ARTHRITIS & RHEUMATISM
Vol. 58, No. 9, September 2008, pp 2835–2844
© 2008, American College of Rheumatology
Proteomic Analysis in Monocytes of
Antiphospholipid Syndrome Patients
Deregulation of Proteins Related to the Development of Thrombosis
Chary L´ opez-Pedrera,1Maria Jos´ e Cuadrado,2Vanessa Hern´ andez,1Paula Buendía,1
Maria Angeles Aguirre,1Nuria Barbarroja,1Luis Arístides Torres,1Jos´ e Manuel Villalba,3
Francisco Velasco,1and Munther Khamashta2
Objective. Antiphospholipid antibodies (aPL) are
closely related to the development of thrombosis, but the
exact mechanism(s) leading to thrombotic events re-
mains unknown. In this study, using proteomic tech-
niques, we evaluated changes in protein expression of
monocytes from patients with antiphospholipid syn-
drome (APS) related to the pathophysiology of the
Methods. Fifty-one APS patients were included.
They were divided into 2 groups: patients with previous
thrombosis, and patients with recurrent spontaneous
abortion. As controls, we studied patients with throm-
bosis but without aPL, and age- and sex-matched
Results. The proteins that were more significantly
altered among monocytes from APS patients with
thrombosis (annexin I, annexin II, protein disulfide
isomerase, Nedd8, RhoA proteins, and Hsp60) were
functionally related to the induction of a procoagulant
state as well as to autoimmune-related responses. Pro-
teins reported to be connected to recurrent spontaneous
abortion (e.g., fibrinogen and hemoglobin) were also
determined to be significantly deregulated in APS pa-
tients without thrombosis. In vitro treatment with IgG
fractions purified from the plasma of APS patients with
thrombosis changed the pattern of protein expression of
normal monocytes in the same way that was observed in
vivo for monocytes from APS patients with thrombosis.
Conclusion. For the first time, proteomic analysis
has identified novel proteins that may be involved in the
pathogenic mechanisms of APS, thus providing poten-
tial new targets for pathogenesis-based therapies for the
Antiphospholipid syndrome (APS) is an acquired
autoimmune disorder of unknown pathogenesis that is
defined by the association of arterial or venous throm-
bosis and/or pregnancy morbidity in the presence of
antiphospholipid antibodies (aPL), i.e., anticardiolipin
antibodies (aCL) and lupus anticoagulant (LAC) (1).
Antiphospholipid antibodies are a heterogeneous family
of autoantibodies whose origin and role are not fully
understood. Many of these autoantibodies are directed
against phospholipid-binding plasma proteins, such as
?2-glycoprotein I (?2GPI) and prothrombin, or phos-
pholipid–protein complexes, located on the surface of
vascular endothelial cells, platelets, or monocytes (2).
Several nonexclusive mechanisms could explain the in-
volvement of aPL in the pathogenesis of thrombosis in
APS, including the induction of tissue factor (TF) ex-
pression by endothelial cells and monocytes (3,4). Intra-
cellular mechanisms underlying aPL-induced TF gene
and protein expression in endothelial cells and mono-
cytes have also been delineated at a molecular level
Supported by the Junta de Andalucı ´a of Spain (grants
PI0014/06 and PI0042/2007). Dr. Lo ´pez-Pedrera’s work was supported
by a postdoctoral contract from the Fundacio ´n Progreso y Salud, Junta
de Andalucia of Spain.
1Chary Lo ´pez-Pedrera, MSc, PhD, Vanessa Herna ´ndez, MSc,
Paula Buendı ´a, MSc, Maria Angeles Aguirre, PhD, Nuria Barbarroja,
MSc, PhD, Luis Arı ´stides Torres, MSc, Francisco Velasco, PhD:
Hospital Universitario Reina Sofı ´a, Co ´rdoba, Spain;
Cuadrado, MD, PhD, Munther Khamashta, MD, PhD, FRCP: St.
Thomas Hospital, London, UK;3Jose ´ Manuel Villalba, MSc, PhD:
Universidad de Co ´rdoba, Co ´rdoba, Spain.
Address correspondence and reprint requests to Chary
Lo ´pez-Pedrera, MSc, PhD, Unidad de Investigacio ´n, Hospital Univer-
sitario Reina Sofı ´a, Avenida Mene ´ndez Pidal s/n, E-14004 Cordova,
Spain. E-mail: email@example.com.
Submitted for publication November 30, 2007; accepted in
revised form May 12, 2008.
2Maria Jose ´
(5,6). Nevertheless, despite these findings, the precise
pathogenesis of thrombotic diathesis associated with
aPL remains unknown.
Genetic factors related to aPL and APS have
been widely investigated. Many candidate genes, includ-
ing HLA class II haplotype, predispose patients to APS
(7). However, genetic risk factors related to aPL devel-
opment and clinical manifestations of APS in these
patients remain elusive because of the heterogeneity in
antigen specificity and in the pathophysiology of throm-
bosis. Moreover, gene expression analysis does not cover
the entire spectrum of proteins produced by the organ-
ism. Due to posttranscriptional and posttranslational
modifications, as well as genomic rearrangements (in the
case of Ig), the number of proteins greatly exceeds the
number of genes. Therefore, expression analysis does
not allow the identification of changes in small subsets of
Ig or T cell receptors, which may lead to the production
of autoreactive T cells that contribute to the develop-
ment of autoimmune diseases. Thus, the aim of our
study was to investigate changes in the proteomic pat-
terns of monocytic cells that could underlie the patho-
genic mechanisms associated with thrombosis in this
autoimmune disease. These studies have helped identify
critical proteins that might be involved in the pathogen-
esis of APS.
PATIENTS AND METHODS
Patients. Fifty-one patients (42 women [none preg-
nant] and 9 men; mean age 43 years [range 18–74]) who
fulfilled the classification criteria for APS (1) were included in
the study after ethics committee approval was obtained. All
patients provided written informed consent. They were divided
into 2 groups: group 1 consisted of 32 patients with previous
thrombotic events (arterial in 20 patients [62.5%] and venous
in 12 patients [37.5%]), and group 2 consisted of 19 patients
with recurrent spontaneous abortion. We excluded all APS
patients with evidence of an underlying systemic rheumatic
disease or antibodies against double-stranded DNA or extract-
able nuclear antigen. Patients were studied for at least 9
months after their latest thrombotic event or spontaneous
abortion. As controls, we studied 20 patients with verified
thrombosis (8) but without aPL (group 3; venous in 10 patients
[50%] and arterial in 12 patients [58%]), and 15 healthy
subjects (group 4). None of the healthy controls had a history
of autoimmune disease, bleeding disorders, thrombosis, or
miscarriage. None of the patients had protein C, protein S, or
antithrombin deficiency, factor V Leiden, or prothrombin
Patients were tested for the presence of aCL and LAC
(9,10). The results for aCL were expressed in IgG and IgM
phospholipid units (GPL and MPL units, respectively), and
were reported as positive if they were present in medium or
high titers (?40 GPL units) (1). IgG and IgM aCL were
positive in 33 and 28 APS patients, respectively, and LAC was
positive in 36 patients.
All APS patients with thrombosis were being treated
with an oral anticoagulant (dicumarol). APS patients without
thrombosis were taking a low dosage of aspirin (125 mg/day) or
received no treatment. Finally, some patients with thrombosis
but without APS were taking oral anticoagulants and some
received no treatment. No patients were being treated with
immunomodulatory agents, since they had no other underlying
systemic autoimmune disease.
Monocyte isolation. Peripheral venous blood samples
were collected in sterile precooled tubes containing 0.129M
sodium citrate (1/9 [volume/volume]; Becton Dickinson, Mey-
lan, France) and centrifuged immediately at 500g for 10
minutes at 4°C to remove platelets. Monocytes were isolated by
depletion of nonmonocytes, using a commercial kit (Miltenyi
Biotech, Bergisch Gladbach, Germany), which enabled us to
obtain monocytes without activation (6). Briefly, peripheral
blood mononuclear cells (PBMCs) were obtained by density
gradient centrifugation over Ficoll-Paque. PBMCs were de-
pleted from T cells, natural killer cells, B cells, dendritic cells,
and basophils by indirect magnetic labeling with a cocktail of
hapten-conjugated CD3, CD7, CD19, CD45RA, CD56, and
anti-IgE antibodies and microbeads coupled to an antihapten
monoclonal antibody. Magnetically labeled cells were removed
by retention on a column in a magnetic field. The purity of
isolated monocytes was evaluated by staining cell aliquots with
a fluorochrome-conjugated antibody against monocytes (fluo-
rescein isothiocyanate [FITC]–conjugated anti-CD14), and
analyzed by flow cytometry. Using this method, a mean ?
SEM 92.7 ? 3% viable monocytic cells were obtained.
Flow cytometric analysis of TF activity. Flow cytomet-
ric analysis was performed using a FACScan (BD Biosciences,
San Jose, CA) (9) and specific monoclonal antibodies to
human TF (clone TF9-6B4, FITC conjugated; American Di-
agnostica, Greenwich, CT) or to human CD14 (phycoerythrin-
conjugated; Caltag, South San Francisco, CA). TF activity on
intact and lysed monocytic cells was determined using a
continuous chromogenic assay (9,11).
IgG purification and in vitro exposure of normal
monocytes to aCL. IgG from the pooled sera of 7 patients with
APS and thrombosis, and from the pooled sera of 10 healthy
controls was purified by protein G–Sepharose affinity chroma-
tography (MabTrap kit; Amersham Biosciences, Uppsala,
Sweden). Endotoxin in IgG fractions was measured by Limulus
amebocyte lysate assay (Amebo-lysate; ICN Biomedicals,
Costa Mesa, CA). Anti-?2GPI activity of purified IgG was
confirmed by enzyme-linked immunosorbent assay and re-
ported semiquantitatively in standard IgG anti-?2GPI units.
IgG and aCL were determined as described above, and titers
were reported in GPL units.
For in vitro studies, normal monocytes purified from
healthy controls were cultured with serum-free RPMI 1640
containing 2 mM L-glutamine, 100 units/ml penicillin, 100
mg/ml streptomycin, and 250 pg/ml Fungizone (BioWhit-
taker/MA Bioproducts, Walkersville, MD) at 37°C in a humid-
ified, 5% CO2atmosphere. Monocytes (1.5 ? 106/ml) were
incubated for 6 hours at 37°C with purified APS patient IgG
(100 ?g/ml) or normal human serum IgG (obtained from
2836L´OPEZ-PEDRERA ET AL
Protein extraction. Purified monocytes were washed
twice with chilled phosphate buffered saline. Then, cell pellets
were resuspended with lysis buffer (200 ?l per 107cells)
containing 7.0M urea, 2M thiourea, 4% CHAPS, 2% carrier
ampholytes, and 1% dithiothreitol (DTT) (12). Extracted
proteins were separated by centrifugation at 12,000g for 15
minutes, and the supernatant was used for 2-dimensional (2-D)
electrophoresis. Protein levels were determined using a protein
assay (Bio-Rad, Hercules, CA). Samples were stored at ?80°C
Two-dimensional gel electrophoresis. Immobilized pH
gradient strips (11 cm, range 3–10 pH; Bio-Rad) were passively
rehydrated with 350 ?g of protein lysate in 150 ?l of rehydra-
tion buffer (7M urea, 2M thiourea, 4% CHAPS, 20 mM DTT,
0.5% Triton X-100, 0.5% Pharmalyte 3–10, and 0.001% brom-
phenol blue) for 12 hours. Isoelectric focusing (IEF) was
performed at 20°C, using a Protean IEF system (Bio-Rad).
Thereafter, the strips were soaked for 10 minutes in equilibra-
tion solution (50 mM Tris HCl, pH 8.8, 6M urea, 30% glycerol,
2% sodium dodecyl sulfate [SDS], and 0.001% bromphenol
blue) that contained 20 mg/ml DTT and then for an additional
10 minutes in equilibration solution that contained 25 mg/ml
iodoacetamide (13). The second-dimension procedure was
performed in 12.5% polyacrylamide gel at 35 mA/gel (Mini-
Protean 3 cell; Bio-Rad).
Detection of protein spots and data analysis. Gels
were silver stained using the Plus One Protein Silver Staining
kit (Amersham Biosciences). Images were obtained using a
GS-800 calibrated densitometer (Bio-Rad) and analyzed with
PDQuest 7.1.0 2-D analysis software (Bio-Rad). Gels were
divided into the 4 groups of study, and protein spots were
matched between gels (performed at least in duplicate for each
patient sample) and between groups. To accurately compare
spot quantities between gels, image spot quantities were
normalized by dividing the raw quantity of each spot in a gel by
the total quantity of all valid spots in that gel. Spots present in
?50% of the control gels and in ?50% of the patient gels were
filtered out of the analyses. In each of the 3 patient groups, the
mean value of the quantity of each remaining spot in the gel
was expressed as a ratio to the mean value in control gels.
Mean values of each spot on the 2 duplicate gels per individual
were calculated and then compared using the Mann-Whitney
2-tailed rank sum test. Samples from healthy controls were
used to generate control maps. Fluctuations in the protein
expression levels among control maps were monitored using
densitometry analysis, and a coefficient of variation (CV) was
generated for the mean value of each spot.
Protein spots of interest were manually excised from
preparative gels and subjected to mass spectrometry (MS)
analysis. Peptides of each sample were analyzed in a 4700
Proteomics Station (Applied Biosystems, Foster City, CA) in
automatic mode at the Proteomics Service of the University of
Co ´rdoba. After drying, samples were analyzed in the 800–
4,000 mass/charge range, with an accelerating voltage of 20 kV.
Spectra were internally calibrated with peptides from trypsin
autolysis. Proteins were identified by peptide mass fingerprint-
ing, and confirmed by matrix-assisted laser desorption ioniza-
tion (MALDI)–time-of-flight MS. The Mascot search engine
(Matrix Science, Boston, MA) was used for protein identifica-
tion over the Mass Spectrometry protein sequence DataBase.
Validation of proteomic data by quantitative real-time
reverse transcriptase–polymerase chain reaction (RT-PCR).
Total cellular RNA from monocytic cells was extracted using
TRI Reagent (Sigma, St. Louis, MO). All PCRs were per-
formed using the LightCycler (Roche Diagnostics, Indianapo-
lis, IN). The primer sequences used in this study and the
theoretical size of the PCR products are as follows: TF forward
CTACTGTTTCAGTGTTCAAGCAGTGA, reverse CAGT-
GCAATATAGCATTTGCAGTAGC (283 bp); Hsp60 for-
ward 5?-ATTCCAGCAATGACCATTGC-3?, reverse 5?-
GAGTTAGAACATGCCACCTC-3? (306 bp); annexin II
forward 5?-ATGTCTACTGTTCACGAAATC-3?, reverse 5?-
AATGAGAGAGTCCTCGTCGG-3? (387 bp); annexin I for-
ward 5?-TTGAGGAGGTTGTTTTAGCTCTG-3?, reverse 5?-
AGTTCTTGATGCCAAAATCTCAA-3? (125 bp); RhoA
forward 5?-CGCTTTTGGGTACATGGAGT-3?, reverse 5?-
GGAGGGCTGTTAGAGCAGTG-3? (247 bp); Nedd8 for-
ward 5?-AGAGCGTGACCGGAAAGGA-3?, reverse 5?-
TCATCATTCATCTGCTTGCCAC-3? (142 bp); protein
disulfide isomerase (PDI) forward 5?-GAATCTTTCT-
GAAGCCACAC-3?, reverse 5?-CATACGACCCAGAAC-
CATC-3? (235 bp); GAPDH forward 5?-TGATGACA-
TCAAGAAGGTGGTGAAG-3?, reverse 5?-TCCTTGG-
AGGCCATGTAGGCCAT-3? (239 bp).
One-step RT-PCR was performed using the Quanti-
Tect SYBR Green RT-PCR kit (Qiagen, Hilden, Germany)
(6). Samples of messenger RNA (mRNA) were analyzed in at
least 3 similar RT-PCR procedures. Negative controls contain-
ing water instead of RNA were run to confirm that samples
were not cross-contaminated. Relative expression was quanti-
fied by the standard curve method, as recommended by the
manufacturer (Qiagen), and the target amount was normalized
to the GAPDH gene. Expression of mRNA was considered
positive in patient samples when the value was ?1.7-fold
higher than in controls.
Validation of proteomic data by Western blotting.
Total cell lysate fractions (50 ?g), prepared by standard
protocols (14), were resolved by SDS–10% polyacrylamide gel
electrophoresis and then transferred to nitrocellulose mem-
branes. Annexin I, annexin II, PDI, Nedd8, RhoA, and Hsp60
protein levels were determined by Western blotting using their
respective monoclonal or polyclonal antibodies (Santa Cruz
Biotechnology, Santa Cruz, CA). Protein levels were quanti-
fied using the image analysis software Intelligent Quantifier,
version 2.1.1 (Bio Image Systems, Jackson, MI). Results were
calculated as integrated optical density and expressed in
Statistical analysis. All data are expressed as the
mean ? SEM, except those in Table 2, which are the mean ?
SD. Statistical analyses were performed using Sigmastat soft-
ware (Jandel Scientific, Erkrat, Germany). Before comparing 2
data groups, normality and equal variance tests were per-
formed. If both tests were passed, comparison was made using
a parametric test (Student’s paired t-test). If the normality
and/or equal variance test was violated, a nonparametric test
(Mann-Whitney rank sum test) was used instead. Groups were
compared by analysis of variance. Correlations were assessed
by Pearson’s product-moment correlation coefficient. P values
less than 0.05 were considered significant.
PROTEOMIC ANALYSIS OF MONOCYTES IN APS PATIENTS2837
Titers of aCL antibodies, and TF expression and
activity in monocytes from APS patients. IgG aCL
antibody isotypes were positive in 25 of the 32 patients
with primary APS with thrombosis (mean ? SEM 83.3
? 10.1 GPL units) and in 8 of the 19 patients with
primary APS without thrombosis (57.2 ? 7.9 GPL
units). IgM aCL antibody isotypes were positive in 17 of
the 32 patients with primary APS with thrombosis (48.9
? 12.3 GPL) and in 11 of the 19 patients with primary
APS without thrombosis (75.3 ? 11.6 GPL).
Mean TF mRNA levels were significantly higher
in monocytes from patients with a history of thrombosis
(relative expression level 3.57 ? 0.6) than in those from
patients without thrombosis (1.57 ? 0.3; P ? 0.025),
from patients with thrombosis but without APS (1.68 ?
0.4; P ? 0.035), and from healthy controls (0.59 ? 0.1;
P ? 0.001). Accordingly, cell surface–associated TF was
significantly increased in patients with thrombosis
(49.7 ? 3.8% positive cells) compared with patients
without thrombosis (22.4 ? 1.9%; P ? 0.001), patients
with thrombosis but without APS (19.9 ? 2.8%; P ?
0.001), and with healthy controls (4.8 ? 0.8%; P ?
There was also a direct correlation between the
degree of TF antigen expression and procoagulant ac-
tivity (PCA) of TF. PCA of TF on intact cells and cell
lysates was significantly higher in group 1 (157.7 ? 12.6
units/105lysate cells) than in the other 3 groups (64.5 ?
12.3, 39.7 ? 7.8, and 27.4 ? 2.1 units/105lysate cells in
groups 2, 3, and 4, respectively; P ? 0.001). Patients with
primary APS with thrombosis showed a significant in-
crease in TF expression among those who were positive
for IgG aCL antibody isotypes (P ? 0.05), but not
among those positive for IgM aCL antibody isotypes or
LAC, thus confirming the results of previous studies (9).
Proteomic findings. At least 2 separate experi-
ments were performed on all samples, and similar
protein spot patterns were obtained. Approximately 500
protein spots were detected on silver-stained gels.
Nearly 85% of all spots were matched on duplicate gels,
and the intensity of the same spot on different gels
showed no significant changes. Representative gels of
the highly expressed proteins in control and APS mono-
cytes are shown in Figure 1. We identified 29 spots in
APS monocytic cells, of which 22 corresponded to
known proteins, whose expression was altered signifi-
cantly between monocytes from APS patients with
thrombosis and the remaining groups (P ? 0.05). These
identified proteins are listed in Table 1.
The proteins identified as being more signifi-
cantly altered between monocytes from APS patients
with thrombosis (group 1) and the remaining groups
were annexin I, annexin II, PDI, Nedd8, RhoA, and
Hsp60. These proteins were functionally related to pro-
cesses associated with the induction of a procoagulant
state, as well as autoimmune-related responses. Further-
more, we found a significant correlation between the
titers of IgG aCL and the expression levels of the
above-mentioned proteins (all P ? 0.05). Variability of
Figure 1. Representative gels of the highly expressed proteins in
monocytes from the 4 groups (G) analyzed. A, Scanned images of the
silver-stained gels used to detect and compare spots. The positions of
proteins that were differentially expressed between groups are circled
and indexed as numbered. Molecular mass standards are shown on the
right. B, Representative close-up views of the differentially expressed
protein spots in the 4 groups analyzed, as indicated in 2-dimensional
echocardiogram maps. C, Relative normalized amounts of identified
proteins. Values are the mean and SEM. ? ? P ? 0.05 versus healthy
donors. The complete list of proteins is shown in Table 1. APS ?
antiphospholipid syndrome; PDI ? protein disulfide isomerase.
2838L´OPEZ-PEDRERA ET AL
the differentially expressed protein spots among the 10
controls used was very low, as demonstrated by CV
analysis. The proteins that were more significantly al-
tered in monocytes from patients with primary APS with
thrombosis compared with the other groups are shown
in Figures 1A and B and are listed in Table 2.
In addition, proteins associated with metabolism,
protein folding/modification, immune response, and
transcriptional factors were readily identified as being
distinctively expressed in monocytes from APS patients
with thrombosis compared with the remaining groups
analyzed. No significant differences were seen in the
expression levels of those proteins between monocytes
from healthy subjects and APS patients without throm-
bosis, between healthy subjects and patients with throm-
bosis but without APS, or between APS patients without
thrombosis and patients with thrombosis but without
APS, with the exception of 2 relevant proteins previously
reported to be connected to recurrent spontaneous
abortion, fibrinogen (spot 4), significantly reduced ver-
sus the remaining groups (P ? 0.05), and hemoglobin
(spot 22), significantly increased versus the remaining
groups (P ? 0.05). Both of these proteins were found to
be significantly deregulated in APS patients without
Proteins differentially expressed in monocytes from APS patients with thrombosis and from healthy subjects*
no.†Protein name Functional classification
mass, kd‡ pI‡Ratio§
Glutathione transferase chain A
Metabolism, response to stress
Autoimmune responses, signal transduction
Blood clot formation
Receptor for aPL induction of cell
Macrophage activity, immune responses,
Modulation of macrophage gene/protein
expression, signal transduction
Immune and inflammatory responses
8 P04083 Annexin I48 38.96.6
Phosphoglycerate mutase 4
Glutathione transferase O1-1
PA28, ? chain
Apolipoprotein A-IV precursor
RS6K 3 (fragment)
NIT 1 protein
Hemoglobin, chain D
* ERp28 ? endoplasmic reticulum protein 28; aPL ? antiphospholipid antibody; PA28 ? proteasome activator 28; PDI ? protein disulfide
isomerase; RS6K ? ribosomal S6 kinase.
† For the identification of the proteins, the repositories used were the Swiss-Prot and TrEMBL databases.
‡ Molecular mass and pI denote experimental values.
§ Expression in antiphospholipid syndrome (APS) patients with thrombosis in relation to the expression in healthy donors. Positive ratios denote
up-regulated proteins; negative ratios denote down-regulated proteins.
proteins found to be highly differentially expressed in monocytes from
patients with primary APS with thrombosis*
Expression levels (versus those in healthy donors) of the 6
Mean ? SD % change in protein
expression vs. healthy donors
180.6 ? 37.5
220.6 ? 51.5
22.7 ? 5.91
887.1 ? 65.1
378.2 ? 47.6
52.2 ? 8.4
97.4 ? 20.3
105.3 ? 30.1
88.7 ? 12.8
101.2 ? 18.8
127.4 ? 15.0
88.7 ? 10.2
112.7 ? 5.4
97.4 ? 11.2
93.2 ? 10.9
98.5 ? 19.1
123.1 ? 29.2
92.4 ? 16.2
* CV ? coefficient of variation (see Table 1 for other definitions).
PROTEOMIC ANALYSIS OF MONOCYTES IN APS PATIENTS2839
thrombosis compared with the other groups of patients
studied (P ? 0.05).
Western blot and RT-PCR analyses of differen-
tially expressed proteins in APS patients with thrombo-
sis. To validate 2-D data, annexin I, annexin II, PDI,
Nedd8, RhoA, and Hsp60 levels were analyzed by
Western blotting (Figures 2A and B). These results
showed that all 6 proteins were correctly identified by
MALDI tandem MS.
RT-PCR analysis was performed on those 6
genes identified by proteomic analysis (Figure 2C).
RT-PCR analysis confirmed the proteomic data for 4 of
these (RhoA, annexin II, PDI, and Hsp60), but mRNA
levels were not consistent with the data obtained for the
remaining 2 (annexin I and Nedd8) by differential
proteomic analysis and Western blotting. These data
strongly suggest that, for a number of the identified
proteins, differences that exist between the 2 cell systems
may not be due to altered gene expression, but rather to
alterations of other control mechanisms, such as mRNA
stability or lifespan, translation efficiency, posttransla-
tional processing, control of protein turnover, or a
combination of these (15,16).
Changes induced by aPL in the proteomic profile
of normal monocytes. All IgG aPL samples from the 7
patients were positive for aCL (?100 GPL units), and all
IgG samples from the healthy controls were negative for
aCL (?10 GPL units). The mean ? SD anti-?2GPI
antibody activities of the IgG aPL and control IgG were
78.4 ? 10.7 GPL units and 4.7 ? 0.8 GPL units,
respectively. All patient IgG selected for this study had
moderate to high anti-?2GPI activity, which correlated
with the level of IgG (P ? 0.0375, r ? 0.917). All IgG
preparations tested negative for lipopolysaccharide in
the Limulus amebocyte lysate assay. To ascertain that
changes observed in proteins related to thrombotic
events in APS patients were directly dependent on
aPL-induced activation, we compared the effects of
patient IgG and control IgG on the expression of
annexin I, annexin II, PDI, Nedd8, RhoA, and Hsp60.
Our results showed that IgG fractions purified from the
plasma of APS patients with thrombosis changed the
protein expression pattern of normal monocytes in the
same way that was observed in vivo for monocytes from
APS patients with thrombosis (Figures 3 and 4). These
Figure 2. Validation of proteomic data by Western blotting and
real-time reverse transcriptase–polymerase chain reaction. A, Repre-
sentative Western blots of 4 samples from each patient group (G) and
2 control samples, performed in triplicate, as described in Patients and
Methods. B and C, Mean and SEM mRNA and protein expression
levels of the 6 proteins identified by proteomic analysis in the 4 groups
of patients and controls. ? ? P ? 0.05 versus healthy donors. AnxI ?
annexin I; AU ? arbitrary units; IOD ? integrated optical density (see
Figure 1 for other definitions).
Figure 3. Antiphospholipid antibody (aPL)–induced changes in the
proteomic profile of normal monocytes. A, Representative gels of
differentially expressed proteins in monocytes treated with normal
human serum (NHS) IgG or antiphospholipid syndrome patient IgG.
B, Representative close-up views of the differentially expressed protein
spots, and relative normalized amounts of these proteins. Values are
the mean ? SEM. ? ? P ? 0.05 versus controls. AnxI ? annexin I;
PDI ? protein disulfide isomerase.
2840L´OPEZ-PEDRERA ET AL
IgG fractions from APS were previously shown to acti-
vate human monocytes (4,6,17).
Several mechanisms may contribute to throm-
botic manifestations of APS. Among these, activation of
monocytic cells, with the attendant loss of anticoagulant
and the gain of procoagulant functions, is likely to be
important. We and others have previously demonstrated
that monocytes are involved in the thrombotic state
characteristic of most APS patients (4,6,9). However, no
study has evaluated the changes in the proteomic pat-
terns that could underlie the pathogenic mechanisms
associated with thrombosis in this disease. Here, we have
identified a pattern of 22 differentially expressed pro-
teins that is characteristic for monocytes of APS patients
when compared with healthy individuals. Moreover,
some of these proteins play a potential role in thrombo-
sis development in APS (18).
Two annexins were up-regulated in monocytic
APS samples: annexin I and annexin II. Annexins are a
family of phospholipids and calcium-binding proteins
that modulate inflammation, immune response, and
blood coagulation. Annexin I may play a role in the
regulation of macrophage activity, and its levels are
raised in some autoimmune diseases (19,20), thus sup-
porting the notion of its involvement in the regulation of
the immune system. Also, annexin I is a substrate of
protein kinase C and protein–tyrosine kinases. Increased
annexin I expression leads to constitutive activation of
ERK1/2 in RAW 264.7 macrophage cell lines (21).
Accordingly, up-regulation of annexin I in APS mono-
cytic cells was accompanied by constitutive activation of
the MEK/ERK pathway (Lo ´pez-Pedrera C: unpublished
observations). Thus, the multifactorial downstream ef-
fects of the action of annexin I, including antiprolifera-
tive and antiinflammatory effects, might be translocated
by the ERK signaling pathway.
Annexin II is a receptor for fibrinolytic activation
localized on the cell surface of endothelial cells, mono-
cytes, and syncytiotrophoblasts (22). Annexin II is di-
rectly involved in the pathogenesis of APS. Binding of
?2GPI to human umbilical vein endothelial cells is
mediated by annexin II (23). By functioning as a recep-
tor for ?2GPI, annexin II is a target not only for
anti–annexin II antibodies but also for anti-?2GPI anti-
bodies, which are direct inducers of TF overexpression
and thus are significantly associated with thrombosis in
the setting of APS (18). Recently, it has also been
demonstrated that annexin II plays an important role in
human monocyte/macrophage-directed migration and
recruitment and that it is activated on progression from
monocytes to macrophages (24). Thus, annexin II might
constitute a common receptor for aPL induction of
TF-induced expression by aPL in monocytes
might be responsible for the prothrombotic state of APS
patients (6). A recent study has shown that the surface-
accessible, extracellular Cys186–Cys209 disulfide bond
of TF is critical for coagulation, and that PDI (a
multifunctional protein catalyzing the oxidation, reduc-
tion, and isomerization of disulfide bridges) disables
coagulation by targeting this disulfide. PDI is associated
with TF on the cell surface when coagulant activity is low
and TF-VIIa signaling is enabled. Moreover, decreased
PDI expression was associated with a 2-fold increase in
TF procoagulant activity (25). Thus, reduced expression
of PDI in monocytes from APS patients with thrombosis
might contribute to their prothrombotic state. Addition-
ally, our in vitro results further suggest that its down-
modulation is promoted by aPL, thus connecting this
protein to the mechanisms of thrombosis associated with
the syndrome. Moreover, overexpression of PDI sup-
presses NF-?B–dependent transcriptional activity (26).
Because aberrant activation of the NF-?B pathway likely
contributes to the development and progression of APS,
Figure 4. Validation of proteomic data from in vitro studies by
Western blotting and real-time reverse transcriptase–polymerase chain
reaction. Values are the mean and SEM mRNA expression levels of
the 6 proteins differentially expressed after in vitro treatment with
normal human serum (NHS) IgG or antiphospholipid syndrome (APS)
patient IgG, as identified by proteomic analysis. A representative
Western blot of each protein, analyzed in 3 independent experiments
with similar results, is also shown. ? ? P ? 0.05 versus controls. AnxI ?
annexin I; PDI ? protein disulfide isomerase; C ? control.
PROTEOMIC ANALYSIS OF MONOCYTES IN APS PATIENTS2841
it could be speculated that the reduced monocyte PDI
expression might be related to the constitutive activation
of NF-?B in APS.
Nedd8 was significantly increased in monocytic
APS cells. This protein is involved in the proteolytic
destruction of I?B (27). A constitutive NF-?B binding
activity has been demonstrated in monocytes from APS
patients in vivo, which was related to aPL-induced TF
expression (6,28), as previously demonstrated in aPL-
induced NF-?B activation of endothelial cells (5,29,30).
Thus, increased expression of Nedd8 might account for
constitutive NF-?B–binding activity in APS monocytes.
RhoA was also significantly increased in the APS
with thrombosis group. RhoA proteins are modulators
of gene expression, adhesion, and migration of activated
macrophages, which also play critical roles in inflamma-
tory signal pathways, such as those required for activa-
tion of NF-?B (31). In addition, inhibition of Rho/Rho
kinase proteins down-regulates the synthesis of TF by
cultured human monocytes, and statins (known immu-
noregulatory and antithrombotic compounds that are
now being experimentally tested in APS patients) sup-
press the synthesis of TF mediated by inhibition of Rho
activity (32). Agonists reported to activate Rho proteins
in vascular cells include thrombin, endothelin 1, and
angiotensin. Our study further indicates that aPL also
contribute to their increased expression, thus suggesting
that RhoA proteins may be directly involved in mono-
cyte APS function.
Hsp60 was down-regulated in monocytes from
APS patients. Hsp60 is a target of autoantibodies and
autoimmune T cells in healthy individuals, as well as in
those with autoimmune diseases (33). Our results are
consistent with the significant reduction in Hsp60 levels
in PBMCs from rheumatoid arthritis patients (34). Ac-
tually, Hsp60 contributes to the suppression of arthritis
by stimulating regulatory suppressive T cells (35). On
the other hand, although Hsp60 expression increases as
a consequence of an inflammatory response, monocyte
activation and thrombosis in APS have been proven to
not be related to an acute inflammatory response (9).
That might explain why in our study Hsp60 was not
overexpressed, and was even reduced, both in monocytes
from APS patients with thrombosis and in aPL-treated
monocytic cells. Nevertheless, the underlying mecha-
nisms explaining this reduced expression remain to be
We should also highlight the identification of 2
proteins (fibrinogen and hemoglobin) that might be
related to the pathogenesis of recurrent spontaneous
abortion in APS. We found that the expression levels of
fibrinogen in patients with APS without thrombosis were
lower than in the other study groups. Recent studies
have suggested that the absence or a significant decrease
in maternal fibrinogen is sufficient to cause rupture of
vasculature, affecting embryonic trophoblast infiltration,
and leading to hemorrhagic miscarriage (36). Thus,
deficiencies of fibrinogen during gestation may lead to
abnormal fetal growth or abortion.
Similarly, a recent study has shown increased
gene expression of hemoglobin in patients with recur-
rent abortion (37). Our data at the protein level further
support this increase. Therefore, these proteins might be
helpful in understanding the molecular mechanisms
involved in recurrent abortion in APS. Recent investi-
gations have shown that in many women with APS who
had miscarriages, thrombosis was not evident in the
placentae. Moreover, a very recent study (38) has shown
that inflammation, specifically activation of complement
with generation of the anaphylatoxin C5a, is an essential
trigger of fetal injury, and that TF expression on neu-
trophils, but not macrophages, is essential to the patho-
genesis of aPL-induced fetal loss, and reveals a func-
tional linkage between C5a, neutrophil activation, and
fetal injury. Collectively, these data might explain why
our proteomic studies of monocytes from APS patients
without thrombosis did not reflect many significant
changes in protein expression, particularly in those
proteins related to procoagulant events.
Gene expression patterns of PBMCs from APS
patients have recently been analyzed, and candidate
clusters of genes that clearly exhibited reliable discrim-
inatory patterns between the APS and non-APS patient
populations with thrombosis were identified (39). Nota-
bly, both apoH (?2GPI) and MEKK1 genes, previously
described in the molecular pathogenesis of APS, were
included in the list of genes found to differ between APS
and non-APS patients with thrombosis. Our proteomic
studies further complement the genetic ones, and thus
both proteomic and transcriptomic approaches may
provide distinct but complementary views in profiling
In summary, this study has identified altered
expression of proteins that might be directly related to
thrombotic events in APS. The development and use of
such proteomic biomarkers for diagnosis, assessing prog-
nosis, and guiding therapy might revolutionize the care
of patients with APS.
We thank all patients and healthy subjects for their
participation in the study.
2842 L´OPEZ-PEDRERA ET AL
Dr. Lo ´pez-Pedrera had full access to all of the data in the
study and takes responsibility for the integrity of the data and the
accuracy of the data analysis.
Study design. Lo ´pez-Pedrera, Cuadrado, Velasco, Khamashta.
Acquisition of data. Herna ´ndez, Buendı ´a, Aguirre, Barbarroja, Torres,
Analysis and interpretation of data. Lo ´pez-Pedrera, Cuadrado, Vil-
Manuscript preparation. Lo ´pez-Pedrera, Cuadrado, Villalba, Kha-
Statistical analysis. Aguirre, Barbarroja, Velasco.
1. Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL,
Cervera R, et al. International consensus statement on an update
of the classification criteria for definite antiphospholipid syndrome
(APS). J Thromb Haemost 2006;4:295–306.
2. Amengual O, Atsumi T, Khamashta MA. Tissue factor in an-
tiphospholipid syndrome: shifting the focus from coagulation to
endothelium. Rheumatology (Oxford) 2003;42:1029–31.
3. Amengual O, Atsumi T, Khamashta MA, Hughes GR. The role of
the tissue factor pathway in the hypercoagulable state in patients
with the antiphospholipid syndrome. Thromb Haemost 1998;79:
4. Zhou H, Wolberg AS, Roubey RA. Characterization of monocyte
tissue factor activity induced by IgG antiphospholipid antibodies
and inhibition by dilazep. Blood 2004;104:2353–8.
5. Vega-Ostertag M, Casper K, Swerlick R, Ferrara D, Harris EN,
Pierangeli SS. Involvement of p38 MAPK in the up-regulation of
tissue factor on endothelial cells by antiphospholipid antibodies.
Arthritis Rheum 2005;52:1545–54.
6. Lopez-Pedrera C, Buendia P, Cuadrado MJ, Siendones E, Aguirre
MA, Barbarroja N, et al. Antiphospholipid antibodies from pa-
tients with the antiphospholipid syndrome induce monocyte tissue
factor expression through the simultaneous activation of NF-?B/
Rel proteins via the p38 mitogen-activated protein kinase pathway,
and of the MEK-1/ERK pathway. Arthritis Rheum 2006;54:
7. Sebastiani GD, Galeazzi M, Tincani A, Scorza R, Mathieu A,
Passiu G, et al. HLA-DPB1 alleles association of anticardiolipin
and anti-?2GPI antibodies in a large series of European patients
with systemic lupus erythematosus. Lupus 2003;12:560–3.
8. Special report from the National Institute of Neurological Disor-
ders and Stroke: classification of cerebrovascular diseases III.
9. Cuadrado MJ, Lopez-Pedrera C, Khamashta MA, Camps MT,
Tinahones F, Torres A, et al. Thrombosis in primary antiphospho-
lipid syndrome: a pivotal role for monocyte tissue factor expres-
sion. Arthritis Rheum 1997;40:834–41.
10. Brandt JT, Triplett DA, Alving B, Scharrer I. Criteria for the
diagnosis of lupus anticoagulants: an update. Thromb Haemost
11. Carson S. Continuous chromogenic tissue factor assay comparison
to clot-based assays and sensitivity established using pure tissue
factor. Br J Haematol 1987;47:379–87.
12. Herbert B. Advances in protein solubilization for two-dimensional
electrophoresis. Electrophoresis 1999;20:660–3.
13. Herbert B, Galvani M, Hamdan M, Olivieri E, MacCarthy J,
Pedersen S, et al. Reduction and alkylation of proteins in prepa-
ration of two-dimensional map analysis: why, when and how?
14. Andrews NC, Faller DV. A rapid micropreparation technique for
extraction of DNA-binding proteins from limiting numbers of
mammalian cells. Nucleic Acid Res 1991;19:2499.
15. Salzano AM, Paron I, Pines A, Bachi A, Talamo F, Bivi N, et al.
Differential proteomic analysis of nuclear extracts from thyroid
cell lines. J Chromatogr B Analyt Technol Biomed Life Sci
16. Kwong KY, Bloom GC, Yang I, Boulware D, Coppola D, Hase-
man J, et al. Synchronous global assessment of gene and protein
expression in colorectal cancer progression. Genomics 2005;86:
17. Wolberg AS. Antiphospholipid antibody effects on monocytes.
Curr Rheumatol Rep 2007;9:198–204.
18. Zhang J, McCrae KR. Annexin A2 mediates endothelial cell
activation by antiphospholipid/anti-?2 glycoprotein I antibodies.
19. Probst-Cousin S, Kowolik D, Kuchelmeister K, Kayser C, Neun-
dorfer B, Heuss D. Expression of annexin-1 in multiple sclerosis
plaques. Neuropathol Appl Neurobiol 2002;28:292–300.
20. Goulding NJ, Dixey J, Morand EF, Dodds RA, Wilkinson LS,
Pitsillides AA, et al. Differential distribution of annexins-I, -II,
-IV, and -VI in synovium. Ann Rheum Dis 1995;54:841–5.
21. Alldridge LC, Harris HJ, Plevin R, Hannon R, Bryant CE.
Regulation of the mitogen activated protein kinase/ERK pathway
by the annexin lipocortin 1. J Biol Chem 1999;274:37620–8.
22. Falcone DJ, Borth W, Faisal Khan KM, Hajjar KA. Plasminogen-
mediated matrix invasion and degradation by macrophages is depen-
dent on surface expression of annexin II. Blood 2001;97:777–84.
23. Cesarman-Maus G, Rios-Luna NP, Deora AB, Huang B, Villa R,
del Carmen Cravioto M, et al. Antibodies against the fibrinolytic
receptor, annexin 2, in antiphospholipid syndrome. Blood 2006;
24. Brownstein C, Deora AB, Jacovina AT, Weintraub R, Gertler M,
Khan KM, et al. Annexin II mediates plasminogen-dependent
matrix invasion by human monocytes: enhanced expression by
macrophages. Blood 2004;103:317–24.
25. Ahamed J, Versteeg HH, Kerver M, Chen VM, Mueller BM,
Hogg PJ, et al. Disulfide isomerization switches TF from coagu-
lation to cell signaling. Proc Natl Acad Sci U S A 2006;103:
26. Higuchi T, Watanabe Y, Waga I. Protein disulfide isomerase
suppresses the transcriptional activity of NF?B. Biochem Biophys
Res Commun 2004;318:46–52.
27. Amir RE, Iwai K, Ciechanover A. The NEDD8 pathway is
essential for SCF?-TrCP- mediated ubiquitination and processing of
the NF-?B precursor p105. J Biol Chem 2002;277:23253–9.
28. Bohgaki M, Atsumi T, Yamashita Y, Yasuda S, Sakai Y, Furusaki
A, et al. The p38 mitogen-activated protein kinase (MAPK)
pathway mediates induction of the TF gene in monocytes stimu-
lated with human monoclonal anti-?2 glycoprotein I antibodies.
Int Immunol 2004;16:1633–41.
29. Dunoyer-Geindre S, de Moerloose P, Galve-de Rochemonteix B,
Reber G, Kruithof EK. NF?B is an essential intermediate in the
activation of endothelial cells by anti-?2-glycoprotein 1 antibodies.
Thromb Haemost 2002;88:851–7.
30. Raschi E, Testoni C, Bosisio D, Borghi MO, Koike T, Mantovani
A, et al. Role of the MyD88 transduction signaling pathway in
endothelial cell activation by antiphospholipid antibodies. Blood
31. Rolfe BE, Worth NF, World CJ, Campbell JH, Campbell GR.
Rho and vascular disease. Atherosclerosis 2005;183:1–16.
32. Nagata K, Ishibashi T, Sakamoto T, Ohkawara H, Shindo J,
Yokohama K, et al. Rho/Rho-kinase is involved in the synthesis of
tissue factor in human monocytes. Atherosclerosis 2002;163:
33. Wallin RP, Lundquist A, More SH, von Bonin A, Kiessling R,
Ljunggren HG. Heat shock proteins as activators of the innate
immune system. Trends Immunol 2002;23:130–5.
34. Dotzlaw H, Schulz M, Eggert M, Neeck G. A pattern of protein
expression in peripheral blood mononuclear cells distinguishes
PROTEOMIC ANALYSIS OF MONOCYTES IN APS PATIENTS2843
rheumatoid arthritis patients from healthy individuals. Biochim
Biophys Acta 2004;1696:121–9.
35. Quintana FJ, Carmi P, Mor F, Cohen IR. Inhibition of adjuvant
arthritis by a DNA vaccine encoding human heat shock protein 60.
J Immunol 2002;169:3422–8.
36. Kim YS, Kim MS, Lee SH, Choi BC, Lim JM, Cha KY, et al.
Proteomic analysis of recurrent spontaneous abortion: identifica-
tion of an inadequately expressed set of proteins in human
follicular fluid. Proteomics 2006;6:3445–54.
37. Baek KH. Aberrant gene expression associated with recurrent
pregnancy loss. Mol Hum Reprod 2004;10:291–7.
38. Redecha P, Tilley R, Tencati M, Salmon JE, Kirchhofer D,
Mackman N, et al. Tissue factor: a link between C5a and neutro-
phil activation in antiphospholipid antibody-induced fetal injury.
39. Potti A, Bild A, Dressman HK, Lewis DA, Nevins JR, Ortel TL.
Gene-expression patterns predict phenotypes of immune-medi-
ated thrombosis. Blood 2006;107:1391–6.
Clinical Image: Stalagmite hips in a patient with systemic sclerosis
The patient, a 61-year-old woman with a 25-year history of systemic sclerosis, presented to our clinic with chronic bilateral hip pain
and severe restriction of movement. Three-dimensional computed tomographic angiography, performed to assess joint replacement
as a treatment approach, provided a vivid illustration of ectopic calcifications that had developed in the course of the disease. The
posteroanterior view of the pelvis presented here reveals entrapment of both hip joints within massive calcific tissue. Calcifications
developed in the periarticular space and within adjacent muscles (quadriceps, gluteal muscles, obturators). Bilateral loss of cartilage
space and erosion of the external side of the acetabulum were also documented, especially on the right side. Ileac arteries and deep
and superficial femoral arteries were unaffected, although smaller branches appeared to be trapped within these enormous calcific
masses. The morphology of ectopic calcifications in this patient resembled the formation of stalagmites.
Christina G. Katsiari, MD, PhD
Athens University Medical School
Eleftherios Dounis, MD
Laiko General Hospital
Athanasios Sissopoulos, MD
Petros P. Sfikakis, MD
Athens University Medical School
2844 L´OPEZ-PEDRERA ET AL