| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Pediatrics, University of Utah School of Medicine, Salt Lake City, UT 84132;
Division of Rheumatology, Scripps Clinic and Research Foundation, La Jolla, CA 92037; and
Department of Pediatric Cardiology and The Oklahoma Children’s Heart Center and
§
Department of Microbiology and Immunology, University of Oklahoma Health Sciences Center, Oklahoma City, OK 73190
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-helical proteins are characteristically found in streptococcal
diseases and sequelae such as acute rheumatic fever
(ARF)3 (1, 2, 3, 4). Abs that
recognize Ags shared between Streptococcus pyogenes and
heart tissues may play a role in the autoimmune cardiac injury
associated with ARF. Previous studies of murine cross-reactive
anti-myosin/anti-streptococcal Abs have classified the Abs into
three groups. The groups include: 1) those
anti-streptococcal/anti-myosin mAbs reacting with -helical
molecules including myosin, tropomyosin, and vimentin (5, 6); 2)
anti-nuclear Abs (5); and 3) mAbs that react with myosin and the
carbohydrate N-acetylglucosamine (7, 8, 9). In humans with ARF,
anti-nuclear Abs are not characteristically found in the serum.
Furthermore, human anti-streptococcal/anti-myosin mAbs produced
from ARF do not have anti-nuclear reactivity but demonstrate strong
reactivity with the group A carbohydrate epitope,
N-acetylglucosamine, as shown in this report. In ARF Abs and
complement were shown to be deposited in hearts of patients who died
from the disease. In our studies, murine
anti-streptococcal/anti-myosin mAb 36.2.2 was shown to be
cytotoxic and to react with extracellular molecules on the surface of
heart cells (10). Therefore, anti-myosin Abs capable of recognizing
surface molecules may play a role in development of disease
manifestations observed in ARF. Only one of the human mAbs reported
herein was cytotoxic for heart cells in culture. Cytotoxic Abs have the
potential for deposition in tissues and initiation of inflammation. Anti-myosin Abs have also been associated with chronic autoimmune heart disease such as myocarditis (11, 12). It was reported that anti-myosin Abs were present in animals developing coxsackieviral-induced or cardiac myosin-induced myocarditis (11, 12, 13, 14, 15). Studies have shown that coxsackievirus, streptococcal M protein, and cardiac myosin share epitopes that contribute to antigenic cross-reactivity as well as disease production (15, 16, 17, 18). Cardiac myosin was shown to play a major role in myosin-induced myocarditis models such as the A/J mouse strain (19, 20). In this model, myocarditis was dependent on the presence of CD4+ lymphocytes (20). However, in the DBA/2 mouse model of myocarditis, it was shown that passive administration of mouse IgG anti-myosin mAb produced myocarditis and concomitant deposition of the Ab (13). Ab deposition in the extracellular matrix of the cardiac tissues was explained by the expression of a 200-kDa molecule in the susceptible tissue. The expression of myosin or a myosin-like molecule was believed to be the susceptibility factor in the DBA/2 animals. Other animals not expressing the molecule did not develop myocarditis after passive administration of the mouse anti-myosin Ab (13).
In this report, we identify the autoantigens recognized by the human
anti-myosin Abs from patients with rheumatic carditis and correlate
this information with the nucleotide sequences utilized for their Ig
genes. We recently reported that an
anti-streptococcal/anti-myosin cross-reactive human mAb
(10.2.3) derived from normal human tonsil was encoded by V3-23
VH and L9 V
variable segments (21) and that
cytotoxic anti-streptococcal/anti-myosin mouse mAb 36.2.2 was
encoded by VH7183 and V8 (10). In our report, human IgM
anti-streptococcal/anti-myosin mAbs 1.C8, 1.H9, 5.G3, 3.B6, and
10.2.5 are shown to react strongly with the
N-acetylglucosamine determinant of the group A streptococcal
carbohydrate, with human cardiac myosin, other related self Ags, and
streptococcal M proteins. Elevated reactivity of ARF sera with the
group A carbohydrate has been reported to be associated with persistent
valvular heart disease (22). The human anti-streptococcal mAbs
1.C8, 1.H9, and 5.G3 have been characterized for their reactivity with
keratin and keratin peptides, a feature of Abs reactive with
N-acetylglucosamine (8). The nucleotide sequences of their
Ig VH and VL region genes now reveal that the
five mAbs analyzed are encoded by a heterogeneous group of
preferentially expressed VH and VL genes.
Rearranged Ig VH and Ig VL genes exhibit little
evidence of Ag-driven somatic mutation irrespective of germline origin
and pathogenic potential.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Patients were treated in the department of pediatric cardiology
at the University of Oklahoma Children’s Heart Center, and ARF was
diagnosed on the basis of the Jones criteria. All patients had carditis
with heart murmur, and in one instance a patient had both carditis and
arthritis. In addition, serology revealed elevated
anti-streptolysin O titers of 
600 in all cases. Two of the
hybridomas, 1.C8 and 1.H9, were generated from the same patient, while
hybridomas 5.G3 and 3.B6 were derived from two other patients. The
hybridoma secreting mAb 10.2.5 was derived from a tonsil of a normal
individual.
Human hybridomas
Human hybridomas utilized in this study were produced as described by Shikhman and Cunningham (2, 3, 8). Human PBL were isolated using Histopaque-1077 Hybri-Max (Sigma, St. Louis, MO) and were stimulated for 1 week at 37°C with streptococcal peptidoglycan-polysaccharide complexes (10 µg/ml) (mAbs 1.C8, 1.H9, 5.G3, and 3.B6) as previously described (8) or with streptococcal M5 membranes (mAb 10.2.5) (3). After 1 week, Iscove’s modified Dulbecco’s medium (IMDM) containing the peptidoglycan-polysaccharide complexes and 10% human AB serum was replaced with IMDM containing 10% serum and pokeweed mitogen (1 µg/ml). The cells were stimulated for an additional week. Cells were then collected by centrifugation, washed three times in IMDM without serum, and fused by standard methods as described with either HMMA2.11TG/O, a rat-human heterohybridoma purchased from Dr. M. R. Posner, Brown University, Providence, RI, or with WI-L2-729HF2, a human lymphoblastoid cell line purchased from Dr. R. Lundak, Techniclone International, Santa Ana, CA (3, 8). Hybridomas were selected on the basis of reactivity with group A streptococci, myosin, and N-acetylglucosamine-BSA with no reactivity with BSA as previously described (7, 8). Positive hybridomas were cloned three times by limiting dilution in 96-well tissue culture plates. Hybridomas were grown in IMDM plus 20% FBS.
Synthetic peptides
The light meromyosin (LMM) tail region of the human cardiac myosin heavy chain was synthesized in overlapping 18-mer peptides with a 5-amino acid overlap. The sequence of human cardiac myosin ß-chain was taken from the sequence published by Vosberg et al. (23). In some cases, the peptides were synthesized longer or shorter to facilitate purification of the peptide by HPLC or to prevent cyclization. All peptides were chemically synthesized on a Dupont RAMPS manual synthesizer using the f-moc strategy (24) and purified by HPLC. The synthetic LMM peptides have been reported previously (21), and their amino acid sequences are shown in Results. The synthetic peptides were tested for reactivity with the human mAbs in immunodot blots as previously described (3).
Enzyme-linked immunosorbent assay
A variety of streptococcal and self Ags were tested against the mAbs in the ELISA. Purified human cardiac myosin was prepared according to a procedure described previously (25). Skeletal myosin from rabbit muscle, tropomyosin from rabbit muscle, vimentin from bovine lens, LMM from rabbit muscle, heavy meromyosin from rabbit muscle, actin from rabbit muscle, keratin from human epidermis, mouse laminin, BSA, ss-DNA and ds-DNA were obtained from Sigma. Synthetic polynucleotides were obtained from Pharmacia (Piscataway, NJ), and elastin from bovine neck ligament was obtained from Elastin Products. Streptococcal recombinant M5 and M6 proteins were gifts from Dr. James B. Dale (VA Medical Center, Memphis, TN) and Dr. Vincent A. Fischetti (Rockefeller University, New York, NY). Phosphorylcholine conjugated with BSA was provided by Dr. David Briles (University of Alabama, Birmingham, AL). N-Acetyl-ß-D-glucosamine (GlcNAc) conjugated to BSA was produced by a two-step reaction as previously described (14). Ags were tested at a concentration of 10 µg/ml in 0.1 M sodium carbonate buffer, pH 9.0, adsorbed to 96-well Immulon-4 microtiter plates (Dynatech, Alexandria, VA). The plates were coated with Ag overnight at 4°C and then blocked with 1% BSA in PBS containing 0.5% Tween 20. The ELISA was performed according to standard protocol. The secondary Ab was peroxidase labeled goat anti-human IgM. H2O2 and o-phenylenediamine (Sigma) were added finally as the peroxidase substrate to detect Ab binding. Results were calculated from triplicate measurements. Assays were performed several times with the human mAbs in the ELISA.
Western immunoblots
Human cardiac myosin as well as other proteins were separated by discontinuous SDS-PAGE by the method described by Laemmli (26). The stacking gel contained 4% acrylamide and the separating gel contained 7.5% acrylamide. Proteins were separated by SDS-PAGE electrophoresis using the SE 280 11-cm vertical slab gel unit (Hoeffer Scientific Instruments, San Francisco, CA) with 1.5-mm-thick gel. After electrophoresis, the gel was overlayed with a sheet of Immobilon-NC (Millipore, Bedford, MA), and the proteins were transferred overnight at 50 mA current in the transfer unit TE22 (Hoeffer Scientific Instruments). Efficiency of the protein transfer was determined by the transfer of prestained m.w. standards (Sigma). In addition, a portion of the blot was stained with amido black, and the transferred bands were observed after destaining. The blot was blocked with PBS containing 2% Tween and 2% BSA and was then washed and subsequently incubated with the human mAbs overnight at 4°C. After washing, a peroxidase-labeled goat anti-human IgM (Kierkegaard and Perry Laboratories, Gaithersburg, MD) was incubated with the blot, and after washing a TMB membrane substrate system (Kierkegaard and Perry) was used to develop the blots.
51Cr release assay for cytotoxicity
A primary rat heart cell line (ATCC-CRL-1446) was cultured overnight in sterile 96-well microtiter plates in IMDM containing 20% FBS. Cells were cultured initially at 1 x 104 cells/well at 37°C and 5% CO2. The cells were labeled with Na251CrO4 (Dupont) at 5 µCi/well (1 Ci = 37 GBq) for 2 h at 37°C. Culture medium was removed and the attached cells were washed 3 times with IMDM plus 20% fetal bovine serum and incubated for 1 h in IMDM with serum. Ab was added to the cells at 50 µl/well for 1 h at 37°C. An equal volume of guinea pig complement (Whittaker Bioproducts, Walkersville, MD) was added, and the mixture was incubated for 1 h at 37°C. Supernatant fluid was harvested from the 96-well plate using a Skatron harvester system (Sterling, VA) and 51Cr release was measured in an LKB gamma-counter (LKB-1282 Compugamma; LKB Instruments, Gaithersburg, MD). Minimum lysis was calculated from measurements of supernatants from cells treated with culture medium alone, and 100% lysis was calculated from measurements of supernatants taken from cells treated with 1 M HCl for maximum 51Cr release. Percent lysis was determined by calculating [(test sample release - minimum release)/(maximum release - minimum release)] x 100. Tests were performed in quadruplicate, and the average and SD determined for each sample tested.
Cloning and sequencing of Ig variable region genes
Ig variable region genes were cloned and sequenced following
amplification by PCR. Total RNA was prepared from 106 to
108 hybridoma cells by the phenol-guanidine isothiocyanate
method (TRIzol Reagent; Life Technologies, Gaithersburg, MD) according
to the manufacturer’s protocol. A 5-µg sample of total RNA was
primed for first-strand synthesis with 10 pmol of an antisense human JH
or antisense human C or C
primer as previously described, or by
using commercially available human Cµ and C primers (Ig-Prime;
Novagen, Madison, WI) (27, 28). First-strand cDNA synthesis was
performed with recombinant Moloney murine leukemia virus reverse
transcriptase (Superscript II; Life Technologies) using the
manufacturer’s protocol, except that DMSP (6%) was added to each
reaction. Variable region genes were amplified as previously described,
using sense primers corresponding to each human heavy chain variable
region (VH) family leader sequence or degenerate primers
homologous to or variable region (V/V) leader sequences
and antisense JH or internal C/C primers, respectively (27, 28).
Amplification products were ligated into pBS phagemid vectors
(Stratagene, La Jolla, CA) using standard protocols and both strands of
inserts sequenced by the dideoxy technique (29). A minimum of two
identical sequences from two different PCR reactions were obtained for
all variable region genes except for the 1.H9 V segment. Variable
region gene sequences were compared with those entered into the GenBank
database.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human mAbs 1.C8, 1.H9, 5.G3, and 3.B6, derived from patients at
the onset of ARF, and human mAb 10.2.5, derived from a human tonsil,
reacted with N-acetylglucosamine, a dominant epitope of the
group A carbohydrate, and with -helical structures such as myosin,
keratin, or streptococcal M protein (Fig. 1
). Only mAb 3.B6 reacted with
tropomyosin (Fig. 1D). Western blots demonstrated the strong
reactivity of mAbs 1.C8, 1.H9, 3.B6, and 10.2.5 with purified human
cardiac myosin in the Western blot, while mAb 5.G3 did not react with
purified human cardiac myosin in the Western blot. Figure 2 shows the reactivity of mAbs 1.C8 and
1.H9 with purified human cardiac myosin in the Western blot. mAbs 3.B6
and 10.2.5 reacted similarly with human cardiac myosin heavy chain in
the immunoblots (data not shown). Differences in reactivity of the two
mAbs 1.C8 and 1.H9 was observed when the Abs were compared for
reactivity with vimentin in the Western blot as shown in Figure 3 and with keratin and human skin as
previously reported (8). mAb 1.C8 reacted with vimentin in the Western
blot (Fig. 3) and also with tissue sections whereas mAb 1.H9 did not
(8). Despite these differences, mAbs 1.C8 and 1.H9 reacted strongly
with human cardiac myosin in Western blots, and the specificity of the
two mAbs for human cardiac LMM synthetic peptides was nearly identical
in an immunodot blot (Table I). Rheumatic
carditis mAbs 1.C8 and 1.H9 reacted most strongly with LMM 21, 28, 33,
and 46, and rheumatic carditis mAb 3.B6 reacted strongly with LMM 1 and
33. All three rheumatic carditis mAbs reacted with LMM peptide 33.
However, mAb 10.2.5 from a normal tonsil reacted most strongly with LMM
10, 23, 27, and 30. Two other mAbs produced from a streptococcal
carrier reacted with the same peptides recognized by the mAb 10.2.5.
Although this is a small number of mAbs to compare, the data suggest
that the peptides recognized by the Abs from rheumatic heart disease
were distinctly different from those mAbs isolated from normal
individuals. Among the different peptides binding a single Ab, none
contained identities in amino acid sequence. However, structural or
conformational similarities may exist among the peptides due to
positions of hydrophobic and hydrophilic amino acid residues
characteristic of helices. The conjugate control was negative with
all of the LMM peptides in the immunoassay. Differences in reactivity
of mAbs 1.C8 and 1.H9 were also observed in their abilities to bind to
GlcNAc:BSA conjugates containing different numbers of GlcNAc residues
conjugated to BSA. Figure 4 shows that
mAb 1.H9 was more efficient at binding lower numbers of GlcNAc
molecules on BSA than 1.C8. In additional studies, the 51Cr
release assay was used to determine whether any of the human mAbs were
cytotoxic for heart cells. mAb 1.H9 was moderately cytotoxic for rat
heart cells in culture in the presence of complement, as shown in
Figure 5. The other human mAbs were not
cytotoxic for rat heart cells.
|
|
|
|
|
|
Germline genes most closely homologous to those encoding mAb IgVH
and IgVL and the nucleic acid and amino acid sequence homology to these
genes are noted in Tables II and
III. In Figure 6A, the nucleotide sequence
and the translated amino acid sequence of the mAb 3.B6 heavy chain
variable region gene is shown. The sequence is identical with the
germline 4–59 VH, DXP'1, and JH4 genes (30, 31). In
Figure 6B, the nucleotide sequence and the translated amino
acid sequence of the mAb 3.B6 light chain variable region gene is
shown. The sequence is compared with the germline A27 and J3 (32).
|
A)
(33, 34, 35). Recently, two laboratories have cloned the entire
VH locus (30, 36). A germline VH segment
identical with VH3-8 was not identified in these studies, but the VH3-8
germline gene and each of these rearranged genes are also 97%
homologous to the V3-11 (DP-35/22-2B). VH3-8 may, therefore, represent
an allele of V3-11. Alternatively, like certain other closely related
VH segments that appear to have arisen through duplication
events (notably V3-30, V3-30b, and V30.3), VH3-8 may be located on an
insertion polymorphism present in only certain haplotypes (36, 37). In
Figure
7B, the nucleotide sequence and the translated
amino acid sequence of the mAb 5.G3 light chain variable region gene is
shown. The sequence is compared with the germline 012/02 VH
and J4 (38).
|
A) (30). The 1.C8 V
segment is 98% homologous to the DPL11 germline V gene segment and
uses a J2 or 3 gene (Fig. 8B) (39, 40). A total of 12
clones from 4 amplification reactions were obtained for the 1.H9 V
gene, the variable segment of which shared 99% homology with the
germline DPL3 gene (39). In each clone, we observed a single base pair
deletion, resulting in a frame shift at the VJ joint (Fig. 8B). This deletion was present on sequencing both sense and
antisense strands and with alternative sequencing methods (PCR). No
other rearrangement was detected in any amplification reaction.
Unfortunately, this cell line was unstable, and we were unable to
examine heavy or light chain rearrangements by Southern blotting (data
not shown). It is not clear, therefore, whether this transcript
represents the functional rearrangement with an unusual sequencing
artifact, whether it is a highly expressed nonproductive rearrangement,
or whether this deletion may represent an acquired mutation in the 1.H9
cell line (which might also affect the stability of the line).
|
genes are encoded
by one of the duplicated L9 germline V segments and J1 genes;
however, the 10.2.5 V segment has three nucleotide differences from
the 10.2.3 gene, including a single coding change in
complement-determining region (CDR) 2 that may be responsible
for the very subtle changes in reactivity between the two mAbs (Fig. 9) (21).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
50 µg/ml IgM and
300 µg/ml IgG, while sera from normal subjects and uncomplicated
streptococcal infection contain 12 µg/ml IgM and IgG
anti-myosin Ab as acertained from human cardiac myosin affinity
column calculations (M. W. Cunningham, unpublished data). In this
report, we investigated anti-streptococcal/anti-myosin mAbs,
produced from patients with rheumatic carditis, for their reactivity
and specificity for various streptococcal and tissue Ags, myosins, and
LMM peptides. In addition, we analyzed the nucleotide sequence of their
variable region VH and VL genes.
All of the human anti-streptococcal/anti-myosin mAbs were
strikingly reactive with the group A streptococcal carbohydrate epitope
N-acetylglucosamine. This is an important feature since
elevated and persistent levels of Abs against group A streptococcal
carbohydrate were previously reported in cases of chronic rheumatic
valvulitis (22). In addition, the mAbs were highly reactive with myosin
and its fragments as well as other -helical coiled-coil proteins.
The mAbs did not react with DNA as might be expected, since patients
with ARF do not characteristically develop antinuclear Abs. The most
interesting mAb was 1.H9 which demonstrated moderate cytotoxic activity
for rat heart cells in the presence of complement. These Abs were
polyreactive with nonidentical peptide structures of human cardiac
myosin and the carbohydrate N-acetylglucosamine. Although
the peptides contained no common sequences, the Ab combining site may
accommodate these Ags due to its flexibility. Recently, it was proposed
that a germline antibody may be polyreactive due to conformational
rearrangement and configurational change, permitting binding of diverse
molecules such as carbohydrates and peptides (42).
Human anti-streptococcal/anti-myosin Abs were encoded by a heterogeneous population of Ig variable region genes. This repertoire, however, was striking in that it was encoded by a group of variable region gene segments, including V3-23, V4-59, V3-11, and V3-30, which are preferentially expressed beginning in early life, and in protective and autoimmune responses (34, 43, 44, 45, 46).
DXP family members are highly represented among the variable region D genes in the human anti-streptococcal mAbs. In the adult repertoire, the DXP family is used almost twice as frequently as expected based on its germline complexity (47). Two of the mAbs, 5.G3 and 3.B6, could not be assigned. The mAb 5.G3 D segment shares some homology with DXP'1, and the short D segment of mAb 3.B6 shares homology with an inverted DIR2 element; however, the germline origin of these D segments is not certain (47, 48). In the case of mAb 5.G3, which has a relatively long CDR-3, the VDJ may have resulted from a complex recombination event such as D-D fusion or inversion. Three of five of the anti-streptococcal mAbs were encoded by JH4, and none by the infrequently rearranged JH1 and JH2 germline joining segments. In all human developmental stages, the JH4 segment is markedly overutilized, encoding 45 to 65% of heavy chain rearrangements (47, 49). The mean length of the anti-streptococcal mAb CDR-3 was 44 nucleotides, with a range of 27 to 63 nucleotides, and in each gene in which the germline D segment is identifiable, potential "N" and "P" segments are present. The molecular characteristics of these cross-reactive anti-streptococcal Ig VH are more like the Ig VH of the normal adult repertoire than the early human Ab repertoire.
Considerably less is known about human light chain expression than that
of heavy chains, but in mice V gene usage appears nonstochastic (50, 51). A recently reported compilation of expressed human V genes also
suggests that some V segments, conspicuously A27 and 012/02, are
highly expressed in random libraries and in protective and autoimmune
responses (52). V genes for A27- and 012/02-encoded VL
chains for mAbs 3.B6 and 5.G3, respectively.
The early Ab repertoire is characterized by striking poly- and
autoreactivity (53). Like the fetal repertoire, the variable region
genes that we have identified in the human cross-reactive
anti-streptococcal/anti-myosin repertoire were encoded by a
small group of preferentially expressed germline genes. Each of these
segments also encodes other autoantibodies, such as anti-DNA, RF,
anti-i, and others (54, 55, 56). This report of cross-reactive Abs
directed against both foreign and self Ags and studies of other
autoimmune and physiologic immune responses suggest that ligand binding
may be an important force shaping the expressed Ab repertoire (2, 21, 57, 58). Ab molecules encoded by preferentially expressed germline
genes may have the potential to reconfigure and bind to a wide
array of three-dimensionally similar ligands ("molecular
mimicry") such as streptococcal M protein, myosin, other
-helical proteins, and carbohydrate epitopes (7, 8, 9, 59).
A disadvantage of a polyreactive repertoire is the potential for the
emergence of autoimmune disease. Despite extensive study, the
relationship among variable region gene usage, "natural"
polyreactive Abs, and pathogenic autoantibodies is unclear.
Conventionally, "natural" autoreactive Abs are of the IgM isotype,
are polyspecific, have low affinity interactions with Ag, and are
encoded by variable region genes identical with or very close to
germline configurations, whereas Abs implicated in the pathogenesis of
autoimmune disease are of the IgG or IgA isotype, are monospecific,
have high affinity, and have evidence of Ag-driven somatic mutation.
The anti-streptococcal/anti-myosin mAbs reported herein are of
the IgM isotype and are encoded by variable region genes closely
homologous to or identical with germline sequences. Overall R:S ratios
range from 0:0 to 10:5, CDR R:S ratios range from 0:0 to 8:5, and
FR R:S ratios range from 0:0 to 2:1. Only genes encoding mAbs
1.C8 and 1.H9 show significant mutation, and only the mAb1.C8 V
segment has some evidence suggestive of Ag selection, with an overall
R:S ratio of 4:1 (but CDR and framework ratios 2:0 and 2:1,
respectively).
Numerous exceptions to generalizations about pathogenic and natural
autoantibodies have been reported. Low affinity autoantibodies may be
pathogenic, and Ag-selected responses need not demonstrate the
molecular characteristics of Ag-driven selection (60). It is possible
that the IgM anti-myosin Abs reported from rheumatic carditis could
be part of the natural autoantibody population and quite distinct from
those involved in disease. However, in Table I, the mAbs 1.C8, 1.H9,
and 3.B6 from rheumatic carditis patients demonstrated a distinctly
different peptide binding pattern from mAb 10.2.5 and other normal
anti-myosin mAbs, which all reacted with the same set of peptides.
In this study, mAb 1.H9 appeared to be cytotoxic for rat heart cells in
culture. There were no obvious features other than heavy-light chain
pairing to distinguish this Ab from the noncytotoxic Abs. We have also
characterized a murine anti-streptococcal/anti-myosin mAb
(36.2.2), which like the human mAbs described here reacts with myosin
and other structurally similar proteins and is cytotoxic to rat heart
cells (3, 16). The IgVH of the murine mAb 36.2.2 is encoded by a member
of the preferentially expressed 7183 VH family and is
highly homologous to genes encoding protective and
anti-carbohydrate immune responses (10). The development of
anti-myosin Abs recognizing cell surface molecules may be important
in the development of disease pathogenesis.
Autoimmune cardiac injury in most animal models occurs by a more complex mechanism than direct Ab-mediated cytotoxicity. Autoimmune myocarditis models induced by cardiac myosin immunization or by coxsackieviral infection have been shown to be dependent on T lymphocytes (20, 61, 62). In mice, myocarditis is strain specific, suggesting that genetic factors influence susceptibility to disease (63). In ARF, the relatively low frequency of disease in individuals exposed to rheumatogenic strains of S. pyogenes and the increased prevalence of ARF in related individuals support the hypothesis that genetic susceptibility plays a role in development of the disease (64, 65, 66). The deposition of Ab and complement in the heart in ARF supports a pathogenic role for Ab in the disease. Passively administered anti-myosin (IgG) Abs were shown to deposit in the hearts of DBA/2 mice which then developed cardiac lesions (13). Only DBA/2 mice expressed a 200-kDa protein in the extracellular matrix of the heart, which was suggested to bind the anti-myosin Ab and to deposit in hearts in the susceptible strain. Anti-myosin Abs of the IgM isotype did not deposit or produce cardiac lesions in the heart tissues of the mice. Ab deposition in the heart may depend on the IgG isotype. Autoantibodies like 1.H9 could potentially bind to epitopes in tissues of susceptible individuals. What is most clear from the studies relating gene expression to normal and pathogenic Ab repertoires is that given the great potential and cross-reactivity between protective and autoimmune Ab repertoires, the most important contributors to Ab-mediated autoimmune disease must be the inability to delete or suppress the expression of pathogenic Abs and the presence of the antigenic epitope in the target tissues. In the absence of the appropriate development of tolerance or clonal deletion, humoral immune responses elicited by streptococcal infection may have the potential to precipitate or exacerbate autoimmune cardiac disease in susceptible individuals.
|
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Madeleine Cunningham, Department of Microbiology and Immunology, University of Oklahoma Health Sciences Center, P.O. Box 26901, Oklahoma City, OK 73190. E-mail address:
3 Abbreviations used in this paper: ARF, acute rheumatic fever; IMDM, Iscove’s modified Dulbecco’s medium; LMM, light meromyosin; GlcNAc, N-acetyl-ß-D-glucosamine; CDR, complement-determining region.
Received for publication January 5, 1998. Accepted for publication April 9, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and V segments and VJ combinations. J. Clin. Invest. 89:729.
locus: characterization of the duplicated A regions. Eur. J. Immunol. 22:1023.[Medline]
-like and V pre-B) and present immunoglobulin V-D-J gene usage strongly biased to a fetal-like repertoire. J. Clin. Invest. 91:1616.
locus: characterization of extended immunoglobulin gene regions by cosmid cloning. Eur. J. Immunol. 193:241.
gene segments. J. Immunol. 23:1456.
light chain locus: organization and DNA sequences of three genomic J regions. Immunogenetics 25:63.[Medline]
chains, unlike genes encoding heavy chains, use variable gene segments dispersed throughout the locus. Proc. Natl. Acad. Sci. USA 86:6744. gene expression does not follow the VH paradigm. J. Exp. Med. 169:1859. segments reveals a strong bias in their usage. Eur. J. Immunol. 24:827.[Medline]
locus: characterization of the partially duplicated L regions. J. Immunol. 23:2860.
This article has been cited by other articles:
|
|
 |
|
  D. C. Kem, X. Yu, E. Patterson, S. Huang, S. Stavrakis, B. Szabo, L. Olansky, J. McCauley, and M. W. Cunningham Autoimmune Hypertensive Syndrome Hypertension, November 1, 2007; 50(5): 829 - 834. [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Kirvan, C. J. Cox, S. E. Swedo, and M. W. Cunningham Tubulin Is a Neuronal Target of Autoantibodies in Sydenham's Chorea J. Immunol., June 1, 2007; 178(11): 7412 - 7421. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Blank, I. Krause, L. Magrini, G. Spina, J. Kalil, S. Jacobsen, H. J. Thiesen, M. W. Cunningham, L. Guilherme, and Y. Shoenfeld Overlapping humoral autoimmunity links rheumatic fever and the antiphospholipid syndrome Rheumatology, July 1, 2006; 45(7): 833 - 841. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Edwards, W. Brouwer, C. H. Y. Choi, J. Ruhno, R. L. Ward, and A. M. Collins Analysis of IgE Antibodies from a Patient with Atopic Dermatitis: Biased V Gene Usage and Evidence for Polyreactive IgE Heavy Chain Complementarity-Determining Region 3 J. Immunol., June 15, 2002; 168(12): 6305 - 6313. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Galvin, M. E. Hemric, S. D. Kosanke, S. M. Factor, A. Quinn, and M. W. Cunningham Induction of Myocarditis and Valvulitis in Lewis Rats by Different Epitopes of Cardiac Myosin and Its Implications in Rheumatic Carditis Am. J. Pathol., January 1, 2002; 160(1): 297 - 306. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. W. Cunningham Cardiac Myosin and the TH1/TH2 Paradigm in Autoimmune Myocarditis Am. J. Pathol., July 1, 2001; 159(1): 5 - 12. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Quinn, S. Kosanke, V. A. Fischetti, S. M. Factor, and M. W. Cunningham Induction of Autoimmune Valvular Heart Disease by Recombinant Streptococcal M Protein Infect. Immun., June 1, 2001; 69(6): 4072 - 4078. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. C. Montalto, C. D. Collard, J. A. Buras, W. R. Reenstra, R. McClaine, D. R. Gies, R. P. Rother, and G. L. Stahl A Keratin Peptide Inhibits Mannose-Binding Lectin J. Immunol., March 15, 2001; 166(6): 4148 - 4153. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Malkiel, L. Liao, M. W. Cunningham, and B. Diamond T-Cell-Dependent Antibody Response to the Dominant Epitope of Streptococcal Polysaccharide, N-Acetyl-Glucosamine, Is Cross-Reactive with Cardiac Myosin Infect. Immun., October 1, 2000; 68(10): 5803 - 5808. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. W. Cunningham Pathogenesis of Group A Streptococcal Infections Clin. Microbiol. Rev., July 1, 2000; 13(3): 470 - 511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. J. Albert and R. D. Inman Molecular Mimicry and Autoimmunity N. Engl. J. Med., December 30, 1999; 341(27): 2068 - 2074. [Full Text] [PDF]
|
|
|
|
|
|
M. W. Cunningham, H. C. Meissner, J. S. Heuser, B. A. Pietra, D. K. Kurahara, and D. Y. M. Leung2 Anti-Human Cardiac Myosin Autoantibodies in Kawasaki Syndrome J. Immunol., July 15, 1999; 163(2): 1060 - 1065. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Quinn, K. Ward, V. A. Fischetti, M. Hemric, and M. W. Cunningham Immunological Relationship between the Class I Epitope of Streptococcal M Protein and Myosin Infect. Immun., September 1, 1998; 66(9): 4418 - 4424. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
