HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chamberlain, G. Articles by Green, E. A. Search for Related Content PubMed PubMed Citation Articles by Chamberlain, G. Articles by Green, E. A. Pubmed/NCBI databases Gene GEO Profiles HomoloGene UniGene Compound via MeSH Substance via MeSH Medline Plus Health Information Diabetes Type 1

A 20-Mb Region of Chromosome 4 Controls TNF--Mediated CD8⁺ T Cell Aggression Toward Cells in Type 1 Diabetes¹

Giselle Chamberlain^*,, Maja Wållberg^*,, Dan Rainbow^*,, Kara Hunter^*,, Linda S. Wicker^*, and E. Allison Green^2,*,

^* Cambridge Institute for Medical Research, Department of Pathology and Juvenile Diabetes Research Foundation/Wellcome Trust Diabetes and Inflammation Laboratory, and Department of Medical Genetics, University of Cambridge, Addenbrooke’s Hospital, Cambridge, United Kingdom

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Identification of candidate genes and their immunological mechanisms that control autoaggressive T cells in inflamed environments, may lead to novel therapies for autoimmune diseases, like type 1 diabetes (T1D). In this study, we used transgenic NOD mice that constitutively express TNF- in their islets from neonatal life (TNF--NOD) to identify protective alleles that control T1D in the presence of a proinflammatory environment. We show that TNF--mediated breakdown in T cell tolerance requires recessive NOD alleles. To identify some of these recessive alleles, we crossed TNF--NOD mice to diabetes-resistant congenic NOD mice having protective alleles at insulin-dependent diabetes (Idd) loci that control spontaneous T1D at either the preinsulitis (Idd3.Idd5) or postinsulitis (Idd9) phases. No protection from TNF--accelerated T1D was afforded by resistance alleles at Idd3.Idd5. Lack of protection was not at the level of T cell priming, the efficacy of islet-infiltrating APCs to present islet peptides, nor the ability of high levels of CD4⁺Foxp3⁺ T cells to accumulate in the islets. In contrast, protective alleles at Idd9 significantly increased the age at which TNF--NOD mice developed T1D. Disease delay was associated with a decreased ability of CD8⁺ T cells to respond to islet Ags presented by islet-infiltrating APCs. Finally, we demonstrate that the protective region on chromosome 4 that controls T1D in TNF--Idd9 mice is restricted to the Idd9.1 region. These data provide new evidence of the mechanisms by which selective genetic loci control autoimmune diseases in the presence of a strong inflammatory assault.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Type 1 diabetes (T1D)³ is an autoimmune disease where the insulin-secreting cells in the islets of Langerhans are destroyed by T cell-mediated mechanisms. Studies in the NOD mouse, a well-described mouse model for spontaneous T1D, have demonstrated that CD4⁺ and CD8⁺ T cell priming to islet Ags occurs in the pancreatic lymph node (PLN). Subsequently, primed T cells migrate to the islets, where cell destruction is thought to be primarily mediated by the CD8⁺ T cells (1).

Analyses of congenic strains of NOD mice, in which defined genetic loci in NOD mice are replaced by resistant alleles from mice not genetically predisposed to developing T1D, such as (C57BL/6 (B6) and C57BL/10 (B10)), have demonstrated that >20 loci are implicated in the progression to T1D (2, 3). The presence of a resistance allele at these insulin-dependent diabetes (Idd) loci has been shown to protect against T1D either alone or in combination with resistance alleles at other Idd loci. For example, resistant alleles at Idd1 and Idd16, both linked to the MHC (2, 3), or Idd9, which has been defined to have three Idd regions, Idd9.1, Idd9.2, and Idd9.3, provide a high degree of protection from T1D (4). The combination of Idd3 (candidate gene Il2 (5)) and Idd5 (candidate genes Ctla4 and Nramp1 for the Idd5.1 (6) and Idd5.2 regions (7), respectively) decrease T1D incidence to <2% as compared with 20% and 40% in mice having resistance alleles at only the Idd3 or Idd5 regions, respectively.

TNF- is thought to be the first proinflammatory cytokine produced by the earliest islet-infiltrating dendritic cells and is strongly linked to the development of T1D (8). Indeed, injection of neonatal NOD mice with TNF- results in rapid progression to T1D, whereas neutralization of TNF- protects (9). Previously, we have shown that transgenic expression of TNF- in neonatal NOD (TNF--NOD) mice increased both the penetrance and kinetics of T1D development compared with nontransgenic littermates (10). In part, this result was due to localized expression of TNF- in the islets enhancing intraislet dendritic cell presentation of islet peptides to CD8⁺ T cells and their subsequent differentiation into CTL (10, 11). In contrast to NOD mice, in neonatal mice with TNF-, CD4⁺ T cells did not have a role to play in T1D progression (11). This finding was recently supported by Rajagopalan et al. (12), who demonstrated that localized expression of TNF- in the islets of MHC class II-deficient mice derived from two successive intercrosses of B6.RIP-TNF and MHC class II-deficient NOD mice rapidly induced T1D development in 74% of the cohort. It is also important to note that the Idd1 regions present in the MHC class II-deficient and MHC class II-expressing mice studied by this group were derived from diabetes-resistant B6 and 129 mice, not from the NOD strain. Thus, the susceptible NOD MHC haplotype at Idd1 is not required for TNF--mediated acceleration of diabetes on the NOD background.

Interestingly, B6 mice that express the same TNF- transgene develop insulitis but not T1D (13). This suggests that B6 mice harbor protective alleles that prevent or delay T1D even in the presence of a strong proinflammatory assault and highlights the attractiveness of using TNF- in neonatal NOD mice for deciphering genes regulating T1D and the immunological mechanisms they control. In this study, using a series of congenic mice with protective alleles that control T1D at the preinsulitis or postinsulitis phases (Idd3.Idd5 and Idd9, respectively), we demonstrate that T1D development in TNF--NOD mice requires recessive NOD alleles. Further, we show that TNF- overrides protection afforded by the combination of T1D-resistance alleles at Idd3 and Idd5. This lack of protection by Idd3.Idd5 relates to TNF- altering the intraislet environment to increase CD8⁺ T cell numbers and decrease putative regulatory CD3⁺CD4^–CD8^– T cells but not CD4⁺ Foxp3⁺ T regulatory (Treg) cells. In contrast, protective alleles at Idd9, which control T1D at the postinsulitis phase, significantly delay T1D in TNF--NOD mice by decreasing the capacity of CD8⁺ T cells to respond efficiently to islet Ag. Finally, we show that the delay of disease mediated by the Idd9 region is attributable to a gene (or genes) present in the Idd9.1 region.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
Mice and genotyping

TNF--NOD mice (N22) has been previously described (10, 11). Only male mice heterozygous for the TNF- transgene were used for breeding to nontransgenic females due to the difficulty of maintaining a breeding colony in which female mice heterozygous for the TNF- transgene rapidly develop diabetes while still pregnant or when suckling. As a consequence, all studies both previously (10, 11) and presently analyze TNF--positive mice heterozygous for the transgene. TNF- transgenic mice were screened by PCR using forward primer 5'-ATTTGAGGGACGCTGTGGGCTCTT-3' and reverse primer 5'-CACCCCGAAGTTCAGTAGACAG-3'. PCR conditions were 94°C for 3 min, 94°C 1 min, 64.5°C 1 min, and 72°C 1 min.

NOD-Idd3.Idd5 (B6 at Idd3, B10 at Idd5.1 and Idd5.2) (14, 15) and NOD-Idd9 (B10 at Idd9.1, Idd9.2, and Idd9.3) (4) mice have been previously described (16) and are available from the Emerging Models Program (Taconic Farms) as lines 6109 and 905, respectively. The Idd9.3 congenic strain (line 1106; Taconic Farms) has been described (17). The Idd9.2/Idd9.3 congenic strain used in this study is unpublished and is defined by the markers D4Mit233 and D4Mit59 and its introgressed interval derived from the B10 strain is 9.8 Mb (see line 907 at http://t1dbase.org/cgi-bin/dispatcher.cgi/DrawStrains/display for further information). The Idd9.1 congenic strain (line 1565; Taconic Farms) is unpublished and is defined by the markers D4Mit200 and D4Mit69 and has a 19.9-Mb introgressed region from the B10 strain. B10.NOD H2^g7 mice (referred to as B10.H2g7 mice) have been published previously (18). Genotyping of F₂ mice used the following markers: D3Nds6 (Idd3); D1Mit303 (Idd5.1); D1Mit46 (Idd5.2); D7Nds6 (Idd7); D14Nds1 (Idd8); D4Mit233 (Idd9); AL607143(CA) (Idd9.3); D4Mit114 (Idd9.4); D3Mit12 (Idd10), and AC102578_3 (Idd18). PCR primers to identify the MIT markers are found at http://www.ensembl.org/Mus_musculus/index.html and the PCR primers to identify the Nuffield Department of Surgery microsatellite markers are found at http://www-gene.cimr.cam.ac.uk/todd/public_data/mouse/NDS/NDSMicrosTop.html. For AL607143(CA), detection was achieved using forward primer 5'-CCTTCGACCTTCACTCTGCT-3' and reverse primer 5'-TCATGCTCCTGTCCTCACAC-3'. For AC102578_3 detection, forward primer 5'-CAGCAGCCTCTCCAAAAAGT-3' and reverse primer 5'-GCTAGTTTTCAACTCATAAGATAGCC-3' were used. In both cases, PCR conditions were 94°C for 1 min, 55°C 1 min, and 72°C 1 min.

All mice were bred and maintained under specific pathogen-free conditions. All procedures were conducted under U.K. Government Home Office guidelines.

Cell preparations and islet Ag presentation assays

T cells were prepared from pooled PLN and peripheral lymph nodes (inguinal, axillary, and lateral axillary) by standard methods. For CD8⁺ T cell purification, single-cell suspensions were incubated with rat anti-CD4 (GK1.5), anti-B220 (RA3-6B2), and anti-DX5 Ab (all from BD Pharmingen). Following incubation with goat anti-rat BioMag beads (Qiagen) at 4°C for 20 min, CD8⁺ T cells were isolated by magnetic separation. Contaminating macrophages and dendritic cells were removed by incubating the suspension on plastic at 37°C for 30 min. For purification of CD4⁺ T cells, the rat anti-CD4 Ab was substituted with rat anti-CD8 Ab. In all cases, purity of the isolated fraction was determined to be >96%.

Islet-infiltrating APCs and islet Ag were prepared using collagenase digestion of pancreata as previously described (10). For islet Ag presentation assays, the islet Ag/APC suspension was irradiated before addition to the assay. Proliferation assays were conducted in triplicate and the assay was pulsed with 1 µCi [methyl-³H]thymidine (Amersham Biosciences) for the last 6 h of a 4-day culture. Cells were harvested and [methyl-³H]thymidine incorporation measured.

Flow cytometry

All Abs used for flow cytometry were purchased from BD Pharmingen unless otherwise stated. For lymph nodes and islet-infiltrating cells, single cell suspensions were prepared as mentioned. For cell surface staining, the cells were stained with fluorochrome-labeled anti-CD4 (GK1.5), anti-CD8 (53-6.7), anti-CD3 (2C11), and anti-CD45 (30-F11). For intracellular staining for Foxp3, a Foxp3 detection kit was used following the manufacturer’s directions (Insight Biotechnology). Cells were acquired using a FACSCaliber (BD Biosciences) and the data were analyzed using FlowJo software (Tree Star).

Diabetes detection

Diabetes was detected by measuring urine glucose levels using Keto-Diastix reagent strips (Bayer Diagnostics) and confirmed when serum glucose levels were >300 mg/dl on two consecutive occasions.

Statistics

Comparisons of disease frequency were performed using the log rank test. Values <0.01 were considered significant. For proliferation assays, statistical significance was measured using the Mann-Whitney nonparametric U test. Values of p < 0.05 were considered significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
TNF--accelerated diabetes is dependent on recessive NOD alleles

Localized expression of TNF- in the islets of B6 mice results in insulitis but not diabetes (13). In contrast, similar expression of TNF- in neonatal NOD (TNF--NOD) mice rapidly promotes disease progression compared with progression in nontransgenic littermates (10). Diabetes development in TNF--NOD mice heterozygous for the TNF- transgene occurs with equal penetrance and kinetics in both male and female mice (10). These findings suggest that NOD alleles are required for TNF- to disrupt peripheral tolerance to islet Ags. Initial investigations to identify the NOD alleles required for T1D in TNF--NOD mice examined the importance of the NOD MHC. TNF- B6 male mice (heterozygous for the TNF- transgene) were mated to B10.H2g7 females that have the NOD MHC but B10 alleles elsewhere. The 50% of F₁ male mice that had inherited the TNF- transgene were mated again to B10.H2g7 females, and TNF--positive progeny homozygous for the NOD MHC but harboring B10 or B6 alleles at non-MHC genes were identified and monitored for T1D. For ease of understanding, we called these TNF--positive progeny TNF--B10.H2g7.

TNF--B10.H2g7 mice developed extensive insulitis (data not shown) but did not progress to diabetes (Fig. 1a). Complete protection from T1D was also observed in (B10.H2g7 x TNF--NOD)F₁ progeny derived from breeding TNF--positive male TNF--NOD mice to female B10.H2g7 mice (Fig. 1b). Such TNF--positive F₁ mice were NOD MHC homozygous and NOD/B10 heterozygous at all non-MHC genes, suggesting that recessive non-MHC NOD alleles are required for T1D development. In all instances, control neonatal NOD mice with TNF- that had NOD alleles throughout the genome rapidly developed T1D between 9 and 17 wk of age as expected (10).

View larger version (16K): [in this window] [in a new window]   FIGURE 1. Recessive NOD alleles are a prerequisite for accelerated T1D in TNF--NOD mice. a, Male TNF- B6 mice were crossed to female B10.H2g7 mice. TNF--positive F1 males were crossed to female B10.H2g7 mice to create TNF- transgenic mice that were homozygous for NOD MHC genes but harbored B6 or B10 alleles at non-MHC genes (termed TNF--B10.H2g7 mice). Diabetes development in TNF--B10.H2g7 mice (n = 18) was compared with TNF--NOD mice (n = 21). Diabetes was detected when serum glucose levels were >300 mg/dl on two consecutive occasions. b, Male TNF- NOD mice were crossed to female B10.H2g7 mice and diabetes development in the TNF--positive (B10.H2g7 x TNF--NOD)F1 progeny (n = 17) was compared with that of TNF--NOD mice (n = 20) as before. c, Male TNF--positive progeny from the (B10.H2g7 x TNF--NOD)F1 cross were either backcrossed to female NOD mice or intercrossed with female nontransgenic littermates. Diabetes development in the TNF--positive ((TNF--NOD x B10.H2g7)F1 x NOD) backcross progeny (n = 30) and ((B10.H2g7 x NOD)F1 x (B10.H2g7 x TNF--NOD)F1) F2 cross progeny (n = 35) was monitored as before. In all experiments, TNF--positive mice analyzed were heterozygous for the transgene. The data are presented as a Kaplan Meier survival curve, and statistical comparisons of disease frequency were performed using the log rank test.

Segregation analysis of alleles that protect from TNF--mediated T1D

From these models, we hypothesized that TNF- is unable to disrupt peripheral tolerance to islet cells unless two doses of at least some non-MHC-linked NOD alleles are present in the mice expressing the TNF- transgene. We further hypothesized that one or more of the B10 or B6 alleles known to protect NOD mice from spontaneous T1D also protect NOD mice from TNF--mediated accelerated T1D. To test these hypotheses we generated two crosses to allow the putative protective alleles to segregate and NOD homozygosity at non-MHC loci to be present. First, we backcrossed TNF--positive (B10.H2g7 x TNF--NOD)F₁ male mice to NOD females. The resulting backcross progeny are all homozygous at the NOD MHC and genotypes at non MHC-linked loci are 50% NOD/NOD and 50% NOD/B10. Secondly, we mated TNF--positive (B10.H2g7 x TNF--NOD)F₁ males with nontransgenic (B10.H2g7 x TNF--NOD)F₁ female littermates. All of the resulting F₂ progeny were homozygous at the NOD MHC and would be expected to randomly segregate non-MHC loci at a ratio of 1:2:1 (NOD/NOD to NOD/B10 to B10/B10). In other words, 75% of the loci in the TNF--positive F₂ progeny have either one or two B10-derived alleles for each gene not linked to the MHC and 25% of the loci have two NOD alleles. The backcross and F₂ cohorts were genotyped for the TNF- transgene, and TNF--positive mice were monitored for the development of T1D.

As shown in Fig. 1c, all TNF--positive backcross mice rapidly progressed to diabetes, with all mice diabetic by 15 wk of age. In addition we observed no sex-bias in the kinetics of diabetes development. Thus, this cohort developed disease at an incidence and frequency identical with that of the TNF--NOD parental strain. In contrast, although diabetes was restored in 60% of TNF--positive F₂ mice, T1D progression was slower because only 30% of the F₂ mice developed T1D by 15 wk of age. Development of, or protection from, T1D in the TNF--positive F₂ cohort was not linked to the sex of the mouse. This pattern of inheritance in the backcross and F₂ generations is not consistent with a simple recessive gene model as the genetic basis of TNF--mediated accelerated T1D. For example, if two doses of one non-MHC NOD gene were needed to cause disease, only 50% of the backcross mice and 25% of the F₂ mice should have developed diabetes. The much higher observed frequency of disease in the backcross and F₂ generations, together with the fact that TNF--positive (B10.H2g7 x TNF--NOD)F₁ mice do not develop disease, strongly suggests that different combinations of genes can cause disease. Overall, our observations that 25% NOD homozygosity at non-MHC genes in the F₂ generation only allowed for rapid progression to T1D in 30% of the mice, as compared with 50% NOD homozygosity allowing all mice to develop disease rapidly, are consistent with a complex genetic control of TNF--mediated acceleration of T1D.

The cohort of 35 TNF--positive diabetic and nondiabetic F₂ progeny was genotyped with one genetic marker at each of the following regions for which B10 mice are known to have a protective allele for spontaneous T1D: Idd3, Idd10, Idd18, Idd5.1, Idd5.2, Idd9.1, Idd9.2, Idd9.3, Idd4, and Idd6 (2). No significant association was observed for any of the loci (data not shown). This result is consistent with the conclusions drawn from the inheritance pattern of the disease phenotype in the F₁, F₂, and backcross generations. Because multiple genes are apparently responsible for allowing TNF--mediated acceleration of T1D on the NOD background, the sample size of 35 F₂ mice had insufficient statistical power to detect an association.

TNF- overpowers the highly protective combination of protective alleles at Idd3 and Idd5

An alternative strategy to ascertain whether known Idd regions having protective B10- or B6-derived alleles can reduce TNF--mediated aggression is to express the TNF- transgene in congenic NOD mice having protective B10 (or B6) Idd alleles at one or more of the Idd regions. For example, very potent protection from spontaneous T1D as well as insulitis is mediated by the combination of two Idd regions, Idd3 and Idd5 (14, 16). Therefore, to test the effect of these two loci on TNF--accelerated disease on the NOD background, TNF--Idd3.Idd5 congenic mice were developed and diabetes progression monitored in comparison with neonatal NOD mice with TNF-.

In contrast to the nearly complete protection of NOD mice from T1D afforded by non-NOD resistance alleles at Idd3 and Idd5 (14) the localized islet-specific expression of TNF- resulted in rapid T1D irrespective of the sex of the mice (Fig. 2a). Further the kinetics and penetrance of disease were identical with that seen for TNF--NOD mice. Taken together, these data demonstrate that although protective B10 and B6 alleles can delay and reduce T1D progression in the presence of a strong inflammatory response (Fig. 1c), particular combinations of the protective alleles are required.

View larger version (26K): [in this window] [in a new window]   FIGURE 2. Protective alleles at both Idd3 and Idd5 are ineffective at controlling TNF- aggression in T1D. a, TNF--NOD mice were crossed to NOD-Idd3.Idd5 mice and male TNF--positive F1 progeny crossed to female NOD-Idd3.Idd5 mice, creating TNF--Idd3.Idd5 congenic mice. TNF--NOD (n = 24) and TNF--Idd3.Idd5 (n = 23) mice were monitored for diabetes as before. All TNF--positive mice analyzed were heterozygous for the transgene. CD4+ T cells (b) and CD8+ T cells (c) were purified from either TNF--Idd3.Idd5 mice () or their nontransgenic littermates (termed NOD-Idd3.Idd5 mice) () and their ability to respond to islet Ag presented by islet-infiltrating APCs from either TNF--Idd3.Idd5 or NOD-Idd3.Idd5 mice was determined. After a 4-day culture, cell proliferation was assessed by measuring the incorporation of [methyl-3H]thymidine following a 6-h pulse period on the last day of culture. Background counts for each islet preparation are shown (). Background counts for T cell responses alone were <200 cpm and have been subtracted from the data. The data are presented as the mean ± SD of triplicate cultures. Statistical significance in responses was determined using the Mann-Whitney U test. The result shown was reproducible in five independent experiments, and all TNF--positive mice analyzed were heterozygous for the transgene.

Recent reports have shown that when protective alleles at Idd3 and Idd5 are present together in NOD mice, accumulation of anti-islet CD4⁺ and CD8⁺ T cells in the PLN is prevented (19). In part, this result was thought to be due to the presence of CD4⁺CD25⁺ Treg cells preventing CD8⁺ T cell priming in the PLN (19). NOD-Idd3.Idd5 mice have substantially reduced insulitis, the number of islet-infiltrating cells recovered was 20% of the number recovered for NOD, TNF--NOD, and TNF--Idd3.Idd5 mice (data not shown). The number of islet-infiltrating cells recovered from NOD, TNF--NOD, and TNF--Idd3.Idd5 mice was equivalent (data not shown).

We tested whether the constitutive presence of TNF- in the islets of NOD-Idd3.Idd5 mice caused increased T cell priming to islet Ag thereby promoting T1D development.

CD4⁺ T cells were isolated from the pooled inguinal, axillary, and lateral axillary lymph nodes and PLN of TNF--Idd3.Idd5 or control nontransgenic littermates (referred to as NOD-Idd3.Idd5 mice). Purified T cells were cultured with islet-infiltrating APCs and islet Ag derived from either TNF--Idd3.Idd5 or NOD-Idd3.Idd5 mice. As shown in Fig. 2b, CD4⁺ T cells isolated from NOD-Idd3.Idd5 mice responded to islet Ag as efficiently as CD4⁺ T cells from TNF--Idd3.Idd5 mice, suggesting that priming and survival of islet-specific T cells in NOD-Idd3.Idd5 mice is not compromised. Further, we observed equal efficacy of islet-infiltrating APCs to present islet Ag to CD4⁺ T cells irrespective of the source of the stimulants.

A similar outcome was demonstrated when CD8⁺ T cells from TNF--Idd3.Idd5 or NOD-Idd3.Idd5 were used as responders in Ag presentation assays, where the stimulators were isolated from either group of mice (Fig. 2c).

Taken together, these data highlight two major points about the protective alleles in NOD-Idd3.Idd5 mice: 1) they do not prevent T cell priming to islet Ag and 2) they do not prevent the occurrence of functional APCs in the islets, at least as assessed ex vivo.

TNF- alters the intraislet cellular composition in disease-resistant NOD-Idd3.Idd5 and prediabetic NOD mice

T1D development is dependent on the infiltration of islets with T cells, particularly CD8⁺ T cells (20). NOD-Idd3.Idd5 mice have substantially reduced insulitis, whereas TNF--Idd3.Idd5 mice have extensive, destructive insulitis (data not shown). It was possible that protective alleles either prevented intraislet T cell entry, or altered the intraislet percentage of defined T cell subsets, in NOD-Idd3.Idd5 mice protecting against T1D. TNF-, therefore, may overcome this regulation step. To address this, we compared T cell percentages in islets of age-matched, sex-matched 8-wk-old neonatal mice with TNF- (TNF--NOD), their nontransgenic littermates (referred to as NOD mice), TNF--Idd3.Idd5 mice, and their nontransgenic littermates (referred to as NOD-Idd3.Idd5 mice). As a control, we included analysis of T cell populations in the PLN.

There were no differences in the percentages of either CD4⁺ or CD8⁺ T cells residing in the PLN of any of the groups of mice examined (Fig. 3, upper panels). As shown in Fig. 3 (lower panels) in the islets, TNF- slightly enhanced the percentage of CD4⁺ T cells (70% vs 52% for TNF--NOD mice and their nontransgenic NOD littermates, respectively, and 73% vs 60% for TNF--Idd3.Idd5 and their nontransgenic NOD-Idd3.Idd5 littermates, respectively). Similarly, the percentage of CD8⁺ T cells was also slightly enhanced in the islets of TNF--positive mice, at 25% vs 15% for TNF--NOD mice and their nontransgenic NOD littermates, respectively, and 19% vs 12% for TNF--Idd3.Idd5 mice and their nontransgenic NOD-Idd3.Idd5 littermates, respectively.

View larger version (40K): [in this window] [in a new window]   FIGURE 3. TNF- alters the islet T cell milieu in prediabetic NOD and diabetes-resistant NOD-Idd3.Idd5 mice. T cells were isolated from the PLN (upper panels) or pancreas (lower panels) of 8-wk-old female TNF--NOD, TNF--Idd3.Idd5 mice, and their nontransgenic littermates (termed NOD and NOD-Idd3.Idd5, respectively). The cell suspension was incubated with anti-CD45, CD3, CD4, and CD8 Abs. The percentage of CD4+, CD8+, and CD4–CD8– T cells in a CD45+CD3+ gate was determined by FACS. Representative dot plots are shown from a single mouse for each strain. The graphs depict the data obtained from groups of mice from each strain analyzed on the same day. The entire FACS experiment was repeated three times using three mice per strain on all occasions (a total of nine mice were examined) with identical results obtained each time. In all cases, the TNF--positive mice analyzed were heterozygous for the transgene.

To further evaluate the changes in the islet T cell composition, we analyzed the percentage of CD4⁺ Foxp3⁺ Treg cells between age-matched, sex-matched 8-wk-old TNF--NOD mice, their nontransgenic NOD littermates, TNF--Idd3.Idd5 mice, and their nontransgenic NOD-Idd3.Idd5 littermates. As a comparison, we measured CD4⁺ Foxp3⁺ T cells in the PLN or peripheral lymph nodes of the respective mice. In all strains examined, we detected no differences in the percentages of CD4⁺ Foxp3⁺ T cells in either the peripheral lymph nodes or PLN (Fig. 4). In the islets, the percentages of CD4⁺ Foxp3⁺ T cells were higher than the percentage in either the peripheral lymph nodes or PLN for all strains of mice. In the presence of TNF-, the percentage of intraislet CD4⁺ Foxp3⁺ T cells was negligibly higher suggesting that TNF- does not decrease this Treg cell population in the NOD mouse as previously suggested (22).

View larger version (44K): [in this window] [in a new window]   FIGURE 4. TNF- does not decrease the percentage of CD4+ Foxp3+ T cells in the PLN or islets. T cells were isolated from the peripheral inguinal lymph nodes, PLN, or pancreas of 8-wk-old female TNF--NOD, TNF--Idd3.Idd5 mice, and their nontransgenic littermates (termed NOD and NOD-Idd3.Idd5, respectively). The cell suspension was incubated with anti-CD45, CD4, Foxp3, or isotype control Abs and the percentage of CD4+ Foxp3+ cells in a CD45+CD3+ gate was determined by FACS. Representative histograms are shown for each strain. The graphs display the total number of animals analyzed for each strain on a single day. The experiment was repeated three times using three mice per mouse strain on all occasions (a total of nine mice was examined) with identical results obtained each time. All TNF--positive mice analyzed were heterozygous for the transgene.

Protective alleles at Idd9 regulate TNF--mediated T1D development

A second Idd region that efficiently protects NOD mice against T1D is Idd9. Several reports have suggested protective alleles present at Idd9 act at the level of the islet, not the priming phases in the PLN (16, 17). This latter point made Idd9 an interesting candidate region for controlling the local inflammatory events in the islets of TNF--NOD mice. Thus, to test the importance of protective alleles at Idd9, TNF--Idd9 congenic mice were developed and monitored for diabetes progression.

Despite the constitutive overexpression of TNF- in the islets, protective alleles at Idd9 significantly delayed diabetes progression (Fig. 5a) in comparison with TNF--NOD mice. This delay in diabetes development was equally apparent in both male and female TNF--Idd9 mice. Control, nontransgenic littermates of TNF--Idd9 mice, rarely progressed to diabetes as expected (4). These data demonstrate that TNF- increases T1D development even in the presence of resistant alleles at Idd9 but that the kinetics of T1D development is significantly reduced compared with the TNF--NOD strain.

View larger version (26K): [in this window] [in a new window]   FIGURE 5. Protective alleles at Idd9 delay TNF--mediated CD8+ T cell aggression against cells. a, Male TNF--NOD mice were crossed to NOD-Idd9 females and male TNF--positive F1 progeny backcrossed to NOD-Idd9 females. Male Idd9 homozygous mice expressing the TNF- transgene were identified from the backcross progeny to initiate the TNF--Idd9 congenic strain, which was maintained by continuously backcrossing male TNF--Idd9 mice to NOD-Idd9 females. Diabetes progression in TNF--Idd9 (n = 25) vs TNF--NOD (n = 26) mice was monitored as before. b, CD8+ T cells were purified from a pool of PLN and inguinal and axillary lymph nodes of either TNF--Idd9 () or TNF--NOD () mice. Purified CD8+ T cells were cultured with islet-residing APCs and islet Ag derived from either TNF--Idd9 or TNF--NOD mice (left) or anti-CD3 Abs (right). After a 4-day culture, cell proliferation was assessed by measuring the incorporation of [methyl-3H]thymidine, which was added for the last 6 h of the culture. Background counts for APCs/islet Ag alone are shown (). Background counts for T cells alone were <200 cpm and have been subtracted from the data. The data are presented as the mean ± SD of triplicate cultures. The statistical significance of differences between responses was analyzed using the Mann-Whitney U test. The results shown were reproducible in four independent experiments. All TNF--positive mice used were heterozygous for the transgene.

T1D development in neonatal NOD mice with TNF- is CD8⁺ T cell-dependent, CD4⁺ T cell-independent (11). We considered two possibilities as to why protective alleles at Idd9 significantly delayed diabetes progression even in the presence of a strong inflammatory cytokine like TNF-. First, that the intraislet cellular composition may be altered by the presence of the protective alleles resulting in decreased T cells, particularly CD8⁺ T cells. However, extensive FACS analysis of islet-infiltrating immune cells could not find any significant differences in T cells, B cells, dendritic cells, or macrophages between TNF--NOD mice and TNF--Idd9 mice (data not shown).

Second, we hypothesized that CD8⁺ T cell responses to islet Ags may be impaired in TNF--Idd9 mice with respect to TNF--NOD mice. To test this hypothesis, we isolated CD8⁺ T cells from TNF--Idd9 and control TNF--NOD mice and cultured them with islet-infiltrating APCs and islet Ag from TNF--NOD or TNF--Idd9 mice. As a positive control, we stimulated purified CD8⁺ T cells from each strain with anti-CD3 Abs.

As shown in Fig. 5b, CD8⁺ T cells from TNF--Idd9 mice demonstrated a significantly decreased response to islet Ag than CD8⁺ T cells isolated from control TNF--NOD mice irrespective of the source of the islet-infiltrating APCs. This decrease in response was not related to an intrinsic incapability of T cells from TNF--Idd9 mice to be stimulated, as mitogenic responses to anti-CD3 Abs were comparable for CD8⁺ T cells isolated from TNF--NOD and TNF--Idd9 mice.

As CD8⁺ T cells from TNF--NOD mice responded equally well to islet Ag presented by APCs from TNF--Idd9 or TNF--NOD mice, our data suggest that protective alleles at Idd9 function to decrease the responsiveness or the number of islet Ag-specific CD8⁺ T cells and not the efficacy of the islet-infiltrating APCs.

Protection mediated by Idd9 is localized to the Idd9.1 subregion

The Idd9 region is composed of three distinct subregions: Idd9.1, Idd9.2, and Idd9.3. Together, protective B10 alleles at the three subregions allow <10% of the congenic NOD mice to develop diabetes. B10 alleles at Idd9.3 alone reduce the rate of diabetes but not to the same extent as Idd9.2 and Idd9.3 together (4). We have recently observed that protective B10 alleles at Idd9.1 provide protection from disease at a level reduced from that observed when protective alleles are present at all three Idd9 subregions (L. S. Wicker and D. Rainbow, unpublished observations).

To determine whether a single Idd9 subregion makes a discernible contribution to the Idd9-mediated protection from TNF--accelerated T1D, the following strains were developed and T1D monitored: TNF--Idd9.3, TNF--Idd9.2/Idd9.3, and TNF--Idd9.1. As shown in Fig. 6, the replacement of B10 alleles at Idd9.3 by NOD alleles restored rapid T1D development in TNF--Idd9.3 mice (Fig. 6a). Thus, TNF- overcomes the partial protection from T1D in NOD mice caused by B10 alleles at Idd9.3 alone. TNF--Idd9.2/Idd9.3 mice also developed T1D as rapidly as neonatal NOD mice with TNF- (Fig. 6b), suggesting that the disease delay seen in TNF--Idd9 mice was either dependent on the presence of all three subregions or restricted to Idd9.1. This latter hypothesis was substantiated by the observation that TNF--Idd9.1 mice developed T1D with significantly delayed kinetics compared with TNF--NOD mice (Fig. 6c). Furthermore, comparison of the delayed disease kinetics observed in TNF--Idd9 (Fig. 5a) vs TNF--Idd9.1 (Fig. 6c) mice revealed no significant difference (p = 0.14), indicating that Idd9.1 accounts for all the Idd9-mediated delay from disease acceleration caused by TNF-.

View larger version (15K): [in this window] [in a new window]   FIGURE 6. The Idd9.1 region controls TNF--mediated T cell aggression toward cells. Male TNF--NOD mice were crossed to NOD-Idd9.3, NOD-Idd9.2/9.3, and NOD-Idd9.1 females. Male TNF--positive F1 progeny were backcrossed to females of the appropriate strains, and the TNF--Idd9.3, TNF--Idd9.2/Idd9.3, and TNF--Idd9.1 congenic strains were developed and maintained as described in Fig. 3. Diabetes development in TNF--Idd9.3 (a), TNF--Idd9.2/Idd9.3 (b), and TNF--Idd9.1 (c) mice was monitored as before. In all cases, TNF--positive mice analyzed were heterozygous for the transgene.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

Although the study by Rajagopalan et al. (12) was primarily focused on the effect of MHC class II expression on T1D mediated by TNF- expressed in the islets, our goal was to dissect the genetic control of protective alleles at non-MHC (non-Idd1) loci. A complex picture emerged from the surprising results obtained from the TNF--positive ((B10.H2g7 x TNF--NOD)F₁ x NOD) backcross progeny, which all rapidly developed diabetes (Fig. 1c). These results suggest that multiple loci are responsible for the complete protection from diabetes observed in TNF--positive (B10.H2g7 x TNF--NOD)F₁ mice. This hypothesis was supported by the fact that although the Idd9 region was later shown to confer protection from TNF--mediated T1D in the TNF--Idd9 strain, no linkage to Idd9 was observed when the F₂ progeny were genotyped at Idd loci, including Idd9, previously shown to provide protection from spontaneous T1D.

As an alternative approach to identify genetic loci that control T1D development in a proinflammatory environment, we crossed neonatal NOD mice with TNF- to two congenic strains having protective alleles at multiple Idd loci. We examined the Idd3.Idd5 and the Idd9 congenic strains as these loci have been shown to control spontaneous T1D before, or following, the development of insulitis, respectively. The Idd3.Idd5 congenic strain was of particular interest. Previous reports have demonstrated such mice have decreased B cell (16) and T cell responses (19) to islet Ag and have substantially reduced insulitis. Considering that rat insulin promoter TNF- B6 mice develop insulitis not T1D, TNF--Idd3.Idd5 congenic mice allowed us to test the ability of these protective alleles to control T1D following insulitis. We established that TNF--Idd3.Idd5 mice progressed to T1D with the incidence and kinetics akin to TNF--NOD mice demonstrating that a proinflammatory environment overrides protection afforded by the Idd3 and Idd5 loci (Fig. 2a). Originally, we speculated that TNF- overcame protection by enhancing presentation of islet Ag to autoreactive T cells. However, islet Ag presentation assays revealed a surprising result. We found equal efficacy of the CD4⁺ or CD8⁺ T cell response to islet Ag between T1D-susceptible TNF--Idd3.Idd5 mice and their T1D-protected nontransgenic NOD-Idd3.Idd5 littermates, irrespective of the source of the islet-infiltrating APCs/Ag (Fig. 2, b and c). These results suggest that priming and accumulation of islet-reactive T cells in the periphery is not impaired in nontransgenic NOD-Idd3.Idd5 mice. In addition, the islets of nontransgenic NOD-Idd3.Idd5 mice contain functional APCs. Based on these observations we analyzed immune cells in the PLN and islets of TNF--Idd3.Idd5 and TNF--NOD mice with their nontransgenic NOD-Idd3.Idd5 and NOD littermates, respectively. Flow cytometric analysis established that nontransgenic littermates of TNF--Idd3.Idd5 mice have islet-associated immune cells although the yield of cells was 80% lower than for TNF--Idd3.Idd5, TNF--NOD, or the nontransgenic littermates of TNF--NOD mice (data not shown). More significantly, the islet composition of immune cells was altered depending on the transgenic status of the mice. Sex-matched 8-wk-old NOD and NOD-Idd3.Idd5 mice had similar percentages of intraislet CD4⁺ T cells and CD8⁺ T cells. Introduction of the TNF- transgene slightly enhanced the percentage of both cell populations, with the increase most evident in the CD8⁺ T cells (Fig. 3). It is not clear from these data whether protective alleles at Idd3.Idd5 control not only the total degree of insulitis, but the percentages of CD4⁺ and CD8⁺ T cells present in the islets as well. Although the percentages of these T cells in young, nontransgenic NOD mice resembled those observed in age-matched nontransgenic NOD-Idd3.Idd5 mice, preliminary studies have shown that as diabetes-resistant, nontransgenic NOD-Idd3.Idd5 mice age, the percentage of CD8⁺ T cells in the islets does not increase as it does in disease-prone nontransgenic NOD mice (data not shown). Nevertheless, a more detailed study using a large cohort of mice is required to ascertain whether the protective effects of the Idd3.Idd5 genetic loci correlate with reduced percentages of intraislet CD8⁺ T cells.

Two candidate genes at the Idd3.Idd5 genetic loci, Il2 and Ctla4, respectively (5, 14), are linked to immune regulatory mechanisms, particularly the homeostasis and function of CD4⁺CD25⁺ Treg cells (23, 24). Thus, the Idd3.Idd5 genetic loci may prevent T1D by increasing CD4⁺CD25⁺ Treg cells in the PLN, stopping autoreactive T cell accumulation (19). In this model, using Foxp3 as a more stringent marker for Treg cells (25), we found that there was no change in the percentage of CD4⁺ Foxp3⁺ T cells in either the peripheral lymph nodes or PLN (Fig. 4) of any mouse strain, suggesting that in young NOD mice, the Idd3.Idd5 genetic loci do not influence Treg percentages. In the islets, not only was the overall percentage of CD4⁺ Foxp3⁺ T cells increased compared with the PLN and peripheral lymph nodes in all strains of mice, animals that expressed TNF- also showed a subtle increase. These findings are slightly at odds with published studies that suggest TNF- decreases the number of Treg cells in the NOD mouse (22). In part, this difference may be due to the use of CD25 instead of Foxp3 as a marker for CD4⁺ Treg cells in the previous publication. Nevertheless, it remains to be seen whether there is a relationship between localized intraislet TNF- and accelerated T1D at the level of the functionality of CD4⁺ Foxp3⁺ T cells, or the sensitivity of the intraislet CD8⁺ T cells to be regulated by them (26).

Interestingly, we documented a substantial decrease in the percentage of CD3⁺CD4^–CD8^– T cells in the islets of TNF--NOD and TNF--Idd3.Idd5 mice compared with their nontransgenic NOD and NOD-Idd3.Idd5 littermates, respectively (Fig 3). It is not clear what the significance is of this finding. Recent reports have documented a similar population of double negative CD3⁺ T cells in the mouse genital tract, and such cells have regulatory properties (27). Similarly, CD3⁺CD4^–CD8^– T cells with regulatory properties have been linked to allogenic skin (21, 28) and xenogenic heart (29) transplant tolerance as well as to the prevention of graft-vs-host disease (30). Future investigations will hopefully clarify whether CD3⁺CD4^–CD8^– T cells have a regulatory role to play in T1D. If such cells are shown to have a critical role in controlling T1D, the fate of CD3⁺CD4^–CD8^– T cells in a proinflammatory environment as well as in diabetes-resistant congenic NOD mice will be paramount to determine.

In contrast to the results obtained with protective alleles at Idd3 and Idd5, protective alleles at Idd9 significantly delayed T1D progression in TNF--Idd9 mice compared with TNF--NOD mice (Fig. 5). Flow cytometric analyses revealed this delay in T1D was not due to alterations in the intraislet B and T cell numbers between the two groups of mice (data not shown). To determine the cellular mechanisms responsible for the delay in T1D development in TNF--Idd9 mice, we examined the capacity of islet-residing APCs in TNF--Idd9 mice to present islet Ag and found no difference when compared with APCs from TNF--NOD mice. Instead our data suggest that CD8⁺ T cells from TNF--Idd9 congenic mice are less capable of responding to islet peptides. Whether this defect is due to dampened signals within the T cell or a decreased pool of islet-reactive T cells capable of responding to critical islet Ags remains to be verified. The Idd9 locus is composed of three subregions, Idd9.1, Idd9.2, and Idd9.3 (4). Recently, Watts and colleagues (17) refined the Idd9.3 subregion to a 1.2-Mb interval containing 15 genes, with Cd137 (41bb) the strongest candidate gene for mediating the 50% protection from T1D seen in NOD-Idd9.3 mice. However, we could not demonstrate a role for Idd9.3 in the control of T1D in the presence of a strong proinflammatory environment because the development of T1D in TNF--Idd9.3 mice was identical with that in the TNF--NOD strain. Instead, our observations support the hypothesis that a gene (or genes) present in the Idd9.1 subregion controls T1D in the TNF- model. Of interest, this locus contains the candidate gene lymphocyte-specific protein tyrosine kinase (Lck), a protein tyrosine kinase of the Src family that is involved in T cell signal transduction (31). Previous reports demonstrated that hyporesponsiveness of thymic CD4⁺ but not CD8⁺ T cells in NOD mice could be linked in part to sequestering of Lck in comparison with non-NOD strains of mice (32). Although functional differences in Lck could contribute to differential T cell activation in mice that harbor NOD vs non-NOD allelic variants of this gene, many other genes located in the 20-Mb Idd9.1 region are also candidates. Future investigations using novel congenic strains that will substantially reduce the size of the Idd9.1 subregion will greatly facilitate the identification of genes that control T1D even in the presence of a strong inflammatory assault.

In conclusion, we have provided new evidence on the genetic loci and immunological mechanisms by which a strong inflammatory assault promotes development of T1D. We have shown that islet-specific TNF- requires recessive NOD alleles to break T cell tolerance to islet Ags and can overcome protection seen in diabetes-resistant NOD-Idd3.Idd5 mice. Further, protective alleles at the Idd9.1 genetic locus significantly delayed T1D in the presence of a proinflammatory environment, by decreasing the number or the capacity of anti-islet CD8⁺ T cells to respond to their cognate Ag. Identification of the gene (or genes) that push the balance of the immune system from tolerance to autoimmunity during inflammation will be beneficial in the design of novel therapeutics for inflammatory-based disorders.

	Acknowledgments

 
We thank Gillian M. Brodie and Sarah Howlett for technical support and Philip Spence for critical reading of the manuscript.

	Disclosures

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 
The authors have no financial conflict of interest.

	Footnotes

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ This work was supported with grants from the Juvenile Diabetes Research Foundation and the Wellcome Trust (to E.A.G. and L.S.W.). The availability of NOD congenic mice through the Taconic Farms Emerging Models Program has been supported by grants from the Merck Genome Research Institute, National Institute of Allergy and Infectious Diseases, and the Juvenile Diabetes Research Foundation. E.A.G. is a Wellcome Trust Senior Research Fellow in Basic Biomedical Science.

² Address correspondence and reprint requests to Dr. E. Allison Green, Cambridge Institute for Medical Research, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 2XY, U.K. E-mail address: allison.green{at}cimr.cam.ac.uk

³ Abbreviations used in this paper: T1D, type 1 diabetes; PLN, pancreatic lymph node; Idd, insulin-dependent diabetes; Treg, T regulatory.

Received for publication March 28, 2006. Accepted for publication August 1, 2006.

	References

Top Abstract Introduction Materials and Methods Results Discussion Disclosures References

 

1. Delovitch, T. L., B. Singh. 1997. The nonobese diabetic mouse as a model of autoimmune diabetes: immune dysregulation gets the NOD. Immunity 7: 727-738. [Medline]
2. Maier, L. M., L. S. Wicker. 2005. Genetic susceptibility to type 1 diabetes. Curr. Opin. Immunol. 17: 601-608. [Medline]
3. Serreze, D. V., E. H. Leiter. 2001. Genes and cellular requirements for autoimmune diabetes susceptibility in nonobese diabetic mice. Curr. Dir. Autoimmun. 4: 31-67. [Medline]
4. Lyons, P. A., W. W. Hancock, P. Denny, C. J. Lord, N. J. Hill, N. Armitage, T. Siegmund, J. A. Todd, M. S. Phillips, J. F. Hess, et al 2000. The NOD Idd9 genetic interval influences the pathogenicity of insulitis and contains molecular variants of Cd30, Tnfr2, and Cd137. Immunity 13: 107-115. [Medline]
5. Lyons, P. A., N. Armitage, F. Argentina, P. Denny, N. J. Hill, C. J. Lord, M. B. Wilusz, L. B. Peterson, L. S. Wicker, J. A. Todd. 2000. Congenic mapping of the type 1 diabetes locus, Idd3, to a 780-kb region of mouse chromosome 3: identification of a candidate segment of ancestral DNA by haplotype mapping. Genome Res. 10: 446-453. [Abstract/Free Full Text]
6. Wicker, L. S., G. Chamberlain, K. Hunter, D. Rainbow, S. Howlett, P. Tiffen, J. Clark, A. Gonzalez-Munoz, A. M. Cumiskey, R. L. Rosa, et al 2004. Fine mapping, gene content, comparative sequencing, and expression analyses support Ctla4 and Nramp1 as candidates for Idd5.1 and Idd5.2 in the nonobese diabetic mouse. J. Immunol. 173: 164-173. [Abstract/Free Full Text]
7. Kissler, S., P. Stern, K. Takahashi, K. Hunter, L. B. Peterson, L. S. Wicker. 2006. In vivo RNA interference demonstrates a role for Nramp1 in modifying susceptibility to type 1 diabetes. Nat. Genet. 38: 479-483. [Medline]
8. Held, W., H. R. MacDonald, I. L. Weissman, M. W. Hess, C. Mueller. 1990. Genes encoding tumor necrosis factor and granzyme A are expressed during development of autoimmune diabetes. Proc. Natl. Acad. Sci. USA 87: 2239-2243. [Abstract/Free Full Text]
9. Yang, X. D., R. Tisch, S. M. Singer, Z. A. Cao, R. S. Liblau, R. D. Schreiber, H. O. McDevitt. 1994. Effect of tumor necrosis factor on insulin-dependent diabetes mellitus in NOD mice. I. The early development of autoimmunity and the diabetogenic process. J. Exp. Med. 180: 995-1004. [Abstract/Free Full Text]
10. Green, E. A., E. E. Eynon, R. A. Flavell. 1998. Local expression of TNF in neonatal NOD mice promotes diabetes by enhancing presentation of islet antigens. Immunity 9: 733-743. [Medline]
11. Green, E. A., F. S. Wong, K. Eshima, C. Mora, R. A. Flavell. 2000. Neonatal tumor necrosis factor promotes diabetes in nonobese diabetic mice by CD154-independent antigen presentation to CD8⁺ T cells. J. Exp. Med. 191: 225-238. [Abstract/Free Full Text]
12. Rajagopalan, G., Y. C. Kudva, R. A. Flavell, C. S. David. 2003. Accelerated diabetes in rat insulin promoter-tumor necrosis factor- transgenic nonobese diabetic mice lacking major histocompatibility class II molecules. Diabetes 52: 342-347. [Abstract/Free Full Text]
13. Picarella, D. E., A. Kratz, C. B. Li, N. H. Ruddle, R. A. Flavell. 1992. Insulitis in transgenic mice expressing tumor necrosis factor (lymphotoxin) in the pancreas. Proc. Natl. Acad. Sci. USA 89: 10036-10040. [Abstract/Free Full Text]
14. 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: 1744-1747. [Abstract]
15. Lord, C. J., S. K. Bohlander, E. A. Hopes, C. T. Montague, N. J. Hill, J. B. Prins, R. J. Renjilian, L. B. Peterson, L. S. Wicker, J. A. Todd, et al 1995. Mapping the diabetes polygene Idd3 on mouse chromosome 3 by use of novel congenic strains. Mamm. Genome 6: 563-570. [Medline]
16. Robles, D. T., G. S. Eisenbarth, N. J. Dailey, L. B. Peterson, L. S. Wicker. 2003. Insulin autoantibodies are associated with islet inflammation but not always related to diabetes progression in NOD congenic mice. Diabetes 52: 882-886. [Abstract/Free Full Text]
17. Cannons, J. L., G. Chamberlain, J. Howson, L. J. Smink, J. A. Todd, L. B. Peterson, L. S. Wicker, T. H. Watts. 2005. Genetic and functional association of the immune signaling molecule 4-1BB (CD137/TNFRSF9) with type 1 diabetes. J. Autoimmun. 25: 13-20. [Medline]
18. Lee, M. S., R. Mueller, L. S. Wicker, L. B. Peterson, N. Sarvetnick. 1996. IL-10 is necessary and sufficient for autoimmune diabetes in conjunction with NOD MHC homozygosity. J. Exp. Med. 183: 2663-2668. [Abstract/Free Full Text]
19. Martinez, X., H. T. Kreuwel, W. L. Redmond, R. Trenney, K. Hunter, H. Rosen, N. Sarvetnick, L. S. Wicker, L. A. Sherman. 2005. CD8⁺ T cell tolerance in nonobese diabetic mice is restored by insulin-dependent diabetes resistance alleles. J. Immunol. 175: 1677-1685. [Abstract/Free Full Text]
20. Mora, C., F. S. Wong, C. H. Chang, R. A. Flavell. 1999. Pancreatic infiltration but not diabetes occurs in the relative absence of MHC class II-restricted CD4 T cells: studies using NOD/CIITA-deficient mice. J. Immunol. 162: 4576-4588. [Abstract/Free Full Text]
21. Zhang, Z. X., L. Yang, K. J. Young, B. DuTemple, L. Zhang. 2000. Identification of a previously unknown antigen-specific regulatory T cell and its mechanism of suppression. Nat. Med. 6: 782-789. [Medline]
22. Wu, A. J., H. Hua, S. H. Munson, H. O. McDevitt. 2002. Tumor necrosis factor- regulation of CD4⁺CD25⁺ T cell levels in NOD mice. Proc. Natl. Acad. Sci. USA 99: 12287-12292. [Abstract/Free Full Text]
23. Read, S., V. Malmstrom, F. Powrie. 2000. Cytotoxic T lymphocyte-associated antigen 4 plays an essential role in the function of CD25⁺CD4⁺ regulatory cells that control intestinal inflammation. J. Exp. Med. 192: 295-302. [Abstract/Free Full Text]
24. Scheffold, A., J. Hühn, T. Höfer. 2005. Regulation of CD4⁺CD25⁺ regulatory T cell activity: it takes (IL-)two to tango. Eur. J. Immunol. 35: 1336-1341. [Medline]
25. Hori, S., T. Nomura, S. Sakaguchi. 2003. Control of regulatory T cell development by the transcription factor Foxp3. Science 299: 1057-1061. [Abstract/Free Full Text]
26. McGregor, C. M., S. P. Schoenberger, E. A. Green. 2004. CD154 is a negative regulator of autoaggressive CD8⁺ T cells in type 1 diabetes. Proc. Natl. Acad. Sci. USA 101: 9345-9350. [Abstract/Free Full Text]
27. Johansson, M., N. Lycke. 2003. A unique population of extrathymically derived TCR⁺CD4^–CD8^– T cells with regulatory functions dominates the mouse female genital tract. J. Immunol. 170: 1659-1666. [Abstract/Free Full Text]
28. Young, K. J., L. Yang, M. J. Phillips, L. Zhang. 2002. Donor-lymphocyte infusion induces transplantation tolerance by activating systemic and graft-infiltrating double-negative regulatory T cells. Blood 100: 3408-3414. [Abstract/Free Full Text]
29. Chen, W., M. S. Ford, K. J. Young, M. I. Cybulsky, L. Zhang. 2003. Role of double-negative regulatory T cells in long-term cardiac xenograft survival. J. Immunol. 170: 1846-1853. [Abstract/Free Full Text]
30. Young, K. J., B. DuTemple, M. J. Phillips, L. Zhang. 2003. Inhibition of graft-versus-host disease by double-negative regulatory T cells. J. Immunol. 171: 134-141. [Abstract/Free Full Text]
31. Alberola-Ila, J., S. Takaki, J. D. Kerner, R. M. Perlmutter. 1997. Differential signaling by lymphocyte antigen receptors. Annu. Rev. Immunol. 15: 125-154. [Medline]
32. Zhang, J., K. Salojin, T. L. Delovitch. 1998. Sequestration of CD4-associated Lck from the TCR complex may elicit T cell hyporesponsiveness in nonobese diabetic mice. J. Immunol. 160: 1148-1157. [Abstract/Free Full Text]

This article has been cited by other articles:

				G. M. Brodie, M. Wallberg, P. Santamaria, F. S. Wong, and E. A. Green B-Cells Promote Intra-Islet CD8+ Cytotoxic T-Cell Survival to Enhance Type 1 Diabetes Diabetes, April 1, 2008; 57(4): 909 - 917. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chamberlain, G. Articles by Green, E. A. Search for Related Content PubMed PubMed Citation Articles by Chamberlain, G. Articles by Green, E. A. Pubmed/NCBI databases Gene GEO Profiles HomoloGene UniGene Compound via MeSH Substance via MeSH Medline Plus Health Information Diabetes Type 1