British Journal of Rheumatology 1997;36:964±968
AUTOANTIBODIES AGAINST OXIDIZED LOW-DENSITY LIPOPROTEIN
IN ANTIPHOSPHOLIPID SYNDROME
O. AMENGUAL, T. ATSUMI, M. A. KHAMASHTA, F. TINAHONES* and
G. R. V. HUGHES
Lupus/Arthritis Research Unit, The Rayne Institute, St Thomas' Hospital, London and *Endocrinology
Department, Hospital Regional Carlos Haya, MaÂlaga, Spain
SUMMARY
The prevalence and clinical signi®cance of anti-oxidized low-density lipoprotein antibodies (anti-ox-LDL) were evaluated
in patients with the antiphospholipid syndrome (APS). Anti-ox-LDL were measured in the sera of 107 patients with APS
(64 primary APS, 43 secondary to systemic lupus erythematosus) by enzyme-linked immunosorbent assay (ELISA) utilizing
malondialdehyde (MDA)-modi®ed LDL as antigen. In the same patients, anticardiolipin antibodies (aCL) and anti-b2-glycoprotein I antibodies (anti-b2GPI) were also measured. A positive titre of anti-ox-LDL was detected in 22% of patients, but
only in 6% of control subjects (w2 = 12, P = 0.0005). Levels of anti-ox-LDL were higher in patients with arterial thrombosis
(n = 58) than in those without (n = 49) (P = 0.0001). Anti-ox-LDL levels correlated weakly with those of aCL (r = 0.196,
P = 0.043), but not with those of anti-b2GPI (r = 0.076). Our ®ndings suggest that elevated levels of anti-ox-LDL may
represent another potential marker of APS, particularly of patients prone to arterial thrombosis.
KEY WORDS: Anticardiolipin antibodies, Anti-b2-glycoprotein I antibodies, Atherosclerosis, Thrombosis, Systemic lupus
erythematosus.
(aCL) [12]. The same study suggested cross-reactivity
between aCL and anti-ox-LDL. Since aCL is one of
the antiphospholipid antibodies associated with the
antiphospholipid syndrome (APS), a thrombophilic
disorder characterized by arterial and venous thrombosis, recurrent fetal losses and thrombocytopenia
[13], we investigated the prevalence and clinical signi®cance of anti-ox-LDL in APS.
LOW-DENSITY lipoprotein (LDL) is a hydrophilic complex of lipids and apoliprotein B100, and represents
one of the major cholesterol-carrier lipoproteins in
plasma. Epidemiological studies have established that
an elevated plasma level of LDL represents one of
the most important risk factors for the development
of atherosclerosis [1]. In vitro studies have shown that
LDL can undergo several chemical modi®cations,
such as acetylation and oxidation [2]. The latter process is of great interest because oxidation of LDL
may occur in vivo [3±5] and may contribute to the
development of atherosclerosis, as suggested by the
presence of oxidized-LDL (ox-LDL) particles in the
early phase of atherosclerotic plaque formation [6, 7].
Structural changes of LDL may enhance LDL uptake
by macrophage scavenger receptors, promoting the
transformation of macrophages into foam cells [8],
and may favour the recruitment and migration of
monocytes and leucocytes within arterial vessels [9].
On the other hand, oxidatively modi®ed LDL is
more immunogenic than its native counterpart, eliciting speci®c anti-ox-LDL antibodies (anti-ox-LDL)
[2]. The latter have been detected in human sera from
a variety of in¯ammatory conditions. Initially
reported in patients with chronic periaortitis [10],
they were subsequently detected in subjects with carotid atherosclerosis where they represented a marker
of progressive disease [11], and they were also found
in 80% of patients with systemic lupus erythematosus
(SLE) with and without anticardiolipin antibodies
PATIENTS AND METHODS
Patients
A total of 107 patients were included in the study
[94 female and 13 male; mean age 41 yr (range 22±
66)]. Of these, 64 patients had primary APS (60%)
and 43 had APS secondary to SLE (40%). Clinical
features of the patients are reported in Table I. All
patients ful®lled the proposed criteria for the APS
[14]. One hundred and four sex- and age-matched
healthy controls were also included.
LDL isolation
Human LDL was isolated from pooled plasma of
healthy fasting adults by density gradient ultracentrifugation with BrK (Beckman L8-70 ultracentrifuge,
rotor VTI 65) at 65 000 r.p.m. for 35 min, followed
by a second ultracentrifugation with BrK at 49 000
r.p.m. for 18 h. The LDL layer was then dialysed for
30 h against phosphate-bu€ered saline (PBS) (0.14 M
NaCl/0.01 M phosphate bu€er). Puri®ed LDL showed
a single band on 1% agarose gel electrophoresis in
borate bu€er.
Modi®cation of LDL
Malondialdehyde (MDA) was freshly generated
from malonaldehyde bis dimethylacetal by acid
hydrolysis as described by Palinski et al. [15]. MDA-
Submitted 31 December 1996; revised version accepted 10 March
1997.
Correspondence to: M. A. Khamashta, Lupus Research Unit,
The Rayne Institute, St Thomas' Hospital, London SE1 7EH.
# 1997 British Society for Rheumatology
964
AMENGUAL ET AL.: ANTI-ox-LDL ANTIBODIES IN APS
TABLE I
Patient characteristics
Primary APS
Secondary APS to SLE
aCL
Lupus anticoagulant
Thrombosis
Arterial
Venous
Arterial + Venous
Recurrent miscarriages
Thrombocytopenia
No.
Percentage
64
43
75
78/95
91
58
55
23
38/94
22
60
40
70
82
84
54
51
21
40
21
APS, antiphospholipid syndrome; SLE, systemic lupus erythematosus; aCL, anticardiolipin antibodies
LDL was prepared by incubating LDL (100 ml/mg)
for 3 h at 378C with 0.5 M MDA. The reaction was
stopped by adjusting the pH to 7.4 with NaOH.
After conjugation, MDA-LDL was dialysed extensively against PBS.
Anti-ox-LDL ELISA
Half of a microtitre plate (Immulon 4; Dynatech
Laboratories Inc., VA, USA) was coated with LDL
and the other half with MDA-LDL, both at 5 mg/ml
in PBS containing 2 mM ethylenediaminetetra-acetic
acid tetrasodium salt (EDTA) (BDH Chemicals Ltd,
Poole) and 20 mM butylated hydroxytoluene (Sigma
Chemical Co., St Louis, MO, USA), and incubated
at 378C for 2 h and then overnight at 48C. Plates
were washed four times with PBS containing 0.05%
Tween 20 (Sigma) (PBS-Tween) and wells were
blocked with 150 ml of 0.5% gelatin (BDH) for 1 h at
378C. After washing, 50 ml of serum diluted with
PBS-Tween containing 1% bovine serum albumin
(BSA) (Sigma) at 1:100 were added in duplicate.
Plates were incubated for 2 h at room temperature,
and washed four times. Fifty microlitres per well of
the appropriate dilution of alkaline phosphatase-conjugated goat anti-human IgG (Sigma) in PBS-Tween
containing 1% BSA were added and incubated for
1 h at room temperature. After four washes, 100 ml/
well of 1 mg/ml p-nitrophenylphosphate disodium
(Sigma) in 1 M diethanolamine bu€er (pH 9.8) were
added. Following colour development, optical density
at 405 nm (OD 405) was measured by a Titertek
Multiskan MC apparatus (Flow Laboratories, Herts).
Results were expressed as OD, and binding to oxLDL was calculated by subtracting OD 405 obtained
by LDL-coated wells from that by MDA-LDLcoated wells. A normal range was established using
104 controls with a cut-o€ of 172 OD units being
2 S.D. above the mean.
Anti-b2-glycoprotein I antibodies (anti-b2GPI)
ELISA
Anti-b2GPI were detected by ELISA as previously
described [16]. Brie¯y, irradiated microtitre plates
(Sumilon Bakelite type C, Tokyo, Japan) were coated
965
with 50 ml of 4 mg/ml of puri®ed human b2GPI in
PBS at 48C overnight. Wells were blocked with 3%
gelatin for 1 h at 378C. After three washes with PBSTween, 50 ml of serum diluted in PBS containing 1%
BSA at 1:50 were added in duplicate. Plates were
incubated for 1 h at room temperature, followed by
alkaline phosphatase-conjugated goat anti-human
IgG and substrate.
Statistical analysis
All statistical analysis was performed by Statview
II (Apple Macintosh software). Comparisons were
determined by w2 test or Mann±Whitney non-parametric test.
RESULTS
Positivity for anti-ox-LDL was found in 22% (24/
107) of APS patients, but in only 6% (6/104) of controls (w2 = 12, P = 0.0005). Likewise, levels of antiox-LDL were higher in patients than in controls
(P = 0.0001) (Fig. 1). In the patient group, a positive
level of anti-ox-LDL was detected in 17% (11/64) of
primary APS patients and in 30% (13/43) of patients
whose APS was secondary to SLE. By correlating
levels of anti-ox-LDL with clinical manifestations of
APS (arterial/venous thrombosis, miscarriages and
thrombocytopenia), it was found that patients with a
history of arterial thrombosis had higher levels of
anti-ox-LDL than patients without (P = 0.0001)
(Fig. 2), but no correlation was found between levels
of anti-ox-LDL and a history of venous thrombosis
or miscarriages (data not shown). A positive titre of
aCL IgG and anti-b2GPI IgG was found in 70%
(75/107) and 49% (52/107) of patients, respectively.
Positivity for anti-ox-LDL was more frequent in
patients with aCL IgG (31%, 23/75) than in those
without aCL IgG (3%, 1/32) (w2 = 9.7, P = 0.0018),
but no di€erence in positivity for anti-ox-LDL was
FIG. 1.Ð Levels of IgG anti-ox-LDL (optical density values) in
patients with APS (n = 107) and in healthy controls (n = 104).
Levels of anti-ox-LDL were higher in APS patients than in controls.
966
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 9
FIG. 2.Ð Presence of anti-ox-LDL in APS patients with or without
arterial thrombosis. APS patients with arterial thrombosis showed
higher levels of anti-ox-LDL than those without.
noted between patients with (29%, 5/52) or without
anti-b2GPI (16%, 9/55). Anti-ox-LDL IgG levels correlated weakly with the titres of aCL IgG (r = 0.196,
P = 0.043), but not with those of anti-b2GPI
(r = 0.076) (Fig. 3).
DISCUSSION
The present study reveals that anti-ox-LDL antibodies are often found in patients with APS, both
primary and secondary. Levels of anti-ox-LDL were
found to be higher in patients with a history of arterial thrombosis compared to those without.
It has been hypothesized that ox-LDL may combine with anti-ox-LDL, leading to the formation of
an immune complex, the uptake of which by Fc
receptors on macrophages occurs in synergy but faster than by the scavenger pathway [2, 17]. An in vitro
study showed that uptake of radiolabelled ox-LDL
by a monocyte/macrophage-like cell line was more
rapid in the presence of anti-ox-LDL than the uptake
of ox-LDL alone [2]. On the other hand, it may be
that the mechanism described contributes to the elimination of excess ox-LDL produced in some disease
states, and that anti-ox-LDL in serum may represent
a marker of inappropriate ox-LDL generation [18].
Antibodies against ox-LDL have been detected in
a variety of diseases, including SLE [12, 19±23].
Although the role of anti-ox-LDL remains unclear,
there are several pieces of evidence supporting their
involvement in the development of atherosclerosis:
(a) the correlation between anti-ox-LDL and disease
progression in carotid atherosclerosis [11]; (b) the
predictive value of anti-ox-LDL for myocardial infarction [18]; (c) the presence of ox-LDL by immunostaining in atherosclerotic plaques of human aortas
[7]; (d) the in vitro e€ects of ox-LDL on monocyte/
macrophages [2]. Clari®cation of the latter issue is of
practical importance because it contributes signi®cantly towards the morbidity and mortality of a€ected
patients. In our study with APS patients, we did not
®nd a convincing relationship between the titres of
aCL and anti-ox-LDL, in agreement with earlier data
from Vaarala et al. [12], eventually implying that
anti-ox-LDL may represent a di€erent autoantibody
in the APS scenario. There are some reports suggesting cross-reactivity between aCL and anti-ox-LDL.
Initially, this was reported by Vaarala et al. [12], in
SLE patients, showing that aCL activity was inhibited by ox-LDL in the ELISA system. Recent studies
proved the cross-reaction using murine monoclonal
antibodies [24]. Monoclonal anti-ox-LDL raised from
apo E-de®cient mice bound to di€erent immunogenic
structures generated by the oxidation of LDL, and
some of them bound strongly to cardiolipin as it was
progressively oxidized [25]. On the other hand, monoclonal aCL established from NZW/BXSBF1 mice
(both b2GPI dependent and independent) bound to
ox-LDL [26]. In another study, puri®ed IgG from
FIG. 3.Ð Correlation of the levels of anti-ox-LDL with those of (A) aCL and (B) anti-b2GPI in sera from patients with APS.
AMENGUAL ET AL.: ANTI-ox-LDL ANTIBODIES IN APS
SLE patients containing aCL was fractionated
according to their anity for cardiolipin and oxLDL binding was found in some fractions, implying
that at least some aCL may have binding activity to
ox-LDL [23]. However, b2GPI-dependent murine
monoclonal aCL were reported to bind ox-LDL only
in the presence of b2GPI, showing that monoclonal
aCL recognize a cryptic epitope on b2GPI which
bound to lipids contained in ox-LDL molecules [27].
Thus, some but not all aCL may cross-react with
ox-LDL.
A number of mechanisms have been hypothesized
in the pathophysiology of thrombosis in APS [28]
and the atherosclerosis process could be implicated as
one of them [29±31]. Oxidative modi®cation of LDL
may occur as a result of free radical generation in the
blood stream, particularly on the arterial side where
primed neutrophils may release reactive oxygen species [29]. In keeping with this concept, we feel that
our data regarding higher anti-ox-LDL levels in APS
patients with a history of arterial thrombosis may
represent a novel marker for arterial disease in APS.
We used MDA-LDL as an antigen in ELISA.
MDA is a highly reactive dialdehyde generated during arachidonic acid catabolism in thrombocytes and
it is also produced from lipid peroxidation during
phagocytosis of monocytes. Thus, MDA-LDL may
be a `physiological' ox-LDL and it is known that
MDA-LDL favours the binding of anti-ox-LDL as
well as Cu2+-oxidized LDL [15]. Our anti-ox-LDL
ELISA was carried out in the absence of b2GPI in
order to reduce the heterogeneity of antibodies
detected by the assay. In that way, we reduced the
chances that our assay could detect b2GPI-dependent
aCL (= anti-b2GPI). In fact, the levels of anti-oxLDL did not correlate with those of anti-b2GPI and
the correlation with aCL was extremely weak.
Therefore, anti-ox-LDL may represent a distinct subset of antibodies co-existing with b2GPI-dependent
aCL (speci®city directed towards the epitope on
b2GPI) and with conventional `aCL' detected by
standard aCL assay [16].
In conclusion, our data from a large number of
APS patients suggest that high titres of anti-ox-LDL
may be a potential marker of arterial disease in APS
and anti-ox-LDL might contribute to a better characterization of this syndrome.
ACKNOWLEDGEMENTS
This work was supported by grants from Fondo de
InvestigacioÂn de la Seguridad Social of Spain (FIS
95/5011, FIS 96/5093), the Daiwa Anglo-Japanese
Foundation and Lupus UK. The authors wish to
thank Dr P. R. J. Ames (Haematology Department,
St Thomas' Hospital, London) for his helpful suggestions and comments on this manuscript.
REFERENCES
1. Brown MS, Goldstein JL. The hyperlipoproteinemias
and other disorders of lipid metabolism. In: Isselbacher
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
967
KJ, ed. Harrison's principles of internal medicine, 13th
edn, Vol. 2. New York: McGraw-Hill, Inc. 1994:2058±
69.
Wu R, Lefvert AK. Autoantibodies against oxidized
low density lipoproteins (oxLDL): characterization of
antibody isotype, subclass, anity and e€ect on the
macrophage uptake of oxLDL. Clin Exp Immunol
1995;102:174±80.
Palinski W, Rosenfeld ME, YlaÈ-Herttuala S et al. Lowdensity lipoprotein undergoes oxidative modi®cations
in vivo. Proc Natl Acad Sci USA 1989;86:1372±6.
Steinberg D, Parthasarathy S, Carew TE, Khoo JD,
Witztum JL. Beyond cholesterol: modi®cations of
low-density lipoprotein that increase its atherogenicity.
N Engl J Med 1989;320:915±24.
Steinberg D, Witztum JL. Lipoproteins and atherogenesis: current concepts. J Am Med Assoc 1990;264:
3047±52.
YlaÈ-Herttuala S, Palinski W, Rosenfeld ME et al.
Evidence for the presence of oxidatively modi®ed lowdensity lipoprotein in atherosclerotic lesions of rabbit
and man. J Clin Invest 1989;85:1086±95.
O'Brien KD, Alpers CE, Hokanson JE, Wang S, Chait
A. Oxidation-speci®c epitopes in human coronary
atherosclerosis are not limited to oxidized low-density
lipoprotein. Circulation 1996;94:1216±25.
Witztum JL, Steinberg D. Role of oxidized low density
lipoprotein in atherogenesis. J Clin Invest 1991;88:
1785±92.
Berliner JA, Territo MC, Sevanian A et al. Minimally
modi®ed low-density lipoprotein stimulates monocyte
endothelial interactions. J Clin Invest 1990;85:1260±6.
Parums DV, Brown DL, Mitchinson MJ. Serum antibodies to oxidized low density lipoprotein and ceroid
in chronic periaortitis. Arch Pathol Lab Med
1990;114:383±7.
Salonen JK, YlaÈ-Herttuala S, Yamamoto R et al.
Autoantibody against oxidized LDL and progression of
carotid atherosclerosis. Lancet 1992;339:883±7.
Vaarala O, Alfthan G, Jauhiainen M, Leirisalo-Repo
M, Aho K, Palosuo T. Crossreaction between antibodies to oxidised low-density lipoprotein and to
cardiolipin in systemic lupus erythematosus. Lancet
1993;341:923±5.
Hughes GRV. The antiphospholipid syndrome: ten
years on. Lancet 1993;342:341±4.
Harris EN. Antiphospholipid antibodies. Br J
Haematol 1990;74:1±9.
Palinski W, YlaÈ-Herttuala S, Rosenfeld ME et al.
Antisera and monoclonal antibodies speci®c for epitopes generated during oxidative modi®cation of low
density lipoprotein. Arteriosclerosis 1990;10:325±35.
Amengual O, Atsumi T, Khamashta MA, Koike T,
Hughes GRV. Speci®city of ELISA for antibody to b2glycoprotein I in patients with antiphospholipid syndrome. Br J Rheumatol, 1996;35:1239±43.
Khoo JC, Miller E, Pio F, Steinberg D, Witztum JL.
Monoclonal antibodies against LDL further enhance
macrophage uptake of LDL aggregates. Arterioscler
Thromb 1992;12:1258±66.
Puurunen M, MaÈnttaÈri M, Manninen V et al.
Antibody against low-density lipoprotein predicting
myocardial
infarction.
Arch
Intern
Med
1994;154:2605±9.
Bellomo G, Maggi E, Poli M, Agosta FG, Bollati P,
Finardi G. Autoantibodies against oxidatively modi®ed
968
20.
21.
22.
23.
24.
25.
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 9
low-density lipoproteins in NIDDM. Diabetes
1995;44:60±6.
Boullier A, Hamon M, Walters-Laporte E et al.
Detection of autoantibodies against oxidized low-density lipoproteins and of IgG-bound low density lipoproteins in patients with coronary artery disease. Clin
Chim Acta 1995;238:1±10.
Maggi E, Bellazzi R, Falaschi F et al. Enhanced LDL
oxidation in uremic patients: An additional mechanism
for accelerated atherosclerosis?. Kidney Int 1994;45:
876±83.
Maggi E, Marchesi E, Ravetta V, Martignoni A,
Finardi G, Bellomo G. Presence of autoantibodies
against oxidatively modi®ed low-density lipoprotein in
essential hypertension: a biochemical signature of an
enhanced in vivo low-density lipoprotein oxidation.
Hypertension 1995;13:129±38.
Vaarala O, Puurunen M, Lukka G et al. Anity-puri®ed cardiolipin-binding antibodies show heterogeneity
in their binding to oxidized low-density lipoprotein.
Clin Exp Immunol 1996;104:269±74.
Palinski W, HoÈrkkoÈ S, Miller E et al. Cloning of
monoclonal autoantibodies to epitopes of oxidized lipoproteins from apolipoprotein E-de®cient mice. J Clin
Invest 1996;98:800±14.
HoÈrkkoÈ S, Miller E, Dudl E et al. Antiphospholipid
antibodies are directed against epitopes of oxidized
phospholipids. J Clin Invest 1996;98:815±25.
26. Mizutani H, Kurata Y, Kosugi S et al. Monoclonal
anticardiolipin autoantibodies established from the
(New Zealand White BXSB)F1 mouse model of antiphospholipid syndrome cross-react with oxidized lowdensity lipoprotein. Arthritis Rheum 1995;38:1382±8.
27. Matsuura E, Katahira T, Igarashi Y, Koike T. b2Glycoprotein I bound to oxidatively modi®ed lipoproteins could be targeted by anticardiolipin antibodies.
Lupus 1994;3:314.
28. Roubey RAS. Autoantibodies to phospholipid-binding
plasma proteins: a new view of lupus anticoagulants
and other `antiphospholipid' autoantibodies. Blood
1994;84:2854±67.
29. Ames PRJ. Antiphospholipid antibodies, thrombosis
and atherosclerosis in systemic lupus erythematosus: a
unifying `Membrane stress syndrome' hypothesis.
Lupus 1994;3:371±7.
30. Asherson RA, Khamashta MA, Baguley E, Oakley
CM, Rowell NR, Hughes GRV. Myocardial infarction
and antiphospholipid antibodies in systemic lupus
erythematosus and related disorders. Q J Med
1989;73:1103±15.
31. Nihoyannopoulos P, Gomez PM, Joshi J, Loizou S,
Walport MJ, Oaley MJ. Cardiac abnormalities in systemic lupus erythematosus. Association with raised
anticardiolipin antibodies. Circulation 1990;82:369±75.