To localize the MHC-linked diabetogenic genes in the nonobese diabetic (NOD) mouse, a recombinational hotspot from the B10.A(R209) mouse was introduced to the region between the MHC class I K and class II A of the NOD mouse with the recombinational site centromeric to the Lmp2/Tap1 complex by breeding the two strains. Replacement of the NOD region centromeric to the recombinational site with the same region in R209 mice prevented the development of diabetes (from 71 to 3%) and insulitis (from 61 to 15%) in the N7 intra-MHC recombinant NOD mice. Similarly, the replacement of the NOD class II A, E and class I D region with the same region in R209 mice prevented the diseases (diabetes, from 71 to 0%; insulitis, from 61 to 3%). In addition to the MHC class II genes, there are at least two MHC-linked diabetogenic genes in the region centromeric to Lmp2.

The nonobese diabetic (NOD)3 mouse develops type 1 (insulin-dependent) diabetes mellitus secondary to β cell destruction by infiltrating immune cells (insulitis) (1). The development of type 1 diabetes is controlled by multiple loci (Idd1–17), including the MHC-linked gene(s) on chromosome 17 (2, 3, 4, 5) and non-MHC-linked diabetogenic genes on other chromosomes (6). The genes in the MHC region are the most important in determining susceptibility or resistance to type 1 diabetes. The number and location of the MHC-linked diabetogenic gene(s) are still unclear. Transgenic studies indicate that the unique MHC class II Aβ (7, 8, 9) and deleted E molecules (7, 10) are important for the development of diabetes in NOD mice. In addition, there is increasing evidence suggesting a second MHC-linked diabetogenic gene outside the MHC class II region in the NOD mouse (11, 12, 13). To identify the MHC-linked diabetogenic genes, it is essential to dissect the MHC region and restrict the responsible region for the development of diabetes. Recombination frequency within the MHC region is too low to fine map a gene by standard linkage analysis in breeding studies between two inbred strains because of strong linkage disequilibria. We report here that a recombinational hotspot was introduced from the B10.A(R209) mouse into the region between the MHC class I K and class II A of the NOD mouse to dissect the MHC region. Identification of all the MHC-linked diabetogenic genes in the NOD mouse should accelerate the studies of homologous genes in human type 1 diabetes.

The NOD/Shi/Jos and B10.A(R209)/Msf/Jos mice were housed at Joslin Diabetes Center Animal Facility (Boston, MA). The animals were screened for the development of diabetes once a week. When glycosuria was found, blood glucose levels were measured (diabetic, >17 mmol/L). To localize the MHC region responsible for diabetes in the NOD mouse (H2g7; MHC haplotype, Kd, Ag7, E, Db), intra-MHC recombinant NOD mice were established by introducing a recombinational hotspot from B10.A(R209) mice into NOD mice. The B10.A(R209) mouse (H2r209; MHC haplotype, at Kwm7, Ak, Ed, Dd) has a hotspot that causes intra-MHC recombinations between the MHC class I K and class II A regions (14, 15). The MHC wm7 haplotype derived from wild mice, Mus musculus molossinus (MOL), enhances meiotic recombination at a 1.3-kb DNA segment adjacent to the gene for low-molecular-mass polypeptide-2 (Lmp2 hotspot) (16, 17).

NOD mice were mated with B10.A(R209) mice (hereafter R209) to produce F1(NOD × R209) mice. Heterozygous F1 mice were mated with NOD mice to produce first backcross (BC1) [(NOD × R209)F1 × NOD] mice. Female BC1 F107 and F144 were intra-MHC recombinants (recombination frequency 2.6% in females). The MHC haplotype of BC1 F107 was g7/g7 at K and g7/r209 at A through D. The MHC haplotype of BC1 F144 was g7/r209 at K and g7/g7 at A through D. The two intra-MHC recombinant BC1 females were backcrossed to NOD mice to produce BC2 and BC3 intra-MHC recombinant mice. The BC3 heterozygous mice were intercrossed to produce the N4 intra-MHC recombinant NOD mice. The N4 intercross animals were typed for the MHC haplotypes and divided into five groups (G1-G5 in Table I).

Table I.

The incidence of diabetes and insulitis in N4 and N7 intra-MHC recombinant NOD mice at 10 mo of agea

MHCN4N7
KA–DDiabetesInsulitis (%)bDiabetesInsulitis (%)
FemaleMaleTotal (%)FemaleMaleTotal (%)
G1c g7 /g7 g7 /g7 7 /10 4 /10 11 /20 (55)c 78 /109 (72)c 17 /19 13 /23 30 /42 (71)d 359 /586 (61)d 
G2 g7 /r209 g7 /g7 1 /5 0 /5 1 /10 (10) 20 /172 (12) 12 /28 1 /23 13 /51 (26) 295 /662 (45) 
G3 r209 /r209 g7 /g7 0 /5 0 /4 0 /9 (0) 27 /186 (15) 1 /17 0 /13 1 /30 (3) 103 /669 (15) 
G4 g7 /g7 g7 /r209 0 /5 0 /5 0 /10 (0) 59 /248 (24) 7 /29 0 /15 7 /44 (16) 102 /772 (13) 
G5 g7 /g7 r209 /r209 0 /5 0 /5 0 /10 (0) 15 /173 (9) 0 /7 0 /12 0 /19 (0) 14 /446 (3) 
MHCN4N7
KA–DDiabetesInsulitis (%)bDiabetesInsulitis (%)
FemaleMaleTotal (%)FemaleMaleTotal (%)
G1c g7 /g7 g7 /g7 7 /10 4 /10 11 /20 (55)c 78 /109 (72)c 17 /19 13 /23 30 /42 (71)d 359 /586 (61)d 
G2 g7 /r209 g7 /g7 1 /5 0 /5 1 /10 (10) 20 /172 (12) 12 /28 1 /23 13 /51 (26) 295 /662 (45) 
G3 r209 /r209 g7 /g7 0 /5 0 /4 0 /9 (0) 27 /186 (15) 1 /17 0 /13 1 /30 (3) 103 /669 (15) 
G4 g7 /g7 g7 /r209 0 /5 0 /5 0 /10 (0) 59 /248 (24) 7 /29 0 /15 7 /44 (16) 102 /772 (13) 
G5 g7 /g7 r209 /r209 0 /5 0 /5 0 /10 (0) 15 /173 (9) 0 /7 0 /12 0 /19 (0) 14 /446 (3) 
a

The degree of insulitis: grade 0, no infiltrating immune cells; grade 1 (periinsulitis), immune cells are adjacent to the islet but do not enter the islet; grade 2, infiltrating cells occupy <25% of the islet area; grade 3, infiltrating cells occupy 25–50% of the islet area; and grade 4, infiltrating cells occupy >50% of the islet area. Immune cell infiltration into the islets with more than grade 2 was regarded as insulitis. Periinsulitis (grade 1) was found at the incidence 8, 4, 6, and 2%, and grade 0 at the incidence 67, 80, 94, and 91% in N4 and N12 G1, G2, G3, and G5 mice at 2 mo of age (G4 not examined), respectively. The incidence of insulitis in N4 mice at 10 mo of age was: grade 0, 26%; grade 1, 2%; grade 2, 16%; grade 3, 7%; and grade 4, 49% in G1 (109 islets, 20 mice); 83, 5, 4, 3, and 5% in G2 (172 islets, 10 mice); 80, 5, 6, 3, and 6% in G3 (186 islets, 9 mice); 69, 7, 14, 5, and 5% in G4 (248 islets, 10 mice); and 87, 4, 5, 2, and 2% in G5 (173 islets, 10 mice). The incidence of insulitis in N10 at 10 mo of age was: grade 0, 24%; grade 1, 15%; grade 2, 16%; grade 3, 11%; and grade 4, 34% in G1 (586 islets, 42 mice); 38, 17, 20, 9, and 16% in G2 (662 islets, 51 mice); 70, 15, 9, 3, and 3% in G3 (669 islets, 30 mice); 77, 10, 6, 3, and 4% in G4 (772 islets, 44 mice); and 94, 3, 1, 1, and 1% in G5 (444 islets, 19 mice).

b

Percent insulitis = [(no. of islets with insulitis)/(total no. of islets)] × 100.

c G1 mice were derived from the BC3 and BC6 intercross animals, but not from regular NOD mice.

c

, p < 0.05 (χ2 test with continuing correction); p = 0.0002 (Log-Rank test) (G1 vs G2–G5).

d

, p < 0.0001 (χ2 test with continuing correction); p = 0.0002 (Log-Rank test) (G1 vs G2–G5).

Cervical lymph node cells were prepared from NOD and R209 mice and incubated with mAbs reacting with the MHC class I and II molecules: mAb SF1-1.1 (Kd), mAb HD25 (Kwm7) (15), mAb 28-8-6S (Db), mAb 34-5-8S (Dd), mAb 10-2-16 (I-Ak,f,r,s) reacting with both NOD and R209, mAb Y3P (I-Ab,f,p,q,r) reacting with R209, and mAb 14-4-4S (I-E). Nonspecific mouse myeloma Ig (RPC5.4) was used as negative control. FITC-conjugated goat anti-mouse IgG (Fc specific) was used as a second Ab. The reactivity was examined by flow cytometric analysis.

DNA was extracted from tails or livers according to the standard protocols using proteinase K and phenol/chloroform. Southern blot analysis was performed according to the previous method (16). The DNA samples were subjected to genotyping with the microsatellite markers (MapPairs; Research Genetics, Huntsville, AL). PCR reactions were performed according to the methods described elsewhere (18). The marker distances from the centromere were based on the data of the MIT F2 intercrosses.

The incidence of diabetes and insulitis in the G1-G5 group of the N4 and N7 intercross mice at 10 mo of age is shown in Table I. Replacement (homologous recombination) of the NOD class I K region with the R209 class I K region prevented the development of diabetes and insulitis in the N4 intercross mice (diabetes, 55% in G1 and 0% in G3; insulitis, 72% in G1 and 15% in G3) and in the N7 intercross mice (diabetes, 71% in G1 and 3% in G3; insulitis, 61% in G1 and 15% in G3). Similarly, the replacement of the NOD A, E, and D region with that of the R209 mouse prevented the development of diabetes and insulitis in N4 G5 mice (diabetes, 0%; insulitis, 9%) and in N7 G5 (diabetes, 0%; insulitis, 3%). The heterozygotes in G2 and G4 of the N4 and N7 generations showed a reduced incidence of diabetes and insulitis in comparison with the homozygotes in G1. The incidence of diabetes in the N10 intercross mice at 6 mo of age was 67% (22/33) in G1, 17% (4/24) in G2, 0% (0/12) in G3, 5% (1/19) in G4, and 0% (0/18) in G5. Thus, replacement of the NOD MHC class I K region with the R209 K region was associated with marked suppression of diabetes in the N4, N7, and N10 intra-MHC recombinant NOD mice in a dose-dependent manner.

The MHC class II molecules on the spleen cells from the N10 G3 intra-MHC recombinant NOD mice at 4 mo of age were measured and compared with those from the N10 G1 mice using the Quantum Fluorescence Kits for molecules of equivalent soluble fluorochromes (MESF) units of FITC (Sigma, St. Louis, MO) and mAb 10-2-16 and 40A reacting with NOD MHC I-Ag7. The level of the MHC class II surface molecules on the spleen cells was not significantly different between the G1 and G3 group in the quantitation assay. This result indicates that the G3 intra-MHC recombinant NOD mice express the MHC class II molecules well, and that the suppression of diabetes in the G3 mice is unrelated to the MHC class II surface expression, but may be caused by novel diabetogenic genes in the region centromeric to the Lmp2 gene.

RFLP analysis showed that the parental BC1 F107 and F144 mice and G3 mice in the N4 intra-MHC recombinant mice had a recombinational site at 2 kb centromeric to Lmp2/Tap1 complex (Fig. 1). Taken the incidence of diabetes and insulitis in the N4 intra-MHC recombinant NOD mice into considerartion (Table I), the results from the RFLP analysis suggests that the region centromeric to Lmp2, including the MHC class I K, is important for the development of diabetes and insulitis as well as the MHC class II genes.

FIGURE 1.

Southern blot analysis for the recombination site between class I K and class II A in the N4 intra-MHC recombinant G1, G2, and G3 animals. Locations of the probes are indicated above the restriction map (16 ). The DNA samples were digested by TaqI or SphI and hybridized with Probe A. (Southern blot analysis with Probe B, data not shown)

FIGURE 1.

Southern blot analysis for the recombination site between class I K and class II A in the N4 intra-MHC recombinant G1, G2, and G3 animals. Locations of the probes are indicated above the restriction map (16 ). The DNA samples were digested by TaqI or SphI and hybridized with Probe A. (Southern blot analysis with Probe B, data not shown)

Close modal

To assess the contribution of the R209 MHC-linked segments to the disease protection, tail or liver DNA samples from NOD, R209, and the BC6 intra-MHC recombinant (heterozygous) mice were examined for homozygosity at Idd3, -4, -5, -6, -7, -8, -9, -11, -12, -13, -14, and -15 loci. All the non-MHC Idd segments carried only the NOD types. This suggests that the contribution to changes in the incidence of diabetes and insulitis was caused by the MHC-linked segments.

Our breeding studies have reached the BC11–13 and N12 generations. The G3 intra-MHC recombinant NOD mice from the intercrosses (N12) of the BC11 mice were examined for the centromeric boundaries of the r209 segment. Two different types of the recombinational sites were found at 4.4 and >6.6 cM centromeric to the MHC K locus. In comparison with the G1 mice, these G3 mice with a different length of homozygous r209/r209 segment should provide information to determine the region containing the second MHC-linked diabetogenic gene. Two N12 congenic lines, whose centromeric boundaries are at different distances, were established and screened for the development of diabetes for 250 days. Of the G3 lines, G3-A line, whose centromeric boundary was at >6.6 cM to Lmp2, did not develop diabetes (incidence: 0/16 females and 0/28 males, p < 4 × 10−10 and p < 1 × 10−10, respectively, vs G1; Fisher’s exact test), while G1 mice developed diabetes at the incidence of 87% (40/46) in females and 78% (32/41) in males by the age of 250 days (Fig. 2). G3-B line, whose centromeric boundary was at 4.4 cM, was protected against diabetes by the age of 250 days (incidence: 4/10 females and 0/8 males, p < 0.005 and p < 1 × 10−4, respectively, vs G1). Comparison of the disease incidence between female lines G3-A and G3-B (p = 0.014) suggests a third MHC-linked diabetogenic gene outside of the MHC region. This is further supported by the analysis of the G3 cogenic mice whose chromosomal segments were homozygous r209/r209 in the region of 4.4 cM centromeric to Lmp2 and heterozygous r209/nod in the different segment of the region of >4.4 cM developed diabetes at 20% incidence (2/10 mice) at the age of 250 days, suggesting a third MHC-linked diabetogenic gene outside of the MHC region.

FIGURE 2.

Development of diabetes in two congenic lines of the G3 intra-MHC recombinant NOD mice. Kaplan-Meier cumulative survival plot at 250 days of age. Two congenic lines with a different length of the r209/r209 segment centromeric to the MHC K gene were established from the G3 mice from the N12 intra-MHC recombinant NOD mice by intercrossing the BC11 mice. Line G3-A showed a centromeric boundary at >6.6 cM to the MHC K gene. Line G3-B showed a centromeric boundary at 4.4 cM to the MHC K gene. Left figure indicates female animals of G3-A (16 females), G3-B (10 females), and G1 (46 females, the whole genetic background: nod/nod) from the N12 mice. Right figure indicates male animals of G3-A (28 males), G3-B (8 males), and G1 (41 males).

FIGURE 2.

Development of diabetes in two congenic lines of the G3 intra-MHC recombinant NOD mice. Kaplan-Meier cumulative survival plot at 250 days of age. Two congenic lines with a different length of the r209/r209 segment centromeric to the MHC K gene were established from the G3 mice from the N12 intra-MHC recombinant NOD mice by intercrossing the BC11 mice. Line G3-A showed a centromeric boundary at >6.6 cM to the MHC K gene. Line G3-B showed a centromeric boundary at 4.4 cM to the MHC K gene. Left figure indicates female animals of G3-A (16 females), G3-B (10 females), and G1 (46 females, the whole genetic background: nod/nod) from the N12 mice. Right figure indicates male animals of G3-A (28 males), G3-B (8 males), and G1 (41 males).

Close modal

The data suggest that, in addition to the MHC class II genes, there is a second major MHC-linked diabetogenic gene(s) in the region of 4.4 cM centromeric to Lmp2, which includes the MHC class I K gene. This second gene delays the onset of diabetes but does not abrogate the disease in the animals. Thus, it may not be the MHC K gene, since NOD mice containing the MHC class I Ld transgene was protective against diabetes (11). A third MHC-linked diabetogenic gene may be located outside of the MHC region. Our results clearly demonstrate that the regions immediately flanking the MHC contains susceptibility genes to the development of diabetes. Given the importance of the MHC region, and notably the MHC class II alleles in many other autoimmune diseases, it is possible that the new gene(s) may influence the other diseases as well as diabetes.

We thank Chie Owa for her technical support and Eric Lander for his encouragement.

1

This work was supported by grants from Juvenile Diabetes Foundation International (JDFI), Diabetes Research and Wellness Foundation, British Diabetes Association, and National Institutes of Health (RO1 DK43613, DK48825, and P30 DK36836). E.Y., M.F., and T.T. were recipients of the JDFI postdoctoral fellowship. T.F. is a recipient of Postdoctoral Fellowship for Research Abroad of Japan Society for the Promotion of Science.

3

Abbreviation used in this paper: NOD, nonobese diabetic.

1
Makino, S., K. Kunimoto, Y. Muraoka, Y. Mizushima, K. Katagiri, Y. Tochino.
1980
. Breeding of a non-obese, diabetic strain of mice.
Exp. Anim.
29
:
1
2
Hattori, M., J. Buse, R. A. Jackson, L. Glimcher, M. E. Dorf, M. Minami, S. Makino, K. Moriwaki, H. Kuzuya, H. Imura, et al
1986
. The NOD mouse: recessive diabetogenic gene in the major histocompatibility complex.
Science
231
:
733
3
Prochazka, M., E. H. Leiter, D. V. Serreze, D. L. Coleman.
1987
. Three recessive loci required for insulin-dependent diabetes in non-obese diabetic (NOD) mice.
Science
237
:
286
4
Wicker, L. S., B. J. Miller, L. Z. Coker, S. E. McNally, S. Scott, Y. Mullen, M. C. Appel.
1987
. Genetic control of diabetes and insulitis in the nonobese diabetic (NOD) mouse.
J. Exp. Med.
165
:
1639
5
Ikegami, H., S. Makino, E. Yamato, Y. Kawaguchi, H. Ueda, T. Sakamoto, K. Takekawa, T. Ogihara.
1995
. Identification of a new susceptibility locus for insulin-dependent diabetes mellitus by ancestral haplotype congenic mapping.
J. Clin. Invest.
96
:
1936
6
Podolin, P. L., P. Denny, C. J. Lord, N. J. Hill, J. A. Todd, L. B. Peterson, L. S. Wicker, P. A. Lyons.
1997
. Congenic mapping of the insulin-dependent diabetes (Idd) gene, Idd10, localizes two genes mediating the Idd10 effect and eliminates the candidate Fcgr1.
J. Immunol.
159
:
1835
7
Lund, T., L. O’Reilly, P. Hutchings, O. Kanagawa, E. Simpson, R. Gravely, P. Chandler, J. Dyson, J. K. Picard, A. Edwards, D. Kioussis, A. Cooke.
1990
. Prevention of insulin-dependent diabetes mellitus in non-obese diabetic mice by transgenes encoding modified I-Aβ chain or normal I-Eα-chain.
Nature
345
:
727
8
Slattery, R. M., L. Kjer-Nielson, J. Allison, B. Charlton, T. E. Mandel, J. F. A. P. Miller.
1990
. Prevention of diabetes in non-obese diabetic I-Ak transgenic mice.
Nature
345
:
724
9
Miyazaki, T., M. Uno, M. Uehira, H. Kikutani, T. Kishimoto, M. Kimoto, H. Nishimoto, J-I. Miyazaki, K-I. Yamamura.
1990
. Direct evidence for the contribution of the unique I-ANOD to the development of insulitis in non-obese diabetic mice.
Nature
345
:
722
10
Nishimoto, H., H. Kikutani, K. Yamamura, T. Kishimoto.
1987
. Prevention of autoimmune insulitis by expression of I-E molecule in NOD mice.
Nature
328
:
432
11
Miyazaki, T., Y. Matsuda, T. Toyonaga, J-I. Miyazaki, Y. Yasaki, K-I. Yamamura.
1992
. Prevention of autoimmune insulitis in nonobese diabetic mice by expression of major histocompatibility complex class I Ld molecule.
Proc. Natl. Acad. Sci. USA
89
:
9519
12
Wicker, L. S., E. H. Leiter, J. A. Todd, R. J. Renjilian, E. Peterson, P. A. Fischer, P. L. Podolin, M. Zilska, R. Jaenisch, L. B. Peterson.
1994
. β2-microglobulin-deficient NOD mice do not develop insulitis or diabetes.
Diabetes
43
:
500
13
Serreze, D. V., E. H. Leiter, G. J. Christianson, D. Greiner, D. C. Roopenian.
1994
. Major histocompatibility complex class I-deficient NOD- β2mnull mice are diabetes and insulitis resistant.
Diabetes
43
:
505
14
Shiroishi, T., T. Koide, M. Yoshino, T. Sagai, K. Moriwaki.
1995
. Hotspots of homologous recombination in mouse meiosis.
Adv. Biophys.
31
:
119
15
Sagai, T., T. Shiroishi, K. Moriwaki, F. Bonhomme, M. L. Petras, M. Thohari, Z.-C. Yu, W.-S. Cho.
1986
. Characterization of newly lsolated monoclonal antibodies against MHC of a Japanese wild mouse.
Immunogenetics
24
:
361
16
Yoshino, M., T. Sagai, K. Fischer-Lindahl, Y. Toyoda, K. Moriwaki, T. Shiroishi.
1995
. Allele-demendent recombination frequency: homology requirement in meiotic recombination at the hot spot in the mouse major histocompatibility complex.
Genomics
27
:
298
17
Mizuno, K., T. Koide, T. Sagai, K. Moriwaki, T. Shiroishi.
1996
. Molecular analysis of a recombinational hotspot adjacent to Lmp2 gene in the mouse MHC: fine location and chromatin structure.
Mamm. Genome
7
:
490
18
Dietrich, W., H. Katz, S. E. Lincoln, H.-S Shin, J. Freidman, N. C. Dracopoli, E. S. Lander.
1992
. A genetic map of the mouse suitable for typing intraspecific crosses.
Genetics
131
:
423