| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Institut de la Santé et de la Recherche Médicale Unité 519, Faculté de Médecine et de Pharmacie, Hôpital Charles-Nicolle, and
Laboratoire de Spectroscopie, Institut Fédératif de Recherches Multidisciplinaires sur les Peptides, Institut de Rechereche en Chimie Organique Fine, Rouen, France.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-enolase, heterogeneous nuclear ribonucleoprotein L, and
nucleophosmin, were the targets of the B cell autoimmune response, and
that autoantibodies to them were synthesized sequentially in an orderly
pattern that recurred in all the male WB mice analyzed:
anti-vimentin first and anti-nucleophosmin last, with
anti-RAD23 and anti-heat shock protein 60, then
anti--enolase and anti-heterogeneous nuclear
ribonucleoprotein L Abs occuring concomitantly. Anti-vimentin
reactivity always appeared before anti-cardiolipin and anti-DNA
Abs, suggesting that vimentin is the immunogen initiating the
autoimmune process. The pattern of HL-60 proteins recognized by female
WB sera differed from that of male sera, indicating that the Y
chromosome-linked autoimmune acceleration gene is not an accelerator
but a strong modifier of the autoimmune response. Thus, 1) combining
two-dimensional PAGE and mass spectrometry constitutes a powerful tool
to identify the set of Ags bound by autoantibodies present in a single
serum and the whole autoantibody pattern of an autoimmune disease; 2)
the diversification of the autoimmune response in male WB mice occurs
in a predetermined pattern consistent with Ag spreading, and thus
provides a useful model to further our understanding of the development
of the autoantibody response in
lupus. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Male (New Zealand White (NZW) x BXSB)F1 (WB) mice develop a lupus-like disease that shares clinical, histopathological, and immunological features with other strains of lupus mice, including the production of anti-nuclear Abs, elevated Ig concentrations, and glomerulonephritis (14). These mice also frequently suffer from coronary artery disease, myocardial infarction, and thrombocytopenia, and produce autoantibodies that bind cardiolipin (CL) in a cofactor-dependent manner (15, 16, 17). The severe disease observed in males results in part from the action of a mutant gene, Y chromosome-linked autoimmune acceleration (Yaa) (18). Therefore, male WB mice are considered to constitute a model of both SLE and secondary anti-phospholipid syndrome. In this study, using a powerful technique combining two-dimensional (2D)-PAGE and mass spectrometry (MS), we showed that the B cell autoimmune response in these mice sequentially targets various cellular autoantigens, thereby defining a specific autoantibody pattern that arises by diversification of the autoimmune response in a preordained manner, highly suggestive of an Ag-spreading phenomenon.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A lysate of the human promyelocytic leukemia cell line HL-60 (European Collection of Animal Cell Cultures, Salisbury, U.K.) was used as the substrate for 2D-PAGE. The cell line was grown at 37°C in a humidified atmosphere (95% air and 5% CO2) in 75-cm2 cell-culture flasks (Corning, Acton, MA) in 25 ml of RPMI 1640 (BioWhittaker, Walkersville, MD) supplemented with 10% v/v heat-inactivated FCS (BioWhittaker), 2 mM L-glutamine (BioWhittaker), and penicillin (100 U/ml)/streptomycin (100 mg/ml) (BioWhittaker). Cells were washed three times with PBS and isolated by centrifugation (1500 rpm, 5 min) at room temperature.
Mice and sera
Female NZW and male BXSB mice, originally purchased respectively from Bomholtgard Breeding and Research Center (Ry, Denmark) and The Jackson Laboratory (Bar Harbor, ME), then maintained in our animal facilities, were crossbred to obtain WB offspring. All mice were housed in the same room and fed the same diet. Mice were bled on days 60, 90, 120, and 180. Sera were stored at -80°C until used.
ELISA
Anti-CL Abs. Flat-bottomed, 96-well polystyrene microtiter plates (Virion, Roche, France) were coated with 30 µl/well of bovine heart CL (100 µg/ml) in absolute alcohol. The plates were incubated overnight at 4°C to allow the ethanol to evaporate and the remaining free binding sites were saturated with 10% FCS for 2 h at room temperature. After washing, the sera (diluted 1/100) were incubated for 2 h at room temperature. After washing the plates three times with PBS, biotin-conjugated goat anti-mouse IgG (Caltag Laboratories, San Francisco, CA) was added for 1 h at 37°C. After three washes, alkaline phosphatase-conjugated streptavidin (Caltag Laboratories) was added and revealed with p-nitrophenol (Sigma-Aldrich, St. Louis, MO). The OD was read at 405 nm with a Titertek Multiskan plate reader (Flow Laboratories, Les Ulis, France).
Anti-DNA Abs. Flat-bottomed 96-well microtiter plates were coated with 10 µg/ml ultrapure DNA (Sigma-Aldrich) in 0.1 M carbonate buffer, pH 9.5, and processed as described above.
2D-PAGE map construction
All reagents and instruments used have been described in detail elsewhere (19).
Protein sample preparation. HL-60 cells were harvested, centrifuged (1,500 rpm, 5 min) at room temperature, and 1 x 108 were precipitated in organic solvent before being lysed in 10 ml of 40 mM Tris buffer (Sigma-Aldrich), 8 M urea (Fluka, Ulm, Germany), 2% (w/v) 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (Sigma-Aldrich), 20 mM spermine tetrahydrochloride (Sigma-Aldrich), and a mixture (10 µg/ml each) of protease inhibitors (benzamidine, pepstatin, leupeptin, EDTA, and EGTA) (Sigma-Aldrich). The lysate was ultracentrifuged (36,000 rpm, 1 h, 4°C) and sonicated (Vibra Cell; Bioblock Scientific, Illkirch, France), and the resulting protein preparation was subjected to 2D PAGE.
Rehydration, first-dimension isoelectric focusing, and equilibration: 125 µl (250 µg) of the protein lysate solution mixed with the appropriate volume of rehydration solution were loaded into a ceramic strip holder (Pharmacia Biotech, Uppsala, Sweden) in the IPGphor system (Pharmacia Biotech). The protocol given is suitable for first-dimension isoelectric focusing of typical analytical quantities. Rehydration was performed overnight. Commercial 7-cm immobilized pH gradient (IPG; 3.5–10) strips (Pharmacia Biotech) were used for the first dimension. The voltage was linearly increased from 200 to 8,000 V. A total of 30 kVh was used for the run. After the first-dimension run, the strips were equilibrated in order to resolubilize the proteins and to reduce disulfide bonds.
Second-dimension electrophoresis.
After equilibration, the
IPG gels were transferred for the second dimension onto vertical 10%
(w/v) polyacrylamide (Sigma-Aldrich) slab mini-gels (9 cm x 7
cm x 1.5 mm) and run with the Laemmli SDS-discontinuous system at
a maximum of 30 mA per gel for 
2 h until the dye front reached the
bottom of the gel. Proteins were detected after Coomassie brillant blue
G-250 (Sigma-Aldrich) staining or gels were electroblotted for 2 h
onto polyvinylidene difluoride (PVDF) membranes (Millipore,
Bedford, MA) stained with Ponceau S before being used for the
immunodetection. The immunoreactive spots were detected using sera
(diluted 1/100), peroxidase-conjugated goat anti-mouse IgG
(Sigma-Aldrich), and an ECL system (Pharmacia Biotech). 2D gels were
scanned and images were obtained with a computing Imaging Densitometer
(Model GS-700; Bio-Rad, Hercules, CA) and digitalization software
(Multi-Analyst PC Software for the Bio-Rad Image Analysis Systems
v1.1). 2D protein pattern recognition was performed by comparing
autoradiography and Ponceau S pattern with the corresponding Coomassie
brillant blue G-250-stained mini gels. Protein spots were excised from
Coomassie-stained gels (18 cm x 16 cm x 1.5 mm).
Protein identification by MS and peptide mass fingerprinting
The immunoreactive spots were excised from polyacrylamide gels with a sterile scalpel and cut into 1 mm x 1 mm cubes which were transferred into microfuge tubes, washed once or twice with water, and then with water/acetonitrile (1/1, v/v) (Sigma-Aldrich) for 15 min/wash, and placed at 56°C for 45 min in 0.01 M DTT and 0.1 M ammonium bicarbonate (Sigma-Aldrich) to be reduced and alkylated. Lastly, the particles were placed in digestion buffer containing 12.5 ng/µl of sequencing grade trypsin (Boehringer Mannheim, Mannheim, Germany), which is the enzyme of choice for subsequent MS peptide sequencing because of its specificity. Peptides extracted from the gel and the samples were dried in a vacuum centrifuge and desalted. For MS, peptides were redissolved in 20 µl of 5% (v/v) formic acid (Sigma-Aldrich). Samples were analyzed with a matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) Reflector (Tof Spectrometer E; Micromass, Manchester, U.K.) to obtain peptide mass information for database searches. The peptide fingerprints obtained were used to search the nonredundant protein databases (SWISS-PROT; NCBInr). Data were matched against the databases using the PeptIdent and/or MS-Fit programs (accessible through the EXPASY server at the following address: http://www.expasy.ch/) and the Protein Prospector server (http://prospector.ncsf.edu/). The peptide masses were matched within the largest window possible for isoelectric point (pI), relative molecular mass, and species specificity (Homo sapiens).
Antigens
Vimentin and -enolase were respectively purified from bovine
lens and rabbit muscle. These two Ags and recombinant heat shock
protein (hsp) 60 were purchased from Sigma-Aldrich. Human nucleophosmin
and RAD23B were produced as recombinant proteins.
Total RNA was extracted from HL-60 cells using TRIzol reagent (Invitrogen, Eragny, France) and reverse transcribed with M-MLV reverse transcriptase (Invitrogen) using an oligo(dT)-primer. A DNA fragment encoding human RAD23B (aa 1–409) was amplified from the cDNA by PCR with the following primers: RAD23-S-Not I, 5'-TGG GCG GCC GCC AGG TCA CCC TGA AGA C-3' (sense) and RAD23-R-Pfl23 II, 5'-GCC GTA CGA TCT TCA TCA AAG TTC TGC TGT AGA AGA-3' (antisense). The PCR product was digested with NotI-Pfl23 II and ligated with NotI-Pfl23 II-cut pIVEX-TAG plasmid. pIVEX-TAG was constructed by replacing the NotI-BamHI fragment of pIVEX-2.3MCS (Roche, Meylan, France) with a synthetic oligonucleotide corresponding to restriction site Pfl23 II, FLAG tag DYKDDDDK and hexahistidine followed by a stop codon and BamHI site. Thereafter, recombinant RAD23B protein was expressed by Rapid Translation System RTS 500 according to manufacturer’s instructions (Roche).
A 879-bp fragment of the nucleophosmin gene, using cDNA as a template,
was amplified by PCR with primers, B23-NdeI 5'-GGA ATT CCA
TAT GGA AGA TTC GAT GGA CAT GGA C-3' (sense) and B23-BamHI
5'-CCG GAT CCT TAC TTG TCA TCG TCG TCC TTG TAG TCC GTA CGA AGA GAC TTC
CTC CAC TGC C-3' designed to create FLAG tag followed by a stop codon
and BamHI site. The resulting PCR product was cloned into
Nde-BamHI-digested pET-15b (Novagen, Madison, WI) downstream
of the histidine tag. Recombinant nucleophosmin B23 was expressed in
liquid culture of Escherichia coli BL21 (DE3) pLysE after
induction with 1 mM isopropyl-
-D-thiogalactopyranoside.
The majority of protein was found in soluble fractions of bacterial
extracts. Therefore, E. coli cells were resuspended in PBS
and disrupted by sonication. After centrifugation, supernatant was
applied to a Ni2+-NTA column (Qiagen, Helden, Germany) and
nucleophosmin was purified according to manufacturer’s instructions.
The expected size of recombinant proteins RAD23B and nucleophosmin B23
was confirmed by Western blotting using anti-FLAG mAb
(Sigma-Aldrich).
One-dimensional Western blot analysis
Commercially available and recombinant proteins were fractionated by electrophoresis on 12% SDS-PAGE, electroblotted onto nitrocellulose membranes, and incubated for 2 h with 1/100 diluted mouse sera or 1/2000 diluted anti-FLAG mAb. After washing, the membranes were probed with peroxidase-conjugated goat anti-mouse IgG (diluted 1/5000) and revealed with an ECL system.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We first characterized the antigenic specificities of
autoantibodies produced by 6-mo-old male WB mice. Four mouse sera
collected on day 180 were incubated on a 2D PAGE separated-HL-60
protein map to identify Ags bound by male WB autoantibodies. As shown
in Fig. 1
each serum recognized a limited
number of proteins. All recurrent protein spots recognized by an
individual serum were excised from the gel, trypsinized, and the
protein digests (20 µl) were subjected to MALDI-TOF reflector
analysis. For example, spectrum obtained with spot c bound by serum
from male WB mouse 1 is shown in Fig. 2.
After introducing the protein pI and relative molecular mass filters
into MS-Fit, the protein is identified as UV excision-repair protein
RAD23B for spot c. Identification of all other spots recognized by the
four sera from the 6-mo-old male WB mice indicated that these sera
contained autoantibodies reacting with a restricted set of six
proteins: vimentin (spot a), hsp 60 (spot b), UV excision-repair
protein RAD23B (spot c), -enolase (spot d), heterogeneous nuclear
RNP (hnRNP) L (spot e), and nucleophosmin/phosphoprotein B23 (spot f),
whose characteristics are summarized in Table I. Some nonrecurrent spots could not be
identified although the immunodetection experiments were done twice
and, when enough serum was available, realized from a zoom area of a 18
cm x 16 cm x 1.5 mm gel.
|
|
|
To study the pattern of evolution of the autoantibody response in
WB males, sequential sera obtained from individual mice on days 60, 90,
120, and 180 of life were analyzed under the same proteomic assay
conditions. We first selected sera from male 5 whose day 180 serum
bound to six distinct proteins and, thus, exhibited a full-blown immune
response. This analysis indicated that the first target of
seroconversion was vimentin (Fig. 3)
which, on day 60, was the only immunoreactive spot on the 2D
PAGE-separated HL-60 protein map. Analysis of the sample collected on
day 90 clearly showed a diversification of the autoantibody response
since, in addition to vimentin, two other proteins, the UV
excision-repair protein RAD23 and hsp 60 were bound by Abs present in
WB male 5 serum (Fig. 3B). By day 120, Abs directed against
-enolase and hnRNP L developed (Fig. 3C) and, in day 180
serum, autoantibodies recognized the last target of the autoimmune
response, nucleophosmin/phosphoprotein B23 (Fig. 3D). It
should be noted that the sequential serum samples from male 5 showed no
loss of Ab positivity. This observation suggests that, in this animal,
vimentin was the first autoantigen to which autoimmune response
developed, followed in a preordained pattern by spreading to other
target Ags: anti-RAD23 and anti-hsp 60, then
anti--enolase and anti-hnRNPL, and, finally,
nucleophosmin.
|
To determine whether the pattern of development of the
autoantibody response observed in male 5 was similar in other male WB
mice, the order of emergence of the different autoantibody populations
was examined in sequential sera obtained from seven male WB mice. One
male died on day 100 and another on day 170, respectively, numbers 2
and 7. Our results clearly indicated an orderly pattern of evolution of
the autoantibody response in male WB mice. First, in all seven males,
vimentin was always recognized by male WB sera on day 60 (Fig. 4), thereby confirming that this protein
constitutes the initial immunogen of the autoantibody response. Fig. 4
shows that Abs to hsp 60 and UV excision-repair protein RAD23 appeared
concurrently for the first time at day 60 in two males and at day 90 in
four animals and, as in the course of WB male 5, were followed by Abs
directed against -enolase and hnRNP L on day 120. Nucleophosmin
reactivity was the last to emerge and was observed in three of the five
surviving mice on day 180. One should note that, as far as assessable,
these mice developed full-blown responses when the reactivities in
sequential sera are considered. However, reactivity was lost in
animals: on day 120, three no longer had anti-vimentin Abs, two had
lost anti-hsp 60 Abs, and two had lost anti-UV excision-repair
RAD23 Abs.
|
Previous studies showed that male WB mice develop anti-CL and
anti-DNA Abs (14, 17). To determine the kinetics of
the different autoantibody populations and, in particular, the temporal
relationship of anti-HL-60 protein Abs with anti-CL and
anti-DNA Abs, CL- and DNA-binding activities were evaluated in
sequential sera obtained from the seven male WB mice under study. On
day 60, when all seven male WB mice had already developed
anti-vimentin Abs, none of them had anti-CL or anti-DNA
Abs. Most (six of seven) male WB mice started to produce anti-CL
Abs by day 90, while anti-DNA Abs were detected later, in day 120
sera, i.e., when anti-hsp 60 and anti-RAD23 Abs had also
already appeared. Fig. 5, A and
B, and B shows the
kinetic of the anti-CL and anti-DNA Ab response in the seven
male WB mice.
|
Analysis of sequential sera collected individually from eight
female WB mice on days 60, 90, 120, and 180 showed that the
autoantibody-response pattern differed from that of WB males (Table I
).
First and surprisingly, none of the 32 sera reacted with Ags targeted
by male WB autoantibodies. Second, female WB sera collected on day 180
mainly recognized four proteins: hsp 71, hsp 27, high mobility group
protein (HMG) 2, and tropomyosin, as illustrated by Fig. 6, which shows results obtained with
female WB mouse 4 serum. Analysis of sequential sera from the other
female WB mice confirmed the above findings and indicated that the
specificity of autoantibodies produced in the WB model during the
course of the spontaneous SLE differed dramatically between males and
females. Furthermore, it also showed that, at a given time, the
autoantibody populations differed among the female mice and that they
appeared in a disordered manner.
|
To unambiguously identify the Ags bound by male WB mouse sera,
sera or pooled sera collected from male WB mice at different ages were
tested against several of the candidate target Ags obtained in purified
form. Vimentin and -enolase were purified from, respectively, bovine
lens and rabbit muscle. hsp 60 was a human recombinant protein
expressed in E. coli. Nucleophosmin and RAD23B were produced
in our laboratory as recombinant proteins. First, a pool of sera
obtained from 6-mo-old male WB mice, at a time when the full-blown
autoimmune response had developed, was tested by Western blotting
against nucleophosmin by SDS-PAGE. Fig. 7
shows that these sera bound to nucleophosmin. Second, sera obtained
from 3-mo-old male WB mice consistently bound to hsp 60, RAD23, and
vimentin. In contrast, no reactivity with purified -enolase was
observed whatever the age of mice. Female WB sera did not bind to any
of the purified Ags recognized by male sera, confirming the results
given by 2D gel and MS analysis. hnRNP L was not available and was not
tested in the assay. Thus, the use of purified Ags allowed us to
confirm most of the Ag-binding properties of male WB mouse sera
identified by 2D gel electrophoresis and MS.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Six proteins constituted the major target Ags of the male autoimmune
response and defined its characteristic pattern. Four of the six
identified proteins, vimentin, hsp 60, -enolase, and nucleophosmin,
were previously shown to be the targets of autoantibodies produced in
human or mouse nonorgan-specific autoimmune diseases. For example,
anti-vimentin Abs have already been described in SLE patients
(20, 21) and their presence was positively correlated with
that of anti-CL Abs (22), an observation similar to
that made in this study for male WB mice. Similarly,
anti-nucleophosmin autoantibodies were previously detected in SLE
patients, and were shown to be constantly associated with the presence
of anti-CL Abs (23). It is worth noting that two
monoclonal autoantibodies, one derived from the PBL of an SLE patient,
and the other from male WB splenocytes, and both selected based on
their capacities to react with anionic phospholipids, also bound to
vimentin (24) and nucleophosmin (19),
respectively. These observations, and the high frequency of
anti-vimentin and anti-nucleophosmin Abs in male WB mice that
produce anti-CL, indicate that the pathophysiological mechanisms
responsible for the production of these antiprotein and
antiphospholipid reactivities are intimately linked. hsp 60 is a stress
protein that plays a crucial chaperone role in the correct folding,
assembly, and compartmentalization of intracellular proteins
(25), is highly immunogenic in response to infection, and
is also a common target of autoantibodies in various rheumatic and
inflammatory diseases (26). Finally, -enolase Abs have
been described in various pathological situations, mainly in
nonorgan-specific autoimmune diseases (27).
Two autoantibody reactivities, so far not described in human or mouse autoimmune diseases, were identified by our technical procedure. UV excision-repair protein RAD23 is a highly conserved 58-kDa protein, located in the nucleus and involved in nucleotide excision-repair and possibly in DNA-damage recognition and/or in altering chromatin structure to allow access by damage-processing enzymes (28). hnRNP L belongs to the large superfamily of hnRNP, which are frequently recognized by autoantibodies present in patients with nonorgan-specific autoimmune diseases: Abs to hnRNP A1, A2, and B have been previously found in patients with rheumatoid arthritis, SLE, and mixed connective tissue diseases (29) and Abs directed to hnRNP I (polypyrimidine tract-binding protein) were detected in the sera of systemic sclerosis patients (30). Notably, the amino acid sequence of hnRNP L is highly homologous to that of hnRNP I which together constitute a new family of hnRNP proteins within the RNP consensus RNA-binding proteins (31).
Thus, male WB mice synthesize autoantibodies directed against a particular set of autoantigens which defines their specific B-cell autoimmune response. Indeed, this pattern of autoantibody reactivities differed markedly from that of female WB mice whose sera bound predominantly to four proteins, hsp 27, hsp 71, tropomyosin, and HMG2. Therefore, the Yaa gene and its product, thought to be expressed at the B-cell surface (32), not only behave as accelerators of the lupus disease occurring in the WB strain, as previously demonstrated by their capacities to enhance autoantibody production in different lupus-prone mice (14, 33), but also as strong modifiers of the autoimmune response. This observation indicates that a single locus can dramatically modify the autoimmune response and phenotype of a nonorgan-specific autoimmune disease.
One key issue to understanding the mechanism and etiology of complex autoimmune diseases like SLE, is to determine what characteristics immunodominant self-Ags, such as those identified in this study may share. In this regard, several studies have indicated that apoptotic cells could constitute a reservoir of self-Ags that elicit an autoimmune response in SLE and contribute to the development of autoimmunity (34). Indeed, many lupus-related autoantigens undergo posttranslational modifications during apoptosis, such as cleavage by caspases (34) or the cytotoxic-granule protein, granzyme B (35), and are relocalized and clustered at the surface of apoptotic blebs or vesicles released by cells that enter a process of programmed death (36). Pertinently, four of the six Ags recognized by male WB mouse sera underwent at least one of these changes. Vimentin was reported to be cleaved by caspases during programmed cell death (37), and, in addition, to be targeted by autoantibodies produced by normal BALB/c mice immunized with apoptotic Jurkat cells (38), thereby suggesting the presence of vimentin at the surface of apoptotic blebs. RAD23 was identified as an apoptosis-associated protein cleaved by caspase 3 in a human Burkitt’s lymphoma cell line incubated with an anti-IgM Ab (39). Nucleophosmin has been shown to undergo modifications during apoptosis (40) and translocation of hsp 60 to the cell surface membrane of T cells undergoing apoptosis was also described recently (41). One should note that, while five of the six proteins can undergo phosphorylation, another modification of cell components that also occurs during programmed cell death and confers the ability to induce autoantibody production in SLE patients (42, 43), and some of them were reported to be cleaved by caspases (37, 39), none of these proteins express the VGPD proteolytic site of granzyme B.
Another characteristic feature of Ags targeted by autoantibodies produced in SLE is that they are often constituted of physically linked components formed by the association of a nucleic acid and a nucleic acid-binding protein, for example, DNA and histones or RNA and RNP. Pertinently, three of the six autoantigens bound by autoantibodies detected in male WB mouse sera, hnRNP L (44), nucleosphosmin (45), and UV excision-repair protein RAD23 (46), are RNA-binding proteins involved in RNA metabolism.
This study also showed that the breaking of the tolerance of B cell clones toward various self-binding properties did not occur at the same time, but in a stepwise manner. In addition, the time of appearance of a given autoantibody population was not random but, in contrast, preordained since it was very similar from one male WB mouse to another. An intriguing observation, which contributes to the understanding of the initiation and expansion of the autoimmune response in this mouse, is the demonstration that vimentin was the first autoantigen to which an immune response developed. Indeed, anti-vimentin Abs preceded the appearance of any other autoantibody populations, including anti-CL and anti-DNA Abs. The search for the initial immunogen in SLE has been the subject of intensive investigation. For example, in MRL lupus mice, while anti-DNA Abs constitute the hallmark of the disease, recent studies showed that anti-nucleosome/subnucleosome Abs appeared before anti-DNA and anti-histone Abs (47), thus suggesting that reactivity to nucleosomes is followed by spreading to histones and DNA. This concept was further supported by the demonstration that preautoimmune lupus-prone (SWR x New Zealand Black)F1 mice, immunized with histone-derived peptides developed higher titers of anti-DNA and anti-nucleosome Abs (12). Much closer to our model is the recent demonstration that nucleolin, a nucleolar phosphoprotein with nucleic acid-binding properties, was the first target Ag of autoantibodies produced in MRL/lpr-lpr lupus-prone mice (48). This latter observation and our present findings lead one to think that the initial immunogen in SLE is made of nuclear or cytoplasmic protein(s) different from nucleosomes and may provide insights into the understanding of the development of the autoimmune response. In particular, the demonstration that immunity to other target Ags in male WB mice develops after a response to vimentin has been established, raises the hypothesis that the evolution of the immune response from vimentin to CL, DNA, and other proteins bound by male WB mouse sera occurs through the process of Ag spreading. However, it is very likely that certain but not all autoantigens have the ability to elicit Ag spreading. In this regard, in female WB mice, no ordered pattern of the autoantibody response was observed suggesting that initial immunity to certain of the targeted autoantigens can or cannot display properties leading to subsequent spreading.
It is widely accepted that Ag spreading involves Ags which are physically linked and clustered within a particular anatomic site (49). In this regard, one should recall, as discussed above, that most of the autoantigens recognized by male WB mouse sera are transported as physically linked epitopes in apoptotic blebs released during programmed cell death. For example, early apoptosis is accompanied by changes in cytoskeletal organization, including vimentin aggregation which occurs simultaneously with phosphatidylserine surface exposure (49). It cannot be excluded that vimentin and CL interact at the surface of apoptotic cells to form a typical SLE immunogenic particle and induce the production of three categories of autoantibodies anti-vimentin, anti-CL, and Abs with dual specificities similar to those of certain categories of anti-nucleosome or anti-Sm murine mAbs derived from lupus mice (50, 51). Reactivity to vimentin could be followed by spreading to other self-Ags present at the surface of apoptotic cells, namely hsp 60 and RAD23, which are both translocated to the surface of these cells (39, 41). However, the particular kinetics of autoantibody specificity appearance remain to be explained. It is very likely that several factors contribute to the orderly pattern of expansion of the male BW mouse autoimmune response including, the colocalization or linkage of self-Ags at the surface of apoptotic blebs, the ability of B cell receptors to acquire self-binding specificity, the number of B cell precursors, and the genetic background, e.g., the Yaa gene and H-2 alleles, which have been demonstrated to play a critical role in the initiation and spreading of self-reactivities (8).
This study also showed that combining 2D PAGE and mass spectrometry
constitutes a powerful tool not only to identify the whole autoantibody
pattern of an autoimmune disease but also to study the diversification
of the autoimmune response over time. However, one may wonder whether
2D PAGE and immunoblotting enabled identification of the full-blown
autoimmune response directed against protein components of the HL-60
cellular extract. Indeed, certain limitations of the technique might
account for incomplete detection of autoantibody reactivities present
in WB sera. First, the prolonged focusing time required for the
first-dimension isoelectric focusing separation renders proteins less
soluble, meaning that hydrophobic components could be absorbed onto the
gel matrix during the procedure and not appear in the second gel.
Second, lupus autoantibodies often react with cellular components which
have extreme pI values and that fall far beyond the range of the pH
gradient of 3.5–10 used in this study. Third, the denaturing
conditions used in 2D PAGE to improve protein solubility may lead to
irreversible loss of native protein conformations and thus
antigenicity. The use of a human promyelocytic cell line HL-60 as
substrate could also incite to consider that tissue- or
species-specific critical autoantigens may have been missed by the
methodological approach. However, 2D PAGE constitutes at present the
most powerful technique to separate and isolate proteins contained in a
cellular extract and thus certainly provides at present an ideal
procedure to identify the different autoantibody populations contained
in a single serum and the diversification of an autoimmune response as
illustrated by the present study. In this regard, the validity of the
procedure was confirmed by the demonstration that four of the six
immunoreactive Ags identified in the HL-60 protein map were also bound
by male WB mouse sera when obtained in a purified form. The lack of
reactivity of WB sera with purified -enolase still remains
unexplained but the identical staining pattern given by male WB sera
and goat anti--enolase Abs on the 2D gel replica argues for the
presence of this autoantibody specificity in WB sera (data not
shown).
Finally, characterization of the self-binding specificities and the kinetics of the autoimmune response in male WB mice provide a useful model to further our understanding of the initiation and development of the autoimmune process in SLE and the anti-phospholipid syndrome.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. François Tron, Institut de la Santé et de la Recherche Medicale Unité 519, Faculté de Médecine et de Pharmacie, 22 boulevard Gambetta, 76183 Rouen Cedex, France. E-mail address: francois.tron{at}chu-rouen.fr
3 Abbreviations used in this paper: SLE, systemic lupus erythematosus; RNP, ribonucleoprotein; Yaa, Y chromosome-linked autoimmune acceleration; CL, cardiolipin; 2D, two-dimensional; MS, mass spectrometry; hsp, heat shock protein; pI, isoelectric point; PVDF, polyvinylidene difluoride; MALDI-TOF, matrix-assisted laser desorption/ionization-time of flight; HMG, high mobility group; IPG, immobilized pH gradient.
Received for publication November 20, 2001. Accepted for publication July 31, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-enolase in systemic autoimmune disorders. J. Rheumatol. 27:109.[Medline]
This article has been cited by other articles:
|
|
 |
|
  Y Dai, C Hu, Y Huang, H. Huang, J Liu, and T Lv A proteomic study of peripheral blood mononuclear cells in systemic lupus erythematosus Lupus, September 1, 2008; 17(9): 799 - 804. [Abstract] [PDF]
|
|
|
|
 |
|
 H. S. Leong, B. M. Mahesh, J. R. Day, J. D. Smith, A. D. McCormack, G. Ghimire, T. J. Podor, and M. L. Rose Vimentin autoantibodies induce platelet activation and formation of platelet-leukocyte conjugates via platelet-activating factor J. Leukoc. Biol., February 1, 2008; 83(2): 263 - 271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Mahesh, H.-S. Leong, A. McCormack, P. Sarathchandra, A. Holder, and M. L. Rose Autoantibodies to Vimentin Cause Accelerated Rejection of Cardiac Allografts Am. J. Pathol., April 1, 2007; 170(4): 1415 - 1427. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Moisan and D. Girard Cell surface expression of intermediate filament proteins vimentin and lamin B1 in human neutrophil spontaneous apoptosis J. Leukoc. Biol., March 1, 2006; 79(3): 489 - 498. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Dugue, R. Perraut, P. Youinou, and Y. Renaudineau Effects of Anti-Endothelial Cell Antibodies in Leprosy and Malaria Infect. Immun., January 1, 2004; 72(1): 301 - 309. [Abstract] [Full Text] [PDF]
|
|
), 90 (
), 120 (•), and 180 (
).
