Iddm30 Controls Pancreatic Expression of Ccl11 (Eotaxin) and the Th1/Th2 Balance within the Insulitic Lesions

Gary Y. C. Chao, Robert H. Wallis, Leili Marandi, Terri Ning, Janice Sarmiento, Andrew D. Paterson and Philippe Poussier
J Immunol April 15, 2014, 192 (8) 3645-3653; DOI: https://doi.org/10.4049/jimmunol.1302383

Abstract

The autoimmune diabetic syndrome of the BioBreeding diabetes–prone (BBDP) rat is a polygenic disease that resembles in many aspects human type 1 diabetes (T1D). A successful approach to gain insight into the mechanisms underlying genetic associations in autoimmune diseases has been to identify and map disease-related subphenotypes that are under simpler genetic control than the full-blown disease. In this study, we focused on the β cell overexpression of Ccl11 (Eotaxin), previously postulated to be diabetogenic in BBDR rats, a BBDP-related strain. We tested the hypothesis that this trait is genetically determined and contributes to the regulation of diabetes in BBDP rats. Similar to the BBDR strain, we observed a time-dependent, insulitis-independent pancreatic upregulation of Ccl11 in BBDP rats when compared with T1D-resistant ACI.1u.lyp animals. Through linkage analysis of a cross-intercross of these two parental strains, this trait was mapped to a region on chromosome 12 that overlaps Iddm30. Linkage results were confirmed by phenotypic assessment of a novel inbred BBDP.ACI-Iddm30 congenic line. As expected, the Iddm30 BBDP allele is associated with a significantly higher pancreatic expression of Ccl11; however, the same allele confers resistance to T1D. Analysis of islet-infiltrating T cells in Iddm30 congenic BBDP animals revealed that overexpression of pancreatic Ccl11, a prototypical Th2 chemokine, is associated with an enrichment in Th2 CD4+ T cells within the insulitic lesions. These results indicate that, in the BBDP rat, Iddm30 controls T1D susceptibility through both the regulation of Ccl11 expression in β cells and the subsequent Th1/Th2 balance within islet-infiltrating T lymphocytes.

Introduction

The BioBreeding diabetes–prone (BBDP) rat is a model for human type 1 diabetes (T1D) (1). The diabetic syndrome of the BBDP rat is a polygenic disease characterized by an early inflammation of the pancreas also known as insulitis that precedes the T cell–mediated autoimmune destruction of insulin-secreting pancreatic β cells (2–7). In our rat colony, 80–90% of BBDP rats spontaneously develop T1D by the age of 120 d, without sex bias. Two T1D susceptibility genes have been previously identified in this animal model, as follows: the MHC (RT1) class II u haplotype (Iddm1) on chromosome 20 (3, 8, 9) and a recessive null mutation of GTPase, IMAP family member 5 (Gimap5; Iddm2; Lyp) on chromosome 4 (3, 10, 11). This Gimap5 mutation is responsible for the premature death of recent thymic emigrants, resulting in severe peripheral lymphopenia that affects all T cell subsets, including regulatory T cells (12, 13). This impaired development of regulatory T cells appears to be the main diabetogenic mechanism contributed by this mutation (12, 14, 15). Although multiple additional Iddm loci linked to either T1D, severity of insulitis, or age of disease onset were subsequently mapped in the BBDP rat, the chromosomal intervals of these loci were large, thus making the search for the causal genes within and the characterization of the underlying pathogenic mechanisms controlled by these genes difficult (2, 16, 17). A successful approach to obtain insight into the genetic basis of autoimmunity has been to identify and map disease-related subphenotypes that are highly penetrant, quantifiable, and under simpler genetic control than the multistep, full-blown autoimmune disease (3, 14, 18–21). The cloning of the lymphopenia-causing gene in the BBDP model is an example of this approach (22, 23). We focused in this study on a subphenotype that was recently observed at the level of the target tissue of the diabetogenic process, namely the upregulation of the chemokine Eotaxin (Ccl11) in the pancreatic β cells of T1D-prone rats (24).

The pancreatic β cell–specific upregulation of Ccl11 was first reported in DR rats, an inbred strain that was derived from the original BBDP/Worcester forebears (24). By using immunohistochemistry on pancreatic sections, Ccl11 was detected in pancreatic β cells as early as 40 d of age, hence prior to any detectable signs of insulitis in lymphopenic and T1D-susceptible DRlyp/lyp rats, and in age-matched, T1D-resistant, and insulitis-free DR+/+ rats that are congenic for wild-type Gimap5 (24). This β cell–specific expression of Ccl11 was, however, not detectable in T1D-resistant Wistar–Furth (WF) rats that share the same MHC RT1u haplotype (24). Together, these observations suggested that β cell overexpression of Ccl11 is an early preinsulitic phenotype that is independent of Iddm2 and possibly under control of the DR rat genetic background.

The observation of chemokine expression by β cells is not new, and, in fact, multiple chemokines are readily detectable in normal rodent and human pancreatic islets (25, 26). The physiological significance of this expression remains elusive, although it has been shown that the interaction of extracellular matrix proteins with β cells results in NF-κB activation and the subsequent synthesis of cytokines by β cells, in vitro (27). Ccl11 is one of the three members of the eotaxin family of C-C chemokines that are potent chemoattractants for Ccr3-expressing cells, which include eosinophils, mast cells, basophils, neutrophils, and Th2 lymphocytes (28–34). Although the presence of eosinophils and mast cells within the insulitic lesions of BBDP rats has been previously reported (35–37), the observation that Ccl11 is upregulated in the target organ of this diabetogenic process remains puzzling (24, 37), particularly because diabetes in the BBDP rat is a Th1-mediated autoimmune disease (38–40). Notwithstanding, Hessner’s group (37) examined the role of pancreatic Ccl11 expression in T1D pathogenesis in DRlyp/lyp rats. Their observation that therapeutic inhibition of mast cell degranulation delayed the onset of disease in these animals provided circumstantial evidence that elevated expression of pancreatic Ccl11 may be pathogenic (37).

In this study, we have tested the hypothesis that pancreatic overexpression of Ccl11 is genetically determined and contributes to the regulation of the diabetogenic process in the BBDP rat. We observed a differential pancreatic expression of Ccl11 between T1D-susceptible BBDP and T1D-resistant ACI.1u.lyp rat strains. Specifically, Ccl11 transcript levels were upregulated in a time-dependent fashion in the BBDP rat pancreas, and this upregulation was confirmed to be Iddm2 independent. Through linkage analysis of a cross-intercross of BBDP and ACI.1u.lyp strains, we mapped this quantitative trait to a region of chromosome 12 that overlaps with a previously identified locus, Iddm30 (2). These linkage results were confirmed through the development and subsequent phenotypic analysis of a BBDP strain congenic for the chromosome 12 interval that controls T1D, insulitis, and Ccl11 expression in pancreatic islets. Importantly, our data show that whereas the BBDP allele at Iddm30 is associated with the expected higher levels of pancreatic Ccl11, this allele confers resistance to T1D. Further analysis of islet-infiltrating T cells in Iddm30 congenic BBDP animals revealed an increased proportion of IL-4–synthesizing CD4+ T cells in rats homozygous for the BBDP allele. Taken together, our results strongly suggest that the elevated recruitment of these Th2 cells was driven by the pancreatic overexpression of Ccl11 and resulted in a shift of the intraislet Th1/Th2 balance toward Th2 that may have contributed to diabetes resistance.

Materials and Methods

Animals

The parental inbred BBDP and ACI.1u.lyp rat strains and the congenic lines derived from them have been maintained at Sunnybrook Research Institute under specific pathogen-free conditions. Inbred BBDP rats originally purchased from Biomedical Research Models (Worcester, MA) have been maintained for >30 generations of brother × sister breeding and are homozygous at every genetic marker tested across the whole genome. Nonlymphopenic, diabetes-resistant BB.ACI-Iddm2 rats (BB.NonLyp) congenic for wild-type Gimap5 have been previously described (15). The inbred ACI.1u and ACI.1u.lyp strains were derived from the August Copenhagen Irish (ACI)/Hsd rats (Harlan Sprague Dawley, Indianapolis, IN) through introgression of the BBDP RT1u (Iddm1) locus alone or in combination with the BBDP Gimap5 (Lyp) locus (Iddm2) (5). To generate the (BBDP × ACI.1u.lyp)F2 cohort, BBDP males were first crossed with ACI.1u.lyp females to generate F1 progenies that were then intercrossed for linkage analysis. The results of this linkage analysis for type 1 diabetes, severity of insulitis, and age of disease onset have been reported previously (2).

Development of congenic inbred lines of BBDP rats

For generation of the congenic inbred BBDP.ACI-Iddm30 (BB.ACI-Iddm30) line, a ∼36-Mb interval of ACI.1u.lyp chromosome 12 that encompasses the 1 log of odds (LOD) interval of Iddm30 as defined by F2 linkage analysis (2) was introgressed into the BBDP background. Specifically, BBDP and ACI.1u.lyp rats were first intercrossed, and this was followed by 10 successive, marker-assisted backcrosses to the BBDP strain. Progenies of the final backcross, which were heterozygous at Iddm30, were then intercrossed to fix the region as ACI homozygous. Diabetes-prone animals were tested for glycosuria and ketonuria three times per week after 60 d of age. Animals found positive were then tested for hyperglycemia and considered as diabetic if blood glucose was >16.6 mM. All animal protocols were approved by the Sunnybrook Animal Care Committee.

Genotyping and linkage analysis

Genotyping and linkage analysis were performed, as previously described (2). Briefly, diabetic F2 animals were genotyped across the genome using 229 microsatellite markers at an average distance of 12.5 Mb apart. Microsatellite genotyping was done using PCR amplification, and PCR products were separated and analyzed on 3% agarose gel for allelic determination. Physical locations and primer information of microsatellite markers were obtained from the Rat Genome Database (http://rgd.mcw.edu/). Segregation analysis of T1D based on comparison of the observed to expected Mendelian inheritance genotype distribution at each marker was then performed, and markers linked to diabetes with a LOD score >1 were genotyped in all nondiabetic animals. Further refinement of linkage was obtained through genotyping of additional markers flanking these peaks of linkage, followed by interval mapping analysis.

RNA extraction from pancreatic tissue and quantitative RT-PCR

Total RNA was isolated from carefully dissected and immediately homogenized rat pancreas using GenElute mammalian total RNA miniprep kit (Sigma-Aldrich, St. Louis, MO). RNA samples were treated with DNase I (Life Technologies, Green Island, NY) to eliminate genomic DNA contamination prior to first-strand cDNA synthesis using oligo-dT primer (Fisher Scientific, Pittsburgh, PA) and Superscript III (Life Technologies), according to manufacturer’s instructions. Quantitative real-time PCR (qRT-PCR) was performed using an ABI Prism 7000 sequence detection system (Life Technologies). PCR consisted of iTaq SYBR Green supermix with ROX (Bio-Rad, Hercules, CA), β-actin primers (forward, 5′-CCC TAA GGC CAA CCG TGA A-3′; reverse, 5′-ACC AGA GGC ATA CAG GGA CAA-3′; Sigma Genosys) as quantification standard, and Ccl11 primers (forward, 5′-CAG CTC TCC ACA GCA CTT CT-3′; reverse, 5′-CTG GTC ATG GTA AAG CAG CA-3′; Sigma Genosys). Triplicate reactions were done for each sample in 20 μL reactions. Specificity for qRT-PCR amplification was verified by melting-curve analysis and through detection of a single PCR product on 3% agarose gel. Data were analyzed on ABI Prism 7000 software using the Δ cycle threshold (dCT) method to calculate the relative transcript copy number of Ccl11 to β-actin. Unless otherwise stated, t test was used for all statistical analyses and a p value <0.05 was considered as significant.

Immunofluorescent staining of pancreatic sections

Two-color immunofluorescent staining of pancreatic sections for rat Ccl11 and pancreatic hormones used goat polyclonal anti-Ccl11 (R&D Systems, Minneapolis, MN) and mouse monoclonal anti-insulin or anti-glucagon Abs (Sigma-Aldrich), and was performed as described previously (37). In brief, freshly excised pancreata were fixed in 10% phosphate-buffered formalin (Cedarlane Laboratories, Burlington, ON, Canada) at 4°C overnight. The following day, tissues were dehydrated by serial incubations in 10, 20, and 30% sucrose solutions. Samples were then embedded in optimal cutting temperature medium (Electron Microscopy Sciences, Hatfield, PA) and snap frozen. Two 4-μm sections cut 250 μm apart were prepared from each pancreas and mounted onto l-lysine–coated microscope slides. Pancreatic sections were first subjected to Ag retrieval using citrate-based Ag retrieval solution (Dako, Carpinteria, CA) at 95°C for 10 min, then rinsed in Tris buffer (pH 7.6), followed by blocking with 5% normal donkey serum (Jackson ImmunoResearch Laboratories, West Grove, PA). Sections were stained overnight in a humidity chamber with appropriate dilutions of primary Ab combinations. Sections were then rinsed and stained with the appropriate secondary Abs, Texas Red–conjugated donkey anti-goat IgG and FITC-conjugated donkey anti-mouse IgG (Jackson ImmunoResearch Laboratories), for 1 h at room temperature. Slides were washed, dried, and mounted using fluorescence mounting medium (Dako). Slide analysis was done on a Zeiss Axiovert inverted fluorescent microscope using Apotome optical sectioning imaging system (Zeiss, Thornwood, NY).

Assessment of Ccl11 protein levels in isolated pancreatic islets by immunoprecipitation and Western blot

Pancreatic islets were isolated from age-matched, nonlymphopenic, and T1D-resistant BB.NonLyp and ACI.1u, as previously described (41). In brief, whole pancreas was distended by collagenase D (Sigma-Aldrich) and digested to homogeneity at 37°C. Digestion was stopped by adding ice-cold RPMI 1640 medium containing 10% FBS (Sigma-Aldrich). Total tissue suspension was then filtered through a 400-μm nylon filter and resuspended in ice-cold fresh RPMI 1640. The dissociated islets were handpicked under a dissection microscope. To maximize the purity of the islet preparation, the handpicking was repeated two more times and the final purity was verified by staining a portion of the isolated islets with 0.01% w/v dithizone (Sigma-Aldrich) in 10 mM HEPES-buffered PBS solution (42). Collected islets were then lysed in 1% Nonidet P-40 (Roche, Indianapolis, IN) lysis buffer with Complete Protease Inhibitor (Roche), and the total protein concentration was determined by bicinchoninic acid assay (Life Technologies). Equal amounts of islet-derived protein extract were incubated with protein A–Sepharose beads (Sigma-Aldrich) precoated with rabbit polyclonal anti-rat Ccl11 Ab (Peprotech, Rocky Hill, NJ) for immunoprecipitation (IP) at 4°C overnight. To determine whether the original lysate had been depleted of Ccl11 during the first IP, supernatant from the first IP was subjected to a second overnight IP under identical conditions. Immunoprecipitates were separated by SDS-PAGE, transferred onto a nitrocellulose membrane, and then blotted with goat polyclonal anti-rat Ccl11 Ab (R&D Systems) overnight at 4°C. The following day, a HRP-conjugated secondary donkey anti-goat Ab was added and proteins were visualized using enhanced chemiluminescence (Life Technologies). To establish the semiquantitative nature of the IP-Western blot assay for Ccl11 in ex vivo islets, we performed the same assay on serial dilutions of lysates from 293T cells transfected with rat Ccl11 as positive controls.

Isolation of islet-infiltrating mononuclear cells, intracellular staining, and FACS analysis

Pancreatic islets were isolated from nondiabetic, age-matched, congenic inbred BBDP and BB.ACI-Iddm30 animals, as described above. The isolated islets were washed once in PBS with 1 mM EDTA and then digested in 0.25% Trypsin-EDTA (Life Technologies) at 37°C for 15 min. Digestion was stopped by adding ice-cold IMDM medium containing 10% FBS, and mononuclear cells were then isolated using a discontinuous Percoll gradient (GE Life Sciences, Pittsburgh, PA). Mononuclear cells were cultured in IMDM containing 10% FBS in the presence of 25 ng/ml PMA, 250 ng/ml ionomycin (Sigma-Aldrich), and brefeldin A (eBioscience, San Diego, CA) for 3 h. Activated mononuclear cells were then stained with CD4-specific, allophycocyanin-conjugated, W3/25 mAb (Sunnybrook mAb core facility, Toronto, ON, Canada), permeabilized using BD Cytofix/Perm kit, and stained for intracellular cytokines using IFN-γ–specific, FITC-conjugated, DB-1 mAb (BioLegend, San Diego, CA) and IL-4–specific, PE-conjugated, OX81 mAb (BioLegend). Stained cells were analyzed by flow cytometry using a FACSCalibur cell analyzer (BD Biosciences), and the proportions of IFN-γ+ and IL-4+ CD4+ T cells were determined using FlowJo software.

Results

We sought to determine whether, similar to the observation made between diabetes-prone DR and diabetes-resistant WF strains, Ccl11 was differentially expressed in the pancreatic islets of diabetes-prone BBDP and diabetes-resistant ACI.1u.lyp animals. If this were the case, we reasoned that we would be able to take advantage of an ongoing cross-intercross between these two parental strains for concurrent linkage analysis of this trait and T1D, whereby allowing us to determine whether pancreatic Ccl11 expression is under genetic control and linked to diabetes.

Overexpression of Ccl11 in the pancreatic β cells of BBDP rats

Using two-color immunofluorescence for Ccl11 and pancreatic hormones, we observed that Ccl11 was detectable in the β cells of virtually all islets from 90-d-old BBDP rats (n = 10). A representative pattern of immunofluorescence is illustrated in Fig. 1A and 1B. However, a similar pattern of expression was observed in age-matched ACI.1u.lyp animals (n = 10) as well as in the nonlymphopenic, Gimap5 (Iddm2) congenic ACI.1u and BB.NonLyp strains (n = 8∼10; data not shown). These results suggested that this trait was shared among the BBDR, BBDP, and ACI strains in an Iddm2-independent manner. To confirm this apparent phenotypic similarity in a more quantitative fashion, we assessed the levels of Ccl11 in lysates of adult pancreatic islets. To avoid any potential influence of ongoing islet inflammation in T1D-prone BBDP rats, this evaluation was performed in islets from nonlymphopenic, age-matched, adult BB.NonLyp and ACI.1u donors that are both T1D and insulitis resistant. The low levels of Ccl11 protein expression required enrichment of the islet extracts by specific IP prior to quantification by Western blot. Validation of this quantitative evaluation was obtained by submitting serial dilutions of lysates from 293T cells transfected with rat Ccl11 to the same assay, as illustrated in Fig. 1C. Furthermore, incomplete depletion of Ccl11 by IP of the original lysates was ruled out by subjecting the supernatants from the first round of IP to an additional round of IP and Western blot for Ccl11 (Fig. 1D). Densitometry consistently showed a ∼2-fold increase in Ccl11 content of pancreatic islets from BB.NonLyp rats compared with their ACI.1u counterparts (Fig. 1D). This differential expression of Ccl11 in the β cells of strains other than DR and WF further supports the notion that this trait is genetically determined. Furthermore, its overexpression in diabetes-prone DR and BBDP strains compared with diabetes-resistant WF and ACI animals, respectively, raises the possibility that its regulation plays a role in disease pathogenesis.

FIGURE 1.
FIGURE 1.

Ccl11 is synthesized by rat pancreatic β cells. Ccl11 protein expression was assessed by two-color immunofluorescence of pancreatic sections (A and B) or IP and Western blot analysis of protein lysates from isolated pancreatic islets (D). (A and B) Representative staining of frozen pancreatic sections from adult BBDP rats with anti-Ccl11 (Texas Red) and anti-insulin [FITC; green (A)] or anti-glucagon [FITC; green (B)] Abs. Images of the individual color channels (red or green) are shown on the left, followed by a merge of the two channels on the right. All images were acquired using Zeiss Apotome optical sectioning system under a 20× dry objective lens (original magnification ×200). (C) Semiquantitative assessment of Ccl11 in 2-fold serial dilutions of lysates from 293T cells transfected with rat Ccl11 by IP and Western blot. Densitometry of the blots was performed and showed a linear relationship between the OD and dilution factor. (D) Representative pattern of IP and Western blot for Ccl11 from pancreatic islet lysates of age-matched adult BB.NonLyp (BB) and ACI.1u (ACI) rats. Protein contents of islet lysates were normalized by bicinchoninic acid assay prior to IP, and a protein lysate from 293T cells transfected with rat Ccl11 (Ccl11) was used as positive control. Densitometry revealed a ∼2-fold increase in Ccl11 levels in BB.NonLyp islet lysates compared with that of the ACI.1u animals. Following IP, islet lysate supernatants were subjected to a second round of IP and Western blot to ensure that all Ccl11 had been captured during the first round. Data are representative of four independent experiments.

Quantitative assessment of islet Ccl11 expression by IP and Western blot is not applicable to a large cohort of animals required for genetic analysis. To circumvent this limitation, we sought to determine whether Ccl11 protein levels in pancreatic islets correlated with those of gene transcription from the whole pancreas using qRT-PCR analysis, a technique that would be amenable to higher throughput for linkage analysis. As illustrated in Fig. 2A, there was a ∼6-fold increase in pancreatic transcript levels of Ccl11 in adult BBDP rats when compared with age-matched ACI.1u.lyp animals (p = 0.0002). This difference was similar to that observed between the pancreata of adult BB.NonLyp and ACI.1u donors (p = 0.0004). Meanwhile, no difference was found between Iddm2 (Gimap5) congenic BBDP and BB.NonLyp or ACI.1u.lyp and ACI.1u animals, confirming that this quantitative trait is not under the control of the Iddm2 locus. These differential transcript levels of Ccl11 also appear to be tissue specific, as Ccl11 transcripts were below the qRT-PCR detection threshold in liver and gut extracts collected from either strains. Next, we sought to determine whether this differential regulation of Ccl11 transcription between BBDP and ACI.1u.lyp rats was age dependent. As shown in Fig. 2B, significantly higher transcript levels of Ccl11 were found in the pancreas of BBDP donors as early as 30 d of age (p = 0.01), hence long before the onset of islet inflammation in these animals, and this difference between the two strains grew progressively larger over time. Having demonstrated that pancreatic expression of Ccl11 is differentially regulated between T1D-prone BBDP and T1D-resistant ACI.1u.lyp animals in a time-dependent fashion, we then sought to determine whether this quantitative trait is genetically determined through linkage analysis of a previously described F2 cohort that was derived from these two parental strains to map novel Iddm loci (2).

FIGURE 2.
FIGURE 2.

Increased Ccl11 mRNA levels in the pancreas of BBDP rats compared with ACI.1u.lyp animals. Ccl11 mRNA levels were assessed by qRT-PCR. The relative copy number of Ccl11 mRNA in each sample was normalized to the copy number of β-actin transcripts ×103 as determined by the dCT method. (A) Comparison of pancreatic Ccl11 transcript levels in 90-d-old BBDP, ACI.1u.lyp, and their respective Iddm2 (lyp) congenic counterparts, BB.NonLyp and ACI.1u animals (individual values of each animal are shown; bars represent mean ± SEM; statistical significance was determined by t test, and significant p values (p < 0.05) are indicated unless no significance [n.s.] in difference was found). (B) Time-dependent upregulation of Ccl11 transcript levels in the pancreas of age-matched BBDP and ACI.1u.lyp rats. Four to twenty-three animals of each strain were analyzed at each time point (statistical significance was determined by t test; **p ≤ 0.01, ***p ≤ 0.001).

Pancreatic transcript levels of Ccl11 are linked to Iddm30

The assessment of Ccl11 transcript levels in the pancreas of (BBDP × ACI.1u.lyp)F2 animals was initiated when the development of this cohort was well underway. F2 animals (n = 574) were followed prospectively until the onset of T1D (n = 127) or up to 165 d for nondiabetic animals. Of the 447 animals that remained diabetes free, we assessed the pancreatic levels of Ccl11 transcripts in the last 117 rats followed. There were two reasons for excluding diabetic rats from our analysis. First, they would not be age matched to nondiabetic rats, and this age difference would be a confounding factor because our data showed that age has a profound impact on the pancreatic levels of Ccl11 transcripts (Fig. 2B). Second, we reasoned that the destruction of β cells by the diabetogenic process would remove the main source of Ccl11 expression in the pancreas.

The (BBDP × ACI.1u.lyp)F2 animals assessed for pancreatic Ccl11 transcript levels were genotyped across the genome using 229 microsatellite markers at an average of one marker every 12.5 Mb, as previously described (2). This was followed by linkage analysis to determine which chromosomal region(s) controls this quantitative trait. Only one region, on chromosome 12, showed significant linkage. The peak of linkage (LOD = 3.65) is located at marker D12Rat51, and the 1-LOD interval of linkage spans 36.4 cM between markers D12Rat28 and D12Rat52 (Fig. 3A). This locus alone accounts for 12.2% of the trait variance observed in this cross by ANOVA. As expected based on parental phenotypes, when the Ccl11 transcript levels of the 117 F2 rats were compared based on the BBDP or ACI genotype of these animals at the peak of linkage (Fig. 3B), those that were BBDP homozygotes had significantly higher levels than ACI homozygotes (p = 0.0006) and heterozygotes (p = 0.0007). In contrast, there was no significant difference in Ccl11 transcription levels between animals that were ACI homozygous and heterozygotes (Fig. 3B), indicating a recessive mode of inheritance for this trait.

FIGURE 3.
FIGURE 3.

Pancreatic transcript levels of Ccl11 are linked to a chromosome 12 region that also controls T1D and severity of insulitis in a cohort of (BBDP × ACI.1u.lyp)F2 animals. (A) Linkage plots for T1D (solid line), insulitis (dashed line), and pancreatic Ccl11 transcript levels (dotted line) on chromosome 12. The peaks of linkage for T1D (LOD = 3.76) and insulitis (LOD = 6.44) are located at the same marker D12Rat28 that is only 2.96 cM away from the peak of linkage for pancreatic transcript levels of Ccl11 (LOD = 3.65) at marker D12Rat51. Chromosomal locations are indicated in cM along the x-axis. (B and C) Influence of the genotype (ACI allele = A; BBDP allele = B) of F2 animals at marker D12Rat51 on pancreatic transcript levels of Ccl11 and T1D, respectively. (B) Individual values of each animal are shown; bars represent mean ± SEM; statistical significance was determined by t test, and significant p values (p < 0.05) are indicated unless no significance (n.s.) in difference was found. (C) Cumulative survival of T1D-free F2 animals according to their genotype at D12Rat51 (statistical analyses were performed by Kaplan–Meier survival analysis, and significant p values are indicated).

Strikingly, the chromosome 12 locus influencing pancreatic transcript levels of Ccl11 overlaps with the Iddm30 locus that was found to control the severity of insulitis and T1D susceptibility in the whole cohort of 574 (BBDP × ACI.1u.lyp)F2 animals (Fig. 3A) (2). The 117 nondiabetic animals assessed for transcript levels of Ccl11 in the pancreas were derived from the last 167 members of the whole (BBDP × ACI.1u.lyp)F2 cohort, of which 45 (26.9%) became diabetic. The original report that Ccl11, a chemotactic factor for mast cells, was overexpressed in the pancreatic islets of T1D-prone DR rats compared with WF animals led to the hypothesis that this trait was diabetogenic (24), and this hypothesis was further supported by the observation by the same group that inhibition of mast cell degranulation slowed down the diabetogenic process (37). In this context, it was important to determine whether at D12Rat51, where the BBDP genotype is associated with high levels of Ccl11 pancreatic transcripts in the (BBDP × ACI.1u.lyp)F2 cohort, the same genotype would also confer a high risk for T1D. As shown in Fig. 3C, this was not the case. Specifically, F2 animals that were BBDP homozygous at D12Rat51 exhibited a profound resistance to T1D with a cumulative incidence of only 8.5% by 165 d compared with 30.3 and 40% observed in heterozygotes and ACI homozygotes, respectively (p = 0.006 and p = 0.0004, respectively, by Kaplan–Meier survival analysis; Fig. 3C). Of note, similar to the levels of Ccl11 expression, the susceptibility to T1D controlled by the chromosome 12 locus appears to be inherited as a recessive trait. Taken together, these results indicate that the regulation of Ccl11 expression in the pancreas is under the control of a chromosome 12 locus that overlaps to a large extent with Iddm30, and the high levels of pancreatic Ccl11 expression are associated with protection from T1D.

The phenotypic analysis of BBDP rats congenic for Iddm30 confirms linkage of T1D and pancreatic Ccl11 expression to this locus

To confirm the results of linkage analyses, an inbred BBDP Iddm30 congenic line named BB.ACI-Iddm30 was developed. Specifically, a <36-Mb interval of chromosome 12 of ACI.1u.lyp origin that encompasses the Iddm30 locus was introgressed into the BBDP genetic background (Supplemental Fig. 1). This congenic line was then followed prospectively for T1D and phenotypically assessed for transcript levels of Ccl11 in the pancreas.

When litters of BBDP and BB.ACI-Iddm30 animals were concomitantly followed for T1D development (Fig. 4A), the cumulative incidence of disease by 140 d was significantly higher in BB.ACI-Iddm30 rats (92%, n = 25) than in BBDP animals (75%, n = 36; p = 0.045 by one-tailed χ2 test). Of note, the age of disease onset was not significantly different between the two Iddm30 congenic strains (78 ± 9.9 d in BBDP versus 82 ± 8.9 d in BB.ACI-Iddm30, mean ± 1 SD; p = 0.4 by Kaplan–Meier survival analysis). These results confirm that homozygosity for the BBDP allele at Iddm30 confers significant resistance to T1D.

FIGURE 4.
FIGURE 4.

Phenotypic analysis of inbred Iddm30 congenic BBDP animals confirms linkage of T1D and pancreatic Ccl11 expression to rat chromosome 12. (A) Cumulative survival of T1D-free BBDP and BB.ACI-Iddm30 animals. Survival of BBDP animals (25%, n = 36) is shown by the solid line, whereas survival of the BB.ACI-Iddm30 animals (8%, n = 25) is shown by the dashed line (statistical analysis of significant difference in survival was performed by χ2 test). There was no difference in age of onset between the two strains (median of 83 and 79 d for BBDP and BB.ACI-Iddm30, respectively). (B) Comparison of pancreatic transcript levels of Ccl11 in 90-d-old BBDP, ACI.1u.lyp, and BB.ACI-Iddm30 animals as assessed by qRT-PCR analysis. (C) Time-dependent upregulation of pancreatic Ccl11 transcript levels among these three strains. There was no significant difference in Ccl11 transcript levels between age-matched ACI.1u.lyp and BB.ACI-Iddm30 rats, whereas levels in BBDP were significantly higher than in the two other strains at each age examined. Bars indicate mean transcript levels ± SEM; statistical analyses were performed by t test, and significant p values (p < 0.05) are indicated unless no significance (n.s.) in difference was found.

We then asked whether the increased susceptibility of BB.ACI-Iddm30 animals to T1D was associated with low levels of Ccl11 transcripts in their pancreas. This parameter was assessed in nondiabetic, age-matched ACI.1u.lyp and BBDP Iddm30 congenic animals. As illustrated in Fig. 4B, the levels of Ccl11 transcription were significantly higher in the pancreas of BBDP rats (456.5 ± 356.7/103 β-actin copies, mean ± 1 SD, n = 19) than in their Iddm30 congenic counterparts (116.3 ± 136.3/103 β-actin copies, mean ± 1 SD, n = 5; p = 0.011) at 90 d of age, whereas expression levels were similar between age-matched ACI.1u.lyp and BB.ACI-Iddm30 animals. Similar to what was observed in the parental strains, the pancreatic levels of Ccl11 expression in BB.ACI-Iddm30 animals increased with age (Fig. 4C). However, although these levels increased by almost an order of magnitude in BBDP rats between the ages of 90 and 150 d (456.5 ± 356.7/103 β-actin copies and 4286 ± 2633/103 β-actin copies, mean ± 1 SD, respectively), the corresponding elevation in ACI.1u.lyp and BB.ACI-Iddm30 animals was more modest, which explains the significant differences observed between BBDP rats and the two other strains at 150 d (353.7 ± 628/103 β-actin copies, mean ± 1 SD, n = 5 in ACI.1u.lyp, p = 0.0087 versus BBDP; 564.5 ± 513.9/103 β-actin copies, n = 9 in BB.ACI-Iddm30 rats, p = 0.012 versus BBDP). Thus, the analysis of Iddm30 congenic inbred animals confirms the linkage results whereby homozygosity for the BBDP allele at Iddm30 is associated with both resistance to T1D and elevated levels of Ccl11 expression in the pancreas.

Increased levels of Ccl11 expression in the pancreas are associated with a Th2 bias among CD4+ T cells present in insulitic lesions

T1D is considered as a prototypical Th1-mediated disease (38–40), and it has been demonstrated that Th2 skewing of the antipancreatic autoimmune response in diabetes-prone animals can result in disease protection (19, 43, 44). Increased expression of Ccl11 typically contributes to the recruitment of Th2 cells that express Ccr3, the chemokine receptor for Ccl11, to the site of inflammation (28, 32, 34, 45). The observation that high levels of pancreatic Ccl11 transcripts are associated with resistance to T1D in BBDP rats led us to determine whether these high levels are also correlated with a Th2 bias in the insulitic lesion of these animals. As illustrated in Fig. 5A, the transcript levels of the Th2 transcription factor Gata3 were significantly higher in the pancreas of 90-d-old BBDP rats (15.1 ± 3.6/103 β-actin copies, mean ± 1 SD) when compared with age-matched BB.ACI-Iddm30 animals (8.5 ± 2.9/103 β-actin copies, mean ± 1 SD; p = 0.0087). Next, we isolated the lymphocytes infiltrating the pancreatic islets of these animals to directly assess the proportion of CD4+ T cells synthesizing IFN-γ and IL-4 by intracytoplasmic staining and flow cytometry. The proportions of CD4+ T cells among the mononuclear cells isolated from the islets of nondiabetic, 90-d-old BBDP (n = 6) and BB.ACI-Iddm30 (n = 6) animals were not significantly different (26.8 ± 17.4% and 18.0 ± 10.8%, mean ± 1 SD, respectively; p = 0.3). Among these cells, the proportion of those synthesizing IFN-γ was also similar (26.2 ± 3.9%, mean ± 1 SD in BBDP versus 29.7 ± 6.5% in BB.ACI-Iddm30 rats; p = 0.5). In contrast, the proportion of CD4+ T cells synthesizing IL-4 among insulitic cells (Fig. 5B) recovered from BBDP animals (30.5 ± 8.4%, mean ± 1 SD) was significantly increased when compared with that from BB.ACI-Iddm30 donors (19.6 ± 5.5%, mean ± 1 SD; p = 0.02). Of note, histological analysis of pancreatic sections from diabetic and nondiabetic Iddm30 congenic BBDP animals revealed very low to undetectable numbers of mast cells within the pancreatic islets of both strains (data not shown). Taken together, our results demonstrate that homozygosity for the BBDP allele of Iddm30 confers resistance to T1D and is associated with an increase in both pancreatic expression of Ccl11 and differentiation of insulitic CD4+ T cells toward the Th2 lineage.

FIGURE 5.
FIGURE 5.

Homozygosity for the BBDP allele at Iddm30 is associated with an enrichment in Th2 cells among islet-infiltrating T cells. (A) The relative mRNA copy number of Th2 differentiation transcription factor, Gata3, was normalized to the copy number of β-actin transcripts ×103 as determined by the dCT method in the pancreas of 90-d-old, nondiabetic Iddm30 congenic BBDP rats. (B) Islet-infiltrating mononuclear cells were isolated from pancreatic islets of 90-d-old inbred Iddm30 congenic BBDP animals and assessed for proportion of CD4+ T cells that synthesized IL-4 by multicolor immunofluorescence and FACS analysis. Individual values of each sample are shown, and bars represent mean ± SEM. Statistical analyses comparing Iddm30 congenic BBDP animals with the BBDP parental strain were performed by t test, and significant p values (p < 0.05) are indicated.

Discussion

Although there has been a plethora of studies characterizing alterations of the immune system that contribute to the loss of tolerance to pancreatic β cells in diabetes-prone individuals, very few have described intrinsic abnormalities of these endocrine cells that may play a role in the pathogenesis of the diabetogenic process. These include the upregulation of type I IFN and MHC class I expression in the β cells of diabetes-prone rodents prior to the detection of islet infiltration by myeloid and lymphoid cells (46, 47), and the evidence that the intense remodeling of pancreatic islets taking place around the time of weaning is dysregulated in these same animals (48). However, to the best of our knowledge, there is no evidence that these β cell abnormalities are genetically determined nor can one rule out the possibility that some of them are the consequence of the autoimmune process. In contrast, the time-dependent expression of Ccl11 in the β cells of diabetes-prone DR rats reported by Hessner et al. (24) appeared genetically determined and unlikely to result from islet inflammation. Furthermore, the same group of investigators showed that increased pancreatic expression of the chemokine Ccl11 in DRlyp/lyp animals was associated with an elevated number of activated mast cells in pancreatic lymph nodes that may have contributed to the acceleration of the diabetogenic process in these animals (37).

In the current study, we tested the hypothesis that pancreatic overexpression of Ccl11 is genetically determined and contributes to the regulation of the diabetogenic process in the BBDP rat. Through genetic linkage analysis and the development of a novel congenic inbred line of BBDP rats, we demonstrate that overexpression of Ccl11 is under the control of a region of rat chromosome 12 that overlaps with one of the previously mapped T1D loci in the BBDP rat, Iddm30 (2). However, we also provide evidence that this trait is associated not with increased susceptibility but rather with resistance to T1D and an increase in Th2 differentiation among the islet-infiltrating T lymphocytes.

The DRlyp/lyp strain is genetically related to, yet exhibits genetic polymorphism with the inbred BBDP strain (17). Despite this polymorphism, it has been demonstrated that these two strains share the same known susceptibility loci for spontaneous T1D (3, 14). Assuming that pancreatic overexpression of Ccl11 is genetically determined, the question followed as to whether this trait was controlled by one or more of the Iddm loci. If it were, the expectation was that we should be able to observe it in BBDP animals. It was, however, unclear whether pancreatic expression of Ccl11 would differ between BBDP and the T1D-resistant strain, ACI.1u.lyp, which we have previously used in genetic analyses.

Immunofluorescence analysis detected a similar expression of Ccl11 in the β cells of BBDP and age-matched ACI.1u.lyp animals. However, the use of qRT-PCR and IP followed by Western blot both showed that this expression was significantly higher in the BBDP strain at all time points analyzed, thus further supporting the notion that it is genetically determined. This differential expression between the BBDP and ACI.1u.lyp strains was also observed in their Gimap5 (Iddm2) congenic counterparts, thereby confirming that pancreatic expression of Ccl11 is neither controlled by this locus nor secondary to islet inflammation. Analysis of a cohort of age-matched, nondiabetic (BBDP × ACI.1u.lyp)F2 animals showed significant linkage of pancreatic expression of Ccl11 only to a chromosome 12 region, which overlapped with the Iddm30 locus previously shown to control T1D susceptibility and the severity of insulitis (2). As expected from the phenotype of the parental strains, the BBDP allele at Iddm30 imparts a significant increase in pancreatic Ccl11 transcription levels compared with the ACI allele. Because at the peak of linkage levels of Ccl11 transcripts are similar in F2 animals that are heterozygotes and ACI homozygotes, the inheritance of this trait appears to be recessive at this locus. Strikingly, however, whereas results of the linkage analysis establish the genetic control of this trait, they also show that, at the peak of linkage, the BBDP genotype that is associated with pancreatic overexpression of Ccl11 confers recessive resistance to T1D and insulitis (2). Although at this stage it is premature to consider that Iddm30 and the chromosome 12 locus controlling the levels of Ccl11 expression are allelic, this observation suggests that pancreatic overexpression of Ccl11 is protective against T1D.

Linkage of pancreatic Ccl11 expression to chromosome 12 (LOD = 3.7) is relatively weaker than that of T1D (LOD = 3.8) and insulitis (LOD = 6.4). However, it is important to note that Ccl11 expression was only assessed in 117 F2 rats, a number considerably smaller than that (n = 574) used for the mapping of the two other traits (2). Furthermore, the proportion of the Ccl11 trait variance accounted for by this locus, 12.2%, is much higher than that previously reported for T1D (3.1%) and insulitis (5.1%), reflecting the strong effect of this locus on pancreatic expression of Ccl11 (2). Although one could argue that the linkages of Ccl11 expression, T1D, and insulitis to chromosome 12 are not perfectly overlapping, one should consider two explanations that may have led to a less accurate mapping of the Ccl11 expression trait. One is the relatively low number of animals assessed for Ccl11 expression, and the second is the way these animals were selected. Specifically, to avoid the confounding effects of age on pancreatic expression of Ccl11, we restricted the assessment of this trait to F2 animals that had remained T1D free until the end of the follow-up period. This selection had two consequences. First, this nondiabetic cohort (n = 117) was enriched in the proportion of F2 animals that were homozygous for the BBDP T1D resistance allele at the peak of Iddm30 (34.8%) when compared with the whole cohort of F2 rats (n = 574), in which the expected Mendelian proportion of BBDP homozygous animals was observed (25.3%). Second, it was most likely that by selecting for nondiabetic F2 animals, we enriched for high expressers of pancreatic Ccl11 among those that were heterozygous and ACI homozygous at the peak of linkage on chromosome 12, thereby minimizing the differential Ccl11 expression between the two genotypes.

The one LOD interval surrounding the peak of linkage of pancreatic Ccl11 (Eotaxin) expression on chromosome 12 is large, 36.4 cM or 20.5 Mb, and therefore contains many candidate genes, including Eotaxin-2 (Ccl24) and Eotaxin-3 (Ccl26), that map in close proximity to the peak of linkage (http://www.t1dbase.org/) and share the same receptor Ccr3 with Ccl11. Given the genomic location of Ccl24 and Ccl26 and the evidence that the regulation of their expression is coordinated with that of Ccl11 in allergic and other Th2-mediated immune responses (32, 34, 45), we sequenced the coding regions of these two genes and assessed their transcript levels in the pancreas of BBDP and ACI.1u.lyp parental strains. Of note, Ccl11 itself is located on chromosome 10, and it was therefore plausible that a chromosome 12 genetic polymorphism in our cross would differentially regulate the expression of Ccl11 in trans and that of the two other eotaxins in cis. However, whereas we detected a nonsynonymous single-nucleotide polymorphism in exon 1 of Ccl24 at position 19 in which a G in BBDP to A in ACI substitution resulted in a conservative V to I substitution at aa 7, there was no evidence for differential expression of Ccl24 and 26 in the pancreas of the two parental strains (data not shown).

Constitutive expression of various cytokines by pancreatic islets has been observed in rodents and humans (25, 26), but its physiological significance remains unknown. However, it has been demonstrated in vitro that interaction of β cells with extracellular matrix proteins results in the induction of chemokine expression in these cells through NF-κB activation (27, 49). This activation of NF-κB is sustained through an IL-1–mediated autocrine loop and appears to be important for optimal insulin secretion and β cell survival (27, 49). In this context, analysis of the promoter region of Ccl11 reveals the presence of a consensus NF-κB binding site (−68 to −59 bp), and NF-κB and STAT6 activation have been shown to induce Ccl11 transcription in airway epithelial cells (50). It is therefore possible that the constitutive activation of NF-κB required for β cell function and survival results in the coordinated expression of cytokines by β cells. The question follows as to whether dysregulated expression of these inflammatory mediators in β cells could have an impact on antipancreatic autoimmunity. The functional analysis of the T cells infiltrating the pancreatic islets of Iddm30 congenic inbred BBDP animals suggests that it does.

To confirm the linkage of T1D and pancreatic expression of Ccl11 to chromosome 12, we developed Iddm30 congenic BBDP rats through the introgression of an ACI-derived, ∼36-Mb interval of chromosome 12 into the BBDP background. The significantly higher susceptibility of the resulting BB.ACI-Iddm30 animals to T1D compared with BBDP rats confirmed the influence of this locus on disease pathogenesis. Likewise, the observation that the levels of Ccl11 transcripts in the pancreas of adult BB.ACI-Iddm30 animals were similar to those of age-matched ACI.1u.lyp but significantly lower than those of BBDP animals confirmed the influence of the Iddm30 locus on the regulation of this trait.

The critical role of Ccl11 in driving the recruitment of Ccr3-expressing Th2 lymphocytes, mast cells, and eosinophils to the site of atopic and other Th2-mediated immune responses is well characterized (32, 34, 45, 51). Through this recruitment and the cytokines released by the attracted cells, Ccl11 contributes to the skewing of Th differentiation away from Th1 to Th2 (34). T1D is a prototypical Th1-driven autoimmune disease (38–40), and it has been shown that genetically determined or therapeutically induced shifts of the intraislet Th1/Th2 balance toward Th2 can prevent the disease in rodent models (19, 38, 43, 52). The differential expression of Ccl11 in the pancreas of T1D-resistant and T1D-prone rats led us to determine whether overexpression of this cytokine was associated with evidence of a Th2 skewing among the CD4+ T cells present in inflamed islets. The proportion of intraislet T cells and, among those, that of CD4+ cells synthesizing IFN-γ upon activation was similar between Iddm30 congenic BBDP animals. In contrast, there was an increased proportion of IL-4–synthesizing CD4+ T cells in the islets of animals homozygous for the BBDP allele at the Iddm30 locus shown in this work to be associated with high levels of pancreatic Ccl11 expression. Consistent with this Th2 shift among intraislet CD4+ T cells, we also detected a 2-fold increase in Gata3 transcript levels in the pancreas of nondiabetic BBDP rats compared with their age-matched BB.ACI-Iddm30 counterparts. Of note, we did not observe a differential recruitment of mast cells known to express Ccr3 to the pancreatic islets. In fact, we could hardly detect mast cells in the insulitic lesions of our diabetic and nondiabetic Iddm30 congenic BBDP animals. Taken together, results presented in this study suggest that the BBDP allele of Iddm30 confers resistance to destructive insulitis and diabetes and preferential recruitment of Th2 CD4+ T cells over Th1 cells as a consequence of increased Ccl11 synthesis by pancreatic β cells.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Dr. Martin J. Hessner for providing the protocols for immunofluorescent staining of pancreatic sections, Kirishanthy Kathirkamathamby from the Sunnybrook Antibody Core Facility, Gisele Knowles from the Sunnybrook Flow Cytometry Core Facility, Catherine Gibson from Sunnybrook Comparative Research for Excellent Animal Care, and Dr. Jayne Danska for a critical review of this manuscript.

Footnotes

  • This work was supported by grants from Genome Canada, the Canadian Institutes of Health Research, and the Juvenile Diabetes Research Foundation. G.Y.C.C. received a studentship from the Banting and Best Diabetes Institute.

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    ACI
    August Copenhagen Irish
    BBDP
    BioBreeding diabetes–prone
    dCT
    Δ cycle threshold
    IP
    immunoprecipitation
    LOD
    log of odds
    qRT-PCR
    quantitative real-time PCR
    T1D
    type 1 diabetes
    WF
    Wistar–Furth.

  • Received September 5, 2013.
  • Accepted February 18, 2014.

References

  1. ↵
    1. Mordes J. P.,
    2. R. Bortell,
    3. J. Doukas,
    4. M. Rigby,
    5. B. Whalen,
    6. D. Zipris,
    7. D. L. Greiner,
    8. A. A. Rossini
    . 1996. The BB/Wor rat and the balance hypothesis of autoimmunity. Diabetes Metab. Rev. 12: 103–109.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Wallis R. H.,
    2. K. Wang,
    3. L. Marandi,
    4. E. Hsieh,
    5. T. Ning,
    6. G. Y. Chao,
    7. J. Sarmiento,
    8. A. D. Paterson,
    9. P. Poussier
    . 2009. Type 1 diabetes in the BB rat: a polygenic disease. Diabetes 58: 1007–1017.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Jacob H. J.,
    2. A. Pettersson,
    3. D. Wilson,
    4. Y. Mao,
    5. A. Lernmark,
    6. E. S. Lander
    . 1992. Genetic dissection of autoimmune type I diabetes in the BB rat. Nat. Genet. 2: 56–60.
    OpenUrlCrossRefPubMed
    1. Like A. A.,
    2. C. A. Biron,
    3. E. J. Weringer,
    4. K. Byman,
    5. E. Sroczynski,
    6. D. L. Guberski
    . 1986. Prevention of diabetes in BioBreeding/Worcester rats with monoclonal antibodies that recognize T lymphocytes or natural killer cells. J. Exp. Med. 164: 1145–1159.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Colle E.,
    2. A. Fuks,
    3. P. Poussier,
    4. P. Edouard,
    5. R. D. Guttmann
    . 1992. Polygenic nature of spontaneous diabetes in the rat: permissive MHC haplotype and presence of the lymphopenic trait of the BB rat are not sufficient to produce susceptibility. Diabetes 41: 1617–1623.
    OpenUrlAbstract/FREE Full Text
    1. Edouard P.,
    2. J. C. Hiserodt,
    3. C. Plamondon,
    4. P. Poussier
    . 1993. CD8+ T-cells are required for adoptive transfer of the BB rat diabetic syndrome. Diabetes 42: 390–397.
    OpenUrlAbstract/FREE Full Text
  7. ↵
    1. Métroz-Dayer M. D.,
    2. A. Mouland,
    3. C. Brideau,
    4. D. Duhamel,
    5. P. Poussier
    . 1990. Adoptive transfer of diabetes in BB rats induced by CD4 T lymphocytes. Diabetes 39: 928–932.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Chao N. J.,
    2. L. Timmerman,
    3. H. O. McDevitt,
    4. C. O. Jacob
    . 1989. Molecular characterization of MHC class II antigens (beta 1 domain) in the BB diabetes-prone and -resistant rat. Immunogenetics 29: 231–234.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Colle E.,
    2. R. D. Guttmann,
    3. T. Seemayer
    . 1981. Spontaneous diabetes mellitus syndrome in the rat. I. Association with the major histocompatibility complex. J. Exp. Med. 154: 1237–1242.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Guttmann R. D.,
    2. E. Colle,
    3. F. Michel,
    4. T. Seemayer
    . 1983. Spontaneous diabetes mellitus syndrome in the rat. II. T lymphopenia and its association with clinical disease and pancreatic lymphocytic infiltration. J. Immunol. 130: 1732–1735.
    OpenUrlAbstract
  11. ↵
    1. Poussier P.,
    2. A. F. Nakhooda,
    3. J. A. Falk,
    4. C. Lee,
    5. E. B. Marliss
    . 1982. Lymphopenia and abnormal lymphocyte subsets in the “BB” rat: relationship to the diabetic syndrome. Endocrinology 110: 1825–1827.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Angelillo M.,
    2. D. L. Greiner,
    3. J. P. Mordes,
    4. E. S. Handler,
    5. N. Nakamura,
    6. U. McKeever,
    7. A. Rossini
    . 1988. Absence of RT6+ T cells in diabetes-prone biobreeding/Worcester rats is due to genetic and cell developmental defects. J. Immunol. 141: 4146–4151.
    OpenUrlAbstract
  13. ↵
    1. Woda B. A.,
    2. A. A. Like,
    3. C. Padden,
    4. M. L. McFadden
    . 1986. Deficiency of phenotypic cytotoxic-suppressor T lymphocytes in the BB/W rat. J. Immunol. 136: 856–859.
    OpenUrlAbstract
  14. ↵
    1. Markholst H.,
    2. S. Eastman,
    3. D. Wilson,
    4. B. E. Andreasen,
    5. A. Lernmark
    . 1991. Diabetes segregates as a single locus in crosses between inbred BB rats prone or resistant to diabetes. J. Exp. Med. 174: 297–300.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    1. Poussier P.,
    2. T. Ning,
    3. T. Murphy,
    4. D. Dabrowski,
    5. S. Ramanathan
    . 2005. Impaired post-thymic development of regulatory CD4+25+ T cells contributes to diabetes pathogenesis in BB rats. J. Immunol. 174: 4081–4089.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Wallis R. H.,
    2. K. Wang,
    3. D. Dabrowski,
    4. L. Marandi,
    5. T. Ning,
    6. E. Hsieh,
    7. A. D. Paterson,
    8. J. P. Mordes,
    9. E. P. Blankenhorn,
    10. P. Poussier
    . 2007. A novel susceptibility locus on rat chromosome 8 affects spontaneous but not experimentally induced type 1 diabetes. Diabetes 56: 1731–1736.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Fuller J. M.,
    2. M. Bogdani,
    3. T. D. Tupling,
    4. R. A. Jensen,
    5. R. Pefley,
    6. S. Manavi,
    7. L. Cort,
    8. E. P. Blankenhorn,
    9. J. P. Mordes,
    10. A. Lernmark,
    11. A. E. Kwitek
    . 2009. Genetic dissection reveals diabetes loci proximal to the gimap5 lymphopenia gene. Physiol. Genomics 38: 89–97.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    1. Morel L.,
    2. B. P. Croker,
    3. K. R. Blenman,
    4. C. Mohan,
    5. G. Huang,
    6. G. Gilkeson,
    7. E. K. Wakeland
    . 2000. Genetic reconstitution of systemic lupus erythematosus immunopathology with polycongenic murine strains. Proc. Natl. Acad. Sci. USA 97: 6670–6675.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Fox C. J.,
    2. J. S. Danska
    . 1997. IL-4 expression at the onset of islet inflammation predicts nondestructive insulitis in nonobese diabetic mice. J. Immunol. 158: 2414–2424.
    OpenUrlAbstract
    1. Fox C. J.,
    2. J. S. Danska
    . 1998. Independent genetic regulation of T-cell and antigen-presenting cell participation in autoimmune islet inflammation. Diabetes 47: 331–338.
    OpenUrlAbstract
  21. ↵
    1. Fox C. J.,
    2. A. D. Paterson,
    3. S. M. Mortin-Toth,
    4. J. S. Danska
    . 2000. Two genetic loci regulate T cell-dependent islet inflammation and drive autoimmune diabetes pathogenesis. Am. J. Hum. Genet. 67: 67–81.
    OpenUrlCrossRefPubMed
  22. ↵
    1. MacMurray A. J.,
    2. D. H. Moralejo,
    3. A. E. Kwitek,
    4. E. A. Rutledge,
    5. B. Van Yserloo,
    6. P. Gohlke,
    7. S. J. Speros,
    8. B. Snyder,
    9. J. Schaefer,
    10. S. Bieg,
    11. et al
    . 2002. Lymphopenia in the BB rat model of type 1 diabetes is due to a mutation in a novel immune-associated nucleotide (Ian)-related gene. Genome Res. 12: 1029–1039.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Hornum L.,
    2. J. Rømer,
    3. H. Markholst
    . 2002. The diabetes-prone BB rat carries a frameshift mutation in Ian4, a positional candidate of Iddm1. Diabetes 51: 1972–1979.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Hessner M. J.,
    2. X. Wang,
    3. L. Meyer,
    4. R. Geoffrey,
    5. S. Jia,
    6. J. Fuller,
    7. A. Lernmark,
    8. S. Ghosh
    . 2004. Involvement of eotaxin, eosinophils, and pancreatic predisposition in development of type 1 diabetes mellitus in the BioBreeding rat. J. Immunol. 173: 6993–7002.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Carvalho-Pinto C.,
    2. M. I. García,
    3. L. Gómez,
    4. A. Ballesteros,
    5. A. Zaballos,
    6. J. M. Flores,
    7. M. Mellado,
    8. J. M. Rodríguez-Frade,
    9. D. Balomenos,
    10. C. Martínez-A
    . 2004. Leukocyte attraction through the CCR5 receptor controls progress from insulitis to diabetes in non-obese diabetic mice. Eur. J. Immunol. 34: 548–557.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Schulte B. M.,
    2. K. H. Lanke,
    3. J. D. Piganelli,
    4. E. D. Kers-Rebel,
    5. R. Bottino,
    6. M. Trucco,
    7. R. J. Huijbens,
    8. T. R. Radstake,
    9. M. A. Engelse,
    10. E. J. de Koning,
    11. et al
    . 2012. Cytokine and chemokine production by human pancreatic islets upon enterovirus infection. Diabetes 61: 2030–2036.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Ribaux P.,
    2. J. A. Ehses,
    3. N. Lin-Marq,
    4. F. Carrozzino,
    5. M. Böni-Schnetzler,
    6. E. Hammar,
    7. J. C. Irminger,
    8. M. Y. Donath,
    9. P. A. Halban
    . 2007. Induction of CXCL1 by extracellular matrix and autocrine enhancement by interleukin-1 in rat pancreatic beta-cells. Endocrinology 148: 5582–5590.
    OpenUrlCrossRefPubMed
  28. ↵
    1. Forssmann U.,
    2. M. Uguccioni,
    3. P. Loetscher,
    4. C. A. Dahinden,
    5. H. Langen,
    6. M. Thelen,
    7. M. Baggiolini
    . 1997. Eotaxin-2, a novel CC chemokine that is selective for the chemokine receptor CCR3, and acts like eotaxin on human eosinophil and basophil leukocytes. J. Exp. Med. 185: 2171–2176.
    OpenUrlAbstract/FREE Full Text
    1. Kitaura M.,
    2. N. Suzuki,
    3. T. Imai,
    4. S. Takagi,
    5. R. Suzuki,
    6. T. Nakajima,
    7. K. Hirai,
    8. H. Nomiyama,
    9. O. Yoshie
    . 1999. Molecular cloning of a novel human CC chemokine (Eotaxin-3) that is a functional ligand of CC chemokine receptor 3. J. Biol. Chem. 274: 27975–27980.
    OpenUrlAbstract/FREE Full Text
    1. Ponath P. D.,
    2. S. Qin,
    3. D. J. Ringler,
    4. I. Clark-Lewis,
    5. J. Wang,
    6. N. Kassam,
    7. H. Smith,
    8. X. Shi,
    9. J. A. Gonzalo,
    10. W. Newman,
    11. et al
    . 1996. Cloning of the human eosinophil chemoattractant, eotaxin: expression, receptor binding, and functional properties suggest a mechanism for the selective recruitment of eosinophils. J. Clin. Invest. 97: 604–612.
    OpenUrlCrossRefPubMed
    1. White J. R.,
    2. C. Imburgia,
    3. E. Dul,
    4. E. Appelbaum,
    5. K. O’Donnell,
    6. D. J. O’Shannessy,
    7. M. Brawner,
    8. J. Fornwald,
    9. J. Adamou,
    10. N. A. Elshourbagy,
    11. et al
    . 1997. Cloning and functional characterization of a novel human CC chemokine that binds to the CCR3 receptor and activates human eosinophils. J. Leukoc. Biol. 62: 667–675.
    OpenUrlAbstract
  32. ↵
    1. Humbles A. A.,
    2. B. Lu,
    3. D. S. Friend,
    4. S. Okinaga,
    5. J. Lora,
    6. A. Al-Garawi,
    7. T. R. Martin,
    8. N. P. Gerard,
    9. C. Gerard
    . 2002. The murine CCR3 receptor regulates both the role of eosinophils and mast cells in allergen-induced airway inflammation and hyperresponsiveness. Proc. Natl. Acad. Sci. USA 99: 1479–1484.
    OpenUrlAbstract/FREE Full Text
    1. Gutierrez-Ramos J. C.,
    2. C. Lloyd,
    3. J. A. Gonzalo
    . 1999. Eotaxin: from an eosinophilic chemokine to a major regulator of allergic reactions. Immunol. Today 20: 500–504.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Sallusto F.,
    2. C. R. Mackay,
    3. A. Lanzavecchia
    . 1997. Selective expression of the eotaxin receptor CCR3 by human T helper 2 cells. Science 277: 2005–2007.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Eastman S.,
    2. H. Markholst,
    3. D. Wilson,
    4. A. Lernmark
    . 1991. Leukocytosis at the onset of diabetes in crosses of inbred BB rats. Diabetes Res. Clin. Pract. 12: 113–123.
    OpenUrlCrossRefPubMed
    1. Kürner T.,
    2. V. Burkart,
    3. H. Kolb
    . 1986. Large increase of cytotoxic/suppressor T-lymphoblasts and eosinophils around manifestation of diabetes in BB rats. Diabetes Res. 3: 349–353.
    OpenUrlPubMed
  37. ↵
    1. Geoffrey R.,
    2. S. Jia,
    3. A. E. Kwitek,
    4. J. Woodliff,
    5. S. Ghosh,
    6. A. Lernmark,
    7. X. Wang,
    8. M. J. Hessner
    . 2006. Evidence of a functional role for mast cells in the development of type 1 diabetes mellitus in the BioBreeding rat. J. Immunol. 177: 7275–7286.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Bradley L. M.,
    2. V. C. Asensio,
    3. L. K. Schioetz,
    4. J. Harbertson,
    5. T. Krahl,
    6. G. Patstone,
    7. N. Woolf,
    8. I. L. Campbell,
    9. N. Sarvetnick
    . 1999. Islet-specific Th1, but not Th2, cells secrete multiple chemokines and promote rapid induction of autoimmune diabetes. J. Immunol. 162: 2511–2520.
    OpenUrlAbstract/FREE Full Text
    1. Zipris D.
    1996. Evidence that Th1 lymphocytes predominate in islet inflammation and thyroiditis in the BioBreeding (BB) rat. J. Autoimmun. 9: 315–319.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Rabinovitch A.,
    2. W. Suarez-Pinzon,
    3. A. El-Sheikh,
    4. O. Sorensen,
    5. R. F. Power
    . 1996. Cytokine gene expression in pancreatic islet-infiltrating leukocytes of BB rats: expression of Th1 cytokines correlates with beta-cell destructive insulitis and IDDM. Diabetes 45: 749–754.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    1. Pipeleers D. G.,
    2. M. A. Pipeleers-Marichal
    . 1981. A method for the purification of single A, B and D cells and for the isolation of coupled cells from isolated rat islets. Diabetologia 20: 654–663.
    OpenUrlPubMed
  42. ↵
    1. Ricordi C.,
    2. D. W. Gray,
    3. B. J. Hering,
    4. D. B. Kaufman,
    5. G. L. Warnock,
    6. N. M. Kneteman,
    7. S. P. Lake,
    8. N. J. London,
    9. C. Socci,
    10. R. Alejandro,
    11. et al
    . 1990. Islet isolation assessment in man and large animals. Acta Diabetol. Lat. 27: 185–195.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Mathieu C.,
    2. K. Casteels,
    3. M. Waer,
    4. J. Laureys,
    5. D. Valckx,
    6. R. Bouillon
    . 1998. Prevention of diabetes recurrence after syngeneic islet transplantation in NOD mice by analogues of 1,25(OH)2D3 in combination with cyclosporin A: mechanism of action involves an immune shift from Th1 to Th2. Transplant. Proc. 30: 541.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Qin H. Y.,
    2. B. Singh
    . 1997. BCG vaccination prevents insulin-dependent diabetes mellitus (IDDM) in NOD mice after disease acceleration with cyclophosphamide. J. Autoimmun. 10: 271–278.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Fulkerson P. C.,
    2. C. A. Fischetti,
    3. M. L. McBride,
    4. L. M. Hassman,
    5. S. P. Hogan,
    6. M. E. Rothenberg
    . 2006. A central regulatory role for eosinophils and the eotaxin/CCR3 axis in chronic experimental allergic airway inflammation. Proc. Natl. Acad. Sci. USA 103: 16418–16423.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    1. Huang X.,
    2. B. Hultgren,
    3. N. Dybdal,
    4. T. A. Stewart
    . 1994. Islet expression of interferon-alpha precedes diabetes in both the BB rat and streptozotocin-treated mice. Immunity 1: 469–478.
    OpenUrlCrossRefPubMed
  47. ↵
    1. Ono S. J.,
    2. B. Issa-Chergui,
    3. E. Colle,
    4. R. D. Guttmann,
    5. T. A. Seemayer,
    6. A. Fuks
    . 1988. IDDM in BB rats: enhanced MHC class I heavy-chain gene expression in pancreatic islets. Diabetes 37: 1411–1418.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    Wang, G. S., J. Karamchandani, O. Pulido, L. Rosenberg, and F. W. Scott. 2002. Altered islet homeostasis before classic insulitis in BB rats. Diabetes Metab. 28: 3S90–3S97; discussion 3S108–3S112.
  49. ↵
    1. Hammar E. B.,
    2. J. C. Irminger,
    3. K. Rickenbach,
    4. G. Parnaud,
    5. P. Ribaux,
    6. D. Bosco,
    7. D. G. Rouiller,
    8. P. A. Halban
    . 2005. Activation of NF-kappaB by extracellular matrix is involved in spreading and glucose-stimulated insulin secretion of pancreatic beta cells. J. Biol. Chem. 280: 30630–30637.
    OpenUrlAbstract/FREE Full Text
  50. ↵
    1. Matsukura S.,
    2. C. Stellato,
    3. J. R. Plitt,
    4. C. Bickel,
    5. K. Miura,
    6. S. N. Georas,
    7. V. Casolaro,
    8. R. P. Schleimer
    . 1999. Activation of eotaxin gene transcription by NF-kappa B and STAT6 in human airway epithelial cells. J. Immunol. 163: 6876–6883.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    1. Tomkinson A.,
    2. C. Duez,
    3. G. Cieslewicz,
    4. E. W. Gelfand
    . 2001. Eotaxin-1-deficient mice develop airway eosinophilia and airway hyperresponsiveness. Int. Arch. Allergy Immunol. 126: 119–125.
    OpenUrlCrossRefPubMed
  52. ↵
    1. Rapoport M. J.,
    2. A. Jaramillo,
    3. D. Zipris,
    4. A. H. Lazarus,
    5. D. V. Serreze,
    6. E. H. Leiter,
    7. P. Cyopick,
    8. J. S. Danska,
    9. T. L. Delovitch
    . 1993. Interleukin 4 reverses T cell proliferative unresponsiveness and prevents the onset of diabetes in nonobese diabetic mice. J. Exp. Med. 178: 87–99.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section