Content uploaded by Masataka Kuwana
Author content
All content in this area was uploaded by Masataka Kuwana on Feb 18, 2015
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
Detection of Circulating B Cells Producing Anti-GPIb
Autoantibodies in Patients with Immune
Thrombocytopenia
Masataka Kuwana
1
*, Yuka Okazaki
1
, Yasuo Ikeda
2
1Division of Rheumatology, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan, 2Faculty of Science and Engineering, Waseda University,
Japan
Abstract
Background:
We previously reported that an enzyme-linked immunospot (ELISPOT) assay for detecting anti-GPIIb/IIIa
antibody-secreting B cells is a sensitive method for identifying patients with immune thrombocytopenia (ITP). Here we
assessed the clinical significance of measuring circulating B cells producing antibodies to GPIb, another major platelet
autoantigen.
Methods:
Anti-GPIb and anti-GPIIb/IIIa antibody-producing B cells were simultaneously measured using ELISPOT assays in
32 healthy controls and 226 consecutive thrombocytopenic patients, including 114 with primary ITP, 25 with systemic lupus
erythematosus (SLE), 30 with liver cirrhosis, 39 with post-hematopoietic stem cell transplantation (post-HSCT), and 18 non-
ITP controls (aplastic anemia and myelodysplastic syndrome).
Results:
There were significantly more circulating anti-GPIb and anti-GPIIb/IIIa antibody-producing B cells in primary ITP,
SLE, liver cirrhosis, and post-HSCT patients than in healthy controls (P,0.05 for all comparisons). For diagnosing primary ITP,
the anti-GPIb ELISPOT assay had 43% sensitivity and 89% specificity, whereas the anti-GPIIb/IIIa ELISPOT assay had 86%
sensitivity and 83% specificity. When two tests were combined, the sensitivity was slightly improved to 90% without a
reduction in specificity. In primary ITP patients, the anti-GPIb antibody response was associated with a low platelet count,
lack of Helicobacter pylori infection, positive anti-nuclear antibody, and poor therapeutic response to intravenous
immunoglobulin.
Conclusion:
The ELISPOT assay for detecting anti-GPIb antibody-secreting B cells is useful for identifying patients with ITP,
but its utility for diagnosing ITP is inferior to the anti-GPIIb/IIIa ELISPOT assay. Nevertheless, detection of the anti-GPIb
antibody response is useful for subtyping patients with primary ITP.
Citation: Kuwana M, Okazaki Y, Ikeda Y (2014) Detection of Circulating B Cells Producing Anti-GPIb Autoantibodies in Patients with Immune
Thrombocytopenia. PLoS ONE 9(1): e86943. doi:10.1371/journal.pone.0086943
Editor: Frederic Rieux-Laucat, Pavillon Kirmisson, France
Received August 29, 2013; Accepted December 17, 2013; Published January 22, 2014
Copyright: ß2014 Kuwana et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by a research grant on intractable diseases by the Japanese Ministry of Health, Welfare and Labor. The funders had no role in
study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: Masataka Kuwana, the corresponding author, currently serves as an Academic Editor of PLOS ONE. This does not alter the authors’
adherence to all the PLOS ONE policies on sharing data and materials.
* E-mail: kuwanam@z5.keio.jp
Introduction
Immune thrombocytopenia (ITP) is an autoimmune disease in
which accelerated platelet consumption and impaired platelet
production are mediated primarily by IgG anti-platelet autoanti-
bodies [1]. This condition is seen in patients with various diseases,
including systemic lupus erythematosus (SLE) and human
immunodeficiency virus infection. It can also occur without an
underlying disease, which is known as primary ITP. The
production of IgG autoantibodies to platelet surface glycoproteins,
such as GPIIb/IIIa and GPIb, is the hallmark of the disease [2].
Several antigen-specific assays for detecting platelet-associated
anti-GPIIb/IIIa and anti-GPIb antibodies are reported to be
useful for identifying patients with ITP [3–5]. However, no
laboratory test for detecting platelet antigen-specific antibodies is
used widely in clinical settings, because these assays require
complicated procedures such as platelet solubilization, the use of
commercially unavailable monoclonal antibodies, and a relatively
large blood sample.
We previously developed an enzyme-linked immunospot (ELI-
SPOT) assay for detecting IgG anti-GPIIb/IIIa antibody-secreting
B cells in the circulation and spleen of patients with primary ITP
[6]. We subsequently showed that the detection of circulating anti-
GPIIb/IIIa antibody-producing B cells is a sensitive, specific, and
convenient method for evaluating the presence or absence of an
anti-platelet autoantibody response [7]. The anti-GPIIb/IIIa
antibody response is very common in patients with primary ITP
as well as those with various forms of secondary ITP, including
thrombocytopenia associated with SLE [8], liver cirrhosis with or
without hypersplenism [9], and post-hematopoietic stem-cell
transplantation (post-HSCT) [10]. These findings led us to
propose preliminary diagnostic criteria for ITP based on a
PLOS ONE | www.plosone.org 1 January 2014 | Volume 9 | Issue 1 | e86943
combination of ITP-associated laboratory findings, including
circulating anti-GPIIb/IIIa antibody-producing B cells, reticulated
platelets, and thrombopoietin [11]. The ELISPOT assay has
several advantages over assays that detect platelet antigen-specific
antibodies, i.e., the results are not influenced by the binding of the
antibodies to platelet surfaces and only a small blood sample
(,3 mL) is required. However, the anti-GPIIb/IIIa antibody
response was not detectable in a small proportion (,20%) of ITP
patients, even if the sensitive ELISPOT assay was used. Thus,
adding a concomitant measurement of B cells producing
antibodies to another major platelet autoantigen, GPIb, may
increase the assay’s sensitivity to the anti-platelet autoantibody
response in patients with ITP. In this study, we established an
ELISPOT assay for detecting anti-GPIb antibody-producing B
cells and evaluated its potential usefulness for the diagnosis, disease
subtyping, and assessment of the anti-platelet autoantibody profiles
in patients with primary ITP and a various forms of secondary
ITP.
Materials and Methods
Subjects
This study included 114 consecutive patients with primary ITP
whose peripheral blood samples had been sent to an autoimmune
laboratory at Keio University Hospital between April 2003 and
March 2005. Eighteen patients were also included in a multicenter
prospective study for verification of our preliminary diagnostic
criteria for ITP [11]. The inclusion criteria were: (i) clinical
diagnosis of primary ITP; (ii) thrombocytopenia (platelet count
#50610
9
/L); (iii) no previous treatment with corticosteroids or
immunosuppressants; and (iv) availability of detailed clinical
information for at least one year after the diagnosis. The clinical
diagnosis of ITP was made by one of the authors (YI) on the basis
of clinical history, physical examination, complete blood test, and
bone marrow findings if available, according to the guidelines
proposed by the American Society of Hematology [12]. The final
diagnosis was re-evaluated, taking into account the clinical course
of the disease over at least one year, especially the therapeutic
responses to corticosteroids, splenectomy, and eradication of
Helicobacter pylori (H. pylori). YI was blinded to the results of the
anti-GPIIb/IIIa and anti-GPIb antibody-producing B cell assays,
so the clinical diagnosis of primary ITP was not influenced by
these laboratory findings. Patients with primary ITP were
classified as having newly diagnosed, persistent, or chronic ITP,
as described previously [13].
Additional thrombocytopenic patients with underlying diseases
that could potentially cause secondary ITP or non-ITP thrombo-
cytopenia were selected from consecutive patients whose periph-
eral blood samples had been sent to the autoimmune laboratory
during the same period, based on the definitive diagnosis of
underlying diseases/conditions and platelet count #50610
9
/L.
SLE and liver cirrhosis were diagnosed according to the published
criteria [14,15]. HSCT recipients were selected based on a lack of
sustained anemia or leukopenia, and no apparent cause for
thrombocytopenia, such as engraftment failure, recurrence of the
underlying hematologic malignancy, microangiopathy, or drugs
[10]. To minimize the potential influence of procedure-related
complications, we selected patients who had survived for
.100 days after HSCT. Patients with aplastic anemia or
myelodysplastic syndrome (MDS) were also enrolled as a non-
ITP disease control. Diagnosis of aplastic anemia and MDS was
based principally on bone marrow findings and cytogenetic
analysis [16,17]. Thirty-two healthy individuals were also included
as a control. All samples were obtained after the subjects gave their
written informed consent, as approved by the ethical committee of
Keio University School of Medicine (Application number 2010-
031-2).
Therapeutic Response
A therapeutic response to intravenous immunoglobulin (IVIG)
was defined as a platelet count .100610
9
/L at one week,
respectively [8], while a response to H. pylori eradication or
corticosteroids (.0.5 mg/kg prednisolone in combination with or
without IVIG) was defined as a platelet count .100610
9
/L at 24
weeks, respectively [18]. We used the published definition for
therapeutic response to splenectomy [19], but complete and
partial responses were combined. That is, a therapeutic response
was defined as a platelet count of $50610
9
/L for 30 days or
longer after splenectomy, with or without other treatment.
H. pylori Infection
H. pylori infection was evaluated with a
13
C urea breath test
using a UBiT tablet (Otsuka Assay, Tokyo, Japan), the detection of
serum IgG anti-H. pylori antibodies using a commercially available
kit (Kyowa Medex Company, Tokyo, Japan), and the detection of
H. pylori antigen in stool samples using ImmunoCardHHpSTH
(Meridian Bioscience, Cincinnati, OH). Patients positive for the
urea breath test plus at least one additional test were regarded as
H. pylori-positive [18].
Antinuclear Antibody (ANA)
ANA was measured by indirect immunofluorescence using
commercially available HEp-2 slides (MBL, Nagoya, Japan) as the
substrate. A positive result was determined as a significant signal
using two different cut-off levels: serum samples diluted 1:40 and
1:160.
Measurement of IgG Anti-GPIIb/IIIa and Anti-GPIb
Antibody-producing B Cells
B cells producing IgG anti-GPIIb/IIIa antibodies were mea-
sured using the ELISPOT assay as described previously [6,7].
Briefly, a polyvinylidene difluoride-bottomed 96-well microplate
(Millipore, Bedford, MA) was activated by incubation with ethanol
(.99.5%) at room temperature for 10 minutes. After extensive
wash with phosphate-buffered saline (PBS) containing 0.5 mM
CaCl
2
(PBS-Ca), the microplates were coated with affinity-purified
human GPIIb/IIIa (purity .80%; Enzyme Research Laborato-
ries, Swansea, UK) dissolved in PBS-Ca at a concentration of
30 mg/mL over night at 4uC. Then, the plates were washed three
times with PBS-Ca, and were subsequently blocked with 1%
bovine serum albumin (Sigma-Aldrich, St. Louis, MO) in PBS-Ca
at room temperature for one hour. Peripheral blood mononuclear
cells (PBMCs), isolated from heparinized peripheral blood by
Lymphoprep (Fresenius Kabi Norge AS, Halden, Norway) density
gradient centrifugation, were re-suspended in RPMI1640 con-
taining 10% fetal bovine serum (Life Technologies, Carlsbad, CA),
and were pipetted into the wells (10
5
/well) and cultured at 37uC
with 5% CO
2
for 4 hours. After washing away the cells with PBS-
Ca containing 0.05% Tween 20, the membranes were incubated
with alkaline phosphatase-conjugated goat anti-human IgG (ICN/
Cappel, Aurora, OH) diluted at 1:1,000 in PBS-Ca at room
temperature for 2 hours, followed by wash two time each with
PBS-Ca with 0.05% Tween 20 and PBS-Ca. Finally, anti-GPIIb/
IIIa antibodies that bound to the membrane were visualized as
spots by incubation with nitro blue tetrazolium (Sigma-Aldrich;
300 mg/mL)/5-bromo-4-chloro-3-indolyl phosphate (Sigma-Al-
drich; 60 mg/mL) in a buffer consisting of 100 mM Tris-HCl
Anti-GPIb ELISPOT
PLOS ONE | www.plosone.org 2 January 2014 | Volume 9 | Issue 1 | e86943
(pH9.5), 100 mM NaCl, 50 mM MgCl
2
at room temperature for
20 minuts. B cells producing IgG anti-GPIb antibodies were also
measured by ELISPOT assay, in which a recombinant GPIba
fragment was used instead of GPIIb/IIIa as the antigen. The
recombinant GPIbafragment, which covered the entire von
Willebrand factor-binding site (residues 1 to 302), was expressed in
Chinese hamster ovary cells [20]. For the anti-GPIb antibody
ELISPOT assay, PBS was used instead of PBS-Ca in the entire
protocol. The plates coated with bovine serum albumin in the
blocking buffer in the absence of GPIIb/IIIa or GPIb were used as
control for the ELISPOT assay. Each assay was conducted in 5
independent wells, and the results represented the mean of the 5
values. The frequency of circulating anti-GPIIb/IIIa or anti-GPIb
antibody-producing B cells was calculated as the number per 10
5
PBMCs. The cut-off value for anti-GPIIb/IIIa antibody-produc-
ing cells was defined as 2.0 per 10
5
PBMCs [7]. The cut-off value
for anti-GPIb antibody-producing cells was set at 5 standard
deviations above the mean value from healthy controls.
Statistical Analysis
All continuous variables were expressed as the mean 6standard
deviation (SD). Comparisons between two groups were tested for
statistical significance using the Mann-Whitney test. Differences in
frequency between two groups were compared using the Chi-
square test or Fisher’s exact test, when applicable. The correlation
coefficient (r) was determined using a single-regression model.
Results
Patient Characteristics
This study enrolled a total of 226 thrombocytopenic patients.
They were composed of 114 with primary ITP, 25 with SLE,
30 with liver cirrhosis, 39 with post-HSCT, and 18 non-ITP
controls, including 4 with aplastic anemia and 14 with myelodys-
plastic syndrome. Table 1 summarizes the sex, age at examination,
and platelet count of thrombocytopenia patients and healthy
controls. Forty-eight patients (42%) with primary ITP were
categorized as having newly diagnosed ITP, while the remaining
patients had persistent or chronic ITP. The etiologies of liver
cirrhosis included hepatitis B virus infection in 5, hepatitis C virus
infection in 21, and alcohol in 4. Of the post-HSCT patients,
37 had received bone marrow transplantation while 2 had
received peripheral blood stem cell transplantation. Compared
with patients with primary ITP, patients with SLE were
predominantly female (P= 0.02) and those with liver cirrhosis
and MDS were older (P= 0.001 and P= 0.03, respectively). There
was no difference in platelet count among the thrombocytopenic
patient groups. The mean follow-up period in patients with
primary ITP was 49626 months.
Detection of IgG Anti-GPIIb/IIIa and Anti-GPIb Antibody-
producing B Cells
Circulating B cells producing IgG anti-GPIIb/IIIa and anti-
GPIb antibodies were simultaneously measured in patients with
primary ITP, various thrombocytopenic conditions, and healthy
controls (Figure 1). No clear spot was detected in the control plates
coated with bovine serum albumin alone. There were significantly
more circulating anti-GPIIb/IIIa antibody-producing B cells in
patients with primary ITP, SLE, liver cirrhosis, and post-HSCT
than in healthy controls (5.464.7, 6.066.4, 10.065.8, and
6.368.3 versus 0.360.4; P,0.05 for all comparisons). In contrast,
there was no difference in anti-GPIIb/IIIa antibody-producing B
cells between the non-ITP disease controls, including aplastic
anemia and MDS, and healthy controls. Among ITP-related
conditions, patients with liver cirrhosis had a greater frequency of
anti-GPIIb/IIIa antibody-producing B cells than did those with
primary ITP, SLE, or post-HSCT (P,0.05 for all comparisons).
Similarly, there were significantly more anti-GPIb antibody-
producing B cells in patients with primary ITP, SLE, liver
cirrhosis, and post-HSCT than in healthy controls (3.063.3,
10.5625.6, 4.865.2, and 3.466.2 versus 0.460.4; P,0.01 for all
comparisons). Again, there was no difference between the non-ITP
disease controls and healthy controls. The frequency of anti-GPIb
antibody-producing B cells tended to be higher in SLE patients
than in those with other ITP-related conditions, but the difference
was not statistically significant.
The circulating frequencies of anti-GPIb and anti-GPIIb/IIIa
antibody-producing B cells were correlated with each other in
patients with primary ITP, SLE, liver cirrhosis, and post-HSCT
(P,0.0003 for all correlations) (Figure 2). Based on the slopes of
the fitted lines obtained by the single regression model, the anti-
GPIIb/IIIa antibody-producing cells exceeded the anti-GPIb
antibody-producing B cells in patients with primary ITP, liver
cirrhosis, and post-HSCT (slope ,1), whereas the anti-GPIb
antibody-producing B cells predominated in SLE patients (slope
.1).
Diagnostic Utility of Anti-GPIIb/IIIa and Anti-GPIb
Antibody-producing B Cells
To evaluate the diagnostic utility of the anti-GPIIb/IIIa and
anti-GPIb ELISPOT assays, the results of these tests were judged
as positive or negative based on being above or below a defined
cut-off level. We used 2.0 per 10
5
PBMCs as the cut-off value for
anti-GPIIb/IIIa antibody-producing B cells, which was deter-
mined in our previous study [7] and 2.4 per 10
5
PBMCs for
circulating anti-GPIb antibody-producing B cells, which was 5
standard deviations above the mean value obtained from healthy
controls. The positive frequencies of circulating anti-GPIIb/IIIa
and anti-GPIb antibody-producing B cells, and their combination
in patients with primary ITP, various thrombocytopenic condi-
tions, and healthy controls are summarized in Table 2. Anti-
GPIIb/IIIa antibody-producing cells were detected in 86% of the
patients with primary ITP, and in 76%, 97%, and 62% of the
patients with SLE, liver cirrhosis, and post-HSCT, respectively. In
contrast, the percentages of patients with a positive frequency of
anti-GPIb antibody-producing B cells were lower (38–50%) than
those of anti-GPIIb/IIIa antibody-producing cells. These anti-
body-producing cells were infrequently detected in patients with
aplastic anemia or MDS. Of 16 patients with primary ITP who
were negative for the anti-GPIIb/IIIa antibody-producing cells, 5
(31%) were positive for the anti-GPIb antibody-producing cells.
Three (50%) out of 6 SLE patients with the negative anti-GPIIb/
IIIa ELISPOT result were positive for the anti-GPIb ELISPOT
assay, but none of the patients with liver cirrhosis or post-HSCT
who showed the negative anti-GPIIb/IIIa ELISPOT result were
positive for the anti-GPIb ELISPOT assay. When the results for
anti-GPIIb/IIIa and anti-GPIb antibody-producing B cells were
combined, the positive frequency was slightly increased in patients
with primary ITP and SLE, but not in those with liver cirrhosis
and post-HSCT, because in the latter cases the anti-GPIb
antibody-producing B cells always coexisted with anti-GPIIb/IIIa
antibody-producing ones.
We then focused on the utility of anti-GPIIb/IIIa and anti-
GPIb antibody-producing B cell measurement for the diagnosis of
primary ITP. This analysis included 114 patients with primary
ITP and 18 with non-ITP thrombocytopenia, including aplastic
anemia and MDS. The anti-GPIIb/IIIa antibody-producing B cell
measurement had a sensitivity of 86%, specificity of 83%, positive
Anti-GPIb ELISPOT
PLOS ONE | www.plosone.org 3 January 2014 | Volume 9 | Issue 1 | e86943
predictive value of 98%, and negative predictive value of 50%. In
contrast, the sensitivity of the anti-GPIb antibody-producing cells
was only 53%, while the specificity was 89% and positive and
negative predictive values were 86% and 20%, respectively. When
the two tests were combined, the sensitivity was slightly improved
to 90% without effectively reducing the specificity or positive
predictive value of the anti-GPIIb/IIIa ELISPOT assay alone
(83%, 97%, respectively). When the same analysis was performed
in patients with SLE, the sensitivity was improved from 76% in
case of the anti-GPIIb/IIIa ELISPOT assay alone to 88% in case
of combining the two tests.
Clinical Characteristics Associated with Anti-GPIIb/IIIa
and Anti-GPIb Antibody-producing B Cells in Patients
with Primary ITP
Patients with primary ITP were stratified into two groups based
on the presence or absence of anti-GPIIb/IIIa or anti-GPIb
antibody-producing B cells (Table 3). Newly diagnosed ITP was
less common in patients with anti-GPIIb/IIIa antibody-producing
B cells, than in those without. The positive anti-GPIb ELISPOT
assay result was associated with a low platelet count, lack of H.
pylori infection, and positive ANA, independent of the cut-off
levels. Therapeutic responses to H. pylori eradication, IVIG, and
splenectomy tended to be worse in patients with a positive anti-
GPIb ELISPOT assay than in those without, but only the
difference in the response to IVIG reached statistical significance.
On the other hand, there were no differences in clinical
Table 1. Demographic features and platelet count of subjects enrolled in this study.
Number Sex (% male) Age at examination (years) Platelet count (610
9
/L)
Primary ITP 114 40% 49.6617.1 28.1611.6
SLE 25 12% 43.6614.6 28.6613.3
Liver cirrhosis 30 53% 63.369.1 35.2611.5
Post-HSCT 39 59% 37.6610.6 31.5611.3
Aplastic anemia 4 25% 46.3623.9 24.5616.2
MDS 14 57% 60.4617.5 26.9613.3
Healthy controls 32 50% 44.1612.2 252.4656.8*
ND: not determined.
*Data were derived from 16 healthy donors.
doi:10.1371/journal.pone.0086943.t001
Figure 1. Anti-GPIIb/IIIa and anti-GPIb antibody-producing B cells in the circulation of patients with various thrombocytopenic
conditions and healthy controls. Cut-off values for anti-GPIIb/IIIa and anti-GPIb antibody-producing B cells were 2.0 and 2.4 per 10
5
PBMCs,
respectively. Bars indicate the mean, and asterisks indicate statistical significance (P,0.05).
doi:10.1371/journal.pone.0086943.g001
Anti-GPIb ELISPOT
PLOS ONE | www.plosone.org 4 January 2014 | Volume 9 | Issue 1 | e86943
characteristics except the ITP classification between patients with
and without anti-GPIIb/IIIa antibody-producing B cells.
Discussion
In this study, we successfully developed an ELISPOT assay for
detecting anti-GPIb antibody-secreting B cells, by applying the
principles of our previously developed anti-GPIIb/IIIa antibody-
producing B cell measurement. Anti-GPIb antibody-secreting B
cells were detected in the circulation of patients with primary ITP
as well as conditions that potentially cause secondary ITP, but
were infrequently found in patients with aplastic anemia or MDS.
Thus, the anti-GPIb ELISPOT assay is useful for identifying
patients with ITP, but its sensitivity was much inferior to that of
the anti-GPIIb/IIIa ELISPOT assay, indicating that detection of
anti-GPIb antibody-producing cells could not replace the anti-
GPIIb/IIIa assessment in ITP diagnosis. In addition, concomitant
measurement of anti-GPIb and anti-GPIIb/IIIa antibody-produc-
ing B cells had limited utility: a slight increase in sensitivity only for
primary ITP and SLE. These findings indicate that, rather
Figure 2. Correlations between circulating anti-GPIb and anti-GPIIb/IIIa antibody-producing B cells in patients with primary ITP,
SLE, liver cirrhosis, and post-HSCT.
doi:10.1371/journal.pone.0086943.g002
Table 2. Positive frequencies of circulating anti-GPIIb/IIIa and anti-GPIb antibody-producing B cells, and their combination in
patients with primary ITP, various thrombocytopenic conditions, and healthy controls.
Primary ITP
(n = 114) SLE (n = 25)
Liver cirrhosis
(n = 30)
Post-
HSCT(n = 39)
Aplastic anemia/
MDS (n = 18)
Healthy
controls
(n = 32)
Anti-GPIIb/IIIa antibody-producing B cells alone 86% 76% 97% 62% 17% 0%
Anti-GPIb antibody-producing B cells alone 43% 40% 50% 38% 11% 0%
Anti-GPIIb/IIIa antibody-producing B cells AND
anti-GPIb antibody-producing B cells
38% 28% 50% 38% 11% 0%
Anti-GPIIb/IIIa antibody-producing B cells OR
anti-GPIb antibody-producing B cells
90% 88% 97% 62% 17% 0%
SLE, liver cirrhosis, and post-HSCT are conditions potentially causing secondary ITP, whereas aplastic anemia and MDS are non-ITP disease controls.
doi:10.1371/journal.pone.0086943.t002
Anti-GPIb ELISPOT
PLOS ONE | www.plosone.org 5 January 2014 | Volume 9 | Issue 1 | e86943
disappointingly, the additional measurement of anti-GPIb anti-
body-producing B cells on top of the anti-GPIIb/IIIa ELISPOT
assay does not improve the diagnostic accuracy for patients
suspected of having ITP. However, it may be worth measuring the
anti-GPIb ELISPOT assay in patients who are suspected to have
primary ITP or secondary ITP in association with SLE, but are
negative for anti-GPIIb/IIIa antibody-producing cells, because
there is a .30% chance for obtaining the positive result.
Several laboratories have reported antigen-specific assays, such
as the monoclonal antibody-specific immobilization of platelet
antigen (MAIPA) assay, for detecting autoantibodies to GPIIb/
IIIa and GPIb, which are either bound to platelet surfaces or
present in plasma, although an international study comparing
these antigen-specific assays revealed that good inter-laboratory
agreement was obtained only when the platelet-associated
antibodies were measured [21]. In this regard, Warner and
colleagues reported that an antigen-specific assay detecting
platelet-associated anti-GPIIb/IIIa antibodies had a sensitivity of
57% and a specificity of 96% for the diagnosis of primary ITP, and
the additional measurement of anti-GPIb antibodies increased the
diagnostic sensitivity to 66%, and retained a specificity of 92% [3].
In another report using a prospective cohort of thrombocytopenic
patients, platelet-associated anti-GPIIb/IIIa and anti-GPIb anti-
bodies detected by direct MAIPA were present in 49% of 93
patients with ITP, including 74 with the primary form, and in only
22% of 54 patients with non-ITP thrombocytopenia [4]. The
platelet-associated antibodies to GPIIb/IIIa (88%) were more
frequently directed than to GPIb (52%), while 40% of patients had
concomitant antibodies to both of these platelet glycoproteins.
Finally, McMillan et al examined platelet-associated anti-GPIIb/
IIIa and anti-GPIb antibodies in 282 patients with primary ITP,
and found that the majority of patient samples contained platelet-
associated antibodies recognizing GPIIb/IIIa alone (52%); fewer
reacted to GPIb alone (12%) or to both complexes (15%) [5]. Our
findings, obtained by measuring the anti-GPIIb/IIIa and anti-
GPIb antibody-producing circulating B cells, were generally
concordant with these results from assays detecting specific
platelet-associated antibodies: GPIIb/IIIa antibodies were pre-
dominantly recognized, while anti-GPIb antibody measurement
contributed minimally to the diagnosis of primary ITP. The
ELISPOT assays appear to be more sensitive than the platelet-
associated antigen-specific assays (90% versus 49–66%), but
prospective studies comparing the ELISPOT assays with other
anti-platelet autoantibody detection tests are necessary to confirm
this.
The main reason for the low utility of anti-GPIb antibody-
producing B cells for the diagnosis of ITP is that circulating anti-
GPIIb/IIIa and anti-GPIb antibody-producing B cells coexist in
the majority of patients with ITP. Only a small number of patients
had anti-GPIb antibody-producing B cells alone. Therefore, in
routine clinical settings, the anti-GPIIb/IIIa ELISPOT assay
appears to be sufficient for the diagnosis of ITP. Interestingly, the
levels of anti-GPIb and anti-GPIIb/IIIa antibody-producing B
cells in the circulation were correlated with each other in ITP
patients, irrespective of whether the diagnosis was primary ITP or
one of the secondary forms. These findings indicate that the
autoimmune response in the majority of ITP patients targets
multiple platelet glycoproteins, which might be a consequence of
‘‘epitope spreading’’ [2]. In this regard, we have proposed a
‘‘pathogenic loop’’ model for the ongoing anti-platelet autoanti-
body response in ITP patients [22]. Namely, macrophages in the
reticuloendothelial system (spleen in the majority of the patients)
capture opsonized platelets, and activate autoreactive T helper
cells that stimulate the B cells to proliferate [23], differentiate into
plasma cells [24], and produce anti-platelet autoantibodies, which
in turn bind to circulating platelets. The continuous destruction of
platelets in the reticuloendothelial system would allow the
processing and presentation of a whole panel of platelet antigens
by macrophages, some of which could elicit additional autoreac-
tive T cell responses, resulting in the production of autoantibodies
against other platelet glycoproteins.
In primary ITP and various forms of secondary ITP, SLE was
unique in having a predominant anti-GPIb antibody-producing B
cell response. SLE is a systemic autoimmune disease characterized
by a loss of tolerance to nuclear and other self-antigens, a
production of pathogenic autoantibodies, and damage to multiple
organ systems [25]. Taken together with the association between
anti-GPIb antibody-producing B cells and the production of
ANAs, even in patients with primary ITP, the anti-GPIb
autoantibody response might be linked to systemic autoimmunity.
Table 3. Clinical findings in patients with primary ITP, stratified by the presence or absence of circulating anti-GPIb or anti-GPIIb/
IIIa antibody-producing B cells.
Anti-GPIIb/IIIa antibody-producing B cells Anti-GPIb antibody-producing B cells
Present (n = 98) Absent (n = 16)
P
Present (n = 49) Absent (n = 65)
P
Sex (% female) 58% 69% 0.60 65% 55% 0.28
Age at examination (years) 50.3617.5 45.4614.1 0.29 50.0617.6 49.3616.9 0.85
Newly diagnosed ITP (%) 37% 75% 0.009 47% 38% 0.47
Platelet count (x 10
9
/L) 27.5611.5 31.7612.1 0.19 19.869.4 34.468.8 ,0.0001
H. pylori infection 28% 19% 0.66 14% 35% 0.01
Positive ANA ($1:40) 26% 44% 0.23 51% 23% 0.002
Positive ANA ($1:160) 17% 13% 0.83 24% 5% 0.004
Therapeutic response
H. pylori eradication 62% (n =26) 100% (n = 3) 0.50 33% (n = 6) 74% (n =23) 0.16
IVIG 65% (n = 40) 56% (n = 9) 0.60 46% (n = 24) 80% (n = 25) 0.03
Corticosteroids 21% (n = 53) 17% (n = 12) 0.91 24% (n = 33) 15% (n = 33) 0.54
Splenectomy 76% (n = 37) 75% (n = 8) 0.97 64% (n = 22) 86% (n = 23) 0.14
doi:10.1371/journal.pone.0086943.t003
Anti-GPIb ELISPOT
PLOS ONE | www.plosone.org 6 January 2014 | Volume 9 | Issue 1 | e86943
Because a significant proportion of SLE patients had anti-GPIb
antibody-producing B cells in the absence of anti-GPIIb/IIIa
antibody-producing B cells, measurement of the anti-GPIb in
addition to anti-GPIIb/IIIa antibody-producing B cells may have
some merit for accurately identifying secondary ITP in patients
with SLE and thrombocytopenia, although the number of patients
analyzed in this study was too small to draw a firm conclusion.
Much effort has been made to identify clinical associations of
individual anti-platelet glycoprotein antibodies, but the clinical
significance of such antibodies remains uncertain. In patients with
primary ITP, the presence of platelet-associated anti-GPIb
antibodies was shown to be associated with a lower platelet count
[26,27] and inadequate responses to corticosteroids [26] and IVIG
[28]. Our results were consistent with these previous observations,
including the low platelet count and poor responses to therapeutic
interventions, especially to IVIG. In this regard, some monoclonal
antibodies against GPIb are known to induce platelet activation,
which may lead to accelerated platelet destruction independent of
the Fccreceptor-mediated process in ITP patients [29]. We
additionally found correlations between anti-GPIb antibody-
producing B cells and a low prevalence of H. pylori infection or a
high frequency of positive ANA. These findings indicate that there
may be a relatively homogeneous subset of primary ITP cases
defined by the anti-GPIb antibody response, and characterized by
severe thrombocytopenia, the absence of H. pylori infection, a
positive ANA, and a poor therapeutic response.
In summary, our ELISPOT assay for detecting anti-GPIb
antibody-secreting B cells is useful for identifying patients with
ITP, but its utility for diagnosing ITP is apparently inferior to the
anti-GPIIb/IIIa ELISPOT assay. Nevertheless, detection of the
anti-GPIb antibody response is useful for subtyping patients with
primary ITP and predicting the therapeutic response.
Acknowledgments
We thank Dr. Mitsuru Murata (Department of Laboratory Medicine, Keio
University School of Medicine) for providing the recombinant GPIb
fragment.
Author Contributions
Conceived and designed the experiments: MK YI. Performed the
experiments: MK YO YI. Analyzed the data: MK YI. Wrote the paper:
MK YO YI.
References
1. Stasi R (2001) Immune thrombocytopenia: pathophysiologic and clinical update.
Semin Thromb Hemost 38: 454–462.
2. McMillan R (2007) The pathogenesis of chronic immune thrombocytopenic
purpura. Semin Hematol 44: S3–S11.
3. Warner MN, Moore JC, Warkentin TE, Santos AV, Kelton JG (1999) A
prospective study of protein-specific assays used to investigate idiopathic
thrombocytopenic purpura. Br J Haematol 104: 442–447.
4. Brighton TA, Evans S, Castaldi PA, Chesterman CN, Chong BH (1996)
Prospective evaluation of the clinical usefulness of an antigen-specific assay
(MAIPA) in idiopathic thrombocytopenic purpura and other immune throm-
bocytopenias. Blood 88: 194–201.
5. McMillan R, Wang L, Tani P (2003) Prospective evaluation of the immunobead
assay for the diagnosis of adult chronic immune thrombocytopenic purpura
(ITP). J Thomb Haemost 1: 485–491.
6. Kuwana M, Okazaki Y, Kaburaki J, Kawakami Y, Ikeda Y (2002) Spleen is a
primary site for activation of glycoprotein IIb-IIIa-reactive T and B cells in
patients with immune thrombocytopenic purpura. J Immunol 168: 3675–3682.
7. Kuwana M, Okazaki Y, Kaburaki J, Ikeda Y (2003) Detection of circulating B
cells secreting platelet-specific autoantibody is a sensitive and specific test for the
diagnosis of autoimmune thrombocytopenia. Am J Med 114: 322–325.
8. Kuwana M, Kaburaki J, Okazaki Y, Miyazaki H, Ikeda Y (2006) Two types of
autoantibody-medi ated thrombocytopenia in patients with systemi c lupus
erythematosus. Rheumatology 45: 851–854.
9. Kajihara M, Kato S, Okazaki Y, kawakami Y, Ishii H, et al (2003) A role of
autoantibody-mediated platelet destruction in thrombocytopenia in patients with
cirrhosis. Hepatology 37: 1267–1276.
10. Yamazaki R, Kuwana M, Mori T, Okazaki Y, Kawakami Y, et al (2006)
Prolonged thrombocytopenia after allogeneic haematopoietic stem cell trans-
plantation: Associations with impaired platelet production and increased platelet
turnover. Bone Marrow Transplant 38: 377–384.
11. Kuwana M, Kurata Y, Fujimura K, Fujisawa K, Wada H, et al (2006)
Preliminary laboratory-based diagnostic criteria for immune thrombocytopenic
purpura: Evaluation by multi-center prospective study. J Thromb Haemost 4:
1936–1943.
12. George JN, Woolf SH, Raskob GE, Wasser JS, Aledort LM, et al (1996)
Idiopathic thrombocytopenic purpura: A practice guideline developed by
explicit methods for the American Society of Hematology. Blood 88: 3–40.
13. Rodeghiero F, Stasi R, Gernsheimer T, Michel M, Provan D, et al (2009)
Standardization of terminology, definitions and outcome criteria in immune
thrombocytopenic purpura of adults and children: report from an international
working group. Blood 113: 2386–2393.
14. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, et al (1982) The 1982
revised criteria for the classification of systemic lupus erythematosus. Arthritis
Rheum 25: 1271–1277.
15. Erlinger S, Benhamou JP. Cirrhosis: clinical aspects. In: Bircher J, Benhamou
JP, McIntyre N, Rizetto M, Rodes J, eds. Oxford textbook of clinical hepatology.
Volume 1. 2nd ed. Oxford: Oxford University Press, 1999: 629–41.
16. Guinan EC. (1997) Clinical aspects of aplastic anemia. Hematol Oncol Clin
North Am 11: 1025–1044.
17. Greenberg P, Cox C, LeBeau MM, Fanaux P, Morel P, et al (1997)
International scoring system for evaluating prognosis in myelodysplastic
syndromes. Blood 89: 2079–2088.
18. Asahi A, Kuwana M, Suzuki H, Hibi T, Kawakami Y, et al (2006) Effects of a
Helicobacter pylori eradication regimen on anti-platelet autoantibody response in
infected and uninfected patients with idiopathic thrombocytopenic purpura.
Haematologica 91: 1436–1437.
19. Kojouri K, Vesely SK, Terrell DR, George JN (2004) Splenectomy for adult
patients with idiopathic thrombocytopenic purpura: a systematic review to assess
long-term plat elet count resp onses, prediction of response, a nd surgical
complications. Blood 104: 2623–2634.
20. Murata M, Ware J, Ruggeri ZM (1991) Site-directed mutagenesis of a soluble
recombinant fragment of platelet glycoprotein Ibademonstrating negatively
charged residues involved in von Willebrand factor binding. J Biol Chem 266:
15474–15480.
21. Berchtold P, Mu¨ller D, Beardsley D, Fujisawa K, Kaplan C, et al (1997)
International study to compare antigen-specific methods used for the
measurement of antiplatelet autoantibodies. Br J Haematol 96: 477–483.
22. Kuwana M, Okazaki Y, Ikeda Y (2009) Splenic macrophages mainta in the anti-
platelet autoimmune response via uptake of opsonized platelets in patients with
immune thrombocytopenic purpura. J Thromb Haemost 7: 322–329.
23. Daridon C, Loddenkemper C, Spieckermann S, Ku¨hl AA, Salama A, et al
(2011) Splenic proliferative lymphoid nodules distinct from germinal centers are
sites of autoantigen stimulation in immune thrombocytopenia. Blood 120: 5021–
5031.
24. Mahe´vas M, Patin P, Huetz F, Descatoire M, Cagnard N, et al (2013) B cell
depletion in immune thrombocytopenia reveals splenic long-lived plasma cells.
J Clin Invest 123: 432–442.
25. Liu Z, Davidson A (2012) Taming lupus-a new understanding of pathogenesis is
leading to clinical advances. Nat Med 18: 871–882.
26. Nomura S, Yanabu M, Soga T, Kido H, Fukuroi T, et al (1991) Analysis of
idiopathic thrombocytopenic purpura patients with antiglycoprotein IIb/IIIa or
Ib autoantibodies. Acta Haematol 86: 25–30.
27. Hou M, Stockelberg D, Kutti J, Wadenvik H (1995) Antibodies against platelet
GPIb/IX, GPIIb/IIIa, and other platelet antigens in chronic idiopathic
thrombocytopenic purpura. Eur J Haematol 55: 307–314.
28. Go RS, Johnston KL, Bruden KC (2007) The association between platelet
autoantibody specificity and response to intravenous immunoglobulin G in the
treatment of patients with immune thrombocytopenia. Haematologica 92: 283–
284.
29. Cauwenberghs N, Schlammadinger A, Vauterin S, Cooper S, Descheemaeker
G, et al (2001) Fc-receptor dependent platelet aggregation induced by
monoclonal antibodies against platelet glycoprotein Ib or von Willebrand factor.
Thromb Haemost 85: 679–685.
Anti-GPIb ELISPOT
PLOS ONE | www.plosone.org 7 January 2014 | Volume 9 | Issue 1 | e86943
