вход по аккаунту


Genetic control of rheumatoid factor production in the mouse Role of Genes Linked to the Immunoglobulin Heavy Chain Locus and to the Major Histocompatibility Complex.

код для вставкиСкачать
Role of Genes Linked to the Immunoglobulin Heavy Chain Locus and to
the Major Histocompatibility Complex
The influence of Igh-1 and H-2 linked genes on
the production of IgG2a-specific rheumatoid factors
(RF) by 129/Sv mice was examined. Heterozygous animals carrying the 129 alleles, H-2b and Igh-la, and
either H-2k, H-2", or Igh-lb (but not Igh-ld or Igh-1')
had considerably reduced RF levels. In contrast,
Igh-lb linked genes failed to suppress RF production by
MRL/MpJ-lpr mice, This lack of suppression was
linked to the lymphoproliferation trait characteristic of
these animals.
Several observations suggest the existence of a
genetic control of rheumatoid factor (RF) production
in human disease. In rheumatoid arthritis, for example, the frequency of HLA-DRw4 is increased in
seropositive patients but normal in seronegative patients ( l ) , and in coal workers' pneumoconiosis the
risk of developing anti-IgG autoantibodies is related to
Gm allotypes (2). However, because of the complexity
of human genetics, it is difficult to examine the mechanisms of such genetic control in humans. We therefore
looked for RF responses in species, such as the
mouse, where the identification of specific genes and
From the Unit of Experimental Medicine, International
Institute of Cellular and Molecular Pathology and Universitt Cathohque de Louvain, Brussels, Belgium.
Supported by grants from the FNRS and FRSM, Belgium.
J. L. Van Snick, MD: Research Associate, FNRS; P. G.
Coulie, MD: Recipient of studentship, Fonds de Dtveloppement
Scientifique, Universitt Catholique de Louvain; M. Stevens, MD.
Address reprint requests to Dr. J . L. Van Snick, Unit of
Experimental Medicine, Universitt Catholique de Louvain, Avenue
Hippocrate 7430, B-1200 Brussels, Belgium.
Submitted for publication January 21, 1983; accepted in
revised form April 6, 1983.
Arthritis and Rheumatism, Vol. 26, No. 9 (September 1983)
their mode of action might be facilitated by the availability of inbred strains.
Studies initiated a few years ago demonstrated
that in certain colonies normal inbred mouse strains
spontaneously produce anti-IgG autoantibodies of the
RF type. For a given strain, the incidence of these RF
varies from one colony to another, and within the
same colony there are large differences between the
titers of individual strains, suggesting that both environmental and genetic factors affect this autoantibody
production (3,4). Although little is known about the
environmental factors responsible for the high incidence of RF in certain colonies, the role of infectious
agents seems important in this respect ( 4 3 . Mice of
the 129/Sv (129) strain are particularly responsive to
such stimuli, as indicated by the observation that in
many colonies they produce 10 to 20 times more RF
than most other strains.
The involvement of a gene linked to the IgG2a
heavy chain locus (Igh-1) in the genetic control of this
R F production was recently demonstrated in crosses
between the 129 and C57B1/6 (B6) strains, which carry
the Igh-1" and Igh-lb alleles, respectively. Among
their backcross and F2 progeny, a considerable reduction of RF titers was indeed observed in heterozygous
Igh-lab mice (6). Interestingly enough, the allotypelinked control of RF mentioned above for pneumoconiotic coal workers seems to operate in a similar way
since in this situation Gm heterozygosity also is associated with a lower RF incidence (2). This similarity in
the control of RF in mice and humans prompted us to
further investigate the genetic factors that influence
the levels of RF in the former species, and more
specifically, to examine the respective roles of
immunoglobulin and H-2 linked genes.
IgA anti-lgG2a(ng/mll
Figure 1. Standard curve of radioimmunoassay for IgA rheumatoid
factor (RF). A purified monoclonal IgA anti-lgG2a RF of 129/Sv
origin (A2902C1) was incubated for 4 hours at 37°C in polyvinyl
wells coated with a 10 pg/ml solution of mouse IgG2a. After
appropriate washing and a further 2-hour incubation with I2’Ilabeled anti-mouse IgA antibodies which had been purified from a
goat antiserum by affinity chromatography, the wells were counted
for radioactivity. Each point represents the mean (? 2 standard
deviations) of triplicate measurements.
Mice. 129/Sv (129) and C57B1/6 (B6) mice were
obtained from the animal colony of the ICP, Brussels,
Belgium; A/HeJ from the Central Laboratory, Nijmegen,
The Netherlands; AKNRholco (AKR) from Iffa-Credo, Les
Oncins, France; and MRL/MpJ-lpr (MRL/l) from the Jackson Laboratory, Bar Harbor, ME. 129 X B6 (129xB)
recombinant inbred (RI) strains developed by Dr. J. L.
GuCnet at the Institut Pasteur, Paris, France (7) were rnaintained at our laboratory under clean conventional conditions. The animals used in the present experiments had been
inbred for 14-19 generations.
Allotype typing. IgG2a of the b allotype was detected
by immunodiffusion with a BALB/c serum directed against
B6 IgG. IgG2a of the d and e allotypes was detected by solid
phase radioimmunoassay (RIA) with a 129 anti-AKR IgG
antiserum. Polyvinyl flexible microtiter plates (Flow Laboratories, VA) coated with polyclonal AKR IgG2a (10 &ml)
were first incubated with 129 anti-AKR IgG serum and, after
washing, with a mixture of 1251-labeledAKR IgG2a and the
serum to be tested. The binding of the label was inhibited by
AKR and N H e J sera which share IgG2a atlotypic specificity
number 5 according to the nomenclature of Herzenberg et al
(8). Therefore the same RIA could be used to analyze the
segregation of both the Igh-ld and Igh-le alleles in 129 x (129
x AKR)Fl and 129 x (129 x A/HeJ)FI backcross populations, respectively.
RF determinations. Anti-IgG2a RF levels were determined either by latex agglutination or by solid phase RIA
using polyvinyl wells as described in detail elsewhere (3,9).
The only modification introduced in the present experiments
was the used of monoclonal IgA and IgM R F to calibrate the
RIA (10). A typical standard curve for IgA anti-IgG2a R F is
shown in Figure I . Sera were usually diluted 11100 in
Tris(hydroxymethy1)-aminomethane-buffered saline supplemented with 5% fetal bovine serum. For crosses involving
the 129 strain only, data concerning IgA R F are shown but
IgM RF levels followed similar distribution patterns. Since it
was previously shown that the anti-IgG2a RF produced by
B6 mice react equally well with 129 as with isologous IgG2a
and that 129 RF has the same affinity for AKR and A/HeJ
IgG as for 129 IgG (4), anti-IgG2a autoantibodies were
measured in all crosses tested here by their binding to a
monoclonal IgG2a protein of 129 origin (1 103G4). This
protein was purified from ascites fluid by affinity chromatography on Protein A-Sepharose (Pharmacia, Uppsala, Sweden) and stepwise elution with 0.1M citrate buffers of
decreasing pH (6, 5 , and 3, respectively). By double
immunodiffusion with subclass-specific rabbit antisera, the
fraction eluted at pH 5 contained only IgG2a.
Statistical tests. Probabilities were calculated from x2
tests. Yates correction was included whenever appropriate.
H-2 typing. Blood drawn from the retroorbital plexus
was collected in tubes containing heparin. After 3 washes
with O.15M NaCl, packed red blood cells (10 pl) were
incubated for 2 hours at room temperature with anti-H-2 sera
and, after 3 further washes, with 12’I-labeledanti-mouse IgG
antibodies which had been purified from a rabbit antiserum
by affinity chromatography. Two hours later the radioactivity of the washed pellet was measured. Anti-H-2 sera were
obtained from Dr. J. G. Ray, National Institutes of Health,
Bethesda, MD. H-2” and H-2k were detected with a (129 x
A.TL) anti-A. AL serum that specifically recognizes Kk.
H-2b typing of 129/Sv cells was done with monoclonal antiH-2Kb antibody B 8-24-3 (1 1).
Influence of the Igh-lb allele on RF levels in
crosses of the 129 strain with 129 X B RI strains. Mice of
the 129 strain maintained in the specific pathogen-free
colonies of the Institut Pasteur, Paris, and of the ICP,
Brussels, spontaneously produce large amounts of RF
specific for autologous IgG2a. B6 mice maintained
under identical conditions do not. Previous titrations
of RF in segregating crosses of 129 and B6 strains,
which carry the Igh-1”” and Igh-lbb genotypes respectively, have demonstrated that high RF levels occur
only in Igh-I”” animals (6). To verify these data and
find possible recombinants that would express the b
‘i 1000
z 500
Igh- 1
/129 x
Figure 2. -1gGZa-specific rheumatoid factor levels in crosses between 129/Sv mice and 129xB R1 strains. Agglutination titers of
IgG2a-coated polystyrene particles measured as described (2) in the
serum of individual mice derived from various crosses between
129/Sv and 129xB strains (0 = females, 0 = males). Mice were
tested when more than 20 weeks old.
Igh-lb allele on R F production by 129 mice, it was of
interest to introduce this gene into the MRL/l genome.
As previously reported by others (14), we found
that (B6 x MRL/l)F, mice had low R F levels. When
backcrossed to MRL/l females, approximately 60% of
the offspring from these F, mice had IgM anti-IgG2a
levels greater than 10 p g / d . Yet no linkage was
detected between low R F levels and Igh-lab genotype
(Figure 3). In contrast, a good correlation was found
between elevated R F levels and the lymphadenopathy
which is a characteristic feature of the MRL/l lupuslike disease. Of 20 mice tested, strong R F responses
were found in 7 of 8 with lymphadenopathy but in only
2 of 12 with normal lymph nodes (x’ = 7.08;P < 0.01).
Influence of the Igh-ld and Igh-1‘ alleles on RF
levels in crosses of the 129 with low RF strains. In order
to find out whether all “non-a” Igh-I alleles would
lead to suppression of high R F levels in the 129 strain,
we tested the influence of the Igh-ld and Igh-1‘ alleles.
129 mice were thus mated to AKR (Igh-ld) and A/HeJ
allotype but fail to suppress high R F titers when
crossed with 129 mice, we analyzed the R F levels in
the progeny of crosses between 129 mice and nine
129xB RI strains. The proportions of animals with
high R F levels ranged from 30 to 80% of the offsprings
in crosses with RI strains that carried the Igh- 1””
genotype but fell to about 1% of the animals derived
from crosses with Igh-lbb RI strains (Figure 2). In two
crosses (129 x G and 129 x R). females apparently had
higher R F titers than males. However these differences were not statistically significant (x2 < 3.8).
Despite their genetic homogeneity, individual offsprings of FI crosses between 129 and Igh-1”” 129xB
RI strains showed considerable variation in their RF
responses. This problem of penetrance is probably
related to the fact that R F production by 129 mice
depends upon their exposure to particular environmental factors (4).
Lack of influence of the Igh-lb allele on RF
production by MRL/l mice. The MRL/l, which like the
129 carry the Igh-la allele, develop a severe lupus-like
disease with massive lymphoproliferation resulting in
a characteristic lymphadenopathy. Mice of this strain
produce large amounts of IgM and IgG R F (12,13). In
view of the considerable suppressive influence of the
.a.. -
Figure 3. IgG2a-specific IgM rheumatoid factor (RF) in crosses
between MRL/l and C57B1/6 mice. IgM anti-IgG2a RF levels were
measured by radioimmunoassay as described in Materials and
Methods. Each point represents an individual mouse (0 = females,
0 = males). Mice were tested when more than 20 weeks old.
Table 1. Influence of Igh-1 genotype on rheumatoid factor (RF) levels in crosses between 129/Sv,
AKR, and AlHeJ strains
No. mice with.RF
No. mice
Igh-l genotype*
<I0 pglml
129 x (129 x A/HeJ)FI
129 X (129
210 &ml
1156 )
* Igh-l genotypes were determined by testing individual sera for the presence of the allotypic forms of
IgG2a contributed by the AKR and A/HeJ parents, i.e., d and e respectively (see Materials and
I’The levels of IgA anti-IgG2a R F were measured as described in Materials and Methods in the serum
of mice more than 20 weeks old.
$ NS = not significant.
(Igh-le) strains, both of which produce little RF. All F1
hybrids had less than 10 pghl IgA anti-IgG2a RF, but
when backcrossed to 129 mice they generated a significant number of animals with high RF levels. For both
backcrosses, approximately one-fourth of the offspring had more than 10 pg/ml IgA RF, 23% for 129 X
(129 x A/HeJ)FI and 26% for 129 x (129 x AKR)FI.
These proportions were similar to that previously
reported for the 129 X (129 x B6)F, backcrosses (6).
However, unlike what had been observed for the latter
mice, no significant differences could be detected
between allotype homozygous and heterozygous offspring (Table 1). Therefore, the ability to suppress
high RF levels in 129 mice seems to be a peculiarity of
a gene closely linked to the Igh-lb allele.
Influence of H-2 genotype on RF levels. AKR
and NHeJ mice differ from the 129 not only by their
Igh-1 genotypes but also by their H-2 genotypes
(H-2kk,H-2”, and H-2bb,respectively). The backcross
mice generated for the R F allotype linkage tests could
thus also be used to detect a possible influence of the
H-2 genotype on RF levels (Table 2). Among the
offspring of the 129 x (129 x AKR)FI backcrosses, the
relative risk of having high RF levels was reduced to
0.07 for H-2bk mice. A greater difference still was
observed between H-2bband H-2baoffspring in the 129
x (129 x A/HeJ)F, crosses, with a relative risk of only
0.013 for the latter. As shown in Figure 4, this influence of H-2ba heterozygasity was especially marked
with female mice.
We have previously shown that in certain specific pathogen-free colonies, mice of the 129 strain
spontaneously produce large quantities of RF-like
anti-IgG autoantibodies that specifically react with
mouse IgG2a, except when the latter is of the b
allotype (3). In young animals these rheumatoid factors mainly belong to the IgM class, but with age there
is a progressive switch to IgA that is paired with a
sharp increase in the titers between 10 and 20 weeks.
This RF production is not associated with any obvious
pathology and affects neither breeding performance
nor life expectancy (unpublished results).
Genetic factors play an important role in the
development of this RF response since 129 mice
frequently have 10-20 times more RF than most other
mouse strains maintained in the same colonies. The
results reported here confirm and extend two observa-
Table 2. Influence of H-2 genotype on rheumatoid factor (RF) levels
No. mice with RFlevels*
No. mice
H-2 genotype
< l o pghl
129 x (129 x A/HeJ)F,
210 pg/ml
* IgA anti-IgG2a R F levels measured by radioiminunoassay in the serum of mice more than 20 weeks
tions made previously in crosses between the 129 and
B6 strains: the recessive character of the 129 high R F
phenotype and the Igh-1' linked suppression of RF.
The former observation was corroborated by the low
RF titers of (129 X AKR)FI and (129 X A/HeJ)FI
hybrids and the second by the dramatic drop of RF
levels in crosses of the 129 strain with 129xB RI
strains that carry the Igh-I " genotype. This Igh-lb
linked suppression did not work in MRL/I mice despite the fact that they have the same Igh-1 genotype
as the 129 strain. It is not known whether this is due to
differences in the stimuli that induce RF in either strain
or to the severe alterations of the immune system of
the MRL/l that may upset normal regulatory pathways
in these mice. Further analysis of the influence of IghI genes on RF levels indicated that Igh-ld and Igh-1"
did not significantly affect RF in mice derived from
backcrosses of the 129 strain with (129 x AKR)FI and
(129 x A/HeJ)FI, respectively. In contrast, in the
same crosses, H-2k and H-2" linked genes were found
to considerably reduce RF levels. However, not all H2 haplotypes were able to suppress RF production in
mice derived from the 129 strain. High RF titers were
recently found in both H-2 heterozygous and homozygous offspring of 129 x (129 x RIIIS/J)FI backcrosses,
indicating that the H-2' haplotype contributed by the
RIIIS/J did not influence RF synthesis (data not
The present data clearly show that RF levels
are genetically controlled in the 129 mouse; however,
they do not indicate whether this genetic control acts
directly on RF synthesis or on the stimuli that induce
RF. Rheumatoid factor production by 129 mice is most
probably related to latent infections since animals
reared in isolation which were offspring of RF-positive
dams are free of R F (manuscript in preparation).
Hence the unusually high R F titers of the 129 strain
could be due to some selective deficiency in their
immune defenses against a particular infectious agent.
It is noteworthy in this respect to recall the observation by Parker and coworkers (15) that 129 mice are
about 10,000 times more sensitive to infection by
Sendai virus than B6 mice. Although Sendai virus has
never been detected in our RF-positive mice, the
extreme sensitivity of the 129 strain to this virus raises
the possibility that these mice would be abnormally
susceptible to some other infectious agents as well.
The reduced R F levels of H-2b" and H-2bk mice
observed here could then merely reflect the existence
of H-2" and H-2k linked immune response genes that
would afford better resistance to the R F inducing agent
C r o s s ( l 2 9 x A / H e J ) F l 129x1 129xA/HeJ)F 1
Figure 4. IgG2a-specific IgA rheumatoid factor (RF) in crosses
between 129/Sv and A/HeJ mice. IgA anti-IgG2a R F levels were
measured by radioimmunoassay as described in Materials and
Methods. Sera were appropriately diluted in Tris-buffered saline
containing 5% fetal bovine serum, and the R F levels were calculated
on a calibration curve run in parallel with a monoclonal IgA R F
specific for IgGZa. Each point represents an individual mouse (0=
females, 0 = males). Mice were tested when 20-24 weeks old.
and consequently reduce RF levels by eliminating the
stimulus for RF production. Such enhanced immunologic surveillance has been reported previously in mice
heterozygous at the H-2 gene complex (16). This
interpretation of the data is further supported by the
apparent dominance of the low RF phenotype observed in our experiments. Such dominance of the
nonresponder trait is at variance with the rules established for H-2 linked immune response genes against
both alienand self antigens (17-19). This discrepancy
would be eliminated if, as suggested above, H-2 linked
genes were to affect RF levels indirectly since in that
case the apparent dominance of the low RF phenotype
would indeed correspond to dominance of the high
response against a foreign antigen.
A similar indirect mechanism implying a better
resistance against infection could be operating for the
Igh-lb linked suppression of RF since variable region
genes encoding antibodies of higher affinity for the RF
inducing agent could be associated with the b allotype.
However, in the case of the Igh-lb linked control,
another possibility may be envisaged that is suggested
by the allotypic specificity of the RF of 129 mice.
These autoantibodies react very well with the a, d, and
e allotypic forms of IgG2a but very poorly if a t all with
the b allotype (4).It is therefore intriguing that t h e Ighlb allele suppresses R F in 129 mice, whereas the Igh-ld
and Igh-1“ alleles d o not. The possibility that the low
RF titers of Igh-lab mice actually result from the
reduced levels of IgG2a” molecules capable of stimulating R F production may therefore be cautiously
considered. Our failure with 129xB RI strains t o
detect any recombination between the capacity t o
suppress R F and t h e Igh-lb gene itself is compatible
with this view.
A t present, it is not possible to exclude the
alternative interpretation that high RF production results from deficient suppressive mechanisms which
could b e restored by genes linked to certain H-2 or
Igh-1 haplotypes (20). Although a definitive answer t o
this issue will probably require the identification of the
elusive environmental agents that induce R F in our
mice, it seems likely that further analysis of the genes
involved in R F control may in turn help to clarify the
origin of R F itself.
We are deeply indebted to Dr. J. L. Guenet for
making the (129xB) RI strains of mice available to us and to
Dr. P. Masson for support and critical reading of the
manuscript. The secretarial assistance of Mrs. N. Joris is
gratefully acknowledged.
1. Dobloug JH, FBrre 0, Kiss E, Thorsby E: HLA antigens and rheumatoid arthritis. Arthritis Rheum 23:309313, 1980
2. Mentnech MS, Pearson DJ, Elliott JA, Taylor G , Major
PC: Correlation of Gm allotype with the incidence of
anti-IgG antibody in coal workers’ pneumoconiosis. Clin
Immunol Immunopathol 17:274-279, 1980
3. Van Snick JL, Masson PL: Age-dependent production
of IgA and IgM autoantibodies against IgG2a in a colony
of 129/Sv mice. J Exp Med 149:1519-1530, 1979
4. Van Snick JL. Masson PL: Incidence and specificities of
IgA and IgM anti-IgG autoantibodies in various mouse
strains and colonies. J Exp Med 151:45-55, 1980
5. Dresser DW, Keeler KD, Phillips JM: Immunoglobulin
allotypes of the mouse. Biochern SOCTrans 4:34-38,
6 . Van Snick JL: A gene linked to the Igh-C locus controls
the production of rheumatoid factor in the mouse. J Exp
Med 153:738-742. 1981
7. Taylor BA: Recombinant inbred strains, Genetic Variants and Strains of the Laboratory Mouse. Edited by M
Green. Stuttgart, Gustav Fischer Verlag, 1981, p 397
8. Herzenberg LA, McDevitt HO, Herzenberg LA: Genetics of antibodies. Ann Rev Genet 2:209, 1968
9. Masson PL, Cambiaso CL, Collet-Cassart D, Magnusson CGM, Richards CB, Sindic CJM: Particle counting
immunoassay (PACIA), Methods in Enzymology. Edited by JJ Langone, H Van Vunakis. New York, Academic Press, 1981. pp 106-139
10. Van Snick JL, Coulie P: Monoclonal anti-IgG autoantibodies derived from lipopolysaccharide-activated spleen
cells of 129/Sv mice. J Exp Med 155:219-230, 1982
11. Kohler G. Fischer-Lindahl K, Heusser C: Characterization of monoclonal anti-H-2Kb antibody, The Immune
System. Vol. 2. Edited by C Steinberg, I Lefiovits.
Basel, S. Karger, 1981, p 202
12. 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 148:1198-1215, 1978
13. Eisenberg RA, Thor LT, Dixon FJ: Serum-serum interactions in autoimmune mice. Arthritis Rheum 22: 10741081, 1979
14. Dixon FJ, Theofilopoulos AN, Izui S, McConahey PJ:
Murine SLE-aetiology and pathogenesis, Fourth International Congress of Immunology. Edited by M Fougereau, J Dausset. London, Academic Press, 1980, pp 959995
IS. Parker J, Whiteman M, Richter C: Susceptibility of
inbred and outbred mouse strains to Sendai virus and
prevalence of infection in laboratory rodents. Infect
Immun 19:123, 1978
16. Doherty PC, Zinkernagel RM: Enhanced immunological
surveillance in mice heterozygous at the H-2 gene complex. Nature 256:50-52, 1975
17. McDevitt HO, Deak BD, Shreffler DC, Klein J , Stimpfling HH, Snell GD: Genetic control of the immune
response: mapping of the Ir-1 locus. J Exp Med
135:1259-1278, 1972
18. Vladutiu A, Rose N: Autoimmune murine thyroiditis
relation to histocompatibility (H-2) type. Science
174~1137-1139, 1971
19. Berman PW, Patrick J: Linkage between the frequency
of muscular weakness and loci that regulate immune
responsiveness in murine experimental myasthenia gravis. J Exp Med 152:507, 1980
20. Owen FL, Riblet R, Taylor BA: The T suppressor
alloantigen Tsud maps near immunoglobulin allotype
genes and may be a heavy chain constant region marker
on a T cell receptor. J Exp Med 153:801, 1981
Без категории
Размер файла
521 Кб
production, major, histocompatibility, locus, complex, chains, mouse, control, factors, immunoglobulin, role, heavy, genes, genetics, rheumatoid, linked
Пожаловаться на содержимое документа