[CANCER RESEARCH 47, 999-1005, February 15, 1987)
Recombinant Human-Mouse Chimeric Monoclonal Antibody Specific for Common
Acute Lymphocytic Leukemia Antigen1
Yushi INishimura, Masatoshi
Yokoyama, Kazuo Araki, Ryuzo Leda, Akira Kudo, and Takeshi Watanabe2
Department of Molecular Immunology, Medical Institute of Bioregulation, Kyushu University, Fukuoka [Y. N., M. Y., K. A., A. K., T. W.J and Laboratory of
Chemotherapy, Aichi Cancer Research Institute, Nagoya [R. U.J, Japan
ABSTRACT
A human-mouse chimeric antibody constructed in the present study
was specific for a human tumor-associated antigen, common acute lymphocytic leukemia antigen. The antibody consisted of human heavy and
light chain constant domains (7, and «type) and mouse heavy and light
chain variable domains, which were derived from human plasma cell
leukemia line (ARI 177) and mouse hybridoma cells (NL-1) specific for
common acute lymphocytic leukemia antigen, respectively. The artifi
cially fused immunoglobulin molecules were producedin mouse myeloma
cells, X63Ag8.6S3 which were transformed with the chimeric heavy and
light chain genes formed by joining the corresponding gene segments in
vitro at the J-C introns. The human heavy chain enhancer element was
Ugated to the chimeric heavy and light chain genes, and this enhancer
appeared to be obligatory for the efficient production of the chimeric
antibody molecules. The stably transformed cells secreted the chimeric
antibody, which specifically bounda common acute lymphocytic leukemia
antigen expressing cell line. The amount of the chimeric antibody pro
duced (10-30 Mg/mlin the serum-free medium) was comparable to that
made by murine hybridoma line, NL-1. The molecular weight of the
chimeric heavy chain molecules was reduced from 54,000 to 50,000 upon
treatment with tunicamycin, suggesting that the peptide was normally
glycosylated in the transformants. The chimeric antibody exhibited com
plement-dependent cytotoxicity, in which glycosylation is thought to be
indispensable. The antibody also mediated antibody-dependent cell-me
diated cytotoxicity to the human target cells. The antibody-dependent
cell-mediated cytotoxicity activity of the chimeric antibody was twice that
of the murine NL-1 monoclonal antibody when human peripheral blood
mononuclear cells were used as effectors.
INTRODUCTION
Monoclonal antibodies specific for tumor-associated antigens
are indispensable in cancer immunotherapy and diagnosis. Hu
man immunoglobulin is presumably superior to that of other
species when administered to humans because it may function
better with the recipient's effector cells and be less immuno-
ulin molecules are probably an important target of antigenic
recognition by the human immune system (1). This portion of
the molecule is responsible for many physiological functions of
immunoglobulins with the exception of specific antigen recog
nition, which is attributable to the variable domain. If the
constant domains of murine antibodies could be replaced by
the human counterparts, the resultant chimeric human-mouse
molecules would be expected to retain the original specificity
but have a much lower antigenicity to humans. Advances in
molecular biology have elucidated the structure and function of
immunoglobulin genes (2). Technology is now available to
prepare artificially fused proteins by gene manipulation. Con
struction of chimeric human-mouse antibodies were first re
ported by Boulianne et al. (3) and Morrison et al. (4) utilizing
hapten specific monoclonal antibodies. They showed assembly
of the chimeric antibodies, and in the former case, antigen
binding activity was retained.
cALLA3 is a well-characterized antigen of non-T, non-B acute
lymphocytic leukemia. NL-1, a mouse hybridoma line, produces
antibody (y^, «)specific for cALLA (5). The structure of the
variable region gene segment of the heavy chain of the antibody
was clarified in our laboratory and found to be highly homolo
gous to the MOPC-21 immunoglobulin heavy chain (6, 7). We
now show the structure of the NL-1 light chain gene and
describe the construction of chimeric heavy and light chain
immunoglobulin genes utilizing a human heavy chain enhancer.
Chimeric antibody molecules were successfully made in mouse
myeloma cells which bound to a cALLA-expressing cell line
and killed the cells by complement-dependent cytotoxicity as
well as ADCC. Our approach to preparing human-mouse chi
meric antibodies to tumor-associated antigens should be advan
tageous in obtaining reagents suitable for clinical use.
MATERIALS AND METHODS
genic. However, few human monoclonal antibodies are available
because of ethical problems in in vivo immunization and diffi
culties in in vitro immunization required for production of
human-human hybridomas. Recent advances in murine hybrid
oma technology have resulted in many promising reagents and
some are already being used in clinical trials. Since multiple
injections of the antibodies are necessary in immunotherapy
protocols, reduction in antigenicity is an important concern.
Administration of heterologous monoclonal antibodies may
lead to allergic reactions and neutralization of the administrated
antibodies.
Immunoglobulin molecules consist of variable and constant
region domains. The constant domains of mouse immunoglobReceived 7/2/86; revised 10/20/86; accepted 11/18/86.
The costs of publication of this article were defrayed in part by the payment
of page charges. This article must therefore be hereby marked advertisement in
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by a grant-in-aid for cancer research from the
Ministry of Education, Science, and Culture of Japan.
2 To whom requests for reprints should be addressed, at Medical Institute of
Bioregulation, Kyushu University, Maidashi 3-1-1, Higashi-ku, Fukuoka 812,
Japan.
Cell Line. A murine hybridoma, NL-1, which secreted monoclonal
antibody (72««)specific for cALLA (5), and Manca (SK-DHL-2,
human B-lymphoblastoid cell line) (5, 8), K562 (human chronic myelogenous leukemia, blast cells), and CCRF-HSB-2 (human T-acute
lymphocytic leukemia) were cultured in RPMI1640 medium containing
10% fetal calf serum. A murine myeloma cell line, X63Ag8.6S3 (9),
was obtained from Dr. G. Köhler.
Screening of Immunoglobulin Genes. Hindlll digests of NL-1 DNA
were used to contract a DNA library for the variable region of the light
chain gene by insertion into the Charon 28 phage vector. The gene was
screened using mouse J.i •,
(provided by Dr. T. Honjo, Kyoto Univer
sity) as a hybridization probe. The constant region gene of the human
ic light chain was screened from a library constructed by insertion of
EcoRI digests of ARH77 DNA in XgtWESXB vector. Mouse C, gene
(provided by Dr. T. Honjo) was used as a cross-hybridization probe.
The human immunoglobulin yt chain was obtained as a 21-kb EcoRI
fragment from a human plasma cell leukemia line, ARH77, which had
3The abbreviations used are: cALLA, common acute lymphocytic leukemia
antigen; ADCC, antibody-dependent cell-mediated cytotoxicity; FITC, fluorescein isothiocyanate; gpt, gene encoding xanthine-guanine phosphoribosyl trans
it-rase: kb, kilobase(s); neo, gene conferring resistance to antibiotic G418; SDS,
sodium dodecyl sulfate.
999
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1987 American Association for Cancer Research.
RECOMBINANT
CHIMERIC
MONOCLONAL
been reported previously (6). The mouse heavy chain variable region
gene of NL-1 hybridoma was cloned previously (7).
Transformationof Mouse Myeloma Cells. The chimeric heavy chain
genes were inserted into pSV2gpt plasmili vector (10), and the chimeric
light chain genes were inserted into pSV2neo vector (11). Transforma
tion was carried out either by protoplast fusion (12) or by a DEAEdextran method (13). Stable transformants were screened by resistance
to mycophenolic acid (Lilly Co., Ltd.) for gpt and G418 (Gibco Labo
ratories) for neo.
Northern Blot Analysis. Total RNA was extracted from the
X68Ag8.6S3 transformants and subjected to Northern blot analysis
after electrophoresis on a denatured gel (14).
DNA Sequencing. DNA sequences were determined by a chain ter
mination method after cloning appropriate DNA fragments on
M13mpl8 and M13mpl9 phage vectors (Pharmacia) (IS).
SDS-Polyacrylamide Gel Electrophoresis. The transformants were
cultured for 8 h in methionine-free RPMI 1640 medium supplemented
with 10% fetal calf serum in the presence of ["SJmethionine (Amer-
a)
EcoRV
BamHI
TO cALLA
VK-NU
BamHI Pvull
\/u5'Pstl\
/n
«Ubai
\/\
J3n/
i1 I1
IJ4
VJiHindi J5rBglllJpstl
/1
ÉÉL
4.2kb
1kb
sham). The secreted immunoglobulins were precipitated from culture
fluid by goat anti-human y antibody or goat anti-human x antibody
which bound to Protein A-Sepharose beads followed by electrophoresis
on SDS-polyacrylamide gel with and without reduction by 2-mercaptoethanol.
Assay of Antigen-binding Activity. The X63Ag8.6S3 transformants
were cultured in RPMI 1640 medium supplemented with 10% fetal
calf serum. Human tumor cell lines or peripheral blood lymphocytes
were incubated with the culture supernatant of the transformants and
then stained with FITC-labeled anti-human y antibody (Behringerwerke
AG). Binding of the chimeric antibody to the cells was examined by
flow cytometry using EPICS V (Coulter).
Assay of Complement Dependent Cytotoxicity. Manca cells (1x1(0
EcoRI
ANTIBODY
b)
TGACTTTATAAGCCGAGAACTCCAAAGACTACTJATTTGCAjfcGTTCATCCTCAGAAACCA
1
20
To
60
CAAATTTCTCACAGTTGTTTTAAAGAGATGCACTTATAGGAAGAGCAATAATTAGTCAGA
80
100
120
MetAspPReHi
sVa1Gln
11ePReSerPReMetLeu11eSe
GACCAGGATCAACAACACAATGGATTTTCATGTGCAGATTTTCAGCTTCATGCTAATCAG
140
160 i
180
rValThrV
'
'
L.
TGTCACAGGTAGAGATGGGTAGGAGTTTGAGTTCAGAAAACAAGGTTCTTTTTCCCAGGA
'
200
220
240
AAGTTTCATAAAGTAGACTTTTTCTTTCCTCTGAATATTGAATGCTATGAAATTATTTTT
260
280
300
GTATCTCTGTCTACTAACACCCTCTATCTCTTTCACTCTCTCTTCTGTTGTTTTTCCATA
EcoRI
320
340
360
allleLeuSerSerGlyGluIleValLeuThrGlnSerProAlaLeuMetAlaAlaSer
GTCATATTGTCCAGTGGAGAAATTGTGCTCACCCAGTCTCCAGCACTCATGGCTGCTTCT
,
i •
380
400
420
ProGlyGlulysValThrlleArgCysSerValSerSerSerlleSerSerSerAsnleu
CCAGGGGAGAAGGTCACCATCCGCTGCAGTGTCAGCTCAAGTATAAGTTCCAGCAACTTG
440
.,
460
480
(21kb)
BomHI
HisTrpTyrGlnGlnLysSerGluThrSerProLysProTrpIleTyrGlyThrSerAsn
CACTGGTACCAGCAGAAGTCAGAAACCTCCCCCAAACCCTGGATTTATGGCACATCCAAC
500
520
540
Mlul
LeuAlaSerGlyValProValArgPheSerGlySerGlySerGlylhrSerTyrPheLeu
CTGGCTTCTGGAGTCCCTGTTCGTTTCAGTGGCAGTGGATCTGGGACCTCTTATTTTCTC
560
580
600
Mlul digestion
BamHI digestion
ThrlleSerSerMetGluAlaGluAspAlaAlaThrTyrTyrCysGlnGlnTrpSerSer
ACAATCAGCAGCATGGAGGCTGAAGATGCTGCCACTTATTACTGTCAACAGTGGAGTAGT
620
640
660
BamHI
TyrProLeuThrPheGlyGlyGlyThrLysLeuGluIleLys
TACCCACTCACGTTCGGAGGGGGGACCAAGCTGGAAATAAAACGTAAGTAGTCTTCTCAA
680
700 ,
720
II
J2
'
Fig. 2. Structure and nucleotide sequences of mouse a chain variable region
gene (V.) derived from mouse hybridoma, NL-1, specific for cALLA. a, restriction
enzyme map of V.N, , gene; />. nucleotide sequences and deduced amino acid
sequences of the variable region of NL-I * chain gene. The consensus octurner
sequence of immunoglobulin gene is boxed.
were labeled with 100 itCi of "Cr eliminate (Japan Atomic Research
Institute) at 37'C for l h in RPMI 1640 complete medium. After
Fig. I. Construction of the chimeric immunoglobulin heavy chain gene. Boxes,
exon regions of human and mouse immunoglobulin heavy chain genes (6, 7);
thick lines, intron DNA; thin lines, vector DNA. ©,position of the human heavy
chain enhancer element. pSV2gpt was used as a vector for transforming mouse
cells.
removing free enrómateby washing, the cells were incubated with the
chimeric antibody and rabbit complements for l h at 37*C. The cytotoxic activity was determined by 51Crrelease to the supernatant. Spon
taneous release was in the range of 6-13% and was subtracted from the
measured values (16).
Antibody-dependentCell-mediated Cytotoxicity. Manca cells (2x1 ()'')
were labeled with s'Cr enrómate and used as target cells. Human
peripheral blood mononuclear cells were obtained from healthy volun-
1000
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1987 American Association for Cancer Research.
RECOMBINANT
mouse
yj
BamHI
5'
~
CHIMERIC
MONOCLONAL
Hindin
| xgCOR|
ANTIBODY
TO CALLA
a)
b)
ÃŽ3'
160k54k
28kBamHI
-
H2L.2
H
•
i
BamHI
-€H
0.9kb
human enhancer
Chimeric light chain gene
(mouse VNL1-human CK)
Fig. 3. Construction of the chimeric immunoglobulin light chain gene, fiymbols are the same as in the legend for Fig. 1. pSV2.net> was used as a plasmid
vector instead of pSV2gpt.
ontlhuman
K
(b)
antihuman
1
K
t
54K
50K
kb
3.5
1.8
1234
<mtihuman
Fig. S. SDS-polyacrylamide gel electrophoresis analysis of translated products
of transformed X63Ag8.6S3 cells. One ml of S x 10' cells of the transformants
was cultured in the presence of 100 ui'i [3SS]methionine for 8 h, and ¡nummo
globulins produced were precipitated with goat anti-human 7 antibody or goat
anti-human * antibody which bound to protein A-Sepharose beads (100 ¡ill.The
precipitates were applied to SDS-polyacrylamide gel electrophoresis with (a) or
without (b) reduction by 2-mercaptoethanol.
(A)
(a)
ontlhuman
1234
(B)
CO
(b)
(a)
- 1.3kb
.
1
2
456
7
8
Fig. 4. Northern blot analysis of the transcripts of X63Ag8.6S3 cells which
were transformed by chimeric heavy and light chain genes. RNA was extracted
and 10 Mgof total RNA were electrophoresed on denatured gels and transferred
to a nitrocellulose filter followed by hybridization with appropriate nick-translated
probes. A, RNA blot analysis of chimeric heavy chain gene expression. The probes
used were (a) mouse VH gene (Pvull-Bglll) fragment and (b) human >, heavy
chain gene (Pstl-Pstl fragment). Lane J (a) NL-1 and (/»)
ARH77; lane 2 (a and
b) X63Ag8.653; lanes 3 and 4 (a and A), transformants. B, RNA blot analysis of
chimeric light chain gene expression, a, mouse V.M , gene (Pstl-Hincll fragment);
b, human C. gene (Hpal-Hpal fragment); c. mouse C, gene used as probes. Lanes
7, 5, and S, chimeric gene transformants; lanes 2 and 4, X63Ag8.6S3; lanes 5 and
7, NL1 cells; lane 6, ARH77. One /ig of polyadenylated RNA was used for
ARH77.
Fig. 6. Effects of tunicamycin treatment of the X63Agg.6S3 transformants on
the produced immunoglobulin. The transformants were cultured in the presence
(lane 2) or absence (lane I) of tunicamycin for 16 h in the medium supplemented
with [35S]methionine. Immunoglobulins were precipitated with goat anti-human
7 antibody as described in Fig. 5. K, thousands.
leers and separated by using lymphocyte separation medium (Bionetics
Laboratory Products). Target cells (2 x IO4; 50 n\) were mixed with
100 M!of effector cells at ratios of 1:50, 1:20, and 1:5 together with 50
Ailof the chimeric or NL-1 antibody solution (1 /in/ml). Incubation was
1001
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1987 American Association for Cancer Research.
RECOMBINANT
CHIMERIC
MONOCLONAL
ANTIBODY
TO cALLA
(A)
(O
i_
<u
.0
E
(D)
(E)
a,b
I
i\
I \
Fluorescence Intensity
Fig. 7. Binding of the chimeric antibody to Manca cells, expressing cALLA. The cells were incubated with supernatants »fNI. l or the chimera- antibody at 4*C
for 30 min, washed, and stained by MR '-labeled goat anti-human y antibody (B-E) or FITC-labeled anti-mouse •>
antibody (A). Stained cells were analyzed by flow
cytometry with EPICS V. A and B Manca cells; C, K562 (human chronic myelogenous leukemia, blast cells); D, CCRF-HSB2 (human T-acute lymphocytic leukemia);
/;'. normal human peripheral blood lymphocyte, a, stained chimeric antibody or NL-1 antibody and FITC-conjugated anti-human -t or anti-mouse y antibody,
respectively; b, stained with medium and FITC-conjugated antibodies. The small positive peak found in normal peripheral blood lymphocytes was due to the presence
of surface IgG* lymphocytes.
done at 37°Cfor 6 h under a 5% CO2 atmosphere. Released "Cr was
measured to determine the cytotoxic activity (17).
RESULTS
Construction of the Chimeric Heavy Chain Gene. The rear
ranged variable region (V-D-J) gene of mouse heavy chain was
cloned from NL-1 hybrido ma cells which secreted monoclonal
antibody specific for cALL antigen, and the constant region
gene of human 71 heavy chain was cloned from human plasma
cell leukemia cells, ARH77. Both genes were previously re
ported (6, 7). The mouse variable region gene was used as an
EcoRl (5' of leader exon)-&zmHI (between J2 and J3 exons)
fragment for constructing a chimeric heavy chain gene (Fig. 1).
The Mlu\-lfam\\\ fragment of human heavy chain constant
region gene was prepared from a 21-kb fragment of the cloned
7i heavy chain gene, which also contained the human heavy
chain enhancer element. The mouse variable region gene was
joined to a 5' site of the human constant region in the same
transcriptional direction. The constructed chimeric gene was
then inserted into vector pSV2gpt at EcoRl-BamHl sites.
Construction of the Chimeric Light Chain Gene. The light
chain variable region gene was cloned from genomic DNA of
mouse hybridoma, NL-1, as a 6.5-kb Hindlll fragment using
the Charon 28 library. The DNA base sequence of the cloned
gene was determined (Fig. 2). The variable region gene had
rearranged to the .I: segment. The constant region gene of the
human K light chain gene was a 2.6-kb EcoRl fragment which
was cloned from the XgtWESXB genomic library of ARH77
using mouse constant region gene as a cross-hybridization
probe. The mouse variable region gene was trimmed to a
BamHl-Hindlll fragment and a small Himil\]~EcoRl fragment
of pBR322 was added to the 3' site of the variable region gene
segment. The resultant EcoRl-tìum\l\ fragment of mouse var
iable region gene was cloned into pSV2neo (Fig. 3). The human
K chain constant region gene was then inserted into an EcoRl
site of pS V2«covector containing mouse variable region gene.
The human heavy chain enhancer segment (Mlul-Hpal) was
ligated to the BamHl site (5' end of the variable region gene)
of the recombinant plasmid with the aid of an oligonucleotide
synthetic linker.
Transformation of Mouse Myeloma Cells. X63Ag8.653, a
nonproducer mouse myeloma cell line (9), was sequentially
transformed by the chimeric heavy and light chain genes. The
chimeric heavy chain gene was introduced into myeloma cells
by a protoplast fusion method (6, 12). Protoplasts (IO9) of
Escherichia coli MC 1000 harboring the plasmid were fused
with 2 x IO6 of X63Ag8.653 cells in the presence of polyeth
ylene glycol 4000 followed by selection with mycophenolic acid
(10). The resultant stable transformants which were obtained
about 3 weeks after the fusion were screened to select clones
producing the gene product of the introduced chimeric heavy
chain gene. Screening was done by staining the cytoplasmic
human 7 chain with FITC-labeled anti-human 7 chain antibody.
Four human 7 chain producing clones were obtained. The
clones producing chimeric heavy chain molecules were then
used as recipient cells for transfection of the chimeric light
chain gene. The chimeric light chain gene was introduced to
the cells producing chimeric 7 chain by a DEAE-dextran
method (13). Stable transformants appearing after about 20
days as G418-resistant clones were screened for the production
of the chimeric light chain gene products. Two positive clones
1002
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1987 American Association for Cancer Research.
RECOMBINANT
CHIMERIC
MONOCLONAL
100
80
60
x
o
40
NL-1
20
Chimera
>>
O
xl
x2
X4
X16
X8
X32
x64
Dilution of supernatant
Fig. 8. Complement-dependent cytotoxicity. Manca cells (100 jil; 4 x 10s)
labeled with "Cr chromate, were incubated for 30 min at 37'C with 100 pi of
diimeric antibody solution (1 /ig/ml) of the X63Agg.6S3 transformants (•)or
NL-1 monoclonal antibody solution (1 jig/ml) (A) in the presence (
) or
absence (
) of rabbit complement. MCr release from the Manca cells was
measured and the percentage of cytotoxic activity was calculated after subtracting
spontaneous "Cr release. Rabbit complement alone caused 11% "Cr release.
30
25.8«
*-*
200o0
10o
o0£•r.«*-l-»1-13.4.5*5B*8.0*ISO
2.3*FI
I
»5rh
20V
20
'i
V9
20V
SO
Y13.6*
50
1
1
Effector Cells
Chimera Sup
Only
Fig. 9. ADCC. Manca cells (2 x 10* in 50 /il) labeled with "Cr chromate were
incubated for 6 h with SOn\ of the chimeric antibody or NL-1 antibody and 100
nl of the human effector cells at effectontumor cell ratios as indicated. Antibody
concentration used was 1 «ig/ml.Amounts of "Cr released were measured in
triplicate assays, and the percentage of cytotoxicity was calculated after subtrac
tion of spontaneous release (6-13%). The chimeric antibody did not show any
ADCC activity against CCRF-HSB2 cells which were cALLA negative. Sup,
supernatant containing antibody (I «tg/ml);bars, SD.
NL-1 Sup
producing chimeric heavy and light chains were established
from 2x10* cells. Products of the recio ned transformants were
used for further analyses.
Analysis of Transcripts. Transcripts
of introduced genes in
ANTIBODY TO cALLA
the stable transformants were analyzed by Northern blot anal
ysis. As shown in Fig. 44, a band of 1.8 kb in length appeared
with RNAs of chimeric gene-transformed cells when the human
C,, probe (ftil-Psil fragment) was used. The probe gave no
such band in RNAs extracted from X63Ag8.6S3. The variable
exon probe of the mouse heavy chain gene (.PrwII-/fa/Il frag
ment) gave the same hybridizing band as the C\, probe. Hence,
the 1.8-kb RNA was assigned to the secretory type mRNA of
the chimeric heavy chain gene. Amounts of mRNAs in chimeric
gene-transformed myeloma cells were 3-4 times higher than
those in hybridomas. Similar observations were obtained for K
chain gene transcripts (Fig. 4«).The human Kconstant region
probe (Hpal-Hpal) gave a band of about 1.3 kb in length. The
band of the same size was observed when the blotted filler was
hybridized with the variable exon probe (Pstl-Hincil) of the
mouse K chain gene. Northern blot analysis revealed that the
introduced chimeric heavy and light chain genes were tran
scribed and processed accurately in X63Ag8.6S3 and might
yield functional mRNAs of the chimeric immunoglobulin genes.
Analysis of Translation Products. Immunoglobulin produc
tion was examined in the culture supernatants of X63Ag8.653
transformed by the chimeric genes. The stable transformants
were cultured in [3SS]methionine-containing medium for 8 h.
An affinity-purified goat antibody specific for human y chain
was bound to Protein A-Sepharose beads and added to the
cultured medium. The precipitates were applied to SDS-polyacrylamide gel electrophoresis giving a band corresponding to
the assembled protein of I I:L: at nonreducing conditions (Fig.
5¿»).
Two bands appeared on SDS-polyacrylamide gel electro
phoresis at the place corresponding to 7 heavy and light chains
when the sample was applied after reduction by 2-mercaptoethanol (Fig. 5a). These results indicated that the introduced
chimeric genes were transcribed, processed, and translated to
yield the chimeric immunoglobulin light and heavy chain peptide, which associated together to form the natural tetrameric
molecules H2L2 of IgG.
Glycosylation of Immunoglobulin y Peptide. Human immu
noglobulin 7i peptides possess sites of in vivo glycosylation that
are sensitive to tunicamycin treatment. The X63Ag8.653 trans
formants were cultured in RPMI 1640 complete medium sup
plemented with 1 Mg/ml tunicamycin and [3SS]methionine for
16 h at 37°C,and the resulting immunoglobulin molecules were
analyzed by SDS-polyacrylamide gel electrophoresis after re
duction with 2-mercaptoethanol (Fig. 6). The band of M, 50,000
in size appeared in addition to a faint band of MT 54,000
corresponding to the 71 chain. The transformants cultured
without addition of tunicamycin gave only the Mr 54,000 band.
In the case of ARH77 cells, where the human 71 gene was
cloned, the M, 54,000 band shifted to M, 50,000 upon addition
of tunicamycin to the culture medium (data not shown). These
data suggested that the products of the chimeric genes were
glycosylated similarly to native human 7 chains.
Binding Activity of the Chimeric Antibody to cALL Antigen.
X63Ag8.653 transformants secreted the assembled antibodies
to the culture fluid. The binding activity of the antibodies to
the cALL antigen was assayed using Manca cells as targets.
This is a human B-lymphoblastoid cell line (8) which expresses
IgM (/¿,K) and cALL antigen. The culture supernatant of the
transformed cells was incubated with Manca cells followed by
staining with FIÌ
('-labeled goat antibody specific for human 7
heavy chain. The staining was then assessed by flow cytometry
(Fig. IB). The chimeric antibody in the culture supernatants of
X63Ag8.653 transformants bound to Manca cells. The binding
activity was comparable to the mouse monoclonal antibody
1003
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1987 American Association for Cancer Research.
RECOMBINANT
CHIMERIC
MONOCLONAL
from NL-1 hybridoma (Fig. 7A). The chimeric antibody did not
react with human tumor cell lines, KS62 (chronic myelogenous
leukemia, blast cells), and CCRF-HSB2 (T-acute lymphocytic
leukemia) or with normal human peripheral blood lymphocytes
(Fig. 1C-E), which were lacking cALLA. No positive binding
was observed in the culture supernatants of X63Ag8.653 cells
or X63Ag8.653 cells transformed with the chimeric heavy chain
gene alone.
Complement-dependent Cytotoxicity. Human IgGl subclass
antibodies mediate complement-dependent
cytotoxicity. The
culture fluids of X63Ag8.653 transformants were added to the
labeled Manca cells together with rabbit complement (Fig. 8).
The complement used in this experiment caused 11% 51Cr
TO CALLA
ment. When the same amounts of antibodies were used, the
NL-1 monoclonal antibody (mouse y •*,
K)showed higher activ
ity in complement-dependent cytotoxicity than did the chimeric
antibody, whereas the same concentration of chimeric antibody
exhibited strikingly higher ADCC activity. This may be due to
the effectiveness of interaction between the human constant
domains and human effector cells. High ADCC activity may be
particularly beneficial for immunotherapy of cancer patients.
Boulianne et al. (3) reported a human-mouse chimeric anti
body (fi, K) specific for haptenic antigen, trinitrophenyl. They
observed that the transformed cells produced pentameric mol
ecules, which exhibited similar specificity to trinitrophenyl with
nearly equal affinity to that of the mouse hybridoma from
which the variable region genes were isolated. Morrison et al.
(4) reported human-mouse chimeric antibody of y-n type. The
variable region genes were obtained from S107, specific for
phosphorylcholine hapten. J558L and P3 mouse myeloma cells
were transformed by the chimeric genes and assembled immu
noglobulins of tetrameric form (H2L2) were produced. The
present study demonstrated the preparation of a human-mouse
chimeric antibody exhibiting specificity to human tumorassociated antigen of potential clinical importance. The
X63Ag8.653 transformants yielded satisfactory amounts of chi
meric antibody (10-30 tig/m\ in the serum-free medium) and
more concentrated chimeric antibodies were obtained in murine
ascites.4 This high expression of chimeric heavy and light chain
release, and the antibody alone exhibited infrequent lytic activ
ity (0.9%) in the absence of complement. The antibody solution
of the chimeric gene-transformed cells were clearly cytotoxic to
Manca cells in the presence of complement. The monoclonal
antibody obtained from murine NL-1 hybridoma showed twice
this level of cytotoxicity. The chimeric or NL-1 antibody was
not cytotoxic to K562 or CCRF-HSB2 (data not shown).
Antibody-dependent Cell-mediated Cytotoxicity by the Chi
meric Antibody. ADCC is an important function of certain
classes of antibodies, including IgG. The chimeric antibody
possessing the constant region of human origin was tested for
ADCC activity using human effector cells. Normal human
peripheral blood mononuclear cells were used as effector cells
(Fig. 9). The antibody of the chimeric gene transformants,
which showed low but significant activity in the complementdependent cytotoxic assay, exhibited twice the ADCC activity
as the same concentration of monoclonal antibody of the mu
rine NL-1 hybridoma. These results suggest that our chimeric
antibodies may function as effector molecules in antibodydependent cell-mediated cytotoxicity in vivo.
genes in murine myeloma cells may be due to the presence of a
human heavy chain enhancer element in the gene constructs
(6). Very recently, a human-mouse chimeric antibody against a
human breast cancer antigen has been reported (18). Assembly
of any other variable region genes to human ( \, and C, genes
with the human enhancer is possible and such human chimeric
antibodies offer a promising replacement for conventional mu
rine monoclonal antibodies.
DISCUSSION
REFERENCES
We have made a human-mouse chimeric antibody consisting
of murine variable region and human constant region domains.
The chimeric immunoglobulins were produced by transformed
mammalian cells. Chimeric immunoglobulin genes of heavy
and light chains were sequentially introduced to a mouse mye
loma. The introduced genes were transcribed and spliced accu
rately, and the resultant mRNAs were translated to immuno
globulin heavy and Klight chain peptides. These were assembled
together to yield a natural form of IgG and secreted into the
culture fluid. The chimeric antibody retained its specificity for
cALLA, which was defined by the variable region genes (VH
and V.) of mouse NL-1 hybridoma, and the antibody molecules
with human 71 and Kconstant domains demonstrated binding
to cells expressing cALLA. The carbohydrate moiety of im
munoglobulin molecules is reported to be indispensable for
some physiological functions, i.e., activation of complements
and ADCC (17). From this point of view, antibodies produced
by mammalian cells should be superior to those produced by
complementary DNA-type gene transfected to bacterial cells.
The molecular weight of the chimeric heavy chain changed
from 54,000 to 50,000 upon treatment of the transformed cells
with tunicamycin, indicating that glycosylation occurred nor
mally. The chimeric antibody caused the release of 5lCr from
labeled cells in the presence of complement, indicating that the
constant region functioned in complement-dependent cytotox
icity. ADCC is an important physiological function attributed
to certain classes of antibodies and has been implicated in self
defense against malignant cells. Thus, retention of this activity
is strongly desirable for antibodies intended for clinical treat
ANTIBODY
1. Miller, R. A., Oseroff, A. R., Stratte, P. T., and Levy, R. Monoclonal antibody
therapeutic trials in seven patients with T-cell lymphoma. Blood, 62: 988995, 1983.
2. Honjo, T. Immunoglobulin genes. Ann. Rev. Immunol., /: 499-528,1983.
3. Boulianne, G. I... Hozumi, V. and Shulman, M. J. Production of functional
chimaeric mouse/human antibody. Nature (Lond.), 312:643-646, 1984.
4. Morrison, S. L., Johnson, M. J., Herzenberg, L. A., and Oi, T. V. Chimeric
human antibody molecules: mouse antigen-binding domains with human
constant region domains. Proc. Nati. Acad. Sci. USA, 81:6851-6855,1984.
5. Ueda, R., Tanimoto, M., Takahashi, T., Ogata, S., Nishida, K., Namikawa,
R., Nishizuka, Y., and Ota, K. Serological analysis of cell surface antigens
of null cell acute lymphocytic leukemia by mouse monoclonal antibodies.
Proc. Nati. Acad. Sci. USA, 79:4386-4391, 1982.
6. Kudo, A., Ishihara, T., Nishimura, Y., and Watanabe, T. A cloned human
immunoglobulin heavy chain gene with a novel direct repeat sequence in 5'
flanking region. Gene (Amst.). 53:181-189, 1985.
7. Kudo, A., Nishimura, Y., and Watanabe, T. The antibody molecule to
common acute lymphocytic leukemia (cALL) antigen used the identical or
closely related V.., gene segment as that of MOPC-21 immunoglobulin heavy
chain. J. Immunol., /35:642-645, 1985.
8. Nishikori, M., Hansen, H., Jhanwar, S., Fried, J., Sondili. P., Koziner, B.,
Lloyd, K., and Clarkson, B. Establishment of near-tetraploid B-cell lym
phoma line with duplication of the 8; 14 translocation. Cancer Genet, t'y in
genet., 12: 39-50, 1984.
9. Kearney, J. F., Radbruch, A., Liesegang, B., and Rajewsky, K. A new mouse
myeloma cell line that has lost immunoglobulin expression but permits the
construction of antibody-secreting hybrid cell lines. J. Immunol., ¡23:15481550, 1979.R. ('.. and Berg, P. Selection for animal cells that express the
10. Mulligan,
EscHerichia coli gene coding for xanthine-guanine phosphoribosyl-transferase. Proc. Nati. Acad. Sci. USA, 78: 2072-2076, 1981.
11. Southern, P. J., and Berg, P. Transformation of mammalian cells to antibiotic
resistance with a bacterial gene under control of the SV40 early region
promotor. J. Mol. Appi. Genet., /: 327-341, 1982.
4 M. Yokoyama, Y. Nishimura, and T. Watanabe. Regression of tumor growth
by in vivo administration of recombinant chimi-rie monoclonal antibody specific
for cALL antigen, manuscript in preparation.
1004
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1987 American Association for Cancer Research.
RECOMBINANT CHIMERIC MONOCLONAL ANTIBODY TO cALLA
12. Sandri-Goldin, R. M., Goldin, A. L., and Levine, M. High frequency transfer
of cloned Herpes simplex virus type 1 sequences to mammalian cells by
protoplast fusion. Mol. Cell. Biol., /: 734-752, 1981.
13. Neuberger M. S. Expression and regulation of immunoglobulinheavy chain
gene transfected into lymphoid cells. EMBO J., 2:1317-1328, 1983.
14. Thomas, P. S. Hybridization of denatured RNA and small DNA fragments
transferred to nitrocellulose. Proc. Nati. Acad. Sci. USA, 77: 520U5205,
19gO
15. Yanisch-Perron, C., Vieira, J., and Messing, J. Improved M13 phage cloning
vector and host strains: nucleotide sequence of the M13 mpl8 and mpl9
vectors. Gene (Amst.), 33:103-119, 1985.
16. Hellstrom, I., Brankvan, V., and Hellstrom, K. E. Strong antitumor activities
of IgG antibodies to a human melanoma-associated ganglioside. Proc. Nati,
Aca(l. Sci. USA, 82:1499-1502,1985.
, 7 Nose M ^ wi
„H Biologiciu significance of carbohydrate chains on
monoclonal antibodies Prnc Nati AcaH Sri I ISA Art-fifiì7-Afi1i,IQ84
monoclonal antibodies. Kroc Nati. Acad. 5>ci.USA^ SO.6632-6636, 1984.
18' Sahagan B. G Dorai H., Saltzgaber-Muller, J., Toneguzzo, F, Gumdon,
C. A., Lilly, S. P., McDonald, K. W., Momssey, D. V., Stone, B. A., Davis,
G- L., Mclntosh, P. K., and Moore, G. P. A genetically engineered murine/
human chimeric antibody retains specificity for human tumor-associated
antigen. J. Immunol., 737:1066-1074, 1986.
1005
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1987 American Association for Cancer Research.
Recombinant Human-Mouse Chimeric Monoclonal Antibody
Specific for Common Acute Lymphocytic Leukemia Antigen
Yushi Nishimura, Masatoshi Yokoyama, Kazuo Araki, et al.
Cancer Res 1987;47:999-1005.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/47/4/999
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
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 June 14, 2017. © 1987 American Association for Cancer Research.