вход по аккаунту


Induction of endothelial cell apoptosis by the binding of antiendothelial cell antibodies to Hsp60 in vasculitis-associated systemic autoimmune diseases.

код для вставкиСкачать
Vol. 52, No. 12, December 2005, pp 4028–4038
DOI 10.1002/art.21401
© 2005, American College of Rheumatology
Induction of Endothelial Cell Apoptosis by the
Binding of Anti–Endothelial Cell Antibodies to Hsp60 in
Vasculitis-Associated Systemic Autoimmune Diseases
Christophe Jamin,1 Christophe Dugué,1 Jean-Éric Alard,1 Sandrine Jousse,1 Alain Saraux,1
Loı̈c Guillevin,2 Jean-Charles Piette,3 and Pierre Youinou1
anti-Hsp60–containing AECA-positive sera and was inhibited by preincubation of the ECs with recombinant
Conclusion. Our data support the notion that
Hsp60 is an important target for AECAs and that such
an interaction contributes to pathogenic effects, especially
in vasculitis-associated systemic autoimmune disease.
Objective. Anti–endothelial cell antibodies
(AECAs), which recognize a number of endothelial
antigens, are seen in patients with systemic autoimmune
diseases, more often in the presence of vasculitis than in
its absence. Some AECAs induce apoptosis of endothelial cells (ECs), but their target antigens remain unknown. The aim of this study was to determine whether
Hsp60 is a target antigen and whether AECAs induce
apoptosis in ECs.
Methods. Two-dimensional electrophoresis and
conventional Western blotting techniques were used to
characterize AECA targets. Hsp60 reactivity was determined by enzyme-linked immunosorbent assay.
Results. Hsp60 was shown to be targeted by a
proportion of AECAs. The level of reactivity was higher
in patients with systemic autoimmune disease and vasculitis than in those without vasculitis and in patients
with systemic lupus erythematosus than in patients with
other systemic autoimmune diseases. Hsp60 was expressed on the plasma membrane of heat-stressed ECs,
and this followed Hsp60 messenger RNA transcription,
confinement of the protein to the cytoplasm, and translocation of the protein to the surface. Shedding of Hsp60
from ECs was induced by stress and resulted in the
binding of soluble Hsp60 to the surface of ECs, particularly stressed ECs. Apoptosis of ECs was triggered by
The 60-kd heat-shock protein (Hsp60) acts as a
chaperone in the folding, translocation, and assembly of
protein complexes throughout their posttranslational
maturation (1). Although originally restricted to the
mitochondrial matrix, Hsp60 is then exported to extramitochondrial sites, most particularly the plasma membrane (2). As a result, it becomes detectable on different
cell types, including oligodendrocytes (3), macrophages
(4), B cell lymphoma cells (5), pancreatic carcinoma
cells (6), epithelial cells (7), and endothelial cells (ECs)
(8). It has also been established that a diverse array of
metabolic stimuli, such as hypoxia, heat shock, fluid
pressure, cytokines, and inflammatory reactions, induce
the expression of Hsp60 on ECs (8,9). In these settings,
Hsp60 expression acts as a protective mechanism to
maintain homeostasis.
This seminal finding prompted studies of the
immunologic consequences of Hsp60 expression, such as
those in which it presents immunodominant antigens
(10). Specific T cell activation and anti-Hsp60 antibody
production have been characterized in infectious diseases, including leprosy, tuberculosis, and malaria (11).
Autoantibodies to Hsp60 have also been identified in
systemic autoimmune diseases, including rheumatoid
arthritis (RA) (12), systemic lupus erythematosus (SLE)
(13), and diabetes mellitus (14). In fact, immune reactivity to Hsp60 has been most extensively analyzed in
atherosclerosis, which has thus been considered to be an
Supported by a grant from the Brest Métropole Océane and
from the Conseil Régional de Bretagne, France.
Christophe Jamin, PhD, Christophe Dugué, PhD, Jean-Éric
Alard, BSc, Sandrine Jousse, MD, Alain Saraux, MD, PhD, Pierre
Youinou, MD, PhD: Brest University Medical School, Brest, France;
Loı̈c Guillevin, MD: Cochin Hospital, Paris, France; 3Jean-Charles
Piette, MD: La Pitié-Salpêtrière Hospital, Paris, France.
Address correspondence and reprint requests to Pierre Youinou, MD, PhD, Immunology, Brest University Medical School Hospital, BP824, Brest F29609, France. E-mail:
Submitted for publication April 19, 2005; accepted in revised
form August 2, 2005.
autoimmune condition (8,15). Although the association
of anti-Hsp60 antibody with disease has not been completely characterized (16), the general perception is that
the interaction of autoantibodies with Hsp60 proteins
might be involved in the pathogenesis of systemic autoimmune diseases. They trigger complement-mediated
cytotoxicity and antibody-dependent cell-mediated cytotoxicity (8) at the expense of macrophages and ECs.
Several investigators, including our group, have
reported the presence of anti-EC antibodies
(AECAs) in autoimmune conditions, such as SLE (17),
polyarteritis nodosa (PAN) (18), and systemic sclerosis
(SSc) (19). Furthermore, AECAs have also been found
in patients with infectious diseases, such as leprosy (20),
hepatitis virus C infection (21), and cytomegalovirusinfected recipients of cardiac or renal allografts (22), as
well as in patients with borderline hypertension (23).
Interestingly, the binding of certain AECAs induces
apoptosis in ECs, both in vitro (24) and in vivo (25).
Given the impressive diversity of diseases associated
with AECAs, their target antigens are likely to be
This study was aimed at determining whether
Hsp60 is one of the target antigens of AECAs and
whether such autoantibodies induce the apoptosis of
ECs. Using recombinant Hsp60, we demonstrated that
reactivity of a group of AECAs with this antigen is more
frequent in patients with systemic autoimmune diseases
who also have vasculitis than in those who do not.
Furthermore, autoantibodies that bind Hsp60 exposed
on the surface of heat-stressed ECs or shed Hsp60 that
is captured by unknown receptors on the surface of ECs
induce apoptosis. These results suggest a role of AECA
interaction with Hsp60 in the pathogenesis of vasculitisassociated systemic autoimmune diseases.
Patients. Sera were obtained from 2 groups of patients
with systemic autoimmune disease. The first group comprised
176 patients with systemic autoimmune disease associated with
vasculitis: 51 with SLE, 45 with Wegener’s granulomatosis
(WG), 36 with PAN, 16 with microscopic polyangiitis (MPA),
16 with Churg-Strauss syndrome (CSS), and 12 with Behçet’s
disease. The second group consisted of 95 patients with
systemic autoimmune disease, but without vasculitis: 26 with
polymyositis, 15 with RA, 41 with primary Sjögren’s syndrome
(SS), and 13 with SSc. All patients fulfilled the criteria for their
respective diseases. Because patients with RA (26) and patients with primary SS (27) can potentially present with
features of vasculitis, we took care that none of them had signs
of vasculitis at the time their serum samples were collected.
Sera from 30 healthy laboratory and clinical staff
members were used as controls to set the cutoff level for
AECA and anti-Hsp60 antibody positivity in our enzymelinked immunosorbent assays (ELISAs). In the 10 patient
subgroups and in the control population, the ratios of women
to men were ⬃3:1, and the mean ages of the subjects ranged
from 23 to 66 years.
Measurement of EC reactivity. EA.hy926 cells were
kindly donated by Dr. Cora-Jean S. Edgell (University of
North Carolina, Chapel Hill, NC). AECAs were identified by
Western blotting on EA.hy926 cell lysates, using a membrane
preparation method described in detail elsewhere (28). The
pattern of reactivity of each serum was determined using a
G-700 imaging densitometer and analyzed with Molecular
Analyst software (both from Bio-Rad, Marnes la Coquette,
France). A serum was considered positive when enhanced
intensity for a given peak (optical density [OD] ⬎0.1) was
identified relative to the controls or when an additional peak
over an OD threshold arbitrarily set at 0.1 was found.
Two-dimensional electrophoresis (2DE) followed by
Western blotting were also performed as previously described
(20). For isoelectric focusing, protein samples were loaded
onto nonlinear, pH 3–10, immobilized pH gradient strips
(Pharmacia, Uppsala, Sweden) and focused for 45,000 V/hour.
The strips were incubated for 20 minutes at room temperature
with 50 ml of an equilibration buffer containing 0.05M Tris
HCl, pH 6.8, 6M urea, 1% sodium dodecyl sulfate, 30%
glycerol, and 0.5% dithiothreitol (DTT). The strips were then
incubated in the equilibration solution, which contained 4.5%
iodoacetamide (Sigma-Aldrich, St. Louis, MO) instead of
0.5% DTT. Then, 5–15% acrylamide–bisacrylamide gradient
gels prepared in our laboratory were used to separate the
proteins. The strips were loaded onto the top of the slab gel,
subjected to electrophoresis for 4 hours at 50 mA per gel, and
transferred to polyvinylidene difluoride (PVDF) membranes.
After 2 hours of blocking with phosphate buffered saline (PBS)
containing 2% milk protein and 0.05% Tween 20, the PVDF
membranes were probed for 2 hours with sera diluted 1:50 in
PBS supplemented with 1% milk protein and 0.05% Tween 20.
Bound AECAs were revealed with biotinylated anti-human
IgG antibody (Jackson ImmunoResearch, West Grove, PA),
and developed with horseradish peroxidase (HRP)–
streptavidin (Amersham Biosciences, Orsay, France). Autoantigen spots were identified in reference to the EA.hy926 2DE
gel map from the Swiss-Prot database (available at http://
Measurement of Hsp60 reactivity. The presence of
anti-Hsp60 antibodies was determined by Western blotting
using recombinant Hsp60 as the substrate (Sigma-Aldrich).
This protein was loaded onto a 10% acrylamide–bisacrylamide
gel, subjected to electrophoresis, and transferred to a PVDF
membrane. Western blotting with AECA-positive sera was
performed as described above. After densitometric analysis,
the mean ⫹ 2SD of the OD value in control sera was taken as
the threshold for positivity.
Hsp60 reactivity was also detected using an ELISA
developed in our laboratory (29). Briefly, ELISA microtiter
wells were sensitized with 0.5 ␮g of recombinant Hsp60, and
blocked with 2% milk protein. After 3 washes with 0.1%
Tween 20 in PBS, 100 ␮l of serum diluted 1:100 in PBS
containing 0.1% Tween 20 was added in triplicate, and plates
were incubated for 2 hours at 37°C. Incubation with alkaline
phosphatase–conjugated anti-human IgG antibody (Zymed,
South San Francisco, CA) was then performed for a further 2
hours at 37°C, followed by incubation with p-nitrophenyl
phosphate. The absorbance was measured at 405 nm, and the
OD of control wells without Hsp60 was subtracted from the
OD of Hsp60-coated wells. The mean ⫹ 2SD of the OD value
in control sera was taken as the threshold for positivity.
Detection of Hsp60 expression. Human umbilical vein
ECs (HUVECs) were prepared by digestion in 0.1% collagenase (Sigma-Aldrich) and grown to confluence, as previously
described (20). The HUVECs were then passaged twice,
harvested, and seeded at 105 cells in 100 ␮l of Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with 10%
fetal calf serum (FCS) and 100 IU/ml of polymyxin B. Resting
HUVECs or HUVECs subjected to stress by exposure to
temperatures of 42.5°C for 1–5 hours were permeabilized with
0.5% saponin (Sigma-Aldrich), then tested for the expression
of Hsp60. Two hundred microliters of this cell suspension was
incubated with anti-Hsp60 monoclonal antibody (mAb; VWR
International, Fontenay-sous-Bois, France) for 1 hour at 4°C.
The cells were then incubated with fluorescein isothiocyanate
(FITC)–conjugated anti-mouse IgG antibody, washed, and
suspended in 400 ␮l of PBS for fluorescence-activated cell
sorter (FACS) analysis in an Epics Elite flow cytometer
(Coulter, Hialeah, FL). Unstained cells and the FITCconjugated developing reagent alone were the controls used to
set the level of positivity.
For indirect immunofluorescence (IIF) staining, subconfluent HUVECs were permeabilized and fixed in 20%
ethanol plus 80% acetone for 5 minutes following application
of stress. Cells were then incubated with anti-Hsp60 mAb and
revealed using FITC-conjugated anti-mouse IgG antibody.
Slides were mounted in glycerol and examined with an Axioplan fluorescence microscope (Zeiss, Le Pecq, France).
Soluble Hsp60 released into the supernatant of cultured cells was detected by dot-blot analysis. Untreated or
heat-stressed HUVECs were incubated for 18 hours at 37°C.
Soluble proteins were precipitated with acetone and loaded
onto a nitrocellulose membrane. The membrane was blocked
with 2% milk protein, incubated with anti-Hsp60 mAb for 1
hour, revealed with biotinylated anti-mouse IgG antibodies
(Jackson ImmunoResearch), and visualized with HRP–
streptavidin. Recombinant Hsp60 was used as a positive control.
Expression of Hsp60 messenger RNA (mRNA) was
evaluated by quantitative reverse transcriptase–polymerase
chain reaction (RT-PCR). HUVECs were subjected to heat
stress or were untreated, and mRNA was purified. For synthesis of complementary DNA (cDNA), 1 ␮g of mRNA was
incubated for 5 minutes at 68°C and for 1 minute on ice with
an oligo(dT) and deoxynucleotide mixture in a final volume of
11 ␮l. Seven microliters of PCR buffer was added and incubated for 2 minutes at 42°C. One microliter of Superscript II
RT (Invitrogen, Carlsbad, CA) was added, and incubation
continued for 50 minutes at 40°C and then for 15 minutes at
70°C. Finally, 1 ␮l of RNase H was added and incubated for 10
minutes at 37°C.
For PCR amplification, the reactions were performed
in an ABI Prism 7000 system (Applied Biosystems, Warrington, UK) with 1 ␮l of cDNA, 0.5 ␮l of Taq DNA polymerase (5 units/␮l; Invitrogen), and 2 ␮l of the primer pairs
Table 1. Frequencies of AECAs and of anti-Hsp60 antibodies in
AECA-positive sera from patients with vasculitic and nonvasculitic
systemic autoimmune diseases*
Frequency (no.) of patients
By Western
SLE (n ⫽ 51)
WG (n ⫽ 45)
PAN (n ⫽ 36)
MPA (n ⫽ 16)
CSS (n ⫽ 16)
BD (n ⫽ 12)
PM (n ⫽ 26)
RA (n ⫽ 15)
Primary SS (n ⫽ 41)
SSc (n ⫽ 13)
74 (38)
80 (36)
53 (19)
50 (8)
62 (10)
75 (9)
38 (10)
67 (10)
24 (10)
77 (10)
39 (15)
44 (16)
42 (8)
25 (2)
22 (2)
30 (3)
10 (1)
10 (1)
10 (1)
76 (29)
56 (20)
79 (15)
62 (5)
20 (2)
44 (4)
30 (3)
20 (2)
20 (2)
10 (1)
* Anti–endothelial cell antibody (AECA) reactivity was determined by
Western blotting. Anti-Hsp60 reactivity was determined by enzymelinked immunosorbent assay (ELISA) and by Western blotting. SLE ⫽
systemic lupus erythematosus; WG ⫽ Wegener’s granulomatosis; PAN
⫽ polyarteritis nodosa; MPA ⫽ microscopic polyangiitis; CSS ⫽
Churg-Strauss syndrome; BD ⫽ Behçet’s disease; PM ⫽ polymyositis;
RA ⫽ rheumatoid arthritis; SS ⫽ Sjögren’s syndrome; SSc ⫽ systemic
GAGGAACACTGCCTTGGGCTTC-3⬘ for Hsp60, and 5⬘GGCTACCACATCCAAGGAAGGCAG-3⬘ and 5⬘-TTGACACTGGCAAAACAATGCAGAC-3⬘ for 18S as follows. After an initial denaturation step, 35 cycles of PCR, consisting of 1
minute at 58°C and 1 minute at 72°C, were performed, with a final
10-minute extension at 72°C. The relative induction or downregulation of gene expression was obtained using the comparative
threshold cycle (Ct) method and the 2–⌬Ct formula, as follows:
⌬Ct ⫽ Ct Hsp60 – Ct 18S. The relative fold induction of Hsp60
mRNA was estimated, with normalization of the values found
after 1–5 hours of heat-induced stress to the value recorded
before the application of stress. RT-PCR products were visualized on 2% agarose gels using ethidium bromide.
Assessment of apoptosis. IgG fractions of control sera
and patient sera positive for anti-Hsp60 were purified over a
protein G–Sepharose column (Sigma-Aldrich). Their concentrations were determined by an ELISA developed in our
laboratory. HUVECs were harvested from the second passage
of the cultured cells, suspended at a density of 104 cells in 100
␮l of DMEM supplemented with 10% FCS and 100 IU/ml of
polymyxin B, and distributed into 3 sets of duplicate wells. Pilot
experiments were conducted to determine how much Hsp60
was required to obtain maximal inhibition of the binding of
IgG to Hsp60 coated onto an ELISA plate. Then, 100 ␮l of
purified IgG at a concentration of 320 ␮g/ml in the culture
medium was preincubated or was not preincubated with 2.5
␮g/ml of Hsp60 and added to a final concentration of 160
␮g/ml of IgG.
Following a 24-hour incubation, apoptosis was evaluated by measuring FITC-conjugated annexin V staining in
association with propidium iodide (PI) to exclude dead cells
(kit obtained from Beckman-Coulter, Villepinte, France). Per-
Figure 1. Pattern of reactivity of sera from patients with different systemic autoimmune diseases in 1-dimensional and 2-dimensional Western blot
analyses. A, EA.hy926 cell lysates were subjected to electrophoresis and then blotted with sera from patients and controls. Representative results
for sera from patients with systemic lupus erythematosus (SLE), Wegener’s granulomatosis (WG), polyarteritis nodosa (PAN), microscopic
polyangiitis (MPA), and Churg-Strauss syndrome (CSS), as well as serum from a healthy control subject are shown. Molecular weight markers are
shown at the left. B, Densitometric analysis of SLE reactivity (thin line) compared with control reactivity (thick line). Antigens strongly recognized
by SLE sera are indicated. A.U. ⫽ arbitrary units. C, Electrophoresis was performed on EA.hy926 cell lysates and proteins blotted with sera from
patients with SLE, WG, PAN, MPA, and CSS, as well as serum from a healthy control subject. Representative patterns for 2 sera for each disease
are shown. Large arrowheads indicate spots for which the molecular weight and the isoelectric point match those of Hsp60. Molecular weight
markers are shown at the left. D, On another membrane, the spot corresponding to the area indicated by the large arrowheads in C was cut out,
the amino acids were sequenced, and the peptide was aligned with the NH2-terminal end of the Hsp60 protein (Swiss-Prot accession no. P10809),
revealing sequence identity with a 19–amino acid segment.
centages of annexin V–positive cells within the PI-negative
population were calculated. Cells in the early stage of apoptosis were positive for annexin V and negative for PI, because
PI-positive cells were recorded as dead cells. As in previous
studies (24,30), these annexin V data were confirmed by
findings of nuclear condensation, hypoploidy of the cells, and
DNA fragmentation, and by TUNEL staining.
Identification of Hsp60 as a target antigen for
AECAs. Using 1-dimensional Western blotting with
EA.hy926 cell lysates as the substrate, EC reactivity was
identified in 176 sera from patients with various systemic
autoimmune diseases associated with vasculitis and in 95
sera from patients with nonvasculitic systemic autoimmune diseases. Among the subgroups of patients with
associated vasculitis, 50–80% were positive for AECAs,
and among the subgroups of patients without associated
vasculitis, 24–77% were positive for AECAs, depending
on the diagnosis (Table 1).
Western blot analyses revealed a range of bands
in the sera of patients with systemic autoimmune diseases (Figure 1A). Densitometric analysis indicated that,
compared with the control sera, there was greater reactivity with several antigens in the SLE sera (Figure 1B)
and the WG sera (results not shown). Among these
reactivities was a 60-kd band. To identify the nature of
the bands, a 2DE was performed on EA.hy926 cell
lysates blotted with randomly selected sera from patients
in each disease subgroup. Noteworthy was the restriction
of 1 spot to SLE and WG sera (Figure 1C). The pI of the
spot was 5.7, and its molecular weight was 60-kd. These
stereotypical characteristics fit the Hsp60 protein in the
databases. To more accurately confirm its nature, a
PVDF membrane was stained with Ponceau S solution,
the corresponding spot was cut out, and the protein was
sequenced by the Edman degradation method. Given
that the identified peptide (Figure 1D) matched exactly
with the NH2-terminal end of Hsp60 (Swiss-Prot accession no. P10809), this was confirmed to be the target for
AECAs in the positive SLE and WG sera. Although
disease activity and drug regimen are most likely to be
influential in Hsp60 reactivity, such associations could
not be established, particularly in SLE, WG, and PAN.
Screening of AECA-positive sera. AECA-positive
sera were further screened by Western blotting using
recombinant Hsp60. Sera were considered positive for
Hsp60 when the OD was ⬎0.170 (mean ⫹ 2SD of the
control) (Figure 2A). Overall, more sera from patients
with vasculitic systemic autoimmune diseases were positive for Hsp60 (Figure 2B), except for sera from the
Figure 2. Reactivity of anti–endothelial cell antibody (anti-AECA)–
positive sera with Hsp60. A, Densitometric analysis of the reactivity of
sera from patients with systemic autoimmune diseases, with and
without vasculitis, against Hsp60 recombinant protein by Western
blotting. Broken line shows the cutoff level for positivity (optical
density 0.170, representing the mean ⫹ 2SD of control values).
B, Percentage of sera positive for AECAs. SLE ⫽ systemic lupus
erythematosus; PM ⫽ polymyositis; SS ⫽ Sjögren’s syndrome; RA ⫽
rheumatoid arthritis; SSc ⫽ systemic sclerosis; PAN ⫽ polyarteritis
nodosa; MPA ⫽ microscopic polyangiitis; WG ⫽ Wegener’s granulomatosis; BD ⫽ Behçet’s syndrome; CSS ⫽ Churg-Strauss syndrome.
CSS patients, than sera from patients with nonvasculitic
systemic autoimmune diseases. The percentage of positivity ranged from 44% to 79% and from 10% to 30% in
the respective groups (P ⬍ 0.01). This Hsp60 reactivity
was confirmed in the ELISA, where sera were considered positive when the OD was ⬎0.150. Although the
sensitivity of this ELISA was not as high as that of the
Western blotting (Table 1), again, there were more
positive sera among the vasculitic than among the nonvasculitic systemic autoimmune disease groups (22–44%
versus 10–30%; P ⬍ 0.01).
Membrane expression of Hsp60. Next, experiments were performed to determine how Hsp60 is
expressed on EC membranes and thereby becomes
accessible to autoantibodies. It was impossible to detect
Hsp60 by FACS analysis in the cytosol or on the cell
Figure 3. Hsp60 expression on endothelial cells (ECs). Permeabilized (⫹ saponin) or nonpermeabilized
(– saponin) human umbilical vein ECs were analyzed by fluorescence-activated cell sorting before (A) and
after exposure to heat-induced stress for 4 hours (B) and 5 hours (C) (see Patients and Methods for
details). The expression of Hsp60 was evaluated using IgG anti-Hsp60 monoclonal antibody (mAb), as
revealed by fluorescein isothiocyanate (FITC)–conjugated antibody. Thin lines indicate fluorescence
obtained with a control antibody. The percentage of positive cells as well as the mean fluorescence
intensity (in parentheses) are indicated. ECs were also analyzed by indirect immunofluorescence before
(D) and after exposure to heat-induced stress for 3 hours (E), 4 hours (F), and 5 hours (G). The presence
of Hsp60 was evaluated using IgG anti-Hsp60 mAb, as revealed by FITC-conjugated antibody. Bar ⫽ 30
␮m. Expression of Hsp60 mRNA was then determined after different durations of heat–induced stress
(H). Total mRNA was extracted from ECs, the quantity of Hsp60 mRNA was normalized against that of
18S, and the numbers of copies were compared with those in freshly isolated ECs.
membrane of freshly isolated cells (Figure 3A). This lack
of Hsp60 detection was verified by IIF analysis of
permeabilized cells (Figure 3D). Given that Hsp60 may
be induced on the surface of arterial (31) and venous
(32) ECs by stress, we analyzed this phenomenon. Hsp60
protein remained undetectable after 3 hours of exposure
to heat (Figure 3E). After 4 hours of heat stress, Hsp60
was demonstrated by FACS (Figure 3B) and by IIF
from ECs and subsequently captured on their surface.
Dot-blot analysis revealed soluble Hsp60 only in culture
supernatants of stressed ECs (Figure 4A). This established their capacity to release newly synthesized Hsp60
protein. Once released, Hsp60 bound to a small proportion of ECs when they were fresh, but to most of them
after exposure to stress (Figure 4B). This suggests that in
Figure 4. Release of Hsp60 from endothelial cells (ECs) and trapping
of soluble Hsp60 on the surface of ECs. ECs were stressed by exposure
to heat or were not stressed. A, Culture supernatants were evaluated
for the presence of soluble Hsp60 by dot-blotting. The protein was
revealed using anti-Hsp60 antibody, followed by horseradish peroxidase (HRP)–streptavidin antibody. Culture medium and recombinant
Hsp60 were used as the negative and positive controls, respectively.
B, ECs were incubated with (⫹) or without (–) recombinant Hsp60,
and protein binding was evaluated by fluorescence-activated cell
sorting using anti-Hsp60 monoclonal antibody, followed by fluorescein
isothiocyanate (FITC)–conjugated antibody. Thin lines indicate fluorescence obtained with a control antibody.
(Figure 3F) in the cytosol of most of saponin-treated
ECs. Hsp60 was seen with the highest fluorescence
intensity on the membrane after 5 hours of heat stress
(Figures 3C and G). Quantitative RT-PCR was then
used to quantify Hsp60 transcripts in freshly isolated and
heat-stressed ECs. The number of Hsp60 mRNA copies
was markedly increased after a 3-hour period of heat
stress (Figure 3H). This rise in the number of transcripts
preceded the intracytoplasmic expression of the protein.
We further assessed whether Hsp60 was shed
Figure 5. Induction of endothelial cell (EC) apoptosis by reactivity
with anti-Hsp60. ECs were stressed by exposure to heat. A, ECs were
incubated with various concentrations of anti-Hsp60 monoclonal antibody (mAb). After 24 hours of incubation, apoptosis was evaluated
by fluorescence-activated cell sorting after staining with fluorescein
isothiocyanate–conjugated annexin V and propidium iodide (PI).
B, ECs were incubated with 8 ␮g/ml of anti-Hsp60 mAb or with 320
␮g/ml of IgG purified from anti-Hsp60–containing anti-EC antibody–
positive sera (samples 1–6 from patients with systemic lupus erythematosus), with (open bars) or without (solid bars) recombinant Hsp60.
Apoptotic cells were the annexin V–positive cells among the PInegative EC population. The percentage of apoptosis was determined
by subtracting the number of cells undergoing spontaneous apoptosis.
C, The percentage inhibition of the apoptotic response was calculated
as described in Patients and Methods.
addition to proteins translocated to the membrane from
the cytosol, other extracellular proteins bind to the
membrane of ECs, possibly in an autocrine manner.
Proapoptotic effect of anti-Hsp60 antibodies. We
next examined the issue of whether the complex of
AECAs and Hsp60 delivers antiapoptotic signals. ECs
were heat stressed to induce Hsp60 expression and then
incubated with an anti-Hsp60 mAb for 24 hours. ECs
displayed a dose-dependent apoptotic response (Figure
5A). The number of ECs undergoing apoptosis increased above that obtained spontaneously in the culture medium, reaching a plateau at 8 ␮g/ml of antibody.
These data suggested that the binding of anti-Hsp60 to
Hsp60 does indeed induce apoptosis.
To confirm that apoptosis occurs in disease
states, ECs were incubated with IgG from anti-Hsp60
antibody–containing AECA-positive sera from 6 patients with SLE. Such treatment triggered apoptosis in
ECs, but the number of apoptotic ECs varied from
patient to patient (Figure 5B). Interestingly, when purified IgG was preincubated with saturating concentrations of recombinant Hsp60, as determined in pilot
experiments, the apoptotic effect was inhibited to various degrees in sera from the different patients studied
(Figure 5B). More than 50% inhibition was observed in
sera from patients 1–3, while there was as little as 20%
inhibition in sera from patient 4 and negligible effects in
sera from patients 5 and 6 (Figure 5C). Overall, our
results confirm that in some patients with systemic autoimmune diseases, Hsp60 constitutes a target antigen for
AECAs and that the binding of these autoantibodies can
induce apoptosis in stressed ECs.
In this study, we sought to determine the level of
Hsp60 reactivity of AECA-positive sera from patients
with vasculitic systemic autoimmune diseases, compared
with sera from patients with nonvasculitic systemic autoimmune diseases. Because AECAs are thought to be
involved in the pathogenesis of such conditions, we also
assessed the contribution of anti-Hsp60 antibody to the
proapoptotic effects of AECAs. Furthermore, experiments were performed to gain insight into the mechanisms used by Hsp60 to become accessible to AECAs.
Although the prevalence of AECAs may vary among
rheumatic diseases (33), our findings (Table 1) are
consistent with those of previous studies of the overall
distribution of these autoantibodies.
Irrespective of AECA activities, there is still
controversy regarding the presence of anti-Hsp60 anti-
bodies in SLE (13–16). Thus, a consensus on the prevalence of AECAs and Hsp60 antibodies in SLE and
other systemic autoimmune diseases has not yet been
reached. As usual, these discrepancies may be ascribed
to variations in patient selection and differences in the
tests used (33,34). Each of the 3 different approaches
used in the current investigation, however, verified the
findings in the other approaches and showed that recombinant Hsp60 reactivity exists in vasculitic and in nonvasculitic systemic autoimmune diseases. Using Western
blotting with Hsp60, there were more positive sera in the
former group of patients than the latter group. Consistent with our previous study (28), the prevalence was
lower in the ELISA with Hsp60 analyses than in the
Western blotting analyses, but still it predominated in
patients with vasculitis. These observations reinforce our
2DE preliminary data indicating that Hsp60 recognition
by AECAs was more often associated with vasculitic
than with nonvasculitic systemic autoimmune diseases.
This finding is consistent with the frequency of antiHsp60 reactivity in SLE, the prototypical vasculitic
systemic autoimmune disease, compared with other systemic autoimmune diseases not associated with vasculitis.
There is every likelihood that patients with
higher reactivity for Hsp60 were those who had active
disease at the time their blood was collected, but we
could not establish any influence of disease activity in
patients with SLE, WG, or PAN. We cannot exclude the
possibility that different epitopes of Hsp60 would be
targeted, depending on the disease severity, as has been
elegantly demonstrated in atherosclerosis (35). However, serial sampling is warranted in order to address this
specific question. While the patients’ drug regimens
could also have an influence, Hsp60 reactivity did not
appear to be affected by corticosteroids or other immunosuppressants. This remains to be confirmed, however.
Due to its location within the cytoplasm, Hsp60 is
not accessible to AECAs unless the cells are specifically
activated. This contention was confirmed by FACS and
IIF analyses showing that Hsp60 was not present on the
membrane of freshly isolated nonpermeabilized ECs or
in the cytosol of freshly isolated permeabilized ECs.
However, when ECs were exposed to heat stress, they
expressed Hsp60. Initially, the level of Hsp60 mRNA
was markedly raised after 3 hours of heat-induced stress,
which was followed 4 hours later by cytosolic detection
of the protein and another 1 hour later by the translocation of the Hsp60 protein to the cell membrane.
Inasmuch as this protein lacks the peptide signal sequence required for cell surface translocation, this transfer must use unconventional (and thus far unknown)
pathways. Our observations are also consistent with
those of previous experiments showing that ECs express
Hsp60 on their surface under stress, both in vitro (8,36)
and in vivo (32). Our findings support the notion that
translocation of Hsp60 to the cell surface plays a role in
the pathogenesis of vasculitis-associated systemic autoimmune diseases (8).
Once expressed, Hsp60 can also be shed from
stressed ECs. The increased level of soluble Hsp60 in the
sera of patients with early cardiovascular disease (37)
might thus result from its accelerated release from ECs.
Also supporting our interpretation is the recent observation that levels of soluble Hsp60 correlate with markers of inflammation in atherosclerosis (36). It is highly
relevant that SLE patients have been shown to be at risk
of developing atherosclerosis (37), such that there has
been a recent increase in their proportionate death from
vascular disease, particularly from accelerated atherosclerosis (38). Interestingly, human and bacterial heatshock proteins have a 55% homology, and so, the
immune response to a bacterial challenge may result in
cross-reactivity by targeting Hsp60-expressing host ECs.
This is particularly true in Mycobacterium infection (20).
Moreover, we found that soluble Hsp60 could be
taken up on the surface of ECs. The capacity of ECs to
bind soluble Hsp60 was enhanced by stress. This newly
identified property may affect pathophysiologic responses. In this setting, 3 important questions arise.
First, what are the pathogenic consequences of the
presence of Hsp60 on the surface of ECs? Second, which
receptor(s) is triggered by soluble Hsp60 that would
initiate its pathogenic effects? Third, do “autochthonous
Hsp60” and passively bound Hsp60 differ in terms of
their immunogenicity, and do these 2 kinds of Hsp60
trigger different signal transduction cascades?
To gain further insight into the pathogenicity of
AECAs directed toward Hsp60, the apoptotic fate of
ECs was evaluated (24,25). All anti-Hsp60–positive sera
tested induced apoptosis. Those experiments suggest
that binding of anti-Hsp60 to its antigen on the EC
surface is sufficient in itself to initiate programmed cell
death. However, inhibition experiments in which purified IgG was preincubated with recombinant Hsp60 only
partly prevented apoptosis, demonstrating that there is
more than 1 candidate autoantibody among apoptosisinducing AECAs. This finding is at variance with that by
Dieudé et al (13), who reported that AECAs from SLE
sera specifically and exclusively recognized Hsp60. In
our experience, this is really exceptional.
The apoptotic pathways activated by the binding
of AECAs to Hsp60 are unknown. Several mechanisms
may be involved in the process. When complexed with
Bax in the cytosol, Hsp60 participates in an antiapoptotic complex. Stress factors dissociate this complex by
translocating Hsp60 to the plasma membrane and directing Bax to the mitochondria (39). It follows that once
apoptosis has been triggered, mitochondrial Hsp60 is
released and that this, in turn, accelerates the activation
of procaspase 3 (40). A role of surface expression of
Hsp60 was also suspected based on the fact that levels of
Hsp60 on the surface of apoptotic cells are increased
(41). All of these observations converge to suggest that
the abnormal expression of Hsp60 on the plasma membrane sensitizes ECs to apoptosis. Furthermore, our
results support the view that direct recognition of Hsp60
on stressed ECs further promotes the process. A positive
feedback regulation may thus be operant, in that stress
factors favor the appearance of Hsp60 on the membrane, and Hsp60-associated AECAs trigger apoptosis,
which results in the release of mitochondrial matrix
Hsp60 into the cytosol (42). Owing to the stress, Hsp60
then moves to the membrane, where it becomes accessible to AECAs.
Despite the results generated in this study, it is
not yet possible to categorically assign Hsp60-induced
apoptosis to antibodies specific for the translocated form
of the protein or to antibodies specific for the soluble
form passively bound to ECs. There are a number of
reports suggesting that certain anti-Hsp60 antibodies
display cross-reactivity with CD51, CD61, and connexin
45 on EC membranes (43). One might thus hypothesize
that AECAs bridge plasma membrane Hsp60 with any
of these molecules on ECs exposed to stress. Such
complexes could be required to trigger the apoptotic
signaling cascade. It is intriguing in this respect that
Toll-like receptor 4 (TLR-4) and TLR-2 mRNA were
shown to be induced by heat stress (Jamin C, et al:
unpublished observations, and Wick G: personal communication). TLR-4 is involved in the signaling of
soluble Hsp60 (44), and TLR-2 is involved in the activation of the apoptotic pathway (45). After exposure to
stress, TLR-4 is likely to be up-regulated, leading to its
passive binding to soluble Hsp60. This would promote
the expression of TLR-2 (46). The Hsp70 ligand for
TLR-2 (47) would then stimulate TLR-2 and enhance
apoptosis (48). Alternatively, a direct recognition of
soluble Hsp60 by AECAs, when bound to EC membranes, might activate the multimolecular complex and
trigger apoptosis.
Taken together, the results of our experiments
establish that the complex mixture of AECAs contain
anti-Hsp60 antibodies. The higher frequency of anti-
Hsp60 antibodies in vasculitic than in nonvasculitic
systemic autoimmune diseases indicates that they play a
deleterious role in the pathophysiology of vasculitis. The
capacity of AECA-positive sera from SLE patients to
induceapoptosisofECsstrengthensthisproposedhypothesis. Overall, Hsp60 seems to be a key antigen involved in
vessel damage during inflammatory responses in the
presence of AECAs.
We thank Dr. Cora-Jean S. Edgell (University of
North Carolina, Chapel Hill, NC) for providing the EA.hy926
cells, Professors Rizgar Mageed (St. Bartholomew’s and the
London Queen Mary School of Medicine and Dentistry,
London, UK) and Georg Wick (Innsbruck Medical University,
Innsbruck, Austria) for insightful review of the manuscript,
and Simone Forest and Cindy Séné for secretarial assistance.
1. Zugel U, Kaufmann SH. Role of heat shock proteins in protection
from and pathogenesis of infectious diseases. Clin Microbiol Rev
2. Soltys BJ, Gupta RS. Immunoelectron microscope localization of
the 60-kDa heat chaperonin protein in mammalian cells. Exp Cell
Res 1996;222:16–27.
3. Freedman MS, Buu NN, Ruijs TC, Williams K, Antel JP. Differential expression of heat shock proteins by human glial cells.
J Neuroimmunol 1992;41:231–8.
4. Wand-Wurttenberger A, Schoel B, Ivanyi J, Kaufmann SH. Surface expression by mononuclear phagocytes of an epitope shared
with mycobacterial heat shock protein 60. Eur J Immunol 1991;
5. Kaur I, Voss SD, Gupta RS, Schell K, Fisch P, Sondel PM. Human
peripheral ␥␦ T cells recognize hsp60 molecules on Daudi Burkitt’s lymphoma cells. J Immunol 1993;150:2046–55.
6. Vendetti S, Cicconi R, Piselli P, Vismara D, Cassol M, Delpino A.
Induction and membrane expression of heat shock proteins in
heat-treated HPC-4 cells is correlated with increased resistance to
LAK-mediated lysis. J Exp Clin Cancer Res 2000;19:329–34.
7. Dziewanowska K, Carson AR, Patti JM, Deobald CF, Bayles KW,
Bohach GA. Staphylococcal fibronectin binding protein interacts
with heat shock protein 60 and integrins: role in internalization by
epithelial cells. Infect Immun 2000;68:6321–8.
8. Wick G, Knoflach M, Xu Q. Autoimmune and inflammatory
mechanisms in atherosclerosis. Annu Rev Immunol 2004;22:
9. Seitz CS, Kleindienst R, Xu Q, Wick G. Coexpression of heatshock protein 60 and intercellular-adhesion molecule-1 is related
to increased adhesion of monocytes and T cells to aortic endothelium of rats in response to endotoxin. Lab Invest 1996;74:241–52.
10. Soltys BJ, Gupta RS. Cell surface localization of the 60-kDa heat
shock chaperonin protein on mammalian cells. Cell Biol Int
11. Young D, Lathigra R, Hendrix R, Sweetser D, Young RA. Stress
proteins are immune targets in leprosy and tuberculosis. Proc Natl
Acad Sci U S A 1988;85:4267–70.
12. Rudolphi U, Rzepka R, Batsford S, Kaufmann SH, von der Mark
K, Peter HH, et al. The B cell repertoire of patients with
rheumatoid arthritis. II. Increased frequencies of IgG⫹ and IgA⫹
B cells specific for mycobacterial heat-shock protein 60 or human
type II collagen in synovial fluid and tissue. Arthritis Rheum
Dieude M, Senecal JL, Raymond Y. Induction of endothelial cell
apoptosis by heat-shock protein 60–reactive antibodies from
anti–endothelial cell autoantibody–positive systemic lupus erythematosus patients. Arthritis Rheum 2004;50:3221–31.
De Graeff-Meeder ER, Rijkers GT, Voorhorst-Ogink MM, Kuis
W, van der Zee R, van Eden W, et al. Antibodies to human HSP60
in patients with juvenile chronic arthritis, diabetes mellitus, and
cystic fibrosis. Pediatr Res 1993;34:424–8.
Schett G, Xu Q, Amberger A, van der Zee R, Recheis H, Willeit
J, et al. Autoantibodies against heat shock protein 60 mediate
endothelial cytotoxicity. J Clin Invest 1995;96:2569–77.
Jarjour WN, Jeffries BD, Davis JS IV, Welch WJ, Mimura T,
Winfield JB. Autoantibodies to human stress proteins: a survey of
various rheumatic and other inflammatory diseases. Arthritis
Rheum 1991;34:1133–8.
D’Cruz DP, Houssiau FA, Ramirez G, Berguley E, McCutcheon J,
Vianna J, et al. Antibodies to endothelial cells in systemic lupus
erythematosus: a potential marker for nephritis and vasculitis. Clin
Exp Immunol 1991;85:254–61.
Salojin KV, le Tonqueze M, Nassonov EL, Blouch MT, Baranov
AA, Saraux A, et al. Anti-endothelial cell antibodies in patients
with various forms of vasculitis. Clin Exp Rheumatol 1996;14:
Salojin KV, le Tonqueze M, Saraux A, Dueymes M, Nassonov EL,
Piette JC, et al. Anti-endothelial cell antibodies and markers of
endothelial cell damage: clinical associations in systemic sclerosis
and Raynaud’s phenomenon. Am J Med 1997;102:178–85.
Dugue C, Perraut R, Youinou P, Renaudineau Y. Effects of
anti-endothelial cell antibodies in leprosy and malaria. Infect
Immun 2004;72:301–9.
Cacoub P, Ghillani P, Revelen R, Thibault V, Calvez V, Charlotte
F, et al. Anti-endothelial cell autoantibodies in hepatitis C virus
mixed cryoglobulinemia. J Hepatol 1999;31:598–603.
Toyoda M, Galfayan K, Galera OA, Petrosian A, Czer LS, Jordan
SC. Cytomegalovirus infection induces anti-endothelial cell antibodies in cardiac and renal allograft recipients. Transpl Immunol
Frostegard J, Wu R, Gillis-Haegerstrand C, Lemne C, de Faire U.
Antibodies to endothelial cells in borderline hypertension. Circulation 1998;98:1092–8.
Bordron A, Dueymes M, Levy Y, Jamin C, Leroy JP, Piette JC, et
al. The binding of some human antiendothelial cell antibodies
induces endothelial cell apoptosis. J Clin Invest 1998;101:2029–35.
Worda M, Sgonc R, Dietrich H, Niederegger H, Sundick RS,
Gershwin ME, et al. In vivo analysis of the apoptosis-inducing
effect of anti–endothelial cell antibodies in systemic sclerosis by
the chorionallantoic membrane assay. Arthritis Rheum 2003;48:
Turesson C, McClelland RL, Christianson TJ, Matteson EL. No
decrease over time in the incidence of vasculitis or other extraarticular manifestations in rheumatoid arthritis: results from a
community-based study. Arthritis Rheum 2004;50:3729–31.
Youinou P, Pennec YL, Katsikis P, Jouquan J, Fauquert P, le Goff
P. Raynaud’s phenomenon in primary Sjögren’s syndrome. Br J
Rheumatol 1990;29:205–7.
Revelen R, d’Arbonneau F, Guillevin L, Bordron A, Youinou P,
Dueymes M. Comparison of cell-ELISA, flow cytometry and
Western blotting for the detection of antiendothelial cell antibodies. Clin Exp Rheumatol 2002;20:19–26.
Revelen R, Bordron A, Dueymes M, Youinou P, Arvieux J. False
positivity in a cyto-ELISA for anti-endothelial cell antibodies
caused by heterophile antibodies to bovine serum proteins. Clin
Chem 2000;46:273–8.
Sgonc R, Gruschwitz MS, Boeck G, Sepp N, Gruber J, Wick G.
Endothelial cell apoptosis in systemic sclerosis induced by anti-
body-dependent cell-mediated cytotoxicity via CD95. Arthritis
Rheum 2000;43:2550–62.
Amberger A, Maczek C, Jurgens G, Michaelis D, Schett G, Trieb
K, et al. Co-expression of ICAM-1, VCAM-1, ELAM-1 and Hsp60
in human arterial and venous endothelial cells in response to
cytokines and oxidized low-density proteins. Cell Stress Chaperones 1997;2:94–103.
Hochleitner BW, Hochleitner EO, Obrist P, Eberl T, Amberger A,
Xu Q, et al. Fluid shear stress induces heat shock protein 60
expression in endothelial cells in vitro and in vivo. Arterioscler
Thromb Vasc Biol 2000;20:617–23.
Meroni PL, Youinou P. Endothelial cell antibodies. In: Peter JB,
Shoenfeld Y, editors. Autoantibodies. Amsterdam: Elsevier; 1996.
p. 245–52.
Youinou P, Meroni PL, Khamashta MA, Shoenfeld Y. A need for
standardization of the antiendothelial cell antibody test. Immunol
Today 1995;91:363–4.
Perschinka H, Mayr M, Millonig G, Mayerl C, van der Zee R,
Morrison SG, et al. Cross-reactive B-cell epitopes of microbial and
human heat-shock protein 60/65 in atherosclerosis. Arterioscler
Thromb Vasc Biol 2003;23:1060–5.
Xu Q, Schett G, Perschinka H, Mayr M, Egger G, Oberhollenzer F,
et al. Serum soluble heat shock protein 60 is elevated in subjects with
atherosclerosis in a general population. Circulation 2000;102:14–20.
Pockley AG, Wu R, Lemne C, Kiessling R, de Faire U, Frostegard
J. Circulating heat shock protein 60 is associated with early
cardiovascular disease. Hypertension 2000;36:303–7.
Borchers AT, Keen CL, Shoenfeld Y, Gershwin ME. Surviving the
butterfly and the wolf: mortality trends in systemic lupus erythematosus. Autoimmun Rev 2004;3:423–53.
Gupta S, Knowlton AA. Cytosolic heat shock protein 60, hypoxia,
and apoptosis. Circulation 2002;106:2727–33.
Samali A, Cai J, Zhivotovsky B, Jones DP, Orrenius S. Presence of
a pre-apoptotic complex of pro-caspase-3, Hsp60 and Hsp10 in the
mitochondrial fraction of Jurkat cells. EMBO J 1999;18:2040–8.
Sapozhnikov AM, Ponomarev ED, Tarasenko TN, Telford WG.
Spontaneous apoptosis and expression of cell surface heat-shock
proteins in cultured EL-4 lymphoma cells. Cell Prolif 1999;32:
Chandra D, Tang DG. Mitochondrially localized active caspase-9
and caspase-3 result mostly from translocation from the cytosol
and partly from caspase-mediated activation in the organelle: lack
of evidence for Apaf-1-mediated procaspase-9 activation in the
mitochondria. J Biol Chem 2003;278:17408–20.
Bason C, Corrocher R, Lunardi C, Puccetti P, Olivieri O, Girelli
D, et al. Interaction of antibodies against cytomegalovirus with
heat-shock protein 60 in pathogenesis of atherosclerosis. Lancet
Bulut Y, Faure E, Thomas L, Karahashi H, Michelsen KS, Equils
O, et al. Chlamydial heat shock protein 60 activates macrophages
and endothelial cells through Toll-like receptor 4 and MD2 in a
MyD88-dependent pathway. J Immunol 2002;168:1435–40.
Aliprantis AO, Yang RB, Weiss DS, Godowski P, Zychlinsky A.
The apoptotic signaling pathway activated by Toll-like receptor-2.
EMBO J 2000;19:3325–36.
Fan J, Frey RS, Malik AB. TLR4 signaling induces TLR2 expression in endothelial cells via neutrophil NADPH oxidase. J Clin
Invest 2003;112:1234–43.
Wagner M, Hermanns I, Bittinger F, Kirkpatrick CJ. Induction of
stress proteins in human endothelial cells by heavy metal ions and
heat shock. Am J Physiol 1999;277:1026–33.
Asea A, Rehli M, Kabingu E, Boch JA, Bare O, Auron PE, et al.
Novel signal transduction pathway utilized by extracellular HSP70:
role of toll-like receptor (TLR) 2 and TLR4. J Biol Chem
Без категории
Размер файла
293 Кб
induction, associates, antibodies, apoptosis, systemic, autoimmune, disease, cells, endothelial, antiendothelial, hsp60, vasculitis, binding
Пожаловаться на содержимое документа