Production and nucleotide sequence of an inhibitory human IgM

From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
1994 84: 1968-1974
Production and nucleotide sequence of an inhibitory human IgM
autoantibody directed against platelet glycoprotein Ia/IIa
H Deckmyn, J Zhang, E Van Houtte and J Vermylen
Updated information and services can be found at:
http://www.bloodjournal.org/content/84/6/1968.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
Production and Nucleotide Sequence of an Inhibitory Human IgM
Autoantibody Directed Against Platelet Glycoprotein Ia/IIa
By H. Deckmyn, J. Zhang, E. Van Houtte, and J. Vermylen
Human B-cell lines were derived by limiting dilutions of Epstein-Barr virus (EBV) transformed peripheral B cells from a
patient with an autoantibody against glycoprotein (GP)la/
Ila, and manifesting defective collagen-induced platelet aggregation and a bleeding problem. Antibody-producing
clones were selected for their reactivity with wholeplatelets
or with affinity-purified GPlallla by enzyme-linked immunosorbent assay (ELISA). One of these cell lines, selected for
further evaluation, produced an IgM (E&) that interfered
with plateletaggregation responses. Polymerase chain reaction (PCR) amplifications with two different sets of primers
specific forhuman K-chains resulted in therescue ofa unique
and identical sequence. The same was true for the p-chain,
from which it was concluded that the cell line was mono-
clonal. Further analysis showed that the K variable domain
sequence is similar to the germline gene A30, to 2E7,an
anti-GPllb human autoantibody, and to HF2-1/17, a systemic
lupus erythematosus (SLE)-associated broad-specificity human autoantibody. Thus, the specificityofour
antibody,
E&, appears to bedetermined by the p-chain, the sequence
of which is encoded by a Vnlll gene segment strongly homologous to the germline gene DP-77, by a D gene that is not
homologous to any of the germline D genes reported to
date, and by JH4gene segment that is germline. All four
mutations versusDP-77 are in CDRs, and result in amino
acid substitutions, which implies that E& may have been
derived from an antigen-driven response.
0 1994 by The American Societyof Hematology.
T
autoantibody specificity. Furthermore, by comparing several
autoantibodies of defined antigen specificity, the contribution
of individual segments of the variable regions to the identified specificity of the autoantibodies may be determined.
HE PLATELET glycoprotein (GP)Ia/IIa complex is a
member of the integrin family, where it is known as
a2PIor VLA-2. In the platelet membrane, as in other cell
membranes, itplaysacrucial
role in the interactionwith
collagen, oneof the more thrombogenic substances
in a damaged blood vessel wall. Its role as a collagen receptor was
first indicated by the finding that platelets lacking GPIa fail
to react properly with collagen." Later, it was shown that
GPIdIIa could be adsorbed specifically onto collagen in a
Mg2+-dependent
whereas
liposomes
incorpowith
rated GPIdIIa bound to collagen.' Several monoclonal antibodiesreacting with GPIa"" or with GPIIa interfere, to a
variable extent, with collagen-induced platelet aggregation.
We havepreviously described apatientwithanacquired
bleeding disorder, whose platelets specifically failed to react
with collagen." Further studyofthispatient
showed the
presence of an IgG autoantibody against GPIa in her plasma,
whichinhibited
collagen-induced aggregation of normal
platelets. To furthercharacterize thisantibody,its
correspondingepitope,and itspotentialantithromboticeffects,
we set off to generate a B-cell line that would produce this
human anti-GPIa/IIa antibody. In addition, elucidation of its
sequence was undertaken to provide better insight into the
utilization of particular immunoglobulin germline genes and
the contributionofindividualvariableregion
segments to
From the Center for Molecular and Vascular Biology, Leuven;
and the Department of Immunology, DrH. Willems Institute, Diepenbeek, Belgium.
Submitted October 11, 1993; accepted May 3, 1994.
Supported by a grant from the Vluums Aktieprogramma Biotechnologie (VLAB/08I). J.V. is holder of the Dr J. Choay Chuir in
Haemostasis Research.
Address reprint requests to H. Deckmyn, PhD, Center for Molecular and Vascular Biology, University of Leuven, Campus Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium.
The publication costs of this article were defrayed in part by page
charge payment. This article must therefore be hereby marked
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
indicate this jact.
0 1994 by The American Society of Hematology.
0006-4971/94/8406-0006$3.00/0
1968
MATERIALS ANDMETHODS
Production of Ig-producing B-cell lines b.y Epstein-Barr virus
of blood from the
transformation. B-cell isolationfrom30mL
patient with the anti-GPla autoantibody was performed essentially
as described." The T-cell-depleted cellular fraction was incubated
with Epstein-Barr virus (EBV)-containing supernatant from a culture
of the marmoset cell line B95.8 for 2 hours at 37°C. The cells were
then pelleted and seeded in microwells.
Culture and cloning of the EBV-infected cells was performed as
previously described." Briefly, cells were seeded in supplemented
RPM1 1640 on a feeder cell layer ofIO4 irradiated (30 Gy) allogeneic
peripheral blood mononuclear cells. Cells were incubated at
37°C
in 5% CO2, with the culture medium being replaced weekly. The
supernatant of each well was screened for specific antibody production by enzyme-linked immunosorbent assay (ELISA) (see below).
Positive cultures were transferred to 24-well plates, and selected for
further cloning at dilutions ranging from 1 to 50 cells per well. in
the presence of IO4 irradiated feeder cells.
Charucterizufion qf untibodies. Screening of cellsupernatants
was performedwith ELISA procedures using microtiter plates coated
with human platelets, purified GPIdIIa, or human plasminogen activator inhibitor-l (PAT-I) as a negative control.
Human platelet-rich plasma (PRP)
was prepared by differential
centrifugation (10 minutes, 150g) of blood anticoagulated with 0.15
vol 25 mmolk trisodium citrate, 7 1 mmol/L citric acid, I 1 1 mmoll
L dextrose, pH 4.5 (ACD). Gel-filtered
platelets (GFP) were obtained
from this PRP by chromatography on a 20 X 2.5 cm Sepharose 2B
column (Pharmacia, Uppsala, Sweden), using5 mmol/L Hepes, 0.34
NaCI. 2.9
m m o l k Na2HP0,,12 mmollL NaHCO,,134mmol/L
mmol/L KCI, and 10% ACD, pH 5.5.''
After two cycles of centrifugation (10 minutes, 800g) and resuspension, 125 X IO" platelets were disrupted by sonication. On
removal of intact platelets during centrifugation at 2,OOOg for 10 minutes, membranes were collected by centrifugation at 50.0001: for I
hour.Themembranepelletwasresuspended,solubilizedwith
10
mmol/L CHAPS (3-[3-~holamidopropyl) dimethylammonioJ- I-proSt Louis, MO), and first-passed
pane sulfonate; Sigma Chemical CO,
over a Sepharose column. The nonbound fraction was then applied
to a CNBr-Sepharose column, to which the monoclonal anti-GPIa
antibody CLB-IOGl I (a gift from Dr Van Mourik. Amsterdam, the
Blood, Vol 84, No 6 (September 15). 1994: pp 1968-1974
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
ANTI-GPIAAIAHUMAN ANTIBODY
Table 1. Oligonucleotide PrimersUsed for PCR Amplification of EIGscDNA
Primer
Sequence
Pst242 (backward)
Pst243 (forward)
Pst244 (forward)
Pst245 (backward)
Pst246 (backward)
Psl247 (forward)
Psr248 (forward)
GGAAAAGGGITGGGGCGGATGC
TITGGGAGGCAGCTCAGCAATC
GAMATYGTGWTGACNCAGTCTCC
GAMATYCAGWTGACNCAGTCTCC
TGAAGACAGATGGTGCAGCCAC
ACACTCTCCCCTGITGAAGCTC
Sequence degeneracies are given in the single letter code: N
=
A/G/CF, R
Netherlands) had been coupled, and was recycled over the column
for 16 hours at 4°C; nonbound material was washed off with 10
~ o V HEPES,
L
pH 7.4,20
MgClz, and 2 m m O V L CHAPS.
Specifically adsorbed material was eluted with the same buffer containing 2 m o m KSCN, dialyzed, and concentrated. Purity of the
material was checked by sodium dodecyl sulfate-polyacrylamide gel
electrophoresis (SDS-PAGE) and silver-staining (Biorad, Richmond,
CA)," and by ELISA assays using monoclonal anti-GPIIbfiIIa13or
anti-GPIa/IIa14 monoclonal antibodies. Only the latter gave a significant signal (A 492-nm >0.3 v <.07). PAL1 was purified as
de~cribed.'~
Two hundred microliters of 150 to 200 X 103/pLGFP, 2 to 5 pg/
mL affinity-purified GPIa/IIa, 5 pg/mL PAI-1, or 5 pg/mL affinitypurified GPIIb/IIIa" was pipetted into the wells of microtiter plates,
and coated overnight at 4°C. Unbound material was washed away,
and remaining binding sites were blocked with 1% bovine serum
albumin in phosphate-buffered saline. After washing, plates were
incubated with 170-pL serial dilutions of cell supernatants or purified
antibody for 1 to 2 hours at room temperature, washed, and further
incubated for 1 hour with 170 pL of a 1:1,OOO dilution of goat
antihuman IgG andlor a 1: 1,000 dilution of goat antihuman IgM
(Sigma Chemicals), both coupled to horseradish peroxidase (HRP).
After an additional washing step, bound antibody was stained with
160 pL of a solution containing 0.4 mg/mL 1,2-ortho-phenylenediamine (Fluka Chemie, Buchs, Switzerland) in 17 mmoVL citric acid,
65 mmovL NaZHP04,and 0.003% HZ02.The reaction was stopped
with 50 pLof 4 m o m &Sod. Absorption was measured at 492 nm
in an ELISA reader (EAR 400 AT, SCT-Labinstmments, Grodig,
Austria). Platelet aggregations were performed either in suspensions
of 200 pL 200,OOO/pL GFP in 5 mmoVL HEPES, 0.34 mmoUL
NazHP04, 12 mmoVL NaHCO,, 134 mmoVL NaCl, 2.9 mmoVL
KCI, I mmoVL MgCI2, 1 mmoVL CaClz, and 50 pL platelet-poor
Table 2. EBV-Transformed B-Cell Lines and Clones Producing
Antibodies Against GPla/lla
Lines and
Clones
2.7
2.5
2.3
A
Patient IgG0.3
0.24
C5 MC4.100
2.2
MC4.100 E 3
F2 MC4.100
D6 MC4.100
0.2
MC F31.1
1.6
0.5 2 0.04
0.43
Negatives
M
Hv-FR1
5' end of MuCl
5' end ofMuC2
KV-FR1
K"-FR1
5' end of &
3' end of &
AGGTGCARCTGSWGSAGTCKGG
B
C
D
E
0.08
3.0
5.4
11.0
3.1
0.53.4
1.5
0.2
1.4
0.5
0.3
5.0
7.1
8.0
?
0.01
0.59
2
7.5
4.0
4.1
5.5
0.02
0.9
0.7
A 492-nm measurement obtained when patient IgG (30 pg/mL final
concentration) or cell supernatant was tested in ELlSAs on purified
GPlallla (A), platelets (B), or PAI-1 (C),and developed with a mixture
of both goat antihuman IgM and IgG conjugated with HRP. For better
comparison, ratios D = A/C, E = B/C are given. Results on negatives
were obtained in a separate experiment.
=
A/G, Y
=
Cfl, S
=
G/C, W
=
M, K = G/T, M
=
A/C.
plasma (from trisodium citrate anticoagulated blood), or in 200 pL
citrated PRP. Aggregation at 37°C was triggered with either collagen
(Hormon Chemie, Munchen, Germany) or adenosine diphosphate
(ADP Sigma Chemicals) after a preincubation with antibodies for
the indicated times at room temperature.
Purification of anti-GPldlla antibody. ELISA-positive cell supernatants were passed over a CNBr-Sepharose (1 mL) column to
which 1.05 mg affinity-purifiedGPIdIa had been coupled according
to the manufacturer's instructions (Pharmacia). Specifically adsorbed
material was eluted with 2 m o m KSCN anddialyzed against HEPES
buffer. An IgM-producing clone (E3G6) was selected for further
study. Patient IgG was purified from plasma using protein A Sepharose as described."
Sequence of the variable regions of fhe K and pFd-chains of E3G6IgM. Polymerase chain reaction (PCR) amplification wasusedto
rescue the entire K-chain and the p-Fd fragment (heavy-chain variable and first constant region) of the E3G6IgM,I6 thus facilitating
eventual production of a recombinant E3G6-derivedFab fragment.
RNAwas extracted from lo4 to lo5 E3G6 cells and mRNA was
selected on oligo (dT)-cellulose. First-strand cDNA, synthetized using the forward primers Psr248 and Psr244 (Table l), and Moloney
murine leukemia virus (MMLV) reverse transcriptase, was amplified
with Taq DNA polymerase using the primer couples Psr245/246 and
Psr248 (entire K-Chain, Psr245/246 is a mixture of two oligonucleotides), and Psf244 and Psr242 (p-Fd heavy-chain fragment) (Table
l). The correctly sized K and p-Fd PCR products were isolated from
agarose gel, and subsequently used as templates for amplification of
the respective variable regions. This was done with the primer couples Pst246 + Psf247 and Psf242 + Psf243, respectively. The reamplifications were performed with 32P-labeledprimers such that the
obtained radiolabeled variable regions could be sequenced with the
chemical degradation method."
RESULTS
Following limiting dilutions of EBV-transformed peripheral B cells derived from a patient with an anti-GPIa autoantibody, five cell lines were selected, producing positive supernatants during ELISA screenings with immobilized
platelets or GPIalIIa, but notwith PAI-1. The cell line,
MC4100 E3, with the highest IalIIa versus PAI-1 and platelet
versus PAI-1 ratio (Table 2 ) was further subcloned, resulting
in six positives again subdivided in 10 subclones, which
were checked for their reactivity in the ELISAs. Clones belonging to that group (G) showing the best relative reactivity
versus GPIa/IIa and platelets were further expanded, and the
clone that proliferated most extensively (G6) was selected
for further characterization. This E3Gs cell line was kept in
culture for several months, and the cell supernatant regularly
tested for the presence of anti-GPIa/IIa antibody.
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
DECKMYN ET AL
1970
Table 3. Determination of Specificity and Subtype of E3G6-Produced Antibody
lallla
Sample
W
IgM
IgG
W
W
IgM
1
2
3
0.060
0.058
0.055
0.579
0.631
0.516
0.132
0.128
0.128
0.438
0.262
0.203
0.000
0.000
0.000
0.092
0.048
0.020
A 492 n m value when different samples of E3Gssupernatant were checked by ELISA on purified GPla/lla, platelets, or PAI-1, and developed
with either goat antihuman IgG or IgM conjugated with HRP.
The antibody produced by E3G6 was of the IgM subtype
as substantiated from its detection by an antihuman IgM, but
not by an antihuman IgG antibody, following its binding to
the GPIa/IIa receptor (Table 3). Furthermore, anti-lightchain antisera identified a K light chain in the E3G6-IgM in
a Western blot.
The GPIa/IIa-specific antibody could be purifiedon a
GPIa/IIa-Sepharose column. Flow-through and the 2-moVL
KSCN elution peak were collected, dialyzed, and analyzed
in an ELISA using purified GPIa/IIa or PAI-1, and goat
anti-IgM-HRP for the detection (Table 4). GPIa/IIa-reactive
material was specifically recovered in the KSCN fraction.
Specificity of binding was checked in a direct comparison
using affinity-purified GPIIb/IIIa13 versus GPIa/IIa inan
ELISA setting. E3Gs binding to GPIa/IIa resulted in a 3.2fold higher signal thanthat obtained with GPIIblIIIa, the
latter being comparable to the nonspecific binding that was
observed when PAI-1 was used as coating material.
Finally, the activity of the E3G6antibody on platelet aggregations was assessed on several occasions. A consistent inhibition of aggregations induced by collagen using GFP (Fig
l), as well as PRP (Fig 2), was observed. On the contrary,
aggregations induced by ADP were not influenced (Fig 1).
PCR ampl$cation of the K - and p-Fd-chains of G6-IgM.
PCR amplification of first-strand cDNA resulted for both Kand p-chains in fragments of the expected size. The reactions
were performed with backward primers that match the 5‘
ends of the respective framework 1 regions, while the forward primers match with sequences in the constant domains.
Sequence analysis yielded a unique sequence for each, implicating that the PCR products contained single distinct fragments. To ensure that the E3G6 cell line was truly monoclonal, additional PCR amplifications were performed using
Table 4. Affinity Purification of E,G6 Supernatant
Over a GPla/lla Column
ELISA
Protein
Sample
GPla/lla
0.032
Pre-
0.005
0.025
Flow-through
Peak
0.473
0.030
0.057
alternative primer couples (a number of which were provided
by Dr K. Thielemans, Hematology-Immunology Unit, Free
University Brussels, Belgium). In the case of theheavy
chain, a set of family-specific backward primers matching
with leader sequences was used; only the use of the VHlll
family-specific primer resulted in a PCR product, the sequence was identical to the one mentioned above. In addition, these experiments have identified the previously uncertain N-terminal sequence of the heavy chain as E-V-Q-L-VE-S-G-G-G. In the case of the light chain, one alternative
backward primer, which like Pst245/246 matches with sequences in framework 1, was tried. The PCR product was
sequenced and found to be identical to the one previously
obtained.
Sequence of E3G6 heavy- and light-chain variable regions.
The nucleotide sequence of the E3G6heavy chain is shown
in Fig 3A, along with the sequence of the germline gene
DP-77,I8 a member of the VHlllfamily, with a nucleotide
similarity of 98.5%. In addition, within the overlapping region, DP-77 is identical to the WHG16I9 germline gene,
and to two fetal liver-derived heavy chains: the hse 54 3.2
mRNAZ0and the L3AloB-cell line.” Whereas no homology
was found between the D region withany of the known
germline sequences, the JH region on the other hand is identical with the JH4 germline gene.
The nucleotide (Fig 4A) and amino acid sequence (Fig
4B) of the E3Gh K-chain isnearly identical tothe one of
2E7, a human autoantibody against the GPIIb subunit present
in the GPIIb/IIIa or cuIIbp3 complex” and HFz-l/l7, abroadspecificity human autoantibody” that cross-reacts with
ssDNA, cardiolipin, and an unidentified platelet antigen,
most likely gly~olipid,”,’~
that is found in patients with systemic lupus erythematosus (SLE).
Both the EIGh and the 2E7 light chains are most homologous tothe A30 germline gene,” with a nucleotide similarity
of 99.6%. The E3G6 and 2E7 light chains differ from A30
by a two-base insertion.
DISCUSSION
PAIL1
0.1314
0.257
Eight milliters of supernatant was passed over the column;
through and the 2 mol/L KSCN elution peak were collected, dialyzed,
and analyzed in an ELISA using purified GPla/lla or PAI-1, and goat
antihuman ISM-HRP for the detection.
We report here on the identification of a monoclonal human IgM autoantibody against platelet GPIa/IIa by a B-cell
line obtained following EBV transformation of B cells from
a previously described patient.” This patient had an acquired
bleeding disorder and defective platelet aggregation towards
the flowcollagen, which was due to the presence of an autoantibody
against GPIa in her plasma.
Following repetitive subcloning, a monoclonal cell line
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
ANTI-GPIMIIA HUMAN ANTIBODY
1971
collagen
1
2
3
Fig 1. Platelet aggregation
tracings of 150 pL GFP with 50
p L plasma preincubatedwith 50
p L medium (A) or EaGs cdl su15 20
pernatant (B) for 5 (l), (2).
(3).or 30 minutes (4). after which
aggregation was induced with 1
pg/mL collagenor 60 pmol/L
ADP.
was obtained. PCR amplifications using two different pairs
of primers for both heavy and light chains yielded identical
sequences. We have no indication that the original antibody
from the patient was monoclonal, eg, no evidence for a
paraprotein was obtained. Therefore, it is possible that the
antibody-producing cell line that was isolated produces only
one of the possible multiple anti-GPIa antibodies present
originally.
In addition, whereas the original plasma antibody was of
the IgG class, the antibody identified here is an IgM. This
was obvious from ELISA data and SDS-PAGE analysis. The
IgM antibody is directed against platelet GPIdIa, because
it bound to affinity-purified G P I d I a both in ELISA and
during affinity chromatography on immobilized SepharoseGPIa/IIa, whereas binding topurified GPIIb/IIIa wasnot
higher than the nonspecific binding to PAI-1.
Furthermore, the anti-GPIa/IIa antibody E3G6 was functionally active, because, when preincubated with normal
platelets in buffer or in plasma, it markedly inhibited platelet
aggregation induced by collagen, but not by ADP. The antibody did not inhibit collagen-induced platelet shape change.
All of these findings are in agreement with our earlier observations with the native antiserum" and, therefore, provide
good evidence that the antibody sequence selected may also
be produced in the patient. However, because the EBV-
transformation technique does not allow differentiation between naturally expressed and nonexpressed antibodies, further work is needed to fully prove their identity. The human
monoclonal antibody obtained nevertheless is inhibiting
G P I d I a function, and sequence analysis was performed to
possibly identify relevant sequences.
Sequence comparison of PCR-amplified fragments indicates that the nucleotide sequence of the K-variable domain
of &G6 is most homologous to the A30 germline gene,26
with a nucleotide similarity of 99.6%. The E3G6light chain
differs from A30 by a two-base insertion, suggesting that
A30 may be a pseudogene. Further comparison showed that
the &G6 K-chain is also highly homologous to the 2E7 light
chain,22which has the GG insertion identical to
It is
likely that the E3G6light chain is encoded by a new germline
VKgene, yet to be identified. &G6 differs from 2E7 in codon
95 CCG CCT and codon 108 CGA -+ CCT (T. Kunicki,
personal communication, April 1993), andonly the latter
results in an amino acid substitution R + P. Despite this,
we can conclude from both the nucleotide and amino acid
sequence of the K-chain of E3G6 and 2E7 that the antibody
light chains probably do not contribute to the specificity of
the antibodies, unless they would be involved in targeting
the antibodies to integrins.
Furthermore, the amino acid sequence of the E3G6 K-chain
+
3
Fig 2. Platelet aggregation
tracings of PRP incubated for 30
minutes with 50 (l), l00
(2). or
150 (3) p L of buffer (A) or GPla/
Ila affinity-purified antibody
from E.G. cell supernatants (B)
and stimulated with 0.8 pg/mL
collagen.
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
DECKMYN ET AL
1972
CDRl
A
E3G6
10
20
*
30
*
40
*
50
*
*
60
70
*
*
80
-*
90
t
"
GAM ATYCAGWTGACNCAGTCTCCATCCTCCCTGTCTGCATCTGTAGGAGACAGAGTCACCATCACTTGCCGGGCAAGTCAGGGCATTAGAAATGAT
CDR2
110
100
120
130
140
*
"
L
*
E3G6TTA
GGC TGG TATCAGCAGAAACCA
GGG AAA GCC C C 1A A G
." ." ". ".
". ."
"ZE7
"_
_" _ _ _
"_
150
"_ "_
- *
CGC CTGATCTATGCTGCATCCAGT
_ " . _ ___ _
"_
170
160
*
"_ "_ _ _ _ "_
TTG CAAAGT
"_
180
GGG GTCCCATCAAGGTTCAGC
". - _ .
_ _ _ _" _ - _ "_
_""_
"-
COR3
*
t
230
220
210
200
240
250
*
*
270
260
*
t
E 3 t 6A G T
GGA TCT GGG ACAGAATTCACTCTCACAATCAGCAGCCTGCAGCCTGAAGATTT1GCAACTTATTACTGTCTACAGCATAATACTTACCCG
." ".
_ __-.
.
"."
".
2E7
~ 3 0
.-.
___
"_
"_"_
". ...
___ ___
."
"__"
"_ "_ "_
_ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _
_ _ _ _" "_ _" "_
___
___ ___ ___ ___
280
"*
_ _ _ "- _ _ - "-"- - _ - ". - - - "_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _
".
"T
"T
JK1
JK
290
-*
300
310
320
*-
330
*
340
*
350
*
360
*
Fig 3. (A) Nucleotide sequenceof EtG. K-chain variable domain. TheVKsequence of €*G. is compared with the 2E7-anti-GPllb autoantibody
and VK germline gene M O . Identical bases are represented by a dash. CDR-1, CDR-2, and CDR-3 regions are indicated. The EtGa and 2E7 JK
genes are most homologous to the JK, gwmline gane, as depicted. (B) Comparison b.twem the deduced amino acid sequence of EsGsVK
segments with those of 2E7, of HF2-1/17 (experimentally determined), and
of the deduced gwmllne JK,-sequence.
is nearly identical to that of HF2-1/17. HF2-1/17 is a broadspecificity human autoantibody that reacts with, for example,
platelets and ssDNA, and for which the amino acid sequence
has been determined directly by protein ~ e q u e n c i n g2E7,
.~~
a human autoantibody directed against GPIIb, a member of
another platelet glycoprotein complex GPIIblIIa, does not
react with S S D N A .Furthermore,
~~
it is unlikely that 2E7,
produced by a human-mouse heterohybridoma cell line generated from the fusion of splenocytes derived from a patient
with immune thrombocytopenia, would cross-react with
GPIa/IIa, because both complexes are on platelets and, therefore, characterization of the antigen would have yielded a
double signal. Likewise, we did not obtain evidence for
cross-reactivity of our antibody with GPIIbnIIa in an ELISA
setting. Hence, despite the extremely high homology of their
light chains, these antibodies have well-defined and different
antigen specificities, which consequently are determined by
their heavy-chain sequence. The heavy variable region of
E3Gs is a member of the VHllrfamily, as evidenced by the
sequence and by the finding that PCR amplification only
was successful with aVHIIlfamily-specific primer. The E3G6
heavy chain has a 98.5% nucleotide similarity to the DP-77
germline gene.18 All four mutations are in CDRs, and all
cause amino acid substitutions, resulting in a replacement to
silent ratio of infinity, which implies that E3G6 may be derived from an antigen-driven re~ponse.~'
The J H segment, on
the other hand, is identical to the germline J,.
The present findings seem to confirm earlier observations22,28,29 that autoantibody sequences are barely differentiated from the germline, potentially explaining why these
rather immature antibodies are less able to differentiate between self- and nonself-antigens. Recently, Denomme et a130
reported on the production of human monoclonal antiplatelet
autoantibodies derived from tonsillar lymphocytes from a
nonthrombocytopenic donor fused to a lymphoblastoid cell.
It is interesting to note that the antiplatelet antibodies obtained could be classified into two different and mutually
exclusive groups. One group shows reactivity to a variety
of platelet proteins, and the other to ssDNA and anionic
phospholipids. Unfortunately, no sequence data on these an-
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
ANTI-GPWIIA HUMAN ANTIBODY
1973
COR1
10
A
40
*
E3G6
GAG GTG CAG CTGGTC
Dp-77
___
_ __ __ _ _
20
30
*
70
*
___
___
___
___
90
80
60
t-
t
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ GCA
___ _ _ _ _ _ _ _ _ _ _ _ -
SAG TCT GGC GGA CCC CTCGTCAACCCT
___ ___
50
___
CCC CCG TCCCTC
ACA CTCTCC
TCT
CCCTCT
CCA T T C ACC TTCACT
ACC TAT
--- - _ _ _-- -C- - - -
COR2
110
100
-*-
120
130
*
*
140
150
160
-*
E3Gg AGC ATG AAC TGG GTC CGC CAG GCT CCA CGC AAC GGG CTC GAG TGG GTCTCATCCATTAGTCCTACTACTAGTTACATATACTAC
DP-77
AC- -G-
___ ___ ___
___ ___ ___ ___ ___ ___ __ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___
200
210
E3G6
AAG GGC CGA TTC ACC ATCTCC
Dp-77
_ _ _ _____
220
*
- t
___ ___
230
240
AGA GAC AACGCCAACAACTCACTCTATCTGCAAATGAAC
___ ___ ___ ___ _____ _______
250
*
t
___ ___
310290
*
E3Gg
DP-77
-
GCG AGA GAT
___
_._
190
*-
CCA GAC TCACTG
270
*
280
*
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___ _ _ _ _ _ _ _ _ _ ___ _ _ _ _ _ _ ___ _ _ _ ___
AGC CTG AGA GCC GAG CACACCCCTCTC
TAT TACTCT
JH
300
GGG GCG TTTTTT
180
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ______ _ _ _ _ _ _ _ _ _
260
*
D
320
170
360
*-
AGC ACT GGC TGC GGC CTT GAC TACTCG
___ ___
JHL
___
CGC CAG GGA ACCCTCCTCACCCTCTCCTCA
___
--A
___
___ ___
_____ ___ ___ ___
Fig4. (A) Nucleotide sequence comparison of the E,Gs-VH chainand the germline VH gene of DP-77,and the germline JH-4. (---l Base
(B)Deduced amino acid sequence
of the EsGsand 2E7 VHgene
identity with the EIGs sequence. CDR-1, CDR-P, D, and JH regions are indicated.
segments comparedwith the experimentally determined protein sequence of HF2-1/17, the deduced sequence of the DP-77 V, segment, and
the Jw germline segment.
tibodies are available at present, excluding a further comparison.
In conclusion, by transforming B cells from a patient with
an acquired autoimmune bleeding disorder due to an inhibitory autoantibody against GPIa, a monoclonal cell line was
selected that produced an inhibitory IgM reacting with the
G P I d I a complex. This antibody shares a K-chain with at
least two other human autoantibodies, and therefore the pchain must be responsible for antigen-recognition specificity.
ACKNOWLEDGMENT
The authors thank P. Stanssens and Y. Gansemans (Corvas International N.V., Gent, Belgium) and P.P. Chen (University of California San Diego, La Jolla, CA) for valuable help in the sequence
determination and comparison, respectively, T.J. Kunicki (Scripps
Research Institute, La Jolla, CA) for critical reading of the manuscript, P.J. Declerck for his contribution in the cell culture, and
T. Mahau andJ. Vangoetsenhoven for careful preparation of the
manuscript.
REFERENCES
1. Nieuwenhuis HK, Sakariassen KS, Houdijk WPM, Nievelstein
PFEM, Sixma JJ: Deficiency of platelet membrane glycoprotein Ia
associated with a decreased platelet adhesion to subendothelium. A
defect in platelet spreading. Blood 68:692, 1986
2. Nieuwenhuis HK, Akkerman JWN, Houdijk WPM, Sixma JJ:
Human blood platelets showing no response to collagen fail to express surface glycoprotein Ia. Nature 318:470, 1985
3. Kehrel B, Balleisen L, Kokott R, Mester R, Stenzinger W,
Clemetson W,van de Loo J: Deficiency of intact thrombospondin
and membrane glycoprotein Ia in platelets with defective collageninduced aggregation and spontaneous loss of disorders. Blood
71:1074, 1988
4. Santoro SA: Identification of a 160,000 dalton platelet membrane protein that mediates the initial divalent cation-dependent adhesion of platelets to collagen. Cell 46:913, 1986
5. Santoro SA, Rajpara SM, Staatz WD, Woods VCJ Jr: Isolation
and characterization of a platelet surface collagen binding complex
related to VLA-2. Biochem Biophys Res Commun 153:217, 1988
6. Staatz WD, Rajpara SM, Wagner EA, Carter WG, Santoro SA:
The membrane glycoprotein Ia-IIa (VLA-2) complex mediates the
Mg*+-dependent adhesion of platelets to collagen. J Cell Biol
108:1017, 1989
7. Kunicki TJ, Nugent DJ, Staats SJ, Orchekowski RP, Wagner
EA, Carter WC: The human fibroblast class I1 extracellular matrix
receptor mediates platelet adhesion to collagen and is identical to
the platelet glycoprotein Ia-IIa complex. JBiol Chem 263:4516,
1988
8. Coller BS, Beer JH, Scudder LE, Steinberg MH: Collagenplatelet interactions: Evidence for a direct interaction of collagen
with platelet GPIdIIa and an indirect interaction with platelet GPIIbl
IIIa mediated by adhesive proteins. Blood 74:182, 1989
From www.bloodjournal.org by guest on October 15, 2014. For personal use only.
1974
9. Sonnenberg A, Modderman PW, Hogervorst F Laminin-receptor on platelets is the integrin VLA-6. Nature 336:487, 1988
10. Deckmyn H, Chew SL, Vermylen J: Lack of platelet response
to collagen associated with an autoantibody against glycoprotein Ia:
A novel cause of acquired qualitative platelet dysfunction. Thromb
Haemost 64:74, 1990
11. Zhang 5, Lambrechts J, Heyligen H, Vandenbark AA, Raus
J: Human B-cell-lines secreting IgM antibody specific for myelin
basic protein. J Neuroimmunol 23:249, 1989
12. Tangen 0, Berman HJ:Gel filtration of blood platelets: A
methodological report. Adv Exp Med Biol 34:235, 1972
13. Deckmyn H, Stanssens P, Hoet B, Declerck PJ, Lauwereys
M, Gansemans Y, Tomai I,Vermylen J: An echistatin-like ArgGly-Asp (RGD)-containing sequence in the heavy chain CDR3 of
a murine monoclonal antibody that inhibits human platelet glycoprotein IIblIIIa function. Br J Haematol (in press)
14. Santoso S, Kiefel V, Mueller-Eckhardt C: Immunochemical
characterization of the new platelet alloantigen system Bf/BP. Br
J Haematol 72: 19I , 1989
15. Alessi MC, Declerck PJ, DeMolM,
Nelles L, Collen D:
Purification and characterization of natural and recombinant human
plasminogen activator inhibitor-l (PAI-l). Eur J Biochem 175531,
l988
16. Marks JD, Tristem M, Karpas A, Winter G: Oligonucleotide
primers for polymerase chain reaction amplification of humanimmunoglobulin variable genes and design of family-specific oligonucleotide probes. Eur J Immunol 21:985, 1991
17. Maxam AM, Gilbert W: A new method for sequencing DNA.
Proc Natl Acad Sci USA 84560, 1977
18. Tomlinson IM, Walter G, Marks JD, Llewelyn MB, Winter
G: The repertoire of human germline VH sequences reveals about
fifty groups of VH segments with different hypervariable loops. J
Mol Biol 227:776, 1992
19. Kuppers R, Fisher U, Rajewsky K, Gause A: Immunoglobulin
heavyand light chain gene sequences of a human CDS positive
immunocytoma and sequences of four novel V,,,, germline genes.
Immunol Lett 3457, 1992
DECKMYN ET AL
20. Cuisinier AM, Gauthier L, Boubli L, Fougereau M, Tonnelle
C: Mechanisms that generate human immunoglobulin diversity operate from the8thweekof
gestation in fetal liver. Eur J Immunol
23:110, 1993
21. Hillson JL, Oppliger IR, Sasso EH, Milner ECB, Wener MH:
Emerging human B cell repertoire. Influence of developmental stage
and interindividual variation. J Immunol 149:3741, 1992
22. Kunicki TJ, Annis DS, Gorski J, Nugent DJ: Nucleotide sequence of the human autoantibody 2E7 specific for the platelet integrin IIb heavy chain. J Autoimmun 4:433, 1991
23. Dersimonian H, Schwartz RS, Barrett KJ, Stollar BD: Relationship ofhuman variable region heavy chain germline genes to
genes encoding anti-DNA autoantibodies. J Immunol 139:2496,
1987
24.Asano T, Furie BC, Furie B: Platelet binding properties of
monoclonal lupus autoantibodies produced by human hybridomas.
Blood 66: 1254, 1985
25. Asano T, Furie BC, Furie B: Glycolipid is the platelet autoantigen of platelet-binding monoclonal lupus autoantibodies produced
by human hybridomas. Clin Res 34:654A, 1986
26. Lautner-Rieske A, Huber C , Meindl A, Pargent W, Schable
KF, Thiebe R, Zocher I, Zachau HG: The human immunoglobulin
kappa locus. Characterization of the duplicated A region. Eur J
Immunol 22: 1023, 1992
27. Olee T, Lu EW, Huang DF, Soto-Gil RW, Deftos M, Kozm
F, Couson DA, Chen PP: Genetic analysis of self-associating immunoglobulin-G-rheumatoid factors from two rheumatoid synovia implicates an antigen-driven response. J Exp Med 175:831, 1992
28. Sanz I, Dang H, Takei M, Tala1 N, Capra JD: V,-sequence
of human anti-Sm autoantibody. Evidence that autoantibodies can
be unmutated copies of germline genes. J Immunol 142:883, 1989
29. Sanz I, Capra JD: The genetic origin of human autoantibodies.
J lmmunol 140:3283,1988
30. Denomme CA, Kelton JG, Bell DA: The production of human
monoclonal antiplatelet autoantibodies derived from human lymphocyte of normal origin: Reactivity to DNA, anionic phospholipids and
platelet proteins. Br J Haematol 81:99, 1992