| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Medicine, Vanderbilt University School of Medicine, Nashville, TN 37232
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cell Ag, such as insulin, NOD mAb
have characteristics of natural autoantibodies that include low avidity
and broad specificity for multiple Ags. Analyses of the variable region
of Ig H chain (VH) and variable region 
L chain
genes expressed by six insulin binding mAb show that V gene
segments are often germline encoded and are identical with those used
by autoantibodies, especially anti-dsDNA, from systemic autoimmune
disease in MRL, NZB/W, and motheaten mice. VH genes used by
four mAb are derived from the large J558 family and two mAb use
VH7183 and VHQ52 genes. The third
complementarity-determining region of Ig H chain of these mAb have
limited N segment diversity, and some mAb contain DNA segments
indicative of gene replacement. Genetic abnormalities in the regulation
of self-reactive B cells may be a feature that is shared between NOD
and conventional systemic autoimmune disorders. In NOD, the large pool
of self-reactive B cells may fuel autoimmune cell destruction by
facilitating T-B cell interactions, as evidenced by the identification
of one mAb that has undergone Ag-driven somatic
hypermutation. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cell destruction and type I or insulin-dependent
diabetes mellitus (T1DM). As in human T1DM, the disorder in NOD is
under polygenic control with a dominant contribution from genes in the
MHC complex (1, 2). The principal mediators of cell
destruction are believed to be T lymphocytes. Although a direct role
for autoantibodies in disease has not been found, detection of Abs to
cell Ags in the prodrome of T1DM suggests that loss of tolerance in
the B cell compartment is one of the earliest indicators of the covert
autoimmune process (3, 4, 5, 6). Autoantibodies to insulin have
predictive value in both NOD and human T1DM (5, 7), and
insulin-specific T cells can transfer disease (8). Studies
using Ab reagents that block B cell development (9, 10)
and genetic deletion of B cells in µMT mice provide evidence that B
lymphocytes are necessary for the development of diabetes in NOD mice
(11, 12). Elegant studies also show that
I-Ag7 deficiency confined to the B cell
compartment protects NOD mice from T1DM and suggest that B cells play a
key role in mediating T cell responses as a consequence of Ag
presentation (13).
Although the importance of B cells for the development of insulitis and
diabetes in the NOD mouse model of type I diabetes is well-recognized,
data on the expressed B cell repertoire in NOD are limited. Studies
from the Holmberg laboratory used in situ hybridization to show that
the neonatal bias for expression of D-proximal variable
region of Ig H chain (VH) genes persists in adult
NOD mice (14, 15). Pleau et al. (16) examined
the VH and variable region L chain
(V) structures of two IgG mAb selected
for binding to insulin and peripherin in diabetic NOD mice and found
nucleotide replacements suggesting Ag-driven somatic mutation. These
observations indicate that both developmental abnormalities and
autoantigen-driven selection may contribute to the functional B cell
repertoire in NOD mice.
To better understand how developmental programs and Ag-driven selection contribute to the B cell repertoire in prediabetic NOD mice, we characterized the structure and function of a panel of autoantibodies generated by hybridoma formation. In the absence of mitogen or Ag stimulation, this technique favors the capture of spontaneously active B cells (17, 18). Using this approach, we find that the B cell repertoire of NOD mice is highly active when compared with that of nonautoimmune mice. This active repertoire is present from an early age (4 wk) and is not dependent on gender. Because insulin is a known autoantigen in T1DM, we examined a large number of NOD hybridomas in search of high-affinity autoantibodies. In contrast to expectation, we find a high proportion of insulin-binding mAb in NOD (including IgG mAb) have characteristics of a natural autoantibody repertoire. This observation is based on the low avidity of insulin-binding in competitive ELISA and on the ability of mAb captured for insulin-binding to interact with multiple other Ags including DNA and Ig. Further support for this conclusion is based on the nucleotide composition of V regions from these mAb. The V region genes expressed by these mAb are principally unmutated (germline) structures that are often identical with V genes expressed by Abs (especially anti-DNA) in systemic autoimmune disease such as murine lupus. Although most V genes are in germline configuration, data on one mAb show clear evidence for somatic hypermutation and indicate that the active B cell repertoire is available to interact with helper T cells. These findings suggest that the genetic background of NOD mice favors generation of autoreactive B cells and imply that defects in B cell selection may be present in NOD. The recognition that lupus-like syndromes may accompany therapies that protect NOD mice from developing diabetes (19) also indicates that the active B cell repertoire in NOD has a broad pathological potential. Thus, genetic abnormalities in the regulation of self-reactive B cells may be shared between NOD mice and strains of mice that have systemic autoimmune disease.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
NOD, BALB/c, and C57BL/6 mice were purchased from Taconic Farms (Germantown, NY) and used between the ages of 4 and 24 wk. To produce hybridomas, spleens were removed from mice following sacrifice and fused with the mouse myeloma line X63Ag8.653 in a ratio of 1:4 using polyethylene glycol (20). No immunization or mitogen activation was used. Following fusion, cells were plated at 2 x 106/ml in 96-well flat-bottom plates and selected in hypoxanthine-aminopterin-Tdr medium (Sigma-Aldrich, St. Louis, MO). After removal of selection media, growth-positive wells were screened for production of IgM or IgG Abs by a capture ELISA. To score for frequency of hybridoma production, the number of growth positive wells that produced Ab on two screens were counted and the results reported as number of hybridomas per 105 spleen cells used in the fusion. In some experiments, supernatants were also screened for insulin binding and positive wells were cloned at limiting dilution.
Immunoassays
ELISAs were used to identify Ab-producing hybridomas and to
measure specific Ag Ab interactions. Briefly, to detect Ab production,
microtiter plates coated with supernatants from growth positive wells
were reacted with goat anti-mouse IgG or IgM Ab conjugated to
alkaline phosphatase (Kirkegaard and Perry Laboratories, Rockville,
MD). Wells producing Ig were cloned at limiting dilution in 96-well
plates coated with 50 µl of FCS and then expanded for additional
studies. For Ag-specific assays, plates were coated with the indicated
Ag, including insulin (pork or human), calf thymus DNA, cardiolipin,
dextran, thyroglobulin, tetanus toxoid, human gammaglobulin, or
glucagon (Sigma-Aldrich). Human sera from patients with lupus or
rheumatoid arthritis and immunized mouse sera were used as positive
controls for multi-Ag panels. Hybridomas that stick to noncoated wells
were excluded from study. For insulin binding assays, 96-well
microtiter plates (Immulon II) were coated overnight at 4°C with 100
µl of human insulin (1 µg/ml) in PBS pH 8.0. Plates were washed
extensively in PBS containing Tween (0.2%) and 1% BSA before use.
Radiolabeled tracer studies indicate that each coated well contains
30 ng of insulin. Hybridoma cultures producing Ig were screened for
insulin binding using 2-fold serial dilutions to identify a
nonsaturated binding (70% maximum binding), usually 1:10. For
competitive inhibition with insulin, supernatants were assayed on
dilutions from the linear portion of the binding curve and preincubated
with insulin at 0.05–100 µg/ml for 2 h before transfer to
insulin-coated plates (21). Insulin binding in the
presence or absence of soluble inhibitor was measured using
isotype-specific goat anti-mouse IgM or IgG conjugated to alkaline
phosphatase. Anti-insulin clone 1F11 (mAb301) was used as a positive
control for IgM anti-insulin (22) and mAb125 was used
as a control for IgGs (23). All dilutions and washes were
conducted at room temperature using 0.2% Tween 20 and 1% BSA. Results
are reported as mean OD405 for triplicate
determinations using paranitrophenyl phosphate (Sigma-Aldrich) as
substrate and an automated ELISA reader. To simplify data presentation,
results are presented as the OD of insulin binding in the absence or
presence of soluble insulin (50 µg/ml) using the dilution of
supernatant at 70% maximal binding.
RT-PCR and nucleotide sequences
Total cellular RNA was isolated from 107 hybridoma
cells using the TRI Reagent protocol (Molecular Research Center,
Cincinnati, OH). Ten micrograms of total cellular RNA was used as a
substrate for cDNA synthesis using RNase H-reverse transcriptase (Life
Technologies, Gaithersburg, MD). Oligo(dT) primers were used in first
strand synthesis and one-tenth volume of the first strand material was
then amplified directly using the previously described primers for
Ig, IgMµ, and IgG1,2,3 (3' end) and degenerate
oligonucleotides for the 5' ends (22). Amplification was
conducted using the PerkinElmer thermocycler (Cetus, Emeryville, CA)
using 30 cycles of 1 min at 94°C, 2 min at 45°C, and 1 min at
72°C. PCR products were then separated on 1% agarose gels and
purified with QIEAXII Ger Extraction kit (Qiagen, Chatsworth,
CA). Cloned PCR products were sequenced using an Applied Biosystems DNA
sequencer (Applied Biosystems, Foster City, CA). Using this approach to
analyze duplicate PCR products, we estimate the error rate is 0.5/400
bp and is consistent with that expected from Taq polymerase
(24). Nucleotide sequences were analyzed using
BLASTN searches of European Molecular Biology Laboratory/GenBank
databases (National Center for Biotechnology Information, Bethesda, MD)
using sequences from V,
VH, and third complementarity-determining region
of Ig H chain (CDRH3) regions. For purposes of comparison, amino acid
residue assignments are based on the nomenclature of Kabat et al.
(25). All sequences are available in GenBank database.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
These studies were initiated with the goal of identifying
spontaneous insulin autoantibodies in NOD mice. In our initial
experiments, we observed that the frequency of hybridoma production in
naive NOD mice approached that of immunized mice. Because activated B
cells are preferentially selected for hybridoma formation, we used this
approach to examine the fusion frequency of spleens from NOD, C57BL/6,
and BALB/c mice of different ages. The data are summarized in Fig. 1
and show results as the frequency of
Ab-producing wells (M + G) per 105 of spleen
cells. The frequency of Ab-producing wells in NOD mice ranged from
4–20/105 spleen cells while the frequency in
naive BALB/c and B6 mice ranged from 0.3–1.5/105
spleen cells. Thus, the frequency of B cell fusion events is 5- to
20-fold greater in unmanipulated NOD mice than in naive BALB/c or B6
mice. Increased B cell activity in NOD was observed from 4 wk (the
youngest age tested) to 6 mo of age and was highest at 14 wk. The sex
of NOD mice does not appear to be a major influence on B cell
activation as indicated by the similar Ab capture frequency in male and
female NOD mice. In diabetic NOD mice (blood sugar >300 mg/dl), the
recovery of spleen cells was reduced 50%; however, the relative
frequency of Ab capture was still greater than in BALB/c or B6 mice.
mAb generated in 4- to 12-wk-old mice were chiefly IgM and <1% IgG.
IgG or double producers (IgG and IgM) were observed in 5% of
Ab-producing wells from older NOD mice (14–25 wk). These findings
suggest that in addition to having a high basal level of B cell
activity, the active B cell repertoire of NOD undergoes qualitative
changes in isotype expression with age.
|
To further characterize the active B cell repertoire in NOD mice,
hybridomas captured from fusion at the peak of the response (14 wk)
were screened for insulin binding in ELISA and cloned at limiting
dilution. One of 12 insulin binding mAb was of the IgG2a isotype (EB6)
and one (1B9) was a double secretor (IgM + IgG2a); the others were IgM
Abs. All of these mAb-bound insulin in ELISA compared with noncoated
wells (OD>1.0), and binding was removed by
absorption on either insulin-conjugated Sepharose or by insulin-coated
microtiter plates. However, at multiple dilutions binding was poorly
inhibited by preincubation with an excess of soluble insulin (1–100
µg/ml). This was also true of the IgG mAbs. The results of several
experiments are summarized in Fig. 2 in
which histograms show binding of individual mAb in the presence and
absence of insulin (50 µg/ml). In contrast to mAb derived from NOD
mice, the binding of anti-insulin IgM hybridoma (1F11) from an
immune BALB/c mouse (22) is totally inhibited by insulin
(IC50, 0.1 µM). For some mAb, the binding to insulin in
ELISA is facilitated when soluble insulin is added as an inhibitor
(e.g., mAb 7F5, Fig. 2). These findings are entirely representative of
mAb generated in multiple fusions and indicate that the majority of
active anti-insulin B cells captured by hybridoma production in NOD
mice have low affinity for insulin in solution (IC50, >50
µM).
|
. Insulin binding at serial dilutions of
mAb 1B9 (Fig. 3, 
) were compared with binding of the same dilution
after preincubated with 100 µg/ml of human insulin (Fig. 3, 
). The
resulting binding profile in the presence or absence of excess insulin
shows that insulin inhibition is <10% at all dilutions. In contrast,
the binding of IgG mAb125 derived from insulin immunization is
effectively inhibited at all dilutions by insulin in solution. These
data indicated that although IgG mAb 1B9 has undergone isotype switch,
its relative affinity is sufficiently low that it only binds insulin
when the Ag is stabily displayed on a surface and not in solution. As
indicated in Fig. 2
, most spontaneous insulin-binding mAb from NOD have
this characteristic avidity-dependent binding.
|
The apparent low avidity of anti-insulin hybridomas captured
from NOD spleen suggest that these mAb are part of the natural
autoantibody repertoire. Because natural autoantibodies are
characterized by low avidity and broad specificity, mAb from NOD were
examined for reactivity on a panel of environmental and autoantigens.
The results for four mAb are shown in Fig. 4. The findings indicate that all these
mAb bind one or more Ags in addition to insulin, including
thyroglobulin, Ig, and DNA. In all, 16 mAb identified by insulin
binding in NOD were found to be widely cross-reactive with several Ags
and are represented by the profiles of mAb 4A11 and F10B. The two mAb
with selective (i.e., insulin preferential binding) represent a
minority of the B cells captured by hybridoma formation in NOD. One
mAb, 1B9, binds insulin preferentially when compared with other Ags
but, as shown above, this binding is not inhibited by soluble insulin.
Another IgG mAb, EB6, was identified by insulin binding but is observed
to be more active as a rheumatoid factor (anti-Ig). Together the
data show that most mAb captured from NOD spleen on the basis of
insulin binding have characteristics of natural autoantibodies that
include low avidity and multireactivity.
|
genes with Abs
found in systemic autoimmune disease
To investigate the genetic origins of autoantibodies in NOD, we
determined the nucleotide sequences of V
region genes expressed by six anti-insulin hybridomas. The
V and J gene segments
and third complementarity-determining region of Ig L chain
region used by these hybridomas are summarized in Table I. Different V2
genes are used by EB6 and F10B while the same
V1c gene is used by clonally unrelated 1B9 and
10E10. Neither V2 nor
V8 genes observed in this study were found in
our previous analyses of 20 anti-insulin mAb from BALB/c (22, 26, 27). The V28 gene, also known as
V ser, has not been observed in other
anti-insulin responses; however, this gene is closely related to
the V19 family that was found to dominate the
secondary anti-insulin response
(27). The V1c
gene used by 1B9 and 10E10 is almost identical with that
expressed by anti-insulin mAb 301 derived from immunization of
BALB/c with Beucella abortus Ag-insulin (Fig. 5). This same gene isalso expressed in
germline (unmutated) configuration in two anti-DNA mAb from lupus
prone mice, mAb 202s.38 and 111.68 (28). Further analysis
shows the V genes used by all NOD hybridomas
were often identical with V genes expressed by
mAb (principally anti-DNA) from systemic autoimmune disease in MRL,
NZB, and motheaten mice (Table II).
|
|
|
Analysis of the nucleotide sequences from the
VH of six NOD hybridomas shows that four Abs were
derived from different germline genes in the large J558 family, and one
each was from the 7183 and Q52 families (Table III). Although a
number of J558 genes and a few 7183 and Q52 genes are used by
anti-insulin mAb from primary and secondary anti-insulin
responses in BALB/c (22, 26, 27), none of the previously
described anti-insulin VH genes were closely
related to the family member expressed by these NOD mAb. Five of the
six NOD VH genes were identical with known
VH gene segments (Table III). Three of
these VH, J558.B10, J558.B18, and J558.B20 were
identified in LPS-activated B cells and are also found in transformed B
cell lines such a 70Z/3 and CH14 (29). MAb 4A11 contained
four nucleotide replacements that resulted in two amino acid
substitutions in second complementarity-determining region of Ig H
chanin (S54N and Y61N) when compared with a known chromosomal
segment (RP23-197F8) and to anti-DNA mAb H11A3 from motheaten mice
(30). The J1 segment of 4A11 also
contains a nucleotide replacement (L104V) in a potential hot spot for
mutation (31). In contrast to the other mAb that are
encoded directly by germline (unmutated) genes, mAb 4A11 has likely
undergone Ag-driven selection and somatic mutation.
|
Generation of the third hypervariable region of H chain Ig
reflects juxtapositioning of D and JH gene
segments followed by the joining of a VH gene
segment to the DJH complex. Extensive diversity
within this region is created by the exonuclease removal of nucleotides
as well as by the addition of palindromic and nontemplated addition on
nucleotides by TdT. Adult repertoires are characterized by the presence
of N segment addition in >75% of HC gene rearrangements (32, 33) while there is a bias against N segment addition in neonatal
repertoires. The nucleotide and predicted amino acid from V-D-J gene
segments that contribute to CDRH3 (residues 95–102) of NOD
anti-insulins are shown in Fig. 6.
The length of CDRH3 varied from 4–10 aa. Three mAb use
JH4, two use JH1, and one
uses JH2. Evidence for both 5' and 3' TdT
activity (N segments) is seen only in mAb 1B9. For three mAb, a D gene
donor segment was readily identified, two were DSP2 genes that used the
third reading frame and one was a DQ52 gene that used the first reading
frame. Two mAb, EB6 and EF10B, have no D segments. An interesting
feature of mAb 10E10 and 4A11 is the potential for non-D gene
nucleotides to contribute to CDRH3 diversity. A 12-nucleotide gene
segment from plasmacytoma variant translocation-1 (pvt-1) can account
for the nucleotides in CDRH3 of 4A11 (34) and the pvt-1
segment contains a possible recombination signal sequences flanking the
donor segment. In addition, a donor sequence from framework three of an
X24 anti-asialo GM1 mAb (35) is identical with the
nucleotide sequence in CDRH3 of 10E10. The third framework
region of this mAb contains an exact heptamer recombination
signal sequence, consistent with a VH gene
replacement event as originally described in B cell lymphomas
(36). These observations suggest that gene replacement
events may contribute to CDRH3 structures in NOD.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cell Ags to disease
promoting T cells. Data addressing Ig V region expression in B cells
from NOD mice are limited. Studies using RNA hybridization and
amplification of genomic DNA show a disproportional retention of
fetal-type VH7183 genes in adult NOD mice and
suggest that regulation of the B cell repertoire may be abnormal in NOD
(14, 15).
In this study, we use recognition of the cell Ag insulin as a tool
to investigate the expressed B cell repertoire in NOD mice. Hybridoma
generation from spleens of naive mice indicates that a larger component
of the endogenous B cell pool is active in NOD compared with B6 and
BALB/c mice. This property of increased B cell activity is present
early in life (4 wk) and persists in older mice of both sexes. These
differences are not easily attributable to qualitative differences in
NOD B cells since markers of B cell developed and differentiation such
as CD21, CD22, CD23, and B220 are not abnormal in NOD (our unpublished
observations). However, we find NOD mice express the infrequent
CD72c allele which is also used by some lupus prone strains and the
functional consequences of CD72c in NOD is currently being
examined.
A large majority of mAb captured from NOD spleen are IgM; however, the
contribution of IgG Abs increases during the period when the autoimmune
process of cell attack is expected to be intense (>14 wk). Even
though selected for insulin binding in ELISA, these mAb have broad
specificity and are of low avidity as indicated by their polyreactive
nature and poor inhibition by soluble insulin. These mAb are
functionally different from either monospecific insulin Abs that follow
immunization or from spontaneous insulin Abs found in the prodrome of
NOD and human T1DM. Rather, in NOD most spontaneously active B cells
may not be Ag-driven but are expanded from a repertoire that produces
natural type autoantibodies (37). Although we did not
formally analyze the frequency of insulin binding in the small number
of hybridomas from B6 and BALB/c mice, we do find anti-insulin
activity as part of the natural mAb repertoire of these strains. Thus,
insulin binding is not excluded from the natural repertoire of
nonautoimmune strains and this is consistent with the detection of
natural insulin autoantibodies in many studies of human and animal sera
(38, 39). Based on our extensive search, we postulate that
the production of high-affinity insulin autoantibody is a rare event in
NOD spleen. Although high-affinity anti-insulin is detected in
sera, B cells producing such autoantibodies may be short lived or
reside in discrete anatomical locations and we are currently examining
those possibilities. In prior studies on IgG anti-insulin mAb125
and 127, we introduced back mutations to germline structures and show
that a single amino acid substitution can impart insulin binding to V
regions with little or no insulin binding (23). Thus, a
large pool of natural type insulin autoantibodies may provide a
substrate to fuel the autoimmune process if appropriate T cell help is
available. When examined in RIA, we find our panel of NOD mAb bind
insulin poorly. However, the binding of some mAb, such as 4A11, is not
negative. For example, 4A11 typically binds 200 cpm above background
compared with 5000 cpm for mAb125. Because structural studies suggest
that 4A11 has undergone somatic mutation, this mAb may have been
captured during its evolution in the context of T1DM.
The proposition that natural Abs provide a starting repertoire for an
insulin-specific response is supported by data on the genetic
components of spontaneous NOD mAb that bind insulin. The V genes from
NOD mAb are shared with monospecific insulin Abs from immunized mice
and with autoantibodies detected in systemic autoimmune disease.
V genes with identical sequence are observed
in other autoantibodies, principally anti-DNA, and the same
V1c gene is used by NOD mAb and by mAb from
insulin immunization. Similarly, the V Ox2
gene reported for a NOD IgG anti-insulin mAb (16) is
the same V gene used by mAb125 from BALB/c
(26). Analysis of the V gene structures used by NOD mAb
indicate they are derived directly from germline-encoded (i.e.,
unmutated) structures, and only one mAb shows evidence for Ag-driven
selection. The VH genes used by insulin binding
NOD mAb are not closely related to any of the previously reported
VH genes expressed by anti-insulin mAb
(22, 26, 27). These VH are often
identical with genes used by anti-DNA mAb from lupus prone mouse
strains and by B cell lymphomas. An important structural feature of NOD
VH regions is that most mAb maintain CDRH3-length
D segments without the use of N segment additions. Lack of N segments
is associated with VH genes that originate in the
fetal period and is consistent with reports on persistent fetal-type
VH genes in NOD mice (15). Because
the B1 subset has many characteristics of the fetal repertoire
(32), we also examined our hybridomas for CD5 expression
by both RT-PCR and FACS and do not find evidence that these mAb are
derived from B1 B cells (our unpublished observations). In
addition, the B1a population (CD5+) in the
spleens of NOD mice is not significantly different from B6 mice (P.
Kendall and J. W. Thomas, manuscript in preparation). Thus,
increases in autoreactive B cells observed in NOD mice are not easily
attributable to expansion of the B1 subset as occurs, for example, in
motheaten mice (40).
Virtually all of the expressed VH and
V genes used by insulin-binding hybridomas
from in NOD are also found in anti-DNA or other autoantibodies from
mouse strains with systemic autoimmune disease such as MRL and NZB/W.
Although databases are biased by the intense investigation of
autoantibodies, the nearly exact nucleotide sequence homologies in many
cases indicates that the same V genes are selected into the repertoire
of NOD mice as in mice with different autoimmune processes.
Interestingly, some treatments that protect NOD mice from developing
diabetes are observed to induce lupus-like syndromes (19)
and support a hypothesis that NOD and lupus prone strains may share
genetic features with immunoregulatory abnormalities in the B cell
repertoire. Among our mAb, 4A11 shows consistent evidence for Ag-driven
somatic hypermutation. Because mAb 4A11 is the most insulin selective
of our panel, the presence of mutations suggests it may have been
evolving into a high-affinity Ab. One other mutated anti-insulin
(Ig2b) from NOD has been reported and it also expresses a
VH gene used by MRL mice (16). Thus,
some components of the highly active natural autoantibody repertoire in
NOD may be selected into T cell-dependent immune responses and fuel the
autoimmune process. The recognition that high-affinity insulin Abs
detected by RIA and not low avidity binding in ELISAs are more
predictive of disease progression suggests that availability of
Ag-specific T cell help is a critical feature of disease
progression.
The abundant repertoire of autoantibodies in NOD mice raises the possibility that selection and regulation of the B cell repertoire is abnormal in NOD. Recent studies in mice that are deficient in key signaling components in the B cell receptor (BCR) pathway, such as lyn and CD45, show that signaling defects favor the selection of BCR that are autoreactive as a means to maintain B cell homeostasis (41, 42, 43). Because abnormalities in signal transduction are recognized in NOD T cells (44), the increased entry of autoreactive B cells into the repertoire of NOD may reflect a similar abnormality in BCR signaling in NOD. Our recent observation that mice harboring an Ig H chain transgene favor the development of insulin-binding B cells in NOD but not in B6 mice also suggests an altered threshold for selection of autoreactive B cells in NOD (45). The previously noted persistence of fetal VH genes in NOD may reflect an increased frequency of autoreactive V regions in the fetal repertoire that aids selection of B cells in the repertoire. A recent case report of a human T1DM patient with Bruton tyrosine kinase deficiency has been used to suggest that the role of B cells in NOD and human T1DM may be different (46). An alternative interpretation is that a signaling defect in Bruton tyrosine kinase deficiency may have favored selection of small numbers of autoreactive B cells in the human disease and our studies suggest that a qualitatively similar defect may be present in the B cell repertoire of NOD. Understanding the mechanisms responsible for these defects may aid in the identification of additional genetic components that contribute to susceptibility in T1DM.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. James W. Thomas, Department of Medicine, Vanderbilt University, T-3219 Medical Center North, Nashville, TN 37232-2681. E-mail address: james.thomas{at}vanderbilt.edu
3 Abbreviations used in this paper: NOD, nonobese diabetic; BCR, B cell receptor; T1DM, type I or insulin-dependent diabetes mellitus; CDRH3, third complementarity-determining region of Ig H chain; V, variable region L chain; VH, variable region of Ig H chain; pvt-1, plasmacytoma variant translocation-1.
Received for publication May 28, 2002. Accepted for publication October 2, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
cells of nonobese diabetic mice. J. Immunol. 163:743.This article has been cited by other articles:
|
|
 |
|
  R. Ekici, M. Sundstrom, B. Thay, and K. Lejon Enhanced capture of extramembranous IgM and IgG on B cells in the NOD mouse--implications for immune complex trapping Int. Immunol., May 1, 2009; 21(5): 533 - 541. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Bour-Jordan, B. L. Salomon, H. L. Thompson, R. Santos, A. K. Abbas, and J. A. Bluestone Constitutive Expression of B7-1 on B Cells Uncovers Autoimmunity toward the B Cell Compartment in the Nonobese Diabetic Mouse J. Immunol., July 15, 2007; 179(2): 1004 - 1012. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. J. Woodward and J. W. Thomas Multiple Germline {kappa} Light Chains Generate Anti-Insulin B Cells in Nonobese Diabetic Mice J. Immunol., July 15, 2005; 175(2): 1073 - 1079. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Sblattero, F. Maurano, G. Mazzarella, M. Rossi, S. Auricchio, F. Florian, F. Ziberna, A. Tommasini, T. Not, A. Ventura, et al. Characterization of the Anti-Tissue Transglutaminase Antibody Response in Nonobese Diabetic Mice J. Immunol., May 1, 2005; 174(9): 5830 - 5836. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Carrillo, M. C. Puertas, A. Alba, R. M. Ampudia, X. Pastor, R. Planas, N. Riutort, N. Alonso, R. Pujol-Borrell, P. Santamaria, et al. Islet-infiltrating B-Cells in Nonobese Diabetic Mice Predominantly Target Nervous System Elements Diabetes, January 1, 2005; 54(1): 69 - 77. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. S. Wong, L. Wen, M. Tang, M. Ramanathan, I. Visintin, J. Daugherty, L. G. Hannum, C. A. Janeway Jr, and M. J. Shlomchik Investigation of the Role of B-Cells in Type 1 Diabetes in the NOD Mouse Diabetes, October 1, 2004; 53(10): 2581 - 2587. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) or presence (
) of soluble insulin, 50 µg/ml. 1B9 and
EB6 are IgG and other Abs are IgM. 1F11 is an IgM anti-insulin mAb
from BALB/c that is >95% inhibitable with soluble insulin. Data are
mean ± SD of binding OD405 on insulin-coated
plates.
