| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Medicine, Division of Rheumatology, and Departments of
Pathology,
Microbiology and Immunology, and
Epidemiology and Social Medicine, Albert Einstein College of Medicine, Bronx, NY 10461
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Anti-dsDNA Abs display molecular features of Abs arising in an Ag-specific response. Pathogenic anti-DNA Abs isolated from murine and human lupus are primarily IgG, they display high affinity for DNA, and exhibit higher than random replacement to silent mutation ratios in H and L chain V regions (5, 6). However, immunization with mammalian DNA fails to elicit lupus-like anti-DNA Abs, or clinical manifestations of the disease (7). Knowing the trigger for anti-DNA Abs in SLE is critical to our understanding of disease pathogenesis, but this has yet to be satisfactorily elucidated. To identify peptide Ags recognized by anti-dsDNA Abs, we screened a 10-mer phage peptide display library with R4A, a murine IgG2b anti-dsDNA Ab that is pathogenic and binds to renal glomeruli (8). After several rounds of selection, the peptide inserts displayed on the surface of the phage clones selected by R4A were determined. We found that phage-bearing peptides displaying a D/E W D/E Y S/G motif were bound by the Ab with the highest affinity. The most frequently R4A-selected phage clone displayed the DWEYSVWLSN peptide. Interestingly, the DWEYS peptide containing the R4A-specific consensus motif is 100% homologous to sequences in bacterial proteins from Haemophilus influenzae and Streptococcus pneumoniae. DWEYS as well as DWEYSVWLSN, purified peptides each containing the R4A-specific motif, inhibited the binding of R4A to dsDNA (8), while the D form of the peptide DNA mimic inhibited R4A deposition in renal glomeruli of SCID mice.
Several models of induction of anti-dsDNA Abs have been described in recent years, using various types of complexes of DNA and DNA-binding proteins. Nonautoimmune mice immunized with calf-thymus DNA in complex with Fus1, a DNA-binding protein derived from Trypanosoma cruzi, developed anti-dsDNA Abs (9). However, immunization with Fus1 alone was much less effective in inducing an autoimmune response, and glomerular Ig deposition was not induced (9, 10). Similarly, a single mutation in the polyoma BK virus T Ag, resulting in loss of its DNA-binding property, almost eliminated an anti-DNA response to immunization with the protein (11). In these models, therefore, a DNA-binding protein in complex with DNA is necessary for induction of autoimmunity, while the protein Ag alone without DNA is not sufficient. We explored a possible role of peptide DNA mimetopes in eliciting an anti-dsDNA Ab response by immunizing nonautoimmune BALB/c mice with multimeric DWEYSVWLSN in adjuvant (12). We found that at 3 mo of age, immunized mice displayed significantly higher titers of anti-dsDNA and other lupus-associated autoantibodies than unimmunized mice, and glomerular deposition of Ig was present. Unique features of this model of peptide-induced autoimmunity are: 1) the peptide used for immunization is a DNA surrogate rather than a DNA-binding protein, and 2) the finding that protein Ag alone without DNA is sufficient to induce a pathogenic anti-dsDNA response. Possible mechanisms by which an anti-peptide immune response can lead to the generation of autoantibodies include somatic mutation with generation of novel autospecificities, B cell Ag presentation of multimolecular nuclear complexes containing DNA and epitope spreading, and perturbation of the idiotypic network (12).
To address a possible role of peptide Ags in accelerating autoimmune manifestations and organ damage in genetically predisposed mice, and explore differences in B cell responses to the peptide dsDNA mimetope between autoimmune and nonautoimmune mice, we immunized (NZB x NZW)F1 (B/W) mice with the peptide and analyzed the peptide-specific and autoantibody responses.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The derivation of DWEYSVWLSN has been described in detail previously (8). SVIWSWMWLD and TIALKWLRWA, 10-mer peptides from the same phage library, were used as controls. Multimeric peptides were prepared for immunization by synthesis on an eight-branched, polylysine backbone (multiple antigenic peptide (MAP); Research Genetics, Huntsville, AL).
B/W mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Immunizations were performed, as previously described (12), with slight modifications. Thirteen-week-old female mice (n = 14) were immunized s.c. with 100 µg of MAP-DWEYSVWLSN emulsified 1:1 in CFA H37 Ra (Difco, Detroit, MI), followed by a boost of 100 µg of MAP-DWEYSVWLSN in IFA s.c., 3, 6, and 9 wk later. Serial bleeds were obtained at baseline (before the first immunization), and at wk +3, +5, +7, +10, +13, and +16. Control mice received MAP-TIALKWLRWA (n = 4) or MAP-SVIWSWMWLD (n = 5) using the same schedule of adjuvant and immunizations.
Generation of hybridomas
Thirteen-week-old B/W mice (n = 3: B/W1, B/W2, B/W3) were immunized with 100 µg of MAP-DWEYSVWLSN in CFA on day 0, and boosted twice with 100 µg of MAP-DWEYSVWLSN in IFA. Ten days after the last boost, the mice were sacrificed and the splenocytes were fused to NSO myeloma cells at a 1:1 ratio using established methodology (13, 14). Spleens from two age-matched unmanipulated female B/W mice at 20 wk of age were separately fused as controls. Supernatants from hybrid-containing wells were screened for binding to peptide and dsDNA by ELISA, as described below. Cells from positive wells were cloned by limiting dilution.
Ab purification
IgM mAbs were purified from hybridoma supernatants on an immobilized mannan-binding protein column, using the ImmunoPure IgM purification kit (Pierce, Rockford, IL), and the manufacturer’s instructions. IgG mAbs were purified from hybridoma supernatants on a protein G column from Amersham Biosciences (Piscataway, NJ).
ELISAs
Salmon sperm dsDNA (Calbiochem Novabiochem, La Jolla, CA) was purified by filtration with a 0.45-µm filter (Millipore, Bedford, MA), and adsorbed to Immulon II 96-well microtiter plates (Dynatech Laboratories, Chantilly, VA) at a concentration of 100 µg/ml in PBS. For the ssDNA ELISA, a solution of salmon sperm DNA (Calbiochem Novabiochem) in PBS was boiled for 15 min, cooled rapidly on ice, and adsorbed to Immulon II plates at a concentration of 100 µg/ml in PBS. The dsDNA and ssDNA plates were dried overnight at 37°C. Before blocking, excess DNA was removed with a 4-min soak in distilled water. Cardiolipin (Fluka, Ronkonkoma, NY) at a concentration of 75 µg/ml in ethanol was adsorbed to Immulon II plates at room temperature overnight. MAP-DWEYSVWLSN (Research Genetics) at 20 µg/ml in PBS, Sm/RNP (Immunovision, Springdale, AR) at 10 µg/ml in PBS, total histones at 10 µg/ml in PBS (Roche, Indianapolis, IN), and laminin at 2 µg/ml in PBS (Sigma-Aldrich, St. Louis, MO) were adsorbed to Immulon II plates at 4°C overnight. BSA (Roche), cytochrome c, lysozyme, and keyhole limpet hemocyanin (Sigma-Aldrich) all at 10 µg/ml in PBS were adsorbed to Immulon II plates at 4°C overnight. Plates (for all Ags) were then blocked with 3% FBS (Gemini Bio-Products, Calabasa, CA) in PBS for 1 h at 37°C, and incubated with serum at a 1/300 dilution, or hybridoma supernatant for 2 h at room temperature. Plates were washed five times with PBS-0.05% Tween 20 with an M96 ELISA washer (Titertek, Huntsville, AL), and alkaline phosphatase-conjugated goat anti-mouse IgM, IgG, or IgG subclasses (Southern Biotechnology Associates, Birmingham, AL) diluted 1/1000 in 3% FBS/PBS were added for 1 h at 37°C. The ELISAs were developed by adding the alkaline phosphatase substrate p-nitrophenyl phosphate (Sigma-Aldrich), and the OD was monitored at 405 nm using a MRX Revelation ELISA Reader (DYNEX Technologies, Chantilly, VA).
For titration experiments, serum at an initial concentration of 1/200 in PBS was serially diluted on dsDNA-coated plates and the ELISA was continued, as described above. The serum was reported as positive at the highest dilution at which the OD was higher than the mean plus 3 SDs of five control sera (3-mo-old preautoimmune female B/W mice) at that dilution, and the titer is given as 1/dilution. For inhibition ELISAs, serum at a dilution resulting in 50% of maximal DNA binding or mAbs at 10 µg/ml was preincubated with a single concentration (125 µmol) (for serum) or serial dilutions (for mAbs) of MAP-DWEYSVWLSN for 1 h at 37°C. The serum-peptide solution was then transferred to preblocked, dsDNA-coated plates, and the ELISA was continued as described above.
Culture supernatants from cloned hybridomas were quantitated and isotyped by ELISA using purified monoclonal mouse Ig from Sigma-Aldrich. The isotypes of selected hybridomas were confirmed using the Isostrip isotyping kit (Roche). Normalized supernatants at 10 µg/ml were assayed for reactivity with the different Ags described above, using isotype-matched mAbs from Sigma-Aldrich (IgM, TEPC 183; IgG1, MOPC 31C, MOPC 21; IgG2a, UPC 10; IgG2b, MOPC 141; IgG3, FLOPC 21), Southern Biotechnology Associates (IgM, 11E10; IgG1, 15H6; IgG2a, HOPC-1; IgG2b, A-1; IgG3, B10), and Accurate Chemical and Scientific (Westbury, NY) (IgG2a, CBL601) as negative controls. At least two control Abs for each isotype were used to define the criteria for positivity, which was an OD higher than the mean of the control isotype-matched Abs plus 2 SDs. Positive clones were selected for sequencing.
Quantitation of the Ig subclasses in the anti-peptide and anti-dsDNA Ab responses was performed as described (12). Briefly, anti-peptide and anti-dsDNA ELISAs on serial dilutions of mouse sera starting at 1/50 were separately developed with alkaline phosphatase-linked Abs to mouse IgM, and to mouse IgG subclasses (IgG1, IgG2a, IgG2b, IgG3) from Southern Biotechnology Associates. The Ig concentration was calculated for each isotype from a standard curve generated concurrently using purified mouse myeloma proteins and the alkaline phosphatase-linked anti-IgM and anti-IgG subclasses described above.
RNA isolation and PCR of Ig genes
RNA isolation and PCR of Ig genes were performed as previously
described (13). Total RNA was extracted from 2 x
107 hybridoma cells, and reversed transcribed
using the Superscript Pre-Amplification System (Life Technologies,
Gaithersburg, MD), and an oligo(dT) primer. PCR of the 
L chain was
performed using a 3' C region primer and a set of seven 5'
degenerate L chain primers (15). PCR of the H chain
was performed using a set of 10 degenerate 5' H chain V region primers,
and a 3' IgM C region primer or 3' universal IgG primer for IgM and IgG
H chain V regions, respectively (15). PCR amplifications
were conducted in a PerkinElmer 9700 thermal cycler (PerkinElmer, Palo
Alto, CA) for 35 cycles, with a hot start at 94°C for 3 min,
denaturation at 94°C for 50 s, annealing at 50°C for 50
s, and extension at 72°C for 40 s, followed by a final extension
at 72°C for 10 min. PCR products were purified using the QIAquick PCR
purification kit (Qiagen, Valencia, CA). Ambiguities in nucleotide
sequence were resolved by cloning selected PCR products using the TOPO
TA Cloning kit (Invitrogen, Carlsbad, CA), and plasmid sequencing using
the M13 (-20) primer.
Sequencing and sequence analysis
Automated sequencing of PCR products was performed at the DNA sequencing facility of the Albert Einstein College of Medicine using an Applied Biosystems 377 sequencer (Applied Biosystems, Foster City, CA). Generated sequences were compared with the GenBank data bank using the Advanced Blast and IgBlast search programs from the National Center for Biotechnology Information, National Institutes of Health.
Renal histopathology
One kidney from each mouse was obtained at sacrifice on wk +16, and embedded in paraffin. Histological sections were obtained by microtome, and stained with H&E and periodic-acid Schiff (PAS) using standard methods. Kidney histology was evaluated blindly by an experienced pathologist (M. Edelman), without knowledge of the treatment assignment. Histopathological changes in renal tissue were quantitated using the scoring method described by Corna et al. (16), with slight changes. Presence of endocapillary disease was scored from 0 to 3 (0, absent; 1, mild; 2, moderate; 3, severe). The presence of crescents, tubular changes (atrophy, casts, dilatation), and interstitial changes (inflammation, fibrosis) was each graded on a scale of 0–3, with 0, absent; 1, present in <25% of the section; 2, present in 25–50% of the section; and 3, present in >50% of the section. The maximum score for each mouse was 12 (maximum 3 points each for endocapillary glomerular changes, crescent formation, tubular changes, and interstitial damage).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To investigate a possible role of peptide Ag in accelerating
autoantibody production in genetically predisposed mice, we immunized
preautoimmune B/W mice with a peptide dsDNA surrogate (DWEYSVWLSN), and
compared the induced responses with mice immunized with control
peptides (SVIWSWMWLD and TIALKWLRWA), and with age- and sex-matched
unmanipulated B/W mice. As there were no significant differences
between the mice immunized with the different control peptides, these
two groups are considered together as MAP-control for the purpose of
the analysis. The IgM anti-peptide response began to appear after
the second immunization, at wk +3. All MAP-DWEYSVWLSN-immunized mice
developed an IgG anti-DWEYSVWLSN response, which was significantly
higher than control-immunized mice at wk +5 (p
< 0.025), wk +10 (p < 0.0001), and wk +16
(p < 0.0001) (Fig. 1
A). Mean IgG
anti-cardiolipin titers in MAP-DWEYSVWLSN-immunized mice were
significantly higher than control-immunized mice at wk +10
(p = 0.0009) and wk +16
(p = 0.02) (Fig. 1B). Mean IgG
anti-dsDNA Ab titers in MAP-DWEYSVWLSN-immunized mice were
significantly higher than control-immunized mice at wk +10
(p = 0.009), and wk +16
(p = 0.02) (Fig. 1C). Mean
anti-laminin titers in MAP-DWEYSVWLSN-immunized mice were
significantly higher than control-immunized mice at wk +5
(p = 0.04), wk +10 (p =
0.005), and wk +16 (p = 0.03) (Fig. 1D). Although anti-laminin titers in
MAP-DWEYSVWLSN-immunized mice were higher at baseline, there was no
difference between the groups at wk +3, after which the
anti-laminin titers began to diverge. IgG anti-histone titers
did not differ significantly between MAP-DWEYSVWLSN-immunized and
control mice; IgG anti-Sm/RNP Ab titers were higher in the
MAP-DWEYSVWLSN-immunized mice only at a single time point (wk +13)
(data not shown).
|
E). IgG
anti-peptide titers in MAP-DWEYSVWLSN-immunized B/W mice were
significantly higher than in MAP-control-immunized and unmanipulated
B/W mice (p < 0.00001). IgG anti-dsDNA Ab
levels in MAP-DWEYSVWLSN-immunized B/W mice (by OD) were significantly
higher than MAP-control-immunized B/W mice (p =
0.04). IgG anti-dsDNA Ab levels in MAP-DWEYSVWLSN-immunized B/W
mice were also higher than in unmanipulated B/W mice, but the
difference did not reach statistical significance
(p = 0.114). IgG anti-cardiolipin titers in
MAP-DWEYSVWLSN-immunized B/W mice were significantly higher than in
MAP-control-immunized (p = 0.0009) and
unmanipulated B/W mice (p = 0.02). Finally, IgG
anti-laminin titers in MAP-DWEYSVWLSN-immunized B/W mice were
significantly higher than in MAP-control-immunized
(p = 0.008) and unmanipulated B/W mice
(p = 0.016).
The difference in IgG anti-dsDNA Ab levels between the
MAP-DWEYSVWLSN-immunized, MAP-control-immunized, and unmanipulated B/W
mice groups was further quantified in titration experiments (Fig. 2A). MAP-DWEYSVWLSN-immunized
mice had an IgG anti-dsDNA titer of 12,357 ± 2,558 (mean
± SEM) at 26 wk of age, compared with titers of 2,156 ± 1,325 in
MAP-control-immunized mice and 4,450 ± 1,888 in unmanipulated B/W
mice (Fig. 2B). The difference in end-point anti-dsDNA
titers between MAP-control-immunized and unmanipulated B/W mice groups
was not significant (p = 0.34; Fig. 2B).
|
A,
DWEYSVWLSN immunization resulted in an IgG1, IgG2a, and IgG2b
anti-peptide response. The increase in anti-DWEYSVWLSN Abs in
MAP-DWEYSVWLSN vs control-immunized B/W mice was significant for the
IgG1 (p = 0.0007), IgG2a
(p = 0.015), and IgG2b isotypes
(p = 0.004; Fig. 3A), while there
was no increase in the IgG3 or IgM anti-peptide responses. The
titer of IgG1 (p = 0.002), IgG2a
(p = 0.024), and IgG2b
(p = 0.005) anti-peptide Abs in
MAP-DWEYSVWLSN-immunized B/W mice was significantly higher than in
unmanipulated B/W mice of the same age, but did not differ
significantly for the IgG3 or IgM isotypes (Fig. 3A). The
increase in anti-DNA Abs in MAP-DWEYSVWLSN- vs control-immunized
B/W mice was significant for the IgG2a (p =
0.008), IgG2b (p = 0.044), and IgG3
(p = 0.046) isotypes, but not for IgG1 or IgM
(Fig. 3B). The titer of IgG2a (p =
0.048) and IgG3 (p = 0.021) anti-DNA Abs in
MAP-DWEYSVWLSN-immunized B/W mice was significantly higher than in
unmanipulated B/W mice of the same age, but did not reach statistical
significance for the IgG1, IgG2b, and IgM isotypes (Fig. 3B). In MAP-DWEYSVWLSN-immunized mice, MAP-DWEYSVWLSN
inhibited 50% of the IgG3 anti-dsDNA response, followed by IgM
(35%), IgG2b (26%), and IgG2a (16%). Interestingly, the IgG1
anti-DNA response that was induced by MAP-DWEYSVWLSN immunization
was not inhibited by peptide (Fig. 4A).
|
|
MAP-DWEYSVWLSN immunization promotes renal disease
Kidney histopathology was evaluated in 29-wk-old,
peptide-immunized B/W mice to determine whether an immune response to
peptide can increase the severity of target organ damage in murine
lupus. Kidney histology in MAP-DWEYSVWLSN-immunized B/W mice was
compared with age-matched, unmanipulated, and control-immunized B/W
mice (Fig. 5). Tubular and interstitial
damage (p = 0.02) as well as the total kidney
damage (which includes the glomerular, crescents, tubular, and
interstitial components of the score) (p =
0.02) were significantly greater in DWEYSVWLSN-immunized B/W mice than
in unmanipulated B/W mice (Wilcoxon rank sum procedure, Kruskal-Wallis
test). Thus, MAP-DWEYSVWLSN-immunized B/W mice had significantly more
advanced renal disease than unmanipulated mice. More severe
kidney disease was also apparent in MAP-DWEYSVWLSN-immunized B/W mice
in comparison with control-immunized mice, although the differences
were less significant (p = 0.07 for tubular and
interstitial damage, and p = 0.08 for total kidney
damage (Wilcoxon rank sum procedure, Kruskal-Wallis test)). Among the
different mouse groups, significant glomerular disease was seen only in
MAP-DWEYSVWLSN-immunized B/W mice. Seven of 14 mice in the
MAP-DWEYSVWLSN-immunized group demonstrated moderate or severe
glomerular disease (a combined glomerular and crescents score of 
3,
range 3–5) as compared with 0 of 7 control-immunized B/W mice
(p < 0.05, Fisher’s exact test), and 0 of 5
unmanipulated B/W mice (p = 0.11, Fisher’s
exact test). The differences in the total kidney damage score and each
of its components between control-immunized and unmanipulated B/W mice
were not significant.
|
Twenty-one mAbs reactive with peptide and/or autoantigen were
derived from peptide-immunized B/W mice, all utilizing the L chain
(IgM-14, IgG-7 (IgG1-1, IgG2a-2, IgG2b-3, IgG3-1) (Table I). As selection of positive hybridomas
for further analysis was performed using labeled anti-IgM and
anti-IgG in combination as the detection reagent, selection may
have been biased toward selecting IgM hybridomas due to higher Ab
avidity. Two major groups can be distinguished between these mAb, based
upon specificity for peptide: 1) 7 mAbs that do not bind peptide
(17-15, 18-2, 27-7, 28-22, 3-4, 30-4, 30-2), and 2) 14 mAbs that bind
peptide as well as one or more autoantigens. Abs reactive against
peptide alone could not be isolated. Fourteen of the twenty-one mAbs
isolated were polyreactive, as defined by their binding to peptide,
dsDNA, and at least one other autoantigen (Table I). To address
the possibility that polyreactivity may be due to nonspecific binding,
we studied whether these polyreactive Abs would bind other,
nonlupus-related Ags. As shown in Table II, Ab 19-2 reacted strongly with all of
the irrelevant Ags. The other polyreactive mAbs displayed for the most
part low-affinity binding, and reacted selectively with the panel of
irrelevant Ags. DNA binding of polyreactive Abs could be significantly
inhibited by peptide (Fig. 4B), suggesting that
cross-reactivity was a property of the Ag-binding site, rather than
mediated via a nonspecific charge interaction. Despite a large number
of surviving hybrids from the fusion of two spleens of age-matched
unmanipulated B/W mice, peptide- and/or dsDNA-specific clones could not
be isolated from these mice at 20 wk of age.
|
|
Several of the cross-reactive Abs bind to histone and dsDNA, nuclear Ags that bear very different charges. To confirm that these Abs were monoclonal and indeed cross-reactive with both dsDNA and histone, the 19-2, 14-1, and 20-3 cell lines that displayed high affinity for both Ags were recloned, and the Abs were purified from supernatant. Supernatants as well as purified Abs from each of the recloned cell lines continued to demonstrate strong affinity for histone as well as for dsDNA.
Gene usage and clonality
The complete H and L chain sequences of mAbs derived from
MAP-DWEYSVWLSN-immunized B/W mice are available through GenBank,
accession numbers AF321931–AF321972. H and L chain gene family usage
of the mAbs is shown in Table III. Most
of the mAbs (16 of 21) used gene segments of the J558 family
to encode the H chain V region. Four mAbs used Q52, and one
mAb used a VH10-derived gene segment. No
apparent bias was found for D and J segment usage. L chain analysis
revealed that 12 of the 21 mAbs used a V1- or
V8-derived gene to encode the L chain V region, 4 used
V21, 2 used V4, and 1 each used
VOx-1, V9, and V32. No obvious
bias was present in J usage.
|
). There is a single silent mutation difference in FR1 between
the L chains of the 12-5 and 11-6 Abs, and six amino acid substitutions
in the H chains, five in FR1 and one in FR3. The small differences in
the FR regions between the 12-5 and 11-6 Abs probably account for the
minor differences between their antigenic specificities (Table I). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have examined in this study the effect of immunization with the
peptide DNA surrogate on the immune response and disease course of mice
genetically predisposed to autoimmunity, and we have analyzed the
molecular genetics of their anti-dsDNA response. B/W mice
spontaneously begin to produce anti-dsDNA Abs at 26–30 wk of age
and develop clinical nephritis at 36–40 wk (20).
Control-immunized B/W mice appeared to be following this schedule. In
contrast, MAP-DWEYSVWLSN-immunized mice developed significantly higher
IgG anti-DNA, anti-cardiolipin, and anti-laminin Abs
starting from 
10 wk after the initial exposure to peptide, and
continuing until the mice were sacrificed at 29 wk of age.
Several important differences in the immune and autoimmune responses to peptide between nonautoimmune BALB/c and autoimmune B/W mice are worth emphasizing. B/W mice had a somewhat delayed anti-peptide response to DWEYSVWLSN. We previously found that the response to DWEYSVWLSN is genetically determined, with several mouse strains not mounting any anti-peptide response to immunization (12). The short delay in the anti-peptide response in B/W mice may be a result of a genetic variability in the response to this peptide. Alternatively, it is possible that the initial peptide immunizations elicited a cross-reactive autoantibody response, which was effectively down-regulated when the B/W mice were young. Peptide-immunized B/W mice, in contrast to our earlier findings in BALB/c mice, did not display a consistent increase in serum IgG anti-Sm/RNP Abs, presumably reflecting the general absence of Abs with these specificities in the B/W lupus strain. In peptide-immunized BALB/c mice, both the anti-peptide as well as the anti-DNA Abs were predominantly of the IgG1 isotype. Moreover, in peptide-immunized BALB/c mice, IgG1 was prominent in renal immune deposits (12). In serum of immunized B/W mice, however, anti-peptide and anti-DNA Abs of the IgG2a and IgG2b isotypes were present. This is consistent with spontaneous lupus in B/W mice, in which anti-DNA Abs of the IgG2a and IgG2b isotypes are most frequent, and the predominant isotype in glomerular deposits is IgG2a (21). The induced isotypes of the anti-dsDNA Ab response in peptide-immunized B/W mice, as opposed to peptide-immunized BALB/c mice, may have contributed to the more severe renal disease induced in B/W mice in response to peptide immunization (22).
Comparison of the isotypes in the induced anti-peptide and anti-dsDNA Ab responses in MAP-DWEYSVWLSN-immunized mice with the spontaneous response in unmanipulated B/W mice reveals some notable differences. The induced anti-MAP-DWEYSVWLSN response in immunized mice, as compared with unmanipulated mice, is primarily IgG1, IgG2a, and IgG2b; this suggests that the anti-peptide response results from activation of follicular B cells, which undergo maturation and class switching in germinal centers. In the isotype analysis of the anti-dsDNA response, however, IgG2a as well as IgG3 anti-dsDNA Abs were significantly higher than in mice with spontaneous disease.
Perhaps the most important difference between MAP-DWEYSVWLSN-immunized BALB/c and B/W mice was found in analysis of target organ damage in the kidney. Although glomerular immune deposits occurred in peptide-immunized BALB/c mice, none of the mice displayed any histologic evidence of renal damage at light microscopy resolution (12). In contrast, DWEYSVWLSN immunization precipitated severe glomerular and to a lesser extent tubulointerstitial disease that was not present in B/W mice of the same age immunized with control peptides. Similarly, global kidney damage (as measured by a composite of the damage scores in the glomerular, crescents, tubules, and interstitium categories) was significantly greater in MAP-DWEYSVWLSN-immunized than in unmanipulated B/W mice.
It is important to note that several of the peptide-immunized mice displayed significant tubulointerstitial disease, including infiltration by mononuclear cells. Although the histopathological lesion in lupus with the greater physiological consequence to renal function is glomerular disease, tubulointerstitial inflammation is quite common in both human (23) and murine (24, 25) lupus nephritis. In a study by Hurd and Ziff (24), 100% of B/W mice at 7 mo of age displayed significant mononuclear cell interstitial infiltration. Therefore, we believe that the interstitial renal disease as well as the glomerulopathy observed in peptide-immunized B/W mice do not represent a novel form of kidney disease, but rather significant acceleration of the type of nephritis due to appear later on in the course of spontaneous murine lupus.
Although anti-histone Abs were not induced in MAP-DWEYSVWLSN-immunized mice, both IgG anti-DNA and IgG anti-laminin Ab titers were significantly higher than in controls. The latter Ab specificities have been closely associated with renal pathogenicity in lupus (2, 3, 4, 26), and probably contributed to the accelerated renal disease seen in immunized mice. Although control-immunized and unmanipulated B/W mice also had measurable autoantibodies in their serum, the titers were lower than in MAP-DWEYSVWLSN-immunized mice.
IgG is the Ab class most closely associated with kidney disease in lupus (2). It is interesting to note that IgG autoantibodies from peptide-immunized B/W mice have been more easily isolated than from BALB/c mice, suggesting that induced pathogenic Abs are more frequent and/or regulated differently in peptide-immunized B/W mice. Differences in antigenic specificity, isotype, or complement-fixating capability of the induced Abs, the cytokine milieu, or T cells may also be possible contributors to the observed pathologic differences between BALB/c and B/W peptide-immunized mice. Alternatively, genetic differences in Ag display at the level of the kidney are responsible for the differing degree of renal damage. These possibilities will be addressed directly in future studies.
Molecular genetic analysis of the peptide and autoantibody responses in B/W mice reveals some novel features. With respect to H chain gene usage, all mAbs showed close homology to spontaneously arising anti-DNA Abs from B/W mice. The H chains of two mAbs (14-1 and 30-4) showed, in addition, close homology to the H chains of F1-3 and F5-40, anti-DNA Abs induced in preautoimmune B/W mice by bacterial DNA immunization (18), respectively. Furthermore, the H chains of eight mAbs (25-32, 11-6, 12-5, 29-2, 24-4, 7-13, 3-4, and 9-18) displayed in addition significant similarities to H chains used in anti-peptide and autoantibodies arising in nonautoimmune BALB/c mice immunized with MAP-DWEYSVWLSN (13). Two-thirds (14 of 21) of the mAbs used a J558 family member to encode for the H chain V region of peptide- and/or DNA-binding Abs in peptide-immunized B/W mice as compared with only 33% (8 of 24) in peptide-immunized BALB/c mice (13), apparently in conformance with the J558 bias in spontaneously arising anti-DNA Abs in B/W mice (27). Q52 (4 of 21 in B/W, 3 of 24 BALB/c) and VH10-derived H chains (1 of 21 B/W, 3 of 24 BALB/c) showed less of a difference in usage frequency between strains. Several H chain gene families represented in the BALB/c response to peptide (S107, 7183, 36-60, 606) were not isolated from mAbs in B/W mice. Although unusual D segment utilization has been reported in murine lupus anti-DNA Abs (28, 29), D-D fusions or inverted Ds were not identified in peptide-immunized B/W mice.
The L chains of all but four mAbs (14-1, 21-28, 24-4, and 3-4) were
closely homologous to L chains used in the spontaneous anti-DNA
response in B/W mice. The L chain of Ab 14-1 used a
V1-c-derived gene segment. The L chains of Abs 21-28 and
24-4 were identical, and were closely related to the L chain of bfd05,
an induced anti-DNA Ab derived from a BALB/c mouse immunized with
the Fus1 peptide (19). The 3-4 L chain was closely related
to the L chain of the 84-39 anti-DNA Ab arising in a transgenic
BALB/c mouse (30). Light chain subfamily usage did not
appear to significantly differ between B/W- and BALB/c-immunized mice:
a V1-a-encoded L chain was used in two mAbs in B/W mice
and three mAbs in BALB/c mice, a V1-c-encoded L chain was
used in three mAbs in B/W mice and two mAbs in BALB/c mice, and a
V1-b L chain in a single B/W-derived mAb.
Nine mAbs (of which eight bound DNA) displayed
VH-VL combinations common
in the spontaneous anti-DNA response in B/W mice (27):
six mAbs displayed a J558-V1 combination, and three mAbs
had a J558-V8 combination. Interestingly, of the six
J558-V1 Abs that bound DNA, only 14-1 and 29-2 also bound
peptide. In contrast, J558-V1 and J558-V8
combinations appeared only in 3 of 24 Abs in peptide-immunized
BALB/c mice.
Since the H chains from the MAP-DWEYSVWLSN-induced mAbs are encoded by
the same genes that are used in spontaneously arising anti-DNA Abs
in B/W mice, it is of particular interest to analyze those mAbs from
peptide-immunized mice that do not bind DNA (Table I). Comparison with
mAbs 165.3 and 165.6 (27), previously described
anti-DNA Abs from B/W mice, is particularly pertinent. The V and J
regions of 18-2 and 165.3 H chains are identical; however, the D region
of 165.3 contains two arginines, an amino acid that is important in
conferring anti-DNA specificity. The 30-4 H chain has a single FR1
substitution as compared with 165.6, and is otherwise identical through
FR3. The D region of 165.6 is 10 aa long and has several charged
residues, while the D region of 30-4 is much shorter, and not charged.
An alternative explanation for the lack of binding to DNA by these mAbs
may lie in the L chain. Although the L chains of 18-2 and 30-4 are
closely homologous or identical to L chains of DNA-binding Abs, it is
possible that small differences in the L chain or in the specific H and
L chain pairing contribute to the loss of specificity for DNA.
Nevertheless, to conclusively demonstrate which amino acids are
important in DNA binding, and whether peptide or DNA binding is encoded
in the germline, further studies by mutation analysis are
necessary.
Results in B/W mice reported in this study are consistent with our findings in analysis of the induced response in peptide-immunized BALB/c mice, namely, that autoantibodies induced in response to peptide immunization show close structural homology to anti-DNA Abs arising in spontaneous murine lupus. Interestingly, this has also been the conclusion of analyses of anti-DNA Abs isolated in two induced disease models that are quite different from our peptide-induced model, specifically immunization with Fus1 (a DNA-binding peptide from Trypanosoma cruzi) in complex with DNA (9) and bacterial DNA (18). Furthermore, Abs induced in the anti-peptide response are homologous to induced anti-DNA Abs in the latter two models. Taken together, these results seem to indicate a common molecular genetic response pathway to a variety of stimuli that share the ability to break tolerance to native, mammalian DNA.
Most of the mAbs induced by immunization of B/W mice with the peptide
DNA surrogate were cross-reactive, binding both peptide and dsDNA
(Table I). We isolated five anti-DNA mAb that do not also bind
peptide: 17-15, 27-7, 28-22, 3-4, and 30-2 (Table I). Of these mAb, the
H chains of 17-15, 27-7, 28-22, and 3-4 are closely related, and
probably derive from the same germline gene as 111.185, an IgM
anti-DNA Ab derived from a B/W mouse (27). Although it
is not known whether the 111.185 Ab binds peptide, the fact that the H
chains of 17-15, 27-7, 28-22, and 3-4 are related, yet the mAbs do not
bind peptide suggests that it is possible that some anti-DNA
Abs in peptide-immunized B/W mice arose spontaneously, not in response
to peptide. Consistent with this suggestion is the finding that while
peptide partially inhibited the binding of sera from peptide-immunized
B/W mice to dsDNA, the degree of inhibition was less than in
peptide-immunized BALB/c mice (12).
How are cross-reactive anti-peptide/anti-DNA Abs being generated in this model? Although we could not isolate germline-encoded, cross-reactive Abs, cross-reactivity may be germline encoded. Alternatively, mutation of spontaneously arising anti-DNA Abs to acquire specificity for peptide, or somatic mutation in the anti-peptide immune response can account for cross-reactive anti-peptide/anti-DNA Abs in peptide-immunized B/W mice. We hope to be able to find direct evidence for either or both of these hypotheses by selective analysis of the peptide-binding splenocyte population in peptide-immunized mice.
There are several possible ways to explain the derivation of Abs that are cross-reactive for both histone and dsDNA from peptide-immunized B/W mice, despite the disparity in charge between these two autoantigens. The immunizing peptide DWEYSVWLSN is a peptide mimic of DNA for R4A, an Ab that is specific for DNA as well as histone. In studies from other investigators as well as in our own work (13), it has been demonstrated that immunization with a peptide mimic can generate Abs with binding sites that are similar to the original Ab used to select the peptide from the phage library. Similarly, immunization with the R4A peptide mimetope elicits cross-reactive anti-dsDNA/anti-histone Abs. Furthermore, Monestier et al. (31) reported that certain anti-histone mAbs isolated from lupus-prone mice can bind to DNA in the absence of histone.
In summary, immunization of mice genetically predisposed to autoimmunity with a peptide surrogate of dsDNA induced the early onset of high titer anti-DNA, anti-laminin, and anti-cardiolipin Abs, and precipitated severe lupus-like glomerular and tubulointerstitial disease. Peptide immunization in B/W mice appeared to have resulted in a markedly increased frequency of B cells secreting pathogenic autoantibodies, resulting in accelerated disease. Interestingly, in only 1 of 14 sera from MAP-DWEYSVWLSN-immunized B/W mice was there any inhibition of the anti-cardiolipin and anti-laminin responses by peptide, supporting the suggestion that peptide immunization induces autoimmune manifestations in B/W mice by promoting the underlying disease. Although there were some differences in autoantibody isotypes, the induced anti-dsDNA Abs displayed close structural similarities to anti-DNA Abs arising in spontaneous, as well as induced murine lupus.
Based on the demonstrated similarity between the spontaneous, and several types of induced anti-DNA Abs, it may be possible to postulate that there may not be a single, triggering Ag for the disease clinically classified as SLE. Furthermore, although direct evidence for this hypothesis is not available at this time, the similarities between the induced autoimmune response to peptide and spontaneous autoimmunity suggest that peptide Ags may induce an autoimmune response in individuals with certain genetic predispositions. Finally, peptides that can up-regulate the autoantibody response in lupus have been used successfully in blocking kidney deposition of Ig, delaying the onset of renal disease, and in prolonging survival in murine lupus. Possible mechanisms to explain this therapeutic benefit include induction of Ag-specific unresponsiveness, activation of regulatory T cells, or interference with binding to a cross-reactive antigenic target (32). Similarly, peptide DNA mimetopes that induce lupus-like autoantibodies by immunization may eventually turn out to have therapeutic potential in SLE.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Chaim Putterman, Division of Rheumatology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461. E-mail address: putterma{at}aecom.yu.edu
3 Abbreviations used in this paper: SLE, systemic lupus erythematosus; FR, framework region; MAP, multiple antigenic peptide; PAS, periodic-acid Schiff.
Received for publication February 16, 2001. Accepted for publication January 28, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Z. Zhao, L. C. Burkly, S. Campbell, N. Schwartz, A. Molano, A. Choudhury, R. A. Eisenberg, J. S. Michaelson, and C. Putterman TWEAK/Fn14 Interactions Are Instrumental in the Pathogenesis of Nephritis in the Chronic Graft-versus-Host Model of Systemic Lupus erythematosus J. Immunol., December 1, 2007; 179(11): 7949 - 7958. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. O. Jacob, L. Pricop, C. Putterman, M. N. Koss, Y. Liu, M. Kollaros, S. A. Bixler, C. M. Ambrose, M. L. Scott, and W. Stohl Paucity of Clinical Disease despite Serological Autoimmunity and Kidney Pathology in Lupus-Prone New Zealand Mixed 2328 Mice Deficient in BAFF J. Immunol., August 15, 2006; 177(4): 2671 - 2680. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Onate, G. Donoso, G. Moraga-Cid, H. Folch, S. Cespedes, and E. Andrews An RNA Vaccine Based on Recombinant Semliki Forest Virus Particles Expressing the Cu,Zn Superoxide Dismutase Protein of Brucella abortus Induces Protective Immunity in BALB/c Mice Infect. Immun., June 1, 2005; 73(6): 3294 - 3300. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Guth, X. Zhang, D. Smith, T. Detanico, and L. J. Wysocki Chromatin Specificity of Anti-Double-Stranded DNA Antibodies and a Role for Arg Residues in the Third Complementarity-Determining Region of the Heavy Chain J. Immunol., December 1, 2003; 171(11): 6260 - 6266. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Hardin Directing Autoimmunity to Nucleoprotein Particles: The Impact of Dendritic Cells and Interferon {alpha} in Lupus J. Exp. Med., March 17, 2003; 197(6): 681 - 685. [Full Text] [PDF]
|
|
|
|
 |
|
 B Deocharan, X Qing, E Beger, and C Putterman Antigenic triggers and molecular targets for anti-double-stranded DNA antibodies Lupus, December 1, 2002; 11(12): 865 - 871. [Abstract] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
