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 Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chiu, P. P. L. Articles by Danska, J. S. Search for Related Content PubMed PubMed Citation Articles by Chiu, P. P. L. Articles by Danska, J. S.

Genetic Control of T and B Lymphocyte Activation in Nonobese Diabetic Mice¹

Priscilla P. L. Chiu^*,,, Anthony M. Jevnikar^¶ and Jayne S. Danska^2,*,,

^* Program in Developmental Biology, Hospital for Sick Children Research Institute, Departments of Surgery and Immunology and Institute of Medical Science, University of Toronto, Toronto, Canada; and ^¶ Departments of Nephrology and Transplantation and Microbiology and Immunology, University of Western Ontario, London, Canada

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Type 1 diabetes in nonobese diabetic (NOD) mice is characterized by the infiltration of T and B cells into pancreatic islets. T cells bearing the TCR V3 chain are disproportionately represented in the earliest stages of islet infiltration (insulitis) despite clonal deletion of most V3⁺ immature thymocytes by the mammary tumor virus-3 (Mtv-3) superantigen (SAg). In this report we showed that a high frequency of NOD V3⁺ T cells that escape deletion are activated in vivo and that this phenotype is linked to the Mtv-3 locus. One potential mechanism of SAg presentation to peripheral T cells is by activated B cells. Consistent with this idea, we found that NOD mice harbor a significantly higher frequency of activated B cells than nondiabetes-prone strains. These activated NOD B cells expressed cell surface molecules consistent with APC function. At the molecular level, the IgH repertoire of activated B cells in NOD mice was equivalent to resting B cells, suggesting a polyclonal response in vivo. Genetic analysis of the activated B cell phenotype showed linkage to Idd1, the NOD MHC haplotype (H-2^g7). Finally, V3⁺ thymocyte deletion and peripheral T cell activation did not require B cells, suggesting that other APC populations are sufficient to generate both Mtv-3-linked phenotypes. These data provide insight into the genetic regulation of NOD autoreactive lymphocyte activation that may contribute to failure of peripheral tolerance and the pathogenesis of type I diabetes.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Elevated frequencies of activated lymphocytes are common features of autoimmune disorders (1, 2, 3, 4, 5, 6, 7, 8, 9). Activated T cells bearing autoreactive TCRs (3) infiltrate tissues that express the self-Ags, recruit inflammatory cells, and provide costimulatory signals that activate B cells (3, 8, 10, 11). Once activated, B cells become competent APCs (12), potentiating autoreactive T cell recognition (3, 6, 13, 14). Hence, T and B cell activation fuels an inflammatory cascade leading to organ destruction. Efforts to identify genes that control autoimmune disease susceptibility have suggested that some of these genetic loci influence lymphocyte activation and homeostasis. For example, in the NZB x NZW mouse model of systemic lupus erythematosus, the sle1 locus has been shown to mediate loss of B cell tolerance and activation of autoreactive T cells (15). The sle2 locus was shown to regulate B cell hyperactivity and autoantibody production (16). In the MRL-lpr mouse model, lpr was identified as a mutation in the Fas gene resulting in persistent activation of autoreactive lymphocytes contributing to autoimmune nephritis (17). Although the genes at some of these loci have yet to be identified, these studies demonstrate that genetic regulation of lymphocyte activation, proliferation, and homeostasis plays a central role in autoimmune pathogenesis.

Insulin-dependent or type I diabetes mellitus (T1D)³ is an autoimmune disease characterized by the destruction of insulin-producing pancreatic islet cells (18). Accumulating evidence suggests that dysregulated lymphocyte activation and homeostasis play a role in diabetes susceptibility. Type I diabetes patients (4, 7) and their nondiabetic monozygotic twins display increased frequencies of activated circulating T cells compared with controls (19, 20). In the biobreeding rat model of T1D, variation at the lyp locus causes T cell lymphopenia that is required for diabetes susceptibility in this strain (21). In the nonobese diabetic (NOD) mouse model (22), two insulin-dependent diabetes (Idd) loci have been shown to influence the apoptotic response of lymphocytes to multiple cytotoxic agents (23, 24). In addition, robust proliferation of autoreactive T cells was observed after immunization of NOD mice with autoantigens (25, 26). Collectively, these results suggest that dysregulated T cell activation and death may contribute to susceptibility to and development of T1D.

Previously, we reported that NOD T cells that express the TCR V3⁺ gene segment are highly represented in early islet infiltrates and that these TCR -chains display restricted CDR3 region diversity, suggesting reactivity for relatively few Ags (27). NOD mice harbor an endogenous murine mammary tumor virus (Mtv)-3-encoded superantigen (SAg) that deletes the vast majority of V3⁺ thymocytes (28). Thus, these results suggested that rare NOD T cells that escape deletion may become activated and contribute to the early stages of autoimmune islet infiltration. Alternatively, V3⁺ T cells that escape deletion could be rendered unresponsive in peripheral tissues, as suggested by a transgenic model involving an Mtv SAg-responsive TCR (29). For these reasons it was of interest to determine whether NOD V3⁺ T cells that escape deletion are responsive to signals through their TCR, and whether they displayed evidence of activation in vivo. In this report we show that a high frequency of NOD V3⁺ T cells are activated in vivo and that this property is genetically linked to the endogenous SAg, Mtv-3, on chromosome 11.

T1D in NOD mice is primarily mediated by T cells (reviewed in Ref. 30). However, recent reports suggest that NOD B cells also play an important role (31, 32, 33, 34), but the mechanism of this effect is unclear. One possibility is that NOD B cells serve to capture and present critical islet Ags to autoreactive T cells (35, 36, 37). Earlier work suggested that B cells were most capable of presenting Mtv SAg to mature T cells (38), although CD8⁺ T cells and dendritic cells can perform this function under certain conditions (39, 40). B cell competence for Ag presentation requires an activation-dependent increase in cell surface expression of costimulatory molecules (41). To address the potential role of B cells as APC in the NOD model, we have evaluated their activation status, cell surface phenotype, and Ig repertoire diversity in vivo. We report that the frequency of activated B cells in NOD mice is significantly higher than that observed in other strains, and that the phenotype of these B cells is consistent with their function as APC. In addition, we provide evidence for genetic linkage of B cell activation to or Idd1, the NOD MHC, particularly to I-A^g7. Interestingly, neither thymic deletion nor peripheral activation of NOD V3⁺ T cells required B cells, suggesting the sufficiency of other APC types in regulating these potentially autoreactive T cells. Heightened levels of T and B lymphocyte activation behave as genetically regulated events in NOD mice that are evident long before diabetes onset. These observations support a central role of dysregulated lymphocyte homeostasis in preclinical stages of diabetes.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Mice

All mice used in these studies were maintained in a pathogen-free facility in which the incidence of diabetes in the NOD colony was 85% for females and 15% for males at 30 wk of age. Female NOD.H-2^q SWR and NOD.H-2^b C57BL/10 (NOD.B10) mice were purchased from The Jackson Laboratory (Bar Harbor, ME). B cell-deficient NOD.µMT mice were obtained from Dr. D. V. Serreze (The Jackson Laboratory) (31). MHC class II-deficient NOD.I-A^-/- mice were obtained from Dr. A. Jevnikar, University of Western Ontario (London, Canada) (42). The mice used in these studies were 2–7 wk old.

Abs and reagents

The Abs used in flow cytometric analyses, magnetic bead depletions, and in vitro lymphocyte activation assays were affinity-purified from tissue culture supernatants: anti-TCR C (H57.597), anti-CD3 (145-2C11), anti-I-A^g7 (10-2.16), anti-B220 (RA3-6B2), anti-Mac-1 (M1/70); and anti-TCR V2 (B20.6), anti-TCR V3 (KJ25), anti-TCR V4 (KT4), anti-TCR V5 (MR9-4), anti-TCR V6 (44-22), anti-TCR V7 (TR310), anti-TCR V8 (F23.1), anti-TCR V9 (MR10-2), anti-TCR V11 (RR3-15), anti-TCR V13 (MR12-4), and anti-TCR V14 (14.2) determinants. Biotinylated anti-K^b (AF6-88.5), FITC-conjugated anti-CD25 (7D4), biotinylated and FITC-conjugated anti-CD69 (H1.2F3), biotinylated anti-IgM (R40-97), biotinylated anti-CD19 (1D3), FITC-conjugated anti-B7-1 (16-10A1), FITC-conjugated anti-B7-2 (GL1), FITC-conjugated anti-ICAM-1 (3E2), allophycocyanin-conjugated RA3-6B2, and FITC-conjugated and biotinylated isotype control Ab were purchased from BD PharMingen (San Diego, CA). Avidin-PE (AV-PE) was purchased from Caltag Laboratories (South San Francisco, CA). Propidium iodide for dead cell exclusion was purchased from Sigma-Aldrich (St. Louis, MO).

Generation and screening of (NOD x B6)F₂ intercross mice

NOD mice have an MHC haplotype designated H-2^g7 (K^d, I-A^d, I-A^g7, I-E^-/-, D^b, L^b). NOD mice were crossed to C57BL/6 (H-2^b, I-E^-/-, Mtv-3^-/-) mice. (NOD x B6)F₂ progeny were screened to determine their MHC and Mtv-3 genotypes. Screening for Mtv-3 was performed by Southern blot as previously described (28) and was confirmed by flow cytometric analysis of PBL using anti-TCR V3 (KJ25) and a pan anti-TCR (H57.597) mAb. Mtv SAg-mediated deletion of T cells behaves as an autosomal dominant phenotype. A PBL KJ25:H57.597 ratio <1%, as seen in NOD mice, was invariably correlated with the presence of the Mtv-3 provirus by Southern blot. The absence of Mtv-3 was indicated by a PBL KJ25:H57.597 ratio >3%, as seen in B6 mice. The MHC haplotypes of (NOD x B6)F₂ mice were determined by flow cytometric analyses of PBL using mAb 10-2.16 (recognizes I-A^g7) and AF6-88.5 (recognizes H-2K^b). H-2^g7 and H-2^b homozygotes with and without the Mtv-3 provirus were identified for analysis.

Flow cytometry

Preparation and staining of lymph node (LN) cell suspensions were performed as previously described (43) using the Ab listed above. Two- and four-color FACS were performed on FACScan and FACSCalibur, respectively (BD Biosciences, Mountain View, CA). For small populations, 1 x 10³–1 x 10⁴ V3⁺ and V7⁺ T cells were counted per sample. Analysis of FACS data was performed using LYSIS II and CellQuest software (BD Biosciences).

Magnetic bead depletion for enrichment of V3⁺ cells in proliferation assays

Enrichment of BALB/c and NOD V3⁺ T cells was achieved by immunodepletion of LN cell suspensions using sheep anti-FITC Ab-coated magnetic beads (BioMag separator; Advanced Magnetics, Cambridge, MA) as directed. LN cells were stained with FITC-conjugated Ab directed against B220 and Mac-1 to deplete B cells and APCs, and anti-TCR V2, -4, -5, -6, -7, -8, -9, -11, -13, and -14 to enrich for V3⁺ T cells. B6 LN cells were depleted of non-T cells by staining with FITC-conjugated anti-B220 and anti-Mac-1 Ab. Immunomagnetic depletion efficiency was confirmed by flow cytometry.

Lymphocyte proliferation assays

Proliferative assays of NOD, BALB/c, and B6 V3⁺ T cells was performed using 96-well flat-bottom microtiter plates coated with KJ25 or isotype control hamster IgG Ab at 10 µg/ml as previously described (44). To assay equivalent numbers of V3⁺ T cells, 3 x 10²–1 x 10⁵ B6 B cell-depleted LN cells and 1 x 10³–3 x 10⁵ BALB/c or NOD V3-enriched LN cells were incubated in triplicate in Ab-coated wells. All samples were incubated with 2 x 10⁵ mitomycin C-treated syngeneic splenocytes at 37°C and 5% CO₂ in 200 µl of RPMI 1640 medium supplemented with 10% FBS, 5 x 10^-3 M 2-ME, 10 mM L-glutamine, 10 mM HEPES, 10 U/ml rIL-2, and 5% penicillin/streptomycin. IL-2 was refreshed every 48 h. Stimulation with 3 µg/ml Con A (Boehringer Diagnostics, La Jolla, CA) was used as a proliferation control. After 72 h the cells were pulsed with 1 µCi of [³H]thymidine (Amersham, Oakville, Canada) for 6 h, and [³H]thymidine incorporation was measured. The mean counts per minute and SD were calculated for each triplicate (mean ± SD). Change in counts per minute was calculated by the difference between the mean counts per minute for the KJ25-treated samples and the mean counts per minute for the isotype control samples.

Cytokine assays

Assays for cytokine production by NOD and BALB/c LN cells were performed using APC-depleted LN cells enriched for V3⁺ T cells. V3-enriched LN cells were incubated in 24-well tissue culture plates coated with KJ25 or 145-2C11 Ab (10 µg/ml) in 1.5 ml of supplemented RPMI 1640 as described above. IL-2 was refreshed after 48 h. Supernatants were removed after 72 h of culture and measured for IL-4 and IFN- in triplicate by ELISA using the DuoSet ELISA development system (R&D Systems, Minneapolis, MN) as directed. Recombinant mouse IL-4 and IFN- supplied by the manufacturer were used to generate the standard curves for each assay. ELISA data were collected and analyzed using SOFTmax Pro software (version 2.1; Molecular Devices, Sunnyvale, CA).

RNA preparation and cDNA synthesis

LN cells were isolated from eight individual (4- to 6-wk-old) NOD mice. CD69⁺ and CD69^- LN cells were separated by magnetic immunodepletion using FITC-conjugated H1.2F3 (anti-CD69) mAb as described above and confirmed by flow cytometric analysis. RNA was prepared from 1 x 10⁶ NOD spleen cells and from 1 x 10⁶ NOD CD69⁺ and CD69^- LN cells using TRIzol as directed (Life Technologies, Burlington, Canada). RNA was reverse transcribed with Superscript II reverse transcriptase (RT; Life Technologies) as directed and resuspended in 20 µl of sterile H₂O. The CD69⁺ and CD69^- LN RNA samples from each animal were coincidentally reverse transcribed using a master RT mix to minimize differences in RT reaction efficiencies between the samples. For every RT reaction set, RNA prepared from NOD spleen cells was used in a mock (without RT) control reaction to provide a negative control in PCR.

Oligonucleotide primers and PCR amplification

In each set of PCRs, titrations of each cDNA sample were analyzed by PCR for -actin expression (45) and for the expression of B cell-specific gene product of IgH locus as previously described (46) using a degenerate V_H 5'-primer, V_HALL, and a Cµ 3'-primer, Cµ2A. The primer sequences used were 5' -actin primer, TGGGTCAGAAGGADTCCTATG; 3' -actin primer, CAGGCAGCTCATAGCTCTTCT; 5' V_HALL primer, AGGT(C/G)(A/C)A(A/G)CTGCAG(C/G)AGTC(A/T)GG; and 3' Cµ2A primer, CTCGATGGTCACCGGATCTG. PCR with both primer sets was performed as follows: 2 min at 85°C followed by 30 cycles of 1.5 min at 94°C, 1.5 min at 55°C, 2.25 min at 72°C, then 10 min at 72°C in a PE 9700 thermal cycler. All PCR sets in which CD69⁺ and CD69^- samples were compared were amplified simultaneously using common stock mixes and identical thermal cycling programs.

Molecular probes

Plasmids containing nine V_H families (V_H3609P, V_HVGam3-8, V_HSM7, V_HJ606, V_HX24, V_H15, V_HQ52N, V_HS107, V_H36-60) were provided by Dr. R. Riblet. Plasmids containing V_H81X and two members of the V_H7183 family, V_H37 and V_H14, were gifts from Dr. A. Feeney. An Igµ probe (45) was used to detect the total Ig amplification products. Plasmid DNA was prepared as previously described (47), and probes were purified using QIAEX II gel extraction (Qiagen, Chatsworth, CA). Fragments were labeled to high specific activity with [-³²P]dCTP (3000 Ci/mmol; Amersham) by random hexamer labeling as previously described (48).

Southern blot analysis

PCR products were separated by agarose gel electrophoresis, transferred to nylon membrane (ZetaProbe; Bio-Rad, Hercules, CA), and immobilized with UV light (Stratagene, La Jolla, CA). [-³²P]-Labeled DNA probes were prepared and hybridized as previously described (27). Each membrane was first probed with one V_H family probe and exposed to phosphorscreen. The probe was subsequently stripped from the membrane as previously described (49), and the membrane was then rehybridized to ³²P-labeled Cµ probe, followed by exposure to phosphorscreen (see Fig. 5A).

View larger version (29K): [in this window] [in a new window]   FIGURE 5. Southern blot analysis of NOD CD69+ and CD69- B cell VH repertoire. LN cells were isolated from eight NOD mice at age 28 days and enriched for CD69+ and CD69- LN cells by magnetic bead depletion as described in Materials and Methods. Postenrichment, CD69 expression was confirmed by flow cytometry (data not shown). Three- to 5-fold dilutions of cDNA from CD69+ and CD69- LN cells of each animal were amplified with VHALL-Cµ2A and -actin primers and transferred to nylon membrane. A, Each blot was first probed with a 32P-labeled VH family probe (VHJ558 probe is shown) and exposed to a phosphorscreen. The VH probe was removed from the membrane, followed by hybridization of the 32P-labeled Ig Cµ probe to the membrane and exposure to phosphorscreen. B, The digitized signal represented by pixel number was quantified by ImageQuant software. The pixel numbers for the VH and Igµ probes for each sample were determined. The results for six VH families from four individual NOD mice are expressed as the mean ratio of CD69+:CD69- samples ± SD for the three dilutions of input cDNA.

Southern blots were digitized by a phosphorscanner (Molecular Dynamics, Sunnyvale, CA). Absolute pixel number (signal intensity) was quantified by volume integration using ImageQuant software (Molecular Dynamics). Quantitation of cDNA titration was used to create a standard curve comparing input templates to signal output. A control sample (without RT) was used to correct for background hybridization for each reaction set. Quantitation of V_H and Igµ probe signals was performed, and the pixel numbers (P_VH and P_Igµ, respectively) were calculated for each sample. The P_VH:P_Igµ ratio represents the V_H family-specific signal compared with the Cµ signal in the cDNA sample. The P_VH:P_Igµ ratios were calculated for CD69⁺ and CD69^- B cells samples for each animal. To compare the representation of V_H families between CD69⁺ and CD69^- B cells, the CD69⁺ P_VH:P_Igµ ratio was expressed relative to the CD69^- P_VH:P_Igµ ratio. A CD69⁺:CD69^- ratio of 1 indicates no difference between V_H signal observed for CD69⁺ and CD69^- B cells.

Statistical analysis

Statistical analysis of one-way ANOVA was performed using SuperANOVA 1.0 (Abacus Concepts, Berkeley, CA) software. Statistical significance was determined using the Tukey-Kramer test with a significance level of p = 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
NOD V3⁺ T cells proliferate in response to signals through the TCR

Previously we reported that NOD V3⁺ T cells, although rare in the periphery, are well represented among the early islet-infiltrating cells (27). These findings suggested that T cells that escape Mtv-3-mediated clonal deletion may become activated and play a role in the initiation of autoimmune disease. However, one TCR-transgenic model suggested that SAg-reactive T cells that are not deleted by endogenous SAg become anergic to TCR ligation in the periphery (29). To determine the potential functionality of NOD V3⁺ T cells that escape Mtv-3 SAg-mediated deletion, lymphocyte proliferation assays were performed. The proliferative response of NOD V3⁺ T cells was compared with V3⁺ T cells from BALB/c and C57BL/6 (B6) mice. In BALB/c mice, Mtv-6 SAg mediates efficient deletion of V3⁺ T cells (50, 51), while B6 has no V3-specific endogenous SAg. Since V3⁺ T cells in NOD and BALB/c are rare, LN T cells expressing V2, -4, -9, -11, -13, and -14 were removed, resulting in enrichment of V3⁺ T cells to 1% of total T cells. In B6 mice, V3⁺ T cells constitute 3% of the T cell population. For stimulation with anti-V3 Ab, B6 LN cells were diluted so that the absolute number of V3⁺ T cells in these cultures was equivalent to NOD and BALB/c V3-enriched LN cells. The LN T cells prepared in this manner were treated either with the polyclonal activator Con A or with KJ25 (anti-TCR V3 Ab) or isotype-matched Ab. T cells from all three strains displayed equally vigorous responses to Con A (Fig. 1A). Consistent with previous results (29), no significant proliferation was detected from BALB/c V3⁺ T cells. In contrast to BALB/c, NOD V3⁺ T cells proliferated following KJ25-mediated receptor ligation (Fig. 1A). Thus, unlike BALB/c V3⁺ T cells that escape endogenous Mtv SAg-mediated deletion, NOD V3⁺ T cells appear competent to respond to TCR-mediated signals.

View larger version (32K): [in this window] [in a new window]   FIGURE 1. Activation and proliferation of B6, BALB/c, and NOD V3+ T cells in vitro by TCR ligation. LN cells were isolated from 4- to 6-wk-old B6, BALB/c, and NOD mice. LN cells were enriched for V3+ T cells by magnetic bead depletion as described in Materials and Methods. A, Equivalent numbers of V3+ T cells from each strain were activated with Con A, immobilized anti-TCR V3 Ab (KJ25), or an isotype-matched control Ab (hamster IgG). After 72 h all samples were pulsed with [3H]thymidine to quantify proliferating cells. Left panel, B6, BALB/c and NOD LN cells proliferated in response to the lectin, Con A. Right panel, B6, BALB/c, and NOD V3-enriched LN cells proliferation in the presence of KJ25 compared with the isotype control Ab. B, Cytokine release by in vitro-activated NOD and BALB/c T cells and V3+ T cells. LN cells from six mice of each strain were enriched for V3+ T cells. Postenrichment, V3+ T cells constituted approximately 1% of the APC-depleted, T cell population. V3-enriched LN cells (1 x 107) were activated with immobilized anti-CD3 (145-2C11) or KJ25 Ab. After 72 h supernatants were analyzed for IFN- and IL-4 by ELISA in triplicate. Because the Ab 145-2C11 activates all of the input cells, whereas KJ25 activates only 1% of the input cells, the table depicts the results as the mean ± SD per 105 responder cells for each assay. The lower limit of detection for IFN- and IL-4 by ELISA was 0.1 pg/ml.

Frequency of activated NOD V3⁺ T cells in vivo

Because NOD V3⁺ T cells responded to TCR ligation in culture (Fig. 1) and were found in the earliest detectable islet infiltrates (27), we postulated that a proportion of these T cells were activated in vivo in young NOD mice. Two cell surface markers were used to examine this idea: CD69, an early activation marker for B and T cells (reviewed in Ref. 54), and CD25, the -chain of the high-affinity IL-2R expressed primarily on activated T cells (reviewed in Ref. 55). Using multiparameter flow cytometry, we first analyzed CD69 and CD25 expression on NOD V3⁺, V7⁺, and V8⁺ T cell subsets (0.3, 3–5, and 15–20% of the T cell population, respectively) in 2- to 7-wk-old NOD mice (n = 20). A higher frequency of V3⁺ LN T cells expressed CD69⁺ compared with V7⁺ or V8⁺ T cell populations from the same animals (p = 0.0001, by one-way ANOVA; Fig. 2, A and B). Similarly, a higher frequency of NOD V3⁺ T cells coexpressed CD25 compared with V8⁺ T cells (p = 0.0001, by one-way ANOVA; Fig. 2B). Multiple LN sites (pancreatic, mesenteric, inguinal, and axillary) were compared for V3⁺ T cell CD69 and CD25 cell surface expression. No significant difference in the percentages of CD69⁺ or CD25⁺ NOD V3⁺ T cells was observed between these LN sites (data not shown). Collectively, these results suggest a higher frequency of NOD V3⁺ T cells are systemically activated in vivo compared with other T cell subsets.

View larger version (30K): [in this window] [in a new window]   FIGURE 2. Expression of activation markers on NOD, BALB/c, or B6 T cells. LN cells (2 x 106) isolated from NOD, B6, and BALB/c mice were stained with FITC-anti-V3 (KJ25), -anti-V7 (TR310), -anti-V8 (F23.1) and biotinylated anti-CD69 (H1.2F3) or anti-CD25 (7D4) Ab, followed by AV-PE. Dead cells were excluded by propidium iodide staining. Results are expressed as the percentage of NOD, BALB/c, or B6 V3+, V7+, or V8+ cells that coexpressed CD69 or CD25. Each bar represents the results from 10 separate experiments performed on NOD (n = 20), BALB/c (n = 10), and B6 (n = 10) mice between 2 and 7 wk of age. The error bars represent the SE. A, FACS histograms of CD69 surface expression on NOD V3+ and V8+ T cells from one representative experiment. B, CD69 and CD25 expression on NOD V3+ and V8+ T cells. *, p = 0.0001, by one-way ANOVA. C, CD69 expression on NOD and BALB/c V3+ and V8+ T cells. *, p = 0.0001, by one-way ANOVA. D, CD25 expression on NOD and B6 V3+ and V8+ T cells. *, p = 0.0001 (by one-way ANOVA).

Next we asked whether the high frequency of in vivo activated V3⁺ T cells observed in NOD mice was evident in diabetes-resistant strains. We compared V3⁺ T cells from NOD to B6, which lacks a V3-specific Mtv, and to BALB/c , in which Mtv-6 SAg mediates deletion of V3⁺ thymocytes (50, 51). Compared with BALB/c , a higher frequency of NOD V3⁺ T cells coexpressed CD69 (p = 0.0001, by one-way ANOVA; Fig. 2C), but no significant differences were observed in V8⁺ T cell CD69 expression between the two strains. Similarly, a higher frequency of NOD compared with B6 V3⁺ T cells expressed CD25 (p = 0.0001, by one-way ANOVA; Fig. 2D) despite the 10-fold greater frequency of V3⁺ T cells in B6 compared with NOD mice. These data suggest that a significantly higher percentage of NOD V3⁺ T cells that escape thymic deletion are activated in vivo, even compared with BALB/c V3⁺ T cells that escape Mtv-6-mediated deletion. Thus, NOD V3⁺ T cells can be activated by signals through their TCR both in vitro and in vivo, a phenotype that was not observed in diabetes-resistant strains.

V3⁺ T cell activation segregates with the Mtv-3 provirus

The systemic activation of NOD V3⁺ T cells suggested a role for a ubiquitously expressed Ag. The V3-specific Mtv-3 SAg was a logical candidate given the distribution of APC, particularly activated B cells, previously shown to present Mtv SAg (38). To examine the requirement for Mtv-3 in the in vivo activation of NOD V3⁺ T cells, a genetic analysis was performed in a series of (NOD x B6)F₂ mice segregating Mtv-3. The Mtv-3 genotype of the F₂ mice was evaluated by Southern blot and FACS analysis (see Materials and Methods). Because of the well-established role of MHC haplotype in T cell repertoire selection, the V3⁺ T cell activation marker phenotype was evaluated in 40 H-2^g7 homozygous F₂ mice; 20 of these were Mtv-3 negative and 20 were Mtv-3 positive (see Materials and Methods). A higher frequency of CD69⁺V3⁺ T cells was observed in Mtv-3⁺ (NOD x B6)F₂ mice compared with the Mtv-3-negative littermates (p = 0.0001, by one-way ANOVA; Fig. 3). In the Mtv-3⁺ (NOD x B6)F₂ mice, the frequency of CD69⁺V3⁺ T cells was also greater than that of V8⁺ T cells (p = 0.0001, by one-way ANOVA; Fig. 3), as observed in NOD animals (Fig. 2). Together, these data suggest that both thymic and peripheral activation of V3⁺ NOD T cells require the Mtv-3 provirus, implicating the endogenous Mtv SAg in the activation of V3⁺ NOD T cells.

View larger version (25K): [in this window] [in a new window]   FIGURE 3. V3+ T cell activation in (NODxB6)F2 mice. (NODxB6)F2 mice were analyzed for H-2 and Mtv-3 genotypes (Materials and Methods). LN cells from H-2g7 homozygotes were stained with KJ25, F23.1, and H1.2F3. A, FACS histograms comparing CD69 surface expression on V3+ (right panels) and V8+ (left panels) T cells from Mtv-3 deficient (upper panels) and Mtv-3+ (lower panels) F2 animals. The solid lines represent isotype control. Dotted lines represent CD69 expression on V-specific subset. B, Comparison of CD69 surface expression on V3+ and V8+ T cells from (NOD x B6)F2 Mtv-3 segregants. Results are expressed as described in Fig. 2. Each bar represents the results from 20 mice. *, p = 0.0001; **, p = 0.0001 (by one-way ANOVA).

These data suggested that Mtv-3 was linked to the high frequency of activated V3⁺ T cells in NOD mice. Because multiple studies have shown that activated B cells are more efficient presenters of Mtv SAg than macrophages and dendritic cells (38, 56, 57), we examined the activation status of NOD B cells by surface expression of CD69 using flow cytometry. Interestingly, a greater percentage of NOD B220⁺ cells coexpressed CD69 than was seen in either BALB/c (Fig. 4A) or B6 mice (p = 0.0001, by one-way ANOVA; Table I), whereas expression of this marker on myeloid APC was equivalent between strains (Fig. 4A). Similarly, a higher frequency of NOD CD19⁺IgM⁺ cells expressed CD69 compared with similar subsets from control mice, confirming the B cell phenotype of these cells (data not shown). This high percentage of activated B cells was observed in NOD mice as young as 14 days of age, 10–15 days prior to the onset of peri-insulitis (data not shown).

View larger version (27K): [in this window] [in a new window]   FIGURE 4. B cell activation in NOD and BALB/c mice. A, LN cells (2 x 106) from NOD and BALB/c mice were stained with FITC-anti-B220 (RA3-6B2), FITC-anti-Mac-1 (M1/70), and biotinylated anti-CD69 (H1.2F3) Ab, followed by AV-PE. Dead cells were excluded by propidium iodide staining. Results are expressed as the percentage of 6B2+ or M1/70+ cells that coexpressed CD69. Each bar represents the average results (± SE) from 23 NOD and 16 BALB/c mice. *, p = 0.0001; **, p = 0.47 (by one-way ANOVA). B, Phenotype of LPS-activated and in vivo-activated NOD B cells. LN cell suspensions were stained with FITC-anti-class II MHC (10-2.16), FITC-anti-B7-1 (16-10A1), FITC-anti-B7-2 (GL1), FITC-anti-ICAM-1 (3E2), biotinylated anti-CD69 (H1.2F3), and allophycocyanin-conjugated anti-B220 (RA3-6B2), followed by AV-PE, and analyzed by three-color FACS. LPS-activated LN cells were used as controls. The histograms depict the fluorescence intensity gated on B220+ LN cells for class II, B7-1, B7-2, or ICAM-1 molecules on LPS-treated or fresh ex vivo NOD LN B cells.

The high frequency of activated NOD LN B cells could represent either polyclonal activation by many Ags or mitogens or an oligoclonal subset activated by Ag-specific interactions. To address this issue, the Ig V_H repertoire expressed by CD69⁺ and CD69^- B cells was compared in individual NOD mice. Magnetic bead immunodepletion was used to partition CD69⁺ and CD69^- NOD LN cells and resulted in >90% purity of these subsets (data not shown). A semiquantitative RT-PCR and Southern blot analysis approach was designed to detect transcripts (cDNA) containing nine different Ig V_H genes in NOD CD69⁺ and CD69^- LN cells (see Materials and Methods). A single pair of IgH primers capable of amplifying a broad repertoire of V_H genes in Igµ transcripts (V_HALL and Cµ2A) (46) was used for each sample. Replicate RT-PCRs were evaluated by Southern blot hybridization with nine V_H family-specific probes and with a Cµ to normalize for total IgH cDNA. Several of the V_H probes were selected because they are infrequently represented in the mature B cell repertoire and were more likely to be absent from an oligoclonal population. Serial dilutions of cDNA demonstrated that the sensitivity of IgH cDNA detection was approximately 10 cells (data not shown). Using this approach, no significant or reproducible difference in V_H family gene usage was detected between CD69⁺ and CD69^- NOD B cells in four individual mice (Fig. 5B). These data were most consistent with polyclonal B cell activation in NOD mice.

Activation of NOD V3⁺ T cells is B cell independent

The correlation between the high frequencies of in vivo activated V3⁺ T cells and B cells in NOD mice suggests that B cells may be involved in Ag and/or SAg presentation to and activation of V3⁺ T cells. To determine whether B cells were required to generate the high frequency of activated V3⁺ T cells and whether thymic deletion of V3⁺ T cells by Mtv-3 was B cell dependent, we used B cell-deficient NOD.µMT mice carrying disruption of the Igµ gene (31). Identical frequencies of V3⁺ T cells were observed in NOD.µMT and NOD mice, suggesting that deletion of NOD V3⁺ T cells by Mtv-3 is not dependent upon mature B cells (Fig. 6A). Interestingly, there were also no significant differences in the percentages of CD69⁺ or CD25⁺V3⁺ T cells between NOD and NOD.µMT mice (Fig. 6B). Together, these results suggest that NOD B cells are not required for either Mtv-3-mediated V3⁺ thymocyte deletion or peripheral activation of V3⁺ T cells.

View larger version (41K): [in this window] [in a new window]   FIGURE 6. Impact of B cells on T cell activation in NOD mice. A, FACS histograms showing the percentage of V3+ T cells from 4-wk-old NOD and NOD.µMT mice. LN cells were stained with FITC anti-TCR C (H57.597) and biotinylated anti-V3 (KJ25), followed by AV-PE. B, Percentages of V3+ and V8+ T cells expressing activation markers in NOD and NOD.µMT mice. Each bar represents the average (± SE) from 10 animals.

A high frequency of activated B cells was observed in NOD mice by14 days of age, suggesting that this phenotype precedes insulitis onset. A critical question was what mechanisms regulate NOD B cell activation. Because B cells can be activated by CD40 signals provided by T cells following interactions between TCR/Ag-MHC (3, 8, 10, 11), we asked whether B cell activation was genetically dependent on the NOD MHC haplotype H-2^g7. To investigate this idea, the frequencies of activated B cells in NOD, (NOD x B6)F₁, (NODxB6)F₂, NOD H-2 congenics NOD.B10 (58), NOD.SWR, and MHC class II-deficient NOD.I-A^-/- (42) were compared by flow cytometry. The frequency distributions of CD69⁺ B cells were similar among all H-2^g7 homozygous (NODxB6)F₂ mice (Table I and Fig. 7) and did not segregate with Mtv-3 (data not shown). Interestingly, H-2^b/g7 (NODxB6)F₁ and H-2^g7 homozygous (NOD x B6)F₂ mice displayed frequency distributions of CD69⁺ B cells similar to NOD mice (p = 0.99 and p = 0.68, respectively, by one-way ANOVA). In contrast, H-2^d (BALB/c), H-2^b (B6), H-2^b (NOD x B6)F₂, H-2^b NOD.B10, and H-2^q NOD.SWR mice all had lower frequencies of CD69⁺ B cells (Table I and Fig. 7). The H-2 haplotype correlated with B cell activation, as CD69 expression on B cells was significantly different in H-2^g7, H-2^b, and H-2^d mice (p = 0001, by one-way ANOVA; Table I). In addition, a significantly lower percentage of activated B cells was observed in NOD.I-A^-/- mice compared with wild-type NOD mice (p = 0.01, by one-way ANOVA). Interpretation of this result should be tempered by the presence of a 19-cM length of 129/Sv-derived chromosome 17 in NOD.I-A^-/- mice that includes H-2K^b, as distinct from the NOD MHC haplotype that is H-2K^d (data not shown). Therefore, we cannot rule out a dampening effect of H-2K^b on the B cell activation phenotype. Collectively, these results suggest that the frequency of activated B cells in NOD mice is genetically linked to the Idd1 haplotype, which exerts a dominant effect on this phenotype. An elegant study showed that MHC class II expression on NOD B cells strongly influences diabetes incidence (59). Our results provide insight into a probable mechanism of this effect, that I-A^g7 expression is required for B cell activation by T cell costimulatory interactions, enabling their APC function in diabetes pathogenesis.

View larger version (25K): [in this window] [in a new window]   FIGURE 7. The high frequency of activated B cells is linked to the NOD MHC. (NOD x B6)F1 and (NOD x B6)F2 mice were bred and screened for H-2 haplotypes as described (see Materials and Methods). LN cells were stained with FITC RA3-6B2 and biotinylated H1.2F3. The average percentage of CD69+B220+ LN cells for each mouse from each strain is shown, with one circle representing one mouse. Each vertical line represents the average percentage of CD69+B220+ LN cells for each strain.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Although T cell surface expression of CD25 (IL-2R -chain) is up-regulated following T cell activation, CD25 may also be a marker for a subset of T cells with regulatory function (reviewed in Ref. 63). Regulatory CD25⁺ T cells derive from CD25⁺CD4⁺CD8^- thymocytes and acquire their characteristic phenotype during thymic maturation (64). In the peripheral lymphoid population, the distinction between CD25⁺ regulatory vs CD25⁺-activated CD4⁺ T cells is based on functional studies. CD25⁺CD4⁺ T cells do not proliferate in response to TCR stimulation (64) and protect against autoimmune disease onset upon cotransfer with pathogenic CD25^-CD4⁺ T cells (64, 65). Thus, the functional significance of this regulatory T cell population distinguishes their identity from other T cell subsets rather than by CD25 expression alone. The significance of this regulatory subset in T1D was described in one recent study that reported the role of NOD splenic CD25⁺CD4⁺ T cells in NOD T1D development (66). Importantly, these CD25⁺CD4⁺ T cells were shown to oppose the effects of islet-reactive CD25^-CD4⁺ T cells in coadoptive transfer experiments, thereby protecting against diabetes onset. Although these results may imply a regulatory role for the CD25⁺V3⁺ LN T cells identified in our study, the available evidence suggests that they are unlikely to be regulatory T cells. First, the distinctive feature of regulatory CD25⁺CD4⁺ T cells is that they proliferate poorly in response to TCR signals (64, 67, 68). In contrast, we found that NOD V3⁺ T cells proliferate vigorously in response to TCR ligation in vitro. Second, one report suggests that CD25⁺CD4⁺ regulatory cells are capable of producing both Th1 and Th2 cytokines, such as IFN- and IL-4 (69). However, IFN-, but not IL-4, was detected following activation of NOD V3⁺ T cells, whereas both cytokines were produced by polyclonal activated NOD T cells. Thus, NOD V3⁺ T cells do not conform to the cytokine profile of regulatory T cells. Finally, these data are consistent with previous evidence that NOD V3⁺ T cells were represented among early islet-infiltrating cells (27), indicating that V3⁺ T cells activated in vivo migrate to sites of autoimmune inflammation, potentially mediating a proinflammatory response. It is possible that both regulatory and effector cells are represented within the NOD V3⁺ T cell subset, as islet Ag-specific, TCR V-specific T cells can mediate both functions (70), and regulatory and effector cells overlap in the expression of cell surface markers such as CD25. However, these data taken together are more consistent with the idea that CD25 expression on NOD V3⁺ T cells reflects their activation rather than a regulatory phenotype. Their proliferative response and cytokine profile suggest that NOD CD25⁺V3⁺ T cells may be autoreactive T cells participating in the pathogenesis of NOD diabetes. More significantly, our results reflect the failure to maintain peripheral tolerance of NOD V3⁺ T cells, indicating the mechanisms by which autoimmune disease develops in the NOD mouse.

The high frequency of activated NOD V3⁺ T cells was an unexpected finding, as V3⁺ T cells constitute <1% of NOD T cells. Given the equally high frequencies of V3⁺ T cells expressing the activation markers CD25 and CD69 in different anatomic sites, the data suggested systemic, rather than pancreas-proximal, activation of this T cell subset. We observed that NOD V3⁺ T cell activation was genetically linked to the Mtv-3 locus, consistent with the idea that TCR interaction with endogenous SAg contributes to this activated phenotype. Although SAg interact with nonpolymorphic residues in TCRV, both the TCR -chain and variable TCR determinants influence specificity for the SAg/MHC complexes (reviewed in Ref. 71). As a result, some V3⁺ T cells may escape SAg-mediated thymic deletion because their TCR -chain and/or other TCR determinants limit affinity/avidity for SAg/MHC complexes (72). Alternatively, some potentially SAg-reactive thymocytes escape because they fail to encounter a sufficient density of SAg to mediate deletion. Whether due to differential TCR affinity for SAg, or stochastic inefficiencies in thymic deletion, V3⁺ T cells may respond to Mtv-3 SAg when presented at sufficient concentration by professional APC (73). Taken together with the previous demonstration that V3⁺ T cells contribute to early NOD insulitis, Mtv-3 SAg-mediated activation of these T cells may promote their involvement in autoimmune inflammation.

Although our results are consistent with a role for Mtv-3 in the activation of NOD V3⁺ T cells, these results cannot rule out that Mtv-3 is a marker for a linked gene. One candidate locus on chromosome 11 is the Idd4, previously shown to affect the frequency and progression of insulitis to overt diabetes (74, 75). Idd4 behaves as a recessive or dominant with low penetrance gene when scoring diabetes in crosses between NOD and C57BL/10 mice (74, 75, 76), but the relevant gene(s) at this locus has not yet been identified. In our study the NOD-derived chromosome 11 carrying Mtv-3 behaved as a fully penetrant dominant locus for both deletion of V3⁺ thymocytes and activation of V3⁺ peripheral T cells. Thymocyte deletion and peripheral T cell activation phenotypes always cosegregated in the (NOD x B6)F₂ progeny. This genetic behavior is identical with that of other Mtv-encoded SAg specific for other TCRV determinants. While it remains possible that Idd4 or another linked locus affected the deletion/activation of V3⁺ T cells, the data are most consistent with the interpretation that Mtv-3 controls these phenotypes in NOD mice.

The idea that viral Ag may be involved in the etiology of T1D has received considerable attention. Common human pathogenic viruses, including coxsackie (77) and congenital rubella (78), have been associated with the onset of human diabetes. It is possible that exposure to specific viral Ag and/or SAg may provoke activation of T cells cross-reactive with self-Ag, leading to the initiation of autoimmune disease in genetically susceptible individuals. Indeed, molecular mimicry between peptides from coxsackie virus and glutamic acid decarboxylase, a neuronal enzyme and an islet Ag (79, 80, 81), has been argued to support a hypothesis that T cell cross-reactivity to viral and self Ag presented on I-A^g7 may promote islet cell destruction in T1D (82, 83). Indeed it has been suggested that viral SAg influence diabetes in NOD mice (84) and in humans (85, 86), although the issue has generated controversy (87) and remains unresolved.

We have shown that a striking frequency of NOD B cells are activated in vivo, a feature observed by 2 wk of age and persisting until T1D onset (data not shown). This finding suggests that NOD B cells may be programmed to activate and proliferate in young NOD mice before the onset of insulitis. Therefore, enhanced lymphocyte activation in NOD mice is not restricted to T cells. We found that activated NOD B cells expressed high levels of MHC class II, costimulatory, and adhesion molecules, predicting that they are competent APC for naive T cells. Analyses of (NOD x B6)F₂ and NOD congenic mice demonstrated the requirement for H-2^g7 to generate the high frequency of activated NOD B cells. Together these data strongly suggest that activated B cells expressing NOD MHC class II molecules may serve as APC in the activation of autoreactive T cells. Indeed, recent reports suggest that NOD B cells are important for the development of T1D (31, 32, 33, 34), serve an important APC function (35, 36, 37), and must express I-A^g7 to contribute to diabetes development (59). The unique NOD class II molecule I-A^g7 has been implicated in the presentation of self-peptides due to promiscuous or altered peptide binding (59, 88, 89). Transgenic expression of MHC class II I-E or alternative alleles of I-A confers strong protection against T1D in NOD mice (reviewed in Ref. 90), suggesting a role for the NOD MHC class II haplotype in the selection and activation of autoreactive T cells (91, 92, 93). Furthermore, the NOD MHC contributes to isotype switch in the production of anti-insulin autoantibodies (94). Therefore, the NOD MHC class II haplotype is pivotal in multiple aspects of T cell and B cell function during progression to T1D.

Previous reports indicate that the NOD splenic B cell V_H repertoire displays a bias toward D-proximal V_H genes compared with B cells from nondiabetes-prone strains (95). We examined the V_H repertoire of the CD69⁺ NOD B cells in individual mice to determine whether they represented expansion of an oligoclonal B cell subset. The repertoire of CD69⁺ NOD B cells did not differ significantly from that of CD69^- B cells in the same animal, suggesting that NOD B cell activation in vivo is stochastic and not Ag restricted. This feature distinguishes NOD CD69⁺ B cells from the CD5⁺ B-1 B cell subset, which are of fetal and peritoneal origin and are characterized by a limited V_H repertoire (reviewed in Ref. 96). However, like the CD5⁺ B-1 subset in the NZB and motheaten models of systemic lupus erythematosus (97), the NOD-activated B cell phenotype showed genetic linkage to a disease susceptibility locus, Idd1 (16). Hence, genetic regulation of B cell activation may contribute directly to the development of diabetes and other forms of autoimmune disease.

The activated V3⁺ T cell subset, the genetic linkage of this phenotype to Mtv-3, and the high frequency of CD69⁺ NOD B cells could be mechanistically related if activated B cells present Mtv-3 SAg to the V3⁺ T cells. However, we found that a similar frequency of CD25⁺CD69⁺V3⁺ T cells was observed in NOD.µMT mice that lack mature B cells. Our data indicate that neither the deletion nor the activation of V3⁺ T cells requires B cells, as these phenotypes did not differ between NOD and B cell-deficient NOD.µMT mice. In contrast to evidence that B cells are required for the negative selection of other subsets of Mtv-reactive T cells (98, 99), presentation by myeloid APC appears adequate to mediate both clonal deletion and activation of NOD V3⁺ T cells. Thus, there may be distinct requirements for B cells in activation of different islet-reactive T cells specificities in NOD mice (35, 36, 59).

	Acknowledgments

 
We thank Valerie Roy, Denise Martin, and Drs. Gillian Wu, Joan Wither, Trang Duong, Rae Yeung, and Roy Riblet for reagents and technical advice. We thank Drs. Philippe Poussier, Joan Wither, Gillian Wu, and Cynthia Guidos for many helpful discussions and the review of this manuscript.

	Footnotes

 
¹ This work was supported by grants from the Canadian Diabetes Association in honor of Madelen Francis Armstrong and the Juvenile Diabetes Research Foundation. J.S.D. is a Research Scientist of the National Cancer Institute of Canada. P.P.L.C. is the recipient of a Juvenile Diabetes Research Foundation Postdoctoral Fellowship.

² Address correspondence and reprint requests to Dr. Jayne S. Danska, Program in Developmental Biology, Hospital for Sick Children Research Institute, University of Toronto, 555 University Avenue, Toronto, Ontario M5G 1X8, Canada. E-mail address: danska{at}sickkids.on.ca

³ Abbreviations used in this paper: T1D, type 1 diabetes mellitus; AV-PE, avidin-PE; LN, lymph node; Mtv, murine mammary tumor virus; NOD, nonobese diabetic; SAg, superantigen; RT, reverse transcriptase.

Received for publication July 24, 2001. Accepted for publication October 5, 2001.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Faveeuw, C., M.-C. Gagnerault, F. Lepault. 1994. Expression of homing and adhesion molecules in infiltrated islets of Langerhans and salivary glands of nonobese diabetic mice. J. Immunol. 152:5969.[Abstract]
2. Shlomchik, M. J., M. P. Madaio, D. Ni, M. Trounstein, D. Huszar. 1994. The role of B cells in lpr/lpr-induced autoimmunity. J. Exp. Med. 180:1295.[Abstract/Free Full Text]
3. Kouskoff, V., A.-S. Korganow, V. Duchatelle, C. Degott, C. Benoist, D. Mathis. 1996. Organ-specific disease provoked by systemic autoreactivity. Cell 87:811.[Medline]
4. Peterson, L. D., G. Duinkerken, G. J. Bruining, R. A. W. van Lier, R. R. P. de Vries, B. O. Roep. 1996. Increased numbers of in vivo activated T cells in patients with recent onset insulin-dependent diabetes mellitus. J. Autoimmun. 9:731.[Medline]
5. Mato, T., K. Masuko, Y. Misaki, N. Hirose, K. Ito, Y. Takemoto, K. Izawa, S. Yamamori, T. Kato, K. Nishioka, et al 1997. Correlation of clonal T cell expansion with disease activity in systemic lupus erythematosus. Int. Immunol. 9:547.[Abstract/Free Full Text]
6. Chan, O., M. J. Shlomchik. 1998. A new role for B cells in systemic autoimmunity: B cells promote spontaneous T cell activation in MRL-lpr/lpr mice. J. Immunol. 160:51.[Abstract/Free Full Text]
7. Gessl, A., W. Waldhausl. 1998. Increased CD69 and human leukocyte antigen-DR expression on T lymphocytes in insulin-dependent diabetes mellitus of long standing. J. Clin. Endocrinol. Metab. 83:2204.[Abstract/Free Full Text]
8. Ishikawa, S., S. Akakura, M. Abe, K. Terashima, K. Chijiwa, H. Nishimura, S. Hirose, T. Shirai. 1998. A subset of CD4⁺ T cells expressing early activation antigen CD69 in murine lupus: possible abnormal regulatory role for cytokine imbalance. J. Immunol. 161:1267.[Abstract/Free Full Text]
9. Odendahl, M., A. Jacobi, A. Hansen, E. Feist, F. Hiepe, G. R. Burmester, P. E. Lipsky, A. Radbruch, T. Dorner. 2000. Disturbed peripheral B lymphocyte homeostasis in systemic lupus erythematosus. J. Immunol. 165:5970.[Abstract/Free Full Text]
10. Croft, M., S. L. Swain. 1995. Recently activated naive CD4 T cells can help resting B cells, and can produce sufficient autocrine IL-4 to drive differentiation to secretion of Th 2-type cytokines. J. Immunol. 154:4269.[Abstract]
11. Bode, U., K. Wonigeit, R. Pabst, J. Westermann. 1997. The fate of activated T cells migrating through the body: rescue from apoptosis in the tissue of origin. Eur. J. Immunol. 27:2087.[Medline]
12. Constant, S., N. Schweitzer, J. West, P. Ranney, K. Bottomly. 1995. B lymphocytes can be competent APCs for priming CD4⁺ T cells to protein antigens in vivo. J. Immunol. 155:3734.[Abstract]
13. Lin, R.-H., R. J. Mamula, J. A. Hardin, C. A. Janeway. 1991. Induction of autoreactive B cells allows priming of autoreactive T cells. J. Exp. Med. 173:1433.[Abstract/Free Full Text]
14. Chan, O. T., L. G. Hannum, A. M. Haberman, M. P. Madaio, M. J. Shlomchik. 1999. A novel mouse with B cells but lacking serum antibody reveals an antibody-independent role for B cells in murine lupus. J. Exp. Med. 189:1639.[Abstract/Free Full Text]
15. Sobel, E. S., C. Mohan, L. Morel, J. Schiffenbauer, E. K. Wakeland. 1999. Genetic dissection of SLE pathogenesis: adoptive transfer of Sle1 mediates the loss of tolerance by bone marrow-derived B cells. J. Immunol. 162:2415.[Abstract/Free Full Text]
16. 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]
17. Watanabe-Fukunaga, R., C. I. Brannan, N. G. Copeland, N. A. Jenkins, S. Nagata. 1992. Lymphoproliferative disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356:314.[Medline]
18. Foster, D. W.. 1983. Diabetes mellitus. E. Braunwald, ed. In Harrison’s Principles of Internal Medicine Vol. 2:1778. McGraw-Hill, New York.
19. Peakman, M., L. Alviggi, M. J. Hussain, D. A. Pyke, R. D. G. Leslie, D. Vergani. 1991. Persistent T cell activation predicts diabetes in unaffected identical co-twins of type 1 diabetics. Diabetologia 34:A56.
20. Kallmann, B. A., E. F. Lampeter, P. Hanifi-Moghaddam, M. Hawa, R. D. Leslie, H. Kolb. 1999. Cytokine secretion patterns in twins discordant for type I diabetes. Diabetologia 42:1080.[Medline]
21. Jacob, H. J., A. Pettersson, D. Wilson, Y. Mao, A. Lernmark, E. S. Lander. 1992. Genetic dissection of autoimmune type I diabetes in the BB rat. Nat. Genet. 2:56.[Medline]
22. Makino, S., K. Kunimoto, Y. Muraoka, Y. Mizushima, K. Katagiri, Y. Tochino. 1980. Breeding of a nonobese diabetic strain of mice. Exp. Anim. 29:1.
23. Penha-Goncalves, C., K. Lejon, L. Persson, D. Holmberg. 1995. Type 1 diabetes and the control of dexamethasone-induced apoptosis in mice maps to the same region on chromosome 6. Genomics 28:398.[Medline]
24. Colucci, F., M.-L. Bergman, C. Penha-Goncalves, C. M. Cilio, D. Holmberg. 1997. Apoptosis resistance of nonobese diabetic peripheral lymphocytes linked to the Idd5 diabetes susceptibility region. Proc. Natl. Acad. Sci. USA 94:8670.[Abstract/Free Full Text]
25. Ridgway, W. M., M. Vasso, A. Lanctot, C. Garvey, C. G. Fathman. 1996. Breaking self-tolerance in nonobese diabetic mice. J. Exp. Med. 183:1657.[Abstract/Free Full Text]
26. Ridgway, W. M., H. Ito, M. Fasso, C. Yu, C. G. Fathman. 1998. Analysis of the role of variation of major histocompatibility complex class II expression on nonobese diabetic (NOD) peripheral T cell response. J. Exp. Med. 188:2267.[Abstract/Free Full Text]
27. Galley, K. A., J.