
Clinical sample 0317 was derived from a male patient with hepato cell carcinoma HBV () stage IV-B.
7288
www.pnas.orgcgidoi10.1073pnas.0712202105 Kurosawa et al.
judged to express the intact molecule is indicated in the column
‘‘Intact’’ in Table 1. A total of 7,272 clones turned out to express
intact scFv molecules on the phage. Each one of those 7,272
clones was sequenced. The number of clones with different
sequences isolated in respective screenings is indicated in the
column ‘‘Kinds’’ in Table 1. Because the same clones were
redundantly isolated from different screenings, the total number
of different clones against the same carcinoma is shown in the
column ‘‘Select’’ in Table 1. Because the same clones were also
isolated from screenings against different types of carcinoma,
7,272 clones were composed of 2,114 different clones indicated
as ‘‘Super Select’’ in Table 1. Of those 2,114 clones 406 were
redundantly present in the 3,225 clones summed up in the
column ‘‘Kinds’’ in Table 1, and 1,708 were isolated only once in
all of the 51-time screenings performed in the present work. The
number of times such redundant clones were isolated ranged
from two to 27.
Those 2,114 mAbs were individually screened using at least
three different fresh tumor tissues for each assay. They were
classified into four groups based on the immunostaining patterns
in the histological sections. When mAbs significantly stained
only the surface of tumor cells but negatively or very weakly
stained the other normal cells, they were classified to group A.
When the strong staining by mAbs was localized on the surface
of malignant cells but a part of the other normal tissue was also
stained, they were classified to group B. When mAbs showed
positive staining patterns both on malignant cells and on normal
cells nonspecifically, they were classified into group C. The
clones that did not give any positive signal were classified into
group D. Of 2,114 mAbs 281 were classified to group A and 384
were classified to group B.
Identification of 21 TAAs. Of the 665 clones, 300 that strongly
stained the malignant cells were chosen for further studies. Each
of the 300 clones was screened against six different tumor cell
lines by using flow cytometry (FCM). They were grouped
according to their staining pattern on the basis of the following
principle. In FCM analysis, the degree of peak shift should
reflect the amount of Ag, the accessibility of Ab, and the strength
of binding. The width and shape of peak should reflect the
degree of homogeneity of expressed Ags in the cell population.
Therefore, if the staining patterns against the six cell lines were
identical or very similar among the clones examined, they were
grouped together. It led to 40 groups made up of 150 clones. The
other 150 clones could not be grouped because of a weak signal
in the FCM.
The cell membrane proteins of carcinoma-derived cell lines
were biotinylated and then individually immunoprecipitated by
the mAbs of the same group and analyzed by SDS/PAGE. When
several mAbs in each group gave rise to the same band on the
gel the bands were cut out and subjected to MS analysis. This
enabled us to identify 21 distinct membrane Ags, which are listed
in Table 2. Those 21 Ags are recognized by 84 of the 300 mAbs
that we studied.
We also synthesized the extracellular portions of nine of the
21 Ags. Using ELISA we tested the 2,114 mAb clones against
those nine synthetic Ags. Of those, 272 clones gave a positive
reading in addition to the 84 clones that had been already
identified by the MS analysis. We are now in the process of
synthesizing the remaining 12 Ags for further screening. To date
356 clones of the 2,114 mAbs isolated in the present study were
revealed to specifically bind to one of the 21 TAAs. Of those 356
clones 156 belonged to the redundantly isolated 406 mAbs.
Expression of Fresh Cancer Tissues. Using representative clones that
specifically bound to 18 TAAs except for three TAAs PTK7,
CD9, and CDCP1, which were recently identified, the immuno-
staining analysis was performed against 24 fresh lung carcino-
mas. Table 3 summarizes the results of eight TAAs that gave
simple patterns showing one of the following two cases: ,
overexpression on malignant cells but no or very weak expression
on the other normal cells; and , no expression on either
malignant or normal cells. In the case of the other 10 TAAs,
although differential expression on malignant cells was distinct,
expression on a part of the normal tissue was also observed.
From these analyses we concluded the following. (i)TAAs
identified in our study were overexpressed in fresh tumors at
some frequency, but there is no TAA that was overexpressed in
all of the fresh tumors. (ii) All of the fresh malignant cells
analyzed to date overexpressed some of 18 TAAs in various
combinations. (iii) Approximately half of them showed distinct
tumor-specific expression pattern.
Antitumor Activity of Complete Human IgG mAbs. After preparing
IgG
1
mAbs we performed an in vitro assay for Ab-dependent
cell-mediated cytotoxicity (ADCC) using 22 clones against 10
Ags (EGFR, ALCAM, ICAM-1, EpCAM, HGFR, TfR, ITGA3,
EMMPRIN, PTP-LAR, and CD44) using the cell lines listed in
Table 4. As can be seen, those mAbs gave a positive reading that
ranged between 5% and 95% for cell killing. The details using
anti-EGFR Abs and anti-EpCAM Ab are in Fig. 1 a and e,
respectively. The degree of ADCC by clone 059-152 was strong
compared with that by cetuximab. Clone 067-153 showed ADCC
activity even at an extremely low concentration such as lower
than picomolar. We also performed an in vivo assay using three
Table 2. Cell-surface Ags identified by the mAbs
Ag MS* ELISA
Growth factor receptor
EGFR 3 6
HER2 1 15
HGFR 3 84
PTK7/CCK-4 1 ND
Transmembrane protein-tyrosine phosphatase
PTP-LAR 5 ND
Adhesion molecule
Ig superfamily
IGSF4 10 13
ALCAM 3 8
ICAM-1 5 17
Lu/BCAM 1 48
CEACAM6 1 ND
Non-Ig family
CD44 3 ND
EpCAM 2 ND
Tetraspanin
CD9 1 ND
Adenosine metabolism
Ecto-5-nucleotidase 1 ND
Complement inhibitor
MCP 8 81
Protease inducer
EMMPRIN 1 0
Iron metabolism
TfR 6 ND
Anoikis regulator
CDCP1 2 ND
Integrin family
3
114ND
v
38ND
6
45ND
*Number indicates that of different clones identified by MS analysis.
Clones identified by MS analysis are not included.
Kurosawa et al. PNAS
May 20, 2008
vol. 105
no. 20
7289
MEDICAL SCIENCES
mAbs against two of the Ags (EGFR and EpCAM) in cancer-
bearing athymic mice. As can be seen in Fig. 1d the anti-EGFR
Abs showed a strong antitumor activity against tumor cell line
A431. When we compared our mAbs (048-006 and 059-152)
against EGFR with cetuximab it appeared that they had a very
similar level of antitumor activity. The anti-EpCAM Ab also
prevented the growth of HT29 (Fig. 1f ).
The two anti-EGFR mAbs were used to analyze the mecha-
nism of their antitumor activities. As can be seen in Fig. 1b mAb
048-006 was very effective in inhibiting the binding of EGF to the
EGFR, whereas mAb 059-152 only partially prevented the
binding reaction. The phosphorylation assay (Fig. 1c) showed
that mAb 048-006 was effective in the inhibition of phosphory-
lation. The mAb 059-152 also gave inhibitory effects on phos-
phorylation, although less effective than 048-006. It suggests that
mechanisms of antitumor activity mediated by these two mAbs
might be different from each other.
Discussion
In the present study we used a phage Ab library that had been
constructed from human B cells (10). It has been suspected that
majority of the clones isolated from phage Ab libraries may not
show high affinity to the Ags because they should be naı¨vetothe
Ags (11). However, as shown in Fig. 1 two anti-EGFR Abs and
one anti-EpCAM Ab showed strong ADCC activity at the
concentration of 0.01–0.1
g/ml, which corresponds to 0.060.6
nM. This strength appeared to be practically strong enough to be
therapeutic agents whereas comparison of nucleotide sequences
of V
H
genes encoding theses three Abs with those of germ-line
genes indicated that mutations had not been introduced (data
not shown). In the screenings of Ab library there were many
cases where the same clones were redundantly isolated from
different screenings against the same carcinoma as well as
against different types of carcinoma. The reason why specific
clones were redundantly isolated might be as follows: (i) amounts
of Ags recognized by them were relatively abundant on the cells
used in the screenings, and (ii) the binding activity to the Ags was
stronger than that of other clones. This interpretation could be
at least partly correct. For example, while anti-EGFR Ab
048-006 was isolated by six time screenings the dissociation
constant (K
d
) of the Ag/Ab complex measured by the BIAcore
instrument was 0.025 nM (data not shown).
Table 3. Hisotological analysis of lung carcinomas with mAbs against TAAs
Clinical sample
ABC D E
12 3 4 5 6789101112131415161718192021222324
Antigen
Stage
IA IA IIIB IB IIIB IA IA IA IB IB IB IB IB IIB IIIA IIIA IIIB IB IIB IIIA IIIA IIIA IIIB IIIBClone Ab
ITGA6 029-023 ⫺⫺⫺⫺⫺
ITGAV 064-139 ⫺⫺⫺⫺⫺
CD147 059-053 ⫺⫺⫺⫺⫹
LAR 064-044 ⫺⫺⫺⫺⫹
IgSF4 076-048 ⫹⫺⫹⫺⫺
EGFR 048-006 ⫹⫹⫹⫺⫹
HER2 015-126 ⫺⫺⫺⫹⫹
HGFR 067-133 ⫺⫺⫺⫺
A, squamous cell carcinoma; B, adenosquamous carcinoma; C, bronchioloalveolar carcinoma; D, adenocarcinoma; E, large-cell carcinoma.
Table 4. ADCC activities of mAbs that have been converted to IgG
Ag Clone Target cells Cell killing,* %
EGFR 048-006 NCI-H1373, CCF-RC1, A431, ACHN 36–80
055-147 CCF-RC1, HT-29, A431 25–95
059-152 NCI-H1373, CCF-RC1, A431, ACHN 35–75
059-173 CCF-RC1, HT-29, A431 35–85
ALCAM 035-234 NCI-H1373, SKOv3, CW-2 8–19
041-118 NCI-H1373, EBC-1 14–18
066-174 NCI-H1373, SKOv3, CW-2 45–59
083-040 NCI-H1373 10
ICAM1 053-042 NCI-H1373 16
053-051 NCI-H1373, NCI-H441, HepG2 5–31
053-059 NCI-H1373, NCI-H441, HepG2 8–39
053-085 NCI-H1373, NCI-H441, HepG2 7–26
EpCAM 067-153 NCI-H1373, MKN45, HT-29, EBC-1 23–80
HGFR 067-133 NCI-H1373, MKN45, EBC-1 19–42
TfR 028-178 MIA Paca2 65
052-138 MIA Paca2 80
041-288 MIA Paca2 30
ITGA3 015-003 ACHN 20
EMMPRIN 059-053 CCF-RC1, ACHN 40
PTP-LAR 064-044 PC14 10
079-085 PC14 32
CD44 064-003 PC14 84
*Percentage of cell killing increased dose-dependently. When it reached plateau, the percentage for cell killing
was indicated.
7290
www.pnas.orgcgidoi10.1073pnas.0712202105 Kurosawa et al.
In this study we isolated 2,114 mAbs with unique sequences
that bound to molecules on the surface of tumor-derived cells
and selected 665 clones that gave tumor-specific immunostaining
patterns. To identify TAAs recognized by the tumor-specific
mAbs we developed two strategies. The grouping of mAbs by
FCM enabled us to achieve efficient identification of TAAs by
the following reasons. (i) FCM analyses against several cell lines
taught us which cell expressed most abundantly the target
molecules. (ii) Because multiple mAbs in each group turned out
to bind to the same Ag in most cases, many clones have been
treated very efficiently by a limited number of experiments. (iii)
Detergent for solubilization of membrane proteins may destroy
the structural integrity with the results of losing the antigenic
structure. When several clones classified into the same group
were analyzed together, some of them could bind to a relatively
detergent-resistant epitope. Screenings of all of the 2,114 mAbs
by ELISA with the polypeptides that correspond to the extra-
cellular portions of the TAAs already identified by MS analysis
also enabled us to efficiently identify the Ags recognized by
respective Abs. Now it is likely that we have already revealed
more than half of TAAs potentially identified according to this
procedure.
We characterized two anti-EGFR mAbs, clone 048-006 and
clone 059-152. Whereas clone 048-006 inhibited both the binding
of EGF to EGFR and the phosphorylation of EGFR, clone
059-152 partly inhibited the binding of EGF to EGFR and gave
inhibitory effects on the phosphorylation of EGFR less effec-
tively. However, both clones showed a strong antitumor activity
in cancer-bearing athymic mice. The characteristics of clone
048-006 appeared to be similar to that of cetuximab (12).
However, to the best of our knowledge there has been no report
describing mAb whose characteristics was similar to that of
059-152. In the present study multiple Abs have been isolated
against respective TAAs. It is possible that these mAbs may have
various characteristics as shown by anti-EGFR mAbs.
As indicated in Table 3, half of the TAAs identified in this
study gave a tumor-specific staining pattern that is overexpres-
sion on malignant cells but no or very weak expression on the
other normal cells in the histological sections. These TAAs could
be candidates to be useful targets for therapeutic Abs. Further-
more, as indicated in Table 4 and Fig. 1, IgG form of mAbs that
bound to them showed strong antitumor activity in vitro and in
vivo. Therefore, we believe that some Ags detected will be useful
targets for cancer therapy and that several mAbs will become
useful therapeutic agents in the foreseeable future.
Materials and Methods
Ab Library and Screening. AIMS-5 library constructed by using a phage-display
system was employed (10). Screenings were performed by a method similar to
the one developed by Giordano et al. (13). In brief, the phages (2 10
13
cfu)
were mixed with cells (0.2–1 10
8
) in 1.6 ml of solution A (1% BSA, MEM, and
0.1% NaN
3
), and Ag–Ab complexes on the cell surface were formed. The cell
and phage suspension was overlaid on the organic solution in an Eppendorf
tube. After the tube was centrifuged, water and organic layers were dis-
carded. The collected cells were suspended in solution A. This process was
repeated three times. Finally, the cells were suspended in PBS and frozen in
liquid nitrogen. The frozen cells were thawed and mixed with Escherichia coli
a
c
e
b
d
f
Fig. 1. Antitumor activities of two anti-EGFR mAbs (clone 048-006 and clone 059-152) and an anti-EpCAM mAb (clone 067-153). (a) ADCC. Target cells: NCI-1373.
Ab: HR1-007 (negative control), mAb 048-006, 059-152, and cetuximab (positive control). (b) Inhibitory effects of mAbs on the binding of EGF to EGFR on cell
line A431. Ab: HR1-007 (negative control), mAb 048-006, and 059-152. (c) Inhibitory effects of mAbs on phosphorylation of EGFR induced by EGF. Upper bands:
Western blot by anti-phosphotyrosine mouse mAb. Lanes 1– 6: incubated for 30 min after addition of EGF at 1
g/ml; lane 1, without Ab; lanes 2– 6, incubated
with Ab for 30 min, then EGF was added; lane 2, HR1-007 (negative control) at 10
g/ml; lanes 3 and 4, 048-006; lanes 5 and 6, 059-152; lanes 3 and 5, 1
g/ml;
lanes 4 and 6, 10
g/ml. Lower bands: control Western blot with rabbit antiserum against
-actin. (d) Inhibitory effects of mAbs on the growth of tumor cell line
A431 in athymic nude mice assayed by the first method described in Antitumor Activity in Vivo. Ab: HR1-007 (negative control), mAb 048-006, 059-152, and
cetuximab. (e) ADCC. Target cells: HT-29. Ab: HR1-007 (negative control) and mAb 067-153. (f) Inhibitory effects of mAb on the growth of a tumor cell line in
athymic nude mice assayed by the alternative method described in Materials and Methods. Ab: HR1-007 (negative control) and mAb 067-153.
Kurosawa et al. PNAS
May 20, 2008
vol. 105
no. 20
7291
MEDICAL SCIENCES
DH12S. The phages were prepared. This screening round was performed
repeatedly three times. After three rounds of screenings, E. coli DH12S in-
fected with recovered phages was spread on plates. Approximately 200
colonies were picked up. Thirty-three cancer cell lines listed in Table 1 were
used as Ags.
Immunostaining of Fresh Tumors. Tumor tissues and the neighboring normal
tissues resected by operation were used for immunostaining. They were fixed
with 4% paraformaldehyde in 0.1 M cacodylic buffer (pH 7.4) by microwave
irradiation as described previously (14).
Identification of Ag. Membrane protein analysis was performed according to
Zhao et al. (15). Proteins present on the cell surface were biotinylated accord-
ing to the manufacturer’s instruction by using the EZ-Link Sulfo-NH-LC Bioti-
nylation kit (Pierce). After the cells were homogenized with a Dounce ho-
mogenizer, the protein–membrane complexes were banded between 0.25 M
and 1.25 M sucrose layers by centrifugation. The complexes were dissolved in
a detergent mixture: 50 mM Hepes (pH 7.6), 150 mM NaCl, 5 mM EDTA, 1%
Triton X-100, and 1%
octyl glucoside. scFv-C
L
fused with cp3 was converted
to scFv-C
L
fused with protein A domains (scFv-C
L
-PP) (16). scFv-PP form was
covalently bound to beads that were CNBr-activated Sepharose 4B (GE Health
Care Bioscience). Ab-bound beads were used for immunoprecipitation as
described by David et al. (17). MS analysis was performed according to Geuijen
et al. (8).
Preparation of IgG
1
. ScFv was converted to IgG
1
and prepared by using a
high-level expression vector (18). Using IgG
1
mAbs we examined ADCC, effects
on binding of EGF to EGFR, effects on phosphorylation of EGFR, and antitumor
activity in athymic nude mice.
ADCC. The enzymatic activity of lactic dehydrogenase released from the target
cells was measured for estimation of ADCC (19). Various cell lines were used as
targets for the mAbs. Cells were derived from the following cancers: NCI-
H1373, lung adenocarcinoma; CCF-RC1, renal clear cell carcinoma; A431, vulva
epidermoid carcinoma; ACHN, renal adenocarcinoma; HT-29, colorectal ade-
nocarcinoma; SKOv3, ovarian adenocarcinoma; CW-2, colorectal adenocarci-
noma; EBC-1, lung squamous cell carcinoma; NCI-H441, lung papillary adeno-
carcinoma; HepG2, hepatocellular carcinoma; MKN45, gastric
adenocarcinoma; MIAPaca-2, pancreatic carcinoma; PC14, ling carcinoma.
Effector cells were prepared from blood of healthy volunteers and used in a
ratio of 100:1 (10
6
to 10
4
in 200
l) (20).
Effects of Anti-EGFR mAbs on the Function of EGFR. Binding of EGF to EGFR on
the cell surface was estimated according to Yang et al. (21). Phosphorylation
of EGFR induced by EGF was measured according to Matar et al. (22).
Antitumor Activity
in Vivo
. Two different methods were adopted. In the first
method each mouse was injected with 5 10
6
cells, and when the tumor grew
to 0.2 cm
3
mAb therapy was initiated. Treatments consisted of twice-weekly
i.p. injections of mAb for 3 weeks. One milligram in 0.5 ml of PBS was used in
each injection. Control animals received injection of PBS. Six mice were used
for each treatment. The alternative method: 1 day after injection of cells mAb
therapy was started. Treatments consisted of twice-weekly i.v. injections of
mAb for 2 weeks. A total of 50
g was used in each injection.
ACKNOWLEDGMENTS. This work was supported in part by a grant-in-aid for
the 21st Century Center of ExcellenceProgram of Fujita Health University from
the Ministry of Education, Culture, Sports, Science, and Technology and by a
grant from the New Energy and Industrial Technology Development Organi-
zation (to Y.K.).
1. Waldmann H, et al. (1984) Elimination of graft-versus-host disease by in-vitro depletion
of alloreactive lymphocytes with a monoclonal rat anti-human lymphocyte antibody
(CAMPATH-1). Lancet ii:483– 486.
2. Carter P, et al. (1992) Humanization of an anti-p185HER2 antibody for human cancer
therapy. Proc Natl Acad Sci USA 89:4285–4289.
3. Reff ME, et al. (1994) Depletion of B cells in vivo by a chimeric mouse human
monoclonal antibody to CD20. Blood 83:435–445.
4. Adams GP, Weiner LM (2005) Monoclonal antibody therapy of cancer. Nat Biotechnol
23:1147–1157.
5. Okabe H, et al. (2001) Genome-wide analysis of gene expression in human hepatocel-
lular carcinomas using cDNA microarray: Identification of genes involved in viral
carcinogenesis and tumor expression. Cancer Res. 61:2129 –2137.
6. Hippo Y, et al. (2002) Global gene expression analysis of gastric cancer by oligonucle-
otide microarrays. Cancer Res 62:233–240.
7. Winter G, Griffiths AD, Hawkins RE, Hoogenboom HR (1994) Making antibodies by
phage display technology. Annu Rev Immunol 12:433– 455.
8. Geuijen CAW, et al. (2005) A proteomic approach to tumour target identification using
phage display, affinity purification and mass spectrometry. Eur J Cancer 41:178 –187.
9. Goenaga AL, et al. (2007) Identification and characterization of tumor antigens by
using antibody phage display and intrabody strategies. Mol Immunol 44:3777–3788.
10. Morino K, et al. (2001) Antibody fusions with fluorescent proteins: A versatile reagent
for a profiling protein expression. J Immunol Methods 257:175–184.
11. Gherardi E, Milstein C (1992) Original and artificial antibodies. Nature 357:201–202.
12. Goldstein NI, Prewett M, Zuklys K, Rockwell P, Mendelsohn J (1995) Biological efficacy
of a chimeric antibody to the epidermal growth factor in a human tumor xenograft
model. Clin Cancer Res 1:1311–1318.
13. Giordano RJ, Cardo-Vila M, Lahdenranta J, Pasqualini R, Arap W (2001) Biopanning and
rapid analysis of selective interactive ligands. Nat Med 7:1249 –1253.
14. Mizuhira V, Hasegawa H (1997) Microwave fixation and localization of calcium in
synaptic terminals using X-ray microanalysis (EDX) and electron energy loss spectros-
copy (EELS) imaging methods. Brain Res Bull 43:53–58.
15. Zhao Y, Zhang W, Kho Y, Zhao Y (2004) Proteomic analysis of integral plasma mem-
brane proteins. Anal Chem 76:1817–1823.
16. Ito W, Kurosawa Y (1993) Development of an artificial antibody system with multiple
valency using an Fv fragment fused with a fragment of protein A. J Biol Chem
268:20668–20675.
17. David GS, Chino TH, Reisfeld RA (1974) Binding of proteins to CNBr-activated sepharose
4B. FEBS Lett 43:264 –266.
18. Bebbington CR, et al. (1992) High-level expression of a recombinant antibody from
myeloma cells using a glutamine synthetase gene as an amplifiable selection marker.
Bio/Technology 10:169 –175.
19. Decker T, Lohmann-Matthes ML (1988) A quick and simple method for the quantitation
of lactose dehydrogenase release in measurements of cellular cytotoxicity and tumor
necrosis factor (TNF) activity. J Immunol Methods 115:61– 69.
20. Boeyum A (1964) Separation of white blood cells. Nature 204:793–794.
21. Yang XD, et al. (1999) Eradication of established tumors by a fully human monoclonal
antibody to the epidermal growth factor receptor without concomitant chemother-
apy. Cancer Res 59:1236 –1243.
22. Matar P, et al. (2004) Combined epidermal growth factor receptor targeting with the
tyrosine kinase inhibitor gefitinib (ZD1839) and the monoclonal antibody cetuximab
(IMC-C225): Superiority over single-agent receptor targeting. Clin Cancer Res 10:6487–
6501.
7292
www.pnas.orgcgidoi10.1073pnas.0712202105 Kurosawa et al.
    • "To overcome the limitations of the pure antigen availability, we used phage display and a cell-based antibody selection (CBAS) approach for generation of fully human antibodies against CCR4. The cell-based panning strategies have been successfully used previously for generating antibodies against bulky membrane antigens, such as EGFR, HER2, ALCAM, EpCAM [51] or c-Met [52]. In general, cell-based screening is often challenging due to the much greater antigen complexity, lower antigen concentration and antigen accessibility. "
    [Show abstract] [Hide abstract] ABSTRACT: Background CC chemokine receptor 4 (CCR4) represents a potentially important target for cancer immunotherapy due to its expression on tumor infiltrating immune cells including regulatory T cells (Tregs) and on tumor cells in several cancer types and its role in metastasis. Methodology Using phage display, human antibody library, affinity maturation and a cell-based antibody selection strategy, the antibody variants against human CCR4 were generated. These antibodies effectively competed with ligand binding, were able to block ligand-induced signaling and cell migration, and demonstrated efficient killing of CCR4-positive tumor cells via ADCC and phagocytosis. In a mouse model of human T-cell lymphoma, significant survival benefit was demonstrated for animals treated with the newly selected anti-CCR4 antibodies. Significance For the first time, successful generation of anti- G-protein coupled chemokine receptor (GPCR) antibodies using human non-immune library and phage display on GPCR-expressing cells was demonstrated. The generated anti-CCR4 antibodies possess a dual mode of action (inhibition of ligand-induced signaling and antibody-directed tumor cell killing). The data demonstrate that the anti-tumor activity in vivo is mediated, at least in part, through Fc-receptor dependent effector mechanisms, such as ADCC and phagocytosis. Anti-CC chemokine receptor 4 antibodies inhibiting receptor signaling have potential as immunomodulatory antibodies for cancer.
    Full-text · Article · Jul 2014
    • "Recently, we isolated a novel fully human monoclonal IgG1 antibody designated as 059-053 against CD147 from a large-scale human antibody library constructed using a phage-display system that incorporated a highly efficient screening method termed isolation of antigen–antibody complexes through organic solvent, with living pancreatic cancer cells [8]. This antibody induces antibody-dependent cell-mediated cytotoxicity (ADCC) and inhibits cell proliferation of pancreatic cancer cells [8], [9]. In the present study, we radiolabeled 059-053, and evaluated the in vitro and in vivo properties as a new positron emission tomography (PET) probe for imaging CD147-expressing tumors in a pancreatic cancer model. "
    [Show abstract] [Hide abstract] ABSTRACT: Introduction: Pancreatic cancer is an aggressive cancer and its prognosis remains poor. Therefore, additional effective therapy is required to augment and/or complement current therapy. CD147, high expression in pancreatic cancer, is involved in the metastatic process and is considered a good candidate for targeted therapy. CD147-specfic imaging could be useful for selection of appropriate patients. Therefore, we evaluated the potential of a fully human anti-CD147 monoclonal antibody 059-053 as a new positron emission tomography (PET) probe for pancreatic cancer. Methods: CD147 expression was evaluated in four pancreatic cancer cell lines (MIA Paca-2, PANC-1, BxPC-3, and AsPC-1) and a mouse cell line A4 as a negative control. Cell binding, competitive inhibition and internalization assays were conducted with (125)I-, (67)Ga-, or (89)Zr-labeled 059-053. In vivo biodistribution of (125)I- or (89)Zr-labeled 059-053 was conducted in mice bearing MIA Paca-2 and A4 tumors. PET imaging with [(89)Zr]059-053 was conducted in subcutaneous and orthotopic tumor mouse models. Results: Among four pancreatic cancer cell lines, MIA Paca-2 cells showed the highest expression of CD147, while A4 cells had no expression. Immunohistochemical staining showed that MIA Paca-2 xenografts also highly expressed CD147 in vivo. Radiolabeled 059-053 specifically bound to MIA Paca-2 cells with high affinity, but not to A4. [(89)Zr]059-053 uptake in MIA Paca-2 tumors increased with time from 11.0±1.3% injected dose per gram (ID/g) at day 1 to 16.9±3.2% ID/g at day 6, while [(125)I]059-053 uptake was relatively low and decreased with time, suggesting that 059-053 was internalized into tumor cells in vivo and (125)I was released from the cells. PET with [(89)Zr]059-053 clearly visualized subcutaneous and orthotopic tumors. Conclusion: [(89)Zr]059-053 is a promising PET probe for imaging CD147 expression in pancreatic cancer and has the potential to select appropriate patients with CD147-expressing tumors who could gain benefit from anti-CD147 therapy.
    Full-text · Article · Apr 2013
    • "The use of antibody technology for phenotypic target discovery has been dominated by the use of hybridoma-based techniques. However, improved antibody isolation and target identification techniques combined with the incorporation of high-throughput functional screens have led to increased success using phage displayderived antibodies [7]. The phenotypic antibody screening approach for target discovery has the advantage that the isolated antibodies can also be used for validation activities and in some instances can even be pursued as therapeutic candidates. "
    [Show abstract] [Hide abstract] ABSTRACT: The continued discovery of therapeutic antibodies, which address unmet medical needs, requires the continued discovery of tractable antibody targets. Multiple protein-level target discovery approaches are available and these can be used in combination to extensively survey relevant cell membranomes. In this study, the MDA-MB-231 cell line was selected for membranome survey as it is a 'triple negative' breast cancer cell line, which represents a cancer subtype that is aggressive and has few treatment options. The MDA-MB-231 breast carcinoma cell line was used to explore three membranome target discovery approaches, which were used in parallel to cross-validate the significance of identified antigens. A proteomic approach, which used membrane protein enrichment followed by protein identification by mass spectrometry, was used alongside two phenotypic antibody screening approaches. The first phenotypic screening approach was based on hybridoma technology and the second was based on phage display technology. Antibodies isolated by the phenotypic approaches were tested for cell specificity as well as internalisation and the targets identified were compared to each other as well as those identified by the proteomic approach. An anti-CD73 antibody derived from the phage display-based phenotypic approach was tested for binding to other 'triple negative' breast cancer cell lines and tested for tumour growth inhibitory activity in a MDA-MB-231 xenograft model. All of the approaches identified multiple cell surface markers, including integrins, CD44, EGFR, CD71, galectin-3, CD73 and BCAM, some of which had been previously confirmed as being tractable to antibody therapy. In total, 40 cell surface markers were identified for further study. In addition to cell surface marker identification, the phenotypic antibody screening approaches provided reagent antibodies for target validation studies. This is illustrated using the anti-CD73 antibody, which bound other 'triple negative' breast cancer cell lines and produced significant tumour growth inhibitory activity in a MDA-MB-231 xenograft model. This study has demonstrated that multiple methods are required to successfully analyse the membranome of a desired cell type. It has also successfully demonstrated that phenotypic antibody screening provides a mechanism for rapidly discovering and evaluating antibody tractable targets, which can significantly accelerate the therapeutic discovery process.
    Full-text · Article · Feb 2013
Show more