| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Simmon’s Arthritis Research Center and Center for Immunology, University of Texas Southwestern Medical School, Dallas, TX 75390; and
Rheumatology Division, Northwestern University Medical School, Chicago, IL 60611
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Of the loci that have been identified in the NZM2410 and BWF1 models, several are NZW derived (such as Sle1, Sle3, and part of Sle2), whereas others such as Nba1 and Nba2 are NZB encoded. Thus, it is clear that the epistatic interaction of NZB-derived and NZW-derived loci is required for the genesis of autoimmunity in these models. However, the NZW strain is not the only strain that can engender lupus in epistasis with NZB; SWR is another strain that does so (26, 27). Thus, (SWR x NZB)F1 (or SNF1) mice develop lupus nephritis that is clinically very similar (in onset, severity, female bias, and pathology) to the BWF1 model. Indeed, studies in this model have clearly demonstrated the relative contributions of Ag-specific and Ag-nonspecific modalities of T cell:B cell crosstalk toward disease pathogenesis (reviewed in Refs. 28 and 29). However, the genetic contributions of the SWR genome to disease in this model are poorly understood. Extrapolating from the findings in the BWF1 and NZM2410 models, it is reasonable to expect several SWR-derived loci to be important for disease development in this model. To elucidate the dominant SWR contributions to lupus, we have chosen to examine a panel of 86 (NZB x SNF1) backcross (BC) mice to ascertain whether the development of autoantibodies, glomerulonephritis (GN), or early mortality could be linked to any specific (dominant) SWR and (recessive) NZB loci, using a panel of 122 microsatellite markers, spanning all 19 autosomes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The derivation of the 86 (NZB x SNF1) BC mice has previously been detailed (27, 30). Thus, all loci in the progeny from this cross bear at least one NZB allele (derived from the NZB parent), while the other allele is either SWR or NZB in origin, as it arises from the F1 parent. Mice were sero-tested at 9–12 mo of age, or earlier if moribund. Seropositive mice were monitored for evidence of proteinuria, at weekly intervals, using albustix strips. Mice that exhibited persistent proteinuria for 2 consecutive wk, or appeared moribund, were sacrificed. Upon sacrifice, their kidneys were examined for evidence of GN. The glomeruli were screened for evidence of hypertrophy, proliferative changes, hyaline deposits, and/or basement membrane thickening, in a blinded fashion. The severity of GN was graded on 0–4 scale, in which the grades 1, 2, 3, and 4 were accorded when 1–10%, 11–25%, 26–50%, and >50% of the glomeruli were affected, respectively, as detailed elsewhere (31).
The following phenotypes were used for linkage analysis: serum IgG
antinuclear autoantibodies (ANAs), IgM Abs, histological GN, and age of
death. The ANA specificities that were examined are detailed below. For
all phenotypes studied, the mice were arbitrarily classified as being
negative, low positives, or strong positives, following the criteria
listed in Table I
. As detailed below,
these phenotypic categories were classified in three different ways,
for the purpose of linkage analysis. No significant sex bias was seen
in any of the phenotypes studied. Therefore, the data from both the
male and female progeny were combined for analysis.
|
Total serum IgM and IgG levels were assayed using a sandwich ELISA. Briefly, goat anti-mouse IgM or IgG (Boehringer Mannheim, Indianapolis, IN) was first coated onto Immulon I plates (Dynatech, Chantilly, VA) and blocked. Serum was diluted serially and added to the plates for 2 h at room temperature. Bound Ig was revealed with alkaline phosphatase-conjugated goat anti-mouse IgM or IgG Abs (Boehringer Mannheim), using p-nitrophenyl phosphate as a substrate. Serial dilutions of isotype-specific Ig standards were also added to each plate for quantitation and interplate standardization.
ELISA for autoantibodies
IgM anti-ssDNA, IgM anti-histone, and IgM
anti-DNP/keyhole limpet hemocyanin Abs were assayed as described
(20). The anti-dsDNA, anti-histone, and
anti-histone/DNA ELISAs were conducted as detailed elsewhere
(32). Briefly, for the anti-dsDNA ELISA, Immulon II
plates (Dynatech) precoated with methylated BSA were coated overnight
with 50 µg/ml dsDNA (Sigma-Aldrich, St. Louis, MO), dissolved in PBS,
and filtered through cellulose acetate before use. For the
anti-histone/DNA ELISA, the dsDNA-coated plates were then
postcoated with 10 µg/ml total histones (a mixture of all histones;
purchased from Boehringer Mannheim) overnight at 4°C. After blocking
with PBS/3% BSA/0.1% gelatin/3 mM EDTA, 1/100 dilutions of the test
serum or 1/2 dilutions of culture supernatants were incubated in
duplicate for 2 h at room temperature. Bound IgG was detected with
alkaline phosphatase-conjugated anti-mouse IgG (Jackson
ImmunoResearch Laboratories, West Grove, PA), using
p-nitrophenyl phosphate as a substrate. Raw OD was
converted to U/ml, using a positive control serum derived from an
NZM2410 mouse, arbitrarily setting the reactivity of a 1/100 dilution
of this serum to 100 U/ml. Sera with reactivities stronger than the
test standard were diluted further and reassayed. The ANA serotiters in
the BC mice were compared with sera from a panel of twelve 9- to
12-mo-old SWR control mice. Serotiters that exceeded the SWR means by 3
or 8 SDs were classified as being low positives, or strong positives,
respectively, as indicated in Table I
.
Genotyping
Oligonucleotide primers flanking microsatellite repeats were synthesized commercially (Integrated DNA Technologies, Coralville, IA). A panel of 122 primers that readily distinguished SWR from NZB alleles was utilized. Together, they spanned 1500 cM of the autosomal genome (covering all 19 Chr), with an average intermarker distance of 10 cM. The sequences of the selected primers are publicly available (http://www-genome.wi.mit.edu/). The chromosomal positions of the primers with respect to the acromere are reported in accordance with the Mouse Chromosome Committee Reports obtained through the Encyclopedia of the Mouse Genome, Mouse Genome Database, The Jackson Laboratory (Bar Harbor, ME) (http://www.imformatics.jax.orgl). Genotyping was performed using tail DNA. PCR amplification was performed in an Eppendorf MasterCycler (Eppendorf Scientific, Westbury, NY), generally using 30 cycles of 30 s at 94°C, 1 min at 55°C, and 30 s at 72°C. For certain primers, the annealing temperature was ramped up to 60°C or ramped down to 50°C, for optimal results. Amplified products were electrophoresed onto 5% agarose gels and visualized by ethidium bromide staining and UV transillumination. The mice were then scored as being "N" (or NN, homozygous for the NZB allele), or "H" (heterozygous for SWR/NZB alleles).
Linkage analysis and statistics
Linkage analysis was performed using two different software
programs, MapManager.QTX
(http://mapmgr.roswellpark.org/mapmgr.html) and Bymarker
(http://www.infosci.coh.org/jal/bymarker02/bymarker02.htm), as
described (33). These two programs use very different
algorithms. The Bymarker software performs marker-specific linkage
analysis and calculates the log likelihood of the odds (LOD)
scores and p values at each locus, using the Pearson’s
2 test. In contrast, MapManager.QTX is a QTL
mapping program that was used to perform interval mapping for each
entire Chr, using the Kosambi mapping function. As evident from the
results, there is good concordance between these two algorithms. The
phenotype data were classified and analyzed in three different modes.
In mode A, all offspring were classified as being either negative or
positive (i.e., including both low positives and strong positives),
following the criteria outlined in Table I. In classification mode B,
only the high positives were considered to be positive, whereas all
other progeny were classified as negative. Finally, classification mode
C entailed extreme phenotype analysis (negatives and strong positives
only), essentially excluding the low positives from analysis. Thus,
whereas classification modes A and B utilized all 86 progeny, mode C
used only 
80–90% of the progeny, having excluded the borderline
positives. As is evident in Results, all three
classification modes were successful in identifying most of the
loci.
Since genome-wide search essentially entails multiple hypotheses
testing, a threshold for suggestive linkage was set at LOD > 1.9,
p < 0.0034 (2 > 8.6, 1 df),
based on the recommendation of Lander and Kruglyak (34).
The threshold for significant or probable linkage was set at a LOD
> 3.3, p < 1 x 10-4
(2 > 15.1, 1 df). In this communication, all
loci showing evidence of linkage to any of the tested lupus traits at
p < 0.01 are reported. SWR-derived loci that were
identified at least at the suggestive level (LOD > 1.9) by both
the Bymarker linkage analysis algorithm and the MapManager interval
mapping algorithm were assigned names.
Correlation between serum ANA, GN, and mortality was performed using
the 2 test after classifying individual mice
as being high positive, low positive, or normal for each given
phenotype. Likewise, epistatic interactions between loci
were tested for using 2 test, with Yate’s
correction for continuity where necessary.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As illustrated in Table II,
20–40% of BC progeny exhibit various component lupus phenotypes,
including autoantibodies, GN, or early mortality. This is
consistent with the notion that two or more loci, in various epistatic
combinations, are likely to be responsible for each of the studied
phenotypes. It is also clear that IgG anti-dsDNA, anti-ssDNA,
and anti-histone/DNA ANAs in these mice correlate with GN scores
and early mortality. Indeed, both the latter phenotypes themselves show
strong correlation with each other (p = 0.001).
These observations are consistent with the notion that the early
mortality in these mice is largely due to IgG ANA-induced GN. However,
it is noteworthy that the IgM Abs assayed do not correlate with the GN
scores or early mortality.
|
As illustrated in Table III, the
most potent locus affecting early mortality is the H2 locus
on proximal Chr 17 (D17mit36, 20 cM, LOD = 4.59 - 5.38),
with a dominant contribution from the SWR allele, as identified by both
algorithms. This allele is clearly linked to several other phenotypes,
as discussed below. The second largest contribution to mortality arises
from the recessive NZB allele on mid-Chr 4 (D4mit147, 55 cM, MapManager
LOD = 1.48, Bymarker LOD = 2.21), which qualifies as a
suggestive locus, at least by the Bymarker algorithm. It is to be noted
that this locus maps to the same position as the
Lbw2/Sbw2 locus (also of NZB origin) described by
Kono et al. (7) and the Sle2 locus (NZB/NZW
origin) described by Morel et al. (11). Finally, a weak
NZB recessive contribution is noted from a locus on mid-Chr
10.
|
. Once again, the H2 locus
plays a dominant role in GN susceptibility (MapManager LOD = 2.37,
Bymarker LOD = 2.70). There is also a dominant SWR contribution to
GN from proximal Chr 14 (D14mit37, 36 cM, MapManager LOD = 2.48,
Bymarker LOD = 2.71). This locus reaches the suggestive level of
significance using both analysis algorithms and is accorded the name
Swrl-2 (SWR lupus locus 2). A weaker contribution
to GN is noted from the SWR allele on mid Chr 6 (MapManager LOD =
1.5, Bymarker LOD = 1.43). This locus has been noted in previous
NZB/B6 crosses (12) and NZB/NZW crosses (Lbw4,
Ref. 7). Finally, the NZB allele on distal Chr 4 also
lights up as a suggestive locus for GN (75 cM, MapManager LOD =
2.15, Bymarker LOD = 2.23). Based on its position, strain of
origin, and linked phenotype, this locus is very likely to be the
Nba1 locus, described by Kotzin and coworkers (6, 8). It is to be noted that the three strongest GN-susceptibility
loci, H2, Swrl-2, and Nba1, are all identified as
being suggestive, both the MapManager and the Bymarker analysis
algorithms, almost irrespective of the phenotype classification mode
selected (Table IV).
|
There are several genetic contributions to IgG ANA formation, as
listed in Table V. Importantly, the three
strongest contributions all arise from the SWR strain. The strongest
SWR contribution arises from the H2 locus (MapManager
LOD = 5.48, Bymarker LOD = 4.92). A second important SWR
contribution arises from distal Chr 1, mapping similarly as
Nba2/Sle1 (6, 9, 11), with a LOD
score of 2.89 - 2.91. Interestingly, this locus shows the
strongest linkage to IgG anti-histone/DNA ANAs, rather than
anti-dsDNA ANAs, with little evidence of linkage to IgG
anti-ssDNA ANAs, GN, or mortality, as diagrammed by the interval
mapping chromosomal scan in Fig. 1. This
locus is assigned the name Swrl-1 (SWR lupus locus
1). Another SWR locus that impacts IgG anti-histone/DNA,
anti-dsDNA, and anti-ssDNA ANAs in a dominant fashion is
located on proximal Chr 18 (LOD > 2). This locus is accorded the
name Swrl-3 (SWR lupus locus 3). This locus appears to be a
novel locus implicated in murine lupus. In addition, as summarized in
Table V, there are also weaker contributions from the NZB genome to IgG
ANA formation, notably on proximal Chr 5 and telomeric Chr 10. Once
again, it is clear that the three strongest ANA -susceptibility loci,
H2, Swrl-1, and Swrl-3, are all identified as
being suggestive, both the MapManager and the Bymarker analysis
algorithms, irrespective of the phenotype classification mode selected
(Table V). Finally, several NZB and SWR loci contribute to the
formation of IgM Abs with specificities for ssDNA, histones, or
DNP/keyhole limpet hemocyanin, as summarized in Table VI. These include suggestive loci on
proximal Chr 1 (similarly positioned as BXSB-derived Bxs1)
and proximal Chr 6.
|
|
|
.
Also indicated in Table VII are the penetrance rates of the different
phenotypes in offspring with the indicated susceptibility genotypes, as
well as a listing of potential candidate genes within these loci. It
should be pointed out that none of these loci are absolutely required
for disease or ANA formation, because at least one phenotype-positive
animal could be identified bearing the nonsusceptible genotype at each
of these five susceptibility loci.
|
We next examined for evidence of epistatic interactions between
the different loci, with respect to the different phenotypes studied.
Although the H2 locus on Chr 17 and the
Sle2/Lbw2/Sbw2 locus on Chr 4 both confer susceptibility to
early mortality, we could not demonstrate any statistically significant
epistatic interactions between these two loci. However, we could
demonstrate epistatic interactions between the loci that were linked to
the other phenotypes. Among the loci that contribute to GN, those on
Chr 4 and Chr 14 demonstrate strong epistatic interaction, as depicted
in Table VIII. Thus, mice that bear both
susceptibility alleles (i.e., NZB homozygous at Nba1 on Chr
4 and NZB/SWR heterozygous at Swrl-2 on Chr 14) have
significantly higher incidence of GN (penetrance = 66%), compared
with progeny with any of the other three potential genotypes at these
loci. However, neither of these two loci shows any epistatic
interactions with the H2 locus. In other words, progeny that
are NZB/SWR heterozygous at H2 have a higher incidence of GN
(penetrance = 51%, Table VIII), irrespective of their genotypes
at the Nba1 and Swrl-2 loci.
|
67–71%, as detailed
in Table VIII. This is significantly higher than the penetrance of ANAs
seen in progeny bearing any of the other three two-locus genotypes.
Finally, Fig. 2 charts the incremental
contribution to disease and phenotype penetrance as increasing numbers
of susceptibility loci are inherited. Thus, mice that inherit three
strong susceptibility loci exhibit penetrance rates of GN and IgG ANAs
that parallel the observed penetrance of these phenotypes in the
original SNF1 lupus strain.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Of all these
loci, the locus that clearly towers above the rest is the H2
locus on Chr 17. Early mortality, GN, and IgG ANA formation are all
strongly facilitated by the SWR/NZB genotype at this locus with LOD
scores in excess of 4 (Tables III-V). This is also the one locus that
has been reproducibly identified in almost all murine lupus genetic
studies to date (1, 2, 3, 4, 7, 8, 11, 35, 36, 37, 38). Indeed,
heterozygosity at the H2 locus has also been implicated in
disease by earlier restriction-fragment length polymorphism studies in
the SNF1 model (30). More recently, the same locus has been mapped as the strongest suppressor locus in NZW x B6.Sle1 crosses (39). The later study demonstrates that the H2z/z genotype (or Sles1NZW/NZW) strongly shuts down Sle1-triggered ANA production. Indeed, the findings of the present study are consistent with all of the above reports. Thus, mice that genotype as H2SWR/NZB exhibit significantly higher penetrance of autoantibodies and disease, compared with mice that type as H2NZB/NZB. Moreover, among mice that are heterozygous at Swrl-1 (i.e., mice bearing the Chr 1 susceptibility locus that maps similarly to Sle1), having the H2NZB/NZB genotype on Chr 17 totally shuts down ANA production. All these observations support the recognition of the H2 locus as one of the most potent determinants of murine lupus susceptibility.
Although it is unequivocally accepted that the H2 locus is a
major genetic determinant, the culprit gene(s) within this locus still
remains undetermined. In most of these studies, heterozygosity at the
H2 locus (e.g., d/q, d/z, or b/z
genotypes) is most strongly associated with disease. The H2
molecules themselves are attractive candidates, and several interesting
ideas and potential mechanisms have been put forth to explain how and
why heterozygosity at this locus might confer disease susceptibility
(40, 41, 42). More recently, elegant I-A or
I-E transgene reconstitution experiments conducted by Kotzin
and coworkers (43, 44) have demonstrated that these
molecules do not really confer disease susceptibility, thus dampening
the earlier enthusiasm for these molecules. The TNF
gene
within this locus still remains as an attractive candidate
(45, 46, 47). Importantly, the H2 locus is richly
packed with several genes of immunological interest, a sampling of
which is listed in Table VIII. Based on these studies, it is reasonable
to expect the culprit gene not to be the H2 molecule itself,
but a linked gene, still shrouded in mystery.
Swrl-1 maps right into a chromosomal interval that is also
certainly not new to lupus researchers. It maps similarly to
Sle1 mapped in NZM2410/NZW strains (11), and
Nba2 mapped in the NZB strain (6, 9). Likewise,
a BXSB locus at this position has also been implicated in disease
(13). Most excitingly, a susceptibility locus for human
SLE has been recognized to be syntenic to these murine Chr 1 loci
(48, 49). It is certainly conceivable that the culprit
molecule(s) within these different murine (and human) loci may be
polymorphic variants of the same gene, but this awaits molecular
confirmation. In the meantime, congenic studies of Sle1, on
the normal B6 background, have revealed that Sle1 is a
potent locus that breaches tolerance, leading to high levels of
anti-histone/DNA (but not anti-dsDNA) ANAs, with little
attendant pathology (21). In this context, it is very
interesting to note that the phenotype that shows the strongest linkage
to Swrl-1 is IgG anti-histone/DNA ANAs, not anti-dsDNA,
anti-ssDNA ANAs, early mortality, or GN, as illustrated in Fig. 1.
These observations suggest that Swrl-1 might function
analogously to Sle1. This hypothesis is currently being
tested by the derivation of congenic strains bearing
Swrl-1.
In contrast to the H2 and Swrl-1 loci, the
Swrl-2 and Swrl-3 loci map to genomic positions
that have been infrequently implicated in murine lupus studies.
Interestingly, Rozzo et al. (9) have linked GN
susceptibility to a similarly positioned NZB locus on Chr 14, notably
in H2z-positive progeny. In view of
the colocalization and the similarity in the associated phenotype
(i.e., GN), Swrl-2 is likely to be a genuine GN
susceptibility locus, in which the SWR allele might be more potent that
the NZB counterpart. This hypothesis needs to be verified by congenic
dissection analysis. In contrast, Swrl-3 appears to be a
novel locus, mapping to a chromosomal position not previously
implicated in murine lupus susceptibility. Although it surfaces as a
suggestive locus using both the analysis algorithms, its authenticity
certainly warrants confirmation in additional, independent data sets.
Since it is linked only to IgG ANAs (but not to any of the other tested
phenotypes), it might play an important role in breaking tolerance to
nuclear Ags. Although there are several potential candidates within
this region (see Table VIII), the functional role of this locus would
become clearer once congenic strains bearing this locus become
available.
In addition to these four SWR loci, the NZB genome appears to be
contributing perhaps two recessive loci, both localized to Chr 4. The
NZB locus positioned similarly to Sle2/Lbw2/Sbw2 is linked
to early mortality, but not to any of the other tested phenotypes
studied (Table III). However, this locus has been linked to several
other phenotypes, including ANAs, splenomegaly, and GN, in crosses
performed with other strain combinations (7, 9, 11, 13, 17). These observations suggest that the expression of the locus
on mid-Chr 4 (55 cM) is heavily influenced by the rest of the genomic
context. However, it should be cautioned that this locus did not reach
suggestive levels of significance using the MapManager interval mapping
algorithm. In contrast, the more distal, NZB-derived Nba1
locus on Chr 4 (identified as being a suggestive locus by both mapping
algorithms) is linked to GN, but not ANAs, in the present study, as
well as in previous mapping studies by other investigators (6, 8). Indeed, it is tempting to speculate that this locus on
distal Chr 4 may be impacting local events in the end organs, but not
systemic immunity. These predictions need to be verified by congenic
studies.
As depicted in Table VIII and Fig. 2, it takes the epistatic
interaction of two to three susceptible loci to attain the high
penetrance rates seen in SNF1 mice. Thus, mice
with the resistant genotype at all three susceptibility loci have
<10% penetrance of GN or IgG ANAs (Fig. 2). With the inheritance of
each additional susceptibility locus, the penetrance (or chance of
expressing this phenotype) escalates by 20–30%. Thus, although
lupus pathogenesis in SWR/NZB mice is truly polygenic, the epistatic
interaction of two to three potent loci is all that may be required for
orchestrating disease. These findings parallel the observations in the
other murine models of lupus (1, 2, 3, 4) and are likely to
represent a generalized paradigm for both murine and human lupus.
Indeed, congenic reconstitution studies have provided
experimental support for this "2–3 gene" paradigm for lupus
susceptibility (4, 24, 25).
Mapping of these five loci (H2, Swrl-1, Swrl-2, Swrl-3, and Nba1) is a first step toward understanding the genetic basis of lupus in the SNF1 model. Although the above studies highlight the most potent dominant SWR loci contributing to SNF1 disease, the (NZB x SNF1) BC may not have been optimal in uncovering all potential loci. For instance, recessive SWR loci and loci that require certain other recessive loci for full expression of disease will not be mapped in this BC panel. Linkage analysis with further panels of (SWR x SNF1) and/or (SWR x NZB)F2 mice will be necessary to uncover these additional loci. Finally, to fathom the functions of these loci, it is instructive to introgress these individual loci onto a normal genetic background for detailed immunophenotyping analysis, as has been accomplished for the NZM2410-derived susceptibility loci. Undoubtedly, the ultimate challenge is to decipher the culprit genes within these loci. Charting out the myriad pathways and genes that can potentially pave the way to lupus, using the different, naturally occurring, spontaneous mouse models of lupus, stands to boost our understanding of this complex, systemic autoimmune disease.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Chandra Mohan, Simmons Arthritis Research Center, Department of Internal Medicine/Rheumatology, University of Texas Southwestern Medical School, Mail Code 8884, Y8.204, 5323 Harry Hines Boulevard, Dallas, TX 75390-8884. E-mail address: chandra.mohan{at}utsouthwestern.edu
3 Abbreviations used in this paper: Chr, chromosome; ANA, antinuclear autoantibody; BC, backcross; GN, glomerulonephritis; LOD, log likelihood of the odds.
Received for publication June 5, 2001. Accepted for publication October 3, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- and I-A -chain genes of normal SWR mice are linked with the development of lupus nephritis in NZB x SWR crosses. Proc. Natl. Acad. Sci. USA 84:6850.z/Ad major histocompatibility complex class II molecule: a restriction element for autoreactive T cells in (NZB x NZW)F1 mice. Immunology 95:577.[Medline]
Z/Ad class II molecule in (NZB x NZW)F1 mice detected by T cell clones. J. Immunol. 150:4777.[Abstract]
in murine autoimmune "lupus" nephritis. Nature 331:356.[Medline]
promoter identifies an allele unique to the NZW mouse strain. J. Exp. Med. 171:2141. alleles from normal and autoimmune-prone MHC haplotypes. J. Inflamm. 45:106.[Medline]
This article has been cited by other articles:
|
|
 |
|
  B R Lauwerys and E K Wakeland Genetics of lupus nephritis Lupus, January 1, 2005; 14(1): 2 - 12. [Abstract] [PDF]
|
|
|
|
 |
|
 J. J. Alexander, A. K. Saxena, L. Bao, A. Jacob, M. Haas, and R. J. Quigg Prominent Renal Expression of a Murine Leukemia Retrovirus in Experimental Systemic Lupus Erythematosus J. Am. Soc. Nephrol., December 1, 2002; 13(12): 2869 - 2877. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Xiu, K. Nakamura, M. Abe, N. Li, X. S. Wen, Y. Jiang, D. Zhang, H. Tsurui, S. Matsuoka, Y. Hamano, et al. Transcriptional Regulation of Fcgr2b Gene by Polymorphic Promoter Region and Its Contribution to Humoral Immune Responses J. Immunol., October 15, 2002; 169(8): 4340 - 4346. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), early mortality (dashed line), IgG
anti-dsDNA ANAs (thin solid line), and IgG anti-histone/DNA ANAs
(thick solid line) were performed using MapManager. QTX.LOD scores
depicted on the y-axis were calculated at 2-cM
intervals. The microsatellite marker loci used and their respective
chromosomal positions (in cM from acromere) are also depicted. Total
progeny examined = 86.