The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Boulard, O.
Right arrow Articles by Garchon, H.-J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Boulard, O.
Right arrow Articles by Garchon, H.-J.
The Journal of Immunology, 2002, 168: 4192-4201.
Copyright © 2002 by The American Association of Immunologists

Genetic Analysis of Autoimmune Sialadenitis in Nonobese Diabetic Mice: A Major Susceptibility Region on Chromosome 11

Olivier Boulard*, Guy Fluteau*, Laure Eloy*, Diane Damotte{dagger}, Pierre Bedossa{ddagger} and Henri-Jean Garchon2,*

* Institut National de la Santé et de la Recherche Médicale, Unité 25, Hôpital Necker-Enfants Malades, Paris, France; {dagger} Service d’Anatomie Pathologique, Hôpital Européen Georges Pompidou, Paris, France; and {ddagger} Laboratoire de Pathologie, Hôpital Bicêtre, Université Paris-Sud, Le Kremlin-Bicêtre, France


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
The nonobese diabetic (NOD) mouse strain provides a good study model for Sjögren’s syndrome (SS). The genetic control of SS was investigated in this model using different matings, including a (NOD x C57BL/6 (B6))F2 cross, a (NOD x NZW)F2 cross, and ((NOD x B6) x NOD) backcross. Multiple and different loci were detected depending on parent strain combination and sex. Despite significant complexity, two main features were prominent. First, the middle region of chromosome 1 (chr.1) was detected in all crosses. Its effect was most visible in the (NOD x B6)F2 cross and dominated over that of other loci, including those mapping on chr.8, 9, 10, and 16; the effect of these minor loci was observed only in the absence of the NOD haplotype on chr.1. Most critically, the chr.1 region was sufficient to trigger an SS-like inflammatory infiltrate of salivary glands as shown by the study of a new C57BL/6 congenic strain carrying a restricted segment derived from NOD chr.1. Second, several chromosomal regions were previously associated with NOD autoimmune phenotypes, including Iddm (chr.1, 2, 3, 9, and 17, corresponding to Idd5, Idd13, Idd3, Idd2, and Idd1, respectively), accounting for the strong linkage previously reported between insulitis and sialitis, and autoantibody production (chr.10 and 16, corresponding to Bana2 and Bah2, respectively). Interestingly, only two loci were detected in the (NOD x NZW)F2 cross, on chr.1 in females and on chr.7 in males, probably because of the latent autoimmune predisposition of the NZW strain. Altogether these findings reflect the complexity and heterogeneity of human SS.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Sjögren’s syndrome (SS)3 is a chronic autoimmune disease characterized by lymphocytic infiltration and destruction of the salivary and lacrimal glands, causing oral and ocular dryness (1). It may develop primarily or be secondary and associated with other autoimmune manifestations, among which systemic lupus erythematosus and rheumatoid arthritis are the most frequent. The etiology of SS is presently unknown. It is likely to involve multiple factors, including immunologic, genetic, hormonal, and perhaps viral factors (1, 2).

When they exist, animal models are important for understanding a human disease. They are useful to clarify its causes, analyze its progression, and test novel treatments. Several mouse models for spontaneous SS are presently available, including the lupus-prone MRL-lpr and (NZB x NZW)F1 mice, and the NOD mouse strain (3, 4, 5, 6). The latter strain is known primarily as a reference for insulin-dependent diabetes mellitus (Iddm). However, NOD mice also develop inflammatory infiltration of exocrine glands, notably including salivary and lacrimal glands (5). This model of sialadenitis has now been well characterized, and two features make it valuable for the study of human SS. First, the incidence of sialitis is gender-biased and is higher in females than in males (Refs. 7, 8 , and present work), as it is in human SS patients. Second, the inflammatory infiltration of the salivary glands is associated with a loss of secretory function (6). This infiltrate consists predominantly of CD4+ T cells that produce proinflammatory cytokines (9, 10) and also of some B lymphocytes that synthesize autoantibodies against components of the salivary glands (10, 11), notably against muscarinic and {beta}-adrenergic receptors (12). Loss of secretory function could be mediated by these autoantibodies as shown by the study of NOD.IgMnull mice (13). Intrinsic anomalies of the salivary gland epithelium of NOD mice that do not depend on lymphoid cell infiltration have also been described previously (14, 15, 16, 17).

The molecular mechanisms that underlie sialadenitis in NOD mice are unknown, but must be determined at least in part by genetic factors, as these mice have been inbred. A susceptibility locus was mapped on chromosome 1 (chr.1) in an initial genetic study from our laboratory (18) and was validated by the analysis of congenic mice (19). At the time of our initial study, few polymorphic genetic markers were available, and coverage of the genome was therefore partial. The genetic control of NOD sialitis, however, could be polygenic, as is the case for most autoimmune diseases and as was recently shown for sialitis in MRL-lpr mice (20). In the present study, we have completed the genome scan that we had initiated in search of sialitis loci in two F2 intercrosses, including a (NOD x B6)F2 cross and a (NOD x NZW)F2 cross. We have also used information from a ((NOD x B6) x NOD) backcross. In addition, for data analysis we have considered sialitis as a quantitative trait.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Mice

NOD, C57BL/6 (B6), and NZW mice were bred and maintained under specific pathogen-free conditions in our facility at Institut National de la Santé et de la Recherche Médicale, Unité 25 (Hôpital Necker). We first generated (NOD x B6)F1 and (NOD x NZW)F1 hybrid mice and then the three crosses used in this study: (NOD x B6)F2, ((NOD x B6) x NOD)BC1, and (NOD x NZW)F2.

C57BL/6 mice congenic for the portion of chr.1 derived from the NOD strain comprised between D1Mit18 and D1Mit26 (both markers included), covering ~30 cM, were obtained after 20 generations of iterative backcrossing with the NOD parent, followed by appropriate sister-brother matings.

Histopathology

Mice were sacrificed at the age of 10 mo. However, in the experiment shown in Fig. 8Go, mice were sacrificed at the age of 5 mo. Submandibulary salivary glands were removed and immersed in Bouin’s fixative. Three serial sections (5-µm thick) were stained with H&E and examined for mononuclear cell infiltration. Inflammatory foci were defined as an accumulation of at least 10 mononuclear cells. This threshold allowed us to account for the full range of infiltration occurring in progeny mice and was therefore appropriate for a quantitative trait analysis. The distribution of foci within the glands was heterogeneous. Therefore, whole sections rather than a selected area unit were examined, and data were expressed as the mean number of foci per section. The approximate area of the sections was 63.0 ± 5.8 mm2 (mean ± SD of 10 glands).



View larger version (97K):
[in this window]
[in a new window]
  		FIGURE 8. Sialadenitis in C57BL/6 mice congenic for the NOD chr.1. Representative sections of submandibulary salivary glands from a NOD mouse (A), a C57BL/6 mouse congenic for the middle region of chr.1 of NOD origin (B), and a C57BL/6 mouse (C). Mice were sacrificed at the age of 20 wk. Sections were stained with H&E. Original magnification, x100.

 
Genotyping

Progeny mice were genotyped for polymorphic microsatellite markers identified in the Massachusetts Institute of Technology database (21), (http://www-genome.wi.mit.edu). For (NOD x NZW)F2 mice, markers had to be tested for polymorphism between NOD and NZW. However, alleles were not sized precisely, i.e., by sequencing gel electrophoresis. Qualitative information on these sizes compared among NZW, NOD, and B6 after agarose gel electrophoresis is available upon request. The following markers were used: D1Nds4, D1Mit18, D1Mit19, D1Mit46, D1Mit44, D1Mit8, D1Mit11, D1Nds2, D1Mit16, D1Mit36, D2Mit2, D2Mit32, D2Nds1, D2Mit58, D2Mit17, D3Nds1, D3Mit10, Adh1(D3Nds9), D3Mit19, D4Nds3, Orm1(D4Nds12), D4Mit15, D4Mit28, D4Mit13, D5Mit48, D5Mit72, D5Nds2, D5Mit30, D6Mit223, Tcrb, D6Nds1, D6Mit4, D6Mit9, D6Mit10, Prp(D6Mit13), D6Mit14, D6Mit15, D7Mit20, Ngfg(D7Nds5), D7Mit213, D7Mit53, D7Mit17, D7Mit242, D8Mit8, Mt2(D8Nds3), D8Mit40, D8Mit12, D8Mit88, D8Mit14, D9Mit42, Thy1(D9Nds3), Cyp1a2(D9Nds5), D9Nds1, D10Nds1, D10Mit2, D10Mit10, D10Mit14, Il4, D11Nds1, Odc(D12Nds11), D12Mit2, D12Mit7, D12Mit17, D13Mit16, D13Nds1, D13Mit9, D13Mit36, D14Mit11, D14Mit41, D15Mit13, D15Mit10, D15Nds1, D15Mit29, Ly6, Hoxc(D15Nds5), D16Mit55, D16Nds2, D16Mit178, D17Mit113, heat shock protein 70-1(D17Nds2), D17Mit36, D18Mit12, D18Mit17, D18Mit9, D18Mit4, D19Mit60, and D19Mit10.

The PCR was conducted following standard conditions as previously described (22), and PCR products were analyzed by electrophoresis on 5–6% agarose gels. At each marker, the NOD, B6, and NZW alleles were designated N, B, and W, respectively. The order of the markers and their distance to the centromere that were used in the text and tables were those indicated in the Mouse Genome Database (MGD) (23) (http://www.informatics.jax.org).

Statistical analysis and quantitative trait loci (QTL) mapping

The number of foci, or the sialitis score, was analyzed as a quantitative trait. The sample distribution was normal in ((NOD x B6) x NOD)BC1 female and (NOD x NZW)F2 female mice. Otherwise, it was normalized by logarithmic transformation.

The association of each polymorphic marker with sialitis was tested with the Mapmaker/QTL program (24). Genetic maps used for QTL analysis were determined directly from the dataset with the Mapmaker/EXP 3.0 program (24) and were integrated into Mapmaker/QTL. There was no major difference between our linkage map and the MGD map. The mode of inheritance of each QTL was determined with the "try" command of Mapmaker/QTL. Data were also analyzed with QTL Cartographer (25, 26). The ZMapqtl module of this software implements composite interval mapping and, under model 6, considers background loci that include unlinked loci and linked loci located beyond a predefined window (10 cM in our case). A permutation test with 10,000 permutations was also performed to assess the probability of the data for each locus comparisonwise. Nonparametrics tests (Kruskal-Wallis or Mann-Whitney U) were systematically conducted to confirm the association and mode of inheritance at each locus. Interactions between loci were tested by ANOVA, and post hoc comparisons of means were performed with the honestly significant difference (HSD) test (Tukey). Calculations were made using STATISTICA for Windows software (Statsoft, Tusla, OK).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Phenotypic analysis of parental and F1 mice

Mice with at least one inflammatory focus in their salivary glands were considered affected. Analysis of sialadenitis in NOD parents indicated complete penetrance at 10 mo of age (Fig. 1Go) in both males (13 of 13 mice affected; mean score ± SD, 5.3 ± 2.6) and females (15 of 15 mice affected; score, 15.6 ± 5.4). Conversely, the infiltrate was minimal, although variable, among C57BL/6 male (8 of 12 mice affected; score, 1.6 ± 1.8) and female (6 of 10 mice affected; score, 1.7 ± 2.3) parents and NZW male (7 of 9 mice affected; score, 0.3 ± 0.5) and female (9 of 9 mice affected; score, 2.2 ± 1.6) parents. The (NOD x B6)F1 hybrids had a phenotype similar to that of their B6 parents (2 of 5 mice affected; score, 0.6 ± 0.9 for males; 3 of 5 mice affected; score, 1 ± 1.2 for females). The (NOD x NZW)F1 hybrids had an intermediate phenotype (5 of 5 mice affected; score, 2.4 ± 1.5 for males; 5 of 6 mice affected; score, 9 ± 2.5 for females). In the F2 generations also, the penetrance of sialitis was higher in females than in males. In fact, it was almost complete in females from the ((NOD x B6)xNOD) backcross and from the (NOD x NZW)F2 intercross (Fig. 1Go). The sialitis scores were also higher in the progeny of the backcross on the NOD parent than in the (NOD x B6)F2 intercross. This was consistent with the low expression of the trait in the (NOD x B6)F1 hybrids and suggested a role for recessive genes. In the (NOD x NZW)F2 female mice, the sialitis scores were higher than in (NOD x B6)F2 female mice, but they did not reach those of their NOD parents. Therefore, the NZW genetic background does not synergize with the NOD background to increase the severity of sialitis as it does with the NZB background to determine severe lupus manifestations (27).



View larger version (20K):
[in this window]
[in a new window]
  		FIGURE 1. Distribution (mean and SD) of sialitis score in parental mice, their F1 hybrids, and their F2 progeny in females (A) and males (B). The number of affected mice (with at least one inflammatory focus) and that of total mice are given above the bars.

 
Analysis of (NOD x B6)F2 cross

A genome-wide scan was performed on the (NOD x B6)F2 progeny with a total of 156 microsatellites markers. Their average spacing was 10 cM, the largest gap being 34 cM. Parametric analysis of the data with Mapmaker/QTL disclosed six chromosomal regions associated with sialitis (Fig. 2Go). Of these, only chr.1 was associated at a significant level (logarithm of odds (LOD) score, 4.8 at D1Mit11) according to proposed criteria (28). This major region extended over 50 cM, from D1Mit300 (MGD position 32.8 cM) to D1Mit104 (MGD position 79 cM), with several peaks having a LOD score >3 (Fig. 2Go). The strongest association was with the D1Mit11 marker (MGD position 58.7 cM) localized 1.1 cM centromeric to the Bcl2 gene, which was previously associated with sialitis (18). The best model to fit the inheritance of this susceptibility locus was recessive. As shown in Table IGo, mice homozygous for the NOD allele at this marker had a mean score of infiltration increased by 2-fold compared with mice carrying one or two B6 alleles.



View larger version (30K):
[in this window]
[in a new window]
  		FIGURE 2. Mapmaker/QTL linkage analysis of sialitis on indicated chromosomes in (NOD x B6)F2 mice. Analysis was model-free (heavy line) or used a dominant (dotted line), an additive (dashed line), or a recessive (thin line) model. Dominance and recessitivity were defined by reference to the NOD allele. The straight dashed line indicated the threshold for suggestive evidence of linkage (2.0) (28 ). The scales under QTL maps show distances in centimorgans.

 

View this table:
[in this window]
[in a new window]
  		Table I. Markers associated with sialitis in the (NOD x B6)F2 cross

 
The other five loci mapped to chr.8, 9, 10, 16, and 17 (Fig. 2Go). Evidence for their association was suggestive and was supported by nonparametric tests and analysis with the QTL Cartographer program using composite interval mapping with permutation tests (Table IGo). The NOD alleles were associated with susceptibility to sialitis at loci on chr.8, 9, and 17, but interestingly were protective at loci on chr.10 and 16. Modeling with the "try" command of the Mapmaker/QTL program indicated different modes of action for the susceptibility loci: additive on chr.8, recessive on chr.9, and dominant on chr.17. Conversely, on both chr.10 and 16 the NOD allele was dominantly protective. These adjustments were confirmed by nonparametric tests, showing smaller p values in each instance, which suggested better models (Table IGo).

Considering the marked sex bias of sialitis in the NOD parent and the (NOD x B6)F2 cross, the QTL analysis was also made separately in male and female mice. The H2 complex on chr.17 was the only region that was strictly dependent on sex; it was associated with sialitis only in females (QTL-LOD score, 3.9; nonparametric p = 6 x 10-4). The best model was dominant susceptibility for the NOD allele at this locus (QTL-LOD score, 3.9; p = 1 x 10-4).

Analysis of the ((NOD x B6) x NOD) backcross

The aim of the study of the ((NOD x B6) x NOD) backcross was to examine the association of loci identified in the F2 cross and also of chromosomal regions previously associated with autoimmune diabetes in the NOD strain. Hence, only markers encompassing these regions were genotyped. The large size of the progeny, including 203 females and 247 males, allowed us to perform a sex-specific analysis.

In females, two loci, on chr.1 and 3, showed linkage to sialitis (Fig. 3Go and Table IIGo). On chr.1, the maximum LOD score was at the D1Mit5 marker (MGD position 32.8 cM), close to the peak mapped in F2 females. There seemed to be a second peak, more proximal, between markers D1Nds4 and D1Mit478 (Fig. 3Go), although more extensive typing of this region is needed to rule out a computational artifact. The shift toward the centromere in backcross mice compared with F2 mice might reflect the presence of a second locus that would be recessive and more efficiently detected by a backcross. Expression of these loci might be also differentially influenced by sex. Congenic mice should provide a suitable tool to dissect this region more finely. The locus on chr.3 was centered on the Il2 gene, which is a candidate for the Idd3 locus for autoimmune diabetes (29). For these two loci, the NOD allele conferred susceptibility to sialitis.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 3. Mapmaker/QTL linkage analysis of sialitis on chr.1 and 3 in ((NOD x B6) x NOD) females and on chr.2 in males. The straight dashed line indicates the threshold for suggestive evidence of linkage (2.0) (28 ). The scales under QTL maps show distances in centimorgans.

 

View this table:
[in this window]
[in a new window]
  		Table II. Markers associated with sialitis in the ((NOD x B6) x NOD) backcross

 
In males, only one locus, on chr.2, showed suggestive linkage (Fig. 3Go and Table IIGo) with a maximum between the D2Mit378 and D2Mit14 markers. However, there was no defined peak, and the association followed a plateau across a large portion of chr.2. Remarkably, the NOD allele conferred protection against sialitis.

No association was found with loci that were identified on chr.8, 9, 10, 16, and 17 in the F2 cross. For loci on chr.10, 16, and 17 this is consistent with their dominant mode of inheritance. For loci on chr.8 and 9 this may be explained by their pattern of interaction with the chr.1 locus, as detailed below.

Conversely, we had detected no association on chr.2 and 3 in the F2 cross despite a marked effect of these regions in the backcross progeny. Reconsidering the F2 cross data, we found no evidence for a role for chr.3 in F2 females by single-locus analysis. However, combined analysis of the effects of chr.1 and 3 by multiple regression disclosed an effect at the D3Nds1 marker in F2 females (p = 0.007). As for chr.2, the QTL-LOD score at the D2Mit58 marker was 1.98 (and nonparametric p = 0.015) and therefore close to the significance threshold. This marker is localized 5.4 cM from D2Mit62, where the association was still significant in the backcross (Fig. 3Go). As was the case in the backcross, the NOD allele was protective against sialitis.

Combined analysis of loci

In an attempt to clarify the relationships between the multiple loci identified above, we examined their combined effects. We focused our analysis on the major locus on chr.1 combined with the other minor loci. First, we tested the chr.1 region vs the 3 dominant loci localized on chromosomes 10, 16, and 17 (Fig. 4Go). As the best model for the chr.1 locus in the whole progeny was recessive, mice with NB and BB genotypes at the D1Mit11 marker were grouped. Likewise, the three minor loci followed a dominant model. Therefore, mice with NN and NB genotypes at the relevant markers were grouped. Both protective loci on chr.10 and 16 interacted with the chr.1 region; their protective effect was seen only in mice with the NB or BB genotypes at D1Mit11. The susceptibility determined by the chr.1 locus was thus dominant over the protection afforded by either locus on chr.10 or 16. Interestingly, the protective effects of these two loci were additive, as best shown by testing a liability model (Fig. 5Go). In this model, mice with BB genotypes at both D10Mit257 and D16Mit195 markers were assigned a value of 0. Mice with a NN or NB genotype at one locus, exclusively of the other locus, were assigned a value of 1. Mice with an NN or NB genotype at both D10Mit257 and D16Mit195 were assigned a value of 2. By linear regression, the p value for the model was 1 x 10-5 and became 2 x 10-10 when the genotype at D1Mit11 was also taken into consideration. In contrast, there was no interaction between chr.1 and 17 loci; their effects were simply additive (Fig. 4Go) even when females only were considered (data not shown).



View larger version (23K):
[in this window]
[in a new window]
  		FIGURE 4. Interactions between chr.1 and chr.10 (A), chr.16 (B), or chr.17 (C) in the (NOD x B6)F2 cross. Mice were grouped depending on their genotypes at D1Mit1 on chr.1 and at D10Mit257 on chr.10 (A), D16Mit195 on chr.16 (B), or heat shock protein 70 on chr.17 (C). The sialitis scores were normalized by logarithmic transformation. Means were calculated using all mice (affected and nonaffected) in each genotypic group. The number of affected mice over the total number of mice is indicated for each group. Comparison of the means was conducted with Tukey’s HSD test. Only significant p values are shown.

 


View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 5. Additive protection of chr.10 and 16 in the (NOD x B6)F2 cross. Mice were grouped depending on their genotypes at D10Mit257 on chr.10 and at D16Mit195 on chr.16. The sialitis scores were normalized by logarithmic transformation. Means were calculated using all mice (affected and nonaffected) in each genotypic group. The number of affected mice over the total number of mice is indicated for each group. Comparison of the means was conducted with Tukey’s HSD test. Only significant p values are shown.

 
Analysis of chr.1 vs the two remaining loci, on chr.8 and 9, disclosed a complex pattern of interaction. All combinations of possible genotypes were considered. As shown in Fig. 6Go, the predisposing effect of the NOD allele at chr.8 or 9 was significant only in mice with a BB genotype at D1Mit11. Therefore, although the chr.1 locus followed a recessive model of inheritance, there was a difference between the NB and BB genotypes regarding the effects of chr.8 or 9 loci. These findings, however, explain well the lack of effect of chr.8 and 9 in backcross mice. In contrast, there was no interaction between chr.1 and 3 loci in female ((NOD x B6) x NOD) backcross mice (Fig. 7Go). Their effects were additive.



View larger version (35K):
[in this window]
[in a new window]
  		FIGURE 6. Interactions between susceptibility loci on chr.1 and chr.8 or 9 in the (NOD x B6)F2 cross. Mice were grouped depending on their genotypes at D1Mit11 on chr.1 (indicated on the left) and at D8Mit190 on chr.8 (left panels) or at D9Nds1 on chr.9 (right panels). Sialitis scores were normalized by logarithmic transformation. Means were calculated using all mice (affected and nonaffected) in each genotypic group. The number of affected mice over the total number of mice is indicated for each group. Means were compared with Tukey’s HSD test. Only significant p values are shown.

 


View larger version (33K):
[in this window]
[in a new window]
  		FIGURE 7. Additive effect of chr.1 and 3 in ((NOD x B6) x NOD) BC1 females. Mice were grouped depending on their genotypes at D1Mit5 on chr.1 and at the Il2 gene on chr.3. Means were calculated using all mice (affected and nonaffected) in each genotypic group. The number of affected mice over the total number of mice is indicated for each group. Means were compared with Tukey’s HSD test.

 
Sialadenitis in C57BL/6 mice congenic for NOD chr.1

In a recent study, two congenic strains of C57BL/6 mice carrying large overlapping segments (44 and 47 cM, respectively) of chr.1 derived from the NOD strain (30) were found to display anomalies of their salivary glands, including inflammatory foci, increased cysteine protease activity, decreased amylase activity, and increased protein concentration in saliva (19). We derived a new C57BL/6 congenic strain carrying a restricted region from NOD chr.1. This region extended from D1Mit18 to D1Mit26, which are 32 cM apart, and corresponded to the overlap of the large segments harbored by the two congenic strains mentioned above. As shown in Fig. 8Go, at 20 wk of age these congenic mice displayed infiltration of their salivary glands by inflammatory foci of significant size (5 of 6 mice affected; mean ± SD number of foci, 3.6 ± 1.9), whereas six B6 mice of the same age showed no infiltration (p < 0.01 for comparison with congenics using Mann-Whitney U test). This observation indicated that the chr.1 region alone was sufficient to trigger an SS-like inflammatory infiltrate.

Analysis of the (NOD x NZW)F2 cross

For the (NOD x NZW)F2 intercross, a genome scan was performed with a more limited set of 87 markers that were polymorphic between the NOD and NZW strains (see detailed list in Materials and Methods). These markers were nevertheless evenly spaced, with a maximum gap of 40 cM. When considering the whole progeny (i.e., both females and males), suggestive evidence of linkage with sialitis was obtained for only two markers, both on chr.7 (Table IIIGo and Fig. 9Go). These two markers, D7Mit20 (centromeric) and D7Mit53 (telomeric), could correspond to two distinct loci, as they were separated by a marker, D7Nds5 (MGD position 23 cM, located in the Ngfg gene), showing no linkage with sialitis. At both markers the NOD allele conferred dominant susceptibility.


View this table:
[in this window]
[in a new window]
  		Table III. Markers associated with sialitis in the (NOD x NZW)F2 cross

 


View larger version (26K):
[in this window]
[in a new window]
  		FIGURE 9. Mapmaker/QTL linkage analysis of sialitis on chr.7 in all (NOD x NZW)F2 mice and in males only. Analysis was model-free (heavy line) or used a dominant (dotted line), an additive (dashed line), or a recessive (thin line) model. Dominance and recessitivity were defined by reference to the NOD allele. The straight dashed line indicated the threshold for suggestive evidence of linkage (2.0) (28 ). The scales under QTL maps show distances in centimorgans.

 
Given the sex bias of sialitis, data were reanalyzed according to sex. In males the same two markers on chr.7 were detected again and flanked the nonassociated D7Nds5 locus, but the distal marker (D7Mit53) was now more significantly associated and showed significant linkage (Table IIIGo and Fig. 9Go). It is therefore possible that, at least in males, a major and a minor locus coexisted on chr.7.

In females, only one locus on chr.1 showed suggestive linkage (Table IIIGo). The maximum LOD score was at the D1Mit8 marker (MGD position 52 cM), which is included in the region associated with sialitis in (NODxB6)F2 females. The best model to fit inheritance at this locus was recessive. Of note, in each instance the NOD allele conferred susceptibility, and no protective locus was detected in this (NOD x NZW)F2 cross. Interestingly, no association was found with the H2 region.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our study yields a complex picture of genetic control of the lymphocytic infiltration associated with SS in mice. Multiple and different loci were detected depending on sex, on the combination of strains used in the matings, and on the mating type itself. Moreover, although the NOD genome as a whole predisposes to SS, it does contain loci that are protective, at least in comparison with the C57BL/6 genome. Despite this complexity, the middle region of chr.1, which we had detected in our initial study (18), plays a major role in many respects. It was observed in all crosses studied to date, including two F2 intercrosses, a (NOD x B6)F2 cross and a (NOD x NZW)F2 cross, and two backcrosses, a ((NOD x B6) x NOD) backcross and a (MRL-lpr x (MRL-lpr x C3H-lpr)) backcross (20). Its effect was most significant in the (NOD x B6)F2 intercross, where a QTL-LOD score of 4.8 was reached. In this cross, it also was dominant over several other loci, including those mapping to chr.8, 9, 10 and 16, in that the effect of these minor loci was observed only in the absence of the NOD haplotype at the major locus on chr.1. Finally and most importantly, the study of C57BL/6 mice congenic for this region transferred from the NOD background has brought definitive confirmation of its implication in SS susceptibility (Ref. 19 and this study). The B6 mice congenic for chr.1 regions derived from the NOD strain that were studied by Brayer et al. (19) exhibited several features of autoimmune sialadenitis, including 1) morphological alterations with ductal hypertrophy, loss of acinar cell structures, and occasional lymphocytic infiltrates; and 2) biochemical modifications with increased protein concentration and decreased amylase activity of saliva, increased cysteine protease and metalloproteinase (MMP2 and MMP9) activities, and expression of the NOD-like 20-kDa isoform of parotid secretory protein in salivary gland extracts. The B6 congenic mice that we studied were derived independently and showed an overt mononuclear cell infiltration of their salivary glands. Taken together, these observations indicated that a single region on chr.1 was sufficient to trigger a spontaneous inflammatory infiltration of salivary glands.

Remarkably, this region of chr.1 was also associated with susceptibility to Iddm and is also known as the Idd5 locus (31). This was also the case for three other SS loci identified in this work. These loci map to regions of chr.3, 9, and 17, which correspond to Idd3, Idd2, and Idd1, respectively (29, 32, 33, 34). These shared regions must account for the strong linkage between SS and Iddm in NOD mice (18). Interestingly, the three loci that protected against SS were also involved in autoimmune phenotypes of NOD mice. The chr.2 region corresponds to Idd13 that predisposes to Iddm in NOD mice (35). Chr.10 and 16 correspond to Bana2 (bacillus Calmette-Guérin-induced anti-nuclear Abs locus 2) and Bah2 (bacillus Calmette-Guérin-induced autoimmune hemolytic anemia locus 2), respectively, that also both protect NOD mice against the expression of defined autoantibodies (36). Whether the genes that control these various phenotypes in these common regions are the same is now open to investigation.

Whereas the study of mice congenic for Idd3 and Idd5 has confirmed the role of these loci in predisposition to SS (19), analysis of NOD.B10-H2 congenics led to the conclusion that the MHC played no role in murine SS (37). This discrepancy with our present data could be explained by the redundancy of SS predisposition loci in the NOD genome. Likewise, NOD mice that were single congenics for Idd5 or Idd3 showed no alteration of their salivary gland infiltration and secretory function compared with their NOD parent, whereas in double congenics the autoimmune exocrinopathy was significantly amended, with an increase in amylase activity and flow rate of saliva, a decrease in cysteine protease and metalloproteinase activities, and a reduced number of inflammatory foci (19). Other Idd loci when studied in single congenics, including B6.NOD-Idd6, NOD.B6-Idd13 and NOD.B6-Idd9, did not modify the genetic background. Whether these loci, notably Idd13 in light of our results, would also interact with the chr.1 locus should be investigated. It remains that the H2-linked locus for murine SS does not have the critical importance it has in Iddm predisposition. It is nevertheless worthy of investigation because it may provide a model for understanding the contribution of HLA to predisposition to human SS (38, 39).

Few loci were identified in the (NOD x NZW)F2 cross. Additional loci could have been missed because of an incomplete genome coverage. Alternatively, few loci might be involved because the NZW strain also has an intrinsic predisposition to autoimmunity, notably including to Sle. At least three chromosomal regions predisposing to immune dysregulation are shared by NOD and NZW mice. Distal chr.1 is associated with defective expression of macrophage type II receptor for the Fc fragment of IgG (Fc{gamma}RII), elevated serum levels of IgG (40), and anti-nuclear Abs in NOD mice (Bana3) (36), whereas it controls breakdown of B cell tolerance in NZW mice and corresponds to sle1 (41, 42, 43). Chr.4 harbors the Idd9 locus in NOD mice (44, 45) and the sle2 locus that determines B cell hyperactivity in NZW mice (41, 46). Finally, the region of the H2 complex on chr.17 predisposes NOD mice to diabetes (32, 33, 34), sialitis (this work), and autoantibody expression (Bah1, Bana1) (36) and is also the sles1 locus that suppresses the latent lupus-like autoimmunity of the NZW strain (47). The marked penetrance of sialitis in (NOD x NZW)F1 hybrids, notably in females, is also consistent with an involvement of few loci.

Except for chr.1, different loci were detected in the NOD strain and in the MRL strain (20). This may be a result of the different combination of investigated strains, as MRL-lpr mice were mated with C3H-lpr mice. Yet this is also consistent with the difference in the SS phenotypes. No loss of secretory function of salivary glands was reported in MRL-lpr mice. Dendritic cells are also present in increased number at an early age in salivary glands of NOD mice compared with MRL-lpr mice (48). This diversity of phenotypes mimics the heterogeneity of SS in humans. Therefore, the comparative study of these two strains of mice and their genetic analysis using congenic lines should provide invaluable insight into the pathogenesis of SS.


   Acknowledgments
 
We are most grateful to Isabelle Cisse for careful mouse breeding.


   Footnotes
 
1 This work was supported by funds from Institut National de la Santé et de la Recherche Médicale. Back

2 Address correspondence and reprint requests to Dr. Henri-Jean Garchon, Institut National de la Santé et de la Recherche Médicale, Unité 25, Hôpital Necker, 161 rue de Sèvres, 75743 Paris Cedex 15, France. E-mail address: garchon{at}necker.fr Back

3 Abbreviations used in this paper: SS, Sjögren’s syndrome; chr., chromosome; HSD, honestly significant difference; Iddm, insulin-dependent diabetes mellitus; LOD, logarithm of odds; MGD, Mouse Genome Database; NOD, nonobese diabetic; QTL, quantitative trait locus; Sle, systemic lupus erythematosus. Back

Received for publication April 3, 2001. Accepted for publication February 5, 2002.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Fox, R. I.. 1996. Clinical features, pathogenesis, and treatment of Sjögren’s syndrome. Curr. Opin. Rheumatol. 8:438.[Medline]
  2. Fox, R. I.. 1997. Sjögren’s syndrome: controversies and progress. Clin. Lab. Med. 17:431.[Medline]
  3. Hoffman, R. W., M. A. Alspaugh, K. S. Waggie, J. B. Durham, S. E. Walker. 1984. Sjögren’s syndrome in MRL/l and MRL/n mice. Arthritis Rheum. 27:157.[Medline]
  4. Jabs, D. A., R. A. Prendergast. 1988. Murine models of Sjögren’s syndrome: immunohistologic analysis of different strains. Invest. Ophthalmol. Vis. Sci. 29:1437.[Abstract/Free Full Text]
  5. Goillot, E., M. Mutin, J. L. Touraine. 1991. Sialadenitis in nonobese diabetic mice: transfer into syngeneic healthy neonates by splenic T lymphocytes. Clin. Immunol. Immunopathol. 59:462.[Medline]
  6. Hu, Y., Y. Nakagawa, K. R. Purushotham, M. G. Humphreys-Beher. 1992. Functional changes in salivary glands of autoimmune disease-prone NOD mice. Am. J. Physiol. 263:E607.[Abstract/Free Full Text]
  7. Verheul, H. A., M. Verveld, S. Hoefakker, A. H. Schuurs. 1995. Effects of ethinylestradiol on the course of spontaneous autoimmune disease in NZB/W and NOD mice. Immunopharmacol. Immunotoxicol. 17:163.[Medline]
  8. Toda, I., B. D. Sullivan, E. M. Rocha, L. A. Da Silveira, L. A. Wickham, D. A. Sullivan. 1999. Impact of gender on exocrine gland inflammation in mouse models of Sjögren’s syndrome. Exp. Eye Res. 69:355.[Medline]
  9. Skarstein, K., M. Wahren, E. Zaura, M. Hattori, R. Jonsson. 1995. Characterization of T cell receptor repertoire and anti-Ro/SSA autoantibodies in relation to sialadenitis of NOD mice. Autoimmunity 22:9.[Medline]
  10. Robinson, C. P., J. Cornelius, D. E. Bounous, H. Yamamoto, M. G. Humphreys-Beher, A. B. Peck. 1998. Characterization of the changing lymphocyte populations and cytokine expression in the exocrine tissues of autoimmune NOD mice. Autoimmunity 27:29.[Medline]
  11. Esch, T. R., M. A. Taubman. 1998. Autoantibodies in salivary hypofunction in the NOD mouse. Ann. NY Acad. Sci. 842:221.[Free Full Text]
  12. Yamamoto, H., N. E. Sims, S. P. Macauley, K. H. Nguyen, Y. Nakagawa, M. G. Humphreys-Beher. 1996. Alterations in the secretory response of non-obese diabetic (NOD) mice to muscarinic receptor stimulation. Clin. Immunol. Immunopathol. 78:245.[Medline]
  13. Robinson, C. P., J. Brayer, S. Yamachika, T. R. Esch, A. B. Peck, C. A. Stewart, E. Peen, R. Jonsson, M. G. Humphreys-Beher. 1998. Transfer of human serum IgG to nonobese diabetic Igµ null mice reveals a role for autoantibodies in the loss of secretory function of exocrine tissues in Sjögren’s syndrome. Proc. Natl. Acad. Sci. USA 95:7538.[Abstract/Free Full Text]
  14. Robinson, C. P., H. Yamamoto, A. B. Peck, M. G. Humphreys-Beher. 1996. Genetically programmed development of salivary gland abnormalities in the NOD (nonobese diabetic)-scid mouse in the absence of detectable lymphocytic infiltration: a potential trigger for sialoadenitis of NOD mice. Clin. Immunol. Immunopathol. 79:50.[Medline]
  15. Robinson, C. P., S. Yamachika, C. E. Alford, C. Cooper, E. L. Pichardo, N. Shah, A. B. Peck, M. G. Humphreys-Beher. 1997. Elevated levels of cysteine protease activity in saliva and salivary glands of the nonobese diabetic (NOD) mouse model for Sjögren syndrome. Proc. Natl. Acad. Sci. USA 94:5767.[Abstract/Free Full Text]
  16. Kong, L., C. P. Robinson, A. B. Peck, N. Vela-Roch, K. M. Sakata, H. Dang, N. Talal, M. G. Humphreys-Beher. 1998. Inappropriate apoptosis of salivary and lacrimal gland epithelium of immunodeficient NOD-scid mice. Clin. Exp. Rheumatol. 16:675.[Medline]
  17. Yamachika, S., J. Brayer, G. E. Oxford, A. B. Peck, M. G. Humphreys-Beher. 2000. Aberrant proteolytic digestion of biglycan and decorin by saliva and exocrine gland lysates from the NOD mouse model for autoimmune exocrinopathy. Clin. Exp. Rheumatol. 18:233.[Medline]
  18. Garchon, H. J., P. Bedossa, L. Eloy, J. F. Bach. 1991. Identification and mapping to chromosome 1 of a susceptibility locus for periinsulitis in non-obese diabetic mice. Nature 353:260.[Medline]
  19. Brayer, J., J. Lowry, S. Cha, C. P. Robinson, S. Yamachika, A. B. Peck, M. G. Humphreys-Beher. 2000. Alleles from chromosomes 1 and 3 of NOD mice combine to influence Sjögren’s syndrome-like autoimmune exocrinopathy. J. Rheumatol. 27:1896.[Medline]
  20. Nishihara, M., M. Terada, J. Kamogawa, Y. Ohashi, S. Mori, S. Nakatsuru, Y. Nakamura, M. Nose. 1999. Genetic basis of autoimmune sialadenitis in MRL/lpr lupus-prone mice: additive and hierarchical properties of polygenic inheritance. Arthritis Rheum. 42:2616.[Medline]
  21. Dietrich, W. F., J. Miller, R. Steen, M. A. Merchant, D. Damron-Boles, Z. Husain, R. Dredge, M. J. Daly, K. A. Ingalls, T. J. O’Connor, et al 1996. A comprehensive genetic map of the mouse genome. Nature 380:149.[Medline]
  22. Garchon, H. J., J. J. Luan, L. Eloy, P. Bedossa, J. F. Bach. 1994. Genetic analysis of immune dysfunction in non-obese diabetic (NOD) mice: mapping of a susceptibility locus close to the Bcl-2 gene correlates with increased resistance of NOD T cells to apoptosis induction. Eur. J. Immunol. 24:380.[Medline]
  23. Blake, J. A., J. T. Eppig, J. E. Richardson, M. T. Davisson. 2000. The Mouse Genome Database (MGD): expanding genetic and genomic resources for the laboratory mouse: The Mouse Genome Database Group. Nucleic Acids Res. 28:108.[Abstract/Free Full Text]
  24. Lander, E. S., P. Green, J. Abrahamson, A. Barlow, M. J. Daly, S. E. Lincoln, L. Newburg. 1987. MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174.[Medline]
  25. Basten, C. J., B.S. Weir, Z.-B. Zeng. 1994. Zmap: a QTL cartographer. C. Smith, and J. S. Gavora, and B. Benkel, and J. Chesnais, and W. Fairfull, and J. P. Gibson, and B. W. Kennedy, and E. B. Burnside, eds. Proceedings of the 5th World Congress on Genetics Applied to Livestock Production: Computing Strategies and Software, Vol. 22 65. The Organizing Committee, 5th World Congress on Genetics Applied to Livestock Production, Guelph, Canada.
  26. Basten, C. J., B. S. Weir, Z.-B. Zeng. 2000. Q. T. L. Cartographer, version 1.14 Department of Statistics, North Carolina State University, Raleigh.
  27. Theofilopoulos, A. N., F. J. Dixon. 1985. Murine models of systemic lupus erythematosus. Adv. Immunol. 37:269.[Medline]
  28. Lander, E., L. Kruglyak. 1995. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat. Genet. 11:241.[Medline]
  29. Denny, P., C. J. Lord, N. J. Hill, J. V. Goy, E. R. Levy, P. L. Podolin, L. B. Peterson, L. S. Wicker, J. A. Todd, P. A. Lyons. 1997. Mapping of the IDDM locus Idd3 to a 0.35-cM interval containing the interleukin-2 gene. Diabetes 46:695.[Abstract]
  30. Yui, M. A., K. Muralidharan, B. Moreno-Altamirano, G. Perrin, K. Chestnut, E. K. Wakeland. 1996. Production of congenic mouse strains carrying NOD-derived diabetogenic genetic intervals: an approach for the genetic dissection of complex traits. Mamm. Genome 7:331.[Medline]
  31. Hill, N. J., P. A. Lyons, N. Armitage, J. A. Todd, L. S. Wicker, L. B. Peterson. 2000. NOD Idd5 locus controls insulitis and diabetes and overlaps the orthologous CTLA4/IDDM12 and NRAMP1 loci in humans. Diabetes 49:1744.[Abstract]
  32. Prochazka, M., E. H. Leiter, D. V. Serreze, D. L. Coleman. 1987. Three recessive loci required for insulin-dependent diabetes in nonobese diabetic mice. Science 237:286.[Abstract/Free Full Text]
  33. Prochazka, M., D. V. Serreze, S. M. Worthen, E. H. Leiter. 1989. Genetic control of diabetogenesis in NOD/Lt mice: development and analysis of congenic stocks. Diabetes 38:1446.[Abstract]
  34. Hattori, M., J. B. Buse, R. A. Jackson, L. Glimcher, M. E. Dorf, M. Minami, S. Makino, K. Moriwaki, H. Kuzuya, H. Imura, et al 1986. The NOD mouse: recessive diabetogenic gene in the major histocompatibility complex. Science 231:733.[Abstract/Free Full Text]
  35. Serreze, D. V., M. Prochazka, P. C. Reifsnyder, M. M. Bridgett, E. H. Leiter. 1994. Use of recombinant congenic and congenic strains of NOD mice to identify a new insulin-dependent diabetes resistance gene. J. Exp. Med. 180:1553.[Abstract/Free Full Text]
  36. Jordan, M. A., P. A. Silveira, D. P. Shepherd, C. Chu, S. J. Kinder, J. Chen, L. J. Palmisano, L. D. Poulton, A. G. Baxter. 2000. Linkage analysis of systemic lupus erythematosus induced in diabetes-prone nonobese diabetic mice by Mycobacterium bovis. J. Immunol. 165:1673.[Abstract/Free Full Text]
  37. Robinson, C. P., S. Yamachika, D. I. Bounous, J. Brayer, R. Jonsson, R. Holmdahl, A. B. Peck, M. G. Humphreys-Beher. 1998. A novel NOD-derived murine model of primary Sjögren’s syndrome. Arthritis Rheum. 41:150.[Medline]
  38. Jean, S., E. Quelvennec, M. Alizadeh, P. Guggenbuhl, B. Birebent, A. Perdriger, B. Grosbois, P. Y. Pawlotsky, G. Semana. 1998. DRB1*15 and DRB1*03 extended haplotype interaction in primary Sjögren’s syndrome genetic susceptibility. Clin. Exp. Rheumatol. 16:725.[Medline]
  39. Rischmueller, M., S. Lester, Z. Chen, G. Champion, R. Van Den Berg, R. Beer, T. Coates, J. McCluskey, T. Gordon. 1998. HLA class II phenotype controls diversification of the autoantibody response in primary Sjögren’s syndrome (pSS). Clin. Exp. Immunol. 111:365.[Medline]
  40. Luan, J. J., R. C. Monteiro, C. Sautes, G. Fluteau, L. Eloy, W. H. Fridman, J. F. Bach, H. J. Garchon. 1996. Defective Fc gamma RII gene expression in macrophages of NOD mice: genetic linkage with up-regulation of IgG1 and IgG2b in serum. J. Immunol. 157:4707.[Abstract]
  41. Morel, L., U. H. Rudofsky, J. A. Longmate, J. Schiffenbauer, E. K. Wakeland. 1994. Polygenic control of susceptibility to murine systemic lupus erythematosus. Immunity 1:219.[Medline]
  42. Morel, L., C. Mohan, Y. Yu, B. P. Croker, N. Tian, A. Deng, E. K. Wakeland. 1997. Functional dissection of systemic lupus erythematosus using congenic mouse strains. J. Immunol. 158:6019.[Abstract]
  43. Mohan, C., E. Alas, L. Morel, P. Yang, E. K. Wakeland. 1998. Genetic dissection of SLE pathogenesis. Sle1 on murine chromosome 1 leads to a selective loss of tolerance to H2A/H2B/DNA subnucleosomes. J. Clin. Invest. 101:1362.[Medline]
  44. Ghosh, S., S. M. Palmer, N. R. Rodrigues, H. J. Cordell, C. M. Hearne, R. J. Cornall, J. B. Prins, P. McShane, G. M. Lathrop, L. B. Peterson, et al 1993. Polygenic control of autoimmune diabetes in nonobese diabetic mice. Nat. Genet. 4:404.[Medline]
  45. Rodrigues, N. R., R. J. Cornall, P. Chandler, E. Simpson, L. S. Wicker, L. B. Peterson, J. A. Todd. 1994. Mapping of an insulin-dependent diabetes locus, Idd9, in NOD mice to chromosome 4. Mamm. Genome 5:167.[Medline]
  46. Mohan, C., L. Morel, P. Yang, E. K. Wakeland. 1997. Genetic dissection of systemic lupus erythematosus pathogenesis: Sle2 on murine chromosome 4 leads to B cell hyperactivity. J. Immunol. 159:454.[Abstract]
  47. Morel, L., X. H. Tian, B. P. Croker, E. K. Wakeland. 1999. Epistatic modifiers of autoimmunity in a murine model of lupus nephritis. Immunity 11:131.[Medline]
  48. van Blokland, S. C., C. G. van Helden-Meeuwsen, A. F. Wierenga-Wolf, H. A. Drexhage, H. Hooijkaas, J. P. van de Merwe, M. A. Versnel. 2000. Two different types of sialoadenitis in the NOD- and MRL/lpr mouse models for Sjögren’s syndrome: a differential role for dendritic cells in the initiation of sialoadenitis?. Lab. Invest. 80:575.[Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
M. A. Salam, K. Matin, N. Matsumoto, Y. Tsuha, N. Hanada, and H. Senpuku
E2f1 Mutation Induces Early Onset of Diabetes and Sjogren's Syndrome in Nonobese Diabetic Mice
J. Immunol., October 15, 2004; 173(8): 4908 - 4918.
[Abstract] [Full Text] [PDF]

Home page
 DiabetesHome page
S.-E. Lamhamedi-Cherradi, S. Zheng, R. M. Tisch, and Y. H. Chen
Critical Roles of Tumor Necrosis Factor-Related Apoptosis-Inducing Ligand in Type 1 Diabetes
Diabetes, September 1, 2003; 52(9): 2274 - 2278.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Boulard, O.
Right arrow Articles by Garchon, H.-J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Boulard, O.
Right arrow Articles by Garchon, H.-J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS