Inhibition of lupus disease by antidouble-stranded DNA antibodies of the IgM isotype in the NZB Ф NZWF1 mouse.

ARTHRITIS & RHEUMATISM
Vol. 52, No. 11, November 2005, pp 3629–3638
DOI 10.1002/art.21379
© 2005, American College of Rheumatology
Inhibition of Lupus Disease by Anti–Double-Stranded DNA
Antibodies of the IgM Isotype in the (NZB ⫻ NZW)F1 Mouse
Sonja Werwitzke,1 David Trick,1 Kenji Kamino,1 Torsten Matthias,2 Katja Kniesch,1
Brigitte Schlegelberger,1 Reinhold E. Schmidt,1 and Torsten Witte1
IgG and IgM and increased expression of Fc␥ receptor
were demonstrated in liver sections from the treated
mice compared with the untreated mice, suggesting an
enhanced clearance of soluble ICs from phagocytic cells
of the reticuloendothelial system.
Conclusion. These data demonstrate the efficacy
of IgM anti-dsDNA treatment in inhibiting the pathologic changes of lupus in (NZB ⴛ NZW)F1 mice. Lower
glomerular IC deposition is associated with a reduced
inflammatory response and impaired organ damage.
The reduced frequency of GN in SLE patients who have
IgM anti-dsDNA antibodies may therefore reflect a
disease-modifying effect of this class of autoantibodies
that has potential therapeutic implications. Our findings should encourage the development of new therapeutic modalities using IgM anti-dsDNA antibodies in
humans with SLE.
Objective. In systemic lupus erythematosus
(SLE), immune complexes (ICs) containing pathogenic
IgG anti–double-stranded DNA (anti-dsDNA) autoantibodies are deposited in renal capillaries and initiate
glomerulonephritis (GN) by the activation of complement and effector cells. In contrast, it has been demonstrated that the presence of IgM anti-dsDNA antibodies
correlates negatively with the development of GN in
SLE. The aim of this study was to determine whether
anti-dsDNA antibodies of the IgM isotype protect
against IC-mediated organ damage in SLE.
Methods. Lupus-prone (NZB ⴛ NZW)F1 mice
(females) were treated with murine monoclonal IgM
anti-dsDNA antibodies. Treatment was delivered by
subcutaneous injection at a dosage of 100 ␮g/week
starting at 16 weeks of age (prophylactic) or at 24 weeks
of age (therapeutic).
Results. Mice treated with IgM anti-dsDNA exhibited a delayed onset of proteinuria and a reduced
degree of renal pathology, which resulted in significantly improved survival as compared with control mice.
Serum concentrations of IgG anti-dsDNA antibodies
were not significantly modified. However, glomerular
deposition of ICs was markedly reduced in both treatment protocol groups. In contrast, higher amounts of
Autoantibodies against double-stranded DNA
(dsDNA) are a hallmark of systemic lupus erythematosus (SLE). The development of anti-dsDNA antibodies
of the IgG class are highly specific for SLE and are
therefore very important in the diagnosis and clinical
monitoring of the disease (1–3). Increasing antibody
titers are associated with disease exacerbation, especially
with the risk of glomerulonephritis (GN) (4–6). GN
requires aggressive immunosuppressive therapy, usually
with cyclophosphamide and corticosteroids (7–10). AntidsDNA antibodies of the IgG isotype are pathogenic,
since transfer of murine monoclonal IgG anti-dsDNA
antibodies or anti-dsDNA–producing hybridomas into
mice induces lupus-like GN (11–13). Moreover, IgG
anti-dsDNA antibodies and immune complexes are detectable in the glomeruli of patients with lupus nephritis
(14,15). Immune complexes (ICs) deposited in renal
capillaries result in a severe inflammatory response that
causes GN by activating the complement system and Fc␥
receptor–bearing effector cells (16–19). The circulating
Supported by Hochschulinterne Leistungsförderung from
Hannover Medical School and from the Germany Ministry of Education and Research (BMBF/DLR Kompetenznetz Rheuma C2.12).
1
Sonja Werwitzke, MD, David Trick, MD, Kenji Kamino,
MD, Katja Kniesch, Brigitte Schlegelberger, MD, Reinhold E.
Schmidt, MD, Torsten Witte, MD: Hannover Medical School, Hannover, Germany; 2Torsten Matthias: Aesku Diagnostics, Wendelsheim, Germany.
Dr. Matthias holds a patent for application of IgM dsDNA
antibodies in humans.
Address correspondence and reprint requests to Torsten
Witte, MD, Hannover Medical School, Department of Clinical Immunology (6830), Carl-Neuberg-Strasse 1, D-30625 Hannover, Germany.
E-mail: witte.torsten@mh-hannover.de.
Submitted for publication December 30, 2004; accepted in
revised form July 21, 2005.
3629
3630
WERWITZKE ET AL
ICs are either passively trapped within the glomerulus or
bind directly to antigens of the basement membrane
(20–22).
In contrast, the pathogenic relevance of antidsDNA antibodies of the IgM isotype has yet to be
elucidated. Low-affinity IgM autoantibodies are typically found in normal subjects, and transfer of autoantibodies of the IgM isotype rarely causes severe autoimmune disorders in murine models (23,24). It has been
further demonstrated that the absence of secreted IgM
leads to a significantly increased production of IgG
autoantibodies specific for dsDNA and histones, accelerated GN, and shortened lifespan in lupus-prone MLR/
lpr mice (25). The presence of IgM anti-dsDNA antibodies does not correlate with disease activity, and so
far, no clinical associations have been defined (26,27).
We and other investigators have previously demonstrated that IgM anti-dsDNA antibodies are negatively
correlated with the development of GN and all associated laboratory features (28,29). The negative correlation became more pronounced with increasing duration
of disease, and statistical analyses of elevated levels of
serum IgM, IgM rheumatoid factor, and IgM anticardiolipin antibodies did not show any influence on lupus
disease, especially on lupus nephritis, retrospectively.
Taken together, these data prompted us to examine the potential protective effect of IgM anti-dsDNA
antibodies in the well-established murine model of SLE
occurring in female (NZB ⫻ NZW)F1 mice. The observation that even low concentrations of IgM autoantibody
had a significant negative association with lupus nephritis in humans defined another encouraging rationale by
which to clarify the impact of IgM anti-dsDNA antibodies on the suppression of lupus disease. (NZB ⫻
NZW)F1 mice spontaneously develop an autoimmune
disease with genetic and clinical features that resemble
those of human SLE, including high concentrations of
IgG anti-dsDNA antibodies and IC-mediated GN (30).
In this study, we demonstrate a new therapeutic approach, the administration of IgM anti-dsDNA antibodies to lupus-prone (NZB ⫻ NZW)F1 mice, based on the
negative correlation seen in humans with SLE. Therapeutic efficacy was associated with a reduction in glomerular IC deposition, less severe inflammatory response and organ damage, delayed onset of proteinuria,
and improved survival.
MATERIALS AND METHODS
Mice and experimental design. Female (NZB ⫻
NZW)F1 mice ages 5–8 weeks were purchased from Harlan-
Winkelmann (Borchen, Germany). Mice were maintained in a
conventional animal housing facility at Hannover Medical
School. Treatment was started either at the age of 16 weeks,
before the expected onset of clinical disease (prophylactic
treatment), or at the age of 24 weeks, at the expected clinical
onset of organ manifestations of lupus-like disease (therapeutic treatment). Mice (n ⫽ 41 in the prophylactic treatment
group; n ⫽ 9 in the therapeutic treatment group) were injected
subcutaneously with 100 ␮g of murine monoclonal IgM antidsDNA antibody each week until the death of the mice.
Control mice received identical concentrations of murine
monoclonal IgM anti-human HLA–B27 as an irrelevant IgM
antibody (n ⫽ 15 in the prophylactic treatment group; n ⫽ 0 in
the therapeutic treatment group) or identical volumes of
phosphate buffered saline (PBS) (n ⫽ 37 in the prophylactic
treatment group; n ⫽ 12 in the therapeutic treatment group)
by subcutaneous injections.
The mean absolute level of urinary protein was comparable in each group at baseline. In the group receiving
prophylactic treatment beginning at week 16, the mean ⫾ SD
values were 21.8 ⫾ 3.9 mg/dl in those treated with IgM
anti-dsDNA antibody, 23.3 ⫾ 5.2 mg/dl in those treated with
the irrelevant IgM control, and 23.2 ⫾ 4.8 mg/dl in those
treated with PBS. In the group receiving therapeutic treatment
beginning at week 24, the mean ⫾ SD values were 61.1 ⫾ 36.9
mg/dl in those treated with IgM anti-dsDNA antibody and
59.2 ⫾ 36.0 in those treated with PBS.
Measurement of proteinuria. Urine was collected every 2 weeks and tested for proteinuria by a standard semiquantitative test using Bayer Multistix dipsticks (Bayer, Fernwald,
Germany). Results were graded according to the manufacturer’s instructions as either negative, slight ⫽ 15–20 mg/dl, ⫹ ⫽
30 mg/dl, ⫹⫹ ⫽ 100 mg/dl, ⫹⫹⫹ ⫽ 300 mg/dl, or ⫹⫹⫹⫹
⬎2,000 mg/dl of albumin. Nephritis was defined semiquantitatively as the presence of ⱖ300 mg/dl of proteinuria.
Purification and characterization of monoclonal IgM
antibodies. American Type Culture Collection hybridoma
HB-8329 secreting a murine monoclonal IgM anti-dsDNA
antibody and hybridoma HB-165 secreting an IgM anti-human
HLA–B27 antibody were purchased from Promochem (Wesel,
Germany). Hybridoma cells were cultured under fetal calf
serum–free conditions with insulin–transferrin–sodium selenite media supplement (ITS supplement; Sigma-Aldrich,
Taufkirchen, Germany). IgM antibodies were isolated from
the supernatant using high-performance liquid chromatography. Protein content was analyzed by spectrophotometry, and
activity of IgM anti-dsDNA antibody was determined by
enzyme-linked immunosorbent assay (ELISA) as described
below, using a peroxidase-labeled goat anti-mouse IgM detection antibody (1:1,000 dilution; Chemicon, Hampshire, UK).
Neither IgM antibody was cross-reactive with cardiolipin or
single-stranded DNA, as demonstrated by ELISA. Rheumatoid factor activity was also excluded using a global latex test
(31) (data not shown). Double-stranded DNA was not recognized by IgM anti–HLA–B27 antibodies. Both IgM antibodies
were dialyzed and sterile-filtered before injection.
Analysis of serum anti-dsDNA antibodies. Concentrations of autoantibodies against dsDNA were measured by
ELISA. Flat-bottomed 96-well plates coated with recombinant
human dsDNA, sample buffer, wash buffer, tetramethylbenzidine substrate, and stop solution (1M HCl) were provided by
INHIBITION OF LUPUS BY IgM ANTI-dsDNA ANTIBODIES
3631
Aesku Diagnostics (Wendelsheim, Germany). Detection antibody (horseradish peroxidase–conjugated goat anti-mouse
IgG) was purchased from Chemicon. Diluted probes (1:500
and 1:5,000; standard probes were serially diluted in sample
buffer in order to establish a standard curve) were incubated
for 1 hour at room temperature and, after 3 washing steps,
were subsequently incubated with detection antibody
(peroxidase-labeled goat anti-mouse IgG, dilution 1:1000 in
diluent buffer, Chemicon) for 15 minutes at room temperature. After 3 additional washing steps, 100 ␮l of tetramethylbenzidine substrate was added. After 15 minutes, the reaction
was stopped by adding 100 ␮l of stop solution (1M HCl).
Finally, optical density at 450 nm was determined with an
ELISA reader (Rainbow reader; SLT Labinstruments, Groding, Austria). Analyses were performed in duplicate. AntidsDNA titers are given as units per milliliter, using a reference
pool of sera from 29–38-week-old female (NZB ⫻ NZW)F1
mice.
Histopathologic and immunohistopathologic analyses.
Kidneys and other organs were removed and prepared for
analysis. Organs used for histopathology were fixed in formalin, embedded in paraffin, and sections were stained with
hematoxylin and eosin according to conventional protocols. At
least 2 sequential sections were obtained from each kidney.
Histopathologic changes were evaluated using a semiquantitative scoring system, where 1 ⫽ no lesions, 2 ⫽ minimal lesions,
3 ⫽ moderate lesions, and 4 ⫽ severe lesions.
Organs used for immunohistopathology were snapfrozen in tubes containing liquid nitrogen and stored at ⫺80°C.
Cryostat sections were prepared, and subsequently, the frozen
sections were air-dried and fixed in acetone for 10 minutes.
Glomerular IgG deposition was evaluated in 5-␮m–thick fro-
Figure 1. Incidence of proteinuria and death in female (NZB ⫻ NZW)F1 mice. Comparison of the cumulative incidences of proteinuria in a, mice treated prophylactically with IgM
anti–double-stranded DNA (anti-dsDNA) antibodies (n ⫽ 41) versus control mice treated
with phosphate buffered saline (PBS; n ⫽ 37) or with irrelevant IgM anti–HLA–B27
antibody (n ⫽ 15) and b, mice treated therapeutically with IgM anti-dsDNA antibodies (n ⫽
9) versus control mice treated with PBS (n ⫽ 12). The onset of proteinuria (defined as ⱖ300
mg/dl of albuminuria) was significantly delayed in both IgM anti-dsDNA antibody–treated
groups compared with their respective control groups (for prophylactic treatment, P ⬍ 0.05
versus PBS controls and P ⬍ 0.005 versus irrelevant IgM controls; for therapeutic treatment,
P ⬍ 0.05, by log-rank test). Comparison of the cumulative percentages of survival in c, mice
treated prophylactically with IgM anti-dsDNA antibodies and d, mice treated therapeutically with IgM anti-dsDNA antibodies compared with their respective control groups.
Differences were significant for both groups (for prophylactic treatment, P ⬍ 0.001 versus
PBS controls and P ⬍ 0.001 versus irrelevant IgM controls; for therapeutic treatment, P ⬍
0.05, by log-rank test).
3632
WERWITZKE ET AL
zen sections stained for 60 minutes at room temperature with
either 10% goat serum with horseradish peroxidase–labeled
goat anti-mouse IgG (Dianova, Hamburg, Germany) or 10%
rat serum with horseradish peroxidase–conjugated monoclonal
antibodies against IgG subclasses (rat anti-mouse IgG1 and rat
anti-mouse IgG2a; Chemicon). Complement component C3
was detected using rat anti-mouse C3 and horseradish
peroxidase–labeled goat anti-rat immunoglobulin as a secondary antibody (mouse adsorbed; Serotec, Wiesbaden, Germany). After washing, 3-amino-9-ethyl-carbazole substrate
(Sigma-Aldrich) was added for 15 minutes at room temperature. After additional washing, sections were stained with
hematoxylin. Kidney sections from BALB/c mice served as
negative control for immunohistochemical studies. IgG deposition was evaluated blindly by 3 independent observers and
was graded as follows: ⫹ ⫽ negative/weakly positive, ⫹⫹ ⫽
slightly positive, ⫹⫹⫹ ⫽ positive, and ⫹⫹⫹⫹ ⫽ strongly
positive.
Statistical analysis. Kaplan-Meier plots for proteinuria and survival rates were calculated by log-rank test.
RESULTS
Figure 2. Serum levels of IgG anti–double-stranded DNA (antidsDNA) antibodies in female (NZB ⫻ NZW)F1 mice. IgG antidsDNA antibodies in sera from control mice (■), mice treated
prophylactically with IgM anti-dsDNA antibody (F), and mice treated
therapeutically with IgM anti-dsDNA antibody (Œ) were measured by
enzyme-linked immunosorbent assay. Results were calculated from a
standard curve obtained using a reference pool of sera from 29–38week-old female (NZB ⫻ NZW)F1 mice. Control mice are those from
the prophylactic treatment arm treated with phosphate buffered saline,
since findings were not significantly different from those in the group
treated with irrelevant IgM anti–HLA–B27 antibody (analyzed up to
week 30). Each data point represents a single mouse. Horizontal lines
show the median.
Significantly delayed onset of proteinuria and
prolonged survival time in IgM anti-dsDNA antibody–
treated mice. To examine the potential protective effect
of IgM anti-dsDNA antibodies, female lupus-prone
(NZB ⫻ NZW)F1 mice were treated with murine monoclonal IgM anti-dsDNA antibodies. Mice that received
prophylactic treatment with IgM anti-dsDNA antibodies
from week 16, before the clinical onset of lupus (n ⫽ 41),
showed a significantly delayed onset of proteinuria as
compared with control mice that received either PBS
(n ⫽ 37; P ⬍ 0.05 by log-rank test) or irrelevant IgM
antibodies (n ⫽ 15; P ⬍ 0.005) (Figure 1). At the age of
32 weeks, 62% of mice in the PBS control group and
60% of mice in the irrelevant IgM control group had
developed proteinuria, as compared with 33% of the
IgM anti-dsDNA antibody–treated animals (Figure 1a).
As shown in Figure 1c, the delayed onset of proteinuria
in the IgM anti-dsDNA antibody–treated group translated to a significantly prolonged survival time as compared with the PBS control group (P ⬍ 0.001) and with
the irrelevant IgM control group (P ⬍ 0.001). By the age
of 36 weeks, ⬎50% of mice in the two control groups,
but only 15% of mice in the IgM anti-dsDNA antibody–
treated group, had died.
Mice that received therapeutic injections of IgM
anti-dsDNA antibodies from week 24, when organ manifestations of lupus-like disease are expected (n ⫽ 9),
had very similar results, with an 11-week delay in the
development of proteinuria compared with the PBS
control group (n ⫽ 12; P ⬍ 0.05). At the age of 31 weeks,
only 11% of mice in the IgM anti-dsDNA antibody–
treated group had developed proteinuria, as compared
with 50% of mice in the PBS control group (Figure 1b).
Kaplan-Meier analysis of cumulative survival time revealed 50% mortality in the PBS control group mice by
39 weeks, as compared with 48 weeks in the therapeutically treated group (P ⬍ 0.05), indicating therapeutic
efficacy of IgM anti-dsDNA antibody treatment even in
mice with active lupus (Figure 1d).
No change in serum concentrations of IgG antidsDNA antibodies after treatment with IgM anti-dsDNA
antibodies. IgG anti-dsDNA autoantibodies are known
to have a crucial pathogenic relevance for the initiation
of GN in (NZB ⫻ NZW)F1 mice. To determine whether
IgM anti-dsDNA antibody treatment alters the serum
concentrations of IgG anti-dsDNA antibodies, we evaluated the serum levels before and during treatment. As
expected, serum IgG anti-dsDNA antibody concentrations increased over time and peaked at the age of 34
weeks. Treatment with IgM anti-dsDNA antibodies did
not significantly modify the concentration of IgG antidsDNA antibodies in either the prophylactically treated
group or the therapeutically treated group compared
with control group (Figure 2). Levels of neither IgG1
nor pathogenic IgG2a anti-dsDNA subclasses were significantly different at week 34 (data not shown). Levels
of anti-dsDNA antibody of the IgM class were not
significantly different among the 3 groups before or
during treatment (data not shown).
Interestingly, the majority of control mice
INHIBITION OF LUPUS BY IgM ANTI-dsDNA ANTIBODIES
3633
showed very low serum concentrations of IgG antidsDNA antibodies at weeks 40 and 43, since mice with
high concentrations had already died. In contrast, IgM
anti-dsDNA antibody–treated mice still demonstrated
high serum levels of IgG anti-dsDNA antibodies at these
time points, which suggests an inhibition of disease
progression despite high levels of pathogenic IgG antidsDNA antibodies in the treated mice.
Suppression of GN and reduction of glomerular
IgG and C3 deposition in IgM anti-dsDNA antibody–
treated mice. Typical histologic characteristics of GN in
(NZB ⫻ NZW)F1 mice include vasculopathy, glomerular enlargement due to proliferative changes, and tubular damage. Kidney sections from nephritic control mice
(PBS-treated and irrelevant IgM–treated groups) and
from age-matched IgM anti-dsDNA antibody–treated
mice were examined at a mean ⫾ SD age of 42 ⫾ 2
weeks using hematoxylin and eosin staining and light
microscopy to perform a representative transverse analysis.
Kidney sections from the control groups revealed
typical signs of advanced GN, including glomerular
sclerosis, tubular dilatation and casts, focal crescent
formation, and mononuclear cell infiltration (Figures 3a
and c). In addition, vascular changes due to wall thickening and perivascular lymphocytic cell infiltration were
observed in control mice. In contrast, kidneys from IgM
anti-dsDNA antibody–treated mice remained almost
intact (Figures 3b and d), with a mean ⫾ SD histopathology score of 2.1 ⫾ 1.1 (n ⫽ 5), as compared with a
score of 3.6 ⫾ 0.9 (n ⫽ 5) in the PBS control group.
Glomerular sclerosis, tubular damage, and crescent formation were not observed in the IgM anti-dsDNA
antibody–treated mice. Consistent with these results,
periodic acid–Schiff–stained paraffin sections revealed
mesangial matrix expansion and enlargement of the
glomeruli in mice of the PBS and irrelevant IgM control
groups, as compared with normal features or only mild
changes in mesangial morphology in mice treated with
IgM anti-dsDNA antibody (results not shown).
Deposition of ICs, as indicated by staining for
IgG, IgG1, IgG2a, and C3, was markedly reduced in
treated mice compared with their respective control
groups at week 42 ⫾ 2 (mean ⫾ SD). Figure 4 shows
representative examples of IgG2a and C3 staining.
Mean ⫾ SD staining scores in the control group were
3.3 ⫾ 0.3 for IgG1 and 3.5 ⫾ 0.2 for IgG2a (n ⫽ 5).
Staining scores in the IgM anti-dsDNA antibody–treated
group were 1.8 ⫾ 0.7 for IgG1 and 2.3 ⫾ 0.4 for IgG2a
(n ⫽ 5). PBS and irrelevant IgM control groups showed
intense staining for IgG, IgG1, IgG2a, and C3 in all
mesangial areas, whereas kidney sections from the IgM
anti-dsDNA antibody–treated group exhibited slight or
no glomerular IgG deposition.
To evaluate the histologic characteristics of advanced GN, we examined prophylactically treated, therapeutically treated, and control mice with comparable
durations of longstanding proteinuria (11–13 weeks).
The histopathologic findings, such as glomerular enlargement, tubular damage, hyalinosis, and influx of
mononuclear cells, in mice treated with IgM antidsDNA antibody were similar to those in the control
mice, but deposition of glomerular immune complexes,
as indicated by staining for IgG, IgG1, IgG2a, and C3,
was still reduced. It is noteworthy that a small percentage of the glomeruli of mice treated with IgM antidsDNA antibody were showed negative or only minimal
staining of IgG and complement components. In contrast, all glomeruli of untreated animals showed intense
staining for immune complexes, and IgG-negative mesangial areas were not observed (results not shown).
To determine the histopathologic features of the
early stage of lupus, we compared kidney sections obtained from the different experimental groups at the
Figure 3. Hematoxylin and eosin staining of kidney sections from
control and IgM anti–double-stranded DNA (anti-dsDNA) antibody–
treated female (NZB ⫻ NZW)F1 mice (mean ⫾ SD age 41 ⫾ 2 weeks).
Kidney sections from a, an irrelevant IgM anti–HLA–B27 antibody–
treated control mouse and c, a phosphate buffered saline–treated
control mouse reveal advanced glomerulonephritis. Glomeruli (G) are
sclerotic, hypocellular, and partially surrounded by infiltration of
mononuclear cells (thick arrow). Note the epithelial proliferation in
Bowman’s space (thin arrow), the tubular damage, and the proteinaceous cast (asterisk). In contrast, the renal architecture is still normal
in kidney sections from mice treated b, prophylactically and d,
therapeutically with IgM anti-dsDNA antibody at comparable ages.
(Original magnification ⫻ 630.)
3634
WERWITZKE ET AL
Figure 4. Immunohistochemical analysis of the glomerular deposition of IgG2a and C3 in the kidneys of female (NZB ⫻
NZW)F1 mice (mean ⫾ SD age 41 ⫾ 2 weeks). Shown are representative histologic sections of kidneys from a and e, irrelevant
IgM anti–HLA–B27 antibody–treated control mice, b and f, phosphate buffered saline (PBS)–treated control mice, c and g,
mice treated prophylactically with IgM anti–double-stranded DNA (anti-dsDNA) antibody, and d and h, mice treated
therapeutically with IgM anti-dsDNA antibody at comparable ages. Sections were stained with antibodies against IgG2a and
C3 (red) and then counterstained with hematoxylin (blue). In sections from control mice (PBS-treated in b and f; IgM
anti–HLA–B27 antibody–treated in a and e), there is an intense granular staining in all mesangial areas of the glomeruli (G),
indicating extensive glomerular deposition of immune complexes. In contrast, in sections from the IgM anti-dsDNA–treated
mice, there is no, or only slight, mesangial staining for IgG2a and C3. (Magnification ⫻ 630.)
onset of nephritis. Therefore, 2 mice from the prophylactic treatment group (one treated with IgM antidsDNA antibody, the other with PBS) were killed within
3 weeks after developing proteinuria, and kidney sections were prepared and examined histologically. Both
mice demonstrated pathologic changes of early GN in
the form of lymphocytic cell infiltration and mesangial
proliferation on sections stained with hematoxylin and
eosin and with periodic acid–Schiff (data not shown).
However, glomerular deposition of IgG, IgG1, IgG2a,
and C3 was markedly reduced in tissues from the IgM
anti-dsDNA antibody–treated animal compared with the
PBS-treated mouse. Representative staining for IgG1
and IgG2a is shown in Figure 5. These data indicate a
reduction in the amount of ICs deposited despite high
levels of circulating IgG anti-dsDNA antibodies, as well
as a less severe inflammatory response occurring independently of the stage of disease, in mice treated with
IgM anti-dsDNA antibodies.
Expression of IgG, IgM, and Fc␥ receptors on
hepatic phagocytes in IgM anti-dsDNA antibody–
treated mice. The liver is the major site of removal of
soluble immune complexes from the circulation via Fc
and complement receptor–bearing cells, namely,
Kupffer cells and sinusoidal endothelial cells. We therefore examined IgG staining in the liver of mice ages
34–49 weeks. In the majority of liver sections from IgM
anti-dsDNA antibody–treated mice, there was staining
for IgG, IgG1, and IgG2a as well as IgM. Figures 6b, d,
and f show a representative example of this staining for
IgG1, IgG2a, and IgM, respectively. As expected, hepa-
Figure 5. Comparison of glomerular staining for IgG1 and IgG2a in
kidneys from female (NZB ⫻ NZW)F1 mice with early glomerulonephritis. Shown are histologic sections of kidneys from a and c, a
phosphate buffered saline (PBS)–treated control mouse and b and d,
a mouse treated prophylactically with IgM anti–double-stranded DNA
(anti-dsDNA) antibody at an early stage of glomerulonephritis (defined as proteinuria of 3 weeks’ duration). Sections from the PBStreated control mouse reveal intense staining for IgG1 and IgG2a in
the glomeruli (G), indicating deposition of immune complexes. In
contrast, in sections from the IgM anti-dsDNA antibody–treated
mouse, there is only slight staining for IgG1 and IgG2a. (Magnification ⫻ 630.)
INHIBITION OF LUPUS BY IgM ANTI-dsDNA ANTIBODIES
3635
monoclonal antibody 2.4G2. As shown in Figures 6g and
h, Fc␥ receptor expression was enhanced in liver sections from IgM anti-dsDNA antibody–treated mice. In
contrast, HLA–B27–treated mice showed weak staining
of Fc␥ receptors. Overall, hepatic IgG and IgM staining
patterns correlated with Fc␥ receptor expression.
Side effects. During the experimental course,
there was no clinical evidence of side effects from the
IgM anti-dsDNA antibody treatment. Hematoxylin and
eosin–stained histologic sections of other organs (spleen,
lung, heart, lymph nodes) did not reveal any morphologic alterations indicative of side effects. The development of an IgG alloantibody against the injected IgM
anti-dsDNA antibody was excluded (results not shown).
During treatment, the mean body weight of the animals
did not differ significantly between the treatment groups
and the control groups. At week 31, control mice
weighed a mean ⫾ SD of 40.4 ⫾ 4.7 gm, prophylactically
treated mice weighed 40.1 ⫾ 4.6 gm, and therapeutically
treated mice weighed 40.2 ⫾ 4.9 gm.
DISCUSSION
Figure 6. Immunohistochemical staining for IgG1, IgG2a, IgM, and
Fc␥ receptors in liver sections from control and IgM anti–doublestranded DNA (anti-dsDNA)–treated female (NZB ⫻ NZW)F1 mice
(mean ⫾ SD age 34 ⫾ 2 weeks). Shown are histologic sections of liver
from a, c, e, and g, mice treated with irrelevant IgM anti–HLA–B27
antibody (control) and b, d, f, and h, mice treated prophylactically IgM
anti-dsDNA. Sections were stained with horseradish peroxidase–
conjugated antibodies against IgG1, IgG2a, IgM, or Fc␥ receptor
(Fc␥R) (red) and with hematoxylin (blue). Sections from IgM antidsDNA antibody–treated mice revealed intense staining for IgG1,
IgG2a, IgM, and Fc␥ receptor in liver phagocytic cells (Kupffer cells
and liver sinusoidal endothelial cells). In contrast, sections from
control mice showed only weak staining for IgG1, IgG2a, IgM, and Fc␥
receptor in phagocytic cells. (Original magnification ⫻ 630.)
tocytes were negative for IgG and IgM, but phagocytic
liver cells revealed intensely positive staining. In contrast, the majority of liver sections from the IgM anti–
HLA–B27 antibody–treated control mice were only
weakly positive or were negative for IgG1, IgG2a, and
IgM (Figures 6a, c, and e), suggesting an altered or
enhanced clearance of ICs in IgM anti-dsDNA
antibody–treated mice.
Fc␥ receptor expression by IgG-positive and
IgM-positive hepatic phagocytes was studied using
Glomerulonephritis is one of the potential lifethreatening manifestations of SLE. It is usually treated
with cyclophosphamide, which often causes severe side
effects (32). Immune complexes consisting of dsDNA
and IgG anti-dsDNA antibodies seem to play a pivotal
role in the pathogenesis of lupus nephritis. They are
present in the glomeruli, they activate immune cells that
express Fc␥ receptors, and the presence and concentration of IgG antibodies against dsDNA correlate with the
prevalence of GN (33,34). The knockout of activating
Fc␥ receptors, which disrupts the interaction of cells
with ICs, can prevent GN in lupus-prone (NZB ⫻
NZW)F1 mice (35). There is evidence that IgG antidsDNA antibodies are pathogenic, especially for renal
tissue damage, because of their high efficiency for
complement fixation, high affinity for antigen, charge,
and cross-reactivity (14,36).
In contrast, the impact of IgM anti-dsDNA antibodies in the pathogenesis of SLE remains unresolved.
These antibodies do not correlate with disease activity or
specific clinical symptoms, they mostly exhibit low affinity, and they are also found in healthy individuals
(27,37). One study revealed that 53% of renal relapses
were preceded by rises in IgM anti-dsDNA antibody
levels, suggesting a pathogenic relevance of IgM immunoglobulins (38). However, this increase in IgM antidsDNA antibody levels was also accompanied by rising
concentrations of IgG anti-dsDNA antibodies and may
3636
therefore indicate a secondary phenomenon. Analysis of
a small number of patients with diffuse proliferative GN
or focal proliferative GN revealed relatively higher
concentrations of IgM anti-dsDNA antibodies in the
sera of patients with diffuse proliferative GN (39).
Nevertheless, results demonstrating the relationship between concentrations of IgG anti-dsDNA antibodies
from kidney eluates and the activity of lupus nephritis
strengthen the pathogenic role of IgG anti-dsDNA
antibodies in lupus nephritis (14,36,40,41).
There is indirect evidence that IgM antibodies
against dsDNA may be beneficial. Recently, investigators studying a murine knockout model of lupus GN
(MRL/lpr) that does not secrete IgM showed the accelerated development of IgG autoantibodies against
dsDNA and histones, more severe GN, more abundant
glomerular IC deposition, and a shortened lifespan
compared with their wild-type littermates (25). The
investigators tried unsuccessfully to restore the normal
development of GN in the mouse model with the use of
a monoclonal IgM antibody against dsDNA. As discussed in the report, the lack of success was probably
due to an IgG response against the injected IgM antibody. Mice with a deficiency in secreted IgM generated
on a normal background exhibit the development of IgG
anti-dsDNA antibodies with age or after repeated injections of bacterial lipopolysaccharide. The development
of severe GN has also been observed in secreted IgM–
deficient mice injected with lipopolysaccharide, suggesting that the absence of secreted IgM may predispose to
autoimmunity (42,43). Furthermore, it has been demonstrated that treatment with IgM-enriched intravenous
immunoglobulin (IVIG), in contrast to treatment with
IgG-containing IVIG, resulted in a diminished deposition of complement in the glomeruli and diminished
albuminuria in a rat model of anti–Thy-1 nephritis (44).
Thus, comparison of IgM-enriched and IgG-enriched
IVIG preparations clearly revealed a greater inhibitory
effect of IgM-containing IVIG on the deposition of
components of the classical pathway of complement,
both in vitro and in vivo.
Serum IgM anti-dsDNA antibodies have been
associated with less active disease and longer survival in
patients with SLE (45–47). We and other investigators
have demonstrated that the presence of IgM antidsDNA antibodies is negatively associated with the
development and severity of GN in SLE patients
(26,28,29). The onset of GN appeared to be predictable
based on the ratio of IgG to IgM anti-dsDNA antibodies. In the present study, we have proven that prophylactic and therapeutic treatment of (NZB ⫻ NZW)F1
WERWITZKE ET AL
mice with murine monoclonal IgM anti-dsDNA antibodies was indeed successful in protecting against the development of GN. The efficacy of IgM anti-dsDNA
antibody treatment was demonstrated by the reduced
glomerular IC deposition associated with impaired organ damage, the significantly delayed onset of proteinuria, and the prolonged survival.
Although the mechanism of action of IgM antibodies against dsDNA may not be entirely clear and was
not addressed in our study, it is likely that the antibodies
inhibit the formation of pathogenic ICs composed of
IgG and dsDNA by binding to and clearing the circulating dsDNA. Administration of IgM anti-dsDNA antibodies did not prevent the production of IgG antidsDNA antibodies. It is possible that high amounts of
ICs deposited in the glomeruli, as seen in the untreated
mice, might lead to reduced concentrations of circulating IgG anti-dsDNA antibodies in the serum. However,
it is apparent that IgM, as a polyvalent immunoglobulin
forming pentamer, results in different characteristics of
the ICs as compared with IgG. The deposition of
IgM-containing ICs might be altered and the trapping of
ICs in the glomeruli might be inhibited because of a
modified size and charge of the IgM ICs and, therefore,
may be reduced in the glomeruli of IgM anti-dsDNA
antibody–treated (NZB ⫻ NZW)F1 mice. Moreover, the
intense staining of IgG, IgM, and Fc␥ receptors observed in liver sections from treated mice might be the
result of an ongoing or even enhanced clearance of
soluble ICs by the hepatic reticuloendothelial system.
This might be induced by a modified pattern of deposited IgM ICs or, alternatively, by a different clearance
pathway of IgM-containing ICs through Fc␮ receptors
expressed on macrophages (48).
The possibility that IgM as a potent activator of
complement may contribute to an accelerated clearance
of ICs in (NZB ⫻ NZW)F1 mice is an attractive
hypothesis. IgM anti-dsDNA antibodies may bind to self
antigens, activate proteins of the complement system,
and as a consequence, promote the clearance of partially
IgG-containing ICs. Therefore, IgM ICs may contribute
to the restoration of the impaired clearance mechanism
through efficient activation of complement in murine
lupus.
A better understanding of the mechanism of the
therapeutic benefit of IgM anti-dsDNA antibodies may
provide novel approaches to the treatment of SLE
glomerulonephritis that are more specific and less toxic
than cyclophosphamide. These results should encourage
the generation of a human monoclonal IgM anti-dsDNA
INHIBITION OF LUPUS BY IgM ANTI-dsDNA ANTIBODIES
3637
antibody that may prove to be an efficient treatment of
SLE.
receptors for IgG complexes (hFc␥RIII-A) and the associated
Fc␧RI ␥-chain. J Immunol 1994;153:1281–92.
Shushakova N, Skokowa J, Schulman J, Baumann U, Zwirner J,
Schmidt RE, et al. C5a anaphylatoxin is a major regulator of
activating versus inhibitory Fc␥Rs in immune complex-induced
lung disease. J Clin Invest 2002;110:1823–30.
Uciechowski P, Schwarz M, Gessner JE, Schmidt RE, Resch K,
Radeke HH. IFN-␥ induces the high-affinity Fc receptor I for IgG
(CD64) on human glomerular mesangial cells. Eur J Immunol
1998;28:2928–35.
Pankewycz OG, Migliorini P, Madaio MP. Polyreactive autoantibodies are nephritogenic in murine lupus nephritis. J Immunol
1987;139:3287–94.
Hahn BH. Antibodies to DNA. N Engl J Med 1998;338:1359–68.
Madaio MP, Carlson J, Cataldo J, Ucci A, Migliorini P, Pankewycz
O. Murine monoclonal anti-DNA antibodies bind directly to
glomerular antigens and form immune deposits. J Immunol 1987;
138:2883–9.
Taki S, Hirose S, Kinoshita K, Nishimura H, Shimamura T,
Hamuro J, et al. Somatically mutated IgG anti-DNA antibody
clonally related to germ-line encoded IgM anti-DNA antibody.
Eur J Immunol 1992;22:987–92.
Pisetsky DS, Jelinek DF, McAnally LM, Reich CF, Lipsky PE. In
vitro autoantibody production by normal adult and cord blood B
cells. J Clin Invest 1990;85:899–903.
Boes M, Schmidt T, Linkemann K, Beaudette BC, MarshakRothstein A, Chen J. Accelerated development of IgG autoantibodies and autoimmune disease in the absence of secreted IgM.
Proc Natl Acad Sci U S A 2000;97:1184–9.
Okamura M, Kanayama Y, Amastu K, Negoro N, Kohda S,
Takeda T, 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.
Krippner H, Merle S, Jorgens K, Pirlet K. Antibodies to dsDNA
and ssDNA in the immunoglobulin classes IgG and IgM: prognostic value in the course of SLE. Z Rheumatol 1984;43:265–71.
Witte T, Hartung K, Sachse C, Matthias T, Fricke M, Deicher H,
et al, and the SLE Study Group. IgM anti-dsDNA antibodies in
systemic lupus erythematosus: negative association with nephritis.
Rheumatol Int 1998;18:85–91.
Forger F, Matthias T, Oppermann M, Becker H, Helmke K.
Clinical significance of anti-dsDNA antibody isotypes: IgG/IgM
ratio of anti-dsDNA antibodies as a prognostic marker for lupus
nephritis. Lupus 2004;13:36–44.
Wakeland EK, Liu K, Graham RR, Behrens TW. Delineating the
genetic basis of systemic lupus erythematosus. Immunity 2001;15:
397–408.
Katsuragawa H, Kanzaki H, Inoue T, Hirano T, Mori T, Rote NS.
Monoclonal antibody against phosphatidylserine inhibits in vitro
human trophoblastic hormone production and invasion. Biol Reprod 1997;56:50–8.
Klippel JH. Indications for, and use of, cytotoxic agents in SLE.
Baillieres Clin Rheumatol 1998;12:511–27.
Lefkowith JB, Kiehl M, Rubenstein J, di Valerio R, Bernstein K,
Kahl L, et al. Heterogeneity and clinical significance of glomerular-binding antibodies in systemic lupus erythematosus. J Clin
Invest 1996;98:1373–80.
Koffler D, Carr R, Agnello V, Thoburn R, Kunkel HG. Antibodies
to polynucleotides in human sera: antigenic specificity and relation
to disease. J Exp Med 1971;134:294–312.
Clynes R, Dumitru C, Ravetch JV. Uncoupling of immune complex formation and kidney damage in autoimmune glomerulonephritis. Science 1998;279:1052–4.
Winfield JB, Faiferman I, Koffler D. Avidity of anti-DNA antibodies in serum and IgG glomerular eluates from patients with
systemic lupus erythematosus: association of high avidity antina-
18.
REFERENCES
1. 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.
2. Hochberg MC. Updating the American College of Rheumatology
revised criteria for the classification of systemic lupus erythematosus [letter]. Arthritis Rheum 1997;40:1725.
3. Smeenk RJ, van den Brink HG, Brinkman K, Termaat RM,
Berden JH, Swaak AJ. Anti-dsDNA: choice of assay in relation to
clinical value. Rheumatol Int 1991;11:101–7.
4. Bootsma H, Spronk P, Derksen R, de Boer G, Wolters-Dicke H,
Hermans J, et al. Prevention of relapses in systemic lupus erythematosus. Lancet 1995;345:1595–9.
5. 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.
6. Weinstein A, Bordwell B, Stone B, Tibbetts C, Rothfield NF.
Antibodies to native DNA and serum complement (C3) levels:
application to diagnosis and classification of systemic lupus erythematosus. Am J Med 1983;74:206–16.
7. Klippel JH, Austin HA III, Balow JE, le Riche NG, Steinberg AD,
Plotz PH, et al. Studies of immunosuppressive drugs in the
treatment of lupus nephritis. Rheum Dis Clin North Am 1987;13:
47–56.
8. Balow JE, Austin HA III, Muenz LR, Joyce KM, Antonovych TT,
Klippel JH, et al. Effect of treatment on the evolution of renal
abnormalities in lupus nephritis. N Engl J Med 1984;311:491–5.
9. Boumpas DT, Austin HA III, Vaughn EM, Klippel JH, Steinberg
AD, Yarboro CH, et al. Controlled trial of pulse methylprednisolone versus two regimens of pulse cyclophosphamide in severe
lupus nephritis. Lancet 1992;340:741–5.
10. Contreras G, Pardo V, Leclercq B, Lenz O, Tozman E, O’Nan P,
et al. Sequential therapies for proliferative lupus nephritis. N Engl
J Med 2004;350:971–80.
11. Tsao BP, Ebling FM, Roman C, Panosian-Sahakian N, Calame K,
Hahn BH. Structural characteristics of the variable regions of
immunoglobulin genes encoding a pathogenic autoantibody in
murine lupus. J Clin Invest 1990;85:530–40.
12. Ohnishi K, Ebling FM, Mitchell B, Singh RR, Hahn BH, Tsao BP.
Comparison of pathogenic and non-pathogenic murine antibodies
to DNA: antigen binding and structural characteristics. Int Immunol 1994;6:817–30.
13. Ehrenstein MR, Katz DR, Griffiths MH, Papadaki L, Winkler TH,
Kalden JR, et al. Human IgG anti-DNA antibodies deposit in
kidneys and induce proteinuria in SCID mice. Kidney Int 1995;48:
705–11.
14. 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.
15. Grootscholten C, van Bruggen MC, van der Pijl JW, de Jong EM,
Ligtenberg G, Derksen RH, et al, for the Dutch Working Party on
Systemic Lupus Erythematosus. Deposition of nucleosomal antigens (histones and DNA) in the epidermal basement membrane in
human lupus nephritis. Arthritis Rheum 2003;48:1355–62.
16. Ravetch JV, Bolland S. IgG Fc receptors. Annu Rev Immunol
2001;19:275–90.
17. Radeke HH, Gessner JE, Uciechowski P, Magert HJ, Schmidt RE,
Resch K. Intrinsic human glomerular mesangial cells can express
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
3638
37.
38.
39.
40.
41.
42.
tive DNA antibody with glomerulonephritis. J Clin Invest 1977;59:
90–6.
Aotsuka S, Okawa M, Ikebe K, Yokohari R. Measurement of
anti-double-stranded DNA antibodies in major immunoglobulin
classes. J Immunol Methods 1979;28:149–62.
Bootsma H, Spronk PE, ter Borg EJ, Hummel EJ, de Boer G,
Limburg PC, 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.
Clough JD, Valenzuela R. Relationship of renal histopathology in
SLE nephritis to immunoglobulin class of anti-DNA. Am J Med
1980;68:80–5.
Koffler D, Schur PH, Kunkel HG. Immunological studies concerning the nephritis of systemic lupus erythematosus. J Exp Med
1967;126:607–24.
Bijl M, Dijstelbloem HM, Oost WW, Bootsma H, Derksen RH,
Aten J, et al. IgG subclass distribution of autoantibodies differs
between renal and extra-renal relapses in patients with systemic
lupus erythematosus. Rheumatology (Oxford) 2002;41:62–7.
Boes M, Prodeus AP, Schmidt T, Carroll MC, Chen J. A critical
WERWITZKE ET AL
43.
44.
45.
46.
47.
48.
role of natural immunoglobulin M in immediate defense against
systemic bacterial infection. J Exp Med 1998;188:2381–6.
Ehrenstein MR, Cook HT, Neuberger MS. Deficiency in serum
immunoglobulin (Ig)M predisposes to development of IgG autoantibodies. J Exp Med 2000;191:1253–8.
Rieben R, Roos A, Muizert Y, Tinguely C, Gerritsen AF, Daha
MR. Immunoglobulin M-enriched human intravenous immunoglobulin prevents complement activation in vitro and in vivo in a
rat model of acute inflammation. Blood 1999;93:942–51.
Pennebaker JB, Gilliam JN, Ziff M. Immunoglobulin classes of
DNA binding activity in serum and skin in systemic lupus erythematosus. J Clin Invest 1977;60:1331–8.
Sontheimer RD, Gilliam JN. DNA antibody class, subclass, and
complement fixation in systemic lupus erythematosus with and
without nephritis. Clin Immunol Immunopathol 1978;10:459–67.
Talal N, Pillarisetty RJ, DeHoratius RJ, Messner RP. Immunologic regulation of spontaneous antibodies to DNA and RNA I:
significance of IgM and IgG antibodies in SLE patients and
asymptomatic relatives. Clin Exp Immunol 1976;25:377–82.
Shibuya A, Sakamoto N, Shimizu Y, Shibuya K, Osawa M,
Hiroyama T, et al. Fc␣/␮ receptor mediates endocytosis of IgMcoated microbes. Nat Immunol 2000;1:441–6.