Vh family utilization by igg anti-dnasecreting lymphocytes derived from autoimmune mrl-lprlpr mice.

56 1
VH FAMILY UTILIZATION BY
IgG ANTI-DNA-SECRETING LYMPHOCYTES
DERIVED FROM AUTOIMMUNE MRL-ZprlEpr MICE
DENNIS M. KLINMAN, DALE K. DELLACQUA, JACQUELINE CONOVER, and KONRAD HUPPI
Objective. To evaluate the heterogeneity of the
IgG anti-DNA autoantibody response of MRLlZpr mice.
Methods. B cell clones were grown on nitrocellulose membranes. Those producing IgG anti-DNA antibodies were identified by a modified enzyme-linked
immunospot assay, and their utilization of IgVH genes
was determined by hybridization.
Results. Four hundred sixty-eight IgG anti-DNAsecreting colonies were derived from 9 autoimmune
MRLlZpr mice. Individual VH families contributed to
anti-DNA production at a frequency roughly proportional to their representation in the expressed repertoire
(except for the more frequent use of VH 7183 and less
frequent use of VH 36-60). In individual mice, anti-DNA
antibodies were encoded by 2-5 different IgVH families,
with no single family constituting more than 2 7 4 5 % of
any animal's anti-DNA response.
Conclusion. The IgG anti-DNA response of individual MRLlZpr mice is oligoclonal.
From the Laboratory of Retrovirus Research, Center for
Biologics Evaluation and Research, Food and Drug Administration,
and the Laboratory of Genetics, National Cancer Institute, National
Institutes of Health, Bethesda, Maryland.
Dennis M. Klinman, MD, PhD: Laboratory of Retrovirus
Research, Center for Biologics Evaluation and Research, Food and
Drug Administration; Dale K. Dellacqua, BS: Laboratory of Retrovirus Research, Center for Biologics Evaluation and Research,
Food and Drug Administration; Jacqueline Conover, BS: Laboratory of Retrovirus Research, Center for Biologics Evaluation and
Research, Food and Drug Administration; Konrad Huppi, PhD:
Laboratory of Genetics, National Cancer Institute, National Institutes of Health.
Address reprint requests to Dennis M. Klinman, MD, PhD,
Laboratory of Retrovirus Research, Center for Biologics Evaluation
and Research, Food and Drug Administration, Building 29A, Room
3 D 10, Bethesda, MD 20892.
Submitted for publication September 22, 1992; accepted in
revised form November 10, 1992.
Arthritis and Rheumatism, Vol. 36, No. 4 (April 1993)
Systemic lupus erythematosus (SLE) is an autoimmune disease characterized by the overproduction
of pathogenic autoantibodies. Mice of the MRLllpr
strain provide a useful model for studying the pathogenesis of this disease. These mice develop an
autoimmune-mediated glomerulonephritis and arthritis
similar to that seen in humans with lupus (1,2). They
also express a number of immune abnormalities, notably the accumulation of atypical T lymphocytes and
the hyperactivation of B lymphocytes (3-9).
Increased production of IgG anti-DNA antibodies, a hallmark of human SLE, also occurs in MRLllpr
mice (10,ll). Autoantibodies of this specificity form
immune complexes that contribute to the development
of life-threatening glomerulonephritis (12,13). The
concentration of IgG anti-DNA antibodies at sites of
tissue injury correlates with disease activity (14-17).
Although B cell secretion of anti-DNA is critical to
disease pathogenesis, the clonal diversity of the B cells
producing these autoantibodies remains uncertain
(3,18; for review, see ref. 19). In an attempt to resolve
this issue, investigators have studied and sequenced
the immunoglobulin heavy chain variable regions
(IgVH) utilized by anti-DNA-secreting B cells (20-26).
Early work by Shlomchik et a1 evaluated the
anti-DNA response in 5 MRLllpr mice, and showed
that individual animals used no more than 2 different
VH families to encode IgG anti-DNA antibodies
(21,27). Based on these data, it was concluded that the
anti-DNA response of individual mice was monoclonal
or pauci-clonal. By comparison, Tillman et a1 studied 9
New Zealand black x New Zealand white (NZB X
NZW) mice and found that individual animals utilized
up to 10 different clones and 4 different VH families to
encode IgG anti-DNA antibodies, leading those inves-
562
KLINMAN ET AL
tigators t o conclude that the anti-DNA response was
oligoclonal (25).
Virtually all of the investigations in this field
have relied upon hybridomas as their source of
autoantibody-secreting cells (21-26,28). Yet recent
evidence suggests that a large fraction of lymphocytes
secreting IgG autoantibodies in vivo may not produce
stable hybrids, and that hybridoma panels may therefore be selectively skewed toward the expression of a
nonrepresentative subpopulation of B cells (29,30).
Thus, an alternative cloning strategy was devised t o
study a more diverse and potentially representative set
of anti-DNA-secreting B cells.
In the present study, colonies of MRLllpr
mouse B cells were grown in short-term culture
(31,32), and their IgVH family utilization was examined. This approach yielded large numbers of antiDNA-secreting colonies per animal and was free of
the selection pressures imposed b y hybridization and
long-term culture. Previous reports have shown that
the IgG antibody-secreting clones identified by this
strategy were representative of B cells that had been
secreting Ig under physiologic conditions in vivo (33).
Our results indicate that 1) within the MRLllpr strain
as a whole, all VH families examined were able t o
encode anti-DNA antibodies, 2) within individual
mice, 2-5 different V H families were used t o produce
IgG anti-DNA antibodies, and 3) individual mice relied
on a single VH family to encode only 27-65%, a t
maximum, of their total anti-DNA response.
MATERIALS AND METHODS
Animals. MRL-lpr/lpr (MRWlpr) mice were obtained
from Jackson Laboratories (Bar Harbor, ME). All mice used
in these experiments were 6-month-old females and had active
glomerulonephritis (documented by >2+ proteinuria).
Colony culture system. The method for growing B cell
colonies from MRLllpr mice was adapted from that previously described by Kelsoe (31,32). Single cell suspensions
were made from the spleens of freshly killed animals in
medium consisting of RPMI 1640 supplemented with 10%
fetal calf serum (FCS), 5 x lOP5M2-mercaptoethanol, 2 mM
L-glutamine, 10 mM HEPES buffer, 0.11 mg/ml sodium
pyruvate, and penicillin (50 unitdml) and streptomycin 0.11
pghl(34). Cells were washed twice and centrifuged over a
discontinuous density gradient. Lymphocytes were isolated
from the 50--60% Percoll fraction and cultured on sterile
Immobilon-P discs (Millipore, Bedford, MA) at a concentration of 5 x lo4 cells/ml in complete medium supplemented
with 15 &ml lipopolysaccharide (LPS) (33). Preliminary
experiments showed that colony formation was maximal
following 6 days of culture at 37°C in a 5% C02-95% air
incubator.
Enzyme-linked immunospot assay. Flat-bottom Immulon I microtiter plates (Dynatech, Alexandria, VA) were
coated with goat anti-mouse Ig (Cappel, Cochranville, PA)
or salmon sperm DNA (Sigma, St. Louis, MO) (after pretreatment with mouse bovine gamma globulin) and then
blocked with 1% bovine serum albumin (BSA) in phosphate
buffered saline (PBS) as previously described (35). Serial
dilutions of single cell suspensions, starting with lo6 cells/
well, were incubated on antigen-coated plates for 7 hours at
37°C in a 5% C02-95% air incubator (34). The cells were
washed away with PBS/O.OS% Tween 20 and the plates
overlaid for 2 hours with phosphatase-conjugated antibodies
to mouse IgM or IgG (Kirkegaard and Perry, Gaithersburg,
MD). The antibodies produced by individual B cells that
bound to the plate were visualized by addition of a 5-bromo3-chloroindolyl phosphate solution (Sigma) in a low-melt
agarose kept at 50°C; phosphatase acts on this substrate to
produce a blue spot that cannot diffuse through the agarose
once it solidifies at room temperature (36). The dilution of
cells producing 20-40 spots per well was used to determine
the total number of antibody-specific B cells per sample. The
sensitivity and specificity of this assay has been documented
in antigen-inhibition tests and in studies involving antigenspecific hybridoma cell lines (34,36,37).
Detection of Ig-secreting colonies. Immobilon P membranes were coated with salmon sperm DNA (Sigma) or goat
anti-mouse Ig (Kirkegaard and Perry), at a concentration of
10 pg/ml in borate buffer, pH 8.4, and then blocked with 2%
BSA plus 1% gelatin in PBS (34,38). Culture discs containing
actively growing B cell colonies were washed twice with
complete medium and then overlaid with an antigen-coated
membrane. Following a 6-hour incubation the membrane
was removed, washed extensively with PBS-0.5% Tween
20, and treated with phosphatase-conjugated antibodies to
mouse IgM or IgG (Southern Biotechnology, Birmingham,
AL). Antibodies secreted by the B cell colonies that bound
to the antigen-coated membrane were visualized by addition
of BCIP (Sigma).
Colony hybridization assay. Discs containing B cell
colonies were washed twice in PBS and then treated with a
fresh solution of 1% buffered neutral formalin (Sigma) in
PBS for 7 minutes at room temperature. Discs were washed
3 times in PBS-Tween 20, immediately prehybridized, and
then hybridized overnight to a nick-translated probe of
specific activity 1.42.0 X 10' counts per minute/pg in 50%
formamide at 42°C (39). After washing in 0 . 1 % ~SSC ( I x
SSC = O.15M NaCl and O.015M sodium citrate) for 15-60
minutes at 65°C the discs were exposed to Kodak AR-2 film
(Eastman-Kodak, Rochester, NY) at -70°C in the presence
of an intensifying screen.
Probes. DNA probes used in this study have been
reported previously (39) and correspond to VH gene families
as follows: X24 is a 500-basepair Eco RI insert, 7183 is a
900-bp Sac I insert containing the entire VH coding region,
J558 is a 700-bp Bum HIIBst EII fragment, 5606 is a 350-bp
Hind IIISac I fragment, S107 is a 250-bp Hinf IIBst EII
fragment, 36-60 is a 2,400-bp Hind I11 fragment, and 3609 is
a 2000 bp Eco RYBarn HI fragment. All VH fragments were
subcloned into pGEM3Z. The derivation of these probes and
their specificity under the hybridization conditions used in
this investigation have been previously described (39).
563
VH FAMILY USAGE BY ANTI-DNA-PRODUCING CELLS
Analysis of VH families used by IgG anti-DNAsecreting colonies. Discs containing B cell colonies were first
analyzed for Ig production by the colorimetric assay described above (to detect IgG anti-DNA-secreting cells) and
then fixed with formalin and hybridized to VH-specific
probes. Results from these 2 analyses were integrated by
juxtaposing antigen-coatedmembranes (which identified IgG
anti-DNA-secreting colonies) over autoradiographs (identifying colonies utilizing specific V H families) through use of
alignment markings placed on all membranes and discs
(Figure 1).
Table 1. Repertoire comparison of in vivo-activated cells versus
in vitro-stimulated colonies*
IgM
% IgM- or IgG-secreting
cells/colonies
In vivo-activated B cellst
LPS-induced B cell
colonies$
% DNA-specific Ig-secreting
cellslcolonies5
In vivo-activated B cellst
B cell colonies$
34.6
52.7
~
65.4 t 11.3
47.3 t 4.5
1.9 t 0.7
2.8 t 0.9
-
* Values are
Isotype of Ig secreted by B cell colonies derived
from MRLllpr mice. Preliminary experiments were
conducted to establish conditions that would maximize the formation of Ig-secreting colonies from MRL/
lpr mice. Minor variations on the method developed
by Kelsoe (31,32) proved optimal. For experiments
reported herein, splenic lymphocytes were plated at a
concentration of 5 X lo4 cellslml in medium supplemented with 10% FCS and 15 pg/ml LPS. This resulted in the formation of discrete Ig-secreting colo-
1) GROW COLONIES IN VlTRO
CULTURE DISC
2) OVERLAY WITH DNA-COATED
MEMBRANE TO IDENTIFY
COLONIES PRODUCING
IgG ANTI-DNA ANTIBODIES
8.3
4.5
1.8 2 0.6
1.6 2 0.5
~~~
RESULTS
5
?
IgG
3) FIX AND HYBRIDIZEWITH
lgVH GENE SPECIFIC PROBE
HYBRIDIZED
CULTURE DISC
DNA-COATED MEMBRANE
I
4) OVERLAY HYBRIDIZEDCULTURE DISC
AND DNA-COATED MEMBRANE TO REVEAL
lgVH GENES USED BY DNA SPECIFIC CLONES
OVERLAY RESULTS
Figure 1. Technique to identify IgVH genes used by IgG antiDNA-secreting colonies. 0 = B cell colony; * = alignment spot; 0
= IgVH positive; 0 = DNA positive; @ = DNA/IgVH positive.
the mean 2 SD percent, from assays performed on
cells from at least 6 mice per group. LPS = lipopolysaccharide.
t Freshly isolated splenic B cells from unmanipulated mice were
analyzed using the enzyme-linked immunospot assay (34).
$ Colonies were grown on Immobilon P membranes for 6 days, as
described in Materials and Methods.
§ % DNA-specific cells was calculated by the formula (no. of
anti-DNA-secreting cells/total no. of Ig-secreting cells) x 100%.
nies by -2.5% of the original B cell inoculum after 6
days of culture. This is similar to the frequency of B
cells actively secreting Ig in vivo in MRLllpr mice of
this age (range 1.4-2.9% [331).
The isotype of Ig produced by these colonies
was determined by overlaying the culture discs with
Immobilon P membranes that had been precoated with
goat anti-mouse Ig. Antibodies secreted by B cell
colonies bound to these anti-Ig-coated membranes
and were detected in an isotype-specific colorimetric
assay (34,36,37). This strategy identified both the
number and the location of Ig-secreting colonies. As
seen in Table 1, slightly fewer than one-half of M R L /
fpr colonies produced IgG. This finding is consistent
with previous reports showing that B cells from adult
MRLllpr mice frequently produce IgG in vivo (40) and
continue to produce that isotype at a modestly lower
frequency when stimulated with LPS in vitro (Table 1
and ref. 33). Additional experiments revealed that the
proportion of colonies secreting IgG antibodies reactive with DNA was similar to the fraction of B cells
producing those antibodies in vivo (Table 1).
IgVH families used by B cell colonies derived
from MRLlZpr mice. Kelsoe previously showed that
treating B cell colonies with formalin would fix the
colonies to the culture disc and permeabilize their
membranes such that cellular messenger RNA could
be detected by conventional gene hybridization techniques (31,32). In the present study, B cell colonies
grown for 6 days were treated with formalin and the
culture discs then hybridized with VH farnily-specific
probes. To ensure a high degree of specificity with
KLINMAN ET AL
564
these probes, high-stringency wash conditions were
rigorously maintained (39). Results from these studies
are shown in Table 2. Of the 1,345 colonies examined,
a majority utilized 5558, followed, in order, by 7183,
36-60, 5606, S107, 3609, and X24.
Analysis of IgG anti-DNA-secreting colonies
from MRLlZpr mice. To identify B cell colonies secreting IgG anti-DNA antibodies, culture discs were overlaid with DNA-coated membranes. From 0.3% to 4.8%
(mean 1.9%) of the IgG-secreting colonies from
6-month-old MRLllpr mice secreted antibodies that
bound to these DNA-coated membranes. This is similar to the previously reported fraction of in vivoactivated B cells producing IgG anti-DNA antibodies
in animals of this strain (0.6-7.7%, mean 2.8%) (33).
The discs were then treated with formalin and probed
to determine the number and location of colonies
utilizing specific IgVH families. As diagrammed in
Figure 1, data from these 2 assays could be integrated
by superimposing the DNA-coated membranes (identifying anti-DNA-secreting colonies) with autoradiographs (identifying colonies utilizing specific IgVH
families). This procedure, accomplished with the aid
of preestablished alignment spots, allowed us to determine which VH gene families were used to encode
anti-DNA antibodies.
Four hundred sixty-eight IgG anti-DNAsecreting colonies from 9 MRLllpr mice were studied.
All available probes were used in the analysis of 6 of
these animals; for technical reasons, the other 3 mice
were studied using only 3-5 probes. VH family utilization was analyzed independently in each animal.
Results were then combined to assess the heterogeneity of the anti-DNA response in this strain as a whole.
Table 2.
mice*
IgVH gene utilization by B cell colonies from MRLllpr
% IgVH-positive colonies
Probe
(n = 1,345)
Total
IgG anti-DNA
(n = 468)
7183
J558
3606
36-60
X24
S107
3609
15.7 t 8.4
51.5 -C_ 11.8
9.3 2 5.9
11.4 2 6.8
2.3 2 1.2
6.1 2 4.4
3.7 2 3.8
29.2 2 9.9
47.2 2 19.1
8.6 2 8.1
5.2 2 6.3
2.8 2 5.3
5.0 t 2.8
2.0 2 1.1
* B cell colonies were hybridized with IgVH-specific probes. The
number of colonies using each VH family was divided by the total
number of hybridizing colonies on that filter to calculate percent VH
utilization. Values are the mean 2 SD of the data from all animals
studied with a particular probe (n = 7-9 animals for each probe).
Table 3. Utilization of IgVH families by IgG anti-DNA-secreting
colonies in 6 MRLllpr mice*
% IgVH-positive colonies
Mouse
Probe
7183
Probe
J558
Probe
5606
Probe
36-60
Probe
X24
Probe
S107
Probe
3609
1
2
3
4
5
6
26.7
30.0
35.7
11.1
11.1
33.3
20.0
22.7
21.2
52.9
65.4
57.7
20.8
0
15.0
0
14.3
4.1
14.3
0
3.8
0
0
0
0
14.3
0
0
0
0
0
0
7.9
0
0
0
0
0
0
0
11.1
5.9
* Values are the percent of I& anti-DNA-secreting colonies from 6
MRLllpr mice hybridizing with this set of VH family-specific
probes. These animals represent the subset of mice presented in
Table 2 for which data using every probe were available. It should
be noted that up to 36% of the colonies from individual mice did not
hybridize with any of the probes examined; these presumably
represented colonies that utilized an IgVH region for which no
probe was available.
As seen in Table 2, 1) all VH gene families examined
were able to encode anti-DNA antibodies, and 2) most
VH families contributed to the production of IgG
anti-DNA autoantibodies at a frequency roughly proportional to their representation in the expressed repertoire, although 3) there was a disproportionate increase in the use of VH 7183 (P < 0.001, by chi-square
test) and decrease in the use of VH 36-60 (P < 0.001)
by colonies secreting anti-DNA antibodies in MRL/lpr
mice.
We then examined the frequency with which
VH gene families were used to produce IgG anti-DNA
antibodies in individual MRLllpr mice. Only those
animals examined with all 7 probes were included in
this analysis. As seen in Table 3, considerable animalto-animal variability was detected. Whereas all mice
used VH 7183 and 5558 to encode anti-DNA antibodies, the frequency with which these families were
utilized in different animals varied from 11% to 36%
for VH 7183 and from 20% to 65% for VH 5558. In
addition, 4 of the 6 animals examined utilized 5606, 2
each utilized S107 and 36-60, and only 1 utilized X24 or
3609 to produce IgG anti-DNA antibodies. Individual
mice used 2-5 different VH families to encode antiDNA antibodies. This represents a minimum estimate
of the diversity of VH family usage, since not all
families were examined.
DISCUSSION
This study examined the frequency with which
specific VH families contribute to the production of
VH FAMILY USAGE BY ANTI-DNA-PRODUCING CELLS
IgG anti-DNA antibodies in autoimmune mice. Initial
studies confirmed that the size, number, and complexity of the VH repertoire expressed by MRLllpr mice
was similar to that reported for normal animals (Table
2 and refs. 39, 41, and 42). The analysis of 468
anti-DNA-secreting colonies revealed that each of the
7 VH families examined in this investigation could
encode IgG anti-DNA. As a general rule, individual
families were utilized at a frequency roughly proportional to their representation in the MRLlIpr repertoire
as a whole (Table 2). There were 2 exceptions to this
generalization: The 7183 family was overutilized (P<
O.OOl), while the 36-60 family was underutilized (P <
O.OOl), for anti-DNA production.
Although all MRLlIpr mice used the 5558 and
7183 families to produce IgG anti-DNA, the frequency
with which these families were used varied considerably among individual animals. VH 5558 contributed
20-65%, and VH 7183 contributed 11-36%, of the total
DNA-specific response of different mice (Table 3).
Other VH families were used with even less consistency: Only 4 of 6 animals used J606,2 of 6 used S107
or 36-60, and only 1 of 6 used X24 or 3609 to encode
IgG anti-DNA.
The high concentration of serum autoantibodies
in lupus mice has led some investigators to postulate
that Ig gene abnormalities play a role in disease
pathogenesis (39,43). Classic mendelian analysis of Ig
heavy chain allotypes and Southern blot analyses
using VH-specific probes have revealed no linkage
between SLE and polymorphisms at the H chain
complex (4446). However a number of studies have
shown that D-proximal VH gene families (7183 and
Q52) were utilized preferentially by neonatal B cells
(41,47-51), and there is some evidence that these
families are overutilized by autoimmune mice (24,52).
The latter finding is controversial, since Kastner et a1
found no evidence that 3’ VH families were systematically overexpressed in mice with autoimmune disease
(39). It was concluded that VH gene utilization by
MRWlpr mice was stochastic-a finding consistent
with that of Teale and Morris, who reported that VK
light chain gene usage was random in this strain (42).
There is considerable controversy concerning
the mechanism(s) responsible for the initiation and
perpetuation of autoantibody production in SLE.
Functional studies of B cell isotype, idiotype, and
cross-reactivity indicate that many different clones
may be stimulated to secrete anti-DNA antibodies in a
single animal (13,34,53,54), but this result has been
challenged by molecular evidence suggesting that
565
antigen-reactive B cells from individual mice may be
clonally related (21,27,55). More recently, it was
shown that disease in Ipr mice results from defective
expression of the f u s gene product (56)-an abnormality that presumably affects B cells of all specificities.
These observations have raised questions concerning
the process(es) responsible for selecting and diversifying the IgG anti-DNA repertoire.
To analyze the diversity of the VH families
contributing to the production of IgG anti-DNA antibodies, previous investigators generated panels of
hybridomas. Recent evidence suggests that lymphocytes secreting IgG autoantibodies in vivo do not
readily form stable hybrids (30), which may explain
why relatively few IgG anti-DNA-secreting hybridomas have been cloned from any single mouse (22,2427,52). Indeed, those hybridomas that have been isolated may represent an atypical or pauci-clonal subset
of a more diverse population of in vivo-activated B
cells (29). In this context, it has been shown that IgG
autoantibody-secreting hybridomas do not accurately
reflect the autoantibody specificities present in the
serum of the mice from which they were derived (30).
To avoid these difficulties and generate a larger
and potentially more representative repertoire of
autoantibody-secreting B cells, we adopted the
method of Kelsoe (31,32). B cell colonies were produced by short-term culture of MRLllpr B cells in the
presence of LPS. Using this approach, an average of
61 IgG anti-DNA-secreting clones were derived per 5
X lo6 splenic lymphocytes per animal. This cloning
efficiency is 1-2 orders of magnitude higher than that
reported for the production of IgG anti-DNAsecreting hybridomas following cell fusion (20-26).
The culture system reported by Kelsoe yielded
Ig-secreting clones whose antigenic specificity was
similar to that of in vivo-activated B cells (Table 1 and
data not shown). Moreover, previous reports showed
that cells induced to secrete IgG following LPS stimulation were derived from in vivo-activated precursors (33). Other investigators have shown that VH
family usage by B cells cultured with LPS is similar to
that of normal unstimulated lymphocytes (42). Thus,
while some degree of repertoire selection might take
place during short-term culture (a disproportionate
fraction of DNA-specific clones produced IgM in
vitro), evidence suggests that the colonies secreting
IgG anti-DNA antibodies were derived from B cells
that had been actively producing Ig of that specificity
and isotype in vivo (33).
We found that the VH repertoire expressed by
KLINMAN ET AL
566
MRLllpr mice was of normal size and complexity, and
that all VH families could contribute to the production
of IgG anti-DNA antibodies. Our analysis of 40-170
IgG anti-DNA-secreting colonies per mouse revealed
that 2-5 different VH families encoded IgG anti-DNA
antibodies in each animal (mean -+ SD 3.7 -+ 1.0) and
that no animal derived more than 2745% of its antiDNA response from a single VH family (Table 3).
Since up to 30% of the anti-DNA-secreting colonies
examined did not hybridize with any of the 7 VH
probes used, this represents a minimum estimate of
the number of VH families involved in the anti-DNA
response.
These data can be compared with those derived
from studies of hybridomas. Shlomchik et a1 analyzed
4-6 IgG anti-DNA-secreting hybridomas from each of
5 MRLllpr mice (21,27). They found that 1) only 1-2
(mean ? SD 1.6 +- 0.5) VH families were involved in
the production of IgG anti-DNA antibodies in a single
animal, and 2) hybridomas from individual mice using
the same VH family were always clonally related. By
comparison, Tillman et al analyzed a mean (+SD) of
7.6 ? 6.7 IgG anti-DNA-secreting hybridomas from 9
NZB X NZW mice (25). They found that 1-4 (mean 2.1
1.1) VH families were involved in the production of
these antibodies, and that multiple members of a single
family were used to produce anti-DNA antibodies (25).
The latter finding has been confirmed by other investigators (26).
Our analysis of a much larger number of antiDNA-secreting colonies per animal supports the conclusion of Tillman et a1 that the anti-DNA response is
oligoclonal. In contrast, our data differ from those of
Shlomchik et al, and do not support the conclusion of
those investigators that this response is pauci-clonal or
monoclonal. Clonal diversity among IgG anti-DNAsecreting cells is consistent with the functional heterogeneity of anti-DNA antibodies (13,34,53,54) and the
complexity of the DNA molecule itself, which, unlike
a simple hapten, is composed of multiple immunogenic
epitopes (57).
Other investigators have also evaluated VH
family utilization in mice with lupus. PanosianSahakian et a1 (20) generated 18 anti-DNA hybridomas
from a single NZB x NZW mouse and found that they
were derived from a minimum of 12 different VH genes
from 6 different VH families. Bona studied a pool of 23
DNA-reactive hybridomas from a group of unrelated
mice and found that they were encoded by the J558,
7183, Q52, and S107 families (52). Smith and Voss (22)
reported that 5 monoclonal anti-DNA antibodies from
*
NZB x NZW mice utilized 3 different VH families,
and Trepicchio and Barrett determined that 11 monoclonal anti-DNA antibodies from MRLllpr mice were
derived from at least 9 different VH genes selected
from 4 VH families (24). These findings are consistent
with our observation that the IgG anti-DNA response
in individual mice can involve multiple different clones
and VH families.
Unfortunately, we were unable to generate sequence data from the small number of cells present in
LPS-stimulated anti-DNA-secreting colonies. However, other investigators have shown that multiple
members of the same VH family are used to encode
IgG anti-DNA antibodies in a single mouse (25,26).
Our detection of VH gene diversity using a short-term
in vitro cloning system extends the findings of Tillman
et a1 (25), and confirms that the IgG anti-DNA response of individual MRL/lpr mice is oligoclonal.
Ongoing studies should help delineate the processes
that regulate the selection and activation of the
autoantibody-secreting lymphocytes.
REFERENCES
1 . Kelley V, Roths JB: Interaction of mutant Ipr gene with
background strain influences renal disease. J Immunol
37:220-229, 1985
2. Hang L, Theofilopoulos AN, Dixon FJ: A spontaneous
rheumatoid arthritis-likedisease in MRL/lpr mice. J Exp
Med 155:169&1701, 1982
3. Theofilopoulos AN, Dixon FJ: Etiopathogenesis of murine systemic lupus erythematosus. Immunol Rev 55:
179-216, 1981
4. Theofilopoulos AN, Shawler DC, Eisenberg RA, Dixon
FJ: Splenic Ig-secreting cells and their regulation in
autoimmune mice. J Exp Med 151:44&466, 1980
5 . Andrews BS, Eisenberg RA, Theofilopoulos AN, Izui S,
Wilson CB, McConahey PJ, Murphy ED, Roths JB,
Dixon FJ: Spontaneous murine lupus-like syndromes:
clinical and immunopathological manifestations in several strains. J Exp Med 148:119&1215, 1978
6. Sobel ES, Katagiri T, Katagiri K, Morris SC, Cohen PL,
Eisenberg RA: An intrinsic B cell defect is required for
the production of autoantibodies in the lpr model of
murine systemic autoimmunity. J Exp Med 173:14411449, 1991
7. Prud’homme GJ, Park CL, Fieser TM, Kofler R, Dixon
FJ, Theofilopoulos AN: Identification of a B cell differentiation factor(s) spontaneously produced by proliferating T cells in murine lupus strains of the Iprhpr
genotype. J Exp Med 157:730-742, 1983
8. Prud’homme GJ, Balderas RS, Dixon FJ, Theofilopou10s AN: B cell dependence on and response to accessory
VH FAMILY USAGE BY ANTI-DNA-PRODUCING CELLS
signals in murine lupus strains. J Exp Med 157:18151827, 1983
9. Theofilopoulos AN, Dixon FJ: Murine models of SLE.
Adv Immunol 37:269-285, 1985
10. Schur PH, Sandson J: Immunological factors and clinical activity in SLE. N Engl J Med 278533-540, 1968
11. Papoian R, Pillarisetty R, Talal N: Immunological regulation of spontaneous antibodies to DNA and RNA. 11.
Sequential switch from IgM to IgG in NZBBZW F1
mice. Immunology 32:75-79, 1977
12. Datta SK, Pate1 H, Berry D: Induction of a cationic shift
in IgG anti-DNA autoantibody. J Exp Med 165:12521268, 1987
13. Ebling F , Hahn BH: Restricted subpopulations of DNA
antibodies in kidneys of mice with systemic lupus:
comparison of antibodies in serum and renal eluates.
Arthritis Rheum 23:392403, 1980
14. Yoshida H, Kohno A, Ohta K , Hirose S, Maruyama N,
Shirai T: Genetic studies of autoimmunity in New
Zealand mice. 111. Associations among anti-DNA antibodies, NTA, and renal disease in (NZB x NZW)F, x
NZW backcross mice. J Immunol 127:433437, 1981
15. Dixon FJ, Oldstone MB, Tonietti G: Pathogenesis of
immune complex glomerulonephritis of New Zealand
mice. J Exp Med 134 (suppl):65S-71S, 1971
16. Tan EM: Autoantibodies to nuclear antigens (ANA):
their immunobiology and medicine. Adv Immunol 33:
167-240, 1982
17. Morimoto C, Sano H, Abe T, Homma M, Steinberg AD:
Correlation between clinical activity of systemic lupus
erythematosus and the amounts of DNA in DNA/antiDNA antibody immune complexes. J Immunol 129:
1960-1965, 1982
18. Hayakawa K, Hardy RR, Parkside DR, Herzenberg LA:
The Ly-1 B cell subpopulation in normal, immunodeficient and autoimmune mice. J Exp Med 157:202-218,
1983
19. Klinman DM: Current advances in murine lupus. Curr
Opin Immunol 1:74&746, 1989
20. Panosian-Sahakian N , Klotz JL, Ebling F, Kronenberg
M, Hahn B: Diversity of Ig V gene segments found in
anti-DNA autoantibodies from a single (NZB X
NZW)F, mouse. J Immunol 142:4500-4506, 1989
21. Shlomchik MJ, Aucoin AH, Pisetsky DS, Weigert MG:
Structure and function of anti-DNA autoantibodies derived from a single autoimmune mouse. Proc Natl Acad
Sci U S A 84:915&9154, 1987
22. Smith RG, Voss EW Jr: Variable region primary structures of monoclonal anti-DNA autoantibodies from
NZBRVZW mice. Mol Immunol27:463-470, 1990
23. Kofler R, Dixon FJ, Theofilopoulos AN: The genetic
origins of autoantibodies. Immunol Today 7:374-378,
1987
24. Trepicchio W Jr, Barrett KJ: Eleven MRL-lpr/lpr antiDNA autoantibodies are encoded by genes from four V,
567
gene families: a potentially biased usage of V, genes. J
Immunol 138:2323-2331, 1987
25. Tillman DM, Jou N-T, Hill RJ, Marion TN: Both IgM
and IgG anti-DNA antibodies are the products of clonally selective B cell stimulation in (NZB X NZW)F,
mice. J Exp Med 176:761-779, 1992
26. Taki S, Hirose S, Kinoshita K , Nishimura H , Shimanura
T , Hamuro J, Shirai T: Somatically mutated IgG antiDNA antibody clonally related to germ-line encoded
IgM anti-DNA antibody. Eur J Immunol 22:987-992,
1992
27. Shlomchik M, Mascelli M, Shan H , Zadic MS, Pisetsky
D, Marshak-Rothstein A, Weigert M: Anti-DNA antibodies from autoimmune mice arise by clonal expansion
and somatic mutation. J Exp Med 171:265-297, 1990
28. Woloski B, Fuller G: Identification and partial characterization of hepatocyte-stimulating factor from leukemia cell lines: comparison with IL-1. Proc Natl Acad Sci
U S A 82~1443-1447, 1985
29. Freitas AA, Guilbert B, Holmberg D, Wennerstron G,
Coutinho A, Avrameas S: Analysis of autoantibody
reactivities in hybridoma collections derived from normal adult BALBlC mice. Ann Inst Pasteur Immunol
137D:3345, 1986
30. Berneman A, Ternynck T , Avrameas S: Natural mouse
IgG reacts with self antigens including molecules involved in the immune response. Eur J Immunol22:625633, 1992
31. Kelsoe G: Cloning of mitogen and antigen reactive B
lymphocytes on filter paper disks: phenotypic and genotypic analysis of B cell colonies. Methods Enzymol
1501287-304, 1987
32. Kelsoe G: Cloning of mitogen- and antigen-reactive B
lymphocytes on filter paper discs: phenotypic and genotypic analysis of B cell colonies. J Immunol Methods
76:345-352, 1985
33. Klinman DM: IgGl and IgG2a production by autoimmune B cells treated in vitro with IL-4 and IFN-y. J
Immunol 144:2529-2534, 1990
34. Klinman DM, Steinberg AD: Systemic autoimmune
disease arises from polyclonal B cell activation. J Exp
Med 165:1755-1760, 1987
35. Klinman DM, Steinberg AD: Proliferation of anti-DNA
producing NZB B cells in a non-autoimmune environment. J Immunol 137:69-75, 1986
36. Sedgwick JD, Holt PG: A solid phase immunoenzymatic
technique for the enumeration of specific antibody secreting cells. J Immunol Methods 57:301-309, 1983
37. Ando DG, Ebling FM, Hahn BH: Detection of native
and denatured DNA antibody forming cells by the
enzyme-linked immunospot assay: a clinical study of
(New Zealand black x New Zealand white)F, mice.
Arthritis Rheum 29: 1139-1 146, 1986
38. Monestier M, Manheimer-Lory A, Bellon B, Painter C,
Dang H , Talal N , Zanetti M, Schwartz R, Pisetsky D,
KLINMAN ET AL
39.
40.
41.
42.
43.
44.
45.
46.
47.
Kuppers R, Rose N, Brochier J, Klareskog L, Holmdahl
R, Erlanger B, Alt FW, Bona CA: Shared idiotypes and
restricted immunoglobulin variable region heavy chain
genes characterize murine autoantibodies of various
specificities. J Clin Invest 78:753-759, 1986
Kastner DL, McIntire TM, Mallett CP, Hartman AD,
Steinberg AD: Direct quantitative in situ hybridization
studies of Ig V, gene utilization: a comparison between
unstimulated B cells from autoimmune and normal mice.
J Immunol 143:2761-2767, 1989
Klinman DM, Eisenberg RA, Steinberg AD: Development of the autoimmune B cell repertoire in MRL-lpr/lpr
mice. J Immunol 144506-512, 1990
Jeong HD, Teale JM: Vh gene family repertoire of
resting B cells. J Immunol 143:2752-2760, 1989
Teale JM, Morris EG: Comparison of V, gene family
expression in adult and fetal B cells. J Immunol 143:
2768-2772, 1989
Barrett KJ: Anti-DNA antibodies, autoimmunity and the
immunoglobulin repertoire. Int Rev Immunol 5:43-78,
1989
Datta SK, Owen FL, Womack JE, Riblet RJ: Analysis
of recombinant inbred lines derived from “autoimmune” NZB and “high leukemia” C58 strains: independent multigenic systems control B cell hyperactivity,
retrovirus expression and autoimmunity. J Immunol
129:153%1544, 1982
Gavalchin J , Nicklas JA, Eastcott JW, Madaio MP,
Stollar BD, Schwartz RS, Datta SK: Lupus prone (SWR
x NZB)F, mice produce potentially nephritogenic autoantibodies inherited from the normal SWR parent. J
Immunol 134:885-892, 1985
Kofler R, Perlmutter RM, Noonan DJ, Dixon FJ, Theofilopoulos AN: Ig heavy chain variable region gene
complex of lupus mice exhibits normal restriction fragment polymorphism. J Exp Med 162336355, 1985
Alt FW, Blackwell TK, Yancopoulos GD: Development
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
of the primary antibody repertoire. Science 238: 10791087, 1987
Tutter A, Brodeur P, Shlomchik M, Riblet RJ: Structure, map position and evolution of two newly diverged
mouse Ig Vh gene families. J Immunol 147:3215-3223,
1991
Yancoupoulos GD, Desiderio SV, Paskind M, Kearney
JF, Baltimore D: Preferential utilization of the most
JH-proximal VH gene segments in pre-B cell lines.
Nature 311:727-733, 1984
Perlmutter RM, Kearney JF, Chang SP, Hood LE:
Developmentally controlled expression of Ig Vh genes.
Science 227: 1587-1590, 1985
Glotz D, Sollazzo M, Riley S, Zanetti M: Isotype, Vh
genes and antigen-binding analysis of hybridomas from
newborn normal BALB/c mice. J Immunol 141 :383-390,
1988
Bona CA: V genes encoding autoantibodies: molecular
and phenotypic characteristics. Annu Rev Immunol
6~327-358, 1988
Hahn BH, Ebling FM: Idiotype restriction in murine
lupus: high frequency of three public idiotypes on serum
IgG in nephritic NZB/NZW F, mice. J Immunol 138:
21 10-21 18, 1987
Yoshida H , Yoshida M, Izui S, Lambert PH: Distinct
clonotypes of anti-DNA antibodies in mice with lupus. J
Clin Invest 76:685-692, 1985
Shlomchik M, Nemazee D, van Snick J, Weigert M:
Variable region sequences of murine IgM anti-IgG monoclonal autoantibodies (rheumatoid factors). J Exp Med
165:970-987, 1987
Watanabe-Fukunaga R, Brannan CI, Copeland NG,
Jenkins NA, Nagata S: Lymphoproliferation disorder in
mice explained by defects in Fas antigen that mediates
apoptosis. Nature 356:314-317, 1992
Stollar BD, Schwartz RS: Monoclonal anti-DNA antibodies: the targets and origins of S L E autoantibodies.
Ann N Y Acad Sci 465:192-199, 1986