ORIGINAL ARTICLE

Annexin A5 and anti-annexin antibodies in patientswith systemic lupus erythematosus

Antoni Hrycek • Paweł Cieslik

Received: 16 June 2010 / Accepted: 16 January 2011 / Published online: 5 February 2011

� Springer-Verlag 2011

Abstract Plasma levels of annexin A5 (ANX A5) and

anti-annexin A5 (aANX A5) antibodies were evaluated in

51 women with systemic lupus erythematosus (SLE). The

results were compared between the total SLE group, sub-

groups on/without immunosuppressive therapy and the

control (28 women). The relationships between ANX

A5/aANX A5 antibodies levels and laboratory vari-

ables (anti-cardiolipin antibodies—aCL, total cholesterol,

thrombocyte count, activated partial thromboplastin time—

APTT, prothrombin time, international normalized ratio–

INR) were performed in the total SLE group and in the

patient subgroups identified as the arithmetic mean of ANX

A5 concentration in the control plus 1–4 standard devia-

tions (SD). The whole SLE group and the subgroup on

immunosuppression showed significantly higher ANX A5

and IgG aANX A5 antibodies concentrations. A weak

positive correlation was found between ANX A5 and

thrombocyte count, a moderate one between IgG and IgM

aANX A5 antibodies, a weak negative correlation between

IgG aANX A5 and APTT in the whole SLE group. SLE

subgroups with ANX A5 concentrations higher than the

control mean plus 3 or 4 SD showed a weak/moderate

negative correlation of this parameter with aANX A5

antibodies, moderate one with IgG aCL antibodies levels, a

moderate positive correlation with cholesterol concentra-

tion, moderate/high positive correlations with thrombocyte

count. The association between plasma ANX A5/IgG

aANX A5 levels and severity of disease was noticed. The

role of aANX A5 and IgG aCL antibodies as causative

factors of increased ANX A5 levels was suggested, and the

relationship between ANX A5 and thrombocyte count was

revealed.

Keywords SLE � Annexin A5 � Antibodies �Associations � Therapy

Introduction

Due to structural similarities, twelve of over a thousand

proteins of the annexin superfamily have been classified in

the annexin A family [1–4]. Under physiological condi-

tions, annexins are involved in cell signaling, membrane

trafficking events, the process of blood coagulation, and

inflammation [1, 2, 5–8]. The best known is annexin A5

(ANX A5—previously also referred to as placental anti-

coagulant protein I, vascular anticoagulant a, calphobindin

I) [3]. It has been found intracellularly in cellular vesicles

and plasma membranes. It has been identified in the

endothelium, chondrocytes, osteoblasts, Schwann cells,

hepatocytes, apical surface of placental syncytiotropho-

blasts, and in heart tissues, optical nerve and bronchi

[1, 7, 9]. Although it is an intracellular protein, its small

amounts are also detected in the plasma, cerebrospinal

fluid, urine from healthy persons [4], amniotic fluid, and

semen plasma [1].

ANX A5 is a natural anticoagulant with a high calcium-

dependent binding affinity for negatively charged phos-

pholipids [10, 11]. It also modulates the activity of

phospholipase A2 and protein kinase C (PKC) both in vitro

and in vivo [7, 8, 12–14]. Through inhibiting prothrombin

activation, ANX A5 prevents arterial and venous thrombus

A. Hrycek � P. Cieslik

Department of Internal, Autoimmune, and Metabolic Diseases,

Medical University of Silesia, ul. Medykow 14,

40-752 Katowice, Poland

A. Hrycek (&)

ul. Tysiaclecia 86a/34, 40-871 Katowice, Poland

e-mail: genmar@ka.onet.pl

123

Rheumatol Int (2012) 32:1335–1342

DOI 10.1007/s00296-011-1793-2

formation [1, 7]. The anti-thrombotic effect of ANX A5 is

mediated by its preferential binding to phosphatidylserine

expressed on the surface during the destruction of the cells

[15]. Phosphatidylserine exposure is the physiological

signal for the processes of coagulation and apoptosis.

Extracellular ANX A5 accumulation helps recognize the

signaling and binds with phosphatidylserine preventing the

coagulation process [1, 7, 11]. Similar to prothrombin,

ANX A5 is phospholipids cofactor, and it is assumed that it

forms phospholipid–protein complex, which is the target

for anti-phospholipid antibodies action [10, 16].

IgG and IgM anti-annexin A5 antibodies (aANX A5

antibodies) constitute another problem; they are among

anti-phospholipid/anti-cofactor antibodies that reduce the

availability of ANX A5. The two most important clinical

types of anti-phospholipid antibodies are anti-cardiolipin

(aCL) antibodies and lupus anticoagulant (LA), frequently

detected in patients with systemic lupus erythematosus

(SLE). Common clinical presentations of patients with aCL

antibodies are venous or arterial thrombosis, thrombocy-

topenia, and pregnancy complications such as intrauterine

fetal demise, etc. It should be emphasized that the signifi-

cance of aANX A5 antibodies as well as ANX A5 in SLE

patients has not been clearly defined [10]. However, it has

been suggested that these antibodies might mediate some

reduction in ANX A5 anticoagulant properties [11].

SLE is an autoimmune disease with disturbances in the

process of clearance of apoptotic cells [8, 17] and the

dominant role of extracellular DNA in pathogenesis of

disease [15]. Apoptotic cells display characteristic mor-

phological and surface changes related to a key functional

phenomenon, which is the above-mentioned phosphati-

dylserine expression [15].

The purpose of the present study was to assess plasma

ANX A5 and aANX A5 antibodies levels and their associ-

ations with the severity of the disease in treated SLE

patients. A further aim of this study was to determine rela-

tionships between plasma concentrations of ANX A5/aANX

A5 antibodies and selected laboratory parameters in the

examined SLE group. It was expected that the obtained

results might allow to find some cause-effect phenomena

between the analyzed parameters, and consequently it might

be advantageous in the management of SLE.

Materials and methods

Patients and controls

In this retrospective study, the investigations were carried

out in 51 women with SLE (mean age 52 ± 16 years)

admitted to the hospital for the routine control who had

already received treatment (of several-month to several-

year duration). The following drugs were applied as single

or in various combinations—prednisone (7.5–15 mg per

day in 39 patients), immunosuppressive drugs—azathio-

prine (50–100 mg per day in 11 patients), cyclophospha-

mide (in cyclic courses 200 mg per day in 1 patient),

cyclosporine A (50–100 mg per day in 2 patients) and

prolactin-suppressive drug, dopamine agonist—quinago-

lide (Norprolac 25–50 lg per day in 5 patients). When

necessary, patients were allowed to take nonsteroidal

anti-inflammatory drugs (8 patients) and analgesics.

In 41 patients, immunosuppressive therapy was applied, 10

patients were without immunosuppression, and they

received Norprolac or nonsteroidal anti-inflammatory

drugs. All study subjects met at least four of the 1982

American Rheumatism Association (ARA) criteria for the

diagnosis of SLE [18], updated in 1997 [19]; most were

considered to have mild-to-moderate disease. The control

group consisted of 28 healthy medication-free women

(mean age 50 ± 17 years i.e. age distribution similar to the

patient group) recruited from members of the staff of

medical department. The patients and control subjects were

questioned in relation to tobacco smoking, contraception or

hormone-replacement therapy, history of thrombosis, and

also pre- or post-menopausal period was determined. In the

patient group, 9 women (18%) and 5 in the controls (18%)

were current smokers, 4 patients (8%) and 3 control sub-

jects (10%) applied oral contraceptive pills or hormone-

replacement therapy, the history of thrombosis was

revealed only in the patient group (5 women), 34 patients

were pre-menopausal, 17 patients post-menopausal and 18

controls/10 controls, respectively.

On the day of blood sampling, clinical evaluation was

performed including examination and calculation of the

SLE disease activity index (SLEDAI) score. Our patients

demonstrated the involvement of three to four of the nine

possible organ systems; all treated subjects had a SLEDAI

score of \10 during this examination. Arithmetic mean

(X) and ± standard deviation (SD) of SLEDAI score in the

subgroup of 41 patients with immunosuppressive therapy

and in 10 patients without immunosuppression were the

following: 7.0 ± 1.52 and 5.80 ± 1.23, respectively,

P \ 0.05. None of the patients presented with severe renal

or hepatic dysfunction.

The serological features were assessed in each patient.

The laboratory indicators of seroactivity in SLE and the

screening for autoimmune disease included erythrocyte

sedimentation rate, anti-DNA antibodies, antinuclear anti-

bodies, aCL antibodies, and complement factor C3 levels

in the serum. These tests were performed using standard

techniques.

All blood samples were collected mid-morning between

7 and 8 am from fasting and resting patients and were

stored at the temperature of below -70�C before use.

1336 Rheumatol Int (2012) 32:1335–1342

123

The plasma samples were randomly coded, and the mea-

surements were performed blind. The study was approved

by the local independent bioethics committee; all patients

gave their written consent.

Laboratory measurements and investigation scheme

Clinical measurements included plasma ANX A5, and IgG

and IgM aANX antibodies levels. The tests were performed

by commercially available enzyme-linked immunoassay

(ELISA) using IMUCLONER kits in a lQuant spectro-

photometer ELISA. The following investigation scheme

was adopted: (1) the comparison of the results (a) between

SLE group (51 treated female patients) and healthy con-

trols (28 women), (b) between separated SLE subgroup

receiving immunosuppressive therapy (41 patients) and

healthy controls, (c) between SLE subgroup without

immunosuppressive therapy (10 women) and the control,

(d) between SLE subgroup receiving immunosuppressive

therapy (41 patients) and SLE subgroup without immuno-

suppression (10 women), (2) the evaluation of the rela-

tionships in the whole SLE group (a) between the

concentrations of ANX A5/aANX A5 antibodies and the

following laboratory variables determined by standard

techniques (aCL antibodies, total cholesterol concentration

in serum, platelet number, activated partial thromboplastin

time—APTT, prothrombin time, international normalized

ratio-INR), (b) between ANX A5 and IgG/IgM aANX A5

antibodies levels, (c) between IgG aANX A5 and IgM

aANX A5 antibodies, (3) the determination of the rela-

tionships between ANX A5 concentrations and the

above-mentioned laboratory variables in SLE subgroups

identified on the basis of the arithmetic mean of ANX A5

concentration in the control group plus 1–4 standard

deviations (SD).

Statistical analyses

The obtained results were analyzed statistically by calcu-

lating X and ±SD and also median values were established

and minimum and maximum ranges were included. The

results were subject to normal distribution analysis by the

Shapiro–Wilk’s test. When the distributions of the exam-

ined variables were normal, Student’s t-test for indepen-

dent data was used. When the distributions of the examined

variables were non-normal, nonparametric tests were per-

formed in further calculations. Differences between vari-

ables were analyzed with the U Mann–Whitney’s test.

Correlations between variables were measured with the

Pearson’s correlation coefficient (r). The difference

between arithmetic means was considered statistically

significant at P \ 0.05 and highly statistically significant at

P \ 0.001.

Results

A highly statistically significant difference was found

between mean plasma level of ANX A5 of the total SLE

group (1.34 ± 0.98 ng/ml, median = 0.95, range

0.52–5.51) when compared with the control group

(0.77 ± 0.22 ng/ml, median = 0.76, range 0.48–1.34);

P \ 0.001. This group also demonstrated a statistically

significantly higher plasma concentration of IgG aANX A5

antibodies than the control (2.35 ± 2.25 and 1.53 ± 0.76

AU/ml, respectively; P \ 0.05, median = 1.95, range

0.63–15.13 and median = 1.52, range 0.53–3.95, respec-

tively). No significant difference was seen in the mean

plasma levels of IgM aANX A5 antibodies between the

study group and the control group (2.23 ± 1.22 AU/ml and

2.30 ± 0.95 AU/ml, respectively, median = 2.05, range

0.60–6.27 and median = 1.96, range 1.21–3.90, respec-

tively) (see also Fig. 1).

SLE patients on immunosuppressive therapy had a

highly statistically significant difference in the mean plasma

concentration of ANX A5 compared to the control

(1.43 ± 1.07 ng/ml and 0.77 ± 0.22 ng/ml, respectively;

P \ 0.001, median = 0.95, range 0.57–5.51 for the sub-

group of patients). Their mean plasma level of IgG aANX

A5 antibodies was significantly higher than that of the

control group (2.23 ± 2.20 AU/ml and 1.53 ± 0.76 AU/ml,

respectively; P \ 0.05, median = 1.95, range 0.63–15.13

for the subgroup of patients). The mean values of IgM aANX

A5 antibodies levels did not show statistically significant

difference between the study subgroup and the control group

(2.20 ± 1.26 and 2.30 ± 0.95 AU/ml, respectively, med-

ian = 2.05, range 0.60–6.27 for the patient subgroup). Also,

no statistically significant differences were found between

plasma IgG and IgM aANX A5 antibodies levels in SLE

subgroup without immunosuppressive therapy and the

control (IgG: 2.66 ± 2.55 and 1.53 ± 0.76 AU/ml,

respectively, median = 1.85, range 0.63–9.41 for the sub-

group of patients; IgM: 2.32 ± 1.09 and 2.30 ± 0.95

AU/ml, respectively, median = 2.00, range 1.03–4.06 for

the subgroup of patients). The mean plasma ANX A5 level

was slightly higher in SLE patients without immunosup-

pression than in the healthy controls (0.99 ± 0.36 and

0.77 ± 0.22 ng/ml, respectively, median = 0.93, range

0.52–1.62 for the patient subgroup); the difference was close

to statistical significance (P \ 0.052). The differences of

mean values concerning ANX A5, IgG, and IgM aANX A5

antibodies did not reveal statistical significances when the

results between the above-mentioned subgroups of SLE

patients were compared (1.43 ± 1.07 ng/ml, 2.23 ± 2.20

AU/ml, 2.20 ± 1.26 AU/ml in the subgroup on immuno-

suppressive therapy and 0.99 ± 0.36 ng/ml, 2.66 ± 2.55

AU/ml, 2.32 ± 1.09 AU/ml in SLE subgroup without

immunosuppression, respectively).

Rheumatol Int (2012) 32:1335–1342 1337

123

The analysis of the relationships, performed in the total

SLE group, revealed a moderate positive correlation

between IgG aANX A5 and IgM aANX A5 antibodies

levels, a weak positive correlation between the levels of

ANX A5 and the peripheral blood thrombocyte count, and

a weak negative correlation between IgG aANX A5 anti-

bodies levels and APTT (Table 1). Correlations with other

laboratory variables tested did not reach the level of sta-

tistical significance (Table 1).

Correlations between ANX A5 level and IgG/IgM

aANX A5 antibodies levels and the laboratory variables

results in the subgroups of the SLE patients with ANX A5

cutoff defined as arithmetic mean of the control plus 1–4

SD are presented in Table 2. The subgroups of SLE

patients with three or four SD above the mean of the

control group had weak/moderate negative ANX A5 cor-

relations with IgG and IgM aANX A5 antibodies, respec-

tively, moderate negative correlation with IgG aCL, and

moderate but positive correlation coefficient with serum

cholesterol level. These SLE subgroups also showed

moderate and high positive correlation coefficients

between ANX A5 and thrombocyte count (Fig. 2). No

other relationships were found between ANX A5 concen-

trations of the above-mentioned SLE subgroups, which

were considered together, and the obtained results of other

laboratory variables.

Discussion

As already mentioned, ANX A5 is a protein with potent

anti-thrombotic properties preventing coagulation pro-

cesses. However, its anti-thrombotic potential has not so

far been used in clinical practice. ANX A5 also plays an

important regulatory role in apoptosis [5]. However,

it should be emphasized that the role of circulating ANX

A5 and aANX A5 antibodies remains to be defined.

Our own investigations have revealed significantly

higher plasma concentration of ANX A5 accompanied by

significantly higher IgG aANX A5 antibodies levels in the

total group of SLE patients and SLE patients undergoing

immunosuppressive therapy, compared to the control.

AN

X A

5 pl

asm

a co

ncen

trat

ion

[ng/

mL]

0,0

0,5

1,0

1,5

2,0

2,5

IgM

aA

NX

A5

antib

odie

s pl

asm

a co

ncen

trat

ion

[AU

/mL]

0

1

2

3

4

5

IgG

aA

NX

A5

antib

odie

s pl

asm

a co

ncen

trat

ion

[AU

/mL]

0

1

2

3

4

5

SLEgroup

SLEgroup

SLEgroup

Controlgroup

Controlgroup

Controlgroup

Fig. 1 Plasma levels of ANX A5 and aANX A5 antibodies

(mean ± SD values are shown) in the total SLE group and in the

controls

Table 1 Correlation coefficients (r) between ANX A5 or IgG and IgM aANX A5 antibodies levels and the results of other laboratory variables

investigated in peripheral blood in the total SLE group

Pearson’s (r)

Parameter aANX A5 IgG aANX A5 IgM aCL IgG aCL IgM Cholesterol Thrombocytes APTT PT INR

ANX A5 N = 51 0.0571 -0.0114 -0.1298 0.1102 0.1268 0.2372 -0.0083 -0.0977 -0.0830

aANX A5 IgG N = 51 _ 0.3265 0.1447 0.0624 -0.1382 -0.0170 20.2214 -0.0875 -0.0875

aANX A5 IgM N = 51 _ _ -0.0935 0.0332 -0.0002 -0.0374 -0.0003 -0.0189 -0.0033

Statistically significant results are in the bold

1338 Rheumatol Int (2012) 32:1335–1342

123

Patients with mild SLE severity treated with dopamine

receptor agonists or nonsteroidal anti-inflammatory drugs

(only those individuals can receive such therapies [20]) did

not show significant differences in plasma IgG and IgM

aANX A5 antibodies levels compared to the control.

Plasma concentration of ANX A5 in the patients treated

with dopamine receptor agonist or nonsteroidal anti-

inflammatory drugs was higher than in the control group

with the difference which was only close to, but did not

reach the level of statistical significance.

It has been well established that plasma ANX A5 ele-

vation in women with recurrent miscarriages might result

from its replacement by anti-phospholipid antibodies; thus,

excess annexin might originate from ANX A5 pool highly

expressed by placental trophoblasts [2, 9, 12]. In the

pathomechanism of thrombus formation, the possibility of

ANX A5 displacement from endothelial surface should be

considered. Animal experiments have shown that, similar

to sensu lato anti-phospholipid antibodies, specific aANX

A5 antibodies are capable of replacing and reducing ANX

A5 activity on cell surfaces [12]. However, it has been

suspected that, opposite to anti-phospholipid antibodies,

specific aANX A5 antibodies might also simultaneously

interfere with ANX A5 function. Their actual role requires

further studies [7] because there has been conflicting evi-

dence [21].

Irrespective of the pathomechanism of ANX A5 and IgG

aANX A5 antibodies actions in SLE, it is worth consid-

ering the mechanisms that could possibly lead to excess

amounts thereof in the plasma of SLE in our study subjects.

Taking into consideration highly statistically significant

increase in ANX A5 concentration in the total group and in

SLE subgroup on immunosuppressive therapy and also

noticeable higher mean level of this protein in the subgroup

without immunosuppression the essential role of ANX A5

may be suggested. ANX A5 as an antigen might stimulate

the production of specific aANX A5 antibodies [7] with a

consequent increase in these antibodies concentrations in

plasma. However, we did not observe any significant

associations between elevated plasma level of ANX A5 and

IgG and IgM aANX A5 antibodies as well as IgG and IgM

aCL levels in the whole SLE group. This is, to some extent,

in accordance with the results of other authors, who did not

find any causative relationship between anti-phospholipid

antibody concentrations and plasma ANX A5 in SLE

patients [12]. An analysis of these relationships in our SLE

subgroups with an ANX A5 cutoff defined as the arithmetic

mean of the control plus 1–4 SD revealed a negative cor-

relation (weak or moderate) between plasma ANX A5 and

IgG/IgM aANX A5 and IgG aCL antibodies concentration

in those SLE patients whose ANX A5 levels were higher

than the control mean plus 3 or 4 SD. Therefore, we think

that the obtained result may indicate the main role of the

antibodies in the pathomechanism of the observed phe-

nomena i.e. the highest statistical difference in relation to

ANX A5 between SLE patients (the total group and the

subgroup on immunosuppressive therapy) and the control,

concomitant higher level of IgG aANX A5 antibodies in

these SLE patients, negative correlation of ANX A5 with

IgG/IgM aANX A5 and IgG aCL antibodies in the separated

subgroups. Thus, it might be suggested that in treated SLE

patients, an increase in production and in binding of these

antibodies in organ cells is associated with a consequent

ANX A5 release from the cell surface. This could account

for the phenomena observed in our investigations.

It has also been found that ischemia might cause ANX

A5 concentrations increase [3]. However, attempts to

determine ANX A5 levels in plasma of patients with ath-

erosclerotic disease did not give convincing results; lower

plasma concentrations were found in the acute stage of

myocardial infarction [22]. Taking into account the role of

cholesterol in atherosclerotic events, its serum level was

determined in our patients. Although the total SLE group

did not show correlations between plasma level of ANX

Table 2 Correlation coefficients (r) between concentrations of ANX A5 expressed as[arithmetic mean plus SD (1–4 SD) for the control and

other laboratory variables results obtained in SLE subgroups

Pearson’s (r)

Parameter aANX A5

IgG

aANX A5

IgM

IgG aCL IgM

aCL

Cholesterol Thrombocytes APTT PT INR

ANX A5; above the control

mean ? 1SD, N = 24

-0.0728 -0.22O1 -0.1266 0.0635 0.0807 0.3268 -0.1049 -0.1426 -0.1543

ANX A5; above the control

mean ? 2SD, N = 19

-0.1967 -0.1306 -0.2980 -0.0321 0.1863 0.3901 -0.0067 -0.1666 -0.1852

ANX A5; above the control

mean ? 3SD, N = 16

20.2922 20.2131 20.4180 -0.0945 0.3936 0.4223 0.0481 -0.1209 -0.2159

ANX A5; above the control

mean ? 4SD, N = 11

20.3649 20.3531 20.5049 -0.2120 0.3720 0.7670 0.2219 -0.0815 -0.1374

Statistically significant results are in the bold

Rheumatol Int (2012) 32:1335–1342 1339

123

A5 and cholesterol concentration, SLE subgroups with an

ANX A5 cutoff defined as the control mean plus 3 or 4 SD

had a positive correlation between the two parameters.

As an anticoagulant protein, ANX A5 also inhibits

activated thrombocytes aggregation or their fusion with

tissue factor-bearing microvesicles [3]. Our total SLE

group showed a weak positive correlation between plasma

ANX A5 and thrombocyte count per unit blood volume.

The correlation coefficient was very high in the SLE sub-

group with the level of ANX A5 over the control mean plus

4 SD, and it was moderate in the subgroups with concen-

tration of ANX A5 above the control mean plus 1–3 SD. It

might be hypothesized that these relationships result from

an anti-aggregation effect of plasma ANX A5 on throm-

bocyte, which, from the clinical point of view, could be

beneficial in SLE patients. Our investigations did not reveal

ANX A5 plasma concentration [ng/mL]

Thr

ombo

cyte

s bl

ood

coun

t [x

109 p

er L

]

100

150

200

250

300

350

400

450

ANX A5 plasma concentration [ng/mL]

Thr

ombo

cyte

s bl

ood

coun

t [x

109 p

er L

]

100

150

200

250

300

350

400

450

ANX A5 plasma concentration [ng/mL]

Thr

ombo

cyte

s bl

ood

coun

t [x

109 p

er L

]

100

150

200

250

300

350

400

450

ANX A5 plasma concentration [ng/mL]

Thr

ombo

cyte

s bl

ood

coun

t [x

109 p

er L

]

100

150

200

250

300

350

400

450

r = 0.2372 r = 0.3268

r = 0.3901 r = 0.4223

a b

c d

ANX A5 plasma concentration [ng/mL]

0 1 2 3 4 5 60 1 2 3 4 5 6

0 1 2 3 4 5 60 1 2 3 4 5 6

0 1 2 3 4 5 6

Thr

ombo

cyte

s bl

ood

coun

t [x

109 p

er L

]

100

150

200

250

300

350

400

450r = 0.7670

e

Fig. 2 Correlations between

plasma ANX A5 concentration

and peripheral blood

thrombocyte count in a total

group of SLE patients and b c de subgroups with SLE identified

as the arithmetic mean of ANX

A5 concentration in the control

plus 1–4 standard

deviations (SD)

1340 Rheumatol Int (2012) 32:1335–1342

123

significant correlations between plasma ANX A5 and

APTT, prothrombin time, and INR both in the total SLE

group and in SLE subgroups with an ANX A5 cutoff

defined as the control mean plus 1–4 SD.

As it was mentioned earlier, we obtained significantly

higher IgG aANX antibodies levels in the plasma of the

total SLE group and in the subgroup on immunosuppres-

sive therapy compared with the control. The phenomenon

had been previously investigated [1], and the results of our

investigations are partly consistent with the observations of

other authors [23, 24], who revealed that these antibodies

were more frequently expressed in patients with thrombotic

complications of SLE and suggested that these antibodies

had LA properties [24]. Our studies have revealed a weak

negative correlation between plasma IgG aANX antibodies

levels and APTT in the whole SLE group. This seems to

contradict LA properties of aANX antibodies since LA

typically leads to APTT elongation in vitro and pro-

thrombotic tendency in vivo. As already mentioned, the

pathogenic role of aANX antibodies in humans has not

been fully clarified yet [1]. Differences between our results

and those of other authors [25] might reflect the lack of

standardized tests, use of different-specificity reagents, or

the fact of study parameters being determined in serum or

plasma samples [24, 26].

We conclude that in SLE patients, the annexinopathy

does exist, and it could be suggested that, in patients treated

for mild-to-moderate SLE, elevated plasma ANX A5 level

and IgG aANX A5 antibodies concentration are associated

with the severity of the disease. The obtained results

indicate that the role of aANX A5 and IgG CL antibodies

should be taken into consideration as causative factors of

the increased plasma ANX A5 levels. The relationships

between plasma ANX A5 level and the peripheral blood

thrombocyte count were revealed. Noticeable negative

correlation between IgG aANX A5 antibody levels with

APTT in the total SLE group seems to contradict the

suggested LA properties of these antibodies. The observed

connection between plasma ANX A5 level and serum

cholesterol concentration deserves further exploration.

These data and their possible advantages for the diagnosis,

management, and prognosis in SLE require further ran-

domized clinical studies carried out on larger groups of

untreated and treated patients with SLE.

Conflict of interest The authors declare that they have no conflict

of interest.

References

1. Bozic B, Irman S, Gaspersic N, Kveder T, Blaz R (2005) Anti-

bodies against annexin A5: detection pitfalls and clinical asso-

ciations. Autoimmunity 38:425–430

2. Ulander V-M, Stefanovic V, Masuda J, Suzuki K, Hiilesmaa V,

Kaaja R (2007) Plasma levels of annexin IV and V in relation to

antiphospholipid antibody status in woman with a history of

recurrent miscarriage. Thromb Res 120:865–870

3. Cederholm A, Frostegard J (2005) Annexin A5 in cardiovascular

disease and systemic lupus erythematosus. Immunobiology 210:

761–768

4. Cederholm A, Frostegard J (2007) Annexin A5 as a novel player

in prevention of atherothrombosis in SLE and in the general

population. Ann N Y Acad Sci 1108:96–103

5. Tripathy NK, Sinha N, Nityanand S (2003) Anti-annexin V

antibodies in Takayasu’s arteritis: prevalence and relationship

with disease activity. Clin Exp Immunol 134:360–364

6. Zabek J, Wojciechowska B, Alekberova Z, Reshetniak T, Bier-

nacka E, Nasonova V, Karlik W (2003) Anti-annexin V anti-

bodies as a risk factor of thrombotic events and fetal loss in sera

of SLE, PAPS and SAPS patients. Reumatologia 41:12–24

7. Esposito G, Tamby MC, Chanseaud Y, Servettaz A, Guillevin L,

Mouthon L (2005) Anti-annexin A5 antibodies: are they pro-

thrombotic? Autoimmun Rev 4:55–60

8. Marchewka Z (2006) Low molecular weight biomarkers in the

nephrotoxicity. Adv Clin Exp Med 15:1129–1138

9. Rand JH, Wu XX, Andree HAM, Ross JBA, Rusinova E, Gas-

con-Lema MG, Calandri C, Harpel PC (1998) Antiphospholipid

antibodies accelerate plasma coagulation by inhibiting annexin-V

binding to phospholipids: a ‘‘lupus procoagulant’’ phenomenon.

Blood 92:1652–1660

10. Bizzaro N, Tonutti E, Villata D, Tampoia M, Tozzoli R (2005)

Prevalence and clinical correlation of anti-phospholipid-binding

protein antibodies in anticardiolipin-negative patients with sys-

temic lupus erythematosus and women with unexplained recur-

rent miscarriages. Arch Pathol Lab Med 129:61–68

11. Asian H, Pay S, Gok F, Baykal Y, Yilmaz MI, Sengul A, Aydin

HI (2004) Antiannexin V autoantibody in thrombophilic Behcet’s

disease. Rheumatol Int 24:77–79

12. Van Heerde WL, Reutelingsperger CPM, Maassen C, Lux P,

Derksen RHWM, De Groot PG (2003) The presence of anti-

phospholipid antibodies is not related to increased levels of

annexin A5 in plasma. J Thromb Haemost 1:532–536

13. Satoh A, Suzuki K, Takayma E, Kojima K, Hidaka T, Kawakami

M, Matsumoto I, Ohsuzu F (1999) Detection of anti-annexin IV

and V antibodies in patients with antiphospholipid syndrome and

systemic lupus erythematosus. J Rheumatol 26:1715–1720

14. Cederholm A, Svenungsson E, Jensen-Urstad K, Trollmo Ch,

Ulfgren A-K, Swedenborg J, Fei G-Z, Frostegard J (2005)

Decreased binding of annexin V to endothelial cells: potential

mechanism in atherothrombosis of patients with systemic lupus

erythematosus. Arter Thromb Vase Biol 25:198–203

15. Su K-Y, Pisetsky DS (2009) The role of extracellular DNA in

autoimmunity in SLE. Scand J Immunol 70:175–183

16. Ishikura K, Wada H, Kamikura Y, Hattori K, Fukuzawa T,

Yamada N, Nakamura M, Nobori T, Nakano T (2004) High

prevalence of anti-prothrombin antibody in patients with deep

vein thrombosis. Am J Hematol 76:338–342

17. Cieslak D, Hrycaj P (2007) From apoptosis to autoimmuniza-

tion—a new view into the pathogenesis of systemic lupus ery-

thematosus. Reumatologia 45:382–385

18. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield

NF, Schaller JG, Talal N, Winchester RJ (1982) The revised

criteria for the classification of systemic lupus erythematosus.

Arthritis Rheum 25:1271–1277

19. Hochberg MC (1997) Updating the American College of Rheu-

matology revised criteria for the classification of systemic lupus

erythematosus. Arthritis Rheum 40:1725

20. Hrycek A (2004) Immunomodulatory properties of prolactin. Pol

Arch Med Wewn 112:269–270

Rheumatol Int (2012) 32:1335–1342 1341

123

21. Rand JH, Wu XX, Quinn AS, Taatjes DJ (2008) Resistance to

annexin A5 anticoagulant activity: a thrombogenic mechanism

for the antiphospholipid syndrome. Lupus 17:922–930

22. Shojale M, Sotoodah A, Roozmeh S, Kholoosi E, Dana S (2009)

Annexin V and anti-annexin V antibodies: two interesting aspects

in acute myocardial infarction. Thromb J 7:13

23. Nojima J, Kuratsune H, Suchisa F, Futsukaichi Y, Yamanishi H,

Machii T, Iwatani Y, Kanakura Y (2001) Association between the

prevalence of antibodies to b2 -glycoprotein I, prothrombin,

protein C, protein S, and annexin V in patients with systemic

lupus erythematosus and thrombotic and thrombocytopenic

complications. Clin Chem 47:1008–1015

24. Kaburaki J, Kuwana M, Yamamoto M, Kawai S, Ikeda Y (1997)

Clinical significance of anti-annexin V antibodies in patients with

systemic lupus erythematosus. Am J Hematol 54:209–213

25. Nakamura N, Kuragaki C, Shidara Y, Yamaji K, Wada Y (1995)

Antibody to annexin V has anti-phospholipid and lupus antico-

agulant properties. Am J Hematol 49:347–348

26. Szodoray P, Tarr T, Tumpek J, Kappelmayer J, Lakos G, Poor G,

Szegedi G, Kiss E (2009) Identification of rare anti-phospholipid/

protein co-factor autoantibodies in patients with systemic lupus

erythematosus. Autoimmunity 42:497–506

1342 Rheumatol Int (2012) 32:1335–1342

123