APCs of the nonobese diabetic (NOD) mouse have a genetically programmed capacity to overexpress IL-12p40, a cytokine critical for development of pathogenic autoreactive Th1 cells. To determine whether a diabetes-associated NOD chromosomal locus (i.e., Idd) was responsible for this defect, LPS-stimulated macrophages from several recombinant congenic inbred mice with Idd loci on a C57BL/6 background or with different combinations of NOD and CBA genomic segments were screened for IL-12p40 production. Only macrophages from the congenic strains containing the Idd4 locus showed IL-12p40 overproduction/expression. Moreover, analysis of IL-12p40 sequence polymorphisms demonstrated that the Idd4 intervals in these strains contained the IL-12p40 allele of the NOD, although further analysis is required to determine whether the IL-12p40 allele itself is responsible for its overexpression. Thus, the non-MHC-associated Idd4 locus appears responsible for IL-12p40 overexpression, which may be a predisposing factor for type 1 diabetes in NOD mice.

Multiple environmental and genetic factors interact to precipitate insulin-dependent diabetes mellitus (IDDM)2, a spontaneous autoimmune disease in humans and in the nonobese diabetic (NOD) mouse (reviewed in Ref. 1). Although the genetic factors responsible for the different stages of IDDM have not yet been elucidated, quantitative trait loci analyses have been used to identify at least 18 major NOD chromosomal intervals, i.e., Idd loci, that contribute to the course of disease (reviewed in Refs. 2 and 3). Among them, the MHC H-2g7 haplotype within the Idd1 locus is unique in that it is necessary (but not sufficient) for diabetes to develop. Although a few candidate genes have been nominated as potential aberrant genetic elements contained in non-MHC-linked Idd loci, their impact on the course of disease or on a disease-associated cellular process has been difficult to ascertain.

A major obstacle in identifying the IDDM-promoting genes within these non-MHC-linked Idd loci has been the lack of adequate disease-associated phenotypes with sufficient penetrance for tracking in recombinant congenic gene mapping studies (2). Most attempts have been made using complex traits such as disease incidence or cellular processes such as insulitis that likely require epistatic interactions among several Idd loci. However, we have identified and characterized a genetically-programmed, cell type-specific phenotype unique to the NOD strain that exists in the absence of the disease process (i.e., it is expressed in young male and female NOD mice well before disease symptoms (4, 5)), and thus is expected to show sufficient penetrance in recombinant congenic gene mapping studies. This defect is a 4- to 25-fold elevation in production of the IL-12p40 cytokine subunit (but not of other cytokines) by activated macrophages (Mφ) from the NOD strain compared with at least eight other non-diabetic control strains (4, 5, 6). This finding has recently been confirmed in dendritic cells by Tisch and coworkers (7), indicating that the defect is expressed by two key APCs of the innate immune system. The IL-12p40 subunit complexes with either the p35 or the p19 subunits to form the functionally active IL-12p70 (8) or IL-23 (9) cytokines, respectively, that are required for development of Th1 responses and play a central role in the autoimmune process. Furthermore, elevated IL-12p40 production has also been associated with individuals at high-risk for type 1 diabetes (10) in which unique sequence polymorphisms near the IL-12p40 coding sequence have been identified in a large cohort of type 1 diabetic families (11).

Here, we determined whether a single Idd locus could be responsible for the IL-12p40 expression defect in NOD Mφ by screening several different recombinant congenic mouse strains carrying Idd loci on either C57BL/6 or CBA backgrounds. Our results show that the Idd4 locus is responsible for the elevated expression of the IL-12p40 gene in NOD mice.

Four- to 6-wk-old male mice of the following strains were purchased from The Jackson Laboratory (Bar Harbor, ME) and were maintained for 1–4 wk after arrival under germfree conditions: A/J, BALB/c, C57BL/6, C57BL/10, C3H/OUJ, CBA, and NOD. The following CBA-NOD recombinant congenic strains that have unique sets of Idd loci (12) were a gift from Dr. E. Leiter (The Jackson Laboratory): NOcCB-1 (Idd1, 2, 4, 5, 6, 7, 8, 13, 14, 15, 16), CBcNO6 (Idd1, 2, 6, 8, 9, 11, 13, 15, 16), CBcNO7-C (Idd1, 6, 7, 9, 15, 16), and CBcNO7-D (Idd1, 6, 7, 9, 10, 13, 15, 16, 17). Breeding pairs of the following C57BL/6 mice congenic for Idd loci were also a gift from Dr. E. Leiter and bred in-house: B6.NODc17 (B6.NODIdd1), D17MIT21-D17MIT10, 19 cM interval; B6.NODc11 (B6.NODIdd4), D11Mit20-D11Mit42, 52 cM interval; B6.NODc17/c11 (B6.NODIdd1/Idd4), D17Mit21-D17Mit10, 19 cM interval, D11Mit20-D11Mit42, 52 cM interval; B6.NODc17/c1c (B6.NODIdd1/Idd5), D17Mit21-D17Mit10, 19 cM interval, D1Mit3-BclII, 44 cM interval; B6.NODc17/c6 (B6.NODIdd1/Idd6), D17MIT21-D17MIT10, 19 cM interval, D6MIT54-D6MIT14, 24 cM interval; B6.NODc17/c3 (B6.NODIdd1/Idd3/Idd10), D17Mit21-D17Mit10, 19 cM interval, D3Mit132-Tshb, 43 cM interval; B6.NODc2 (B6.NODIdd13), D2Mit17-D2Mit48, 32 cM interval (13).

Thioglycollate-elicited peritoneal exudate Mφ were obtained by peritoneal lavage and isolated by adherence to plastic as previously described (4). Adherent Mφ (2 × 105) were stimulated with LPS (LPS; Escherichia coli: 0111:B4; Sigma-Aldrich, St. Louis, MO) in RPMI 1640 medium supplemented with 2 mM l-glutamine, 0.5% HEPES (Cellgro, Herndon, VA), 5 μg/ml penicillin and 100 U/ml streptomycin (Life Technologies, Grand Island, NY), and 10% FBS (BioWhittaker, Walkersville, MD). The culture-conditioned medium was collected at 24 h at which time Mφ were replenished with fresh stimuli and cultured for another 48 h before collection of conditioned medium (i.e., 24–72 h). Culture-conditioned medium was stored at −20°C for assessment of cytokine levels using sandwich ELISA (BD PharMingen, San Diego, CA).

RNA was extracted from LPS-stimulated adherent Mφ (107) using a total RNA Isolation Kit (BD PharMingen) and quantified by UV spectrophotometry. Total RNA (5 μg) from each sample was used in an RNase protection assay (Riboquant mCK-2b; BD PharMingen) which was resolved using a 0.4 mm urea-polyacrylimide gel and visualized by autoradiography. Samples were quantified by densitometry using Image-Pro Plus software.

Two T→C sequence polymorphisms exist in the cDNA sequence (accession number NM_008352) of the murine IL-12p40 allele at nucleotide positions 506 and 880 down-stream from the transcription start site and distinguish the C57BL/6 and CBA (i.e., T) from the NOD (i.e., C) sequences (14). These polymorphisms were determined in several congenic strains via RT-PCR amplification of total RNA from LPS-stimulated Mφ and sequencing of the PCR product using the following primers:_GACTTTCCTGAAGTGTGAAGCACC and GCTGACCTCCACCTGTGAGTTC which flank the polymorphism at 506 position, and GCAGCAGAATAAATATGAGAAC and CCTTTCCAACGTTGCATCCTAGG which flank the polymorphism at 880 position.

We previously reported the IL-12p40 overexpression defect in LPS-stimulated Mφ from young (pre-disease) NOD mice (4, 5). Here, we determined whether one of the known Idd loci harbored the defect by screening several congenic strains that contained Idd loci on a C57BL/6 background. We first confirmed and extended our previous studies (4, 5) by performing a comprehensive analysis of IL-12p40 levels produced by Mφ from the C57BL/6 background strain relative to other disease-resistant strains (i.e., BALB/c, A/J, C57BL/10, C3H/OuJ) and to the autoimmune-prone NOD strain. Peritoneal Mφ from young mice of all strains were activated with LPS, and conditioned medium was collected in two sequential intervals to quantitatively and qualitatively evaluate IL-12p40 production. Cytokine levels for each strain were normalized to those of C57BL/6 Mφ and the mean of these relative values from several experiments are reported (Fig. 1). LPS-stimulated Mφ from the NOD strain produced 4- to 12-fold greater levels of IL-12p40 than did Mφ from the five normal strains tested during the first 24 h of culture (Fig. 1,A). This difference increased to 10- to 25-fold greater IL-12p40 levels in the NOD during the next 48 h of culture (i.e., 24–72 h; Fig. 1 B). Differences in Mφ cytokine production among all strains were not associated with a variation in viability or metabolic activity as measured by conversion of the redox reagent, Alamar Blue (data not shown). This dysregulation in cytokine expression appears specific to IL-12p40 because NOD Mφ levels of other cytokines are within the range of normal strains (4).

FIGURE 1.

Elevated production of IL-12p40 by Mφ from the NOD strain. Thioglycollate-elicited peritoneal Mφ (2 × 105) from five control strains (BALB/c, A/J, C57BL/6 (B6), C57BL/10 (B10), and C3H/OuJ) and the diabetes-prone NOD strain were activated with LPS (100 ng/ml) and incubated for 24 h (A) and from 24 to 72 h (B). (At the end of 24 h, conditioned medium was removed and cultures were replenished with fresh medium and LPS for an additional 48 h.) Total IL-12p40 levels were measured using an ELISA. Shown are means ± SEM of several experiments per strain (N denoted in parentheses) of relative IL-12p40 levels from each strain normalized to B6 levels which were given the value of 1. The mean ± SEM of B6 Mφ IL-12p40 levels from several experiments (n = 8) was 3.4 ± 0.9 ng/ml for 24-h cultures, and 0.6 ± 0.1 ng/ml for 24- to 72-h cultures.

FIGURE 1.

Elevated production of IL-12p40 by Mφ from the NOD strain. Thioglycollate-elicited peritoneal Mφ (2 × 105) from five control strains (BALB/c, A/J, C57BL/6 (B6), C57BL/10 (B10), and C3H/OuJ) and the diabetes-prone NOD strain were activated with LPS (100 ng/ml) and incubated for 24 h (A) and from 24 to 72 h (B). (At the end of 24 h, conditioned medium was removed and cultures were replenished with fresh medium and LPS for an additional 48 h.) Total IL-12p40 levels were measured using an ELISA. Shown are means ± SEM of several experiments per strain (N denoted in parentheses) of relative IL-12p40 levels from each strain normalized to B6 levels which were given the value of 1. The mean ± SEM of B6 Mφ IL-12p40 levels from several experiments (n = 8) was 3.4 ± 0.9 ng/ml for 24-h cultures, and 0.6 ± 0.1 ng/ml for 24- to 72-h cultures.

Close modal

LPS-stimulated Mφ from seven recombinant congenic strains on a C57BL/6 background containing single, double, or triple Idd loci were screened for the IL-12p40 overproduction defect. Of the seven congenic strains screened, only those carrying the Idd4 locus, i.e., B6.NODIdd4 and B6.NODIdd1/Idd4, showed significantly elevated levels of IL-12p40 production relative to the parental C57BL/6 strain (Fig. 2), demonstrating that the Idd4 locus is the factor responsible for the IL-12p40 overproduction defect. The 20% difference in IL-12p40 levels between the B6.NODIdd4 and NOD strains is most likely due to elevated IL-10 levels (an inhibitory cytokine) in the C57BL/6 background: C57BL/6 and the B6.NODIdd4 congenic Mφ produce 4-fold more IL-10 than NOD Mφ (Ref. 4 and data not shown). Moreover, Ab-mediated neutralization of IL-10 restored IL-12p40 levels of the Idd4 congenic strains (but not C57BL/6) to those of the NOD strain (data not shown). As a control for the Idd4-specific effects on the IL-12p40 gene, TNF-α levels were similar among all strains (i.e., less than 2-fold differences; data not shown), which is consistent with our previous observations of C57BL/6 and NOD Mφ TNF-α levels (4).

FIGURE 2.

Elevated production of IL-12p40 by Mφ from the B6.NODIdd4 and B6.NODIdd1/Idd4 congenic strains. Thioglycollate-elicited peritoneal Mφ (2 × 105) from seven B6.NODIdd-congenic strains containing single (Idd1, Idd4, Idd13), double (Idd1/4, Idd1/5, Idd1/6, Idd1/7) or triple (Idd1/3/10) Idd loci on a C57BL/6 background, and from the normal C57BL/6 and diabetes-prone NOD control strains were activated with LPS (100 ng/ml) and incubated for 24 h. Conditioned medium was collected and IL-12p40 levels were measured by ELISA. Shown are means ± SEM of several experiments per strain (N denoted in parentheses) of relative IL-12p40 levels from each strain normalized to those of the NOD which were given the value of 1. ∗, Significantly (p < 0.05) different from C57BL/6 mean values. The mean ± SEM of NOD Mφ IL-12p40 levels from several experiments (n = 11) was 14.8 ± 2.3 ng/ml for 24-h cultures.

FIGURE 2.

Elevated production of IL-12p40 by Mφ from the B6.NODIdd4 and B6.NODIdd1/Idd4 congenic strains. Thioglycollate-elicited peritoneal Mφ (2 × 105) from seven B6.NODIdd-congenic strains containing single (Idd1, Idd4, Idd13), double (Idd1/4, Idd1/5, Idd1/6, Idd1/7) or triple (Idd1/3/10) Idd loci on a C57BL/6 background, and from the normal C57BL/6 and diabetes-prone NOD control strains were activated with LPS (100 ng/ml) and incubated for 24 h. Conditioned medium was collected and IL-12p40 levels were measured by ELISA. Shown are means ± SEM of several experiments per strain (N denoted in parentheses) of relative IL-12p40 levels from each strain normalized to those of the NOD which were given the value of 1. ∗, Significantly (p < 0.05) different from C57BL/6 mean values. The mean ± SEM of NOD Mφ IL-12p40 levels from several experiments (n = 11) was 14.8 ± 2.3 ng/ml for 24-h cultures.

Close modal

We confirmed that the IL-12p40 overproduction defect was associated with the Idd4 locus by comparing mRNA levels expressed among the C57BL/6, B6.NODIdd4, B6.NODIdd1/Idd4, and NOD strains (Fig. 3). Maximal expression of IL-12p40 mRNA occurred at 4 h in Mφ from all four strains; Mφ from the NOD, B6.NODIdd4, and B6.NODIdd1/Idd4 strains expressed similar levels, each of which were 6- to 12-fold greater than those of the C57BL/6 at 4 h. Maximum IL-12p35 mRNA expression among all strains occurred at 4 h at which time differences among strains were roughly 2- to 3-fold. Moreover, a distinguishing qualitative difference is that Mφ expression of IL-12p40 mRNA was substantially greater than that of IL-12p35 mRNA only in the NOD, B6.NODIdd4, and B6.NODIdd1/Idd4 strains, whereas mRNA levels of these two subunits were similar in the C57BL/6 (Fig. 3) and in other control strains as well (4, 5). Note that we have previously shown that the IL-12p40 expression defect by NOD Mφ leads to overproduction of the functional heterodimer, IL-12p70 (4, 5). These results further support the association of the IL-12p40 defect with the Idd4 locus.

FIGURE 3.

Elevated expression of IL-12p40 mRNA by the B6.NODIdd4 and B6.NODIdd1/Idd4 congenic strains. Thioglycollate-elicited peritoneal Mφ (107) from C57BL/6, NOD, B6.NOD Idd4, and B6.NOD Idd1/Idd4 strains were activated with LPS (100 ng/ml) for 4 and 16 h. Total RNA was extracted and mRNA levels of IL-12p40 and IL-12p35 subunits were assessed in an RNase protection assay. Samples were separated on a sequencing gel and visualized by autoradiography. The intensities of each IL-12p40 and p35 mRNA band were normalized to the loading control L32 mRNA band via densitometry analysis using the ImagePro software. Results represent one of at least three experiments that had similar results.

FIGURE 3.

Elevated expression of IL-12p40 mRNA by the B6.NODIdd4 and B6.NODIdd1/Idd4 congenic strains. Thioglycollate-elicited peritoneal Mφ (107) from C57BL/6, NOD, B6.NOD Idd4, and B6.NOD Idd1/Idd4 strains were activated with LPS (100 ng/ml) for 4 and 16 h. Total RNA was extracted and mRNA levels of IL-12p40 and IL-12p35 subunits were assessed in an RNase protection assay. Samples were separated on a sequencing gel and visualized by autoradiography. The intensities of each IL-12p40 and p35 mRNA band were normalized to the loading control L32 mRNA band via densitometry analysis using the ImagePro software. Results represent one of at least three experiments that had similar results.

Close modal

Although the Idd4 locus (15) and the IL-12p40 gene, i.e., IL-12b (16), map to murine chromosome 11, it is unclear whether the IL-12b gene, which maps to the 19 cM position, is contained within the Idd4 interval of the B6.NODIdd4 congenic strain, which starts around the 16–20 cM position (Ref. 13 and see Table I). To address this, we made use of two different single-nucleotide polymorphisms located within the IL-12p40 cDNA that distinguish the C57BL/6 (and the CBA) from the NOD sequence (i.e., T→C (B6→NOD) at both 506 and 880 nucleotide positions, see Materials and Methods and Ref. 14). Sequencing of the two single-nucleotide polymorphisms in cDNA from all strains studied in this report showed that strains containing the Idd4 interval (and therefore the IL-12p40 defect) also contained the NOD IL-12b allele, and that those strains that did not contain the Idd4 locus contained the B6 (and CBA, see below) IL-12b allele (Table I).

Table I.

Genotypic characterization of the IL-12b allele and Idd4 intervals in recombinant congenic murine strains

StrainaIL-12p40 AllelebIL-12p40 Expression DefectIdd4 Locusc
506 nt880 ntInterval (cM)Markers
C57BL/6 − None None 
B6.NOD.Idd1 − None None 
B6.NOD.Idd1/3/10 − None None 
B6.NOD.Idd4 20–72 D11Mit20 
     D11mit42 
B6.NOD.Idd1/Idd4 20–72 D11Mit20 
     D11mit42 
B6.NOD.Idd1/Idd5 − None None 
B6.NOD.Idd1/Idd6 − None None 
B6.NOD.Idd13 − None None 
CBA − None None 
CBc.NO7C − 40–48 D11Acrb 
     D11Mit36 
CBc.NO6 − 40–70 D11Acrb 
     D11Mit214 
NOcCB-1 17–40 D11Mit84 
     D11Acrb 
NOD n.a. n.a. 
DBA/2 Cd Cd d n.a. n.a. 
StrainaIL-12p40 AllelebIL-12p40 Expression DefectIdd4 Locusc
506 nt880 ntInterval (cM)Markers
C57BL/6 − None None 
B6.NOD.Idd1 − None None 
B6.NOD.Idd1/3/10 − None None 
B6.NOD.Idd4 20–72 D11Mit20 
     D11mit42 
B6.NOD.Idd1/Idd4 20–72 D11Mit20 
     D11mit42 
B6.NOD.Idd1/Idd5 − None None 
B6.NOD.Idd1/Idd6 − None None 
B6.NOD.Idd13 − None None 
CBA − None None 
CBc.NO7C − 40–48 D11Acrb 
     D11Mit36 
CBc.NO6 − 40–70 D11Acrb 
     D11Mit214 
NOcCB-1 17–40 D11Mit84 
     D11Acrb 
NOD n.a. n.a. 
DBA/2 Cd Cd d n.a. n.a. 
a

Recombinant congenic strains with Idd4 loci on a C57BL/6 background (13 ) or with roughly 50% CBA and 50% NOD genomes with different combinations of Idd loci (12 ).

b

Polymorphism in the IL-12p40 allele that distinguishes the NOD from the C57BL/6 and CBA alleles (14 ).

c

Idd4 intervals on murine chromosome 11 defined by different microsatellite markers (1213 ).

d

Refs. 6 and 14 .

To confirm the association of the IL-12p40 functional defect with the Idd4 locus, IL-12p40 production was assessed in four recombinant congenic strains containing roughly 50% NOD and 50% CBA genomes but varied in the combinations of Idd loci (12). Only one recombinant congenic strain, NOcCB-1, showed elevated-production (Fig. 4,A) and elevated-expression (Fig. 4,C) of IL-12p40 relative to the CBA control strain. As previously noted for the NOD, B6.NODIdd4, and B6.NODIdd1/Idd4 strains (see Fig. 3), expression levels of IL-12p40 mRNA were substantially greater than those of IL-12p35 mRNA only in the NOD and NOcCB-1 strains (Fig. 4,C). TNF-α levels were similar among all strains tested (Fig. 4,B), confirming that the underlying molecular aberrancy does not effect the general Mφ activation pathway, but rather is focused on IL-12p40 expression in the NOcCB-1 and NOD strains. The NOcCB-1 strain was the only one of the four congenic strains to contain the entire Idd4 locus (see Materials and Methods and Ref. 12 for a detailed comparison of strains) and the NOD IL-12b allele (Table I). These results confirm that the IL-12p40 expression defect is caused by the Idd4 locus, which encompasses the NOD IL-12b allele.

FIGURE 4.

Elevated production of IL-12p40 by a CBA-NOD recombinant congenic strain containing the Idd4 locus. Thioglycollate-elicited peritoneal Mφ (2 × 105) from four CBA-NOD recombinant congenic strains (CBcNO6, CBcNO7-C, CBcNO7-D, and NOcCB-1), each containing a different set of Idd loci, were activated with LPS (100 ng/ml) and incubated for 24 h. Conditioned medium was collected and IL-12p40 (A) and TNF-α (B) levels were measured by ELISA. Shown are mean ± SEM values of triplicate cultures from one of two experiments that had similar results. C, mRNA levels of 4-h LPS (100 ng/ml)-activated Mφ (107) from each strain that were assessed using the RNase protection assay. The intensities of each IL-12p40 and p35 mRNA band were normalized to the loading control L32 mRNA band via densitometry analysis using the ImagePro software.

FIGURE 4.

Elevated production of IL-12p40 by a CBA-NOD recombinant congenic strain containing the Idd4 locus. Thioglycollate-elicited peritoneal Mφ (2 × 105) from four CBA-NOD recombinant congenic strains (CBcNO6, CBcNO7-C, CBcNO7-D, and NOcCB-1), each containing a different set of Idd loci, were activated with LPS (100 ng/ml) and incubated for 24 h. Conditioned medium was collected and IL-12p40 (A) and TNF-α (B) levels were measured by ELISA. Shown are mean ± SEM values of triplicate cultures from one of two experiments that had similar results. C, mRNA levels of 4-h LPS (100 ng/ml)-activated Mφ (107) from each strain that were assessed using the RNase protection assay. The intensities of each IL-12p40 and p35 mRNA band were normalized to the loading control L32 mRNA band via densitometry analysis using the ImagePro software.

Close modal

It is unclear whether the NOD IL-12b allele is responsible for the overexpression of IL-12p40 because we have previously reported (6) that the DBA/2, NZB, NZW, and NZB/W strains, which have the NOD IL-12b allele (14), do not overexpress IL-12p40 (see Table I). Further, we observed no sequence polymorphisms in either the IL-12p40 promoter region (i.e., 800 bases upstream from the transcriptional start site) or the 3′ untranslated region (i.e., 750 bases down-stream from the TAG translational stop codon) that were unique to the NOD strain relative to five normal strains (i.e., BALB/c, C57BL/6, C57BL/10, C3H/Ouj, A/J; data not shown). These observations suggest that the underlying genetic element responsible for the IL-12p40 overexpression defect lies outside of the IL-12b gene but within the Idd4 locus. Interestingly, Delovitch and coworkers (15) have mapped the T lymphocyte hypoproliferative phenotype to the Idd4 locus using the NOD background congenic for a B6 resistant locus which spans a 5.2 cM interval at positions 43.8 to 49 cM (i.e., markers D11Nds1 and D11Mit38/325, respectively). However, it is unlikely that this interval confers IL-12p40 overexpression because the CBcNO-6 and CBcNO-7 strains contain this 5.2 cM NOD interval (see Table I), but produced normal levels of IL-12p40 (see Fig. 4). Therefore, the aberrant genes causing the IL-12p40 expression defect are most likely within other subintervals of the Idd4 interval, which could be further addressed by creating congenic strains containing a finer resolution of the interval. In addition, a comparative microarray mRNA expression analysis of LPS-stimulated Mφ from Idd congenic and normal strains may reveal the genes responsible for the IL-12p40 expression defect. As the mouse genome sequence is near completion, this genomic information should be useful in uncovering the genetic aberrancies within disease-associated loci of autoimmune-prone strains.

We thank Dr. Ed Leiter (The Jackson Laboratory, Bar Harbor, ME) for generously providing the C57BL/6 and CBA recombinant congenic strains, Rick Alvarez and Richard Godoy for breeding mice, and Julianne Eggold for preparation of figures.

2

Abbreviations used in this paper: IDDM, precipitate insulin-dependent diabetes mellitus; NOD, nonobese diabetic; Mφ, macrophage.

1
Delovitch, T. L., B. Singh.
1997
. The nonobese diabetic mouse as a model of autoimmune diabetes: immune dysregulation gets the NOD.
Immunity
7
:
727
.
2
Lyons, P. A., L. S. Wicker.
1999
. Localizing quantitative trait loci in the NOD mouse model of type 1 diabetes.
Curr. Dir. Autoimmun.
1
:
208
.
3
Wicker, L. S., J. A. Todd, L. B. Peterson.
1995
. Genetic control of autoimmune diabetes in the NOD mouse.
Annu. Rev. Immunol.
13
:
179
.
4
Alleva, D. G., R. P. Pavlovich, C. Grant, S. B. Kaser, D. I. Beller.
2000
. Aberrant macrophage cytokine production is a conserved feature among autoimmune-prone mouse strains: elevated interleukin (IL)-12 and an imbalance in tumor necrosis factor-α and IL-10 define a unique cytokine profile in macrophages from young nonobese diabetic mice.
Diabetes
49
:
1106
.
5
Alleva, D. G., E. B. Johnson, J. Wilson, D. I. Beller, P. J. Conlon.
2001
. SJL and NOD macrophages are uniquely characterized by genetically programmed, elevated expression of the IL-12(p40) gene, suggesting a conserved pathway for the induction of organ-specific autoimmunity.
J. Leukocyte Biol.
69
:
440
.
6
Alleva, D. G., S. B. Kaser, D. I. Beller.
1998
. Intrinsic defects in macrophage IL-12 production associated with immune dysfunction in the MRL/++ and New Zealand Black/White F1 lupus-prone mice and the Leishmania major-susceptible BALB/c strain.
J. Immunol.
161
:
6878
.
7
Weaver, D. J., Jr, B. Poligone, T. Bui, U. M. Abdel-Motal, A. S. Baldwin, Jr, R. Tisch.
2001
. Dendritic cells from nonobese diabetic mice exhibit a defect in NF-κB regulation due to a hyperactive IκB kinase.
J. Immunol.
167
:
1461
.
8
Gately, M. K., L. M. Renzetti, J. Magram, A. S. Stern, L. Adorini, U. Gubler, D. H. Presky.
1998
. The interleukin-12/interleukin-12-receptor system: role in normal and pathologic immune responses.
Annu. Rev. Immunol.
16
:
495
.
9
Cua, D. J., J. Sherlock, Y. Chen, C. A. Murphy, B. Joyce, B. Seymour, L. Lucian, W. To, S. Kwan, T. Churakova, et al
2003
. Interleukin-23 rather than interleukin-12 is the critical cytokine for autoimmune inflammation of the brain.
Nature
421
:
744
.
10
Szelachowska, M., A. Kretowski, I. Kinalska.
1997
. Increased in vitro interleukin-12 production by peripheral blood in high-risk IDDM first degree relatives.
Horm. Metab. Res.
29
:
168
.
11
Morahan, G., D. Huang, S. I. Ymer, M. R. Cancilla, K. Stephen, P. Dabadghao, G. Werther, B. D. Tait, L. C. Harrison, P. G. Colman.
2001
. Linkage disequilibrium of a type 1 diabetes susceptibility locus with a regulatory IL12B allele.
Nat. Genet.
27
:
218
.
12
Reifsnyder, P. C., J. C. Flynn, A. L. Gavin, E. A. Simone, J. J. Sharp, L. Herberg, E. H. Leiter.
1999
. Genotypic and phenotypic characterization of six new recombinant congenic strains derived from NOD/Shi and CBA/J genomes.
Mamm. Genome
10
:
161
.
13
Yui, M. A., K. Muralidharan, B. Moreno-Altamirano, G. Perrin, K. Chestnut, E. K. Wakeland.
1996
. Production of congenic mouse strains carrying NOD-derived diabetogenic genetic intervals: an approach for the genetic dissection of complex traits.
Mamm. Genome
7
:
331
.
14
Ymer, S. I., D. Huang, G. Penna, S. Gregori, K. Branson, L. Adorini, G. Morahan.
2002
. Polymorphisms in the Il12b gene affect structure and expression of IL-12 in NOD and other autoimmune-prone mouse strains.
Genes Immun.
3
:
151
.
15
Grattan, M., Q. S. Mi, C. Meagher, T. L. Delovitch.
2002
. Congenic mapping of the diabetogenic locus Idd4 to a 5.2-cM region of chromosome 11 in NOD mice: identification of two potential candidate subloci.
Diabetes
51
:
215
.
16
Schoenhaut, D. S., A. O. Chua, A. G. Wolitzky, P. M. Quinn, C. M. Dwyer, W. McComas, P. C. Familletti, M. K. Gately, U. Gubler.
1992
. Cloning and expression of murine IL-12.
J. Immunol.
148
:
3433
.