Thesis

APOPTOSIS AND AUTOANTIBODIES IN
SYSTEMIC LUPUS ERYTHEMATOSUS
Marc Bijl
The studies in this thesis were supported by grant NSN C95.1482 from the Dutch Kidney
foundation.
The publication of this thesis was financially supported by:

Boehringer Ingelheim B.V.
Groningen Utrecht Institute for Drug Exploration GUIDE
Merck Sharp & Dohme B.V.
Novartis Pharma B.V.
Pharmacia B.V.
Roche Nederland B.V.
Stichting Ontwikkeling Klinische Immunologie Groningen
Wyeth-Lederle B.V.
Cover illustration: apoptotic Jurkatt cells binding to and internalised by cultured human
macrophages.
Apoptosis and autoantibodies in Systemic Lupus Erythematosus

Marc Bijl
Thesis University Groningen
ISBN 90-367-1431-1
© M. Bijl
All rights reserved
Printed by Ridderprint B.V., Ridderkerk, The Netherlands

RIJKSUNIVERSITEIT GRONINGEN
APOPTOSIS AND AUTOANTIBODIES IN

SYSTEMIC LUPUS ERYTHEMATOSUS
PROEFSCHRIFT
ter verkrijging van het doctoraat in de

Medische Wetenschappen
aan de Rijksuniversiteit Groningen
op gezag van de
Rector Magnificus, dr. D.F.J. Bosscher,
in het openbaar te verdedigen op
woensdag 30 mei 2001
om 16.00 uur
door
Marc Bijl
geboren op 13 december 1961
te Arnhem
Promotor: Prof. dr. C.G.M. Kallenberg
Co-promotor: Dr. P.C. Limburg
ISBN: 90-367-1431-1
Beoordelingscommissie: Prof. dr. J.H.M. Berden
Prof. dr. J.W.J. Bijlsma
Prof. dr. M.H. van Rijswijk
Paranimfen: Gerda Horst

Siebren Schaafsma
Contents
page
Chapter I: General introduction and aim of the thesis 1
Chapter II: Patients with systemic lupus erythematosus with high plasma
levels of sFas risk relapse
J Rheumatol 1999;26:60-7 9
Chapter III: Do elevated levels of serum soluble Fas contribute to the

persistence of activated lymphocytes in systemic lupus
erythematosus?
J Autoimmunity 1998;11:457-63 25
Chapter IV: Fas expression on peripheral blood lymphocytes of systemic

lupus erythematosus (SLE) patients: relation to lymphocyte
activation and disease activity
Submitted for publication 39
Chapter V: Effects of smoking on activation markers, Fas expression

and apoptosis of peripheral blood lymphocytes
Eur J Clin Invest, accepted for publication 51
Chapter VI: Anti-CD3-induced and anti-Fas-induced apoptosis in systemic

lupus erythematosus (SLE)
Clin Exp Immunol 2001;123:127-32 61
Chapter VII: Expression of costimulatory molecules on peripheral blood

lymphocytes of patients with systemic lupus erythematosus
Ann Rheum Dis, accepted for publication 75
Chapter VIII: IgG subclass distribution of autoantibodies differs between
renal and extra-renal relapses in patients with systemic lupus
erythematosus
Chapter IX: Fc receptor polymorphisms in systemic lupus erythematosus:

association with disease and in vivo clearance of immune
complexes
Arthritis Rheum 2000;43:2793-800 101
Chapter X: New insights into the pathogenesis of systemic lupus erythematosus

(SLE): the role of apoptosis (review)
Chapter XI: Summary and conclusions 133
Chapter XII: Nederlandse samenvatting 141
Dankwoord 153
Curriculum Vitae 156

Chapter I
Introduction
Chapter I
Systemic lupus erythematosus
Systemic lupus erythematosus (SLE) is the prototype of a systemic autoimmune disease

and is characterised by the presence of multiple autoantibodies. Especially antibodies to
dsDNA are specific for the disease, and changes in their levels are a sensitive tool for the
detection of disease activity [1-3]. A rise in the level of these antibodies precedes the
development of a clinical relapse in many cases [1] suggesting a particular role of these
antibodies in the pathogenesis of the disease. This concept is supported by the presence of
antibodies to dsDNA in affected kidneys [4,5]. Furthermore, antibodies to dsDNA in SLE
patients, in contrast to those found in healthy controls, are high-affinity antibodies of the
IgG subclass carrying T cell-dependent somatic hypermutations [6-8]. The rise of
antibody levels preceding a clinical relapse is associated with an increase in the proportion
of B cells and T-helper cells expressing activation markers [9]. Taken together, these data
are consistent with the concept that the anti-dsDNA response in SLE is an antigen driven,
T cell-dependent process [10].
Although the presence of antibodies to nuclear (and cytoplasmic) antigens and their levels
have been shown to be of value for the clinician in diagnosing and monitoring SLE
patients, it is poorly understood why these antibodies develop. However, in recent years
some intriguing new concepts concerning the development of autoantibodies and the
etiopathogenesis of SLE have emerged.
Apoptosis
Every day billions of cells are eliminated by a very tightly orchestrated process called
programmed cell death or apoptosis. In contrast to cell death by necrosis, apoptosis is
characterised by a stereotypical pattern of morphological changes, irrespective of the
initiating event [11]. These changes include cellular and nuclear condensation and
fragmentation, flip flop, (that is, negatively charged phosphatidylserine moves from the
inner to the outer surface) and blebbing of the cell membrane [12]. In these blebs
autoantigens are exposed. It has been shown that in apoptotic keratinocytes many nuclear
and cytoplasmic antigens, that are characteristic targets for autoantibodies, are clustered in
high concentrations in the cell membrane [13]. The presentation of nuclear antigens at the
outer surface of the cell might give a clue to the question why in autoimmunity antibodies
are produced that are directed towards intracellular antigens. It does, however, not answer
the question why antibodies arise anyhow as apoptotic cells are phagocytosed very rapidly
[14].
2
Introduction
Phagocytosis
Phagocytosis of apoptotic cells is a complex process in which many cells, membrane

receptors and serum molecules are involved [15]. The involvement of receptors like CD14
on professional phagocytes insures a non-inflammatory process because apoptosis via this
receptor leads to the production and release of TGFß, a potent anti-inflammatory cytokine
[16]. The phagocytosis of apoptotic cells is a self regulating process [17] as previous
uptake of apoptotic cells downregulates the phagocytic capacity of macrophages. It can be
speculated that the increased presence of apoptotic cells thereby results in a declining
phagocytic capacity of macrophages, and so introduces a vicious circle leading to the
persistence of non-ingested apoptotic cells presenting intracellular antigens on their
surface.
Inflammation
The presence of apoptotic cells in sufficient amounts can break tolerance characterised by
the production of autoantibodies [18]. As this production is probably T cell dependent a
proper T and B cell interaction is necessary. A proper interaction means that, next to
binding of the T cell receptor (TCR) with the major histocompability complex (MHC)
molecule complexed with the antigen, additional signalling through costimulatory
molecules occurs. The autoantibodies produced play a central role in the initiation of
inflammation, as has been shown in the kidney in particular [19,20]. Autoantibodies can
bind to the antigenic structures that are presented on the surface of apoptotic cells during
the apoptotic process [21,22]. In stead of the normal opsonization with serum components
like CRP, SAP and complement proteins [23,24] the apoptotic cell now will be opsonised
with autoantibodies. The constant domain of these antibodies will allow Fcγ-receptor
bearing cells to interact. Interaction with Fcγ-receptors will lead to secretion of pro-
inflammatory cytokines, resulting in attraction of neutrophils and monocytes, eventually
leading to inflammation that is characteristic for autoimmune diseases like SLE. The
important role of Fcγ-receptors is strongly supported by studies performed in γ-chain
knock out mice in which it was shown that, despite the presence of autoantibodies in the
kidneys, no inflammation occurred [25].
Aim of the thesis
The general objective of the studies in this thesis is to elucidate the role apoptotic cells
play in the etiopathogenesis of SLE. First, we focus on the function of soluble Fas (sFas),
a molecule which has been described to interfere with the induction of Fas-mediated
apoptosis. It has been suggested that elevated levels of sFas hamper apoptosis of
3
Chapter I
(autoreactive) lymphocytes, subsequently resulting in autoimmunity. In chapter II we

analyse the fluctuations in sFas levels in the serum of SLE patients in time, in relation to
disease activity. In addition, to evaluate whether increased sFas levels are associated with
lymphocyte activation, we cross-sectionally analysed sFas levels in relation to the
activation status of the respective peripheral blood lymphocyte populations. Results are
presented in chapter III. The interference of sFas with the induction of Fas-mediated
apoptosis is only one possible mechanism accounting for the presence of apoptotic cells in
the peripheral blood of lupus patients. The susceptibility of peripheral blood lymphocytes
(PBL) in SLE patients for Fas-mediated apoptosis is further dependent on the expression
of membrane Fas. In chapter IV we describe the results of Fas expression on PBL of SLE
patients. As a spin off, in chapter V we evaluate the effects of cigarette smoking on Fas
expression in healthy subjects. Finally, the question whether there are intrinsic defects in
the Fas-mediated apoptosis in SLE patients is subject of investigation in chapter VI.
The development of an inflammatory process is one of the hallmarks of SLE. The T cell-
dependent production of autoantibodies plays a central role in the initiation of
inflammation. We measured the expression of costimulatory molecules on PBL in SLE
patients in conjunction with disease activity to evaluate whether expression of these
molecules in SLE is increased, a factor which might facilitate the production of
autoantibodies. In chapter VII the results of this analysis are described. Autoantibodies
are primarily of the IgG isotype. Of the different IgG subclasses, IgG1 and IgG3 activate
complement more efficiently than IgG2 while IgG4 does not activate complement at all.
The IgG subclass distribution of autoantibodies therefore can be of relevance in the
development of inflammation, in lupus nephritis in particular. In chapter VIII we studied
IgG subclass profiles of autoantibodies to nucleohistone and to dsDNA in patients prior to
a renal relapse in comparison to patients developing an extra-renal relapse. Finally, we
analysed whether the Fcγ receptor polymorphism, by their differential interaction with the
respective IgG autoantibodies, has influence on the course of the disease. Results are
presented in chapter IX. This study was conducted because it was shown that the
different Fcγ-receptor isotypes described have different affinity for the several IgG
subclasses. An overview of the current ideas about the pathogenesis of SLE is given in
chapter X. The results of this thesis are summarised in chapter XI.
References
1. ter Borg EJ, Horst G, Hummel EJ, Limburg PC, Kallenberg CG. Measurement of increases
in anti-double-stranded DNA antibody levels as a predictor of disease exacerbation in
systemic lupus erythematosus. A long-term, prospective study. Arthritis Rheum
1990;33:634-43.
2. ter Borg EJ, Horst G, Hummel E, Limburg PC, Kallenberg CG. Rises in anti-double
stranded DNA antibody levels prior to exacerbations of systemic lupus erythematosus are
4
Introduction
not merely due to polyclonal B cell activation. Clin Immunol Immunopathol 1991;59:117-
28.
3. Bootsma H, Spronk P, Derksen R et al. Prevention of relapses in systemic lupus
erythematosus. Lancet 1995;345:1595-9.
4. Amoura Z, Chabre H, Koutouzov S et al. Nucleosome-restricted antibodies are detected
before anti-dsDNA and/or antihistone antibodies in serum of MRL-Mp lpr/lpr and +/+ mice,
and are present in kidney eluates of lupus mice with proteinuria. Arthritis Rheum
1994;37:1684-8.
5. van Bruggen MC, Kramers C, Hylkema MN, Smeenk RJ, Berden JH. Significance of anti-
nuclear and anti-extracellular matrix autoantibodies for albuminuria in murine lupus
nephritis; a longitudinal study on plasma and glomerular eluates in MRL/l mice. Clin Exp
Immunol 1996;105:132-9.
6. Davidson A, Manheimer-Lory A, Aranow C et al. Molecular characterization of a
somatically mutated anti-DNA antibody bearing two systemic lupus erythematosus-related
idiotypes. J Clin Invest 1990;85:1401-9.
7. Winkler TH, Fehr H, Kalden JR. Analysis of immunoglobulin variable region genes from
human IgG anti-DNA hybridomas. Eur J Immunol 1992;22:1719-28.
8. van Es JH, Gmelig Meyling FH, van de Akker WR et al. Somatic mutations in the variable
regions of a human IgG anti-double-stranded DNA autoantibody suggest a role for antigen
in the induction of systemic lupus erythematosus. J Exp Med 1991;173:461-70.
9. Spronk PE, Horst G, Van Der Gun BT, Limburg PC, Kallenberg CG. Anti-dsDNA
production coincides with concurrent B and T cell activation during development of active
disease in systemic lupus erythematosus (SLE). Clin Exp Immunol 1996;104:446-53.
10. Herrmann M, Voll RE, Kalden JR. Etiopathogenesis of systemic lupus erythematosus.
Immunol Today 2000;21:424-6.
11. Kerr JF, Wyllie AH, Currie AR. Apoptosis: a basic biological phenomenon with wide-
ranging implications in tissue kinetics. Br J Cancer 1972;26:239-57.
12. Wyllie AH, Kerr JF, Currie AR. Cell death: the significance of apoptosis. Int Rev Cytol
1980;68:251-306:251-306.
13. Casciola-Rosen LA, Anhalt G, Rosen A. Autoantigens targeted in systemic lupus
erythematosus are clustered in two populations of surface structures on apoptotic
keratinocytes. J Exp Med 1994;179:1317-30.
14. Raff M. Cell suicide for beginners. Nature 1998;396:119-22.
15. Aderem A, Underhill DM. Mechanisms of phagocytosis in macrophages. Annu Rev
Immunol 1999;17:593-623:593-623.
16. Fadok VA, Bratton DL, Konowal A et al. Macrophages that have ingested apoptotic cells in
vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms
involving TGF-beta, PGE2, and PAF. J Clin Invest 1998;101:890-8.
17. Erwig LP, Gordon S, Walsh GM, Rees AJ. Previous uptake of apoptotic neutrophils or
ligation of integrin receptors downmodulates the ability of macrophages to ingest apoptotic
neutrophils. Blood 1999;93:1406-12.
5
Chapter I
18. Mevorach D, Zhou JL, Song X, Elkon KB. Systemic exposure to irradiated apoptotic cells
induces autoantibody production. J Exp Med 1998;188:387-92.
19. Spronk PE, Limburg PC, Kallenberg CG. Serological markers of disease activity in
systemic lupus erythematosus. Lupus 1995;4:86-94.
20. Smeenk RJ. Antinuclear antibodies: cause of disease or caused by disease? [editorial].
Rheumatology (Oxford) 2000;39:581-4.
21. Manfredi AA, Rovere P, Heltai S et al. Apoptotic cell clearance in systemic lupus
erythematosus. II. Role of beta2-glycoprotein I. Arthritis Rheum 1998;41:215-23.
22. Manfredi AA, Rovere P, Galati G et al. Apoptotic cell clearance in systemic lupus
erythematosus. I. Opsonization by antiphospholipid antibodies. Arthritis Rheum
1998;41:205-14.
23. Korb LC, Ahearn JM. C1q binds directly and specifically to surface blebs of apoptotic
human keratinocytes: complement deficiency and systemic lupus erythematosus revisited. J
Immunol 1997;158:4525-8.
24. Gershov D, Kim S, Brot N, Elkon KB. C-Reactive Protein Binds to Apoptotic Cells,
Protects the Cells from Assembly of the Terminal Complement Components, and Sustains
an Antiinflammatory Innate Immune Response. Implications for systemic autoimmunity. J
Exp Med 2000;192:1353-64.
25. Clynes R, Dumitru C, Ravetch JV. Uncoupling of immune complex formation and
kidney damage in autoimmune glomerulonephritis. Science 1998;279:1052-4.
6
Chapter II
Patients with Systemic Lupus Erythematosus with

High Plasma Levels of sFas Risk Relapse
Thea van Lopik1
Marc Bijl2
Margreet Hart1
Leonie Boeije1
Tom Gesner3
Abla A. Creasy3
Cees G. M. Kallenberg2
Lucien A. Aarden1
Ruud J. T. Smeenk1
1
CLB, Sanguin Blood Supply Foundation, Laboratory of Experimental and Clinical
Immunology, Academical Medical Center, University of Amsterdam, Amsterdam, The
Netherlands
2
Department of Clinical Immunology, University Hospital, Groningen, The Netherlands
3
Chiron, Emeryville, CA, USA.
J Rheumatol 1999;26:60-7
Chapter II
Summary
We related soluble Fas (sFas) levels to the Systemic Lupus Erythematosus Disease
Activity Index (SLEDAI) in longitudinal series of plasma samples of SLE patients to
evaluate the relation between excessive production of sFas and disease activity. Therefore,
21 monoclonal antibodies against Fas were raised. Two of these were used to develop and
validate a sensitive sandwich ELISA for the longitudinal analysis of sFas levels in plasma
of 30 patients and 25 controls. At the start of follow-up, a significant elevation (p<0.0001)
was found in sFas levels in SLE (1167±347 pg/ml sFas) compared to controls (618±98
pg/ml sFas). Also, at the start of the follow-up a significant difference (p=0.0028) existed
between patients who were going to have a relapse (1236±402 pg/ml sFas) during follow-
up and patients who were not (809±276 pg/ml sFas). While sFas did not fluctuate with
disease activity in individual patients, we found a strong correlation (r=0.75, p<0.0001)
between sFas and SLEDAI, but only at the time of relapse, when we analysed the patients
as a group. In individual SLE patients, sFas does not fluctuate with disease activity.
However, patients with high plasma levels of sFas are at risk of relapse.
Introduction
A role for Fas in the pathogenesis of systemic lupus erythematosus (SLE) is cogently
demonstrated in murine models of autoimmunity, the MRL-lpr/lpr -mouse and the MRL-
gld/gld-mouse. The phenotypic characteristics of these single gene mutations in the Fas
gene and the Fas ligand gene, respectively, together with an appropriate genetic
background, mimic many of the features of human SLE [1-3].
The Fas protein denotes the CD95 receptor and is a 45-48 kDa type 1 membrane protein
with sequence homology to the TNF-receptor family [3-5]. Membrane Fas (mFas) is
expressed on many different tumour cells [6,7] and on various normal human tissues [8,9].
Triggering of mFas by its ligand results in rapid induction of programmed cell death,
apoptosis, in susceptible cells [8,10]. The tissue distribution of Fas and Fas ligand
suggests that the Fas receptor / ligand system plays an important role in various aspects of
mammalian development, especially in the homeostasis of the immune system [11-18].
The susceptibility of cells to apoptosis-induced cell death is regulated through their state
of activation, activity of down-stream ICE-like proteases (caspases) and activity of
members of the bcl-2/ bax family [19-25].
Following in vitro activation of T-cells isolated from peripheral blood mononuclear cells
(PBMC) of SLE patients, no defects in up regulation of the receptor or in the ability of the
activated cells to undergo apoptosis are detected, suggesting that in the majority of SLE
patients it is unlikely that SLE is caused by defects in the Fas receptor [26]. Two studies
reveal human Fas messenger RNA variants capable of encoding sFas molecules lacking
either the transmembrane domain or the extracytoplasmic region [4,27]. Patients with SLE
display elevated levels of sFas [4,28-31], but not all studies can confirm this elevation
10
Soluble Fas in SLE
[32,33]. Soluble isoforms of cell surface receptors, generated either by proteolytic

cleavage of the transmembrane form or by alternative splicing, regulate receptor function
in a number of biological systems. Such soluble receptors most commonly retain ligand-
binding activity and compete for ligand [34-36]. A soluble form of the Fas receptor might
prevent binding of FasL to Fas thereby blocking Fas-mediated apoptosis. When Fas
mediated apoptosis is disturbed, autoreactive lymphocytes might persist and give rise to
excessive autoantibody levels in the blood. Therefore, the rise in anti-DNA antibodies just
before a disease exacerbation in SLE [37] might thereby be preceded by a rise in soluble
Fas levels. In this study we tested this hypothesis by studying sFas levels in relation to
SLEDAI scores, in longitudinal plasma samples.
Materials and methods
Generation of monoclonal antibodies against the Fas receptor.

Human recombinant Fas (rFas) cloning and expression were as follows: the cDNA for the
extracellular domain of Fas was obtained by RT-PCR. Primers were designed based on
the published sequence [5]. The 3' primer was designed to encode for a C-terminal glu-glu
epitope tag [38]. The cDNA was ligated into the pAcC13 vector [39]. Two microgram
plasmid DNA and 0.5 µg linearized wild type viral DNA were cotransfected into Sf9 cells
[40]. Recombinant virus was isolated by plaque purification [41]. Supernatants from Sf9
cells were 10 times concentrated and passed over an anti-glu-glu/protein G column.
Recombinant Fas was eluted with glu-glu peptide and peak fractions were pooled and
dialyzed versus phosphate buffered saline (PBS). Purity was more than 95%. Protein
determination was made by amino acid analysis.
Two balb-c mice were immunised and 2 times boostered with 10 µg of this preparation of
rFas in montanide (Seppic, France). Mouse sera were tested for the presence of Fas
antibody by radio immunoassay (RIA). In that assay mouse sera were incubated with rat-
anti-mouse kappa-light chain coupled to Sepharose (Pharmacia, Sweden) in the presence
of 125I labeled rFas. Spleen cells of the mouse displaying a clear anti-Fas signal were
fused with non-secreting mouse myeloma cells Sp2/0 under standard conditions [42].
Positive wells were cloned twice by limiting dilution cloning and 21 stable Fas antibody
producing clones were isolated. Culture supernatants of these clones were concentrated
and affinity purified on protein-A Sepharose CL-4B (Pharmacia, Uppsala, Sweden).
Reactivity of these antibodies with wild type mFas was analysed by FACS analysis. Anti-
Fas antibodies were biotinylated following manufacturer's instruction using LC-biotin-N-
hydroxysuccinimide ester (Pierce chemical Co., Rockford, IL) and assayed on Jurkat and
SKW6.4 cells at 5 µg/ml and 200.000 cells per sample. Streptavidin-coupled fluorescein
isothiocyanate (Becton Dickinson Immunocytochemistry Systems, San José, CA) was
used (1:50) as fluorescent marker. All 21 anti-Fas antibodies reacted with Jurkat and
SKW6.4 cells. At least six of them showed bioactivity: CLB-CD95/10, /15 and /18
11
Chapter II
induced apoptosis while CLB-CD95/2, CLB-CD95/6 and CLB95/19 prevented apoptosis

induced by either anti-Fas antibody or TCR triggering of Jurkat cells. The monoclonal
antibodies to Fas will be described in detail elsewhere.
ELISA for soluble Fas.

To develop a sensitive sandwich ELISA for the detection of soluble Fas different
combinations of monoclonal anti-Fas antibodies were evaluated. The use of CLB-CD95/2
to coat plates and biotin-coupled CLB-CD95/ 6 as a conjugate resulted in the most
sensitive assay. The ELISA was performed as follows. Microtiter plates (Nunc-Immuno
plate Maxisorp surface, Nunc, Denmark) were coated with 100 µl/well CLB-CD95/2 (2
µg/ml) in 0.1 M NaHCO3/Na2CO3 buffer (pH=9.6) overnight at room temperature.
Coated plates were washed 5 times with 100 µl/well of phosphate buffered saline (PBS)
containing 0.02% Tween-20 (PBST). Samples and standards were diluted in high
performance ELISA (HPE) buffer (CLB, Amsterdam, The Netherlands) and 100 µl of
each sample dilution was added to the plate. To each sample dilution, 10 l of a 10 g/ml
solution of biotin-coupled CLB-CD95/6 was added and the plate was incubated for 2
hours at room temperature. After 5 washes with 100 µl/well PBST 100 µl of streptavidine-
polyHRP (CLB, Amsterdam, The Netherlands) diluted 1:10,000 in PBS containing 2%
whole milk was incubated for 30 minutes at room temperature. The plates were washed 5
times with 100 µl/well PBST, and developed with 100 µl substrate solution (0.1 mg/ml
3,5,3',5'-tetramethylbenzidine, Merck, Darmstadt, Germany), containing 0.003% H202 in
0.11 M NaAc (pH=5.5)) for 10 minutes. The enzyme reaction was stopped with 100 µl 2
M H2SO4. Plates were read at 450 nm in a Titertek Multiskan reader (Labsystems
Multiskan Multisoft, Helsinki, Finland).
Patients and plasma samples.

This study concerned a cohort of SLE patients fulfilling the 1982 revised ACR criteria
[43] who participated in prospective longterm clinical follow-up studies [44,45]. Patients
were evaluated at least every three months at the outclinic. The SLE disease activity index
[46] was calculated at every outclinic visit from signs and symptoms and routine
laboratory tests.
Special attention was paid to the occurrence of infections in order to discriminate between
infection and disease activity. Dosages of drugs were recorded and exacerbations were
defined as described previously [43,44]. Blood samples were drawn in EDTA monthly
and plasma was stored at -80ºC. Exacerbations are defined as described previously [37]
and are shown in table 1. For this study the first 20 patients (median age: 30.5±9 years,
19 females, 1 male) who experienced an exacerbation (12 major and 8 minor) during this
prospective observation were randomly selected. Analysed were blood samples from 6
monthly time points before the exacerbation (t=-6,-5,-4,-3,-2,-1), from the moment of the
exacerbation (t=0), and from one month afterwards (t=1). In addition, from the patients
who did not experience an exacerbation during the follow-up studies, the first 10 patients
12
Soluble Fas in SLE
(median age: 42±11 years, 9 females, 1 male) were randomly selected. During that period
doses of immunosuppression and/or corticosteroids were not changed. From those patients
Table 1:
Criteria for major and minor disease exacerbations in SLE :
Criteria for major exacerbation: fulfilling one or more of the following *:

1. severe renal disease:
a. recent renal biopsy showing active proliferative lupus nephritis (>50% of glomeruli
affected, and/or
b. decrease of creatinin clearance of >25% within 4 months, accompanied by an active
sediment (>5 ery's, h.p.f., and/or casts) and by proteinuria of >0.5 gram/day)
2. severe central nervous system disease: seizures, cerebral vascular accident, coma,
transverse myelitis, psychosis, choreathetosis, central nerve palsy
3. hematological disease: hemolytic anemia (Hb<3.8 mmol/l) and/or thrombocytopenia
(<50x109 /l)
4. severe serositis: pericarditis with (impending) tamponade and/or massive pleural effusion
5. uveitis and/or retinal vasculitis
6. myocarditis with arrhythmia and/or congestive heart failure
7. severe myositis with proximal muscle weakness
8. lung involvement with hemoptysis
9. major vasculitis: with ulcerations and/or mononeuritis multiplex
10. miscellaneous: fever (>38oC rectally), serositis, hemolytic anemia (>3.8 mmol/l) or throm-
bocytopenia (>50x109/l), all without improvement after prednisolone in a maximum dosage
of 30 mg/day during at least one week
Criteria for minor exacerbation: fulfilling all of the following items:

1. increase of activity index by > 2 points within 6 months with a minimal activity score of 3
points accompanied by:
2. the clinically judged necessity to start prednisolone at a dosage of at least 10 mg/day, or to
increase the prednisolone dosage with > 5 mg/day, or to start with anti-malarials, or
immunosuppressive drugs, and:
3. not fulfilling the criteria for a major exacerbation.
• Only features occurring within 2 weeks of the outclinic visit or admission under consideration
are taken into account
6 consecutive monthly samples were analysed for sFas. As a control group, samples from
25 healthy laboratory donors (median age: 42±12 years, 14 females, 11 males) were
assayed for sFas.
13
Chapter II
Statistical analysis.
Comparisons between patient groups and healthy donors were calculated using the Mann-
Whitney U test (2-tailed). Spearman's test was applied for detecting correlation between
SLEDAI and sFas. P values less than 0.05 were considered significant.
Results
Combination of CLB-CD95/2 as coating antibody with biotinylated CLB-CD95/6 as

conjugate resulted in a sensitive sandwich ELISA with a detection limit for recombinant
Fas of 4 pg/ml (twice the background). Titration of sera of 5 healthy blood bank donors in
this assay resulted in dose-response curves parallel to the rFas curve as demonstrated in
figure 1, indicating comparable affinity for rFas and sFas. The median sFas level in 62
healthy donors was 647±176 pg/ml. The mean interassay variation of this ELISA was
15.9 % and the mean intra assay variation was 6.7 %. Assay readings were not affected by
the use of plasma instead of serum (data not shown).
Figure 1: Comparison of recombinant Fas and soluble Fas in serum samples of 5 healthy blood
donors (bd5). Soluble Fas levels were determined in duplicate. Detection limit was 4 pg/ml.
Soluble Fas levels at t=-6 were analysed for 25 control donors (C) and the 30 SLE
patients (S) (figure 2a). The latter were divided in patients who developed a clinical
exacerbation (E) six months later, and those with quiescent disease (Q) (figure 2a).
Soluble Fas was higher in all SLE patients relative to controls (1167±347 versus 618±98
pg/ml, p<0.0001). Most important, patients who are going to have a relapse, have higher
sFas levels than those who are not (1236±402 versus 809±276 pg/ml, p=0.028), while
14
Soluble Fas in SLE
SLEDAI scores (figure 2b) and immunosuppressive medication did not differ between
both groups at that time point. At all other time points (t=-5, -4, -3 ,-2, -1, 0, 1) the result
of this analysis was identical (data not shown).
Figure 2: (A) Comparison of recombinant sFas levels at the start of the follow-up between
controls (C), 30 SLE patients (S), the subgroup of SLE patients who experienced an exacerbation
during follow-up (E), and the subgroup of exacerbation patients who were exacerbation-free (Q)
during follow-up. ○: quiescent patients, ●: patients who were going to have a relapse at t = 0. P
values of comparisons between the different groups are given. (B) Comparison of SLEDAI scores
of the subgroup of exacerbation patients (E) and quiescent patients (Q) at start and at the time of
a relapse (for exacerbation patients) or at the end of follow-up (for quiescent patients).
Individual sFas curves for each patient group were analysed and no correlation was found
between sFas and SLEDAI. In figure 3a, three examples of sFas curves and SLEDAI
scores of exacerbation patients (E1,E2,E3) are shown. Most sFas curves were similar to
the one of patient E1, but different curves existed too, like E2 and E3 . Furthermore, in
figure 3b, 3 examples of non-exacerbation patients (Q1,Q2,Q3) and in figure 3c, 3
examples of controls (C1,C2,C3) are shown. When we looked at the SLE patients as a
group we found a strong correlation (r=0.75, p<0.0001, n=30) between SLEDAI and sFas
at t=0, the moment of relapse (figure 4a) , but no correlation at all at t=-6, when all
patients were quiescent (figure 4b). The most interesting aspect of our study is that there
was also a correlation between soluble Fas levels at t=-6 and SLEDAI scores at t=0
(figure 4c). These results suggest that sFas levels in SLE patients are rather constant, and
might be related to the severity of a coming relapse.
Discussion
For the retrospective analysis of soluble Fas (sFas) levels in SLE blood samples from a
prospectively designed longitudinal study, we developed a sensitive and reproducible
sandwich ELISA using a combination of 2 monoclonal anti-Fas antibodies, CLB-CD95/2
15
Chapter II
and CLB-CD95/6. Affinities of the anti-Fas antibodies for recombinant Fas and blood
sFas were comparable and levels in serum and plasma of the same donor were equal.
Figure 3: Soluble Fas curves for (A) 3 exacerbation patients (E1-3); (B) for 3 non-exacerbation
patients (1-3); and (C) (opposite) for 3 controls (C1-3). sFas values (●) are given on the left y-
axis and SLEDAI values (□) on the right y-axis.
We analysed sFas in plasma samples in patients with SLE. We found that soluble Fas
levels were elevated in patients compared to healthy controls, as in 5 other studies [4,28-
31]. A difference in patient population is not a very likely explanation why Knipping et al.
[32] and Goel et al. [33] did not find elevated sFas levels in SLE, as both investigators
16
Soluble Fas in SLE
(right) Figure 4: correlation for the group of 30 SLE patients between sFas levels and SLEDAI
scores. ●: exacerbation patients, ○: quiescent patients. (A) correlation between sFas and SLEDAI
at time t = 0 (r=0.75, p<0.0001). (B) correlation between sFas and SLEDAI at time t = -6 (r =
0.31, p = not significant. (C) correlation between sFas at t = -6 and SLEDAI t = 0 (r = 0.58, p
=0.0009).
analysed both active and inactive patients. In addition, in patients with quiescent SLE,
sFas levels were higher than in controls. Since the assay used by Knipping et al [32] has a
detection limit of 2 ng/ml, which also appears to be the detection limit in the assay used
17
Chapter II
by Goel et al [33] it is likely that these assays are not sensitive enough to detect elevated
serum levels of sFas.
We observed that already at the start of the follow-up there was a significant difference in
sFas levels between the group of patients who were going to have an exacerbation later
on, and the group of patients who were not, although both groups were clinically
quiescent at that stage and did not differ in their use of immunosuppressives and/or
corticosteroids.
This difference was found for all time points analysed, including the moment that all 20
patients had active disease (t=0). This confirms the study by Jodo et al. [28], who
compare sFas levels between active SLE (SLEDAI >9) and inactive SLE (SLEDAI<10).
In individual patients we did not find a correlation between sFas and SLEDAI, but sFas
did correlate to SLEDAI scores in the whole group of patients, when we looked at one
time point (t=0).This shows that the relation between disease activity and sFas is
completely dependent on the patient population blood samples are drawn from, i.e., the
greater the number of patients with active disease, the greater the likelihood there will be a
relation between sFas and SLEDAI. That is why Jodo et al. [28] found a correlation
between sFas and SLEDAI also.
Remarkably, sFas levels at t=-6 correlate with the SLEDAI score at t=0, indicating that
sFas levels in SLE patients are rather constant, and might be related to the severity of a
coming relapse. Renal abnormalities, as occur in SLE patients may influence levels of
small sized circulating proteins in various ways. Proteinuria may result in loss of sFas
from the circulation, whereas renal failure may decrease, to some extent, the clearance of
this polypeptide. We found no relation between creatinin clearance or proteinuria and
soluble Fas levels. In addition, we compared sFas values of patients who had an
exacerbation with renal involvement (n=7) to patients who had not (n=13) and found no
significant difference in sFas levels between patients with renal involvement and others
(data not shown).
Although sFas levels in SLE patients were elevated, the amounts of sFas circulating are
fairly low, rendering a pathophysiological role for soluble Fas doubtful: Papoff et al. [27]
found that a concentration of 20-100 ng/ml sFas is necessary for inhibition of apoptosis in
two cell lines. However, the expression levels of Fas ligand on cells certainly play an
important role. Furthermore, it cannot be excluded that sFas is produced in specific tissues
in larger amounts and exerts its effects there locally.
Elevated soluble Fas levels have been measured in tissues of in a considerable number of
diseases [4], [28-31], [32-34,47-52], [53-58]. From this it can be concluded that elevation
of soluble Fas is not specific for rheumatic diseases.
Regarding the source of sFas in serum, it is suggested by Cheng et al. [4] and by Papoff et
al. [27] that sFas is produced by alternative splicing. However, this could not be
confirmed by Rose et al. [31] who do not find a difference in the production of mRNA for
soluble Fas between controls and SLE patients. Possibly, the Fas receptor is released from
cells upon immune activation, as is described for the TNF receptors and the endothelial
18
Soluble Fas in SLE
adhesion molecules sICAM1, E-selectin and sVCAM1 after recombinant human tumor
necrosis factor treatment of cancer patients [59].
We conclude that sFas levels are elevated in SLE patients, but in individual patients sFas
does not fluctuate with disease activity. Our results indicate that sFas levels are constantly
elevated and that sFas levels might be indicative for the severity of an exacerbation. It
remains to be determined from which cells sFas originates and whether there is a relation
with activation markers on immune cells. The question at what moment sFas levels get
elevated in active SLE is currently under investigation in a prospective study, covering
more than 6 months prior to an exacerbation.
References
1. Lynch, D. H., Watson, M. L., Alderson, M. R., Baum, P. R., Miller, R. E., Tough, T.,
Gibson, M., Davis, S. T., Smith, C. A., Hunter, K. 1994. The mouse Fas-ligand gene is
mutated in gld mice and is part of a TNF family gene cluster. Immunity 1:131.
2. Theofilopoulos, A. N. and Dixon, F. J. 1985. Murine models of systemic lupus
erythematosus. Adv Immunol 37:269.
3. Nagata, S. and Suda, T. 1995. Fas and Fas ligand: lpr and gld mutations. Immunol Today
16:39.
4. Cheng, J., Zhou, T., Liu, C., Shapiro, J., Brauer, M., Kiefer, M., Barr, P., and Mountz, J.
1994. Protection from Fas-mediated apoptosis by a soluble form of the Fas molecule.
Science 263:1759.
5. Itoh, N., Yonehara, S., Ishii, A., Yonehara, M., Mizushima, S., Sameshima, M., Hase, A.,
Seto, Y., and Nagata, S. 1991. The polypeptide encoded by the cDNA for human cell
surface antigen Fas can mediate apoptosis. Cell 66:233.
6. Leithauser, F., Dhein, J., Mechtersheimer, G., Koretz, K., Bruderlein, S., Henne, C.,
Schmidt, A., Debatin, K. M., Krammer, P. H., and Moller, P. 1993. Constitutive and
induced expression of APO-1, a new member of the nerve growth factor/tumor necrosis
factor receptor superfamily, in normal and neoplastic cells. Lab Invest 69:415.
7. Debatin, K. M., Goldman, C. K., Waldmann, T. A., and Krammer, P. H. 1993. APO-1-
induced apoptosis of leukemia cells from patients with adult T-cell leukemia. Blood
81:2972.
8. Trauth, B. C., Klas, C., Peters, A. M., Matzku, S., Moller, P., Falk, W., Debatin, K. M., and
Krammer, P. H. 1989. Monoclonal antibody-mediated tumor regression by induction of
apoptosis. Science 245:301.
9. Moller, P., Henne, C., Leithauser, F., Eichelmann, A., Schmidt, A., Bruderlein, S., Dhein,
J., and Krammer, P. H. 1993. Coregulation of the APO-1 antigen with intercellular adhesion
molecule-1 (CD54) in tonsillar B cells and coordinate expression in follicular center B cells
and in follicle center and mediastinal B-cell lymphomas. Blood 81:2067.
19
Chapter II
10. Suda, T., Takahashi, T., Golstein, P., and Nagata, S. 1993. Molecular cloning and
expression of the Fas ligand, a novel member of the tumor necrosis factor family. Cell
75:1169.
11. Watanabe, F. R., Brannan, C. I., Itoh, N., Yonehara, S., Copeland, N. G., Jenkins, N. A.,
and Nagata, S. 1992. The cDNA structure, expression, and chromosomal assignment of the
mouse Fas antigen. J Immunol 148:1274.
12. Giese, T. and Davidson, W. F. 1992. Evidence for early onset, polyclonal activation of T
cell subsets in mice homozygous for lpr. J Immunol 149:3097.
13. Singer, G. G. and Abbas, A. K. 1994. The fas antigen is involved in peripheral but not
thymic deletion of T lymphocytes in T cell receptor transgenic mice. Immunity 1:365.
14. Crispe, I. N. 1994. Fatal interactions: Fas-induced apoptosis of mature T cells. Immunity
1:347.
15. Alderson, M. R., Tough, T. W., Davis, S. T., Braddy, S., Falk, B., Schooley, K. A.,
Goodwin, R. G., Smith, C. A., Ramsdell, F., and Lynch, D. H. 1995. Fas ligand mediates
activation-induced cell death in human T lymphocytes. J Exp Med 181:71.
16. Brunner, T., Mogil, R. J., LaFace, D., Yoo, N. J., Mahboubi, A., Echeverri, F., Martin, S. J.,
Force, W. R., Lynch, D. H., Ware, C. F., and et, a. l. 1995. Cell-autonomous Fas
(CD95)/Fas-ligand interaction mediates activation-induced apoptosis in T-cell hybridomas.
Nature 373:441.
17. Dhein, J., Walczak, H., Baumler, C., Debatin, K. M., and Krammer, P. H. 1995. Autocrine
T-cell suicide mediated by APO-1/(Fas/CD95). Nature 373:438.
18. Ju, S. T., Panka, D. J., Cui, H., Ettinger, R., el, K. M., Sherr, D. H., Stanger, B. Z., and
Marshak, R. A. 1995. Fas(CD95)/FasL interactions required for programmed cell death
after T-cell activation. Nature 373:444.
19. Miyawaki, T., Uehara, T., Nibu, R., Tsuji, T., Yachie, A., Yonehara, S., and Taniguchi, N.
1992. Differential expression of apoptosis-related Fas antigen on lymphocyte
subpopulations in human peripheral blood. J Immunol 149:3753.
20. Itoh, N., Tsujimoto, Y., and Nagata, S. 1993. Effect of bcl-2 on Fas antigen-mediated cell
death. J Immunol 151:621.
21. Klas, C., Debatin, K. M., Jonker, R. R., and Krammer, P. H. 1993. Activation interferes
with the APO-1 pathway in mature human T cells. Int Immunol 5:625.
22. Henkart, P. A. 1996. ICE family proteases: mediators of all apoptotic cell death? Immunity
4:195.
23. Muzio, M., Chinnaiyan, A., Kischkel, F., O'Rourke, K., Shevchenko, A., Ni, J., Scaffidi, C.,
Bretz, J., Zhang, M., Gentz, R., Mann, M., Krammer, P., Peter, M., and Dixit, V. 1996.
FLICE, a novel FADD-homologous ICE/CED-3-like protease, is recruited to the CD95
(Fas/APO-1) death-inducing signaling complex. Cell 85:817.
24. Boldin, M. P., Goncharov, T. M., Goltsev, Y. V., and Wallach, D. 1996. Involvement of
MACH, a novel MORT1/FADD-interacting protease, in Fas/APO-1- and TNF receptor-
induced cell death. Cell 85:803.
20
Soluble Fas in SLE
25 Alnemri, E. S., Livingston, D. J., Nicholson, D. W., Salvesen, G., Thornberry, N. A., Wong,
W. W., and Yuan, J. 1996. Human ICE/CED-3 protease nomenclature [letter]. Cell 87:171.
26. Mysler, E., Bini, P., Drappa, J., and et, a. 1994. The apoptosis-1/Fas protein in human
systemic lupus erythematosus. J Clin Invest 93:1029.
27. Papoff, G., Cascino, I., Eramo, A., Starace, G., Lynch, D. H., and Ruberti, G. 1996. A N-
terminal domain shared by Fas/APO-1 (CD95) soluble variant prevent cell death in vitro. J
Immunol 156:4622.
28. Jodo, S., Kobahashi, S., Kayagaki, N., Ogura, N., Feng, Y., Amasaki, Y., Fujisaki, A.,
Azuma, M., Yagita, H., Okumura, K., and Koike, T. 1997. Serum levels of soluble
Fas/APO-1 (CD95) and its molecular structure in patients with systemic lupus
erythematosus (SLE) and other patients. Clin Exp Immunol 107:89.
29. Tokano, Y., Miyake, S., Kayagaki, N., Nozawa, K., Morimoto, S., Azuma, M., Yagita, H.,
Takasaki, Y., Okumura, K., and Hashimoto, H. 1996. Soluble Fas molecule in the serum of
patients with systemic lupus erythematosus. J Clin Immunol 16:261.
30. Nozawa, K., Kayagaki, N., Tokano, Y., Yagita, H., Okumura, K., and Hasimoto, H. 1997.
Soluble Fas (APO-1, CD95) and soluble Fas ligand in rheumatic diseases. Arthritis Rheum
40:1126.
31. Rose, L., Latchman, D., and Isenberg, D. 1997. Elevated soluble fas production in SLE
correlates with HLA status not with disease activity. Lupus 6:717.
32. Knipping, E., Krammer, P. H., Onel, K. B., Lehman, T. J., Mysler, E., and Elkon, K. B.
1995. Levels of soluble Fas/APO-1/CD95 in systemic lupus erythematosus and juvenile
rheumatoid arthritis. Arthritis Rheum 38:1735.
33. Goel, N., Ulrich, D. T., St, C. E., Fleming, J. A., Lynch, D. H., and Seldin, M. F. 1995.
Lack of correlation between serum soluble Fas/APO-1 levels and autoimmune disease.
Arthritis Rheum 38:1738.
34. Kohno, T., Brewer, M. T., Baker, S. L., Schwartz, P. E., King, M. W., Hale, K. K., Squires,
C. H., Thompson, R. C., and Vannice, J. L. 1990. A second tumor necrosis factor receptor
gene product can shed a naturally occurring tumor necrosis factor inhibitor. Proc Natl Acad
Sci U S A 87:8331.
35. Digel, W., Porzsolt, F., Schmid, M., Herrmann, F., Lesslauer, W., and Brockhaus, M. 1992.
High levels of circulating soluble receptors for tumor necrosis factor in hairy cell leukemia
and type B chronic lymphocytic leukemia. J Clin Invest 89:1690.
36. Fukunaga, R., Seto, Y., Mizushima, S., and Nagata, S. 1990. Three different mRNAs
encoding human granulocyte colony-stimulating factor receptor. Proc Natl Acad Sci U S A
87:8702.
37. Swaak, A. J., Aarden, L. A., Statius, van, Eps, Lw, and Feltkamp, T. E. 1979. Anti-dsDNA
and complement profiles as prognostic guides in systemic lupus erythematosus. Arthritis
Rheum 22:226.
38. Grussenmeyer, T., Scheidtmann, K. H., Hutchinson, M. A., Eckhart, W., and Walter, G.
1985. Complexes of polyoma virus medium T antigen and cellular proteins. Proc Natl Acad
Sci U S A 82:7952.
21
Chapter II
39. Munemitsu, S., Innis, M. A., Clark, R., McCormick, F., Ullrich, A., and Polakis, P. 1990.
Molecular cloning and expression of a G25K cDNA, the human homolog of the yeast cell
cycle gene CDC42. Mol Cell Biol 10:5977.
40. Kitts, P. A., Ayres, M. D., and Possee, R. D. 1990. Linearization of baculovirus DNA
enhances the recovery of recombinant virus expression vectors. Nucleic Acids Res 18:5667.
41. Smith, G. E., Summers, M. D., and Fraser, M. J. 1983. Production of human beta interferon
in insect cells infected with a baculovirus expression vector. Mol Cell Biol 3:2156.
42. Helle, M., Boeije, L., de, G. E., de, V. A., and Aarden, L. 1991. Sensitive ELISA for
interleukin-6. Detection of IL-6 in biological fluids: synovial fluids and sera. J Immunol
Methods 138:47.
43. Tan, E. M., Cohen, A. S., Fries, J. S., Masi, A. T., McShane, D. J., Rothfield, N. F., Green
Schaller, J., Talal, N., and Winchester, R. J. 1982. The 1982 revised criteria for the
classification of systemic lupus erythematosus. Arthritis Rheum 25:1271.
44. Spronk, P. E., Bootsma, H., Huitema, M. G., Limburg, P. C., and Kallenberg, C. G. 1994.
Levels of soluble VCAM-1, soluble ICAM-1, and soluble E-selectin during disease
exacerbations in patients with systemic lupus erythematosus (SLE); a long term prospective
study. Clin Exp Immunol 97:439.
45. Bootsma, H., Spronk, P., Derksen, R., de, B. G., Wolters, D. H., Hermans, J., Limburg, P.,
Gmelig, M. F., Kater, L., and Kallenberg, C. 1995. Prevention of relapses in systemic lupus
erythematosus. Lancet 345:1595.
46. Bombardier, C., Gladman, D. D., Urowitz, M. B., Caron, D., Hsing Chang, C., and Comm
Prog Sty SLE. 1992. Derivation of the SLEDAI: A disease activity index for lupus patients.
Arthritis Rheum 35:630.
47. Hasunuma, T., Kayagaki, N., Asahara, H., Motokawa, S., Kobata, T., Yagita, H., Aono, H.,
Sumida, T., Okumura, K., and Nishioka, K. 1997. Accumulation of soluble Fas in inflamed
joints of patients with rheumatoid arthritis. Arthritis Rheum 40:80.
48. Bansil, S., Holtz, C. R., Cook, S. D., and Rohowsky, K. C. 1997. Serum sAPO-1/Fas levels
in multiple sclerosis. Acta Neurol Scand 95:208.
49. Inoue, A., Koh, C. S., Sakai, T., Yamazaki, M., Yanagisawa, N., Usuku, K., and Osame, M.
1997. Detection of the soluble form of the Fas molecule in patients with multiple sclerosis
and human T-lymphotropic virus type I-associated myelopathy. J Neuroimmunol 75:141.
50. Jiang, J. D., Schlesinger, M., Sacks, H., Mildvan, D., Roboz, J. P., and Bekesi, J. G. 1997.
Concentrations of soluble CD95 and CD8 antigens in the plasma and levels of
CD8+CD95+, CD8+CD38+, and CD4+CD95+ T cells are markers for HIV-1 infection and
clinical status. J Clin Immunol 17:185.
51. Seishima, M., Takemura, M., Saito, K., Ando, K., and Noma, A. 1997. Increased serum
soluble Fas (sFas) concentrations in HCV-positive patients with liver cirrhosis [letter]. J
Hepatol 27:424.
52. Nishigaki, K., Minatoguchi, S., Seishima, M., Asano, K., Noda, T., Yasuda, N., Sano, H.,
Kumada, H., Takemura, M., Noma, A., Tanaka, T., Watanabe, S., and Fujiwara, H. 1997.
22
Soluble Fas in SLE
Plasma Fas ligand, an inducer of apoptosis, and plasma soluble Fas, an inhibitor of
apoptosis, in patients with chronic congestive heart failure. J Am Coll Cardiol 29:1214.
53, Okuyama, M., Yamaguchi, S., Nozaki, N., Yamaoka, M., Shirakabe, M., and Tomoike, H.
1997. Serum levels of soluble form of Fas molecule in patients with congestive heart
failure. Am J Cardiol 79:1698.
54. Kamihira, S., Yamada, Y., Hiragata, Y., Yamaguchi, T., Izumikawa, K., Matsuo, Y.,
Sugahara, K., Tsuruta, K., Atogami, S., Tsukasaki, K., Maeda, T., and Tomonaga, M. 1997.
Serum levels of soluble Fas/APO-1 receptor in human retroviral infection and associated
diseases. Intern Med 36:166.
55. Munker, R., Midis, G., Owen-Schaub, L., and Andreff, M. 1996. Soluble Fas (CD95) is not
elevated in the serum of patients with myeloid leukemias, myeloproliferative and
myelodysplastic syndromes. Leukemia 10:1531.
56. Midis, G. P., Shen, Y., and Owen, S. L. 1996. Elevated soluble Fas (sFas) levels in
nonhematopoietic human malignancy. Cancer Res 56:3870.
57. Mogi, M., Harada, M., Kondo, T., Mizuno, Y., Narabayashi, H., Riederer, P., and Nagatsu,
T. 1996. The soluble form of Fas molecule is elevated in parkinsonian brain tissues.
Neurosci Lett 220:195.
58. Shinohara, K., Ayame, H., and Yoshida, T. 1997. Increased levels of soluble Fas antigen in
the plasma of the patients with refractory anemia [letter]. Int J Hematol 65:413.
59. Boehme, M. W., Waldherr, R., Kist, A., Riedl, S., Walter, K. R., Kempeni, J., and Raeth, U.
1996. Kinetics of soluble TNF-receptors and soluble adhesion molecules ICAM-1, E-
selectin and VCAM-1 under systemic rhTNF alpha therapy. Eur J Clin Invest 26:404.
23
Chapter III
Do elevated levels of serum soluble Fas contribute to

the persistence of activated lymphocytes in
Systemic Lupus Erythematosus?
Marc Bijl1
Thea van Lopik2
Pieter C. Limburg1
Peter E. Spronk1
Sonja M.H.J. Jaegers3
Lucien A. Aarden2
Ruud J. T. Smeenk2
1
Department of Clinical Immunology, University Hospital, Groningen
2
CLB, Sanguin Blood Supply Foundation, Laboratory of Experimental and Clinical
Immunology, Academical Medical Center, University of Amsterdam, Amsterdam
3
Department of Research and development, Rehab Center Beatrixoord, Haren, The
Netherlands
J Autoimmunity 1998;11:457-63
Chapter III
Summary
Systemic lupus erythematosus (SLE) is characterised by generalised immune activation.

Part of this might be explained by a decreased rate of apoptosis, possibly related to eleva-
ted levels of soluble Fas (sFas) which can inhibit Fas mediated apoptosis of lymphocytes.
In order to substantiate the relation between levels of sFas and lymphocyte activation in
SLE we monitored sFas levels, lymphocyte activation and disease activity in twenty five
SLE patients. SLEDAI-scores were registered and sera were assayed for sFas levels by an
enzyme-linked immunosorbent assay. Flow cytometry was used to monitor the state of
activation of lymphocyte subsets. Eighteen healthy, age-matched, volunteers served as
controls. Soluble Fas levels were elevated in SLE patients (n=25) compared to healthy
controls (n=18, p=0.002). Soluble Fas levels correlated with SLEDAI-scores (r=0.45,
p=0.02). Levels of sFas correlated with the percentages of activated B cells defined as
CD20+CD38+ cells (r=0.47, p=0.009). Percentages of CD20+CD38+ cells were increased
in quiescent SLE compared to healthy controls (p=0.003). The expression of activation
markers on CD4+ T-lymphocytes (IL-2R, p=0.04; HLA-DR, p=0.01) and CD8+ T-
lymphocytes (HLA-DR, p=0.007) was increased as well in quiescent SLE compared to
controls. Activation markers on all lymphocyte subsets tended to increase further during
disease activity. No correlation was observed between percentages of activated T-
lymphocyte subsets and levels of sFas. In conclusion, soluble Fas levels are increased in
SLE patients and correlate with disease activity as measured by the SLEDAI-score. B and
T cell subsets are activated even during quiescent SLE. Serum levels of sFas do correlate
with percentages of activated B cells but not with that of activated T cells.
Introduction
Systemic lupus erythematosus (SLE) is a systemic autoimmune disorder characterised by

polyclonal activation of B lymphocytes and the production of a wide range of
autoantibodies. B cell proliferation in SLE has been suggested to be T cell dependent and
persistence of autoreactive B and T-lymphocytes is thought to be responsible for
hypergammaglobulinaemia and autoantibody production in SLE [1]. One of the
mechanisms by which elimination of autoreactive lymphocytes takes place is programmed
cell death (apoptosis). A defect in apoptosis may thus contribute to the development of
autoimmune disease [2]. One of the major routes which triggers apoptosis is the Fas
pathway. Fas, also known as APO-1 or CD95, is a cell surface protein of 45-48 kD
belonging to the tumor necrosis factor (TNF)/nerve growth factor (NGF) receptor family
[3,4]. Fas is expressed on rapidly proliferating cells and is strongly up-regulated on
lymphocytes upon activation [5]. Cross-linking of Fas by Fas ligand (FasL), a membrane
glycoprotein of 40 kD [4], triggers apoptotic cell death with characteristic cytoplasmic
and nuclear condensation and DNA fragmentation [6]. In the lpr and gld murine models
26
Serum-soluble Fas in SLE
the development of SLE is directly linked to a defect in the Fas apoptotic pathway. It has
been shown that in lpr and gld mice mutations in Fas and FasL, respectively, underlie the
development of an autoimmune syndrome with loss of T cell tolerance, B cell defects,
hypergammaglobulinaemia and autoantibody production [7-10]. In humans, recent
findings illustrate the importance of Fas in the development of autoimmune and
lymphoproliferative diseases [11,12]. Up till now, no common defects in the function or
expression of the Fas antigen or FasL itself have been found in human SLE [13,14]. These
findings do not imply, however, that apoptosis is normal in lupus patients. Lymphocytes
from human SLE patients display an increased rate of spontaneous apoptosis in vitro
compared to lymphocytes of healthy donors and patients with rheumatoid arthritis [14,15].
The persistent occurrence of activated B and T cells even in quiescent disease in vivo [16]
points, however, to a decreased rate of apoptosis. These seemingly conflicting data can be
explained by the occurrence of elevated levels of soluble Fas (sFas) in human SLE [17-
21]. Soluble Fas results from alternative splicing and deletion of the transmembrane exon
of the Fas molecule and is able to inhibit apoptosis [17,18]. Although the presence of
elevated levels of sFas in SLE has been debated [22,23], we recently confirmed the data
of Cheng and Jodo, showing increase in levels of sFas during active disease [24].
In this study we tested the hypothesis that elevated levels of sFas, even in quiescent SLE,
contribute to the persistence of activated lymphocytes, by inducing a decreased rate of
apoptosis. We monitored sFas and related sFas levels to disease activity and lymphocyte
activation.
Materials and Methods
Patients and controls

Consecutive outclinic patients fulfilling at least four revised American College of Rheu-
matology (ACR) criteria for the diagnosis of SLE [25] could participate in this study. Pa-
tients who were pregnant or those who showed signs or symptoms suggestive of an
infection at the moment of blood sampling were excluded. Eighteen healthy, age-matched,
volunteers served as controls.
Study design
Patients were seen at the outpatient department for clinical evaluation. Disease activity
was scored and the SLE Disease Activity Index (SLEDAI) was calculated [26]. According
to pre-defined criteria (table 1) it was decided whether or not the patient had a relapse of
the disease [27]. Attention was paid to the occurrence of infections. Dosages of pred-
nisolone and/or immunosuppressives were recorded. Blood samples were drawn in EDTA
(Vacutainer; Becton Dickinson, Mountain View, CA) and heparin (Vacutainer, BD)
between 8.30 and 10.00 a.m. to minimize circadian variations of circulating lymphocyte
subsets. In case patients used corticosteroids medication was temporarily withdrawn for at
27
Chapter III
least 24 hours before blood sampling to exclude direct effects of corticosteroids. Samples
were double-stained and analysed by flow cytometry (FACStar, BD). Plasma was stored
at -200 C until needed for measurement of sFas. Blood samples from patients with a
relapse were drawn before immunosuppressives were started.
Table 1:
Criteria for major and minor disease exacerbations in SLE:

affected, and/or
(<50x109 /l)

28
Antibodies and staining procedure

Phycoerythrin (PE) or fluorescein isothiocyanate (FITC) conjugated monoclonal
antibodies (MoAb) were used for staining blood samples (table 2). In brief, 10µl MoAb
was added to 100 µl of whole blood as described [16]. After incubation for 15 minutes at
room temperature in the dark, 2µl FACS-lysing solution (BD), diluted 1:10 in milipore
water, was added. Samples were incubated for another 10 minutes and washed with
phosphate buffered saline (PBS)/heparin. After the addition of 150 µl PBS/heparin, 104
cells per sample were measured on a FACStar. Cells were gated for lymphocyte
characteristics using both forward and right angle scatter, as well as a dual staining using
CD14 and CD45 (Leukogate, BD) for 'back-gating'. In case of severe lymphopenia, a
'live-gate' was placed to obtain maximal results. Background red and green fluorescence
was determined using combinations of CD20 (PE) with CD4 or CD8 (FITC) as a control
for calculations in the CD4+ and CD8bright subsets respectively. As a control for
calculations within the CD20+ subset, dual-staining for CD4 (PE) and CD20 (FITC) was
used. Positive cells were determined by setting a region with reference to these controls.
The list-mode data were analysed by the Lysis Stat software package (BD) for the
calculation of percentages of populations. All analyses were performed consistently by the
same person, without prior knowledge of the subjects' clinical status.
Measurement of anti-dsDNA antibodies and sFas

Anti-dsDNA antibodies were detected by Farr-assay using 125I-labelled recombinant ds-
DNA (Diagnostic Products Corporation, Los Angeles, USA) which is free of con-
tamination with ssDNA. Farr assay was performed according to the manufacturer's in-
struction and positive samples were measured at different dilutions to obtain
measurements within the range of the assay. Results of this assay were expressed in IU/ml
using Wo/80 as the ultimate standard [28]. Normal value of this Farr-assay in our
laboratory is < 10 IU/ml; intra- and interassay variations are both less than 10 %.
Soluble Fas was detected by sandwich ELISA as described [24]. The anti-human Fas
specific monoclonal antibodies CLB-CD95/2 and CLB-CD95/6 that were used in this
assay, had been generated by immunizing adult Balb/c mice with recombinant human Fas.
In brief, microtitre plates (NUNC maxisorp, Nunc, Denmark) were coated with 100
µl/well CLB-CD95/2 (2 µg/ml) in 0.1 M NaHCO3/Na2CO3 buffer (pH=9.6) overnight at
room temperature. Coated plates were washed 5 times with 100 µl/well phosphate buf-
fered saline (PBS) containing 0.02% Tween-20 (PBST). Samples and standards were
diluted in high performance ELISA buffer (CLB, Amsterdam, The Netherlands) and 100
µl of each sample dilution was added to the plate. Next, 10 µl of a 10 µg/ml solution of
biotin-labelled CLB-CD95/6 was added and the plate was incubated for 2 hours at room
temperature. After washing (5 times with 100 µl/well PBST) 100 µl of streptavidine-
polyHRP diluted 1:10.000 in PBS containing 2% whole milk was incubated for 30
minutes at room temperature. The plates were washed 5 times with 100 µl/well PBST and
29
Chapter III
developed with 100 µl substrate solution (0.1 mg/ml 3,5,3',5'-tetramethylbenzidine,

Merck, Darmstadt, Germany), 0.003% H2O2 in 0.11 M NaAc (pH=5.5) for 10 minutes.
The enzyme reaction was stopped by adding 100 µl 2 M H2S04. Plates were read at 450
nm in a Titertek Multiscan reader (Labsystems Multiskan Multisoft, Helsinki, Finland).
Background absorbance at 540 nm was subtracted. Intra- and interassay variations were
6.7% and 15.9%, respectively.
Table 2:
Monoclonal antibodies used for the detection of surface markers on circulating
Lymphocytes
mAb specificity Label Recognised subsets Source

CD45/CD14 FITC/PE All leukocytes/monocytes ('leukogate') BD
CD3 FITC T cells BD
CD4 FITC/PE Helper/inducer T cells MCA
CD8 FITC Cytotoxic/suppressor T cells/NK cells MCA
CD20 FITC/PE B cells MCA
CD25 PE IL-2 receptor; activated T and B cells BD
CD38 PE Activated T and B cells BD
HLA-DR PE Activated T and B cells, and others BD
BD, Becton Dickinson, Mountain View, CA; MCA, MCA Development BV, Groningen, The
Netherlands. FITC, fluorescein isothiocyanate, PE, phycoerythrin.
Statistics
Differences between groups were calculated using the Student t-test (2-tailed). Correlat-
ions were determined by the Pearson correlation coefficient. P values less than 0.05 were
considered significant.
Results
During the study period between January 1992 and June 1994, 25 patients (22 female, 3
male) and 18 controls (16 female, 2 male) were included. Their mean age was 34 (range
21-55) and 30 (19-45) years, respectively. SLE was diagnosed mean 9 years before entry
of the study. Fourteen patients were analysed at the time of relapse. Five of the relapses
were classified as major relapse and 9 as minor relapse. Further characteristics of the
relapses are given in table 3. Eleven patients were analysed during quiescent disease.
SLEDAI was increased in patients with active disease (p=0.002) and levels of anti-
dsDNA tended to be higher (p=0.07). There were no differences in the use of prednisone
or azathioprine between patients with quiescent or active disease (table 4). None of the
patients was treated with cyclosporin or cyclophosphamide.
30
Table 3:
Characteristics of 14 relapses in patients with systemic lupus erythematosus
Number of patients with symptom present

major exacerbations minor exacerbations
(n=5) (n=9)
renal disease 5 -
central nervous system disease 2 1
serositis - 5
haematological disease 3 3
polyarthritis 1 2
skin involvement 1 1
fever - 1
Soluble Fas
Levels of sFas in controls and patients are shown in figure 1. In controls (n=18) levels of
soluble Fas were 628 ± 142 pg/ml (mean ± SD). Compared to controls, levels of sFas in
patients with quiescent SLE were significantly higher (p=0.02, table 4). Soluble Fas levels
were comparable between patients with quiescent disease and those with a minor relapse
(866 ± 313 and 907 ± 443, respectively).
Table 4:
Levels of soluble Fas, leucocyte counts and percentages of lymphoid subsets in controls and
clinically quiescent and active SLE patients
controls quiescent disease active disease

(n=18) (n=11) (n=14)
Prednisolon (mg/day) 3.9 ± 4.9 5.9 ± 6.4
Azathioprine (mg/day) 13.4 ± 37.7 15.4 ± 42.7
*
sFas (pg/ml) 628 ± 142 866 ± 313 998 ± 375#
Leucocytes (x 106/ml) 6.8 ± 1.8 7.0 ± 5.6 5.1 ± 2.7
6
Lymphocytes (x 10 /ml) 1.6 ± 0.45 1.4 ± 1.0 0.55 ± 0.36###
% CD4+HLA-DR+ 5.2 ± 1.6 11.3 ± 6.6** 15.7 ± 11.5##
% CD4+IL-2R+ 24.9 ± 7.9 34.1 ± 12.3* 39.4 ± 14.4##
% CD8+HLA-DR+ 14.6 ± 8.7 23.1 ± 6.8** 29.2 ± 20.2#
% CD20+CD38+ 58.0 ± 13.7 78.3 ± 15.7** 85.1 ± 11.5###
Mean ± SD values of different patient characteristics; corresponding p values were calculated by
Student t-test. Percentages are expressed as percentages of total CD4+, CD8bright, and CD20+
subpopulations.* p<0.05, ** p<0.01, comparison between controls and quiescent SLE. # p<0.05, ##
p<0.01, ### p<0.001, comparison between controls and active SLE.
31
Chapter III
Levels of sFas in patients with a major relapse were higher (1160 ± 113), but this
difference was not significant when compared to sFas levels in patients with a minor
relapse or patients with quiescent disease (p=0.06). Between active (all minor and major
relapses together) and clinically quiescent patients no differences in sFas levels could be
observed. As sFas tended to increase with disease activity we looked for a correlation
between sFas and the SLEDAI score (n=24). A correlation did exist (r=0.45, p=0.02), as
shown in figure 2. This correlation was due to the relation between sFas levels and
SLEDAI scores in patients with active disease (r=0.68, p=0.05).
p<0.01
p=0.02 ns
ns ns
2000 p=0.03
sFas (pg/ml)
1000
0
Control Quiescent Minor Major
Figure 1: Serum levels of soluble Fas in healthy controls (n=18), clinically quiescent SLE
(n=11), and clinically active SLE represented as minor relapses (n=9) or major relapses (n=5).
Analysis of lymphocyte subsets

The number of leukocytes and lymphocytes did not differ between healthy controls (n=18)
and SLE patients with quiescent disease (n=11), (table 4). The expression of activation
markers on CD4+ T-lymphocytes (IL-2R, HLA-DR) and on CD8+ T-lymphocytes (HLA-
DR) was increased. Also, percentages of activated CD20+ B-lymphocytes were higher in
quiescent SLE compared to healthy controls. The number of lymphocytes in active
disease was lower (p=0.003) compared to quiescent disease. No significant changes were
detected in the percentages of activated B and T-lymphocytes between quiescent and
active disease, although the expression of activation markers on all lymphocyte subsets
further increased during active disease.
Relation between sFas and lymphocyte activation

As lymphocyte activation was increased in SLE and activated lymphocytes express Fas
we sought for correlations between sFas levels and activation markers on lymphocyte
32
subsets. Levels of soluble Fas did not correlate with the number of leukocytes or
lymphocytes. Neither did we find an association between the percentages of CD4+ IL-2R+
lymphocytes and levels of sFas nor between the percentages of CD4+HLA-DR+
lymphocytes and levels of sFas. Percentages of activated CD8+ lymphocytes did not
correlate with levels of sFas. There was, however, a significant correlation between sFas
levels and the percentages of CD20+CD38+ cells (r=0.47, p=0.009).
1800
1600
sFas (pg/ml) 1400
1200
1000
800
600
400
200
0
0 10 20 30 40
SLEDAI
Figure 2: Correlation between levels of sFas and SLEDAI-scores in SLE patients (n=24),
Pearson correlation coefficient = 0.45 (p = 0.01)
Discussion
We demonstrate that levels of sFas in patients with SLE are increased, even during quies-
cent disease. Levels of sFas correlated with disease activity scores and with the extent of
activation of circulating B cells.
In several studies elevated levels of sFas in lupus patients have been reported [17-21]. Al-
though other studies could not confirm these results [22,23], we recently also found
increased levels of sFas in SLE [24]. In the present study, dealing with a group of SLE
patients different from that studied previously [24], similar results were observed. Diffe-
rent affinities of antibodies generated with recombinant human Fas for catching native Fas
may be responsible for the discrepancies in the published data concerning sFas levels in
SLE patients. Furthermore, the existence of multiple forms of sFas cannot be excluded.
Finally, patient selection and use of immunosuppressive drugs may have influenced
results as corticosteroids lower levels of sFas [17]. In the present study this may have
resulted in a masking of even more pronounced elevation of sFas, as patients with
quiescent disease as well as those with active disease used immunosuppressives (table 4).
As Fas expression increases at the moment of lymphocyte activation [5,29] and sFas
levels were found to be elevated in SLE patients we related sFas levels to disease activity
and the percentages of activated peripheral blood lymphocytes. Indeed, a correlation
33
Chapter III
between sFas levels and SLEDAI-scores was observed, in agreement with the studies of
Jodo et al and Tokano et al [19,20]. In our previous study we correlated SLEDAI-scores
with sFas levels of individual patients, longitudinally monitored [24]. No association was
found. This discrepancy can be explained by fluctuations of sFas levels in SLE patients in
time independent of disease activity. Levels of sFas in controls were more stable in time
(618 ± 98 pg/ml, mean ± SD) compared to patients with quiescent disease (809 ± 276
pg/ml). Indeed, when patients of the previous study were analysed during active disease
separately, a correlation was found between sFas levels and SLEDAI-score (p=0.02,
r=0.66, Pearson correlation coefficient). Other studies did not demonstrate an association
between disease activity and sFas levels [21-23] and differences in genetic background
were suggested to explain this discrepancy. As the genetic background of our patients
(mainly Caucasian) was probably to that of the patients in the European studies mentioned
before [21-23], other factors seem to be of influence. Particularly, differences in the assay
used for the measurement of sFas may be relevant, which argues for standardisation of
sFas assays.
Percentages of activated (CD38+) CD20+ B-lymphocytes in quiescent SLE were increased
compared to normal controls. Also, the percentages of activated (IL-2R+, HLA-DR+)
CD4+ T-lymphocytes were higher in quiescent SLE patients compared to controls. In
patients with active disease percentages of CD4+HLA-DR+ T cells, CD8+HLA-DR+ T
cells and CD20+CD38+ B cells were even higher although not statistically significant. This
probably is due to the small number of patients studied, as in a previous study dealing
with more data, significantly higher percentages of activated lymphocytes were observed
in active disease compared to quiescent disease [16].
We found a correlation between sFas levels and activated B cells but not between sFas
levels and percentages of activated T cells. The absence of a correlation between sFas
levels and activation state of T cells might be explained by the fact that percentages of
activated CD4+ and CD8+ T-lymphocytes were lower than that of B cells (table 4). Fur-
thermore, a relation can be masked because most sFas is probably bound to membrane
bound Fas ligand (or soluble Fas ligand) of which increased expression in activated T
cells has been reported [30,31]. Finally, lymphocyte activation was assessed in the periph-
eral blood only, whereas the majority of activated lymphocytes that contribute to sFas
production are localised in the extravascular compartment.
So far the significance of elevated sFas levels is unknown. Soluble Fas might play a
pathophysiological role in the persistence of lymphocyte activation in SLE as it has been
shown that sFas is functional and able to block Fas-mediated apoptosis [16,17].
Correlations between sFas levels and lymphocyte activation were sought because Fas
expression and soluble Fas can be considered as markers of lymphocyte activation com-
parable with other cell-activation markers such as IL-2R as suggested by Ohsako et al [5].
Many of these markers are shed upon cell activation and can be detected in the circulation
as soluble molecules [32-34]. For example soluble IL-2R levels are increased in SLE
patients [32]. Increase of soluble Fas might also be the result from shedding of Fas which
34
is upregulated on activated cells. Alternatively, sFas can arise by concomitant

upregulation of the alternative splicing pathway on cell activation [17]. In this view
Kovacs et al described one SLE patient out of 16 whose T cells secreted sFas lacking the
transmembrane domain [18].
As B cells, particularly in SLE patients, are activated it might be assumed that the
correlation between sFas levels and the percentage of activated B-lymphocytes found is a
reflection of lymphocyte activation. A contributing role of sFas to persistence of B cell
activation is suggested by our findings but needs further study.
In conclusion, this study shows that soluble Fas levels in SLE are increased and are
correlated with disease activity as measured by SLEDAI. B and T cell subsets are
activated even during quiescent SLE. Levels of sFas do correlate with percentages of
activated B cells but not with that of activated T cell subsets. The exact role of soluble Fas
is unknown and has to be elucidated. A prospective study in which, next to soluble Fas,
also soluble FasL, lymphocyte activation, Fas and FasL expression and apoptosis are
measured in relation to disease activity, will give further insight in the possibly disturbed
balance of apoptosis in SLE.
References
1. Linker-Israeli M, Quismorio FP, Horwitz DA. CD8+ lymphocytes from patients with
systemic lupus erythematosus sustain, rather than suppress, spontaneous polyclonal IgG
production and synergize with CD4+ cells to support autoantibody synthesis. Arthritis
Rheum 1990;33:1216-25.
2. Mountz JD, Wu J, Cheng J, Zhou T. Autoimmune disease. A problem of defective
apoptosis. Arthritis Rheum 1994;37:1415-20.
3. Nagata S, Golstein P. The Fas death factor. Science 1995;267:1449-56.
4. Suda T, Nagata S. Purification and characterization of the Fas-ligand that induces apoptosis.
J Exp Med 1994;179:873-9.
5. Ohsako S, Hara M, Harigarai M, Fukasawa C, Kashiwazaki S. Expression and function of
Fas antigen and bcl-2 in human systemic lupus erythematosus lymphocytes. Clin Immunol
Immunopathol 1994;73:109-14.
6. Itoh N, Yonehara S, Ishii A, et al. The polypeptide encoded by the cDNA for human cell
surface antigen Fas can mediate apoptosis. Cell 1991; 66:233-43.
7. Zhou T, Bluethmann H, Eldridge J, Berry K, Mountz JD. Abnormal thymocyte
development and production of autoreactive T cells in TCR transgenic autoimmune mice. J
Immunol 1991;147:466-74.
8. Sobel ES, Cohen PL, Eisenberg RA. Lpr T cells are necessary for autoantibody production
in lpr mice. J Immunol 1993;150:4160-7.
9. Reap EA, Sobel ES, Cohen PL, Eisenberg RA. Conventional B cells, not B-1 cells are
responsible for producing autoantibodies in lpr mice. J Exp Med 1993;177: 69-78.
35
Chapter III
10. Ravirajan CT, Sarraf CE, Anilkumar TV, Golding MCH, Alison MR, Isenberg DA. An
analysis of apoptosis in lymphoid organs and lupus disease in murine systemic lupus
erythematosus (SLE). Clin Exp Immunol 1996;105:306-12.
11. Rieux-Laucat F, Le Deist F, Hivroz C, et al. Mutations in Fas associated with human
lymphoproliferative syndrome and autoimmunity. Science 1995;268:1347-9.
12. Fisher GH, Rosenberg FJ, Straus SE, et al. Dominant interfering Fas gene mutations impair
apoptosis in a human autoimmune lymphoproliferative syndrome. Cell 1995;81:935-46.
13. Mysler E, Bini P, Drappa J, et al. The apoptosis-1/Fas protein in human systemic lupus
erythematosus. J Clin Invest 1994;93:1029-34.
14. Emlen W, Niebur J, Kadera R. Accelerated in vitro apoptosis of lymphocytes from patients
with systemic lupus erythematosus. J Immunol 1994;152:3685-92.
15. Lorenz HM, Grunke M, Hieronymous T, et al. In vitro apoptosis from patients with
systemic lupus erythematosus and other autoimune diseases. Arthitis Rheum 1997;40:306-
17.
16. Spronk PE, Horst G, Gun BTF van der, Limburg PC, Kallenberg CGM. Anti-dsDNA
disease in systemic lupus erythematosus. Clin Exp Immunol 1996;104: 446-53.
17. Cheng J, Zhou T, Liu C, et al. Protection from Fas-mediated apoptosis by a soluble form of
the Fas molecule. Science 1994;263:1759-62.
18. Kovacs B, Szentendrei T, Bednarek JM, et al. Persistent expression of a soluble form of
Fas/APO1 in continuously activated T cells from a patient with SLE. Clin Exp Rheumatol
1997;15:19-23.
19. Jodo S, Kobayashi S, Kayagaki N, et al. Serum levels of soluble Fas/APO-1 (CD95) and its
molecular structure in patients with systemic lupus erythematosus (SLE) and other
autoimmune diseases.. Clin Exp Immunol 1997;107:89-95.
20. Tokano Y, Miyake S, Kayagaki N, et al. Soluble Fas molecule in the serum of patients with
systemic lupus erythematosus. J Clin Immunol 1996;16:261-5.
21. Rose M, Latchman DS, Isenberg DA. Elevated soluble fas production in SLE correlates
with HLA status not with disease activity. Lupus 1997;6:717-22.
22. Knipping E, Krammer PH, Onel KB, Lehman TJA, Mysler E, Elkon KB. Levels of soluble
Fas/APO-1/CD95 in systemic lupus erythematosus and juvenile reumatoid arthritis.
Arthritis Rheum 1995;38:1735-7.
23. Goel N, Ulrich DT, St. Clair EW, Fleming JA, Lynch DH, Seldin MF. Lack of correlation
between serum soluble Fas/APO-1 levels and autoimmune disease. Arthritis Rheum
1995;38:1738-43.
24. Lopik Th v, Bijl M, Smeets-Boeije L, et al. Changing soluble Fas levels in longtitudinal
series of plasma samples of systemic lupus erythematosus (SLE) patients. J Rheumatol.
1999;26:60-7.
25. Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the classification of
systemic lupus erythematosus. Arthritis Rheum 1982;25:1271-7.
36
26. Bombadier C, Gladman DD, Urowitz MB, Caron D, Chang CH. Derivation of the SLEDAI:
a disease activity index for lupus patients. Arthritis Rheum 1992;35:630-40.
27. Bootsma H, Spronk PE, Limburg PC, et al. Prevention of relapses of systemic lupus
erythematosus (SLE) by treatment with corticosteroids based on changes of anti-dsDNA
levels: a long-term controlled prospective study. Lancet 1995;345:1595-9.
28. Feltkamp TEW, Kirkwood TBL, Maini RN, Aarden LA. The first international standard for
antibodies to double stranded DNA. Ann Rheum Dis 1988;47:740-6.
29. Miyawaki T, Uehara T, Nibu R, et al. Differential expression of apoptosis-related Fas
antigen on lymphocyte subpopulations in human peripheral blood. J Immunol 1992;149:
3753-8.
30. Dhein J, Walczak H, Bäumler C, Debatin KM, Krammer PH. Autocrine T-cell suicide
mediated by APO-1/(Fas/CD95). Nature 1995;373:438-41.
31. Kovacs B, Liossis SNC, Dennis GJ, Tsokos GC. Increased expression of functional Fas-
ligand in activated T cells from patients with systemic lupus erythematosus. Autoimmunity
1997;25:213-21.
32. Rubin LA, Kurman, CC, Fritz ME, et al. Soluble interleukin 2 receptors are released from
activated human lymphoid cells in vitro. J Immunol 1985;135:3172-7.
33. Spronk PE, ter Borg EJ, Limburg PE, Kallenberg CGM. Changes in levels of soluble T-cell
activation markers, sIL-2R, sCD4, and sCD8, in relation to disease exacerbations in patients
with systemic lupus erythematosus; a prospective study. Ann Rheum Dis 1994;53:235-9.
34. Boehme MWJ, Waldherr R, Kist A, et al. Kinetics of soluble TNF-receptors and soluble
adhesion molecules ICAM-1, E-selectin and VCAM-1 under systemic rhTNFa therapy. Eur
J Clin Invest 1996;26:404-10.
37
Chapter IV
Fas expression on peripheral blood lymphocytes in

Systemic Lupus Erythematosus (SLE): relation to
lymphocyte activation and disease activity
Marc Bijl1
Gerda Horst1
Pieter C. Limburg2
1
2
Department of Pathology and Laboratory Medicine, University Hospital, Groningen,
The Netherlands
Submitted
Chapter IV
Summary
Levels of apoptotic lymphocytes have been found increased in SLE and persistence of
apoptotic cells has been associated with autoantibody production. Increased lymphocyte
Fas (CD95) expression due to lymphocyte activation may account for increased
susceptibility for Fas-mediated apoptosis in SLE. Flow cytometry was performed to
evaluate membrane expression of Fas in combination with the activation markers CD25,
HLA-DR, and CD38 on, respectively, CD4+, CD8+, and CD19+ lymphocytes of SLE
patients with inactive (n=20) and with active disease (n=13). SLEDAI-scores were
calculated. Healthy volunteers (n=14) served as controls. Percentages of CD4+ T-cells
expressing CD25 and CD19+ B-cells expressing CD38 were increased in patients with
active disease compared to controls (p=0.03, p=0.04, respectively). In contrast to CD4+
and CD8+ cells, percentages of CD19+ cells expressing Fas were increased in SLE patients
with active disease (p=0.0002, versus controls). In these patients percentages of cells
double positive for both CD38 and Fas were increased compared to patients with inactive
disease (p=0.006) and controls (p=0.0007). Percentages of CD19+ cells expressing Fas
correlated with SLEDAI-scores.
In SLE patients, percentages of Fas expressing B-lymphocytes are increased, are related to
the state of lymphocyte activation, and correlate to disease activity. We suggest that
increased Fas expression results in a higher susceptibility for Fas mediated apoptosis. This
might contribute to the increased levels of apoptotic lymphocytes in SLE patients.
Introduction
Systemic lupus erythematosus (SLE) is an autoimmune disease characterised by the

presence of autoantibodies to nuclear and cytoplasmic antigens. The etiopathogenesis of
the disease is unknown. Recently, a possible role of apoptotic cells in the pathogenesis of
SLE has been suggested [1]. First, it has been demonstrated that during the apoptotic
process autoantigens are exposed at the outer surface of the apoptotic cell. This might
allow the development of autoantibodies directed to intracellularly localised antigens [2].
Secondly, injection of large numbers of apoptotic cells in animal models was shown to
induce autoimmunity [3]. Indeed, in SLE patients increased levels of apoptotic
lymphocytes have been detected in the peripheral blood [4]. Why these cells are present in
the circulation is presently unknown. One explanation might be that phagocytosis of
apoptotic cells is impaired in SLE patients [5]. Alternatively, or in conjunction, the
production of apoptotic cells might be increased in lupus patients leading to an overflow
of the phagocytic capacity. It has been suggested that the increased production of
apoptotic cells results from increased lymphocyte activation [6] as is present in SLE
patients even during clinically inactive disease [7,8]. The major route for induction of
apoptosis in activated lymphocytes is mediated by Fas. Fas, also known as APO-1 or
40
Fas expression in SLE
CD95, is a cell surface protein of 45-48 kD belonging to the tumour necrosis factor
(TNF)/nerve growth factor (NGF) receptor family [9,10]. Cross-linking of Fas by Fas
ligand (FasL), a membrane glycoprotein of 40 kD, triggers apoptotic cell death with
characteristic cytoplasmic and nuclear condensation and DNA fragmentation [11]. Fas-
mediated apoptosis is functionally intact in SLE patients in vitro [12,13]. We wondered
whether the increased levels of apoptotic lymphocytes in vivo could be attributed to
increased lymphocyte activation concurrent with increased Fas expression in vivo. To
elucidate these questions Fas membrane expression was measured on peripheral blood
lymphocyte subpopulations in lupus patients in comparison to healthy controls, and results
were related to the state of lymphocyte activation and disease activity.
Patients and Methods
Patients and blood samples

Patients eligible for this study fulfilled at least 4 American College of Rheumatology
(ACR) criteria for SLE [14], were non-smokers, and had either active or inactive disease.
Patients with active disease had to fulfil criteria as shown in table 1. Inactive disease was
defined as the persistent absence of clinical disease activity for at least a 4 month period
while patients were without or on a constant dose of immunomodulating drugs. To
evaluate changes due to disease activity, reevaluation of patients with initially active
disease took place at a moment of inactive disease as well. The SLEDAI score was
calculated for each patient [15]. Healthy non-smoking volunteers, matched for age and
sex, were included as control.
Blood samples for anti-dsDNA detection were drawn in EDTA and were subsequently
analysed as described [16]. Heparinized blood was drawn for lymphocyte subset analysis.
Samples were stained and analysed the same day by flow cytometry.
Cell isolation and staining procedure

Peripheral blood mononuclear cells (PBMC) were isolated from heparinized blood by
lymphoprep density gradient centrifugation. For surface staining the following phycoe-
rythrin (PE) or fluorescein isothiocyanate (FITC) and allophycocyanine (APC) conjugated
monoclonal antibodies (mAb) were used: CD4 (APC), CD19 (APC), CD25 (FITC/PE),
CD38 (FITC/PE), HLA-DR (FITC/PE) and CD95 (PE) (Becton Dickinson, Mountain
View, CA); CD8 (APC) (Pharmingen, Hamburg, Germany). After washing with
phosphate buffered saline (PBS) supplemented with 1% bovine serum albumin (BSA)
staining of 106 cells was done by adding labelled mAb with subsequent incubation for 45
minutes on ice in the dark. Samples were washed with PBS/BSA. Three-colour
immunofluorescence analysis was performed on a Coulter Epics-Elite equipped with a
gated amplifier (Coulter Electronics, Mijdrecht, The Netherlands). Cells were gated for
41
Chapter IV
lymphocyte characteristics using both forward and sideward scatter. For final analysis
only those samples were included in which more than 500 cells of the respective subsets
were counted.
Table 1:

affected, and/or
(<50x109 /l)

Statistics
Differences between groups were evaluated using the student’s t-test or Mann-Whitney U
test when appropriate. Paired samples were analysed separately using the Wilcoxon rank
42
sum test. Spearman’s rank correlation was applied for detecting correlations between
different study parameters. A p value less than 0.05 was considered statistically
significant.
Results
Twenty SLE patients were studied during inactive disease and 13 patients at a moment of
active disease. Fourteen healthy volunteers served as control. Characteristics of the
patients are given in table 2.
Table 2:
Characteristics of patients with Systemic Lupus Erythematosus
inactive active
(n=20) (n=13)
Male/Female 3/17 1/12
age (years, mean, range) 43.4 (21-75) 39.2 (20-74)
Disease duration (years, mean, range) 10.9 (1-38) 7.5 (0-20)
SLEDAI (mean ± SEM) 2.6 ± 0.5 10.7 ± 1.4
Farr (IU/ml, mean ± SEM) 44.7 ± 13.1 149.2 ± 67.0
minor/major exacerbation - 7/6
renal disease - 5
central nervous system disease - 1
Serositis - 3
Hematological disease - 5
Polyarthritis - 4
skin involvement - 2
fever - 3
CD4+ T-lymphocytes of SLE patients with active disease showed signs of increased
activation reflected by increased expression of CD25, the interleukin-2 receptor (p=0.03
versus controls, p=0.01 versus patients with inactive disease, figure 1a). Percentages
CD8+ T-lymphocytes expressing the activation marker HLA-DR tended to be higher in
patients with active disease (p=0.08 versus controls). Percentages CD19+ B-lymphocytes
expressing the activation marker CD38 were increased in patients with active disease
compared to controls (p=0.04, figure 1c). Fas expression was measured on CD4+, CD8+ as
well as on CD19+ lymphocytes. Fas expression on CD4+ T-lymphocytes did not show
significant differences between patients and controls, irrespective of disease activity
43
Chapter IV
p=0.03
(figure 2). (a)
30 p=0.01
The percentage of Fas expressing CD8+
T-lymphocytes tended to be increased in
% CD25 positive CD4+

patients with active disease (p=0.06
lymphocytes
20
versus controls, figure 2). However, the
percentage of Fas expressing CD19+ B-
10
lymphocytes was strongly increased in
patients with active disease (p=0.0002,
versus controls). Also, patients with 0
control inactive active
active disease showed a higher (b) SLE SLE
percentage of Fas expressing B- 40 p=0.08
% HLA-DR positive CD8+

lymphocytes than patients with inactive
30
disease (p=0.02).
lymphocytes
To investigate whether Fas expression
20
was associated with lymphocyte
activation we analysed Fas expression in 10
conjunction with CD38 expression as a
membrane marker for B cell activation 0
and related the data to disease activity. In SLE SLE
SLE patients with active disease the p=0.04
percentage lymphocytes double positive (c) p=0.05

for CD38 and Fas was significantly 100
increased compared to patients with
% CD38 positive CD19+
90
inactive disease (p=0.006) as well as
lymphocytes
compared to controls (p=0.0007, figure 80

3). Consequently, lymphocytes double 70
negative for these markers were
significantly decreased in SLE patients 60
with active disease (p=0.0001 versus 50
controls, p=0.01 versus patients with SLE SLE
inactive disease). Furthermore, when we Figure 1: Percentages of activated
analysed the relation between Fas lymphocytes of the subpopulations analysed
expression and disease activity as in controls and SLE patients. A: percentage
measured by the SLEDAI-score a of CD4+ cells expressing the activation
significant correlation between the marker CD25.. B: percentage of CD8+ cells
percentage of Fas expressing CD19+ B expressing the activation marker HLA-DR.
cells and disease activity was found C: percentage of CD19+ cells expressing the
(figure 4). No correlation was found activation marker CD38. Horizontal lines
between disease activity and the denote the median. Only results of those
percentages Fas expressing CD4+- or samples are shown in which the number of
CD8+-T-lymphocytes. cells measured exceeded 500.
44
Discussion
(a)
In this study we found an increased 100

percentage of Fas expressing B-
CD4+ lymphocytes
75
% Fas positive
lymphocytes in SLE patients in vivo.
This confirms in vitro results in which it 50
was shown that Fas is properly
upregulated upon activation in SLE 25
[12,13] and it extends these findings to
the in vivo situation. In addition, using 0
three-colour flowcytometry, we showed SLE SLE
that Fas expression on B-lymphocytes is (b)

p=0.06
related to the state of activation of these 100

cells as well as to disease activity.
CD8+ lymphocytes
The relation between Fas expression,
% Fas positive 75
lymphocyte activation, and apoptosis is
50
very complex, in SLE patients in
particular. Many factors have been 25
identified which potentially influence the
interrelation of the factors mentioned. 0
For example, antibodies to FasL and to SLE SLE
poly (ADP-ribose) polymerase (PARP)
p=0.0002
have been found in SLE patients both of (c)
p=0.02
which may inhibit apoptosis [17,18]. 100
Elevated levels of sFas in the serum of
CD19+ lymphocytes
75
% Fas positive
SLE patients [8,19-23] also can inhibit

the induction of Fas-mediated apoptosis. 50
Furthermore, the use of medication can
influence the data obtained, especially 25
since many immunosuppressives which
are used in the treatment of SLE, such as 0
corticosteroids and azathioprine, induce SLE SLE
apoptosis thereby indirectly changing the

characteristics of the circulating cell
populations left [24]. Figure 2: Fas (CD95) expression on CD4+-
B cell activation, resulting in the (panel A), CD8+- (panel B) and CD19+-
production of autoantibodies, plays a lymphocytes (panel C) in controls and SLE
central role in the pathogenesis of SLE. patients with inactive and active disease.
In previous reports we showed that, prior Horizontal line denotes the median. Only
to a relapse, an oligoclonal expansion of results of those samples are shown in which
B cells capable of spontaneously the number of cells measured exceeded 500.
45
Chapter IV
producing anti-dsDNA in vitro can be

demonstrated [7].
p=0.04 p=0.01
Recently, the dominant primary role of B (a)
p=0.0001
cells has been demonstrated in the 30
negative CD19+ lymphocytes

MRL/+ Fas intact mice model [25]. In
% Fas negative and CD38

this model, a B cell-deficient strain was 20
created. Mice from this strain, in contrast
to the wild type MRL/+ mice with 10
functionally intact B cells, did not
develop nephritis or vasculitis. In line
0
with this role for B cells we showed that control inactive active
SLE SLE
in SLE patients, during disease activity,
indicators of lymphocyte activation are (b) p=0.0007
most pronounced for B cells. In addition, % Fas positive and CD38 positive 100 p=0.006
we demonstrated that upregulation of Fas

CD19+ lymphocytes
75
occurs in conjunction with this B cell
activation. 50
We could not demonstrate a significant
relation between membrane Fas 25
expression and the state of activation for
0
the T lymphocyte subpopulations control inactive active
SLE SLE
analysed. However, a tendency of
increased Fas expression on CD8+-
subpopulations of T-lymphocytes was Figure 3: Fas (CD95) expression on CD19+
present in patients with active disease. B-lymphocytes in relation to the expression
This is consistent with the results of of the activation marker CD38 in controls
others reporting increased expression of and SLE patients with inactive and active
Fas on the total lymphocyte population disease.
[26] as well as on T cells [27]. Horizontal lines denote the median.
Furthermore, also among CD4+- and
CD8+- subpopulations of T-lymphocytes
increased percentages of Fas positive
cells have been described [12,28].
46
We propose that, in SLE, Fas membrane expression increases upon cell activation which
renders peripheral blood lymphocytes susceptible for the induction of Fas-mediated
apoptosis.
As it has been demonstrated that the Fas pathway is functionally intact in SLE [12,
13],upregulation of membrane Fas probably will result in increased induction of apoptotic
lymphocytes after engagement of the Fas receptor. This will contribute to, or even cause,
lymphopenia, which is present in many SLE patients and aggravates during disease
activity. Indeed, Amasaki et. al. found an inverse relation between Fas antigen intensity
on CD4+ lymphocytes and absolute numbers of these cells [28].
A factor contributing to the presence of apoptotic cells might be defective clearance of
these cells [5]. Together, increased induction of apoptosis and a decreased clearance
capacity will result in the accumulation of apoptotic cells. During the process of apoptosis
cells present nuclear antigens on their surface whether or not in modified form [2].
Continuous overpresentation of these antigens through the accumulation of apoptotic cells
might break tolerance resulting in the autoimmune phenomena that characterise SLE or
may boost an already existing autoimmune response.
In conclusion, we demonstrate in this study that the percentage peripheral blood B
lymphocytes (CD19+) expressing Fas is increased in SLE patients and is related to the
state of lymphocyte activation and the extent of disease activity. These data underscore
the central role of B-lymphocytes in the pathogenesis of SLE. In addition, the data are
compatible with the concept that upregulation of membrane Fas renders lymphocytes
more susceptible for apoptosis resulting in increased rates of apoptotic PBL as described
in SLE. Persistence of these apoptotic cells may play a primary role in the
perpetuation of autoimmune responses.
r=0.52, p=0.0028
Figure 4: Correlation between the 100
percentage of Fas (CD95) expressing B
CD19+ lymphocytes
(CD19+) cells and disease activity as

% Fas positive
measured by the SLEDAI-score. All

moments of evaluation were included 50
(n=33)
0
0 10 20 30
SLEDAI-score
Acknowledgement: This study was supported by the Dutch Kidney Foundation (grant
NSN C95.1482).
47
Chapter IV
References
1. Andrade F, Casciola-Rosen L, Rosen A. Apoptosis in systemic lupus erythematosus.

Clinical implications. Rheum Dis Clin North Am 2000;26:215-27.
4. Perniok A, Wedekind F, Herrmann M, Specker C, Schneider M. High levels of circulating
early apoptic peripheral blood mononuclear cells in systemic lupus erythematosus. Lupus
1998;7:113-8.
5. Herrmann M, Voll RE, Zoller OM et al. Impaired phagocytosis of apoptotic cell material by
monocyte-derived macrophages from patients with systemic lupus erythematosus. Arthritis
Rheum 1998;41:1241-50.
6. Lorenz HM, Grunke M, Hieronymus T et al. In vitro apoptosis and expression of apoptosis-
related molecules in lymphocytes from patients with systemic lupus erythematosus and
other autoimmune diseases. Arthritis Rheum 1997;40:306-17.
8. Bijl M, van Lopik T, Limburg PC et al. Do elevated levels of serum-soluble fas contribute
to the persistence of activated lymphocytes in systemic lupus erythematosus? J Autoimmun
1998;11:457-63.
9. Trauth BC, Klas C, Peters AM et al. Monoclonal antibody-mediated tumor regression by
induction of apoptosis. Science 1989;245:301-5.
11. Suda T, Takahashi T, Golstein P, Nagata S. Molecular cloning and expression of the Fas
ligand, a novel member of the tumor necrosis factor family. Cell 1993;75:1169-78.
12. Mysler E, Bini P, Drappa J et al. The apoptosis-1/Fas protein in human systemic lupus
13. Laderach D, Bach JF, Koutouzov S. Nucleosomes inhibit phagocytosis of apoptotic
thymocytes by peritoneal macrophages from MRL+/+ lupus-prone mice. J Leukoc Biol
1998;64:774-80.
14. Tan EM, Cohen AS, Fries JF et al. The 1982 revised criteria for the classification of
15. Bombardier C, Gladman DD, Urowitz MB, Caron D, Chang CH. Derivation of the
SLEDAI. A disease activity index for lupus patients. The Committee on Prognosis Studies
in SLE. Arthritis Rheum 1992;35:630-40.
48

1990;33:634-43.
17. Suzuki N, Ichino M, Mihara S, Kaneko S, Sakane T. Inhibition of Fas/Fas ligand-mediated
apoptotic cell death of lymphocytes in vitro by circulating anti-Fas ligand autoantibodies in
patients with systemic lupus erythematosus. Arthritis Rheum 1998;41:344-53.
18. Decker P, Isenberg D, Muller S. Inhibition of caspase-3-mediated poly(ADP-ribose)
polymerase (PARP) apoptotic cleavage by human PARP autoantibodies and effect on cells
undergoing apoptosis. J Biol Chem 2000;275:9043-6.
19. Cheng J, Zhou T, Liu C et al. Protection from Fas-mediated apoptosis by a soluble form of
20. Nozawa K, Kayagaki N, Tokano Y et al. Soluble Fas (APO-1, CD95) and soluble Fas
ligand in rheumatic diseases. Arthritis Rheum 1997;40:1126-9.
21. Papo T, Parizot C, Ortova M et al. Apoptosis and expression of soluble Fas mRNA in
22. Courtney PA, Crockard AD, Williamson K et al. Lymphocyte apoptosis in systemic lupus
erythematosus: relationships with Fas expression, serum soluble Fas and disease activity.
Lupus 1999;8:508-13.
23. van Lopik T, Bijl M, Hart M et al. Patients with systemic lupus erythematosus with high
plasma levels of sFas risk relapse. J Rheumatol 1999;26:60-7.
24. Thompson CB. Apoptosis in the pathogenesis and treatment of disease. Science
1995;267:1456-62.
25. Chan OT, Madaio MP, Shlomchik MJ. B cells are required for lupus nephritis in the
polygenic, Fas-intact MRL model of systemic autoimmunity. J Immunol 1999;163:3592-6.
26. Courtney PA, Crockard AD, Williamson K et al. Increased apoptotic peripheral blood
neutrophils in systemic lupus erythematosus: relations with disease activity, antibodies to
double stranded DNA, and neutropenia. Ann Rheum Dis 1999;58:309-14.
27. Ohsako S, Hara M, Harigai M, Fukasawa C, Kashiwazaki S. Expression and function of Fas
antigen and bcl-2 in human systemic lupus erythematosus lymphocytes. Clin Immunol
28. Amasaki Y, Kobayashi S, Takeda T et al. Up-regulated expression of Fas antigen (CD95)
by peripheral naive and memory T cell subsets in patients with systemic lupus
erythematosus (SLE): a possible mechanism for lymphopenia. Clin Exp Immunol
1995;99:245-50.
49
Chapter V
Effects of smoking on activation markers, Fas

expression and apoptosis of peripheral blood
lymphocytes
Marc Bijl1
Gerda Horst1
Pieter C. Limburg2
1
2
The Netherlands
Eur J Clin Invest, accepted for publication

Chapter V
Abstract
Smoking influences numbers and function of peripheral blood lymphocytes (PBL) by a

process that is badly understood. We conducted this study to evaluate whether the immune
impairment of smoking might be related to changes in the expression or functionality of
Fas, a cell surface molecule that plays a central role in immune homeostasis and cytotoxic
activity.
PBL from 10 smoking and 10 non-smoking healthy volunteers were isolated.
Flowcytometry was performed to measure the state of activation, Fas expression and
apoptosis of PBL. Functionality of Fas was tested by assessing apoptosis after incubation
of isolated lymphocytes with agonistic anti-Fas antibodies in 4 smoking and 4 non-
smoking individuals.
Smoking was associated with an increase in the percentage of Fas expressing CD4+ T- and
B-lymphocytes. A decrease in the percentage of activated (CD38+) B cells was observed.
In vitro Fas-induced apoptosis did not appear different between smokers and non-
smokers. No differences in the percentages of circulating apoptotic lymphocytes could be
demonstrated between smoking and non-smoking individuals. Smoking is associated with
increased Fas expression on PBL in general, and on B cells in particular. This might
render these cells more susceptible for apoptosis. As Fas is functionally intact this may
also explain the reduced percentage of activated (CD38+) B cells found in smoking
individuals. The latter may contribute to the reduced humoral immune response observed
in smokers.
Introduction
Chronic cigarette smoking results in an impaired immune response. T cell responsiveness

is affected and T cell dependent antibody responses are decreased [1;2]. Numbers of
peripheral blood leukocytes cell counts change. For example, smoking results in
decreased numbers of natural killer (NK) cells [3]. Data concerning the influence of
smoking on lymphocyte subsets are conflicting. Tollerud et al. found a significant increase
in the percentage of CD4+ T-lymphocytes in smokers, without changes in the percentage
of B-lymphocytes [4]. Hockertz et al., however, demonstrated a decrease in the percentage
of T- as well as B-lymphocytes in smokers [5]. The mechanisms underlying these changes
in immune function and peripheral blood lymphocyte (PBL) distribution are badly
understood.
Fas, (APO-1, CD95) is a membrane receptor that plays a central role in immune
homeostasis and mediates essential signals for apoptosis in activated mature lymphocytes
[6]. In order to get more insight in possible mechanisms underlying selective losses and
impairment of immune function of lymphocytes we hypothesized that chronic cigarette
smoking is associated with increased Fas expression on PBL rendering these cells more
52
Fas expression in smokers
susceptible for apoptosis. Fas expression and the state of activation of PBL was analyzed
in smoking and non-smoking individuals. In addition, we assessed the occurrence of
apoptosis of PBL in vivo as well as the functionality of the Fas-mediated apoptotic
pathway in vitro.
Materials and Methods
Twenty healthy individuals were studied. All denied the presence of clinical symptoms or
signs, in particular wheezing, fever or cough, and physical examination was
unremarkable. Ten of them were cigarette smokers (age 42.9 ± 6.9 years, mean ± SD),
smoking 18.0 ± 2.8 cigarettes per day, with a number of 22.6 ± 8.6. pack-years. Ten were
non-smokers (age 30.9 ± 8.3 years).
FACS analysis and apoptosis assay

Peripheral blood mononuclear cells were isolated by lymphoprep density gradient
centrifugation within one hour after collection from heparinised blood. For surface
staining the following phycoerythrin (PE) or fluorescein isothiocyanate (FITC) and
allophycocyanine (APC) conjugated monoclonal antibodies (mAb) were used: CD4
(APC), CD19 (APC), CD25 (PE), CD38 (FITC), CD95 (PE) and HLA-DR (PE) (Becton
Dickinson, Mountain View, CA); CD8 (APC) (Pharmingen, Hamburg, Germany). After
washing cells with phosphate buffered saline (PBS) supplemented with 1% bovine serum
albumin (BSA) staining of 106 cells was done by adding labeled mAb with subsequent
incubation for 45 minutes on ice in the dark. Samples were washed with PBS/BSA. After
the addition of 222.5 µl binding buffer (10 mM hepes pH=7.4, 140 mM NaCl, 5 mM
CaCl2), 25 µl propidiumiodide (10 µg/ml, Molecular Probes, Leiden, The Netherlands)
and 2.5 µl FITC-labeled annexin V (Nexins Research BV, Hoeven, The Netherlands)
diluted 1:50 were added. Immediately after incubation for 10 minutes on ice in the dark,
three-color immunofluorescence analysis was performed on a Coulter Epics-Elite
equipped with a gated amplifier (Coulter Electronics, Mijdrecht, The Netherlands). Cells
were gated for lymphocyte characteristics using both forward and sideward scatter.
Percentages of cells staining positive for a particular marker within the population
analyzed were compared between smokers and non-smokers. In addition, we compared
mean fluorescence intensity (MFI) of staining between both groups. MFI represents the
intensity of the signal on those cells positively staining for the respective marker. All
analyses were performed consistently by the same person.
Apoptosis induction
After isolation (t0), cells were resuspended in medium (RPMI 1640 + 2 mM glutamin +
60 µg/ml gentamycin + 5 % human poolserum) with and without the addition of 5 µg/ml
anti-Fas (CLB-CD95/15), kindly provided by dr. L. Aarden (CLB, Amsterdam, The
53
Chapter V
Netherlands). Subsequently, cells were cultured for 24 hours (t1). Both immediately after
isolation and after 24 hours of culture cells were analysed for apoptosis assessment.
Statistics
Differences in parameters between smokers and non-smokers were evaluated by the
Mann-Whitney U test (2-tailed). Results of apoptosis induction were analysed within
groups using the Wilcoxon signed rank test. A p value ≤ 0.05 was considered significant.
Results
Smoking individuals had increased numbers of neutrophils and total lymphocyte counts in
their peripheral blood compared to non-smokers (table 1a). Absolute numbers and
percentages of circulating peripheral blood CD4+, CD8+ and CD19+ lymphocytes in
smokers were not significantly different compared to non-smoking individuals (table 1).
Interestingly, the percentages of Fas (CD95) expressing B-lymphocytes and CD4+ T-
lymphocytes were increased in smokers compared to non-smokers (table 1b).
Table 1a:
Absolute numbers of neutrophils, lymphocytes, and lymphocyte subsets
in smokers and non-smokers
non-smokers smokers
(n = 10) (n = 10)
Neutrophils 6.33 ± 2.02 8.42 ± 1.64*
Lymphocytes 1.80 ± 0.61 2.48 ± 0.41*
CD19+ lymphocytes 0.16 ± 0.08 0.24 ± 0.09
+
CD4 lymphocytes 0.78 ± 0.32 1.11 ± 0.32
+
CD8 lymphocytes 0.58 ± 0.24 0.79 ± 0.15
*
Mean ± SD values; p<0.05, p values were calculated using the Mann-Whitney U test (2-tailed).
As Fas is upregulated upon cell activation we analysed the relation between the
expression of Fas and the expression of activation markers on the various lymphocyte
subsets. Percentages of activated B-lymphocytes (CD38+), CD4+ T-lymphocytes (CD25+)
and CD8+ T-lymphocytes (HLA-DR+) are shown in table 1b. Percentages of activated B
cells were significantly decreased in smoking individuals. The difference in Fas
expression on B cells between smokers and non-smokers could be accounted for by an
increased percentage of Fas expressing, non-activated (CD38-) B-lymphocytes in smokers
(table 2).
Besides an increase in the percentage of Fas expressing CD4+ T-lymphocytes also mean
fluorescence intensity (MFI) of Fas on these cells was increased in smokers compared to
54
non-smokers (median value of 187 units and 129 units, respectively, p<0.05). MFI of Fas
on CD19+ cells and CD8+ cells and MFI-values of activation markers on the respective
subpopulations of lymphocytes were comparable between both groups.
Table 1b:
Percentages of lymphocyte subsets and expression of activation markers
and Fas (CD95) within these subsets in smokers and non-smokers
non-smokers smokers
(n = 10) (n = 10)
+
% CD19 7.4 ± 2.2 8.4 ± 3.1
+
% CD38 73.4 ± 6.1 54.0 ± 14.7***
% CD95+ 18.8 ± 6.5 36.3 ± 10.4***
% CD4+ 40.9 ± 4.9 44.6 ± 6.2
+
% CD25 4.0 ± 1.3 4.9 ± 1.5
+
% CD95 43.5 ± 9.1 57.7 ± 8.4**
% CD8+ 32.4 ± 6.1 30.2 ± 6.7
+
% HLA-DR 5.6 ± 2.5 4.2 ± 2.4
+
% CD95 55.8 ± 11.2 62.5 ± 10.6
+ + +
Mean ± SD values; percentages CD19 , CD4 , and CD8 are expressed as percentages of total
lymphocytes. Activation markers and Fas expression are expressed as percentage cells positive
within the CD19+, CD4+, and CD8+ subpopulations; **p<0.01, *** p<0.001, p values were
calculated using the Mann-Whitney U test (2-tailed).
To analyse whether the Fas-mediated pathway of apoptosis induction was functionally

intact in smokers, lymphocytes from 4 non-smokers and 4 smokers were incubated with
an agonistic anti-Fas mAb. Results of anti-Fas induced apoptosis of CD19+ cells are
shown in figure 1. The percentage Fas expressing lymphocytes increased in medium after
24 hours; increase of Fas expression occurred also after incubation with anti-Fas, though
less pronounced (data not shown). Percentages of apoptotic B cells were low directly after
isolation, increased after 24 hours incubation in medium alone, and, compared to the latter
value, reached significantly higher percentages after incubation with anti-Fas,
demonstrating the functionality of the Fas-induced apoptotic pathway (figure 1).
In smokers the percentage Fas expressing B cells after isolation was higher. Despite this
difference, after apoptosis induction in B cells percentages of apoptosis did not differ
between smokers and non-smokers (figure 1). Also, percentages of apoptosis reached in T
cells after apoptosis induction were comparable between smokers and non-smokers (data
not shown).
55
Chapter V
Table 2:
Co-expression of activation markers and Fas in CD19+ B-lymphocytes from
non- smokers and smokers
non-smokers smokers
(n = 10) (n = 10)
+ +
CD95 CD95 10.0 ± 4.7 12.4 ± 4.7
+ -
CD95 CD38 7.5 ± 3.8 22.7 ± 9.8***
CD95- CD38+ 69.3 ± 11.3 42.5 ± 17.8***
CD95- CD38- 13.3 ± 7.4 22.6 ± 10.4*
Mean ± SD values; percentages are expressed as percentage cells positive for the respective
activation markers within the subpopulations; *p<0.05, *** p<0.001; p values were calculated
using the Mann-Whitney U test (2-tailed).
As we found Fas-induced apoptosis similar in smokers and non-smokers in vitro, we

wondered whether differences in percentages of apoptotic lymphocytes could be
demonstrated between smoking and non-smoking individuals in vivo. Percentages of
circulating apoptotic lymphocytes were assessed by annexin V staining. Although
percentages of apoptotic CD19+ cells, CD4+ cells and CD8+ T cells were higher in
smokers versus non-smokers, differences were not statistically significant (table 3).
Table 3:
Percentages of total lymphocytes and lymphocyte subsets staining with annexin V
in smokers and non-smokers
non-smokers smokers p value

(n = 10) (n = 10)
+
% lymphocytes annexin V 4.0 ± 3.4 5.6 ± 8.9 0.82
+ +
% CD4 annexin V 2.0 ± 0.8 3.2 ± 4.3 0.97
+ +
% CD8 annexin V 2.0 ± 4.3 4.3 ± 5.2 0.28
+ +
% CD19 annexin V 7.9 ± 3.2 9.4 ± 13.4 0.19
Mean ± SD values; p values were calculated using the Mann-Whitney U test (2-tailed).
Discussion
The influence of smoking on numbers and function of PBL has been demonstrated in
several studies. T cell dependent and independent antibody responses are decreased. The
mechanisms responsible for these effects, however, are not well understood [2;3]. We
show that smoking is associated with increased expression of Fas on PBL, on B-
lymphocytes in particular.
56
As can be observed from the proportions of peripheral blood lymphocyte subsets which
express Fas, relatively larger effects of smoking were found on B cells. In smokers, the
percentage Fas expressing B cells was nearly doubled compared to non-smokers whereas
in the CD4+ subpopulation the proportion Fas positive cells only rised from 43.4 up to
57.7 percent. Thus, differential effects of smoking on lymphocytes subpopulations can not
be excluded. Differential effects of smoking on B- and T-lymphocytes might be explained
through fundamental differences between these cells. Apoptosis of B and T-lymphocytes
seems to be initiated via different pathways [7]. B cells need stimulation via the B cell
receptor in conjunction with a costimulatory signal for proliferation and differentiation
[8]. Inadequate signaling results in apoptosis that ensures a specific immune response
because bystander B cells will be eliminated. For example, stimulation of resting B cells
via CD40 ligand only leads to upregulation of Fas and increased susceptibility to Fas-
mediated cell death [9]. It is tempting to speculate that smoking serves as an inadequate
activating signal for lymphocytes, resulting in increased Fas expression. A higher
percentage of resting B cells will become Fas positive and will, upon further activation,
disappear by apoptosis. This hypothesis is supported by our in vitro results showing
comparable Fas-mediated apoptosis between smokers and non-smokers, although the
numbers of individuals studied were small. The decreased percentage of CD38+ B cells
that we found in smokers might be explained by elimination of CD38+, Fas expressing
cells via apoptosis. Seemingly in contradiction with this hypothesis is the lack of evidence
that we found for increased apoptosis of lymphocytes in smoking individuals, as
percentages of annexin V staining lymphocytes did not differ between smokers and non-
smokers. However, as apoptosis is an ongoing event and cigarette smoking in all
participants in this study could be considered a chronic stimulus, smoking might result in
slightly elevated levels of apoptotic lymphocytes only (table 3), due to continuous
clearance of apoptotic cells from the circulation. Only in cases when decreased clearance
of apoptotic cells has been suggested, such as in systemic lupus erythematosus [10], a
significant increase of circulating apoptotic cells will be observed. Nevertheless, the
absolute as well as relative numbers of circulating B cells were not decreased in smokers
although we are not informed about possible differences in the total (intra- and
extravascular) numbers of (B-) lymphocytes between smokers and non-smokers.
Fas expression on PBL of smoking individuals has been reported to be unaltered in one
other study [11]. However, this study measured Fas expression on total lymphocytes only,
and no analyses of subpopulations were performed. As B cells constitute a minority of
PBL, changes in Fas expression on these cells may have been missed. It is noteworthy that
in the same study it is demonstrated that Fas ligand is constitutively expressed on the
majority of T cells of individuals with chronic cigarette smoking. This supports our
concept that smoking stimulates lymphocytes since FasL is expressed on T cells after
activation only [12].
57
Chapter V
30 p=0.0078
p=0.04
% Apoptotic CD19+ lymphocytes

p=0.0078
20
10
timepoint t0 t1 t1a
hours 0 24 24
a Fas - - +
Figure: Percentages of apoptotic CD19+ B-lymphocytes non-smokers (n=4, open circles) and
smokers (n=4, black circles), analysed at: t0, immediately after lymphocyte isolation; t1, after 24
hours incubation in medium; t1a, after 24 hours incubation with anti-Fas following incubation
with anti-Fas antibodies. Apoptotic cells are depicted as those staining positive with annexin V.
In conclusion, our data demonstrate that smoking is associated with elevated percentages
of Fas expressing CD4+ and CD19+ PBL in healthy individuals. Probably, as the Fas
pathway did not seem to differ between smokers and non-smokers, activated CD19+ PBL
are eliminated at a higher rate. The increase in Fas expression on these lymphocyte
subsets therefore might have consequences for the immune response and might contribute
to the decreased (humoral) responsiveness of smoking individuals.
References
1. Savage SM, Donaldson LA, Cherian S et al. Effects of cigarette smoke on the immune
response. II. Chronic exposure to cigarette smoke inhibits surface immunoglobulin-
mediated responses in B cells. Toxicol Appl Pharmacol 1991;111:523-9.
2. Kalra R, Singh SP, Savage SM, Finch GL, Sopori ML. Effects of cigarette smoke on
immune response: chronic exposure to cigarette smoke impairs antigen-mediated signaling
in T cells and depletes IP3-sensitive Ca(2+) stores. J Pharmacol Exp Ther 2000;293:166-71.
3. Holt PG. Immune and inflammatory function in cigarette smokers. Thorax 1987;42:241-9.
4. Tollerud DJ, Clark JW, Brown LM et al. The effects of cigarette smoking on T cell subsets.
A population-based survey of healthy caucasians. Am Rev Respir Dis 1989;139:1446-51.
58
5. Hockertz S, Emmendorffer A, Scherer G et al. Acute effects of smoking and high

experimental exposure to environmental tobacco smoke (ETS) on the immune system. Cell
Biol Toxicol 1994;10:177-90.
6. Walczak H, Krammer PH. The CD95 (APO-1/Fas) and the TRAIL (APO-2L) apoptosis
systems. Exp Cell Res 2000;256:58-66.
7. Scott DW, Grdina T, Shi Y. T cells commit suicide, but B cells are murdered. J Immunol
1996;156:2352-6.
8. Akagi T, Yoshino T, Kondo E. The Fas antigen and Fas-mediated apoptosis in B-cell
differentiation. Leuk Lymphoma 1998;28:483-9.
9. Schattner EJ, Elkon KB, Yoo DH et al. CD40 ligation induces Apo-1/Fas expression on
human B lymphocytes and facilitates apoptosis through the Apo-1/Fas pathway. J Exp Med
1995;182:1557-65.
10. Herrmann M, Voll RE, Zoller OM, Hagenhofer M, Ponner BB, Kalden JR. Impaired
phagocytosis of apoptotic cell material by monocyte-derived macrophages from patients
with systemic lupus erythematosus. Arthritis Rheum 1998;41:1241-50.
11. Suzuki N, Wakisaka S, Takeba Y, Mihara S, Sakane T. Effects of cigarette smoking on
Fas/Fas ligand expression of human lymphocytes. Cell Immunol 1999;192:48-53.
12. Suda T, Takahashi T, Golstein P, Nagata S. Molecular cloning and expression of the Fas
ligand, a novel member of the tumor necrosis factor family. Cell 1993;75:1169-78.
59
Chapter VI
Anti-CD3-induced and anti-Fas-induced apoptosis

in systemic lupus erythematosus (SLE)
Marc Bijl1
Gerda Horst1
Pieter C. Limburg2
1
2
The Netherlands
Clin Exp Immunol, 2001;123:127-32

Chapter VI
Summary
Disturbances in apoptosis or in the clearance of apoptotic material might result in

increased presentation of autoantigens which could be relevant to the pathogenesis of
SLE. Data concerning defects in apoptosis in SLE are conflicting. To determine whether
intrinsic defects in apoptosis induction occur in SLE irrespective of disease activity we
examined anti-CD3 and anti-Fas induced apoptosis in vitro in SLE patients with inactive
disease. Isolated peripheral blood lymphocytes (PBL) from 13 SLE patients and 14
healthy controls were incubated with anti-CD3, and, subsequently, after up-regulation of
membrane Fas following anti-CD3 incubation, with anti-Fas. Expression of Fas and levels
of apoptosis as detected by annexin V and propidiumiodide staining were assessed by
flowcytometry before and after the respective incubations. Fas expression on freshly
isolated lymphocytes of SLE patients was increased whereas levels of circulating
apoptotic cells were comparable between patients and controls. Stimulation with anti-CD3
resulted in up-regulation of membrane Fas in patients and in controls. In vitro induction of
apoptosis by anti-CD3 as well as by anti-Fas occurred both in SLE patients and controls,
and was higher in SLE patients after incubation with both anti-CD3 as well as with anti-
Fas. We conclude that Fas expression and in vitro induction of apoptosis is increased in
SLE even in the absence of disease activity.
Introduction
SLE, the prototype of a systemic autoimmune disease, is associated with various

immunologic abnormalities such as increased number of activated B lymphocytes and
production of multiple autoantibodies. Autoantibody production has been shown to be
selective, T cell dependent, and antigen-driven. Even in inactive disease peripheral blood
lymphocytes (PBL) show signs of activation [1]. Persistent autoantigen presentation may
be responsible for this continuous state of lymphocyte activation. Presentation of cryptic
antigens and modification, including cleavage and phosphorylation, can occur during
apoptosis resulting in autoantibody production [2,3]. Therefore, it can be speculated that
increased apoptosis of peripheral blood mononuclear cells (PBMC) contributes to
persistent lymphocyte activation and autoantibody production in SLE. Indeed, in vitro as
well as in vivo, the level of apoptosis of PBMC of lupus patients has been shown to be
elevated [4-7]. It has been suggested that this is due to increased in vivo activation of
PBMC in SLE resulting in increased activation induced cell death [5]. Whether the level
of apoptosis in SLE patients is in balance with the increased lymphocyte activation is,
however, not clear as many other factors than cell activation do influence apoptosis. For
example, apoptosis inhibiting factors such as soluble Fas and antibodies against Fas ligand
have been shown to be elevated in SLE patients [8-10]. On the other hand, apoptosis can
be facilitated and initiated by the use of immunosuppressive medication [11]. Studies
62
Apoptosis induction in SLE
dealing with apoptosis in SLE have shown conflicting results. In certain strains of mice it
has been demonstrated that mutations in the Fas (CD95) receptor or its ligand results in
defective Fas mediated apoptosis and induce autoimmune phenomena [12,13]. Although a
lymphoproliferative syndrome with the production of autoantibodies resulting from a
genetic defect in the Fas system has been described in a few patients [14-16], up till now a
FasL mutation has been described in SLE patients only incidentally. Common defects do
not seem to occur[17]. Abnormalities in Fas-induced apoptosis have not been
demonstrated [18]. Next to Fas signalling, induction of apoptosis can occur by signalling
via the CD3 receptor. Kovacs et al. demonstrated defective induction of apoptosis by anti-
CD3 in short term T (CD8+) cell lines of SLE patients [19]. The conflicting results
concerning apoptosis induction in SLE between study populations can be accounted for by
differences in disease activity, use of medication, or other exogenous factors influencing
apoptosis, such as smoking habits. Therefore, to avoid the presence of these confounding
factors, we tested whether anti-CD3 induced or anti-Fas-induced apoptosis are altered in
inactive, non-smoking lupus patients using no or minor amounts of corticosteroids only.

Patients eligible for this study fulfilled at least 4 American College of Rheumatology
(ACR) criteria for SLE [20], were non-smokers, and had inactive disease. Inactive disease
was defined as the persistent absence of clinical disease activity with a maximum
SLEDAI score [21] of 4 for at least a 4 month period prior to blood sampling. Patients
were allowed to use a constant dose of maximum 5 mg prednisolone a day. Patients using
other immunomodulating drugs were excluded. Tabacco smoking was an exclusion
criterium for this study as it augments Fas expression and thereby, probably, influences
the susceptibility for apoptosis induction (manuscript submitted). The SLEDAI score was
calculated for each patient. Fourteen healthy non-smoking volunteers, 3 males and 11
females, median age 27 years (range 22-39), were included as controls. For measurement
of complement C3 and C4, and detection of anti-dsDNA antibody levels EDTA blood was
drawn. Heparinized blood (30 ml) was drawn for cell isolation and culture.
Measurement of complement, C-reactive protein, and detection of antibodies to

double-stranded DNA
Complement C3 and C4 were measured by nephelometry (normal values 0.64-1.20 g/l and
0.11-0.40 g/l, respectively). Serum levels of C-reactive protein, normal value <3 mg/l,
were measured as previously described [22]. Anti-dsDNA antibodies were detected by
Farr-assay using 125I-labeled recombinant ds-DNA (Diagnostic Products Corporation, Los
Angeles, USA) which is free of contamination with ssDNA. Farr assay was performed
according to the manufacturer's instruction and positive samples were measured at
63
Chapter VI
different dilutions to obtain measurements within the assay range. Results of this assay
were expressed in IU/ml using Wo/80 as the ultimate standard [23]. Normal value of this
Farr-assay in our laboratory is <10 IU/ml; intra- and interassay variations are both < 10 %.
Study design
Cells were stained to measure Fas expression and the level of apoptosis immediately after
isolation (t0). To obtain optimal Fas expression, cells were cultured for different time
intervals under different circumstances: cells were incubated with or without 5 µg/ml anti-
CD3 to evaluate stimulation via the CD3-receptor. IL-2 (100 IU/ml) was added to
evaluate whether IL-2 deprivation could influence apoptosis results. Optimal up-
regulation of Fas was achieved after 48 hours incubation with anti-CD3 (data not shown).
Furthermore, the addition of IL-2 did not alter the level of apoptotic lymphocytes reached,
indicating that in our culture system sufficient IL-2 was present (data not shown).
Therefore, in all further stimulation experiments cells were cultured for 48 hours with
anti-CD3 only. After 48 hours incubation with anti-CD3 (t1), Fas expression and level of
apoptosis was measured . Next, anti-Fas or medium was added and after a further 24 hrs
(t2) and 48 hrs (t3) Fas expression and levels of apoptosis were detected again.
Cell isolation and culture

Within 60 minutes after blood drawing peripheral blood mononuclear cells (PBMC) were
isolated by lymphoprep density gradient centrifugation under sterile conditions. Cells
were resuspended in medium (RPMI 1640 + 2 mM glutamin + 60 µg/ml gentamycin + 5
% human poolserum) at a concentration of 0.5 x 106/ml with 5 µg/ml anti-CD3 (mAb
WT32). All cultures were performed in a culture flask at 370C in a humidified atmosphere
containing 5% CO2. After 48 hours incubation with anti-CD3 (t1) cells were washed twice
with RPMI 1640 and resuspended in medium with and without the addition of 5 µg/ml
anti-Fas (CLB-CD95/15), kindly provided by dr. L. Aarden, CLB, Laboratory of
Experimental and Clinical Immunology, Amsterdam, The Netherlands). Subsequently,
cells were cultured for another 24 hours (t2) and 48 hours (t3). At all time points indicated
part of the cells was used for apoptosis assessment as described below.
Staining procedures, FACS analysis and apoptosis assays

For staining we used the following conjugated monoclonal antibodies (MoAbs): anti-CD4
(allophycocyanine, APC) and anti-Fas/CD95 (phycoerythrin, PE), (Becton Dickinson,
Mountain View, CA). Apoptosis was detected using fluorescein isothiocyanate (FITC)
labeled annexin V, (Nexins, Hoeven, The Netherlands) and propidiumiodide (Molecular
Probes, Leiden, The Netherlands). Apoptosis was defined as positive annexin V staining
of the subpopulation of lymphocytes analyzed [24]. Annexin V binds to
phosphatidylserine which is exposed at the outer surface of the cell membrane already at
early stages of apoptosis and remains so during the subsequent process of apoptosis. By
defining apoptotic cells as those cells staining with annexin V, irrespective of
64
propidiumiodide staining, we were able to detect early as well as late apoptotic cells. After
washing with phosphate buffered saline (PBS) supplemented with 1% bovine serum
albumin (BSA), staining of 106 cells was done by adding labelled MoAb with subsequent
incubation for 45 minutes in the dark on ice. Samples were washed with PBS/BSA. After
the addition of 222.5 µl binding buffer (10 mM hepes, pH=7.4/140 mM NaCl/2.5mM
CaCl2) 25 µl propidiumiodide (10 µg/ml) and 2.5 µl annexin V diluted 1:50 were added
[24]. Immediately after incubation for 10 minutes in the dark on ice, three-color
immunofluorescence analysis was performed on a Coulter Epics-Elite equipped with a
gated amplifier (Coulter Electronics, Mijdrecht, The Netherlands). Cells were gated for
lymphocyte characteristics using both forward and sideward scatter. All studies were
performed with gates set for the total number of lymphocytes and for CD4+ lymphocytes
separately. As control for the level of apoptosis DNA content of cells was measured by
the method of Nicoletti [25]. Apoptosis was defined as the percentage of cells with a
fractional DNA content less than that of intact G1 cells (subdiploid peak). To exclude
cellular debris, fragments with the 20% lowest DNA content were not included in the
results [26]. In brief, 106 cells were washed (PBS/BSA) and resuspended in 0.3 ml
hypotonic propidiumiodide solution (50 µg/ml) with 0.1 % sodium citrate (Janssen,
Tilburg, The Netherlands) plus 0.1 % triton X-100 (Sigma, St Louis, USA) and 0.1 µg/ml
RNAse-A (Sigma, St Louis, USA). Analysis of DNA content was performed by
flowcytometry.
Statistical analysis
Differences between groups were evaluated with the two-tailed Mann-Whitney U-test.
Paired samples were analysed by Wilcoxon matched pairs signed ranks test. P< 0.05 was
considered statistically significant.
Results
Thirteen SLE patients, 11 female and two male, were included. Their median age was 33
years (range 22-76) with a disease duration of median 9 years (range 0-21). Cumulative
ACR criteria of the patients are shown in table 1. Two patients were on a low maintenance
dose of prednisolone (one was on 5 mg/day, the other on an alternating scheme of 2.5/5.0
mg/day). All patients were clinically quiescent: 9 patients had a SLEDAI-score of 0; 3
patients had a SLEDAI-score of 2, which in one patient was due to slightly increased
levels of antibodies against dsDNA (13 IU/ml) and in the others to decreased C4 levels
(0.06 and 0.08 g/l, respectively). The remaining patient had a SLEDAI score of 4 due to a
decreased level of complement C3 (0.58 g/l) and increased levels of antibodies against
dsDNA (19 IU/ml). Levels of C3 and C4 were 0.78 g/l (0.58-1.04) and 0.19 g/l (0.06-
0.40), respectively, for the total group of patients. None of the patients showed signs of
infection. All patients had a C-reactive protein level < 3 mg/l.
65
Chapter VI
Spontaneous and in vitro expression of Fas

Percentages of freshly isolated CD4+ lymphocytes positively staining for Fas did not
differ between healthy controls and SLE patients (figure 1a). Mean fluorescence intensity
(MFI) of Fas positive cells was higher in SLE patients (p=0.002, figure 1b). Incubation
with anti-CD3 resulted in an increase of Fas positive CD4+ lymphocytes after 48 hours (t1,
figure 1a). In controls the percentage of Fas-expressing CD4+ lymphocytes after CD3
stimulation was higher compared to SLE patients (figure 1a). Also, MFI of Fas-positive
cells in SLE patients and controls was significantly higher after incubation with anti-CD3
than at baseline (figure 1b). After 48 hours incubation with anti-CD3 (t1), cells were
Table 1:
Characteristics of 13 patients with systemic lupus erythematosus
Cumulative ACR criteria n (%)
1. malar rash 5 (38)

2. discoid rash 4 (31)
3. photosensitivity 9 (69)
4. oral ulcers 2 (15)
5. arthritis 6 (46)
6. serositis 2 (15)
7. renal involvement 5 (38)
8. cerebral involvement 2 (15)
9. hematologic abnormalities 7 (54)
10. immunologic abnormalities 9 (69)
11. ANA 13 (100)
washed, suspended in medium with or without anti-Fas, and cultured for another 24 (t2) or
48 (t3) hours. Culturing for another 24 hours without anti-Fas did not change Fas
expression in SLE patients nor in controls (figure 1 a,b, t2 versus t1). In contrast, the
addition of anti-Fas resulted in a decrease in the percentage of Fas positive cells and in the
MFI of Fas positive cells; this decrease was similar in patients and controls (figure 1 a,b,
t2a versus t2). Analysis of the total lymphocyte population for Fas expression showed
results comparable to that of the CD4+ population for all experiments mentioned before.
Apoptosis
Apoptosis of CD4+ lymphocytes was detected using annexin V staining. Percentages of
apoptotic lymphocytes were detected by annexin V staining as well as by measuring the
subdiploid peak using propidiumiodide. Neither the percentages of apoptotic CD4+
66
lymphocytes nor that of total lymphocytes differed between SLE patients and controls
immediately after lymphocyte isolation (t0, figure 2, 3a).
After activation with anti-CD3 (t1) levels of apoptotic T lymphocytes as well as levels of
apoptotic CD4+ T cells significantly increased both in SLE patients and in controls. After
anti-CD3 stimulation the percentage CD4+ apoptotic lymphocytes tended to increase in
SLE patients (p=0.09 versus controls, figure 2) reaching significance for the total
lymphocyte populations (p=0.03, figure 3a). Similar results were obtained after further
incubation in medium (t2) or with anti-Fas antibodies (t2a), showing increased apoptosis
of total lymphocytes in SLE patients compared to controls in both assays for apoptosis
detection used (figure 3a,b). The increase in the percentage of apoptotic cells after
(a) (b)
***
*** #
***
100 *** 750
Percent Fas + CD4+ lymphocytes
MFI Fas of Fas + CD4+

75 ***
**
lymphocytes
500 ***
***
50
250
25
##
0 0
timepoint t0 t1 t2 t2a timepoint t0 t1 t2 t2a
hours 0 48 72 72 hours 0 48 72 72
α CD3 - + + + α CD3 - + + +
α Fas - - - + α Fas - - - +
Figure 1: Percentages of Fas (CD95) positive lymphocytes (a) and mean fluorescence intensity
(MFI) of these Fas positive lymphocytes (b) from controls (open bars) and SLE patients (black
bars) at: t0, immediately after lymphocyte isolation; t1, after 48 hours incubation with anti-CD3;
t2, after another 24 hours incubation in medium or, t2a, following incubation with anti-Fas
antibodies. Results are expressed as mean " SD. #p<0.05, ##p<0.01, patients compared to
controls, **p<0.01,***p<0.001 comparing different time points of evaluation.
incubation with anti-Fas was significantly higher compared to the control experiment in
which cells were incubated in medium only. As expected, increase in annexin V staining
occurred earlier in time compared to increase in the propidiumiodide staining. After
incubation with anti-Fas a peak in annexin V binding cells could be observed after 24
hours (t2), whereas propidiumiodide staining peaked after 48 hours (t3, figure 3 a,b).
67
Chapter VI
Discussion
In this study we demonstrate that the percentage of apoptotic lymphocytes in peripheral

blood in clinically quiescent SLE patients is unaltered compared to healthy controls
although the MFI of Fas-positive lymphocytes is increased. Furthermore, the increase in
the percentage of apoptotic lymphocytes after in vitro apoptosis induction with anti-CD3
and anti-Fas demonstrates that there are no defects in either anti-CD3 or anti-Fas induced
apoptosis in SLE patients. In contrast, we found increased levels of apoptotic lymphocytes
in SLE patients after stimulation with anti-CD3 as well as after 24 h incubation with anti-
Fas compared with controls. Thus, we show increased levels of in vitro apoptosis in SLE
in a patient population with inactive disease and without use of immunosuppressive
medication, factors which might have influenced results from previous studies [4-7].
Although the number of patients studied was limited, patients included in this study were
a representative selection of our total population of SLE patients as cumulative ACR
criteria in this group were similar to those of our whole cohort [27].
40 ***
Percent apoptotic CD4+
***
lymphocytes (Ann V+ )
30 **
*
***
20
***
10
0
timepoint t0 t1 t2 t2a
hours 0 48 72 72
α CD3 - + + +
a Fas - - - +
Figure 2: Percentages of apoptotic CD4+lymphocytes from controls (open bars) and SLE patients
(black bars) at: t0, immediately after lymphocyte isolation; t1, after 48 hours incubation with
anti-CD3; t2, after another 24 hours incubation in medium or, t2a, in the presence of anti-Fas
antibodies. Apoptosis was assessed by positive staining with annexin V. Results are expressed as
mean " SD. *p<0.05, **p<0.01, ***p<0.001 comparing different time points of evaluation.
Alterations in Fas and FasL expression and function resulting in defective apoptosis as
found in murine autoimmune mouse strains [12,13] have only incidentally been detected
in human SLE [17]. Nevertheless, changes in the expression of apoptosis-related
molecules and levels of apoptosis have been demonstrated in SLE. Fas expression has
been reported to be elevated in SLE patients [5,18]. As Fas is up-regulated upon in vitro
cell activation [18,28,29], we carefully excluded such confounding factors in the present
68
study. In our patients, we isolated PBL within one hour after blood sampling . We found
the percentage of Fas positive, CD4+ lymphocytes and total lymphocytes to be similar
compared with controls. MFI of Fas on freshly isolated lymphocytes from patients was,
however, increased compared with controls, indicating a higher membrane expression for
Fas on positively staining cells in SLE than in controls. This is probably due to the
increased state of lymphocyte activation which can be found in vivo, even in SLE patients
with inactive disease [1]. Upon stimulation with anti-CD3, Fas expression increased in
controls and patients, in concordance with other studies [18]. The percentage Fas-
expressing lymphocytes after anti-CD3 stimulation was somewhat lower in SLE patients
which probably reflects a loss of in vivo preactivated lymphocytes. After incubation with
anti-Fas antibodies Fas expression diminished, most likely due to the binding of these
antibodies to their targets on the cell surface thereby inducing apoptosis. As apoptosis is a
continuous process we choose to detect apoptotic cells by two different methods so
minimising the influence of different detection techniques on the results.
(a) (b)
40 ***
40 ***
Percent apoptotic lymphocytes
***
Percent apoptotic lymphocytes
***
30 30
(Ann V +)
*
** #
(PI)
# ***
20 20 #
*
10 ** 10 ** **
** #
#
0 0
timepoint t0 t1 t2 t2a timepoint t0 t1 t2 t2a t3 t3a
hours 0 48 72 72 hours 0 48 72 72 96 96
α CD3 - + + + α CD3 - + + + + +
a Fas - - - + a Fas - - - + - +
Figure 3: (a) Percentages of apoptotic cells of the total lymphocyte population from controls
(open bars) and SLE patients (black bars) at: t0, immediately after lymphocyte isolation; t1, after
48 hours incubation with anti-CD3; t2, after another 24 hours incubation in medium or, t2a, in
the presence of anti-Fas antibodies. Apoptosis was assessed by positive staining with annexin V.
Results are expressed as mean " SD. #p<0.05, patients compared to controls; *p<0.05, **p<0.01,
***
p<0.001 comparing different time points of evaluation. (b) Percentages of apoptotic cells of the
total lymphocyte population from controls and SLE patients at the same time points. The
percentage apoptotic cells was defined as the percentage lymphocytes of the total lymphocyte
population containing decreased levels of DNA (subdiploid peak). Results are extended with
measurement after 48 hours incubation in medium or in the presence of anti-Fas antibodies (t3
and t3a respectively).
69
Chapter VI
By measuring the DNA content with propidiumiodide late apoptotic cells were detected
and by measuring annexin V binding early as well as late phases of apoptosis could be
visualised. In freshly isolated peripheral blood cells we analysed the percentage of
apoptotic CD4+ and total lymphocytes. Neither method detected differences in the
percentage of apoptotic lymphocytes between patients and controls, despite an increased
MFI of Fas on isolated lymphocytes in SLE patients. Comparing MFI of Fas staining on
freshly isolated lymphocytes from inactive SLE patients with that on lymphocytes after
stimulation with anti-CD3, the former MFI proved rather low, suggesting a minor in vivo
state of activation only that did not result in significant apoptosis. However, as we show in
our induction experiments, the higher initial Fas expression in SLE might explain the
higher percentages of apoptotic lymphocytes we found upon further activation with anti-
CD3 in patients compared with controls. A contributing factor which leads to increased
levels of apoptotic cells might have been a decreased clearance which has been described
in SLE patients [30].
In agreement with Lorenz et al. we did not find increased levels of apoptosis of freshly
isolated lymphocytes in SLE patients [5]. Others demonstrated increased levels of
apoptotic peripheral blood mononuclear cells in SLE [4,6,7]. This controversy might be
explained by differences in methodology. Lymphocytes of SLE patients may be
preactivated in vivo. Therefore, any delay after blood sampling in detection of apoptosis
will result in increased levels of apoptotic cells. We isolated lymphocytes and detected
apoptosis within 1 h after blood sampling. Additional factors might explain the
discrepancy of our results with those of others. As our study population had inactive
disease and no or minor use of corticosteroids only, the results of our study are most likely
representative for the in vivo situation in patients with quiescent SLE. In other studies
patients were included irrespective of disease activity [4-7]. Although in these studies
only a weak [4] or no [5-7] correlation between the level of apoptosis and disease activity
was found, no conclusions can be drawn as data may have been influenced by the use of
medication as well. Also differences in cell isolation and cell culture conditions can
influence apoptosis [5]. For example, suboptimal concentrations of growth factors such as
IL-2, can induce apoptosis. As addition of IL-2 did not influence levels of apoptosis, we
feel that conditions were optimal in our study. In vitro activation-induced apoptosis in
SLE patients did not seem to be disturbed. After stimulation with anti-CD3 apoptosis
increased in controls and even more so in SLE patients. We, therefore, could not confirm
defective CD3-mediated cell death of lymphocytes from SLE patients as reported by
others [19]. In the latter study, based on experiments with T cell lines consisting
predominantly of CD8+ cells, it was found that activated T cells from patients with SLE
were relatively resistant to anti-CD3 induced apoptosis. We analysed the CD4+ and total
T lymphocyte population. Neither in the whole T cell population nor in the CD4+
subpopulation defects in anti-CD3 induced apoptosis could be found. A selective defect in
activation induced cell death of CD8+ lymphocytes can, however, not be excluded but
seems unlikely.
70
Finally, we showed that anti-Fas induced apoptosis in SLE patients is functionally intact
which confirms earlier studies [18,28]. We extend these data by showing that both
activation-induced Fas expression and anti-Fas-induced apoptosis is normal in SLE
patients with inactive disease.
In conclusion this study shows that in quiescent SLE patients, activation-induced and Fas-
induced apoptosis are functionally intact. As in SLE patients Fas expression is increased
on freshly isolated PBL, the slightly higher levels of apoptotic lymphocytes as seen after
stimulation with anti-CD3 or anti-Fas might well represent minor changes due to
preactivation of PBL of SLE patients in vivo.
NSN C95.1482).
References
2. Utz PJ, Anderson P. Posttranslational protein modifications, apoptosis, and the bypass of
tolerance to autoantigens. Arthritis Rheum 1998;41:1152-60.
5. Lorenz HM, Grunke M, Hieronymus T et al. In vitro apoptosis and expression of apoptosis-
related molecules in lymphocytes from patients with systemic lupus erythematosus and
other autoimmune diseases. Arthritis Rheum 1997;40:306-17.
1998;7:113-8.
Lupus 1999;8:508-13.
71
Chapter VI
11. Thompson CB. Apoptosis in the pathogenesis and treatment of disease. Science
1995;267:1456-62.
12. Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA, Nagata S.
Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates
apoptosis. Nature 1992;356:314-7.
13. Takahashi T, Tanaka M, Brannan CI et al. Generalized lymphoproliferative disease in mice,
caused by a point mutation in the Fas ligand. Cell 1994;76:969-76.
14. Rieux-Laucat F, Le Deist F, Hivroz C et al. Mutations in Fas associated with human
15. Fisher GH, Rosenberg FJ, Straus SE et al. Dominant interfering Fas gene mutations impair
16. Drappa J, Vaishnaw AK, Sullivan KE, Chu JL, Elkon KB. Fas gene mutations in the
Canale-Smith syndrome, an inherited lymphoproliferative disorder associated with
autoimmunity. N Engl J Med 1996;335:1643-9.
17. Wu J, Wilson J, He J et al. Fas ligand mutation in a patient with systemic lupus
erythematosus and lymphoproliferative disease. J Clin Invest 1996;98:1107-13.
19. Kovacs B, Vassilopoulos D, Vogelgesang SA, Tsokos GC. Defective CD3-mediated cell
death in activated T cells from patients with systemic lupus erythematosus: role of
decreased intracellular TNF-alpha. Clin Immunol Immunopathol 1996;81:293-302.
20. Tan EM, Cohen AS, Fries JF et al. The 1982 revised criteria for the classification of
22. Spronk PE, ter Borg EJ, Kallenberg CG. Patients with systemic lupus erythematosus and
Jaccoud's arthropathy: a clinical subset with an increased C reactive protein response? Ann
Rheum Dis 1992;51:358-61.
23. Feltkamp TE, Kirkwood TB, Maini RN, Aarden LA. The first international standard for
24. Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C. A novel assay for apoptosis.
Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using
fluorescein labelled Annexin V. J Immunol Methods 1995;184:39-51.
25. Nicoletti I, Migliorati G, Pagliacci MC, Grignani F, Riccardi C. A rapid and simple method
for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J
Immunol Methods 1991;139:271-9.
26. Darzynkiewicz Z, Juan G, Li X et al. Cytometry in cell necrobiology: analysis of apoptosis
and accidental cell death (necrosis). Cytometry 1997;27:1-20.
27. Spronk PE, Bootsma H, Horst G et al. Antineutrophil cytoplasmic antibodies in systemic
lupus erythematosus. Br J Rheumatol 1996;35:625-31.
72
28. Miyawaki T, Uehara T, Nibu R et al. Differential expression of apoptosis-related Fas

antigen on lymphocyte subpopulations in human peripheral blood. J Immunol
1992;149:3753-8.
Rheum 1998;41:1241-50.
73
Chapter VII
Expression of costimulatory molecules on peripheral

blood lymphocytes of patients with
Systemic Lupus Erythematosus
Marc Bijl1
Gerda Horst1
Pieter C. Limburg2
Cees G.M. Kallenberg1.
1
2`
The Netherlands
Ann Rheum Dis, in press

Chapter VII
Summary
In systemic lupus erythematosus (SLE) autoantibody production is T cell dependent. For a

proper T and B cell interaction signalling of costimulatory molecules on these cells is
necessary. The expression of costimulatory molecules on peripheral blood lymphocytes in
patients with SLE in conjunction with disease activity was measured to evaluate whether
expression of costimulatory molecules in SLE is increased. Thirteen patients with SLE
with active disease, 10 patients with inactive disease, and 14 controls entered the study. In
addition, samples from 10 of the 13 patients with active disease could be studied at a
moment of inactive disease as well. Isolated peripheral blood lymphocytes were stained
for the lymphocyte subset markers CD4, CD8, CD19, their respective activation markers
CD25, HLA-DR, CD38, and the costimulatory molecules CD40L, CD28, CD40, CD80
and CD86. Expression was measured by flow cytometry. Peripheral blood lymphocytes of
patients with SLE showed signs of increased activation at the moment of active disease.
Almost all CD4+ T cells expressed CD28, both in patients and in controls. CD80
expression on CD19+ B cells was low in both groups and did not correlate with disease
activity. In contrast, the percentage CD19+ B cells expressing CD86 was increased in
patients with SLE even in patients with inactive disease (p=0.04) and correlated with the
SLEDAI-score (p=0.0005) and levels of anti-dsDNA (p=0.006). No changes in CD40 or
CD40L expression were found in the patients with SLE. In patients with SLE the
expression of CD86 on CD19+ B cells is increased and is associated with disease activity,
B cell activation, and levels of anti-dsDNA. The increased CD86 expression will render
(autoreactive) B cells more susceptible for T cells. This can facilitate autoantibody
production and might be a target for immunosuppressive treatments.
Introduction
Autoimmune diseases such as systemic lupus erythematosus (SLE) are associated with
various immunologic abnormalities such as increased numbers of activated B cells and
autoantibody production. Much of our understanding of the pathogenetic mechanisms
involved are derived from animal studies. In lupus prone MRL-lpr/lpr mice the selective
B cell proliferation seen is antigen-driven and T cell dependent [1], and treatment of those
mice with anti-Thy-1.2 antibodies abrogated IgG anti-dsDNA production [2]. Activation
of T cells requires interaction of the T cell receptor (TCR) with the major
histocompability complex (MHC) molecule complexed with the antigen in the presence of
additional signalling through costimulatory molecules. Of these, the CD28-CD80/CD86
and CD40-CD40L pathways are well known. The relevance of costimulation in the
development of autoimmune disease in animal models is supported by several studies in
which costimulation antagonists have been used as an effective approach to treat these
diseases [3,4].
76
Costimulatory molecules in SLE
The tight and carefully orchestrated interaction between T and B cells ensures that
bystander lymphocytes are not activated. Whenever, for example, a B cell is signalled via
CD40 and fails to receive a proper signal through its B cell receptor the cell will die by
apoptosis [5]. Adequate stimulation, however, will result in lymphocyte activation and
proliferation by a sequence of activating and regulating signals. It can be suggested that an
increase in the expression of costimulatory molecules facilitates T and B cell activation
and survival, increases autoantibody production, and so may influence the development of
autoimmune disease. In line with this hypothesis we measured the expression of
costimulatory molecules on peripheral blood lymphocytes in lupus patients and healthy
controls in relation to the state of lymphocyte activation to consider the following
questions: (a) Is there an increase in expression of costimulatory molecules in patients
compared with controls? (b) Are these changes related to the state of lymphocyte
activation and levels of antibodies to dsDNA? (c) Is the expression of these molecules in
lupus patients dependent on disease activity?

Outpatients eligible for this study fulfilled at least four American College of
Rheumatology (ACR) criteria for SLE [6], were non-smokers, and had either active or
inactive disease. Patients with active disease had to fulfil predefined criteria [7]. Inactive
disease was defined as the persistent absence of clinical disease activity for at least four
months while patients were without, or on a constant dose of, immunomodulating drugs.
To evaluate changes due to disease activity, re-evaluation of patients with initially active
disease took place at a moment of inactive disease as well. The SLEDAI score was
calculated for each patient [8]. Healthy volunteers, matched for age and sex, were
included as controls.
Blood samples for anti-dsDNA detection were drawn in EDTA and anti-dsDNA were
detected by the Farr-assay using 125I-labelled recombinant dsDNA (Diagnostic Products
Corporation, Los Angeles, USA) as described [7].
Cell isolation and staining procedures

Peripheral blood mononuclear cells were isolated from heparinised blood by lymphoprep
density gradient centrifugation. For surface staining the following phycoerythrin (PE) or
fluorescein isothiocyanate (FITC) and allophycocyanine (APC) conjugated monoclonal
antibodies were used: CD4 (APC), CD19 (APC), CD25 (FITC/PE), CD38 (FITC/PE),
CD40 (PE), CD40L (PE), CD80 (PE), CD86 (PE), and HLA-DR (FITC/PE) (Becton
Dickinson, Mountain View, CA); CD8 (APC) (Pharmingen, Hamburg, Germany). After
washing with phosphate buffered saline (PBS) supplemented with 1% bovine serum
albumin (BSA), 106 cells were stained by adding labelled monoclonal antibodies in the
77
Chapter VII
relevant combinations, with subsequent incubation for 45 minutes on ice in the dark.
Samples were washed with PBS/BSA. Three colour immunofluorescence analysis was
performed on a Coulter Epics-Elite equipped with a gated amplifier (Coulter Electronics,
Mijdrecht, The Netherlands). Cells were gated for lymphocyte characteristics using both
forward and sideward scatter. For final analysis only those samples were included in
which more than 500 cells of the respective subsets were counted.
Statistics
Differences between groups were evaluated using the student’s t test or Mann-Whitney U
test when appropriate. Paired samples were analysed separately using the Wilcoxon rank
sum test. Spearman’s rank correlation was applied for detecting correlations between
different study parameters. A p value less than 0.05 was considered statistically
significant.
Results
Ten SLE patients (two male, eight female) with inactive SLE entered the study. Their
mean (SD) age was 48.5 (1.8) years. Of thirteen different patients (one male, twelve
female) aged 39.2 (16.7) years, blood samples could be drawn at the moment of active
disease. In addition, from ten of these thirteen patients with active disease samples could
be analysed at a moment of inactive disease as well. This resulted in a total of twenty
patients evaluated at the moment of inactive disease.
Table 1
Expression of activation markers and costimulatory molecules on
peripheral blood lymphocytes. Results are given as mean (SD)
Control Inactive Active

+
CD4/CD25 (%) 4.0 (1.4) 4.4 (3.2) 8.5 (6.3)*
+
CD8/HLA-DR (%) 5.6 (2.5) 8.1 (5.3) 10.6 (8.3)
CD19/CD38+ (%) 79.7 (8.7) 84.3 (7.7) 86.5 (9.8)*
CD4/CD28+ (%) 95.3 (3.8) 93.6 (7.9) 95.0 (3.5)

+
CD19/CD40 (%) 85.0 (9.3) 81.3 (11.2) 81.9 (12.4)
+
CD19/CD80 (%) 1.7 (0.9) 3.1 (5.7) 4.4 (5.8)
+ *
CD19/CD86 (%) 2.9 (1.4) 6.9 (7.2) 12.5 (8.9)***
*
p<0.05, ***p<0.001 compared with controls. Percentages are expressed as proportion of total
CD4+, CD8+, and CD19+ subpopulations, from healthy controls (n=10), patients with inactive
disease (n=20), and patients with active disease (n=13).
78
Lymphocytes of SLE patients showed signs of increased activation at the moment of

active disease. The percentage of CD4+ lymphocytes expressing CD25 was higher
(p=0.03) in patients with active disease than in controls (table 1). Furthermore, the
percentage of activated (CD38+) B lymphocytes was raised in patients with active disease
(p=0.04 v controls). Expression of costimulatory molecules was measured on CD4+ T
cells and on CD19+ B cells. Almost all CD4+ T cells expressed CD28 on their surface,
which was comparable between controls and patients with SLE (table 1). CD40L was
hardly detectable on CD4+ T cells of controls and patients (data not shown). Regarding B
cells, the percentage of cells expressing CD40 was comparable between controls and
patients. The percentage of CD19+ B cells expressing CD80 was low in controls as well as
%CD86+ of CD19+ lymphocytes
40 r=0.57, p=0.0005
30
20
10
0
0 10 20 30
SLEDAI
Figure 1: Correlation between the percentage of CD86 expressing B (CD19+) cells and disease
activity as measured by the SLEDAI score analysed for all samples taken (n=33).
patients. A correlation between CD80 expression and the disease activity of patients with
SLE was not found. In contrast, CD86 expression differed between the groups
investigated. Compared to controls the percentage of CD19+ B cells expressing CD86 was
increased in patients with SLE with inactive disease (p=0.04) and was increased even
more in patients with active disease (p=0.0003 v controls). Furthermore, CD86 expression
correlated with the SLEDAI-score (figure 1) and levels of antibodies to dsDNA (r=0.47,
p=0.006).To analyse the relation between CD86 expression and cell activation we
concentrated on activated-that is, CD38+ B cells. Indeed, the percentage of B cells positive
for both markers, though low, was substantially increased among patients with active
disease (p=0.006 v controls).
For all costimulatory molecules paired samples from active and inactive disease were
analysed separately. This analysis revealed similar results, showing increased expression
of CD86 but not of CD80 or the other costimulatory molecules measured (figure 2).
79
Chapter VII
Discussion
In this study we show that CD86 expression on CD19+ B cells from patients with SLE is
increased in patients with inactive disease and increases further in conjunction with
disease activity and B cell activation. Such a relation could not be shown for CD80
expression. No changes were found in the expression of the CD80/CD86 counterreceptor
CD28 on CD4+ T cells. Furthermore, expression of CD40 on CD19+ B cells and CD40L
on CD4+ T cells was comparable between patients and healthy controls.
40
p=0.0004
% CD86+ of CD19+
p=0.02
30
lymphocytes
20
10
0
SLE SLE
Figure 2: Changes in CD86 expression in paired observations (n=10) of patients measured on

the moment of active and inactive disease, respectively. Results for CD86 expression on B cells of
controls (n=13) are also shown.
In SLE, dsDNA antibody production is T cell dependent. Studies in lupus prone mice
indicate that costimulatory molecules play an essential part in the interaction between B
and T cells. Extrapolating these data to lupus patients, it can be speculated that aberrant
expression of costimulatory molecules provides a condition in which autoantibody
production is facilitated, possibly because autoreactive memory B cells and T cells can
expand in the absence of adequate apoptotic signalling [9].
CD86 interacts with CD28 on the T cell. This interaction regulates the activation induced
apoptosis in thymocytes. In vivo, stimulation of CD28 prevents apoptosis of thymocytes
in mice[10]. Therefore, increased expression of CD86 might promote lymphocyte survival
intervening with the elimination of autoreactive lymphocytes. Furthermore, it may render
B cells more susceptible for T cell help. These changes may facilitate autoantibody
production. Indeed, in murine lupus the production of anti-dsDNA seems to be dependent
on CD86 costimulation. Injection of CTLA-4 immunoglobulin in NZB/W F1 mice, which
neutralises both CD80 and CD86, blocked the production of dsDNA antibodies and
80
prolonged life [3]. In MRL-lpr/lpr mice treatment with anti-CD86 alone inhibited anti-
dsDNA expression while treatment with anti-CD80 did not change anti-dsDNA
expression[4]. Furthermore, it was shown that the presence of costimulatory molecules
influenced the course of the disease. CD86 deficient MRL-lpr/lpr mice had significantly
milder glomerulonephritis than wild-type mice[4].
Aberrant CD86 expression on B lymphocytes has been reported in SLE. Folzenlogen et al.
analysed CD80 and CD86 expression on B cells of 13 patients with inactive SLE and
found CD86 increased in patients compared with controls[11]. We confirm these data and
show that during disease activity CD86 is further up regulated. In addition, our data
suggest a discrepancy in the expression of CD80 and CD86. Costimulatory signals of
CD80 and CD86 lead to differential polarisation of T helper cell responses. There is
evidence that CD80 preferentially acts as a costimulus for the generation of Th1 cells
while CD86 costimulates and induces Th2 cells [12]. Our data, showing a difference in
the expression of CD80 and CD86, are in line with the concept of SLE being a
predominantly Th2 mediated disease.
In T cells, ligation of CD28 stimulates membrane expression of another accessory
molecule, CD40 ligand (CD40L) [10]. CD40L is necessary for T cell help for B cells as
shown by defective antigen-specific T cell responses in CD40L deficient mice [13]. For
example, CD40L influences apoptosis when ligation with its counterreceptor CD40 takes
place. Ligation of CD40 by CD40L is able to block B cell apoptosis induced by antigen
receptor cross linking [14]. In line with the CD28-CD80/86 pathway, aberrant expression
of CD40 or CD40L may be a relevant factor in the development of autoimmune disease.
Indeed, in murine lupus, treatment with anti-CD40L reduced the level of antibodies to
dsDNA, and delayed the development of nephritis [9]. Furthermore, a dysregulation of
CD40L expression in human lupus has been described [15]. We did not find changes in
CD40L expression on CD4+ T lymphocytes in patients with SLE. CD40L expression was
low in patients with inactive disease and controls and, unlike the results of others, did not
increase during active disease. In experiments in which we stimulated isolated peripheral
blood mononuclear cells with anti-CD3, we could detect proper increase of CD40L
expression in conjunction with T cell activation, excluding the possibility that our CD40L
monoclonal antibody was inappropriate (data not shown). Also, CD40 expression on B
cells, being high in lupus patients, irrespective of disease activity, did not differ from that
in healthy controls.
In conclusion, our data clearly show that in patients with SLE, even at the moment of
inactive disease, the expression of CD86 on CD19+ B cells is increased, and is associated
with B cell activation and levels of antibodies to dsDNA. Whether aberrant expression of
CD86 in patients with SLE is constitutive or merely results from increased lymphocyte
activation needs further investigation. Nevertheless, our data provide increasing evidence
that manipulating costimulatory pathways in lupus may be a potentially beneficial
therapeutic strategy.
81
Chapter VII
NSN C95.1482).
References
1. Peng SL, Madaio MP, Hughes DP, Crispe IN, Owen MJ, Wen L, et al. Murine lupus in the
absence of alpha beta T cells. J Immunol 1996;156:4041-9.
2. Seaman WE, Wofsy D, Greenspan JS, Ledbetter JA. Treatment of autoimmune MRL/Ipr
mice with monoclonal antibody to Thy-1.2: a single injection has sustained effects on
lymphoproliferation and renal disease. J Immunol 1983;130:1713-8.
3. Finck BK, Linsley PS, Wofsy D. Treatment of murine lupus with CTLA4Ig. Science
1994;265:1225-7.
4. Liang B, Gee RJ, Kashgarian MJ, Sharpe AH, Mamula MJ. B7 costimulation in the
development of lupus: autoimmunity arises either in the absence of B7.1/B7.2 or in the
presence of anti-b7.1/B7.2 blocking antibodies. J Immunol 1999;163:2322-9.
5. Schattner EJ, Elkon KB, Yoo DH, Tumang J, Krammer PH, Crow MK, et al. CD40 ligation
induces Apo-1/Fas expression on human B lymphocytes and facilitates apoptosis through
the Apo-1/Fas pathway. J Exp Med 1995;182:1557-65.
6. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, et al. The 1982 revised
criteria for the classification of systemic lupus erythematosus. Arthritis Rheum
1982;25:1271-7.
1990;33:634-43.
9. Mohan C, Shi Y, Laman JD, Datta SK. Interaction between CD40 and its ligand gp39 in the
development of murine lupus nephritis. J Immunol 1995;154:1470-80.
10. Shi Y, Radvanyi LG, Sharma A, Shaw P, Green DR, Miller RG, et al. CD28-mediated
signaling in vivo prevents activation-induced apoptosis in the thymus and alters peripheral
lymphocyte homeostasis. J Immunol 1995;155:1829-37.
11. Folzenlogen D, Hofer MF, Leung DY, Freed JH, Newell MK. Analysis of CD80 and CD86
expression on peripheral blood B lymphocytes reveals increased expression of CD86 in
lupus patients. Clin Immunol Immunopathol 1997;83:199-204.
12. Kuchroo VK, Das MP, Brown JA, Ranger AM, Zamvil SS, Sobel RA, et al. B7-1 and B7-2
costimulatory molecules activate differentially the Th1/Th2 developmental pathways:
application to autoimmune disease therapy. Cell 1995;80:707-18.
82
13. Grewal IS, Xu J, Flavell RA. Impairment of antigen-specific T-cell priming in mice lacking
CD40 ligand. Nature 1995;378:617-20.
14. Tsubata T, Wu J, Honjo T. B-cell apoptosis induced by antigen receptor crosslinking is
blocked by a T-cell signal through CD40. Nature 1993;364:645-8.
15. Desai-Mehta A, Lu L, Ramsey-Goldman R, Datta SK. Hyperexpression of CD40 ligand by
B and T cells in human lupus and its role in pathogenic autoantibody production. J Clin
Invest 1996;97:2063-73
.
83
Chapter VIII
IgG subclass distribution of autoantibodies differs

between renal and extra-renal relapses in patients
with systemic lupus erythematosus
Marc Bijl1
Hilde M. Dijstelbloem1
Wia W. Oost1
Hendrika Bootsma2
Ronald H.W.M. Derksen3
Jan Aten4
Pieter C. Limburg1,2
Cees G.M. Kallenberg1
1
Department of Clinical Immunology, University Hospital, Groningen
2
Department of Rheumatology, University Hospital, Groningen
3
Department of Clinical Immunology and Rheumatology, University Hospital, Utrecht
4
Department of Pathology, Academical Medical Center, Amsterdam, The Netherlands.
Submitted
Chapter VIII
Summary
IgG subclasses of autoantibodies differ in their potential to induce an inflammatory

response as they differentially interact with complement and Fcγ-receptors. IgG subclass
distribution of anti-nucleohistone and anti-dsDNA antibodies were analysed
longitudinally in SLE patients prior to and at the moment of an extra-renal (n=23) or a
renal relapse (n=17). Kidney biopsy specimens of patients with a renal relapse were
analysed for IgG subclass deposition. IgG1 anti-nucleohistone and IgG1 anti-dsDNA
antibodies were present in plasma of 39 out of 40 patients. At the moment of a relapse,
IgG2- and IgG3 anti-nucleohistone and IgG2 anti-dsDNA antibodies were more
frequently present in patients with renal disease compared to those with extra-renal
disease. Increase in levels of IgG1 anti-dsDNA were observed in 10 out of 11 patients
prior to a renal relapse but occurred in only 10 out of 22 patients with an extra-renal
relapse (p=0.02). Rises in IgG2 anti-dsDNA occurred at an equally low rate both prior to
renal and extra-renal relapses. A rise in IgG2 anti-nucleohistone antibodies preceded a
renal relapse in 8 of 11 patients and an extra-renal relapse in only 4 out of 22 patients
(p=0.006). In kidney biopsies all IgG subclasses could be detected. IgG1- and IgG2-
subclass antibodies to nucleohistone and to dsDNA are the predominant subclasses found
in plasma of lupus patients with renal disease. The frequent occurrence of a rise of IgG2
anti-nucleohistone and IgG1 anti-dsDNA in patients prior to a renal relapse, suggest that,
besides IgG1 subclass autoantibodies, IgG2-subclass antibodies to nucleohistone have a
particular pathophysiological role in lupus nephritis.
Introduction
Systemic lupus erythematosus (SLE) is a multi-system autoimmune disease associated

with a multitude of autoantibodies. The kidneys are frequently involved, presumably due
to autoantibody deposition at the glomerular basement membrane (GBM) which results in
complement activation and attraction and influx of inflammatory cells [1]. Deposition of
antibodies at the GBM can occur via trapping of circulating immune complexes within the
glomerulus, via nucleosomes that bind to the GBM and act as a substrate for autoantibody
binding, or through cross-reactivity of antinuclear antibodies with glomerular structures
[1]. Antibodies most closely associated with lupus nephritis are anti-dsDNA antibodies
[1]. In addition, there is cumulating evidence for an important role of nucleosomes and
anti-nucleosome antibodies in the pathogenesis of SLE. Nucleosome specific antibodies
emerge before anti-dsDNA antibodies in lupus-prone mice. Both antibodies to dsDNA as
well as to nucleosomes can be eluted from kidney specimens of lupus mice with overt
glomerulonephritis [2]. IgG class antibodies seem most relevant as there is a close relation
between levels of IgG anti-dsDNA and histologic activity scores in patients with lupus
86
IgG subclasses in SLE
nephritis [3]. Furthermore, in the majority of patients, a renal relapse is preceded by a

significant rise of IgG anti-dsDNA as detected by ELISA [4].
All IgG subclasses can be found in kidney biopsies [5]. Of the different subclasses, IgG1
and IgG3 activate complement more efficiently than IgG2 while IgG4 does not activate
complement at all [6]. The IgG subclass distribution of autoantibodies could therefore be
of relevance in the pathogenesis of lupus nephritis. To evaluate their possible
nephritogenic role, we monitored levels of total IgG and IgG subclasses of antibodies to
nucleohistone and to dsDNA in time in patients who suffered a renal relapse. IgG subclass
distribution in kidney biopsies of patients with a renal relapse was assessed as well.
Serological data obtained from patients with renal relapses were compared with those of
lupus patients who developed an extra-renal relapse, in order to evaluate the specificity of
the findings for lupus nephritis.
Patients fulfilling at least four revised American College of Rheumatology (ACR) criteria
for the diagnosis of SLE [7] could participate in this study. Patients originated from the
cohorts of SLE patients (n=156) who participated in a prospective study at the University
Hospitals of Groningen (AZG) and Utrecht (AZU), the Netherlands. Patients belonging to
this cohort are seen at the outpatient department at least every 3-4 months for clinical
evaluation. At every visit disease activity is scored and the SLE Disease Activity Index
(SLEDAI) is calculated [8]. Attention is paid to the occurrence of infections. Of all
patients who developed a relapse of the disease in the period between September 1991 till
September 1997 (AZG) and September 1991 till January 1995 (AZU) data from their first
relapse were included. Criteria for a relapse were pre-defined (table 1). Patients were
classified into those with a renal relapse, defined as a biopsy proven WHO class III, IV or
Vd lupus nephritis according to the World Health Organization criteria [9], with or
without extra-renal symptoms, and those with an extra-renal relapse of the disease. Blood
samples were drawn monthly in EDTA (Vacutainer; Becton Dickinson, Mountain View,
CA) during the study period. Plasma was stored at -800C until needed.
Measurement of anti-dsDNA and anti-nucleohistone antibodies

Anti-dsDNA antibodies were detected by Farr-assay using 125I-labeled recombinant ds-
DNA (Diagnostic Products Corporation, Los Angeles, USA) which is free of con-
tamination with ssDNA. Farr assay was performed according to the manufacturer's in-
struction and positive samples were measured at different dilutions to obtain
measurements within the range of the assay. Results of this assay were expressed in IU/ml
using Wo/80 as the ultimate standard [10]. Normal value of this Farr-assay in our
laboratory is < 10 IU/ml; intra- and interassay variations are both less than 10 %.
Antibodies to dsDNA were also detected by ELISA as described [11]. For detection of
87
Chapter VIII
IgG subclasses, microtiter plates (Nunc-Immuno plate Maxisorb, Nunc, Denmark) were
precoated with 100 µl/well protamine sulfate (500 µg/ml in millipore water) at 40C for 45
minutes. Plates were washed with millipore water. Coating was performed with DNA (10
µg/ml DNA, Sigma, St Louis, USA, in 10 mM Tris, pH 8.0, 0.15 M NaCl) overnight at
40C.
Table 1:

affected, and/or
(<50x109 /l)

* Only features occurring within 2 weeks of the outclinic visit or admission under consideration
88
After washing with washing buffer (0.01 M Tris, pH=8.0, 0.15 M NaCl, 0.05% Tween 20)
plasma samples (diluted 1:60 in phosphate buffered saline (PBS, pH 8.0) with 0.05%
Tween 20, 0,2% BSA) were added and plates were incubated for 1 hour at 370C and
subsequently during 2 hours at 40C. Plates were washed and mouse anti-human
monoclonal antibodies in dilution buffer were added (anti-IgG 1:6000, clone HP6017,
anti-IgG1 1:1000, clone 8c/6-39, anti-IgG2 1:6000, clone HP6014, anti-IgG3 1:5000,
clone HP6050, anti-IgG4 1:6000, clone HP6025; Sigma, St Louis, USA). After incubation
for 1 hour at room temperature plates were washed and alkaline-phosphatase labeled
sheep anti-mouse IgG (Sigma; 1:3000 dilution buffer) was added. After final incubation
for 1 hour at room temperature plates were washed and para-nitrophenyl-phosphate
(Sigma nr104-40T, 1 tablet suspended in 40 ml 10% (w/v) diethanolamin, buffer pH 9.8)
was added for development. After 30 minutes incubation in the dark at room temperature
the reaction was stopped with NaOH.
Total IgG and IgG subclasses of anti-nucleohistone antibodies were also detected by
ELISA. Microtiter plates (Nunc Maxisorb) were coated with 10 µg/ml nucleohistone
(Sigma; in 0.15 M NaCl, 0.15 M trisodiumcitrate, pH 7.0) for 1 hour. This nucleohistone
preparation contains only the core histone subcomponents and H1 [12]. After washing
plasma was added (diluted 1:60 in PBS, 0.05% Tween 20, 0.2% BSA) for incubation at
room temperature during 1 hour. Further development of the ELISA was similar to the
anti-DNA ELISA as described.
For both assays normal values were below the detection threshold. Values were expressed
as arbitrary units. To prevent interassay variation serial samples of individual patients as
well as all exacerbation samples were analyzed in one assay. Intra-assay variation was less
than 10%. A rise in antibody level of at least 25% within a period of maximal 4 months
was arbitrarily considered significant.
Biopsies
Renal biopsy specimens from lupus patients with a renal relapse obtained in the study
period were used. From 10 patients with a renal relapse tissue specimens were available.
Cryosections (4 µm) were defrosted and fixed in 100 % acetone. After preincubation with
10% v/v normal goat serum (DAKO, Denmark) in PBS, sections were incubated for 3
hours at room temperature with the same monoclonal mouse anti-human antibodies as
used in the ELISA. Anti-human total IgG (1:3000), anti-human IgG1 (1:600), anti-human
IgG2 (1:600), anti-human IgG3 (1:250) and anti-human IgG4 (1:1000), were diluted in
PBS. Specimens of lupus nephritis class IV were used as positive control. Renal biopsy
specimens from patients with minimal change nephropathy and acute transplant rejection
served as negative control. Possible endogenous peroxidase activity was blocked by
incubation for 15 minutes with 0.1% NaN3 and 0.3% H2O2 in PBS. After washing sections
were incubated with HRP-conjugated goat-anti-mouse IgG1 (Southern Biotechnology
Associates, Birmingham, AL, USA) or HRP-conjugated goat-anti-mouse IgG2a (SBA).
All goat-anti-mouse monoclonal antibodies were used at a 1:100 dilution in PBS enriched
89
Chapter VIII
with 10% normal human serum (CLB, Amsterdam) to block crossreactivity with human
IgG in the biopsies. HRP activity was detected by the addition of 3-amino-ethyl-
carbazole. Finally, specimens were counterstained with hematoxyline. Staining was
scored by two independent observers on a semiquantative scale of 0 (no staining), 1+
(weak or scarce, but unmistakenly present) till 6+ (diffuse and global, strong intensity).
Discrepancies between observers were resolved by scoring specimens together, thereby
reaching consensus.
Statistics
GraphPad Instat (GraphPad Software, San Diego, CA) was used for statistical calcula-
tions. Differences in parameters between groups were evaluated with unpaired t-test when
normal distribution could be assumed, otherwise the Mann-Whitney U test was applied.
Fisher’s exact test was used for comparison of differences in prevalence. Spearman's test
was applied for detecting correlations between different study parameters. A p value
<0.05 was considered significant.
Results
In the study period, 40 relapses were encountered. Characteristics of the patients and the
relapses are shown in tables 2 and 3. Seventeen relapses were classified as renal, all with
biopsy proven proliferative lupus nephritis (WHO class III, IV or Vd).
In a minority of patients the renal relapse was accompanied by other manifestations of the
disease, among which hematological abnormalities were most frequently seen (table 3).
The relapses without renal involvement were classified as extra-renal. Seven patients were
included at the moment of relapse (6 renal, 1 extra-renal); therefore antibody levels of
these patients could not be monitored prior to the relapse.
Plasma anti-nucleohistone antibodies

Total IgG as well as IgG1-subclass antibodies to nucleohistone could be detected in all but
one of the patients (table 2). Antibodies of the IgG2 subclass were more frequently
present in patients with a renal relapse than in those with extra-renal relapses (p=0.01).
Also, IgG3 anti-nucleohistone antibodies could be detected at a higher frequency in
patients with a renal relapse (p=0.009).
A weak correlation was found between levels of total IgG anti-nucleohistone antibodies
and anti-dsDNA as measured by Farr assay (r=0.44, p=0.02). Furthermore, a correlation
was found between levels of total IgG anti-nucleohistone antibodies and anti-
nucleohistone antibody levels of the IgG1 (r=0.87, p<0.001) and the IgG2 subclass
(r=0.69, p<0.001).
90
The presence of the various anti-nucleohistone IgG subclasses during specific disease
manifestations is shown in table 4. No specific association with particular disease
manifestations could be demonstrated, although IgG2 anti-nucleohistone antibodies were
present in most patients with nephritis, serositis as well as hematological abnormalities.
Serial plasma samples were available in 11 patients with a renal and in 22 patients with an
extra-renal relapse. A rise in IgG2 anti-nucleohistone antibodies occurred more frequently
in patients with a renal relapse than in those with an extra-renal relapse (p=0.006). With
respect to the other subclasses, no differences could be found (table 5).
Table 2:
Clinical and serological characteristics of 40 SLE patients at the moment of relapse.
Clinical manifestation Renal relapse Extra-renal relapse

Number of patients 17 23
Male/Female 3/14 1/22
SLEDAI score 15.5 (6-25)*** 9.6 (3-16)
Farr assay (IU/l) 339 (10-1701) 155 (1-726)
Anti-nucleohistone antibodies: IgG1 (+/-) 17/0 22/1

#
presence (+) or absence (-) IgG2 (+/-) 13/4 8/15
##
of IgG subclasses IgG3 (+/-) 5/12 0/23
IgG4 (+/-) 1/16 0/23
Anti-dsDNA antibodies: IgG1 (+/-) 17/0 22/1
#
presence (+) or absence (-) IgG2 (+/-) 13/4 10/13
of IgG subclasses IgG3 (+/-) 7/10 4/19
IgG4 (+/-) 0/17 0/23
Data are expressed as mean (range); *** p<0.001, unpaired t-test; # p<0.05, ## p<0.01 Fisher’s
exact test.(+) denotes presence, (-) denotes absence of resp. subclass at the moment of relapse.
Plasma anti-dsDNA antibodies

In patients with renal relapses, levels of antibodies to dsDNA as measured by Farr assay
tended to be higher compared to those in patients with an extra-renal relapse (p=0.08,
table 2). Levels of IgG-class antibodies to dsDNA as measured by ELISA showed a
similar trend (p=0.07) and correlated with the results of the Farr assay (r=0.63, p<0.0001).
All but one of the patients had antibodies to dsDNA of the IgG1 subclass. The patient
without IgG1 antibodies had an extra-renal relapse. Antibodies to dsDNA of the IgG2
subclass were found in increased frequency in patients with a renal relapse compared to
those with an extra-renal-relapse (13 of 17 versus 9 of 23, respectively, p=0.02). In less
than half of the patients, anti-dsDNA antibodies of the IgG3 subclass were present (table
2), showing a similar trend of a higher frequency in patients with renal disease (p=0.07).
None of the patients had IgG4 anti-dsDNA antibodies.
91
Chapter VIII
Table 3:
Clinical characteristics of 40 relapses
Clinical manifestation Renal relapses Extra-renal relapses

(n = 17) (n = 23)
Renal 17 (100) 0 (0)
Serositis 0 (0) 6 (26)
Musculoskeletal 2 (12) 11 (48)
Cerebral 2 (12) 4 (17)
Haematological 5 (29) 5 (22)
Skin 1 (6) 6 (26)
Vasculitis 1 (6) 5 (22)
Fever 1 (6) 6 (26)
Number and percentage ( ) of patients in subgroup
Levels of total IgG anti-dsDNA correlated with that of the IgG1 subclass (r=0.88,
p<0.0001) as well as with that of IgG2 anti-dsDNA (r=0.86, p<0.0001). Longitudinal
studies revealed that 9 out of 11 (82%) of the patients with a renal relapse and 13 out of
22 (67%) with an extra-renal relapse had a significant rise of antibodies to dsDNA by Farr
assay preceding the clinical relapse by a median period of 3 months (range 1-5).
Table 4:
Clinical manifestations of relapses and presence of IgG-subclasses of anti-nucleohistone
antibodies
Clinical manifestation IgG-subclass of anti-nucleohistone antibodies

IgG1 IgG2 IgG3 IgG4
Renal (n=17) 17 (100) 13 (76) 5 (29) 3 (18)
Serositis (n=6) 6 (100) 5 (83) 0 (0) 0 (0)
Musculoskeletal (n=13) 13 (100) 7 (54) 2 (15) 1 (11)
Cerebral (n=6) 13 (100) 2 (33) 1 (17) 0 (0)
Haematological (n=10) 10 (100) 8 (80) 1 (10) 1 (10)
Skin (n=7) 6 (86) 3 (43) 1 (14) 0 (0)
Vasculitis (n=6) 6 (100) 4 (67) 1 (17) 0 (0)
Fever (n=7) 7 (100) 4 (57) 1 (14) 0 (0)
Number of relapses and percentage ( ) of manifestation with a particular IgG subclass being
positive for anti-nucleohistone antibodies.
Table 5 shows the presence of significant rises of total IgG and IgG subclasses of anti-
dsDNA as measured by ELISA in relation to relapses. The majority (91%) of renal
relapses were preceded by a significant rise in antibodies to dsDNA of the IgG1 subclass.
92
In contrast, only 45% of patients with an extra-renal relapse had a significant rise of IgG1
anti-dsDNA antibodies preceding their relapse (p=0.02). Significant rises of the other IgG
subclasses of anti-dsDNA before disease exacerbation occurred in a minority of patients
only.
Table 5:
Significant rises of anti-nucleohistone and anti-dsDNA antibodies occurring
in a period of 6 months prior to a relapse of the disease
Renal relapse Extra-renal relapse

Significant rise of antibodies prior to a relapse
Yes No Yes No
Anti-nucleohistone
total IgG 9 (82)# 2 (18) 9 (41) 13 (59)
IgG1 8 (73) 3 (27) 14 (64) 8 (36)
##
IgG2 8 (73) 3 (27) 4 (18) 18 (72)
IgG3 2 (18) 9 (72) 3 (14) 19 (86)
Anti-dsDNA
total IgG 9 (82) 2 (18) 10 (45) 12 (55)
#
IgG1 10 (91) 1 (9) 10 (45) 12 (55)
IgG2 4 (36) 7 (64) 6 (27) 16 (73)
IgG3 3 (27) 8 (73) 2 (10) 20 (90)
Number and percentage ( ) of patients in subgroup; # p<0.05, ## p<0.01, Fisher’s exact test
Renal deposition of total IgG and IgG subclasses

In all biopsies immunoglobulins of each IgG subclass could be detected (figure 1).
Notably, in contrast to subclass analysis of plasma samples, even IgG3 and IgG4 were
detected in every kidney specimen. No correlation was found between the relative amount
of any of the IgG subclasses in the kidney biopsy and plasma levels of IgG antibody
subclasses to anti-dsDNA or anti-nucleohistone (data not shown).
Discussion
In this study we cross-sectionally as well as longitudinally analysed plasma IgG subclass

levels of antibodies to nucleohistone and to dsDNA as well as the IgG subclass
distribution in kidney biopsies from SLE patients. We performed this study in order to
obtain more insight in the differential role of these autoantibodies and their IgG subclasses
in the pathogenesis of SLE. Antibodies to nucleohistone as well as to dsDNA were found
in plasma of 39 of 40 patients. These antibodies were mainly of the IgG1 subclass, and
93
Chapter VIII
less frequently of the IgG2 subclass. Antibodies of the IgG3 subclass directed to
nucleohistone or to dsDNA could be detected in less than half of the patients.
5
Arbitrary fluoresence intensity
0
IgG1 IgG2 IgG3 IgG4
Figure 1: Semi-quantitative analysis of IgG subclass deposition in kidney biopsy specimens from
patients with proliferative lupus nephritis. Staining was scored by two independent observers on a
semi-quantitative scale from 0 (no staining), 1+ (weak, but unmistakenly present) till 6+ (diffuse
and global, strong intensity). Horizontal lines denote the median.
Finally, antibodies of the IgG4 subclass directed to nucleohistone were present in only 1
out of 40 patients analysed, whereas those directed to dsDNA were not found at all.
We found that anti-nucleohistone antibodies can be detected in most of our lupus patients.
Interestingly, IgG2 and IgG3 anti-nucleohistone antibodies were present more often in
lupus patients with a renal relapse compared to patients with an extra-renal manifestation
of the disease only. Furthermore, a significant rise in IgG2 anti-nucleohistone preceded
78% of the renal relapses, in contrast to the extra-renal relapses in which only 18% were
preceded by a significant rise of IgG2 anti-nucleohistone antibodies. These data support
the hypothesis that antibodies to nucleohistone are involved in the pathogenesis of lupus
nephritis. In addition, our results suggest that anti-nucleohistone antibodies of the IgG2
subclass play a particular role. The predominance of IgG2 in autoantibody subclass
distribution has been described for other autoantibodies in lupus patients. Indeed,
antibodies of the IgG2 subclass to C1q are overrepresented in lupus nephritis patients [13
14]. Furthermore, we observed that antibodies to dsDNA of the IgG2 subclass were
overrepresented in nephritis patients as well. It should be stated that our nucleohistone
ELISA not only measures antibodies specific for the conformational structure of
nucleohistones but, probably, also antibodies to the individual components of
nucleohistones, that is DNA and histones. As such, it is conceivable that the subclass
94
distribution of anti-dsDNA antibodies was comparable with that of anti-nucleohistone

antibodies, that is, IgG1 being present in the majority of patients, both with renal and
extra-renal relapses, and a skewed distribution to IgG2 in patients with a renal relapse. In
cross-sectional studies, others also reported a restriction in subclass distribution of
antibodies to dsDNA. In accordance with the present study IgG1 has been demonstrated to
be the anti-dsDNA antibody subclass most frequently encountered [15-19]. In addition, in
support of a pathophysiological role for the IgG1 subclass, we observed that a significant
rise of IgG1 anti-dsDNA occurred in 91% of the patients prior to a renal relapse. In
contrast to our results, next to IgG1, the IgG3 isotype of anti-dsDNA as well as of anti-
nucleosome has been described to be present most frequently in patients with renal
disease [18-20]. These discrepancies might, at least in part, be explained by differences in
the affinity of the monoclonal antibodies to IgG-subclasses used in the various studies
[21].
The longitudinal data on IgG-subclass distribution of anti-nucleohistone and anti-dsDNA
antibodies as observed in this study are unique and are therefore difficult to compare with
previous studies. Several others reported stable isotype profiles of anti-dsDNA as well as
anti-nucleohistone antibodies independent of antibody fluctuations as measured by the
Farr assay and independent of disease activity [17,22]. In contrast, Devey et al. found an
increase of the IgG1- and IgG3-subclass, but not of the IgG2 anti-dsDNA antibody
subclass with increased disease severity in patients with renal disease [19]. Since, in all of
the aforementioned studies, time intervals between blood sampling were much longer than
the monthly period that we consistently have used, the possibility cannot be excluded that
rises in levels of IgG subclasses prior to a relapse have been missed in these studies. We
found a subclass-specific rise in levels of anti-nucleohistone and anti-dsDNA occurring
over a period ranging from 1 to 5 months preceding a relapse, which is comparable with a
previous study from our group in which total levels of anti-dsDNA were evaluated [11].
Also in other studies, all IgG subclasses could be detected in renal biopsies from lupus
patients [5 23 24]. Imai et al. analysed glomerular immunofluorescence intensity of IgG
subclasses in patients with membranoproliferative glomerulonephritis, membranous
nephropathy and lupus nephritis and demonstrated increased intraglomerular deposition of
IgG1 and IgG2 in lupus nephritis [5]. This might reflect the predominance of the IgG1 and
IgG2 subclasses of anti-nucleohistone and anti-dsDNA antibodies that we found in plasma
of patients with a renal relapse. Others, however, found IgG3 to be the dominant isotype
in kidney biopsies of patients with class IV lupus nephritis [24]. Interestingly, IgG4 could
be detected in all kidney specimens of our patients with a renal relapse, while, in plasma,
IgG4 anti-nucleohistone antibodies were found in one patient only. The lack of correlation
between levels of the measured autoantibodies in the serum and their deposition in renal
tissue suggest that the autoantibodies deposited in the kidney may include other antigenic
specificities than the ones measured in the sera. The deposition of antibodies might also
be dependent on other factors than their serum concentration such as their affinity to their
95
Chapter VIII
respective antigen, the degree of presence of the antigen along the basement membrane
and epitope specificity.
For many years, attention has been paid to IgG subclasses in SLE. Each IgG isotype has
different biological and functional properties. Subclass distribution might therefore
influence the course of SLE [6]. Recently, the role of Fc-gamma-receptors (FcγR) has
renewed the interest in IgG subclasses. First, the important role of FcγR in the mediation
of an inflammatory reaction by immunecomplexes has been shown in NZB/NZW mice.
These mice spontaneously develop glomerulonephritis. Deficiency of the γ chain of the
FcγR protected them from severe nephritis although immunecomplex deposition and
complement activation were unaltered [25]. Secondly, polymorphisms of FcγR can
influence the interaction with IgG subclasses, in particular with IgG2 [26]. This is highly
relevant for the second Fcγ-receptor (FcγRIIa) in which the presence of either arginin or
histidin at position 131 (FcγRIIa-R131 and FcγRII-H131, respectively) determines the
interaction with IgG2 and, to a lesser extent, IgG3. In contrast to FcγRIIa-H131, the
FcγRIIa-R131 isoform can not interact with IgG2 and has less affinity for IgG3. It has
been suggested that patients homozygous for FcγRIIa-R131 will be prone to the
development of lupus nephritis due to decreased clearance of complexed IgG2 and IgG3-
subclass antibodies. Indeed, in several studies the FcγRIIa-R/R131 genotype was sig-
nificantly more present in patients with proliferative lupus nephritis. However, there is
still controversy, as in most studies no skewing of FcγRIIa-R131 was observed in lupus
patients with compared to those without nephritis [27].
In conclusion, IgG1- and IgG2-subclass antibodies to nucleohistone and to dsDNA are the
predominant subclasses found in plasma of lupus patients with renal disease. The frequent
occurrence of a rise of IgG2 anti-nucleohistone and IgG1 anti-dsDNA in patients prior to
a renal relapse, suggest that, besides IgG1 subclass autoantibodies, IgG2-subclass
antibodies to nucleohistone have a particular pathophysiological role in lupus nephritis.
References
1. Lefkowith JB, Gilkeson GS. Nephritogenic autoantibodies in lupus: current concepts and
continuing controversies. Arthritis Rheum 1996;39:894-903.
2. Amoura Z, Chabre H, Koutouzov S, et al. Nucleosome-restricted antibodies are detected
before anti-dsDNA and/or antihistone antibodies in serum of MRL-Mp lpr/lpr and +/+ mice,
and are present in kidney eluates of lupus mice with proteinuria. Arthritis Rheum
1994;37:1684-8.
3. Okamura M, Kanayama Y, Amastu K, et al. Significance of enzyme linked immunosorbent
assay (ELISA) for antibodies to double stranded and single stranded DNA in patients with
lupus nephritis: correlation with severity of renal histology. Ann Rheum Dis 1993;52:14-20.
96
4. Bootsma H, Spronk PE, ter Borg EJ, et al. The predictive value of fluctuations in IgM and
IgG class anti-dsDNA antibodies for relapses in systemic lupus erythematosus. A
prospective long-term observation. Ann Rheum Dis 1997;56:661-6.
5. Imai H, Hamai K, Komatsuda A, Ohtani H, Miura AB. IgG subclasses in patients with
membranoproliferative glomerulonephritis, membranous nephropathy, and lupus nephritis.
Kidney Int 1997;51:270-6.
6. Maran R, Dueymes M, Le Corre R, et al. IgG subclasses of human autoantibodies. Ann Med
Interne (Paris) 1997;148:29-38.
7. Tan EM, Cohen AS, Fries JF, et al. The 1982 revised criteria for the classification of
9. Churg J, Bernstein J, Glassock RJ. Lupus nephritis. In: Churg J, Bernstein J, Glassock RJ,
editors. Renal disease: classification and atlas of glomerular diseases. 2nd ed. New York:
Igaku-Shoin; 1995. p. 151-80.
10. Feltkamp TE, Kirkwood TB, Maini RN, Aarden LA. The first international standard for
1990;33:634-43.
12. Suenaga R, Abdou NI. Anti-(DNA-histone) antibodies in active lupus nephritis. J
Rheumatol 1996;23:279-84.
13. Haseley LA, Wisnieski JJ, Denburg MR, et al. Antibodies to C1q in systemic lupus
erythematosus: characteristics and relation to Fc gamma RIIA alleles. Kidney Int
1997;52:1375-80.
14. Norsworthy P, Theodoridis E, Botto M, et al. Overrepresentation of the Fcgamma receptor
type IIA R131/R131 genotype in caucasoid systemic lupus erythematosus patients with
autoantibodies to C1q and glomerulonephritis. Arthritis Rheum 1999;42:1828-32.
15. Dueymes M, Barrier J, Besancenot JF, et al. Relationship of interleukin-4 to isotypic
distribution of anti-double-stranded DNA antibodies in systemic lupus erythematosus. Int
Arch Allergy Immunol 1993;101:408-15.
16. Blanco F, Kalsi J, Ravirajan CT, et al. IgG subclasses in systemic lupus erythematosus and
other autoimmune rheumatic diseases. Lupus 1992;1:391-9.
17. Winkler TH, Henschel TA, Kalies I, et al. Constant isotype pattern of anti-dsDNA
antibodies in patients with systemic lupus erythematosus. Clin Exp Immunol 1988;72:434-
9.
18. Rubin RL, Tang FL, Chan EK, et al. IgG subclasses of autoantibodies in systemic lupus
erythematosus, Sjogren's syndrome, and drug-induced autoimmunity. J Immunol
1986;137:2528-34.
97
Chapter VIII
19. Devey ME, Bleasdale K, Isenberg DA. Antibody affinity and IgG subclass of responses to
tetanus toxoid in patients with rheumatoid arthritis and systemic lupus erythematosus. Clin
Exp Immunol 1987;68:562-9.
20. Amoura Z, Koutouzov S, Chabre H, et al. Presence of antinucleosome autoantibodies in a
restricted set of connective tissue diseases: antinucleosome antibodies of the IgG3 subclass
are markers of renal pathogenicity in systemic lupus erythematosus. Arthritis Rheum
2000;43:76-84.
21. Seppala IJ, Routonen N, Sarnesto A, Mattila PA, Makela O. The percentages of six
immunoglobulin isotypes in human antibodies to tetanus toxoid: standardization of isotype-
specific second antibodies in solid-phase assay. Eur J Immunol 1984;14:868-75.
22. Fellows G, Gittoes N, Scott DG, et al. Individual variation in the isotype profile of anti-
histone autoantibodies in systemic lupus erythematosus. Clin Exp Immunol 1988;72:440-5.
23. Haas M. IgG subclass deposits in glomeruli of lupus and nonlupus membranous
nephropathies. Am J Kidney Dis 1994;23:358-64.
24. Iskandar SS, Falk RJ, Jennette JC. Clinical and pathologic features of fibrillary
glomerulonephritis. Kidney Int 1992;42:1401-7.
25. Clynes R, Dumitru C, Ravetch JV. Uncoupling of immune complex formation and kidney
damage in autoimmune glomerulonephritis. Science 1998;279:1052-4.
26. Rascu A, Repp R, Westerdaal NA, Kalden JR, van de Winkel JG. Clinical relevance of Fc
gamma receptor polymorphisms. Ann N Y Acad Sci 1997;815:282-95:282-95.
27. Tan SY. FcgammaRIIa polymorphism in systemic lupus erythematosus. Kidney Blood
Press Res 2000;23:138-42.
98
Chapter IX
Fcγγ receptor polymorphisms in systemic lupus

erythematosus: association with disease and in vivo
clearance of immune complexes
Hilde M. Dijstelbloem1
Marc Bijl1
Rob Fijnheer2
Ronald H.M. Scheepers2
Wia W. Oost1
Marc D. Jansen2
Wim J. Sluiter1
Pieter C. Limburg1
Ronald H.W.M. Derksen2
Jan G.J. van de Winkel2
Cees G.M. Kallenberg1
1
University Hospital, Groningen, The Netherlands
2
University Medical Center Utrecht, The Netherlands.

Chapter IX
Summary
Fc receptors for IgG (FcγR) play a prominent role in the clearance of immune complexes
in systemic lupus erythematosus (SLE). Polymorphisms of FcγR have been proposed as
genetic factors that influence susceptibility to SLE. We analysed three functional FcγR
polymorphisms in a strictly Caucasian population of SLE patients, and determined the
influence of these polymorphisms on the clearance of immune complexes in vivo.
Genomic DNA was isolated from 230 Caucasian patients with SLE and 154 controls.
Amplification of FcγR-genomic regions in allotype-specific polymerase chain reactions
was used to distinguish the genotypes. In addition, we analysed the FcγR genotypes of 13
patients with SLE who participated in a study determining the half-life of IgG-coated
erythrocytes in the blood.
We found a strong trend toward skewing of FcγRIIa, with an enrichment of the
homozygous FcγRIIa-R/R131 genotype in patients compared with controls. We did not
find a correlation between this genotype and the development of lupus nephritis.
However, we established that the half-life of IgG-coated erythrocytes in the blood was
prolonged in patients expressing the FcγRIIa-R/R131 genotype. The homozygous
FcγRIIIa-F/F158 genotype was found more frequently in patients with arthritis and/or
serositis. In Caucasian populations, the R/H polymorphism of FcγRIIa is a minor
determinant in susceptibility to SLE, whereas the V/F polymorphism of FcγRIIIa is
associated with a set of disease manifestations. Notably, the R/H polymorphism of
FcγRIIa affects the clearance of immune complexes in vivo, which may influence the
course of a disease such as SLE.
Introduction
Systemic lupus erythematosus (SLE) represents the prototype of an immune complex-

mediated autoimmune disease. Impaired handling and subsequent tissue deposition of
immune complexes are believed to play an important role in the pathogenesis of the
disease. Clearance of immune complexes mainly depends on the mononuclear phagocyte
system (MPS), in which erythrocytes transport complexes bound via complement receptor
1 to mononuclear phagocytes located in liver and spleen. Abnormalities in complement-
mediated clearance have been described in SLE, as well as abnormalities in receptors for
the Fc fragment of immunoglobulin G (FcγR) on mononuclear phagocytes [1].
Three distinct classes of FcγR are recognised, which vary in ligand-binding affinity and
specificity, as well as in cell distribution [2]. Liver and spleen phagocytes, important
constituents of the MPS, express members of all three classes of leukocyte FcγR: FcγRIa,
FcγRIIa, and FcγRIIIa. In contrast, neutrophils constitutively express FcγRIIa and
FcγRIIIb, and can be induced to express FcγRIa. Many studies have recently identified
polymorphisms of FcγR as genetic factors influencing susceptibility to SLE and other
102
FcγR polymorphisms in SLE
autoimmune diseases. FcγRIIa, FcγRIIIa as well as FcγRIIIb display functional

polymorphisms, including an arginine (R) or histidine (H) at amino acid position 131 for
FcγRIIa, a valine (V) or phenylalanine (F) at amino acid position 158 for FcγRIIIa, and a
4-amino acid variation termed neutrophil antigen (NA) polymorphism (NA1 and NA2) for
FcγRIIIb [3,4]. Indeed, these polymorphisms have a profound influence on human IgG
binding; homozygosity for R/R of FcγRIIa, F/F of FcγRIIIa, and NA2/NA2 of FcγRIIIb
lessens the ability to interact with specific IgG subclasses [4-9]. Thus, the effect of these
polymorphisms on the handling of immune complexes in vivo may have implications for
a disease like SLE, although this has not yet been investigated.
In association studies, evidence in support of a role for the FcγRIIa polymorphism in
susceptibility to SLE has been controversial [10-20], whereas evidence for the FcγRIIIa
polymorphism has been less contradictory [6,16,18,20]. No skewing of the FcγRIIIb
polymorphism in patients with SLE has been reported thus far [16]. The discrepancies
between studies are striking, in particular for the FcγRIIa-R/H131 polymorphism.
Inconsistencies may arise from variation in population sizes, as became apparent in
studies by Song et al [13] and Salmon et al [18]. In addition, variation in the clinical
criteria used, specifically with regard to the classification of nephritis, may affect the
results. A third influence may be ethnic differences between tested populations [21].
Although most studies are confined to populations of a single ethnic background, some
studies investigated populations of mixed ethnic backgrounds [6,14]. Differential
distribution of FcγR polymorphisms among various ethnic populations has been described
extensively, for FcγRIIa in particular [14-15,19,21].
In the present study, we determined the influence of all three functional FcγR
polymorphisms on the susceptibility to SLE in a strictly Caucasian population. We studied
230 Caucasian patients with SLE from 2 large medical centers, and 154 regionally
matched controls. We performed our study in a stable, mostly nonmigrating, population,
in contrast to several studies conducted previously. To evaluate associations between
FcγR polymorphisms and specific clinical parameters of the disease, the well-defined and
uniformly acknowledged SLE criteria of the American College of Rheumatology (ACR)
were used, as well as the World Health Organization (WHO) classification criteria for
nephritis based on the histologic findings on kidney biopsy. Furthermore, we determined
the influence of different FcγR genotypes on the clearance of immune complexes in 13
patients with SLE, who had participated in a previous study determining the half-life of
injected IgG-coated erythrocytes [22].
Study population
Power analysis revealed that 216 patients and 144 controls were required to demonstrate a
15% increase in frequency of the homozygous FcγRIIa-R/R131 genotype in SLE patients
103
Chapter IX
compared with controls (power of 85%, α = 0.05). Subsequently, 230 consecutive,

unrelated Caucasian patients with SLE (24 males, 206 females) who were followed up
prospectively at the University Hospitals of Groningen and Utrecht in The Netherlands,
were included in the study. A total of 154 healthy, unrelated Caucasian blood donors were
used for comparison. All patients fulfilled at least 4 of the ACR 1982 revised criteria for
the classification of SLE [23]. The median age of the patients was 37 years (range 21-78),
with a median disease duration of 9 years (range 1-40). Patient characteristics are shown
in table 1.
Table 1:
Clinical characteristics of patients with systemic lupus erythematosus (SLE) in the Fcγ receptor
(FcγR) polymorphism association study and the immune complex (IC) clearance
study
Characteristic FcγR association study IC clearance study

(n=230) (n=13)
Male / female 24 / 206 4/9

Age (years)
Median (range) 37 (21-78) 38 (24-59)
Disease duration (years)
Median (range) 9 (1-40)
ACR criteria (%)
Malar rash 111 (48) 8 (62)
Discoid rash 49 (21) 3 (23)
Photosensitivity 113 (49) 10 (77)
Oral ulcers 36 (16) 4 (31)
Arthritis 179 (78) 11 (85)
Serositis 76 (33) 2 (15)
Renal involvement 108 (47) 6 (46)
CNS involvement 20 (9) 2 (15)
Haematologic abnormalities 163 (71) 7 (54)
Immunologic abnormalities 209 (91) 12 (92)
Anti-nuclear antibodies 226 (98) 12 (92)
FcγR genotyping
FcγRIIa, FcγRIIIa and FcγRIIIb genotyping was performed on genomic DNA of patients
and controls using polymerase chain reaction (PCR)-based genotyping methods. The
FcγRIIa-131R/H genotyping assay was performed as described previously [17]. In brief, 2
FcγRIIa-specific primers were used to amplify a 1,000-bp fragment from the FcγRIIA
gene, containing the polymorphic site. Subsequently, the amplified fragment served as a
template to amplify a 278-bp fragment in an allele-specific primer reaction. The
genotyping methods for FcγRIIIa and FcγRIIIb consisted of allele-specific primer
104
reactions only, and were performed according to methods described earlier with minor
modifications [6,25].
In each experiment, sequence-verified genomic DNA of all 3 genotypes was included and
served as controls. Amplified products were analysed by gel electrophoresis (1.5%
agarose), stained with ethidium bromide, and visualised under ultraviolet light.
Clearance studies
Clearance studies were performed as described previously [22]. In short, erythrocytes
were coated with anti-rhesus antiserum (mainly IgG1 and IgG3, Central Laboratory of the
Netherlands Red Cross Blood Transfusion Service, Amsterdam, The Netherlands) and
labeled with 500 µCi 99mTc. Labeling efficiency was ~90%. After washing in saline, cells
were injected intravenously in 60 seconds. The injected dose was determined by
measuring the syringe before and after injection in a common dose calibrator. Blood
samples were taken at 0, 3, 8, 13, 18, and 23 min, and the half-life of IgG-coated
erythrocytes in the blood was calculated by regression analyses.
Statistical analyses
Phenotype and genotype frequencies in patients and controls were compared with 3 x 2
and 2 x 2 contingency tables (chi-square test with continuity correction or Fisher’s exact
test, as appropriate). The 95% confidence intervals (95% CI) of odds ratios (OR) were
calculated with the approximation of Woolf. Differences in age at diagnosis between
groups were tested by Mann-Whitney U test. Univariate associations between Fc R
genotypes and half-life of IgG-coated erythrocytes in the blood were compared with
Spearman’s rank correlation test. Two-sided P values less than 0.05 were considered
significant. Corrections for multiple comparisons were not made due to unknown
associations between clinical manifestations.
Results
Distribution of genotypes
In this study, 230 Caucasian patients with SLE (24 males and 206 females), as well as 154
Caucasian controls, were included for FcγR genotype analysis. The median age of the
patients was 37 years (range 21-78), with a median disease duration of 9 years (range 1-
40). The median age at diagnosis for women was slightly lower than the median age at
diagnosis for men (26 years (range 8-73) versus 32 years (range 14-68), p not significant
[NS] ).
FcγRIIa, FcγRIIIa, and FcγRIIIb genotypes were determined for all patients and controls,
using amplification of FcγR-genomic regions in allotype-specific PCRs. No significant
skewing of FcγRIIa, FcγRIIIa, or FcγRIIIb polymorphisms could be observed in this SLE
cohort (table 2).
105
Chapter IX
Table 2:
Distribution of FcγR genotypes in Caucasian patients with SLE and Caucasian controls (%)
FcγRIIa FcγRIIIa FcγRIIIb

R/R 131 R/H 131 H/H 131 V/V 158 V/F 158 F/F 158 NA1/1 NA1/2 NA2/2
Controls 32 (21) 80 (52) 42 (27) 15 (10) 73 (47) 66 (43) 27 (17) 66 (43) 61 (40)
(n=154)
SLE 68 (30) 108 (47) 54 (23) 30 (13) 108 (47) 92 (40) 42 (18) 101 (44) 87 (38)
(n=230)
Values are the number (%) of patients. Fcγ = Fcγ receptor; SLE = systemic lupus erythematosus
However, there was a strong trend toward skewing of the homozygous FcγRIIa-R/R131
genotype, with an enrichment of this genotype in patients with SLE compared with
healthy controls (68 of 230 versus 32 of 154, respectively; OR 1.60, 95% CI 0.99-2.59, p
= 0.071) (table 2). In addition, the median age at diagnosis for patients with the FcγRIIIa-
F/F158 genotype was slightly lower than that of patients with other genotypes (25 years
(range 8-70) versus 28 years (range 12-73), p = 0.082). No differences in sex distribution
were observed between the different FcγR genotypes.
Clinical presentation
Among the SLE patients, 108 patients (47.0%) developed symptoms of nephritis during
the course of the disease. The median age at diagnosis in the group of patients with
nephritis was slightly lower than in the group without nephritis (25 years (range 8-68)
versus 29 years (range 10-73), p = NS). No differences in sex distribution were observed
between these 2 groups. When the distribution of FcγR genotypes between patients with
nephritis and healthy controls was analysed, no significant skewing was observed for
FcγRIIa, FcγRIIIa, or FcγRIIIb (table 3). Remarkably, significant skewing toward the
FcγRIIIb-NA1 allele could be detected in patients with nephritis compared with patients
without nephritis (frequency of 0.46 versus 0.35, respectively; OR 1.61, 95% CI 1.11-
2.35, p = 0.016) (table 3).
A renal biopsy was performed in 81 of the 108 patients who presented with nephritis, and
the findings were characterised according to WHO classification criteria for nephritis.
Within this classification, WHO classes III and IV are of particular interest for this study,
since these forms of nephritis are associated with proliferative inflammation, high titers of
antinuclear antibodies, as well as complement consumption [26]. However, no significant
skewing was observed for FcγRIIa, FcγRIIIa, or FcγRIIIb within this patient group. In
contrast, in patients who presented with WHO class II nephritis, marked skewing toward
the FcγRIIIb-NA1 allele could be detected. This skewing did not reach statistical
significance (frequency of 0.54 versus 0.35 in nonrenal SLE patients; OR 2.18, 95% CI
0.97-4.93, p = 0.085) (table 3). Analysis of patients with WHO class V and class VI was
not performed due to the small numbers of patients in these groups.
106
Table 3:
Distribution of FcγR gene frequencies in controls, SLE patients without nephritis, and SLE
patients with nephritis, classified according to World Health Organization (WHO) criteria
FcγRIIa FcγRIIIa FcγRIIIb

R131 H131 V158 F158 NA1 NA2
Controls (n=154) 0.47 0.53 0.33 0.67 0.39 0.61
Non-renal SLE (n=122) 0.54 0.46 0.37 0.63 0.35 0.65
Renal SLE (n=108) 0.52 0.48 0.36 0.64 0.46 0.54
WHO III + IV (n=59) 0.53 0.47 0.36 0.64 0.43 0.57
WHO II (n=13) 0.38 0.62 0.42 0.58 0.54 0.46
Values are the frequencies of each allele
With regard to the other ACR criteria, no differences in the distribution of FcγR
genotypes were observed in terms of skin manifestations (malar rash, discoid rash,
photosensitivity, oral ulcers), neurologic disorders, or immunologic abnormalities,
including the presence of antinuclear antibodies (table 4). However, in patients who
presented with symptoms of inflammation associated with an acute-phase response [27,
28], that is, arthritis and/or serositis, significant skewing toward the homozygous
FcγRIIIa-F/F158 genotype could be observed in comparison with patients without these
inflammatory manifestations (82 of 190 versus 10 of 40, respectively; OR 2.28, 95% CI
1.05-4.93, p = 0.035) (table 4). For arthritis, the FcγRIIa polymorphism could play an
important role additionally, since 19.6% of these patients expressed the combined
homozygous FcγRIIIa-F/F158 and FcγRIIa-R/R131 genotype, compared with 5.9% of
patients without arthritis (35 of 179 versus 3 of 51, respectively; OR 3.89, 95% CI 1.14-
13.22, p = 0.019). Finally, there was a strong trend toward enrichment of the homozygous
FcγRIIIa-F/F158 genotype in patients with autoantibody-associated cell destruction, that
is, haematologic cytopenias, compared with patients without these conditions (72 of 163
versus 20 of 67, respectively; OR 1.86, 95% CI 1.01-3.42, p = 0.062) (table 4).
Clearance studies
The half-life of IgG-coated erythrocytes in the blood, as a measure of Fc receptor
function, was determined at the University Hospital of Groningen in a previous study
[22]. Thirteen patients who participated in this study were included in our study for FcγR
genotype analysis. One of these 13 patient was deficient for FcγRIIIb.
When patients with the homozygous FcγRIIa-R/R131 genotype were compared with
patients expressing other FcγRIIa genotypes, a significant increase in the half-life of IgG-
coated erythrocytes in the blood was observed (Spearman’s r = 0.61, p = 0.027) (figure
1a). No differences in half-life of IgG-coated erythrocytes were observed for either
FcγRIIIa or FcγRIIIb genotypes (figures 1b and 1c). Notably, 2 patients with FcγRIIa-
H/H131, 2 patients with FcγRIIIa-V/V158 as well as 2 patients with FcγRIIIb-NA1/NA1
had accelerated clearance rates (figure 1), but these are different patients in each graph.
107
Chapter IX
Table 4:
Distribution of FcγR genotypes in patients with SLE classified according to American College of
Rheumatology (ACR) criteria
ACR criterion FcγRIIa FcγRIIb FcγRIIIb

R/R H/H + R/H OR F/F V/V + V/F OR NA2/2 NA1/1 +1/2 OR
Malar rash + 32 79 0.93 48 63 1.30 40 71 0.86

- 36 83 44 75 47 72
Discoid rash + 12 37 0.72 17 32 0.75 23 26 1.62
- 56 125 75 106 64 117
Photo-sensitivity + 30 83 0.75 45 68 0.99 41 72 0.88
- 38 79 47 70 46 71
Oral ulcers + 7 29 0.53 13 23 0.82 13 23 0.92
- 61 133 79 115 74 120
Arthritis, serositis + 58 132 1.32 82 108 2.28* 71 119 0.90
- 10 30 10 30 16 24
Renal involvement + 31 77 0.92 44 64 1.06 30 78 0.44†
- 37 85 48 74 57 65
Neurologic symptoms + 3 17 0.39 10 10 1.56 6 14 0.68
- 65 145 82 128 81 129
Haematologic symptoms + 48 115 0.98 72 91 1.86‡ 63 100 1.13
- 20 47 20 47 24 43
Immunologic symptoms + 61 148 0.82 86 123 1.75 79 130 0.99
- 7 14 6 15 8 13
Anti-nuclear antibodies + 66 160 0.41 91 135 2.02 84 142 0.20
- 2 2 1 3 3 1
OR = Odds Ratio; * 95% CI 1.05-4.93, P = 0.035; †95% CI 0.25-0.76, P = 0.004; ‡95% CI 1.01-3.42, P = 0.062.
Discussion
In the present study, we determined the influence of three functionally relevant FcγR
polymorphisms on susceptibility to SLE and the development of clinical disease
manifestations in a strictly Caucasian population. In addition, we determined the influence
of different genotypes on the clearance of immune complexes in vivo in a pilot study.
Analysis of the FcγRIIa-R-H131, FcγRIIIa-V-F158 and FcγRIIIb-NA1-NA2 polymor-
phisms demonstrated that none of these genotypes constitutes a genetic risk factor for the
development of SLE. However, we did find a strong trend toward skewing of FcγRIIa,
with an enrichment of the homozygous FcγRIIa-R/R131 genotype in patients with SLE
compared to healthy controls.
108
Remarkably, this genotype was also (a) FcγRIIa
associated with decreased clearance of 70
IgG-coated erythrocytes in the blood. 60
The results of our study are in 50
T1/2 (min)
accordance with most association studies 40
on the FcγRIIa-R/H131 polymorphism 30
and susceptibility to SLE, performed in 20
at least five different ethnic backgrounds 10
[11, 14-20]. In contrast, some studies 0
H/H131 H/R131 R/R131
found an apparent correlation between
the homozygous FcγRIIa-R/R131 FcγRIIIa
(b)
genotype and the occurrence of SLE [10, 70
12-13]. Notably, a recent study in a 60
Caucasian German population did not 50
demonstrate an association of the T1/2 (min)
40
homozygous FcγRIIa-R/R131 genotype
30
and susceptibility to SLE, but showed
20
that this genotype was significantly
10
correlated with higher frequencies of
0
various clinical and serological V/V158 V/F158 F/F158
parameters, including a younger age at
FcγRIIIb
disease onset [17]. We could not (c)
70
reproduce these associations in our study,
60
despite our finding of a trend toward
50
enrichment of the homozygous FcγRIIa-
T1/2 (min)
R/R131 genotype in our patient 40
population in general. Therefore, in 30
agreement with that previous study [17], 20
we conclude that in Caucasian 10
populations, the R/H polymorphism of 0

NA1/NA1 NA1/NA2 NA2/NA2
FcγRIIa does not represent a strong
genetic risk factor for SLE, but
constitutes a relatively minor determinant Figure 1: Half life (T1/2) of IgG-coated
in susceptibility to the disease. We erythrocytes in the blood of patients with
speculate that this polymorphism may systemic lupus erythematosus expressing
different Fcγ receptor (FcγR) genotypes.
influence the course of the disease by Half life was prolonged in patients
influencing the clearance of immune expressing the FcγRIIa-R/R131 genotype (a),
complexes, since we found an apparent demonstrating an effect of this genotype on
correlation between these parameters in the clearance of immune complexes in vivo.
our in vivo clearance studies. No differences were observed for either
FcγRIIIa (b) or FcγRIIIb (c) genotypes.
109
Chapter IX
Our results do, however, challenge 2 of 3 studies on the FcγRIIIa-V-F158 polymorphism

and susceptibility to SLE. In a population of mixed ethnic background, as well as in a
Caucasian population, apparent skewing was observed toward the homozygous FcγRIIIa-
F/F158 genotype [6, 16]. We could not reproduce these results in our population, which is
consistent with the finding of a recent study in an African American population [20].
Finally, we did not find an association between the FcγRIIIb-NA1/NA2 polymorphism
and susceptibility to SLE, as one other study previously described [16].
We also assessed the influence of Fc R polymorphisms on the development of clinical
manifestations. Of particular interest in our analysis was glomerulonephritis, since this
complication is associated with the deposition of immune complexes and autoantibodies
at the glomerular basement membrane [29]. In our patient population, we could not
demonstrate a correlation between the occurrence of nephritis and the R/H polymorphism
of FcγRIIa, or the V/F polymorphism of FcγRIIIa. This finding is in agreement with most
studies on FcγRIIa [14-20], but in contrast to 3 studies on FcγRIIIa [6,16,18]. It should be
noted, however, that the latter studies were performed in 3 different ethnic backgrounds.
In a previous study conducted at our hospitals [11], glomerulonephritis was found to be
significantly associated with the FcγRIIa-R/R131 genotype compared with healthy
controls. Although there is some overlap between the cohorts of patients and controls that
participated in the 2 studies, we could not reproduce this association in the present study.
The observed difference may be attributable to differences in the criteria used to define
renal involvement, as well to the different sample sizes between these studies. Notably, no
significant association was found upon comparison of patients with and without nephritis
in either study.
Remarkably, we did find significant skewing toward the FcγRIIIb-NA1 allele in patients
with nephritis compared with patients without nephritis. This skewing was not apparent in
patients who presented with WHO class III or IV nephritis, but was notable in patients
with WHO class II nephritis. Because we would argue that the low-binding NA2 allele of
FcγRIIIb is associated with reduced clearance of immune complexes, this finding is rather
unexpected. A previous study in a Caucasian population did not demonstrate a correlation
between nephritis and FcγRIIIb alleles [16].
In patients who presented with symptoms of inflammation associated with an acute-phase
response [27,28], that is, arthritis or serositis, we observed significant skewing toward the
homozygous FcγRIIIa-F/F158 genotype. These manifestations are often associated with
raised C-reactive protein (CRP) levels, as opposed to cerebral lupus or renal involvement
[27,28]. Recently, CRP has been demonstrated to interact with FcγR on phagocytic cells
[30], and this interaction might be polymorphism-dependent [31]. Indeed, this finding
may have important implications for the inflammatory potential of immune complexes in
patients who are homozygous for certain FcγR alleles. For the development of arthritis,
the FcγRIIa polymorphism could play an important role additionally, since 1 in 5 patients
with this disease complication were shown to express the combined FcγRIIIa-F/F158 and
FcγRIIa-R/R131 genotype. This combined FcγR genotype has also been shown to
110
influence the relapse rate in patients with other autoimmune diseases [32], and a critical
role for at least 2 FcγR has been demonstrated in models of collagen-induced arthritis
using FcR-deficient mice [33].
A strong trend toward enrichment of the homozygous FcγRIIIa-F/F158 genotype was also
observed in patients with haematologic abnormalities, that is, hemolytic anemia,
leukopenia, lymphopenia, and thrombocytopenia. The prognosis of SLE is negatively
influenced by the occurrence of autoimmune cytopenias, as has been clearly demonstrated
for anemia and thrombocytopenia [34-36]. Indeed, the V-F polymorphism of FcγRIIIa
might influence the clearance of autoantibody-opsonized blood cells, leading to
subsequent Fc receptor-mediated destruction of these cells. Notably, a critical role for
FcγR has been demonstrated in models of experimental hemolytic anemia and
thrombocytopenia in FcR γ-chain-deficient mice that lack FcγRI and FcγRIII expression
[37].
Finally, SLE patients with the homozygous FcγRIIIa-F/F158 genotype presented with
disease at a slightly younger age than did patients with other FcγR genotypes, although
this difference did not reach statistical significance. In view of these findings, we
conclude that in our Caucasian population, the FcγRIIIa polymorphism constitutes a factor
that influences the clinical disease manifestations and the course of SLE, without
representing a genetic risk factor for susceptibility to SLE. However, since corrections for
multiple testing were not made due to the explorative nature of this study, and because
there may be unknown associations between clinical manifestations, the observed
associations deserve to be interpreted with caution.
Genetic linkage studies have recently provided evidence for an SLE-associated locus on
the long arm of chromosome 1, either including [38] or excluding [39] the FcγR-encoding
gene cluster (mapped to 1q23-24). Despite the fact that the genes for FcγRIIa, FcγRIIIa
and FcγRIIIb are clustered in close proximity on chromosome 1, there is no evidence for
non-random distribution of FcγR genotypes within healthy control populations [40].
Accordingly, we did not find evidence for a skewed distribution of combinations of FcγR
genotypes between SLE patients and healthy controls in our analyses.
In conclusion, the present data support the notion that in Caucasian populations, the allelic
variant of FcγRIIa is a relatively minor determinant in susceptibility to SLE, whereas the
allelic variants of FcγRIIIa may influence the development of clinical disease
manifestations. Notably, the R/H polymorphism of FcγRIIa affects the clearance of
immune complexes in vivo, which may indeed profoundly influence the course of a
disease like SLE.
Acknowledgements: The authors thank Fredriek G.J. Leppers–van de Straat for

excellent technical support.
111
Chapter IX
References
1. Kimberly RP, Salmon JE, Edberg JC, Gibofsky A. The role of Fcγ receptors in
mononuclear phagocyte system function. Clin Exp Rheumatol 1989;7/S-3:103-08.
2. Ravetch JV. Fc receptors. Curr Opin Immunol 1997;9:121-25.
3. Rascu A, Repp R, Westerdaal NAC, Kalden JR, van de Winkel JGJ. Clinical relevance of
Fcγ receptor polymorphisms. Ann N Y Acad Sci 1997;815:282-95.
4. Pol WL van der, van de Winkel JGJ. IgG receptor polymorphisms: risk factors for disease.
Immunogenetics 1998;48:222-32.
5. Parren PWHI, Warmerdam PAM, Boeije LCM, Arts J, Westerdaal NAC, Vlug A, et al. On
the interaction of IgG subclasses with the low affinity FcγRIIa (CD32) on human
monocytes, neutrophils, and platelets. J Clin Invest 1992;90:1537-46.
6. Wu J, Edberg JC, Redecha PB, Bansal V, Guyre PM, Coleman K, et al. A novel
polymorphism of FcγRIIIa (CD16) alters receptor function and predisposes to autoimmune
disease. J Clin Invest 1997;100:1059-70.
7. Koene HR, Kleijer M, Algra J, Roos D, von dem Borne AEGKr, de Haas M. FcγRIIIa-
158V/F polymorphism influences the binding of IgG by NK cell FcγRIIIa, independently of
the FcγRIIIa-48L/R/H phenotype. Blood 1997;90:1109-14.
8. Salmon JE, Edberg JC, Brogle NL, Kimberly RP. Allelic polymorphisms of human Fc
gamma receptor IIa and Fc gamma receptor IIIb. Independent mechanisms for differences
in human phagocyte function. J Clin Invest 1992;89:1274-81.
9. Bredius RGM, Fijen CAP, de Haas M, Kuijper EJ, Weening RS, van de Winkel JGJ, et al.
Role of the neutrophil FcγRIIa (CD32) and the FcγRIIIb (CD16) polymorphic forms in
phagocytosis of human IgG1- and IgG3-opsonized bacteria and erythrocytes. Immunology
1994;83:624-30.
10. Blasini AM, Stekman IL, Leon-Ponte M, Caldera D, Rodriguez MA. Increased proportion
of responders to a murine anti-CD3 monoclonal antibody of the IgG1 class in patients with
systemic lupus erythematosus (SLE). Clin Exp Immunol 1993;94:423-28.
11. Duits AJ, Bootsma H, Derksen RHWM, Spronk PE, Kater L, Kallenberg CGM, et al.
Skewed distribution of IgG Fc receptor IIa (CD32) polymorphism is associated with renal
disease in systemic lupus erythematosus patients. Arthritis Rheum 1995;39:1832-36.
12. Salmon JE, Millard S, Schachter LA, Arnett FC, Ginzler EM, Gourley MF, et al. FcγRIIA
alleles are heritable risk factors for lupus nephritis in African Americans. J Clin Invest
1996;97:1348-54.
13. Song YW, Han C-W, Kang S-W, Baek H-J, Lee E-B, Shin C-H, et al. Abnormal
distribution of Fcγ receptor type IIa polymorphisms in Korean patients with systemic lupus
erythematosus. Arthritis Rheum 1998;41:421-26.
14. Botto M, Theodoridis E, Thompson EM, Beynon HLC, Briggs D, Isenberg DA, et al.
FcγRIIa polymorphism in systemic lupus erythematosus (SLE): no association with disease.
Clin Exp Immunol 1996;104:264-68.
112
15. Smyth LJC, Snowden N, Carthy D, Papasteriades C, Hajeer A, Ollier WER. FcγRIIa
polymorphism in systemic lupus erythematosus. Ann Rheum Dis 1997;56:744-46.
16. Koene HR, Kleijer M, Swaak AJG, Sullivan KE, Bijl M, Petri MA, et al. The FcγRIIIA-
158F allele is a risk factor for systemic lupus erythematosus. Arthritis Rheum
1998;41:1813-18.
17. Manger K, Repp R, Spriewald BM, Rascu A, Geiger A, Wassmuth R, et al. Fcγ receptor IIa
polymorphism in caucasian patients with systemic lupus erythematosus: association with
clinical symptoms. Arthritis Rheum 1998;41:1181-89.
18. Salmon JE, Sonia Ng, Yoo D-H, Kim T-H, Kim SY, Song GG. Altered distribution of Fcγ
receptor IIIA alleles in a cohort of Korean patients with lupus nephritis. Arthritis Rheum
1999;42:818-19.
19. Yap SN, Phipps ME, Manivasagar M, Tan SY, Bosco JJ. Human Fc gamma receptor IIA
(FcγRIIA) genotyping and association with systemic lupus erythematosus (SLE) in Chinese
and Malays in Malaysia. Lupus 1999;8:305-10.
20. Oh M, Petri MA, Kim NA, Sullivan KE. Frequency of the FcγRIIIA-158F allele in African
American patients with systemic lupus erythematosus. J Rheumatol 1999;26:1486-89.
21. Osborne JM, Chacko GW, Brandt JT, Anderson CL. Ethnic variation in frequency of an
allelic polymorphism of human FcγRIIA determined with allele specific oligonucleotide
probes. J Immunol Methods 1994;173:207-17.
22. Woude FJ van der, van der Giessen M, Kallenberg CGM, Ouwehand W, Beekhuis H,
Beelen JM, et al. Reticuloendothelial Fc receptor function in SLE patients. I. Primary HLA
linked defect or acquired dysfunction secondary to disease activity? Clin Exp Immunol
1984;55:473-80.
23. Tan EM, Cohen AS, Fries JF, Masi AT, McShane DJ, Rothfield NF, et al. The 1982 revised
criteria for the classification of systemic lupus erythematosus. Arthritis Rheum
1982;25:1271-77.
24. Churg J, Bernstein J, Glassock RJ. Lupus nephritis. In: Churg J, Bernstein J, Glassock RJ,
editors. Renal disease: classification and atlas of glomerular diseases. 2nd ed. New York:
Igaku-Shoin; 1995. p. 151-80.
25. Haas M de, Kleijer M, van Zwieten R, Roos D, von dem Borne AEGKr. Neutrophil
FcγRIIIb deficiency, nature, and clinical consequences: a study of 21 individuals from 14
families. Blood 1995;86:2403-13.
26. Huong DL, Papo T, Beaufils H, Wechsler B, Bletry O, Baumelou A, et al. Renal
involvement in systemic lupus erythematosus: a study of 180 patients from a single center.
Medicine Baltimore 1999;78:148-66.
27. Ter Borg EJ, Horst G, Limburg PC, van Rijswijk MH, Kallenberg CGM. C-reactive protein
levels during disease exacerbations and infections in systemic lupus erythematosus: a
prospective longitudinal study. J Rheumatol 1990;17:1642-48.
28. Spronk PE, ter Borg EJ, Kallenberg CGM. Patients with systemic lupus erythematosus and
Jaccoud’s arthropathy: a clinical subset with an increased C reactive protein response? Ann
Rheum Dis 1992;51:358-61.
113
Chapter IX
29. Berden JH, van Bruggen MC. Nucleosomes and the pathogenesis of lupus nephritis. Kidney
Blood Press Res 1997;20:198-200.
30. Bharadwaj D, Stein MP, Volzer M, Mold C, Du Clos TW. The major receptor for C-
reactive protein on leukocytes is Fcγ receptor II. J Exp Med 1999;190:585-90.
31. Stein MP, Edberg JC, Kimberly RP, Mangan EK, Bharadwaj D, Mold C, et al. C-reactive
protein binding to FcγRIIa on human monocytes and neutrophils is allele-specific. J Clin
Invest 2000;105:369-76.
32. Dijstelbloem HM, Scheepers RHM, Oost WW, Stegeman CA, van der Pol WL, Sluiter WJ,
et al. Fcγ receptor polymorphisms in Wegener’s granulomatosis: risk factors for disease
relapse. Arthritis Rheum 1999;42:1823-27.
33. Yuasa T, Kubo S, Yoshino T, Ujike A, Matsumura K, Ono M, et al. Deletion of Fcγ
receptor IIB renders H-2(b) mice susceptible to collagen-induced arthritis. J Exp Med
1999;189:187-94.
34. Ginzler EM, Diamond HS, Weiner M, Schlesinger M, Fries JF, Wasner C, et al. A
multicenter study of outcome in systemic lupus erythematosus. I. Entry variables as
predictors of prognosis. Arthritis Rheum 1982;25;601-11.
35. Jacobsen S, Petersen J, Ullman S, Junker P, Voss A, Rasmussen JM, et al. A multicentre
study of 513 Danish patients with systemic lupus erythematosus. II. Disease mortality and
clinical factors of prognostic value. Clin Rheumatol. 1998;17:478-84.
36. Abu-Shakra M, Urowitz MB, Gladman DD, Gough J. Mortality studies in systemic lupus
erythematosus. Results from a single center. II. Predictor variables for mortality. J
Rheumatol 1995;22:1265-70.
37. Clynes R, Ravetch JV. Cytotoxic antibodies trigger inflammation through Fc receptors.
Immunity 1995;3:21-26.
38. Moser KL, Neas BR, Salmon JE, Yu H, Gray-McGuire C, Asundi N, et al. Genome scan of
human systemic lupus erythematosus: evidence for linkage on chromosome 1q in African-
American pedigrees. Proc Natl Acad Sci USA 1998;95:14869-74.
39. Tsao BP, Cantor RM, Kalunian KC, Chen CJ, Badsha H, Singh R, et al. Evidence for
linkage of a candidate chromosome 1 region to human systemic lupus erythematosus. J Clin
Invest 1997;99:725-31.
40. Lehrnbecher T, Foster CB, Zhu S, Leitman SF, Goldin LR, Huppi K, et al. Variant
genotypes of the low-affinity Fcγ receptors in two control populations and a review of low-
affinity Fcγ receptor polymorphisms in control and disease populations. Blood
1999;94:4220-32.
114
Chapter X
NEW INSIGHTS INTO THE PATHOGENESIS OF

SYSTEMIC LUPUS ERYTHEMATOSUS (SLE):
the role of apoptosis
Marc Bijl1
Pieter C. Limburg2
Cees G.M. Kallenberg1.
1
2`
The Netherlands
Submitted
Chapter X
Introduction
Systemic lupus erythematosus (SLE) is an autoimmune disease with a wide spectrum of

clinical and immunological abnormalities. The prevalence of the disease ranges from 14.6
till 50.8 per 100.000 persons [1]. SLE develops predominantly in women of childbearing
age. The presence of autoantibodies, especially those directed to dsDNA, is characteristic
for the disease. The cause of SLE is unknown. Insight into the pathogenesis of the disease,
however, has deepened in recent years. In particular studies on apoptosis and clearance of
apoptotic cells in lupus have shed a new and intriguing light on the development and
course of the disease. In this review we will focus on apoptosis, on the elimination of
apoptotic cells and on the consequences that disturbances in one of these processes might
have on the presentation of autoantigens to the immune system in such a way that
tolerance can be broken and autoimmunity occurs.
Apoptosis
Apoptosis or programmed cell death is recognised as being fundamental to maturation and

homeostasis of the immune system. During maturation of the immune system apoptosis of
autoreactive lymphocytes in the central lymphoid organs underlies the development of
tolerance. Furthermore, due to apoptosis the size of the peripheral lymphoid and myeloid
compartments is limited. Activated lymphocytes are deleted through apoptosis following
an immune response [2]. Apoptosis is initiated through the ligation of specific death
receptors on the cell surface. Ligation of the death receptors is followed by a cascade of
enzymatic activations, morphologically accompanied by condensation and fragmentation
of cells and nuclei, and blebbing of the plasma membrane [3]. In these blebs autoantigens
are exposed. Casciola-Rosen et al. showed that in apoptotic keratinocytes many nuclear
and cytoplasmic antigens, that are characteristic targets for autoantibodies, are clustered in
high concentrations in the cell membrane [4]. Furthermore, alterations of cellular
constituents during apoptotic degradation, occurring via posttranslational modifications
such as oxidation, phosphorylation, or citrullination, could induce immunogenicity [5,6].
These findings made it understandable why in autoimmunity an immune response can
develop against intracellular, cryptic epitopes, and stresses the important role apoptotic
cells might play in the pathogenesis of autoimmune disorders. Whenever apoptotic cells
do accumulate by an increased rate of apoptosis, decreased elimination, or a combination
of both, it can be imagined that, in the proper environment, tolerance can be broken.
For the initiation of the apoptotic process ligation of the cell death receptors is mandatory.
Cell death receptors belong to the tumor necrosis factor (TNF) receptor superfamily and
share a similar, cysteine rich extracellular domain. In addition, the death receptors contain
a homologous cytoplasmic sequence called the death domain. The best characterised death
receptors are Fas and TNFR1. The Fas-mediated apoptosis pathway has been extensively
118
Pathogenesis of SLE
studied and is crucial for the development of immune tolerance and privilege (reviewed in
[7] and [8]). Fas, also known as APO-1 or CD95, is a cell surface protein of 45-48 kD [9].
Fas is constitutively expressed on subpopulations of peripheral blood lymphocytes,
Fas Ligand
Fas
Cell m em brane
Death domain
FADD (MORT1)
Active
caspase-8 Bid
Caspase-8
(FLICE)
Mitochondria
Proteolysis
of PARP
Caspase 9
Caspase-3
Bcl-2
Active Cytochrom e C
caspase 9
Apoptosis
Figure 1: Apoptotic signal transduction induced by binding of Fas ligand (FasL) to its receptor
Fas (CD95). FasL is a homotrimeric molecule. Binding to Fas results in trimerization of Fas. The
Fas cytoplasmatic region carries a death domain. Trimerization of this death domain recruits
caspase 8 via an adaptor called FADD (Fas associated death domain)/MORT1. Upon recruitment
by FADD, caspase-8 (also called FLICE) drives its activation through self-cleavage. Activated
caspase-8 then degrades poly (ADP-ribose) polymerase (PARP), an enzyme that is thought to be
involved in DNA repair, and activates downstream effector caspases committing the cell to
apoptosis with the characteristic degradation of chromosomal DNA and morphological changes.
Furthermore, caspase-8 cleaves Bid, a pro-apoptotic member of the BH3 subfamily. Cleaved Bid
stimulates the release of cytochrome C from mitochondria which activates caspase-9, which
subsequently activates caspase-3. The release of cytochrome C into the cytoplasm can be
inhibited by bcl-2 which resides on the cytoplasmic face of the mitochondrial outer membrane.
119
Chapter X
predominantly memory T cells, and is strongly upregulated on T- as well as B-

lymphocytes upon activation [10]. Cross-linking of Fas by Fas ligand (FasL), a membrane
glycoprotein of 40 kD, induces trimerization of the receptor. This recruits caspase-8 via an
adaptor called FADD/MORT1. Aggregation of caspase-8 causes self-activation and
activates pro-caspase-3 by proteolytic cleavage that cleaves various other cellular
substrates, which finally results in DNA fragmentation (figure 1). Whether apoptosis is
induced after engagement of the Fas receptor depends on many different factors such as
the stage of the cell in ontogeny and the state of activation of the cell involved. Many of
these factors influence the expression of the Bcl-2 protein. This protein belongs to a
family of related cytoplasmic proteins that are key regulators of apoptosis. The family
consists of at least 15 members, divided into three subfamilies [11]. The Bcl-2 cohort
promotes survival, whereas the Bax and BH3 subfamilies facilitate apoptosis.
Disturbances in one of the many factors that regulate the apoptotic process, might change
the balance present in the immune system and may predispose for the development of
autoimmune phenomena. In recent years many studies have been performed, both in
animal models and in patients with autoimmune diseases, in which the role of these
different factors has been investigated.
Animal models
The importance of functionally intact apoptotic pathways for maintenance of immune

homeostasis and prevention of autoimmunity was highlighted by some intriguing studies
in lupus prone mice. Historically, experimental studies in animals designed to obtain
further insight in the pathogenesis of SLE are based on the MRL-lpr/lpr
(lymphoproliferation), CBA-lprcg/lprcg and C3H/HeJ-gld/gld (generalised lymphoproli-
ferative disease) mice models. These animals spontaneously develop lymphadenopathy as
well as autoimmune features characterised by the presence of autoantibodies to nuclear
antigens [12]. It has been recognised that the molecular basis of these abnormalities is a
defect in Fas mediated apoptosis. The MRL-lpr/lpr mice have a mutation in the Fas
molecule [13] that, in the proper genetic background, results in the generation of
abnormally spliced mRNA and a severe reduction in functional Fas mRNA. The
C3H/HeJ-gld/gld mice are deficient in functional Fas ligand (FasL) due to a single base
mutation [14]. The defects in Fas and FasL, respectively, result in an incomplete
elimination of peripheral autoreactive cells as this elimination occurs predominantly by
Fas mediated apoptosis [15]. Other evidence for the importance of Fas induced apoptosis
in the development of autoimmune disease was delivered by the discovery of soluble Fas
[16]. Soluble Fas (sFas) is the transcript of Fas mRNA lacking the transmembrane domain
because of the deletion of the exon encoding this region. Injecting sFas into female mice
of a certain mouse strain (CD1) resulted in the development of autoimmune features due
to blocking of Fas induced apoptosis [16]. The importance of Bcl-2 expression for
lymphocyte survival and the development of autoimmune disease was supported by
120
Pathogenesis of SLE
experiments with transgenic mice. In a model of host-versus-graft disease, a self-limited

model of systemic autoimmune disease, it was shown that overexpression of Bcl-2
provides survival signals for autoreactive B cells. This resulted in the production of
pathogenic autoantibodies [17]. Furthermore, it has been shown that Bcl-2 protects
thymocytes from radiation and glucocorticoid-induced cell death [11].
In summary, data from animal models demonstrate that alterations in apoptosis induction
can contribute to the development of autoimmunity. In particular, lack of induction of
apoptosis of autoreactive lymphocytes can induce autoimmunity in certain animal models.
Human SLE
Following the demonstration that the autoimmune lymphoproliferative syndrome in the

lpr and gld mouse strains was based on a single gene defect, studies in humans were
performed to detect similar functional defects in Fas or FasL. Indeed, some families have
been described with a mutation in Fas [18-21] or FasL [22]. Also in humans, defects in the
Fas apoptotic pathway result in splenomegaly and lymphadenopathy called the
autoimmune lymphoproliferative syndrome (ALPS), formerly named the Canale-Smith
syndrome [2,18]. In contrast to the mouse models, autoimmune phenomena in affected
human individuals are rare [20]. Moreover, in two studies a total amount of 218 lupus
patients were screened for a defect in the FasL gene. This revealed only one affected
patient suggesting that such mutations are an uncommon cause of the disease [22,23].
Additional studies performed to detect functional defects in the apoptotic machinery in
SLE did not demonstrate any abnormalities [24,25]. Several studies have been performed
in which sFas was measured in SLE patients. These studies demonstrated elevated levels
of sFas in lupus patients and a relation between sFas levels and disease activity [26-32].
Elevated levels of sFas were found prior to relapse of the disease [32] and correlated with
the state of activation of B-lymphocytes [31]. In contrast to the study of Cheng et al. in
which it was shown that sFas inhibited Fas-mediated apoptosis, the presence of sFas does
not seem to block Fas-induced apoptosis in significant amounts in SLE patients. Levels of
apoptotic lymphocytes in SLE patients in vitro as well as in vivo were higher than in
healthy controls [24,33-35]. Even in patients with elevated levels of sFas increased
numbers of apoptotic lymphocytes could be demonstrated in the peripheral blood [36].
Probably, levels of sFas reflect increased lymphocyte activation. It has been shown that
even during inactive disease increased proportions of activated T- as well as B-
lymphocytes are present in the peripheral blood of SLE patients[31,37]. Probably, this is
most pronounced in patients with grumbling disease at risk for a relapse. During disease
exacerbations the state of lymphocyte activation increases further [37]. Lymphocyte
activation results in increased transcription of membrane Fas as well as the alternative
splice form sFas, and explains the relation found between sFas levels and disease activity.
This suggests that, in the long term, an association between sFas levels and organ damage
will occur in patients with remitting disease activity. Indeed, sFas levels correlated with
121
Chapter X
organ damage as measured by the SLICC/ACR score [38]. The hypothesis that sFas is just
a reflection of lymphocyte activation and is therefore not specific for SLE is supported by
the presence of elevated sFas levels in rheumatoid arthritis patients [28,36]. Finally, it
must be stressed that the relation between lymphocyte activation, sFas and apoptosis is
complex, especially in SLE patients. Many factors influence this interrelation. Increased
expression of the aforementioned proto-oncogenes such as Bcl-2 have been found in SLE
[39-41]. Furthermore, increased levels of soluble Fas ligand [42], and the presence of
antibodies to poly (ADP-ribose) polymerase (PARP) can influence the results in studies
dealing with apoptosis in SLE. PARP is necessary for the repair of DNA breaks and is
inactivated early in apoptosis in order to enable the execution of the apoptotic machinery.
This inactivation is blocked by the presence of antibodies to PARP, which have been
detected in SLE patients [43].
So, although suggested by animal studies, no genetically defined defects in the major, Fas-
mediated, apoptotic pathway have been detected in human SLE. Therefore, a hereditary
defect in the induction of apoptosis as a mechanism of elimination of autoreactive
lymphocytes does not seem a prerequisite factor for the development of human
autoimmune disorders. In contrast, studies in human SLE have demonstrated increased
levels of apoptotic cells even in the presence of factors that can inhibit apoptosis induction
like elevated levels of sFas and increased expression of Bcl-2. The increased levels of
apoptotic cells found might be due to increased in vivo activation and Fas-mediated
apoptosis of lymphocytes. Recently, it was found that chlorpromazine, a drug known to
induce lupus erythematosus, induces apoptosis in lymphoblasts independently of Fas [44].
Taken together, data from lupus patients supply evidence that the increased induction of
apoptosis via Fas or via other pathways can trigger the development of autoimmunity.
Phagocytosis of apoptotic cells
The increased presence of apoptotic cells as demonstrated in the peripheral blood of SLE
patients can be accounted for by an increased level of activation induced cell death.
However, apoptosis is a physiological mechanism and occurs continuously in impressive
amounts. For example, every day 2x109/kg body weight apoptotic neutrophils are
removed from the blood stream [45]. Removal of apoptotic cells occurs very effectively
via phagocytosis by bystander (semi-professional) or professional phagocytes like
monocytes and macrophages [27]. Rapid elimination of apoptotic cells is important as it
prevents the release of (toxic) cell constituents like cytolytic enzymes. Furthermore, it has
become clear that during the process of apoptosis antigens are newly exposed in the cell
membrane of the apoptotic cell [4]. Adequate removal of apoptotic cells therefore also
seems important for the prevention of (excessive) autoantigenic exposure.
The interaction between apoptotic cells and other cells, necessary for their elimination, is
very complex [46]. Monocytes and macrophages constitutively express several receptors
122
Pathogenesis of SLE
like CD14, CD36 and scavenger receptors, all involved in the recognition, binding and
internalisation of apoptotic cells [46-55]. Next to binding of apoptotic cells to these
receptors several serum proteins play a role in this process of elimination (figure 2).
Already early in the apoptotic cascade the cell membrane changes. For example,
phosphatidylserine is exposed at the outer surface of the cell membrane. Due to these
changes several serum constituents like complement C1q, C3 and C4, C-reactive protein
(CRP), serum amyloid protein P (SAP) and phospholipase A2 can bind to the apoptotic
cell and facilitate, by yet unknown mechanisms, the interaction with phagocytes [56-59].
Furthermore, phagocytosis itself modulates phagocyte behaviour. The ingestion of
apoptotic cells in vitro promotes the secretion of anti-inflammatory cytokines in human
macrophages [60]. Furthermore, uptake of apoptotic neutrophils by macrophages reduced
the uptake of the former cells when a rechallenge was performed after 48 hours [61]. In
this context, it can be speculated that an increased rate of apoptosis could lead to an
overflow of the phagocytic system with apoptotic cells through this negative feedback
loop.
Thus, as will be discussed, next to increased induction of apoptotic cells an intrinsic
decreased clearance capacity of the phagocytic system, possibly in combination with
defects in the production of anti-inflammatory mediators by macrophages, might be an
important pathogenic factor in the development of SLE.
Animal models
Several animal studies have deepened our understanding of the role of apoptotic cells and
disturbances in the clearance of these cells for the development of autoimmune
phenomena. First, it has been shown that the presence of large numbers of apoptotic cells
can evoke an immune response. Mevorach et al. demonstrated that the intravenous
injection of apoptotic thymocytes resulted in the production of autoantibodies to nuclear
antigens in the majority of normal mice [62]. In addition, the consequences of a
disturbance in the removal of apoptotic cells have been addressed in several studies. As
discussed before, several serum proteins bind to the surface blebs of apoptotic cells. C1q
binds to apoptotic keratinocytes [59]. SAP can bind to extracellular dsDNA and
chromatin, also present in the apoptotic blebs, under physiological circumstances [56].
The functional impact of these findings has subsequently been demonstrated in mice using
selective gene deletion [63-66]. The important role C1q plays in the removal of apoptotic
cells was demonstrated in C1q-knock out mice. C1q-/- mice had higher titres of
autoantibodies and higher mortality compared with strain-matched controls. Furthermore,
25% of C1q-/- mice had glomerulonephritis with immune deposits and multiple apoptotic
cell bodies. In C1q-/- mice without glomerulonephritis, significantly greater numbers of
glomerular apoptotic bodies were detected than in controls [64]. Similar findings have
been reported in SAP-deficient mice. In mice with a targeted deletion of the SAP gene
autoimmune disease characterised by the presence of autoantibodies to DNA and
123
Chapter X
chromatin and severe glomerulonephritis developed spontaneously [65]. Further

indications that the removal of apoptotic cells and their antigenic structures is relevant in
the pathogenesis of SLE are delivered by a recent study performed with Dnase1-deficient
mice [67]. Dnase1 might have a protective task in the removal of DNA from
nucleoprotein complexes so preventing immune stimulation. Indeed, deletion of Dnase1
resulted in the occurrence of classical symptoms of SLE, including the production of anti-
nuclear antibodies and the development of glomerulonephritis.
Human SLE
Evidence that disturbed phagocytosis of apoptotic cells might be a relevant factor in the
pathogenesis of human autoimmune disease was demonstrated by Hermann et al [68].
They showed that phagocytosis of apoptotic cells by in vitro differentiated macrophages
obtained from SLE patients was impaired compared to phagocytosis by macrophages
obtained from healthy controls.
Interestingly, in families with congenital deficiencies of the complement factors C1q, C2
or C4, the fast majority of the affected members spontaneously develop SLE [69]. The
important role of complement factors is further supported by data showing that the
addition of complement components to an in vitro phagocytosis assay using human
monocyte-derived macrophages and apoptotic lymphocytes provided a more than
threefold increase in the uptake of apoptotic cells [70]. Next to deficiencies in the
complement factors also the CRP response in SLE patients seems to be impaired [71].
Recently, we studied whether in human SLE SAP production is (relatively) deficient
compared to disease controls and healthy controls but could not find significant
differences [72]. Further studies are underway to investigate whether SAP is functionally
intact and influences phagocytotic capacity in SLE patients. Finally, Napirei et al.
measured Dnase1 activity in the sera of SLE patients and found lower levels than in
normal controls [67].
As pointed out, next to a decrease in the concentrations of serum proteins with a
functional role in the process of phagocytosis of apoptotic cells, one might hypothesise
that changes in the expression of membrane receptors necessary for binding and
internalisation of apoptotic cells can influence the phagocytotic capacity. Indeed,
monocytes in SLE patients expressed significantly lower levels of CD14 [73].
Interestingly, this receptor mediates clearance of apoptotic cells without inciting
inflammation [54,74]. A lowered expression of CD14, therefore, can tilt phagocytosis of
apoptotic cells, which normally is non-inflammatory, towards an inflammatory process in
SLE patients. In addition, apoptotic cells can be opsonised by autoantibodies present in
patients with autoimmune disease which are directed to auto-antigens presented on
apoptotic cells [75,76]. Ligation of the Fc fragment of these antibodies with Fcγ-receptors
on macrophages than induces the secretion of pro-inflammatory cytokines, so contributing
to the process of inflammation which is characteristic for autoimmune diseases (figure 2).
124
Pathogenesis of SLE
Macrophage
ABC1
d
Lectin
CR1,3,4
MR
CD14 FcγR
αVβ3 CD36
Anti-inflammatory Pro-inflammatory
cytokines cytokines
TSP PS Sugar
Complement opsonized Autoantibody

apoptotic cell opsonized apoptotic cell
Complement binding Auto antibody binding
PS
CRP, SAP, TSP Apoptotic cell
Drugs/UVB
Normal SLE
Figure 2: Phagocytosis of apoptotic cells by macrophages. Cells die by apoptosis during tissue
turnover or at the end of an immune response. In SLE patients additive stimuli like UVB (inducing
apoptotic keratinocytes) or certain drugs (like chlorpromazine, inducing apoptosis in
lymphoblasts) can increase the number of apoptotic cells generated. Cells undergoing apoptosis
display an orderly process of nuclear condensation and fragmentation and cytoplasmic
contraction. Phosphatidylserine (PS) that normally resides at the inside of the cell membrane,
flips to the outside to become exposed on the cell membrane. Furthermore, surface blebbing and
packaging of cellular components within membranes occurs before their budding from the cell. In
the smaller surface blebs fragmented endoplasmic reticulum and ribosomes are concentrated. The
larger blebs contain nucleosomal DNA. In SLE patients autoantibodies can bind to these antigens
whenever exposed. This will enable FcγR-bearing cells to interact with antibody opsonized
apoptotic cells via the FcγR resulting in the release of pro-inflammatory cytokines. Without the
presence of autoantibodies the apoptotic cell is only opsonized with SAP, complement proteins
and other serum factors like thrombospondin (TSP). Interaction of these molecules facilitates
non-inflammatory phagocytosis mediated by a variety of membrane receptors on phagocytes like
the ATP binding cassette transporter ABC1, complement receptors (CR1, 3, 4), the vitronectin
receptor (αv3β3), CD36, CD14, lectins, and the mannose receptor (MR).
125
Chapter X
In their study on binding of anti-phospholipid antibodies to apoptotic cells, Manfredi et al.

found massive secretion of TNF alpha when apoptotic cells opsonised with these
antibodies were internalised by macrophages [75].
Conclusion
There is increasing evidence that the presence and accumulation of apoptotic cells can
result in autoimmunity. Whether this accumulation in SLE patients is due to increased
production of apoptotic cells, results from decreased phagocytic capacity, or from the
combination of both has to be proven. Nevertheless, it has been shown that tolerance can
be broken due to increased amounts of apoptotic cells. Alternatively, or in conjunction,
posttranslational modifications occurring during the process of apoptosis of cellular
antigens can bypass tolerance. Breaking tolerance induces the production of
autoantibodies that will bind to their antigens whenever exposed on the cell membrane.
The presence of these autoantibodies will allow interaction with Fc gamma receptor
binding cells. This will than result in Fc receptor mediated phagocytosis, which induces
the release of pro-inflammatory cytokines. Finally, this cascade of events will lead to the
development of inflammation characteristic for many autoimmune diseases.
Understanding the processes underlying these inflammatory lesions will allow us in the
near future to intervene therapeutically much more specific in autoimmune mediated
disorders so reducing morbidity and mortality of patients suffering from these diseases.
References
1. Wallace DJ, Hahn BH. Dubois’ Lupus Erythematosus. 5th edition. Philadelphia, London,
Lea & Febiger, 1997.
2. Ashkenazi A, Dixit VM. Death receptors: signaling and modulation. Science
1998;281:1305-8.
3. Wyllie AH, Kerr JF, Currie AR. Cell death: the significance of apoptosis. Int Rev Cytol
1980;68:251-306:251-306.
5. Utz PJ, Hottelet M, Schur PH, Anderson P. Proteins phosphorylated during stress-induced
apoptosis are common targets for autoantibody production in patients with systemic lupus
erythematosus. J Exp Med 1997;185:843-54.
6. Utz PJ, Anderson P. Posttranslational protein modifications, apoptosis, and the bypass of
tolerance to autoantigens. Arthritis Rheum 1998;41:1152-60.
7. Nagata S. Fas ligand-induced apoptosis. Annu Rev Genet 1999;33:29-55:29-55.
126
Pathogenesis of SLE
8. Famularo G, Nucera E, Marcellini S, De Simone C. Fas/Fas ligand on the road: an apoptotic

pathway common to AIDS, autoimmunity, lymphoproliferation and transplantation. Med
Hypotheses 1999;53:50-62.
10. Miyawaki T, Uehara T, Nibu R et al. Differential expression of apoptosis-related Fas
antigen on lymphocyte subpopulations in human peripheral blood. J Immunol
1992;149:3753-8.
11. Adams JM, Cory S. The Bcl-2 protein family: arbiters of cell survival. Science
1998;281:1322-6.
12. Cohen PL, Eisenberg RA. The lpr and gld genes in systemic autoimmunity: life and death in
the Fas lane. Immunol Today 1992;13:427-8.
13. Watanabe-Fukunaga R, Brannan CI, Copeland NG, Jenkins NA, Nagata S.
Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates
apoptosis. Nature 1992;356:314-7.
14. Takahashi T, Tanaka M, Brannan CI et al. Generalized lymphoproliferative disease in mice,
caused by a point mutation in the Fas ligand. Cell 1994;76:969-76.
15. Russell JH, Rush B, Weaver C, Wang R. Mature T cells of autoimmune lpr/lpr mice have a
defect in antigen-stimulated suicide. Proc Natl Acad Sci U S A 1993;90:4409-13.
17. Lopez-Hoyos M, Carrio R, Merino R et al. Constitutive expression of bcl-2 in B cells causes
a lethal form of lupuslike autoimmune disease after induction of neonatal tolerance to H-2b
alloantigens. J Exp Med 1996;183:2523-31.
18. Fisher GH, Rosenberg FJ, Straus SE et al. Dominant interfering Fas gene mutations impair
19. Rieux-Laucat F, Le Deist F, Hivroz C et al. Mutations in Fas associated with human
20. Bettinardi A, Brugnoni D, Quiros-Roldan E et al. Missense mutations in the Fas gene
resulting in autoimmune lymphoproliferative syndrome: a molecular and immunological
analysis. Blood 1997;89:902-9.
21. Vaishnaw AK, Toubi E, Ohsako S et al. The spectrum of apoptotic defects and clinical
manifestations, including systemic lupus erythematosus, in humans with CD95 (Fas/APO-1)
mutations. Arthritis Rheum 1999;42:1833-42.
22. Wu J, Wilson J, He J et al. Fas ligand mutation in a patient with systemic lupus
erythematosus and lymphoproliferative disease. J Clin Invest 1996;98:1107-13.
23. Kojima T, Horiuchi T, Nishizaka H et al. Analysis of fas ligand gene mutation in patients
with systemic lupus erythematosus. Arthritis Rheum 2000;43:135-9.
25. Bijl M, Horst G, Limburg PC, Kallenberg CGM. Anti-CD3-induced and anti-Fas-induced
apoptosis in systemic lupus erythematosus (SLE). Clin Exp Immunol 2001;123:127-32.
127
Chapter X
26. Papo T, Parizot C, Ortova M et al. Apoptosis and expression of soluble Fas mRNA in
27. Rose LM, Latchman DS, Isenberg DA. Elevated soluble fas production in SLE correlates
with HLA status not with disease activity. Lupus 1997;6:717-22.
28. Nozawa K, Kayagaki N, Tokano Y et al. Soluble Fas (APO-1, CD95) and soluble Fas ligand
in rheumatic diseases. Arthritis Rheum 1997;40:1126-9.
29. Jodo S, Kobayashi S, Kayagaki N et al. Serum levels of soluble Fas/APO-1 (CD95) and its
molecular structure in patients with systemic lupus erythematosus (SLE) and other
autoimmune diseases. Clin Exp Immunol 1997;107:89-95.
30. Tokano Y, Miyake S, Kayagaki N et al. Soluble Fas molecule in the serum of patients with
systemic lupus erythematosus. J Clin Immunol 1996;16:261-5.
31. Bijl M, van Lopik T, Limburg PC et al. Do elevated levels of serum-soluble fas contribute
to the persistence of activated lymphocytes in systemic lupus erythematosus? J Autoimmun
1998;11:457-63.
34. Courtney PA, Crockard AD, Williamson K et al. Increased apoptotic peripheral blood
neutrophils in systemic lupus erythematosus: relations with disease activity, antibodies to
double stranded DNA, and neutropenia. Ann Rheum Dis 1999;58:309-14.
1998;7:113-8.
Lupus 1999;8:508-13.
38. Al Maini MH, Mountz JD, Al Mohri HA et al. Serum levels of soluble Fas correlate with
indices of organ damage in systemic lupus erythematosus. Lupus 2000;9:132-9.
40. Aringer M, Wintersberger W, Steiner CW et al. High levels of bcl-2 protein in circulating T
lymphocytes, but not B lymphocytes, of patients with systemic lupus erythematosus.
41. Gatenby PA, Irvine M. The bcl-2 proto-oncogene is overexpressed in systemic lupus
erythematosus. J Autoimmun 1994;7:623-31.
128
Pathogenesis of SLE

43. Decker P, Isenberg D, Muller S. Inhibition of caspase-3-mediated poly(ADP-ribose)
polymerase (PARP) apoptotic cleavage by human PARP autoantibodies and effect on cells
undergoing apoptosis. J Biol Chem 2000;275:9043-6.
44. Hieronymus T, Grotsch P, Blank N et al. Chlorpromazine induces apoptosis in activated
human lymphoblasts: a mechanism supporting the induction of drug-induced lupus
erythematosus? Arthritis Rheum 2000;43:1994-2004.
45. Cohen JJ, Duke RC, Fadok VA, Sellins KS. Apoptosis and programmed cell death in
immunity. Annu Rev Immunol 1992;10:267-93:267-93.
46. Aderem A, Underhill DM. Mechanisms of phagocytosis in macrophages. Annu Rev
Immunol 1999;17:593-623:593-623.
47. Savill J, Dransfield I, Hogg N, Haslett C. Vitronectin receptor-mediated phagocytosis of
cells undergoing apoptosis. Nature 1990;343:170-3.
48. Savill J, Hogg N, Ren Y, Haslett C. Thrombospondin cooperates with CD36 and the
vitronectin receptor in macrophage recognition of neutrophils undergoing apoptosis. J Clin
Invest 1992;90:1513-22.
49. Fadok VA, Savill JS, Haslett C et al. Different populations of macrophages use either the
vitronectin receptor or the phosphatidylserine receptor to recognize and remove apoptotic
cells. J Immunol 1992;149:4029-35.
50. Fadok VA, Voelker DR, Campbell PA et al. Exposure of phosphatidylserine on the surface
of apoptotic lymphocytes triggers specific recognition and removal by macrophages. J
Immunol 1992;148:2207-16.
51. Fadok VA, Warner ML, Bratton DL, Henson PM. CD36 is required for phagocytosis of
apoptotic cells by human macrophages that use either a phosphatidylserine receptor or the
vitronectin receptor (alpha v beta 3). J Immunol 1998;161:6250-7.
52. Platt N, Suzuki H, Kurihara Y, Kodama T, Gordon S. Role for the class A macrophage
scavenger receptor in the phagocytosis of apoptotic thymocytes in vitro. Proc Natl Acad Sci
U S A 1996;93:12456-60.
53. Luciani MF, Chimini G. The ATP binding cassette transporter ABC1, is required for the
engulfment of corpses generated by apoptotic cell death. EMBO J 1996;15:226-35.
54. Devitt A, Moffatt OD, Raykundalia C et al. Human CD14 mediates recognition and
phagocytosis of apoptotic cells. Nature 1998;392:505-9.
55. Moffatt OD, Devitt A, Bell ED, Simmons DL, Gregory CD. Macrophage recognition of
ICAM-3 on apoptotic leukocytes. J Immunol 1999;162:6800-10.
56. Pepys MB, Butler PJ. Serum amyloid P component is the major calcium-dependent specific
DNA binding protein of the serum. Biochem Biophys Res Commun 1987;148:308-13.
57. Gershov D, Kim S, Brot N, Elkon KB. C-Reactive Protein Binds to Apoptotic Cells,
Protects the Cells from Assembly of the Terminal Complement Components, and Sustains
129
Chapter X
an Antiinflammatory Innate Immune Response. Implications for systemic autoimmunity. J

Exp Med 2000;192:1353-64.
58. Hack CE, Wolbink GJ, Schalkwijk C et al. A role for secretory phospholipase A2 and C-
reactive protein in the removal of injured cells. Immunol Today 1997;18:111-5.
59. Korb LC, Ahearn JM. C1q binds directly and specifically to surface blebs of apoptotic
human keratinocytes: complement deficiency and systemic lupus erythematosus revisited. J
Immunol 1997;158:4525-8.
60. Fadok VA, Bratton DL, Konowal A et al. Macrophages that have ingested apoptotic cells in
vitro inhibit proinflammatory cytokine production through autocrine/paracrine mechanisms
involving TGF-beta, PGE2, and PAF. J Clin Invest 1998;101:890-8.
61. Erwig LP, Gordon S, Walsh GM, Rees AJ. Previous uptake of apoptotic neutrophils or
ligation of integrin receptors downmodulates the ability of macrophages to ingest apoptotic
neutrophils. Blood 1999;93:1406-12.
63. Mitchell DA, Taylor PR, Cook HT et al. Cutting edge: C1q protects against the
development of glomerulonephritis independently of C3 activation. J Immunol
1999;162:5676-9.
64. Botto M, Dell'Agnola C, Bygrave AE et al. Homozygous C1q deficiency causes
glomerulonephritis associated with multiple apoptotic bodies. Nat Genet 1998;19:56-9.
65. Bickerstaff MC, Botto M, Hutchinson WL et al. Serum amyloid P component controls
chromatin degradation and prevents antinuclear autoimmunity. Nat Med 1999;5:694-7.
66. Taylor PR, Carugati A, Fadok VA et al. A hierarchical role for classical pathway
complement proteins in the clearance of apoptotic cells in vivo. J Exp Med 2000;192:359-
66.
67. Napirei M, Karsunky H, Zevnik B et al. Features of systemic lupus erythematosus in
Dnase1-deficient mice. Nat Genet 2000;25:177-81.
Rheum 1998;41:1241-50.
69. Kovacs B, Szentendrei T, Bednarek JM et al. Persistent expression of a soluble form of
Fas/APO1 in continuously activated T cells from a patient with SLE. Clin Exp Rheumatol
1997;15:19-23.
70. Mevorach D, Mascarenhas JO, Gershov D, Elkon KB. Complement-dependent clearance of
apoptotic cells by human macrophages. J Exp Med 1998;188:2313-20.
71. ter Borg EJ, Horst G, Limburg PC, van Rijswijk MH, Kallenberg CG. C-reactive protein
levels during disease exacerbations and infections in systemic lupus erythematosus: a
prospective longitudinal study. J Rheumatol 1990;17:1642-8.
72. Bootsma H, Geld YM, Geerligs H, et al. Serum amyloid P in systemic lupus erythematosus.
Arthritis Rheum 2000; 43 (abstract).
130
Pathogenesis of SLE
73. Steinbach F, Henke F, Krause B et al. Monocytes from systemic lupus erythematous
patients are severely altered in phenotype and lineage flexibility. Ann Rheum Dis
2000;59:283-8.
74. Savill J. Apoptosis. Phagocytic docking without shocking. Nature 1998;392:442-3.
75. Manfredi AA, Rovere P, Galati G et al. Apoptotic cell clearance in systemic lupus
erythematosus. I. Opsonization by antiphospholipid antibodies. Arthritis Rheum
1998;41:205-14.
76. Manfredi AA, Rovere P, Heltai S et al. Apoptotic cell clearance in systemic lupus
erythematosus. II. Role of beta2-glycoprotein I. Arthritis Rheum 1998;41:215-23.
131
Chapter XI
Summary and general discussion

Chapter XI
Introduction
This thesis studies several aspects of systemic lupus erythematosus dealing with the
central question why autoantibodies, responsible for the development of disease
manifestations, do occur and what the consequences of their presence are (summarised in
chapter I). The development of IgG autoantibodies, as a consequence of a break-through
of tolerance, seems to be associated with disturbances in the elimination of cells.
Elimination occurs by programmed cell death (apoptosis). Several mouse strains, as the
MRL-lpr/lpr (lymphoproliferation), and C3H/HeJ-gld/gld (generalised lymphoprolife-
rative disease) mice serve as the classical animal models for human SLE. These mice
spontaneously develop lymphadenopathy as well as autoimmune features characterised by
the presence of autoantibodies to nuclear antigens. It has been recognised that the
molecular basis of these abnormalities is a defect in Fas mediated apoptosis. The MRL-
lpr/lpr mice have a mutation in the Fas molecule that, in the proper genetic background,
results in the generation of abnormally spliced mRNA and a severe reduction in functional
Fas mRNA. The C3H/HeJ-gld/gld mice are deficient in functional Fas ligand (FasL) due
to a single base mutation. The defects in Fas and FasL, respectively, result in an
incomplete elimination of peripheral autoreactive cells as this elimination occurs
predominantly by Fas mediated apoptosis. Initially, based on these animal models, it was
supposed that defective apoptosis of, naturally occurring, autoreactive lymphocytes
resulted in the development of SLE like features. However, in human SLE mutations in
Fas or FasL have been found only incidentally.
Animal studies suggested that, next to the above mentioned mutations, incomplete
elimination of (autoreactive) lymphocytes might be due to the presence of soluble Fas
(sFas). Soluble Fas is the transcript of Fas mRNA lacking the transmembrane domain
because of the deletion of the exon encoding this region. Injecting sFas into female mice
of a certain mouse strain resulted in the development of autoimmune features due to
blocking of Fas induced apoptosis.
As a first approach to unravel disturbances in the elimination of apoptotic cells in human

SLE, in chapter II we evaluated whether levels of sFas are increased prior to a clinical
relapse. Assuming that increased levels of sFas, through blocking the Fas-FasL
interaction, hinder the induction of Fas-mediated apoptosis in peripheral blood
lymphocytes, we hypothesized that in SLE patients prior to a relapse sFas is already
elevated. This in turn would than result in the persistence of activated autoreactive
lymphocytes followed by an increase in the production of autoantibodies, finally
cumulating in a relapse of the disease. Indeed, sFas levels were elevated already 6 months
in advance in those patients going to have a relapse. Soluble Fas levels were constantly
elevated and might be indicative for the severity of an exacerbation. From this study it
could not be determined from which cells sFas originated and whether there is a relation
of sFas levels with activation markers on immune cells.
134
Therefore, in chapter III we evaluated the relation between sFas levels, disease activity,
and lymphocyte activation markers. We demonstrated that levels of sFas in patients with
SLE are increased, even during quiescent disease. Levels of sFas correlated with disease
activity scores and with the extent of activation of circulating B cells. These findings
suggest that sFas levels are a reflection of cell activation. Since elevated sFas levels have
been found in a considerable number of other, non-autoimmune diseases it can be
concluded that elevation of sFas is not specific for SLE or rheumatic diseases in general.
Furthermore, although sFas levels in SLE patients were elevated, the amounts of
circulating sFas were fairly low. Taken together, these data render a pathophysiological
role for sFas in SLE doubtful.
Blockade of Fas-mediated apoptosis induction by sFas in SLE does not seem to be a very
relevant factor in the etiopathogenesis of the disease. As a consequence it can be
speculated that in SLE a defective elimination of autoreactive lymphocytes is not
contributing to or even causing the disease. Nevertheless disturbances in apoptosis might
have pathophysiological consequences. Especially the finding that on apoptotic cells
autoantigens, whether or not modified, are presented had great impact on our
understanding of the disease. It can be hypothesized that the persistence of apoptotic cells,
either by increased production or by decreased clearance results in continuous
presentation of autoantigens so breaking tolerance. In line with this hypothesis it might be
assumed that in SLE increased apoptosis occurs because factors promoting apoptosis-
induction are present. This theory was supported by studies reporting increased
proportions of apoptotic lymphocytes and neutrophils in the peripheral blood of SLE
patients. To analyse the susceptibility of lymphocytes for apoptosis induction mediated by
Fas, we analysed membrane Fas expression in SLE patients and controls (chapter IV).
We showed that in SLE patients the percentage peripheral blood B-lymphocytes (CD19+)
expressing Fas was increased and was related to the state of lymphocyte activation and the
extent of disease activity. In addition, the results are compatible with the concept that
upregulation of membrane Fas renders B-lymphocytes more susceptible for apoptosis
resulting in increased rates of apoptotic peripheral blood lymphocytes as described in
SLE.
During the studies dealing with Fas expression on peripheral blood lymphocytes it was
noticed that Fas expression of smoking controls, especially on B cells, was higher
compared to that in non-smokers. Differences in Fas expression, as stated, might influence
susceptibility for Fas-mediated apoptosis. As alterations in the humoral immune response
and increased susceptibility for infections have been demonstrated in smokers, we
wondered whether smoking changed Fas expression and, subsequently, levels of apoptosis
after Fas-induced apoptosis. Indeed, the initial observation of increased Fas expression on
peripheral blood B-lymphocytes in smoking individuals could be reproduced in 10,
otherwise healthy, smokers (chapter V). Furthermore, we demonstrated that smoking was
135
Chapter XI
also associated with elevated percentages of Fas expressing CD4+ T-lymphocytes. In

addition, we showed that, using agonistic anti-Fas monoclonal antibodies, the Fas
pathway in smokers is functionally intact. These findings might suggest that activated
lymphocytes will be eliminated at a higher rate in smokers. The increase in Fas expression
on these lymphocyte subsets therefore might have consequences for the immune response
of smoking individuals. Furthermore, these findings can be of relevance interpreting data
of Fas expression and Fas-induced apoptosis of lymphocytes in various populations of
patients.
So far, we investigated whether sFas levels, smoking and activity of SLE could be of
influence in the induction of apoptosis by ligation of the Fas receptor. In chapter VI we
analysed whether there are intrinsic abnormalities in activation-induced and Fas-induced
apoptosis in SLE patients. To avoid the influence of disease activity only SLE patients
with inactive disease were included for this study. As expected, stimulation with anti-CD3
resulted in up-regulation of membrane Fas in patients and in controls. After proper up-
regulation of Fas, cells were incubated with anti-Fas. In vitro induction of apoptosis by
anti-CD3 as well as by anti-Fas occurred both in SLE patients and controls, and was
higher in SLE patients after incubation with both anti-CD3 as well as with anti-Fas. Fas
expression and in vitro induction of apoptosis therefore are increased in SLE even in the
absence of disease activity. As in SLE patients, even during inactive disease, elevated
proportions of activated peripheral blood lymphocytes can be demonstrated it seems
prudent to conclude that these results reflect increased in vivo activation of peripheral
blood lymphocytes and that no intrinsic defect in the apoptotic machinery of SLE patients
is present.
Antibody production in SLE is antigen-driven and T cell dependent. Activation of T cells

requires interaction of the T cell receptor (TCR) with the major histocompability complex
(MHC) molecule complexed with the antigenic peptides in the presence of additional
signalling through costimulatory molecules. Of these, the CD28-CD80/CD86 and CD40-
CD40L pathways are well known. The relevance of costimulation in the development of
autoimmune disease in animal models is supported by several studies in which
costimulation antagonists have been used as an effective approach to treat these diseases.
We hypothesized that, in SLE patients, an increase in the expression of costimulatory
molecules facilitates T- and B-cell activation and survival, increases autoantibody
production, and may so influence the development of autoimmune disease. In line with
this hypothesis we measured the expression of costimulatory molecules on peripheral
blood lymphocytes in lupus patients and healthy controls in relation to the state of
lymphocyte activation to address the following questions: is there an increase in
expression of costimulatory molecules in patients compared to controls and are these
changes related to the state of lymphocyte activation, disease activity and levels of
antibodies to dsDNA. Results are given in chapter VII. In SLE patients, the expression of
136
CD86 on CD19+ B cells was increased and was associated with disease activity, B cell
activation and levels of anti-dsDNA. We suppose that the increased CD86 expression will
render (autoreactive) B cells more susceptible for T cell-help. This might prolong survival
of activated B cells and will facilitate autoantibody production resulting in increased
amounts of autoantibodies produced.
Of the autoantibodies produced, those directed to nucleohistone and their components

(e.g. dsDNA) are strongly implicated in the pathogenesis of SLE. For example, these
antibodies can be eluted from kidney specimens of lupus mice with overt
glomerulonephritis. IgG class antibodies seem most relevant as there is a close relation
between levels of IgG anti-dsDNA and histologic activity scores in patients with lupus
nephritis. Furthermore, in the majority of patients, a renal relapse is preceded by a
significant rise of IgG anti-dsDNA.
All IgG subclasses can be found in kidney biopsies. Of the different subclasses, IgG1 and
IgG3 activate complement more efficiently than IgG2 while IgG4 does not activate
complement at all. The IgG subclass distribution of autoantibodies could therefore be of
relevance in the pathogenesis of lupus nephritis. In chapter VIII we describe the
evaluation of their possible nephritogenic role. We monitored levels of total IgG and IgG
subclasses of antibodies to nucleohistone and to dsDNA in time in patients who suffered a
renal relapse. IgG subclass distribution in kidney biopsies of patients with a renal relapse
was assessed as well. Serological data obtained from patients with renal relapses were
compared with those of lupus patients who developed an extra-renal relapse, in order to
evaluate the specificity of the findings for lupus nephritis.
IgG1- and IgG2-subclass antibodies to nucleohistone and to dsDNA were the predominant
subclasses found in plasma of lupus patients with renal disease. The frequent occurrence
of a rise of IgG2 anti-nucleohistone and IgG1 anti-dsDNA in patients prior to a renal
relapse suggests that, besides IgG1 subclass autoantibodies, IgG2-subclass antibodies to
nucleohistone have a particular pathophysiological role in lupus nephritis.
The important role that IgG2 autoantibodies to nucleohistone might play in the initiation
and outcome of lupus nephritis leads to the question whether the handling of the specific
IgG subclasses differs between SLE patients and controls and between the individual SLE
patients. Handling of IgG is accomplished by Fcγ receptors (FcγR). Several
polymorphisms of these receptors have been recognised, each with its own functional
consequences. In particular the polymorphism of the second Fcγ-receptor (FcγRIIa),
caused by the presence of either arginin or histidin at position 131 (FcγRIIa-R131 and
FcγRII-H131, respectively) determines the interaction with IgG2 and, to a lesser extent,
IgG3. In contrast to FcγRIIa-H131, the FcγRIIa-R131 isoform can not interact with IgG2
and has less affinity for IgG3. The FcγRIIIa-receptor has a valine (V) to phenylalanine (F)
change at amino acid position 158. Individuals homozygous VV for FcγRIIIa bind more
IgG1 and IgG3 compared with those homozygous FF for FcγRIIIa. Finally, the four amino
137
Chapter XI
acid changes termed neutrophil antigen polymorphism (NA1 and NA2) for FcγRIIIb have
consequences in the handling of IgG subclasses as individuals homozygous NA1
phagocytose IgG1 and IgG3 opsonized particles more efficiently than individuals
homozygous NA2 for FcγRIIIb. To evaluate associations between FcγR polymorphisms
and disease susceptibility we determined FcγR polymorphisms in a strictly Caucasian
SLE population and matched controls from the same region. Furthermore, we analysed
whether the respective FcγR polymorphisms were associated with specific disease
manifestations and whether these polymorphisms determined the clearance of immune
complexes in vivo. In chapter IX we show that of the known FcγR polymorphisms only
the R-H polymorphism of FcγRIIa is a relatively minor determinant of susceptibility to
SLE and affects the clearance of immune complexes in vivo. The V-F polymorphism of
FcγRIIIa influences the development of a set of clinical disease manifestations including
arthritis, serositis and hematological abnormalities.
In summary, no defects in the initiation of apoptosis or in the apoptotic machinery could

be found in human SLE. In contrast, susceptibility of peripheral blood lymphocytes for
(anti-CD3- and anti-Fas-induced) apoptosis in SLE seem to be increased, probably due to
their increased in vivo activation. Next to the increased production of apoptotic cells a
decreased clearance of these cells might be present in SLE. This combination can result in
the accumulation of apoptotic cells, subsequently resulting in the formation of
autoantibodies directed to the presented (nuclear) antigens. Among these, the antibodies to
nucleohistones and dsDNA are most important. Their production might be facilitated
through increased expression of CD86 on the cell membrane of B cells in SLE patients.
Of the autoantibodies produced the IgG1- and IgG2 subclasses are probably most
relevant. Their presence might have implications for the course of the disease as the
handling of the IgG subclasses is dependent on genetically determined Fcγ receptor
polymorphism. In chapter X the evolving concepts about the pathogenesis of SLE are
reviewed.
Crucial in research dealing with SLE are two questions. First, what induces autoimmunity,
that is what genetical factors predispose an individual to develop SLE and which
environmental factors are, in addition, necessary to break tolerance. Secondly, which
factors determine the very heterogeneous disease manifestations.
Concerning the first question many studies are nowadays being performed in which
genome scans of lupus patients are made to analyse which genetical factors might
predispose an individual for autoimmune diseases. Progress has been made and several
regions have been identified which contribute to disease susceptibility. Among these are
the HLA loci, the complement genes, and, possibly, Fcγ-receptors. It is still speculative,
but there are compelling data that the gene products of these loci might have a function in
(disturbed) elimination of apoptotic cells by phagocytosis and presentation of
autoantigens. At present, we are performing studies in SLE patients in which in vivo as
138
well as in vitro phagocytosis of apoptotic cells is analysed. The data from genome scans
gathered so far support the hypothesis that multiple genes are involved and that the
genetics of SLE resembles that of many other complex genetic diseases. Fine mapping
and candidate gene sequencing efforts in the chromosomal intervals identified are at
present performed in our own laboratory of medical genetics. Hopefully these studies will
lead to the identification of major genes and their respective products, predisposing to
human SLE. Furthermore, knowledge about these genetic factors makes it possible to
analyse much more specifically which environmental factors play an additional role in the
development of SLE. Among these, sunlight exposure and the use of certain drugs are
commonly accepted.
It is important to disclose whether prior to or at the moment SLE gets clinically manifest,
alterations in the immune system are present and can be demonstrated. This touches the
important dilemma how to interpret the immunological disturbances found in SLE
patients. Are they cause of the disease or caused by the disease? This is not easily
resolved. However, whenever it is possible to identify individuals at risk to develop SLE
analysis of alterations in the immune system already present before disease expression can
be evaluated. Many items dealt with in this thesis hopefully can be answered whenever
studies in persons susceptible for autoimmune disease can be performed. That is, are
increased levels of apoptotic cells present in the peripheral blood or in the tissues already
prior to the formation of autoantibodies? Are signs of lymphocyte activation and
disturbances in the expression of cell membrane molecules such as Fas, CD80 and CD86
already present in these individuals?
This brings up the second question: which factors determine disease manifestations? It is
suspected that also genetical factors might influence the course of the disease as we have
shown for the polymorphism of the Fcγ-RIIIa receptor. Probably, patients at risk to
develop severe organ involvement can be identified which gives the opportunity to treat
immunological disturbances in advance. Recently, the introduction of biologicals has been
a great therapeutical step forward in the treatment of many disorders, including
autoimmune diseases. Intervention with monoclonal antibodies will allow us in the near
future to intervene specifically, and to correct immunological disturbances also in patients
suffering from SLE.
In conclusion, increasing knowledge about the course of SLE in combination with the
availability of therapeutics which make intervention in the immune system much more
specific will profoundly change follow-up and treatment of patients with SLE. In the near
future major progress in our understanding of the pathogenesis of SLE is expected. This
will result not only in changes in treatment and follow-up of those patients already known
with SLE but will also have its impact on their family members. It is supposed that early
detection of family members at risk for developing autoimmune diseases will give us the
opportunity to intervene at an early phase thereby preventing (overt) disease.
139
Chapter XII
Nederlandse Samenvatting
(voor leken)
Chapter XII
Systemische lupus erythematodes (SLE) is een ziekte welke bij 1 op de 4000

Nederlanders voorkomt. De ziekte treft met name vrouwen in de leeftijdsfase van 18 tot
60 jaar en wordt gekenmerkt door de aanwezigheid van eiwitten (antistoffen) in het bloed
die gericht zijn tegen lichaamseigen (=auto) structuren. Deze antistoffen worden daarom
ook wel autoantistoffen genoemd. Ziekten waarbij deze autoantistoffen voorkomen
worden derhalve autoimmuunziekten genoemd. Binding van de autoantistoffen aan de
structuren van het eigen lichaam kan leiden tot een ontstekingsreactie, welke zich bij SLE
patiënten overal in het lichaam kan manifesteren.
Dat iemand antistoffen maakt is niet zo vreemd. Ons afweer (=immuun) systeem is hierin
gespecialiseerd. Door de vorming van antistoffen zijn wij in staat ziekten veroorzaakt
door bijvoorbeeld bacteriën te overwinnen. Doordat de antistoffen zich aan een bacterie
binden wordt deze bacterie vervolgens door andere cellen van het immuunsysteem
aangevallen en verwijderd. Het is echter wel vreemd dat iemand antistoffen maakt die
gericht zijn tegen het eigen lichaam. Het immuunsysteem is namelijk uitstekend in staat
om onderscheid te kunnen maken tussen het eigen lichaam ('zelf') en lichaamsvreemd
(‘niet-zelf’). Het is dan ook de vraag hoe deze ontsporing van het immuunsysteem kan
optreden.
Bij SLE patiënten lijkt het ontstaan van een stoornis in het immuunsysteem bepaald te
worden door een combinatie van factoren. Enerzijds zijn dat erfelijke factoren, waarvan
men aanneemt dat deze noodzakelijk zijn om de ziekte te kunnen ontwikkelen. Er is
echter meer dan een erfelijke aanleg nodig om de ziekte ook te krijgen. Zo krijgen
ééneiige tweelingen niet altijd beiden SLE. Ook komen in de familie van een patiënt met
SLE niet altijd andere patiënten met SLE voor. Er moeten dus andere factoren in het spel
zijn die er voor zorgen dat iemand, met een erfelijke aanleg voor SLE, de ziekte ook
daadwerkelijk krijgt. Van zonlicht en het gebruik van bepaalde medicijnen is inderdaad
aangetoond dat dit SLE kan uitlokken.
De vraag welke stoornis de afwijkingen van het immuunsysteem verklaart is tot op heden
niet beantwoord. De laatste jaren zijn er diverse aanwijzingen gevonden die er op duiden
dat de stoornis waarnaar gezocht wordt mogelijk te maken heeft met een stoornis in de
natuurlijke, geprogrammeerde dood van cellen. Deze geprogrammeerde celdood wordt
apoptose genoemd. De celdood door middel van apoptose is een proces wat een
gestructureerd beloop kent en als zodanig goed is te onderscheiden van de ongeordende
celdood (necrose) zoals die optreedt bij ernstige ongelukken of infecties. Apoptose wordt
gekenmerkt doordat de betreffende cel wat kleiner wordt, de celkern van aspect gaat
veranderen (de kern gaat condenseren) en het kernmateriaal bestaande uit DNA
zorgvuldig in kleine onderdelen wordt geknipt. Vervolgens worden deze geknipte brokjes
DNA en andere delen van de cel naar de buitenzijde van de cel getransporteerd. In kleine
pakketjes worden de onderdelen vervolgens van de cel afgescheiden net zolang tot de cel
in zijn geheel is verdwenen. De losse onderdelen worden door naburige cellen direct
142
Samenvatting
opgenomen (gefagocyteerd). Dit proces voorkomt dat allerlei schadelijke bestanddelen

welke zich in de cel bevinden kunnen vrijkomen hetgeen aanleiding zou geven tot een
ontstekingsreactie.
Apoptose is van groot belang voor een goede aanleg en ontwikkeling van het menselijk
lichaam. Daarnaast is apoptose essentieel om het menselijk lichaam op een goede wijze te
laten functioneren in evenwicht met de omgeving. Tijdens de aanleg- en
ontwikkelingsfase speelt apoptose een belangrijke rol in bijvoorbeeld de vorming van de
handen en de voeten. Doordat op bepaalde plaatsen cellen kunnen doorgroeien en op een
andere plaats cellen juist door apoptose ten gronde gaan ontstaat de vorm waarmee wij
worden geboren. In de aanleg van het immuunsysteem speelt apoptose eveneens een
belangrijke rol. In principe worden tijdens onze ontwikkeling een veelheid aan bepaalde
witte bloedcellen (lymfocyten) gevormd die nog moeten leren wat lichaameigen en wat
lichaamsvreemd is. Dit gebeurt doordat de lymfocyten die sterk binden aan lichaamseigen
structuren, de autoreactieve lymfocyten genaamd, door apoptose worden geëlimineerd.
Ook wanneer aanleg en ontwikkeling zijn voltooid blijft apoptose voor het
immuunsysteem van belang. Lichaamsvreemde structuren zoals een bacterie worden door
de lymfocyten ‘herkend’ indien de antilichamen welke aan de buitenzijde van de
lymfocyten aanwezig zijn zich kunnen binden aan eiwitstructuren, de antigenen genaamd,
op de buitenzijde van de bacterie. De binding tussen de lymfocyt en het antigeen leidt er
o.a. toe dat de lymfocyt wordt geactiveerd en antistoffen gaat produceren. Deze
antistoffen zijn gericht tegen de herkende antigenen en zijn nodig zijn om de bacterie te
verwijderen en te doden. Aangezien op het moment dat de lymfocyt een antigen herkend
de lymfocyt wordt geactiveerd wordt tegelijkertijd het apoptose mechanisme in gang
gezet. De lymfocyt zal dan binnen bepaalde tijd sterven. Hierdoor dooft de afweerreactie
weer uit waarmee het lichaam voorkomt dat deze respons zich ongebreideld zou
voortzetten.
Stoornissen in apoptose zouden in theorie op een tweetal wijzen kunnen leiden tot de
ontwikkeling van autoimmuniteit. Enerzijds kan men zich voorstellen dat cellen van het
immuunsysteem onvoldoende in apoptose gaan. Wanneer tijdens de ontwikkelingsfase (of
daarna) door een stoornis in apoptose de eliminatie van autoreactieve lymfocyten niet
plaats vindt zullen deze lymfocyten aanwezig blijven. De autoreactieve lymfocyten
kunnen vervolgens autoantistoffen gaan produceren waarna een autoimmuunziekte
ontstaat. Daarnaast kan men zich voorstellen dat er juist een overmaat aan apoptotische
cellen aanwezig is. Het immuunsysteem wordt dan als het ware gebombardeerd met een
grote hoeveelheid antigenen waardoor de drempel om antistoffen te gaan produceren
wordt overschreden en autoimmuniteit ontstaat. Een overmaat aan apoptotische cellen kan
zich voordoen doordat de productie ervan sterk is toegenomen en/of het opruimen (door
fagocytose) sterk is vertraagd.
Veel van de kennis over stoornissen van het immuunsysteem bij SLE is afkomstig uit
143
Chapter XII
onderzoek aan proefdieren. Een aantal muizenstammen ontwikkelt spontaan SLE. De

oorzaak hiervan bleek te berusten op een defect van de lymfocyten om in apoptose te
gaan. Daarnaast werd in een ander onderzoek aangetoond dat de aanwezigheid van
bepaalde oplosbare eiwitten (soluble Fas) apoptose kan remmen. Toevoegen van het
soluble Fas eiwit aan een (andere) muizenstam deed eveneens SLE-achtige ziekten
ontstaan. Overigens werd ook de tweede theorie, namelijk dat SLE het gevolg zou zijn
van een overmaat aan apoptotische cellen, gesteund door bevindingen uit
muizenmodellen. Indien muizen afkomstig uit een bepaalde muizenstam apoptotische
cellen krijgen ingespoten gaan deze dieren autoantistoffen produceren als teken van de
ontwikkeling van autoimmuniteit. Deze bevindingen zijn van grote invloed geweest op het
denken over de ontstaanswijze van SLE.
In dit proefschrift is dan ook onderzocht of er stoornissen zijn in de apoptose van

lymfocyten bij SLE patiënten. Voorts is onderzocht of er factoren aanwezig zijn welke de
drempel om autoantistoffen te gaan produceren verlagen. Tenslotte is onderzocht wat de
rol is van de autoantistoffen in de ontwikkeling van de voor SLE kenmerkende
ontstekingsreacties.
In hoofdstuk I worden de bovengenoemde theorieën besproken en wordt in het kort de

doelstelling van dit proefschrift toegelicht. De eerste hoofdstukken behandelen met name
de rol die stoornissen in apoptose zouden kunnen spelen in de ontwikkeling en het beloop
van SLE. In het tweede deel van het proefschrift wordt meer in gegaan op de wijze
waarop autoantistoffen een invloed kunnen hebben op het ziektebeloop van SLE
patiënten.
Een belangrijke stimulus voor cellen om in apoptose te gaan is de binding van het
zogenaamde Fas ligand (FasL) aan het Fas. Zowel het Fas als het FasL zijn eiwitten welke
op de buitenzijde van de cellen, de celmembraan, aanwezig zijn. Bindt nu een cel met het
FasL aan een andere cel welke Fas op de celmembraan draagt dan gaat de Fas dragende
cel door deze binding in apoptose. Van het Fas eiwit zijn oplosbare vormen beschreven
welke in het bloed van iedereen kunnen worden aangetoond. Dit oplosbare Fas eiwit
wordt soluble Fas (sFas) genoemd. Het is grotendeels identiek aan het Fas eiwit wat aan
de celmembraan is gebonden maar mist het ‘anker’ waarmee de binding aan de
celmembraan plaats vindt. Soluble Fas kan dus bijvoorbeeld ook het FasL binden hetgeen
de binding van het FasL met het Fas kan blokkeren. FasL wordt door de binding van het
sFas als het ware afgeschermd waardoor de binding met het membraan gebonden Fas niet
kan plaatsvinden. In theorie is het dus voorstelbaar dat indien sFas sterk is verhoogd de
apoptose van Fas dragende lymfocyten wordt geblokkeerd waardoor deze langer
overleven en bijvoorbeeld meer (auto)antistoffen kunnen produceren.
In hoofdstuk II wordt de betekenis van soluble Fas (sFas) onderzocht op het
ziekteverloop van SLE patiënten. In deze studie vonden wij inderdaad dat bij SLE
144
Samenvatting
patiënten tot zeker 6 maanden voordat de ziekte actief werd, een verhoging van het sFas
aanwezig is. Dit zou er op kunnen duiden dat reeds lang voordat een patiënt daadwerkelijk
bemerkt dat de ziekte actief is een ontsporing van het immuunsysteem aanwezig is. De
toegenomen hoeveelheid sFas zou de apoptose van (geactiveerde) lymfocyten kunnen
remmen en daarmee de ziekteactiviteit kunnen bevorderen.
In hoofdstuk III wordt de relatie tussen sFas spiegels in het bloed, de ernst van de
ziekteactiviteit en de mate van activatie van de lymfocyten onderzocht. Bij SLE patiënten
blijkt, in vergelijking tot gezonde controle personen, sFas verhoogd te zijn indien de
ziekte niet actief is. Bij toename van de ziekteactiviteit nemen ook de sFas spiegels toe.
Tevens blijkt de hoogte van de sFas spiegels te zijn gerelateerd aan de mate van activatie
van de B-lymfocyten. Deze lymfocyten zijn bij SLE patiënten het meest geactiveerd en
zijn verantwoordelijk voor de productie van de autoantistoffen. In andere studies is echter
aangetoond dat ook bij andere ziektebeelden dan SLE sFas verhoogd kan zijn. Daarnaast
is de hoogte van de sFas spiegels bij SLE patiënten niet dusdanig dat daaruit een
belangrijke rol voor sFas bij SLE mag worden verondersteld.
De hoogte van de sFas spiegels is niet dusdanig gebleken dat dit in het menselijk lichaam
zou kunnen leiden tot een verstoorde eliminatie van autoreactieve lymfocyten middels
apoptose. Het is echter niet uitgesloten dat stoornissen in de apoptose van cellen een rol
speelt in de ontwikkeling van SLE. Met name het gegeven dat op het oppervlak van
apoptotische cellen allerlei antigenen aanwezig zijn die normaal binnen in de cel zijn
opgeborgen, pleit voor een rol van apoptotische cellen bij het ontstaan van
autoimmuniteit. Het zijn immers deze antigenen waartegen bij SLE patiënten vele
autoantistoffen zijn gericht Momenteel wordt verondersteld dat een verhoogd voorkomen
van apoptotische cellen tot autoimmuniteit aanleiding kan geven. Een verhoogd aantal
apoptotische cellen kan ontstaan enerzijds doordat het aanbod is toegenomen, anderzijds
doordat de verwijdering is vertraagd.
In hoofdstuk IV is onderzocht of de lymfocyten bij SLE patiënten mogelijk gevoeliger
zijn om in apoptose te gaan hetgeen zou kunnen leiden tot een toename van apoptotische
cellen. Hiervoor is de membraan expressie van het eerder genoemde Fas eiwit gemeten bij
SLE patiënten aangezien de mate waarin dit eiwit aanwezig is de gevoeligheid voor
apoptose kan bepalen. Zowel op het moment van rustige ziekte als ten tijde van
ziekteactiviteit werd Fas op de celmembraan van lymfocyten gemeten. Deze resultaten
werden vergeleken met de Fas expressie bij gezonde personen alsmede met de mate van
lymfocytactivatie. In deze studie bleek dat bij SLE patiënten de Fas expressie inderdaad is
verhoogd op het oppervlak van B-lymfocyten. De B-lymfocyten zijn de producenten van
de (auto)antistoffen. Tevens bleek dat de Fas expressie is gerelateerd aan de mate van
lymfocytactivatie alsmede aan de mate van ziekteactiviteit.
145
Chapter XII
Als nevenbevinding bleek in deze studie dat bij rokende gezonde controle personen de Fas
expressie op lymfocyten duidelijk was verhoogd. Aangezien de Fas expressie gevolgen
kan hebben voor het in apoptose te gaan van cellen kan dit de afweer beïnvloeden. In
hoofdstuk V werd deze bevinding nader onderzocht. Dit temeer omdat bij rokers een
afgenomen afweer is waargenomen en een verhoogde vatbaarheid voor infecties. Voor dit
onderzoek werd bij rokers en niet-rokers bloed afgenomen waarin de mate van Fas
expressie op de lymfocyten werd bepaald. Onze eerste observatie, dat Fas expressie op de
B-lymfocyten van rokers is verhoogd, kon in deze studie inderdaad worden bevestigd.
Tevens bleek de Fas expressie bij rokers ook op andere lymfocyten, de zogenaamde CD4+
lymfocyten, te zijn verhoogd. Vervolgens werd onderzocht of deze lymfocyten in staat
zijn om door stimulatie met een anti-Fas eiwit ook in apoptose te gaan. Het anti-Fas eiwit
dat werd gebruikt, is een eiwit wat in staat is om het Fas eiwit wat op de celmembraan
aanwezig is te binden. Door deze binding wordt het Fas eiwit gestimuleerd hetgeen de
apoptose van de cel in gang zet. Deze route om apoptose te initiëren bleek bij rokende
individuen volledig intact. De toename van de Fas expressie op diverse subpopulaties
lymfocyten kan derhalve consequenties hebben voor de afweerreactie van rokers.
Uitgaande van de eerder beschreven bevinding dat de Fas expressie op de lymfocyten bij
SLE patiënten is verhoogd werd in hoofdstuk VI onderzocht of dit ook consequenties
heeft voor de gevoeligheid van lymfocyten van SLE patiënten om in apoptose te gaan.
Deze vraag werd beantwoord door de lymfocyten van SLE patiënten uit het bloed te
isoleren en vervolgens op te kweken door middel van diverse stimuli. De patiënten die aan
deze studie mee deden waren allen in een rustige fase van de ziekte. Na isolatie van de
lymfocyten werden de lymfocyten allereerst geactiveerd (met het zogenaamde anti-CD3
eiwit) om de Fas expressie verder te verhogen. Op het moment dat deze maximaal was, na
2 dagen, werd vervolgens anti-Fas toegevoegd om te beoordelen in hoeverre de
lymfocyten gevoelig waren voor deze stimulus om in apoptose te gaan. In deze
experimenten werd hetzelfde anti-Fas eiwit gebruikt als beschreven in hoofdstuk V.
Zowel de Fas expressie alsmede het percentage van de lymfocyten wat in apoptose was
gegaan werden direct na isolatie uit het bloed almede na de toevoegingen van
respectievelijk anti-CD3 en anti-Fas geanalyseerd. In deze studie bleek dat er bij SLE
patiënten, in vergelijking tot gezonde controle personen, een hoger percentage van de
lymfocyten in apoptose gaat zowel na activatie met anti-CD3 als na toevoeging van anti-
Fas. Aangezien ook in deze studie direct na isolatie van de lymfocyten bij SLE patiënten
de Fas expressie hoger bleek, ligt de veronderstelling voor de hand dat de verhoogde Fas
expressie de oorzaak is voor een toegenomen gevoeligheid om in apoptose te gaan.
In het volgende deel van het proefschrift worden studies beschreven gecentreerd rond de
rol van autoantistoffen bij SLE. Allereerst is onderzocht of er factoren aanwezig zijn
welke de drempel om autoantistoffen te gaan produceren verlagen. Voorts is bestudeerd
146
Samenvatting
wat de rol is van de autoantistoffen in de ontwikkeling van de voor SLE kenmerkende

ontstekingsreacties
Zoals beschreven worden de (auto)antistoffen door B-lymfocyten geproduceerd. De

antistof productie vindt echter pas plaats nadat de B-lymfocyten hiervoor hulp hebben
gekregen van de zogenaamde T-lymfocyten. Voor deze hulp is een intensieve interactie
tussen de T- en B-lymfocyten een vereiste. Deze interactie vindt plaats via zogenaamde
costimulatoire moleculen. De bekendste zijn CD28 en CD 40L op de T-lymfocyt welke
kunnen binden met respectievelijk CD80/CD86 en CD40 op de B-lymfocyt. Een
verandering in de expressie van een of meerdere costimulatoire moleculen bij SLE
patiënten kan leiden tot een intensievere interactie tussen T- en B-lymfocyten hetgeen de
productie van autoantistoffen kan doen toenemen. In hoofdstuk VII is onderzocht of de
expressie van genoemde costimulatoire moleculen bij SLE patiënten is veranderd. Wij
vonden dat de CD28, CD40, CD40L en CD80 expressie op de celmembraan van
lymfocyten bij SLE patiënten hetzelfde was als bij gezonde controle personen. Opvallend
was dat de expressie van CD86 op de B-lymfocyten van SLE patiënten sterk was
toegenomen. Deze expressie bleek gerelateerd aan de mate van B cel activatie en de
ziekteactiviteit van de SLE bij de patiënten. Aangezien er tevens een relatie bestond met
de hoeveelheid antistoffen tegen dsDNA lijkt de suggestie gerechtvaardigd dat de
toegenomen expressie van CD86 op de B-lymfocyten van SLE patiënten deze cellen
gevoeliger maakt voor de hulp van T-lymfocyten waardoor de autoantistof productie zal
toenemen.
De aanwezigheid van autoantistoffen kan leiden tot de ontwikkeling van

ontstekingsreacties op diverse plaatsen in het lichaam. Verondersteld wordt dat deze
ontsteking wordt veroorzaakt doordat de antistoffen zich direct binden aan bepaalde
structuren zoals die in organen als de nier aanwezig zijn. Anderzijds wordt verondersteld
dat antistoffen zich in het bloed reeds kunnen binden aan de eiwitstructuur waartegen ze
zijn gericht. Hierdoor ontstaat een in het bloed circulerend immuuncomplex wat
vervolgens in de organen kan neerslaan. De aanwezigheid van antistoffen in de organen
leidt tot een ontstekingsreactie. Dit komt omdat het lichaam de antistoffen en de cellen
waaraan deze gebonden zijn als lichaamsvreemd herkend en dus wil opruimen. Hiertoe
worden de antistoffen gebonden door Fc gamma-receptoren (FcγR) welke op de
celmembraan van vele cellen van het immuunsysteem aanwezig zijn. Binding van de
antistoffen aan de FcγR leidt tot een ontstekingsreactie. Deze binding is afhankelijk van
zowel het type antistof alsmede van het soort receptor. Op de verschillende typen
receptoren zal in het volgende hoofdstuk nader worden ingegaan. De meeste
autoantistoffen welke bij SLE voorkomen zijn gericht tegen eiwitten in de kern van de
cellen. Met name zijn de autoantistoffen gericht tegen het nucleohiston en de
afzonderlijke onderdelen van het nucleohiston. Het nucleohiston is opgebouwd uit de
histon eiwitten waarom de strengen DNA zijn gewikkeld. De antistoffen tegen het
147
Chapter XII
nucleohiston en zijn afzonderlijke componenten zijn immunoglobuline G (IgG)

antistoffen. Van dit IgG zijn een viertal subklassen bekend, elk met verschillende
eigenschappen met betrekking tot het vermogen een ontstekingsreactie te doen ontstaan.
IgG1 en IgG3 zijn hierin heel effectief, terwijl IgG2 dat minder doet, en IgG4 in het
geheel niet leidt tot een ontstekingsreactie. Indien bij een patiënt autoantistoffen aanwezig
zijn kan het dus van belang zijn te weten uit welke IgG subklassen deze IgG antistoffen
bestaan. Dit kan immers (mede) bepalen of, en in welke mate, een ontstekingsreactie zal
ontstaan. Wij veronderstelden dat er een verschil in de samenstelling van de IgG
subklassen van de autoantistoffen zou zijn tussen de SLE patiënten die ziekteactiviteit in
de nieren ontwikkelen en degenen die ziekteactiviteit buiten de nier ontwikkelen. Daartoe
werden de IgG subklassen van antistoffen tegen nucleohiston en van antistoffen tegen
dubbelstrengs DNA (dsDNA) maandelijks bepaald in het bloed van SLE patiënten in de
periode van 6 maanden voorafgaande aan ziekteactiviteit tot het moment van manifeste
ziekte. Tevens werden in de nierbiopten van de patiënten die een nierontsteking
ontwikkelden de IgG subklassen bepaald. De resultaten zijn beschreven in hoofdstuk
VIII. In deze studie bleek dat in de nierbiopten alle IgG subklassen kunnen worden
aangetroffen. In het bloed van de patiënten bleek dat bij degenen met een nierontsteking,
in vergelijking tot de patiënten die geen nierontsteking ontwikkelden, de IgG1 en IgG2
subklassen van de antistoffen tegen nucleohiston en dsDNA vaker aanwezig zijn. Tevens
bleken de IgG2 antistoffen tegen nucleohiston en de IgG1 antistoffen tegen dsDNA zeer
frequent te stijgen voordat er een nierontsteking ontstaat hetgeen kan duiden op een
belangrijke rol van deze subklassen bij de ontwikkeling van een nierontsteking.
Het feit dat naast IgG1 ook IgG2 blijkbaar een belangrijke rol speelt in de ontwikkeling
van nierontsteking bij SLE doet de vraag rijzen of er tussen de patiënten verschillen zijn
wat betreft de binding van de respectievelijke IgG subklassen aan de receptoren op
ontstekingscellen. Zoals beschreven worden de IgG antistoffen gebonden door FcγR. Nu
zijn er diverse FcγR bekend. Voorts zijn van elk van deze receptoren isovormen
beschreven. Elke isovorm heeft zijn eigen bindingseigenschappen wat betreft de binding
met de diverse IgG subklassen. De isovormen verschillen van elkaar door een enkele
verandering in de eiwit (aminozuur) samenstelling van deze receptoren. Dit wordt
polymorfismen genoemd. Meest uitgesproken is het polymorfisme voor de tweede FcγR
(FcγRIIa). Van deze receptor zijn een tweetal isovormen beschreven doordat op
aminozuur positie 131 een arginine of een histidine aanwezig kan zijn. De FcγRIIa met
een arginine op positie 131 kan geen IgG2 binden. Indien histidine op deze positie
aanwezig is kan IgG2 wel worden gebonden en is er tevens sprake van een sterkere
binding van IgG3 in vergelijking tot de vorm waarbij arginine op positie 131 zit. Het type
FcγR welke iemand heeft is erfelijk vastgelegd en zal de interactie met de antistoffen
bepalen. Indien iemand bijvoorbeeld (een combinatie van) FcγR heeft die de antistoffen
slecht kunnen binden kan men zich voorstellen dat dit leidt tot een verminderd vermogen
om deze antistoffen te verwijderen. Wellicht maakt dit iemand gevoeliger om SLE te
148
Samenvatting
krijgen. Voor SLE patiënten zou dit in theorie kunnen leiden tot een ander ziektebeloop.
Het is voorstelbaar dat bij SLE patiënten met een verminderd vermogen om antistoffen te
binden, en deze zo te verwijderen, hogere spiegels autoantistoffen aanwezig zijn. Dit kan
uiteindelijk leiden tot een toename van neerslagen van de autoantistoffen in de organen.
Juist deze SLE patiënten zouden dan bijvoorbeeld vaker een nierontsteking kunnen
ontwikkelen.
Deze complexe vragen werden onderzocht in hoofdstuk IX. Uit dit onderzoek bleek dat
alleen de tweede FcγR (FcγRIIa) van geringe invloed is op de gevoeligheid om SLE te
krijgen. Deze receptor kwam in de FcγRIIa R/R isovorm vaker voor bij SLE patiënten dan
bij gezonde controle personen. Dit is de isovorm met een arginine op positie 131 welke
antistoffen van de IgG2-, en in mindere mate van de IgG3-subklasse minder goed kan
binden. De tweede receptor bleek tevens van belang voor het opruimen van
immuuncomplexen. Daarentegen bleek uit dit onderzoek dat de derde FcγR (FcγRIIIa) de
ontwikkeling van bepaalde ziekteverschijnselen beïnvloed. SLE patiënten met FcγRIIIa-
F/F158 hadden vaker gewrichtsontstekingen, ontstekingen van het hart- of longvlies en
tekorten aan bloedcellen dan patienten met de FcγRIIIa-V/V158 isovorm.
Tenslotte wordt in hoofdstuk X een overzicht gegeven over de theorieën zoals die nu
aanwezig zijn met betrekking tot het ontstaan van SLE. In het ontstaan van SLE wordt aan
de aanwezigheid van apoptotische cellen een grote betekenis toebedeeld. De argumenten
hiervoor zijn in de eerste plaats dat gedurende het proces van apoptose allerlei
eiwitstructuren (antigenen), die normalerwijs in de cel zijn opgeborgen, op de
celmembraan tot expositie komen. Veel van deze eiwitstructuren blijken typische
doelwitten te zijn voor de antistoffen zoals die bij SLE patiënten worden aangetroffen. Dit
maakt voor het eerst inzichtelijk hoe het mogelijk is dat juist deze antistoffen kunnen
worden gevormd. In de tweede plaats is in dieronderzoek gebleken dat het inderdaad
mogelijk is de productie van autoantistoffen op te wekken door grote hoeveelheden
lichaamseigen apoptotische cellen toe te dienen.
Uit de in dit proefschrift weergegeven onderzoeken is gebleken dat bij SLE patiënten
weliswaar een apoptose blokkerend eiwit (sFas) in verhoogde spiegels in het bloed
aanwezig is doch dat deze spiegels waarschijnlijk onvoldoende zijn om het in gang zetten
van apoptose ook daadwerkelijk te blokkeren. Meer waarschijnlijk is bij SLE patiënten
sprake van het tegendeel en zijn er in het bloed juist meer apoptotische lymfocyten
aanwezig. Dit lijkt te worden veroorzaakt doordat de lymfocyten bij SLE patienten, zelfs
indien de ziekte volledig rustig is, meer zijn geactiveerd. Als uiting van deze activatie is
het Fas op de celmembraan van de lymfocyten in expressie toegenomen. Dit leidt tot een
verhoogde gevoeligheid van de lymfocyten om in apoptose te gaan. Daarnaast vonden wij
dat er bij SLE patiënten andere veranderingen zijn op de lymfocyten die er wellicht toe
leiden dat de productie van autoantistoffen is toegenomen. Tenslotte is gebleken dat de
interactie tussen de antistoffen en de FcγR, welke de autoantistoffen binden, van belang is
149
Chapter XII
voor het beloop van de SLE. In deze interactie lijkt zowel de IgG subklasse van de
autoantistof alsmede de soort FcγR een rol te spelen in de ontwikkeling van de
ontstekingsreacties bij SLE.
De beschreven onderzoeken beantwoorden een aantal vragen doch leiden, zoals vaak, tot
evenveel nieuwe. Inzicht verkrijgen in de ontstaanswijze en het ziektebeloop van SLE is
en blijft een zeer complexe materie. Desalniettemin zijn er de afgelopen jaren grote
vorderingen gemaakt en zijn in de komende jaren vele nieuwe ontwikkelingen en
inzichten te verwachten. Met name de ontwikkelingen op het gebied van de genetica en de
biotechnologie zijn zeer hoopgevend. Het ligt daarmee in de verwachting dat behandeling,
prognose, en kwaliteit van leven van SLE patienten verder zullen verbeteren.
150
Dankwoord
(Acknowledgements)
Dankwoord
Het gereedkomen van dit proefschrift stemt mij tot grote tevredenheid. De mogelijkheid
om de in dit proefschrift beschreven onderzoeken te verrichten werd mij geboden op de
onderafdeling Klinische Immunologie (hoofd: prof.dr. T.H. The), sinds kort geïntegreerd
met de onderafdeling Algemene Interne Geneeskunde (hoofd: prof.dr. R.O.B. Gans) in het
Academisch Ziekenhuis Groningen.
Dit proefschrift is tot stand gekomen door de inbreng van velen. In de eerste plaats wil ik
de patiënten bedanken voor hun niet aflatende bereidheid om mee te doen aan de
onderzoeken als zij daarvoor werden benaderd. Het is opvallend en bewonderenswaardig
hoe groot deze bereidheid is en het toont aan hoe betrokken patiënten en ook hun naasten
zijn.
Prof.dr. C.G.M. Kallenberg. Beste Cees. Wij werken ondertussen al jaren samen en dat is
voor mij een groot genoegen. Je gaf mij het vertrouwen en de ruimte de onderzoeken
zoals die in dit proefschrift staan vermeld uit te voeren. Het is voor mij niet te begrijpen
hoe je de vele taken die op je schouders rusten met elkaar weet te verenigen en altijd de
gelegenheid weet te vinden extra tijd vrij te maken. In korte tijd wist je manuscripten te
voorzien van nuttig commentaar en je deur stond altijd open voor overleg. Zoniet, dan
wist ik de deurkruk te vinden.
Dr. P.C. Limburg. Beste Piet. Ook jouw inbreng is voor het tot stand komen van dit
proefschrift van onschatbare waarde geweest. We kennen elkaar vanaf het moment dat ik
het lab betrad om te zoeken naar anti-histon, hetgeen ik overigens nooit heb mogen
vinden. In alle rust wist jij (bijna alle) misstappen te voorkomen en indien ze reeds waren
gemaakt, daar toch nog weer een positieve draai aan te geven. Ik bewonder je kennis, je
logische wijze van redeneren en met name de wijze waarop je anderen daar van laat
profiteren.
Gerda Horst heeft enorm veel werk verzet. Beste Gerda, ik waardeer het zeer de afgelopen
jaren met je te hebben mogen samenwerken. Het feit dat je paranimf wilde zijn zie ik als
een soort bekroning van deze samenwerking. Van jouw grote ervaring met de FACS heb
ik dankbaar gebruik gemaakt.
Ook de anderen op het lab wil ik bedanken voor de gezelligheid en prettige werksfeer die
jullie hebben geboden. Met name Hannie, Johan, Wia, Minke, Geke en Greet waren altijd
bereid om plasmamonsters op te sporen, plaatjes te scannen, de PC even af te staan.
Tevens wil ik de AIO’s van de klinische immunologie noemen. Agnieska, Beatriz, Eliane,
Maarten, Raoul dank ik voor hun inbreng tijdens de werkbesprekingen. Hilde, jij verdient
specifieke vermelding. Jouw enthousiasme en organisatorisch talent hebben mij meer dan
eens een duwtje in de rug gegeven.
154
Dankwoord
Kiki, het is heerlijk om met iemand zaken te kunnen doen die aan twee woorden genoeg
heeft en vervolgens alles in drie tellen perfect voor je regelt.
Daarnaast wil ik de (ex) collega’s bedanken voor hun inbreng. Dr. Peter Spronk was
degene die mij wegwijs maakte in databestanden en patiëntengegevens zoals die in de
loop der jaren in het AZG waren verzameld. Peter, de PC was even wennen maar
uiteindelijk heb ik veel gehad aan de gegevens die je mij hebt verstrekt. Lngklinex.xls
wordt nog steeds gebruikt. Dr. Hendrika Bootsma was en is nauw betrokken bij de zorg
van SLE patiënten Het is prettig om samen met je onderzoek te doen en ik ben blij dat
sinds kort de samenwerking in de vorm van de opgezette polikliniek systeemziekten is
geïntensiveerd. Je was altijd bereid te participeren, gegevens aan te dragen en mee te
denken.
Veel werk wat is verricht wordt nooit gepubliceerd. Nu horen onderzoeken die in de
prullenbak zijn beland dan ook niet in een proefschrift thuis. Zo heeft hoofdstuk ? het niet
gehaald. Desalniettemin wil ik dr. Harry van Goor bedanken voor de tijd en energie die hij
heeft gestoken in de kleuringen en analyse van de nierbiopten. Harry, de samenwerking
heb ik zeer op prijs gesteld. Ik hoop dat een deel van de data toch nog gebruikt kunnen
worden en dat we in de toekomst een vervolg aan onze samenwerking kunnen geven.
Prof.dr. L. A. Aarden en dr. R.J.T. Smeenk deelden en delen de interesse in apoptose

en autoimmuniteit. Zij stonden aan de wieg van de sFas hoofdstukken waaraan dr.
Thea van Lopik tevens een belangrijke bijdrage leverde.
Uiteraard dank ik de andere, niet allen met name te noemen, co-auteurs voor hun
inbreng.
Prof.dr. J.H.M. Berden, prof.dr. J.W.J. Bijlsma en prof.dr. M.H. van Rijswijk wil ik
bedanken voor hun bereidheid zitting te nemen in de beoordelingscommissie.
Siebren, als betrokken vriend wist jij dat promoveren niet vanzelf gaat. Je was/bent
opbeurend en onbezorgd. Jouw regelmatige bezoeken aan ons gezin heb ik niet alleen zeer
op prijs gesteld. Dank voor je bereidheid mij als paranimf bij te staan.
Last but not least: lieve Sandra, Martijn, Steven en Fabian. Jullie zijn voor mij ontzettend
belangrijk en dat wil ik op deze plaats nog eens benadrukken. Het boekje is af, de PC kan
nu nog meer gebruikt worden!
155
Curriculum Vitae
De schrijver van dit proefschrift is op 13 december 1961 te Arnhem geboren. Na behalen

van het VWO diploma startte hij in 1980 met de studie geneeskunde aan de
Rijksuniversiteit Groningen. In 1987 werd het artsexamen behaald. Hierna volgde de
militaire dienstplicht welke werd vervuld op de afdeling orthopedie van het Militair
Hospitaal A. Mathijsen in Utrecht. De wachttijd tot de start met de opleiding inwendige
geneeskunde werd vervuld als AGNIO op de afdeling Interne Geneeskunde van het
Academisch Ziekenhuis Groningen. In 1992 werd gestart met de opleiding tot internist in
het Medisch Centrum Leeuwarden (opleider dr. M.P. Leemhuis). In 1994 werd deze
opleiding voortgezet in het Academisch Ziekenhuis Groningen (opleider prof.dr. W.D.
Reitsma). Na voltooiing van de opleiding tot internist werd in juni 1996 gestart met de
opleiding reumatologie in het Academisch Ziekenhuis Groningen (opleider prof.dr. M. H.
van Rijswijk). In deze periode werd tevens begonnen met de in dit proefschrift beschreven
onderzoeken. In juni 1999 werd de opleiding reumatologie afgerond. Sindsdien is hij als
internist-reumatoloog werkzaam als staflid binnen de onderafdeling Algemene Interne van
het AZG (hoofd prof.dr. R.O.B. Gans). Hij is gehuwd met Alexandra Steensma en vader
van 3 zonen, Martijn, Steven en Fabian.
156

Thesis

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Thesis

Uploaded by

Copyright:

Available Formats

APOPTOSIS AND AUTOANTIBODIES IN

SYSTEMIC LUPUS ERYTHEMATOSUS

The publication of this thesis was financially supported by:

Apoptosis and autoantibodies in Systemic Lupus Erythematosus

Printed by Ridderprint B.V., Ridderkerk, The Netherlands

APOPTOSIS AND AUTOANTIBODIES IN

ter verkrijging van het doctoraat in de

Co-promotor: Dr. P.C. Limburg

Paranimfen: Gerda Horst

Chapter III: Do elevated levels of serum soluble Fas contribute to the

Chapter IV: Fas expression on peripheral blood lymphocytes of systemic

Chapter V: Effects of smoking on activation markers, Fas expression

Chapter VI: Anti-CD3-induced and anti-Fas-induced apoptosis in systemic

Chapter VII: Expression of costimulatory molecules on peripheral blood

Chapter IX: Fc receptor polymorphisms in systemic lupus erythematosus:

Chapter X: New insights into the pathogenesis of systemic lupus erythematosus

Chapter XI: Summary and conclusions 133

Chapter XII: Nederlandse samenvatting 141

Curriculum Vitae 156

Systemic lupus erythematosus

Systemic lupus erythematosus (SLE) is the prototype of a systemic autoimmune disease

Phagocytosis of apoptotic cells is a complex process in which many cells, membrane

Aim of the thesis

(autoreactive) lymphocytes, subsequently resulting in autoimmunity. In chapter II we

Patients with Systemic Lupus Erythematosus with

Thea van Lopik1

[32,33]. Soluble isoforms of cell surface receptors, generated either by proteolytic

Materials and methods

Generation of monoclonal antibodies against the Fas receptor.

induced apoptosis while CLB-CD95/2, CLB-CD95/6 and CLB95/19 prevented apoptosis

ELISA for soluble Fas.

Patients and plasma samples.

Criteria for major exacerbation: fulfilling one or more of the following *:

Criteria for minor exacerbation: fulfilling all of the following items:

Combination of CLB-CD95/2 as coating antibody with biotinylated CLB-CD95/6 as

Do elevated levels of serum soluble Fas contribute to

Thea van Lopik2

Sonja M.H.J. Jaegers3

Systemic lupus erythematosus (SLE) is characterised by generalised immune activation.

Systemic lupus erythematosus (SLE) is a systemic autoimmune disorder characterised by

Materials and Methods

Patients and controls

Criteria for major exacerbation: fulfilling one or more of the following *:

Criteria for minor exacerbation: fulfilling all of the following items:

Antibodies and staining procedure

Measurement of anti-dsDNA antibodies and sFas

developed with 100 µl substrate solution (0.1 mg/ml 3,5,3',5'-tetramethylbenzidine,

mAb specificity Label Recognised subsets Source

Number of patients with symptom present

controls quiescent disease active disease

Analysis of lymphocyte subsets

Relation between sFas and lymphocyte activation

is upregulated on activated cells. Alternatively, sFas can arise by concomitant

Fas expression on peripheral blood lymphocytes in

Systemic lupus erythematosus (SLE) is an autoimmune disease characterised by the

Patients and Methods

Patients and blood samples

Cell isolation and staining procedure

Criteria for major exacerbation: fulfilling one or more of the following *:

Criteria for minor exacerbation: fulfilling all of the following items:

% CD25 positive CD4+

percentage of Fas expressing B- 40 p=0.08

% HLA-DR positive CD8+