Thymectomy and Radiation-Induced Type 1 Diabetes in Nonlymphopenic BB Rats Sheela Ramanathan,1,2,3 Marie-Therese Bihoreau,4 Andrew D. Paterson,5 Leili Marandi,1,2,3 Dominique Gauguier,4 and Philippe Poussier1,2,3

Spontaneous type 1 diabetes in BB rats is dependent on It remains unclear when and how this peripheral T- the RT1u MHC haplotype and homozygosity for an allele lymphopenia contributes to the development of diabetes, at the Lyp locus, which is responsible for a peripheral although its contribution is likely multifactorial. T-lymphopenia. Genetic studies have shown that there Our understanding of the pathogenic role of the BBDP are other, as yet unidentified, genetic loci contributing T-lymphopenia is further complicated by the demonstra- to diabetes susceptibility in this strain. BB rats carrying tion that autoimmune diabetes can develop in nonlym- wild-type Lyp alleles are not lymphopenic and are resis- tant to spontaneous diabetes (DR). Here we show that phopenic BBDR rats, a strain that is genetically related to thymectomy and exposure to one sublethal dose of BBDP rats (6,7; www-genome.wi.mit.edu/rat/public/). ␥-irradiation (TX-R) at 4 weeks of age result in the BBDR rats are not lymphopenic and do not develop rapid development of insulitis followed by diabetes in diabetes when maintained in a specific pathogen–free ؉ ؊ 100% of DR rats. Administration of CD4 45RC T-cells (SPF) environment (8). Although there is Ͼ15% genetic from unmanipulated, syngeneic donors immediately af- polymorphism between BBDP and BBDR rats (www- ter irradiation prevents the disease. Splenic T-cells genome.wi.mit.edu/rat/public/), spontaneous diabetes co- from TX-R–induced diabetic animals adoptively transfer type 1 diabetes to T-deficient recipients. ACI, WF, WAG, segregates as a single gene with the T-lymphopenia BN, LEW, PVG, and PVG.RT1u strains are resistant to between these two lines (9). Experimental induction of a TX-R–induced insulitis/diabetes. Genetic analyses re- peripheral T-lymphopenia in BBDR rats, through the ad- vealed linkage between regions on chromosomes 1, 3, 4, ministration of a depleting monoclonal antibody, cyclo- 6, 9, and 16, and TX-R–induced type 1 diabetes in a phosphamide, or sublethal ␥-irradiation, is followed by the ؋ cohort of nonlymphopenic F2 (Wistar Furth BBDP) rapid development of diabetes in a conventional environ- animals. This novel model of TX-R–induced diabetes in nonlymphopenic BB rats can be used to identify envi- ment (10). However, diabetes can also occur in unmanipu- ronmental and cellular factors that are responsible for lated BBDR rats after infection with Kilham’s rat virus the initiation of antipancreatic autoimmunity. Diabetes (KRV), a single-stranded DNA parvovirus with no tropism 51:2975–2981, 2002 for ␤-cells, or injection of polyinosinic-polycytidylic acid (poly[I:C]), an interferon-␣–inducing agent (11,12). Induc- tion of diabetes by administration of poly(I:C) is observed he BioBreeding (BBDP) rat spontaneously devel- both in SPF conditions and in conventional environment ops a T-cell–mediated, autoimmune diabetic syn- (12,13). Importantly, susceptibility to these experimentally drome that is similar to that observed in NOD induced type 1 diabetic syndromes is not restricted to BB- mice and humans (1). Two of the diabetes sus- related strains as long as the animals are haploidentical to T BB rats at the MHC class II locus (12,14). This observation ceptibility loci of the BB rat have been identified, Iddm1, which maps to the Lyp locus on chromosome 4, and suggests that diabetes susceptibility alleles are widespread Iddm2, which maps to the MHC class II haplotype RT1u of among laboratory rats. this animal (2,3). The Lyp allele carried by the BBDP rat This interpretation is further supported by the dem- shortens the life span of naı¨ve T-cells, resulting in a 5- to onstration that another type 1 diabetic syndrome can 10-fold decrease in the number of peripheral T-cells (4,5). be induced in various strains of rats, including some that do not carry the RT1u MHC haplotype (15). Specif-

From the 1Sunnybrook and Women’s College Health Sciences Centre, Univer- ically, adult thymectomy followed by four subsequent, sity of Toronto, Toronto, Ontario, Canada; the 2Department of Medicine, sublethal doses of ␥-irradiation (TX-R) given 2 weeks University of Toronto, Toronto, Ontario, Canada; the 3Department of Immu- apart results in the development of diabetes 10 weeks nology, University of Toronto, Toronto, Ontario, Canada; 4the Welcome Trust Centre for Human Genetics, University of Oxford, Oxford, U.K.; and the after irradiation. Strains susceptible to TX-R–induced 5Program in Genetics and Genomics Biology, Hospital for Sick Children, diabetes include PVG (RT1c), PVG (RT1u), WAG (RT1u), Toronto, Ontario, Canada. ϫ c/u Address correspondence and reprint requests to Philippe Poussier, Sunny- and (PVG WF)F1 (RT1 ) (15,16). Therefore, it seems brook and Women’s Health Sciences Centre, 2075 Bayview Ave., Room A3 38, that most of the experimentally induced type 1 diabetic Toronto, Ontario, Canada M4N 3M5. E-mail: [email protected]. syndromes require the BB rat MHC class II haplo- utoronto.ca. Received for publication 1 April 2002 and accepted in revised form 26 June type, but none requires the BB rat genetic background 2002. exclusively. Here we describe a novel model of experi- FACS, fluorescence-activated cell sorting; KRV, Kilham’s rat virus; mAb, monoclonal antibody; MNC, mononuclear cell; PE, phycoethrin; SPF, specific mentally induced diabetes that is restricted to nonlym- pathogen free; TX-R, ␥-irradiation; VAF, virus antibody free. phopenic, BB-related strains.

DIABETES, VOL. 51, OCTOBER 2002 2975 THYMECTOMY AND RADIATION-INDUCED TYPE 1 DIABETES

RESEARCH DESIGN AND METHODS TABLE 1 Animals. Diabetes-resistant BBDR and diabetes-prone BBDP rats were TX-R induces type 1 diabetes in nonlymphopenic rats with a BB purchased from BRM (Worchester, MA) and maintained in our colony under genetic background SPF and virus antibody–free (VAF) conditions. Specifically, all immunocom- petent sentinels remained serologically negative for the KRV as well as for Mean day other viruses (SDAV, Sendai, PVM, Reo3, TMEV, GDVII, MAD1, parvovirus, MHC Type 1 of onset and LCMV) during the course of the study. Diabetes-prone BB.7b rats that are Strain haplotype diabetes Insulitis after R congenic for the RT7b allele of rat CD45 were derived in our laboratory by u/u introducing the RT7b allele of Wistar Furth (WF) rats into BBDP rats, followed BBDR RT1 31/31 NA 28 Ϯ 6 u/u by Ͼ10 backcrosses to BBDP rats (17). The cumulative incidence of diabetes DR.Lyp/ϩ RT1 6/6 NA 34 Ϯ 5 u/u in BB.7b animals is similar to that observed in BBDP rats originating from the Non-lyp BB/W RT1 7/7 NA 31 Ϯ 3 BRM colony (data not shown). Nonlymphopenic and hence diabetes-resistant ACI RT1a/a 0/5 0/5 (nonlyp BB/W) rats have been generated by introgressing the wild-type Lyp RT1u/u 0/7 0/7 allele from BBDR into BBDP rats, followed by systematic backcrossing to BN RT1n/n 0/5 0/5 BBDP rats. Nonlyp BB/W animals used in this study were the progeny of the LEW RT1l/l 0/4 0/4 seventh backcross. Diabetes-prone and lymphopenic DR.Lyp/Lyp as well as PVG RT1c/c 0/4 0/4 Ϯ diabetes-resistant and nonlymphopenic DR.Lyp/ congenic lines were ob- RT1u/u 1/6 1/5 47 tained from Dr. A. Lernmark (Washington University, Seattle, WA) and have WAG RT1u/u 0/3 0/3 been previously described (9). Other rat strains were purchased from Harlan u/u u u WF RT1 0/9 0/9 Sprague-Dawley (Indianapolis, IN). ACI rats congenic for RT1 and ACI.1 .lyp u/u u (WF ϫ BBDP) RT1 7/8 1/1 48 Ϯ 8 rats congenic for both RT1 and the Lyp allele of BBDP rats have been u/u previously described (18). F and F rats were generated in our animal colony. (BBDP ϫ WF) RT1 4/4 NA 45 Ϯ 4 1 2 u/u Thymectomy, sham thymectomy, and whole-body irradiation of rats were (WF ϫ BBDR) RT1 6/16 5/10 37 Ϯ 1 performed as previously described (5). Rats were tested three times a week (WF ϫ DR.Lyp/ϩ) RT1u/u 8/11 0/3 35 Ϯ 5 for the presence of glycosuria and ketonuria. Once the animals became (ACI.1* ϫ BBDR) RT1u/u 0/4 0/4 glycosuric, the diagnosis of type 1 diabetes was made on the basis of (ACI.1u ϫ BBDR) RT1u/u 0/5 0/5 hyperglycemia (blood glucose Ͼ16.7 mmol/l) for two consecutive days. Diabetic rats were treated with subcutaneous implants of insulin (Linplant; *Insulitis in nondiabetic animals. NA, not applicable. University of Toronto, ON, Canada). After the rats were killed, , lung, kidney, and liver were fixed in 10% formalin for histology. All of the animal protocols were approved by the Animal Care Committee of our institution. noted. Probabilities were independently confirmed using ANOVA. Basically, it Monoclonal antibodies, three-color immunofluorescence, and fluores- is an ANOVA test followed by permutation tests that provide exact signifi- cence-activated cell sorter analysis. The monoclonal antibodies (mAbs) cance thresholds for each pair of genotype/phenotype, regardless of the used in this study have been previously described (5). Suspensions of MNC phenotype distribution in the cross. This method that is therefore particularly were incubated with a biotinylated mAb, followed by streptavidin-phycoethrin appropriate for nonparametric analyses was developed by Churchill and (PE)/Texas Red Tandem. PE-labeled and FITC-conjugated mAbs were then Doerge (22) and was used recently by Martin et al. (23) in their QTL analyses. added simultaneously. Cells were then analyzed by flow cytometry on a We have conformed to the guidelines proposed by Lander and Kruglyak (24) Becton Dickinson (San Jose, CA) FACScalibur or sorted using a Moflo in interpreting the statistical significance of the findings. (Cytomation, Denver, CO). At least 104 cells/sample were acquired for analysis. RESULTS Isolation of T-cell subpopulations. T-cells were enriched by negative selection using a rosetting procedure as previously described (19,20). The Thymectomy and sublethal irradiation induce diabe- purity of T-cells obtained from nonlymphopenic animals by rosetting was tes in BBDR rats in an age-dependent manner. In the routinely Ͼ98%. Furthermore, different subsets of CD4ϩ T-cells and T- course of adoptive transfer experiments using bone mar- depleted splenocytes were purified by fluorescence-activated cell sorting row radiation chimeras, we observed that BBDR rats (FACS). ϫ exposed to TX-R rapidly developed diabetes with a high Genetic analyses. The T-lymphopenic status of F2(WF BBDP) animals was determined on peripheral blood by immunofluorescence and FACS analysis. incidence. We decided to further characterize this exper- Genomic DNA was prepared from tail snips using standard methods. Nonlym- imentally induced diabetic syndrome. phopenic F animals (n ϭ 102) were followed for type 1 diabetes and insulitis 2 When 4-week-old BBDR rats were thymectomized and, 1 induced by TX-R. The animals were killed when they developed type 1 diabetes or 2 months after irradiation. After the rats were killed, the pancreata week later, received one sublethal dose of 5 Gy of TX-R, were fixed in 10% formalin for histological analysis. All microsatellite primers 100% of the animals (31 of 31) developed diabetes 21–35 used in this study were obtained from Genosys, Sigma (Cambridge, U.K.). The days after irradiation with a mean onset of 28 Ϯ 6 days genetic map location of the markers was obtained from the Whitehead (Table 1). Both sexes were susceptible. The diabetic Institute/MIT Center for Genome Research web site (www.genome.wi.mit- syndrome was characterized by the acute development of .edu/rat/public). Primers were tested against parental DNA samples to confirm the polymorphism between BBDP and WF rats. Polymorphic primers were polyuria, polydipsia, weight loss, hyperglycemia, glycos- then used in a genome-wide screen. Information on the primers used and uria, and ketonuria. Diabetic animals required daily insulin tested is available at www.well.ox.ac.uk/rat_mapping_resources/. The ge- injections to survive. Prospective, histological analysis of nome scan was performed at ϳ15- to 20-cM interval. Genotyping was the pancreas performed weekly after irradiation showed performed by PCR amplification on 50 ng of genomic DNA. PCR products were separated by electrophoresis on standard denaturing sequencing gels that diabetes was preceded by the development of insulitis and transferred onto nylon membranes. The membranes were hybridized with (Fig. 1). Infiltration of pancreatic islets by MNCs was a late a primer labeled with (␣-32P) dCTP using terminal transferase (21). Genotypes and rapid event that became detectable only a few days were independently scored by two observers (M.-T.B. and S.R.). Crosses were before and disappeared rapidly after the onset of diabetes, validated by verifying the linkage of lymphopenia with the appropriate leaving in place end-stage islets with no insulin-containing markers on chromosome 4 (2). Statistical analyses. A genetic map was built using MAPMAKER EXP based cells (data not shown). No inflammation was observed in on the order provided at www.genome.wi.mit.edu/rat/public. Quantitative the exocrine pancreas, lungs, kidneys, and liver of thymec- trait linkage analysis was performed using the MAPMAKER EXPv3.0 and tomized and irradiated rats. QTLv1.1suite of programs (www-genome.wi.mit.edu/genome_software/). The As illustrated in Table 2, susceptibility to TX-R–induced marker data were analyzed for linkage to each of the traits studied using MAPMAKER QTL, with the genetic model with the highest LOD score chosen diabetes was age dependent. Specifically, thymectomy had as the best model. The appropriate LOD score, proportion of the trait variance to be performed between 3 and 5 weeks, and sublethal explained at the locus, and the markers that showed LOD scores Ͼ1.5 were irradiation had to be performed within 1 week after

2976 DIABETES, VOL. 51, OCTOBER 2002 S. RAMANATHAN AND ASSOCIATES

TABLE 3 Adoptive transfer of type 1 diabetes by T-cells from diabetic TX-R, RT7a, BBDR rats Donor cells Recipient IDDM Insulitis* None 5Gy BB-7b 0/5 0/5 Splenic T-cells 5Gy BB-7b 4/4 NA WAG rnu/rnu 4/4 NA T-depleted splenocytes 5Gy BB-7b 0/4 0/4 T-cells were enriched by rosetting and were injected intravenously into WAG rnu/rnu rats or 4-week-old, sublethally irradiated BB7b rats. *Insulitis in nondiabetic animals.

onset of diabetes. Both spontaneous diabetes in BBDP rats and the diabetic syndrome induced in PVG.RT1u rats by adult thymectomy and multiple, low doses of TX-R are also associated with T-lymphopenia, and, in both cases, it FIG. 1. Progression of insulitis after TX-R in BBDR rats. Hematoxylin has been demonstrated that the lymphopenia plays a and eosin staining of pancreatic sections of BBDR rats thymectomized central role in disease pathogenesis (25,26). Furthermore, and irradiated at 1 month of age (A, B, and C) and at 2 months of age it has been shown that a lack of regulatory T-cells is one of (D). Pancreatic tissue was collected 2 (A),3(B),4(C), and 6 (D) weeks after irradiation. the pathogenic mechanisms of the T-lymphopenia in both BBDP and PVG.RT1u rats (25,26). Specifically, reconstitu- thymectomy. Exposure to TX-R or thymectomy alone tion of prediabetic rats with RT6ϩ, CD4ϩ T-cells, in the consistently failed to induce diabetes. case of BBDP rats, and CD45RCϪ, CD4ϩ T-cells, in the TX-R–induced diabetes is a T-cell–mediated autoim- case of PVG.RT1u animals, prevented diabetes (25,26). mune disease. The presence of insulitis before the devel- These observations led us to assess the ability of various opment of the disease strongly suggested that TX-R– T-cell subsets to modulate TX-R–induced diabetes in induced diabetes was autoimmune in nature. To determine BBDR rats. TX-R rats received unfractionated T-cells or whether this is the case, we performed an adoptive purified T-cell subsets isolated from adult, unmanipulated transfer of MNC from acutely diabetic TX-R rats to nondi- BBDR donors immediately after irradiation. The T-cell abetic recipients. We used recipients that were both MHC subsets consisted of CD4ϩ, CD8ϩ, CD45RCϪ CD4ϩ,or identical to the donors and T-cell–deficient, WAG rats CD45RCϩ CD4ϩ T-cells. As illustrated in Table 4, as few as homozygous for the nude allele, and 4-week-old, suble- 2 ϫ 105 unfractionated T-cells, CD4ϩ T-cells, and thally irradiated BB-7b rats. Adoptively transferred popu- CD45RCϪ CD4ϩ T-cells afforded protection from diabetes lations of lymphocytes (2 ϫ 106 cells intravenously) in 100% of the recipients. In contrast, reconstitution of included sorted, CD3Ϫ splenocytes and splenocytes en- TX-R BBDR rats with up to 5 ϫ 106 CD45RCϩ CD4ϩ T-cells riched (70–80%) in T-cells by rosetting. All of the recipi- or up to 2 ϫ 106 CD8ϩ T-cells was not protective. ents of splenic T-cells developed diabetes within 4 weeks Importantly, the ability of unfractionated T-cells to pre- after transfer (Table 3), whereas none of the animals that vent diabetes was lost when T-cell reconstitution was received CD3Ϫ splenocytes or were left untreated did. No delayed by 1 and 2 weeks, suggesting that the autoimmune insulitis was found in the recipients that had not become process is initiated soon after irradiation and/or expansion diabetic and were killed 8 weeks after transfer (data not of regulatory T-cells is required before the initiation of the shown). These results demonstrate that TX-R–induced diabetogenic process. These results demonstrate that the diabetes is a T-cell–mediated autoimmune disease. diabetic syndrome induced by TX-R in BBDR rats is .Syngeneic CD45RC؊, CD4؉ T-cells prevent induction amenable to T-cell regulation of TX-R–induced diabetes. As expected, TX-R resulted in peripheral T-lymphopenia in BBDR rats. Specifically, T-cells accounted for 6 Ϯ 1.3% and 22 Ϯ 4% (n ϭ 7) of splenic and MNCs, respectively (Fig. 2), at the

TABLE 2 TX-R–induced type 1 diabetes is age-dependent in BB rats

Age (weeks) Type 1 TX 5 Gy diabetes Insulitis* 4 5 31/31 NA 4 8 0/5 0/5 4 6 0/5 0/5 6 8 0/5 0/5 8 9 0/4 0/5 4 — 0/4 0/4 FIG. 2. Proportion of T-cells among splenic and lymph node MNC in one — 5 0/5 0/5 representative TX-R BBDR rat at the time of diabetes onset. Unfrac- tionated MNC were stained with FITC-R73, an mAb specific for the rat *Insulitis in nondiabetic animals. TcR␣␤, and analyzed flow cytometrically.

DIABETES, VOL. 51, OCTOBER 2002 2977 THYMECTOMY AND RADIATION-INDUCED TYPE 1 DIABETES

TABLE 4 TABLE 5 TX-R–induced diabetes is prevented by syngeneic T-cells Genetic linkage analysis of TX-R–induced type 1 diabetes Day of Type 1 Type 1 Cells injected Cell no. injection* diabetes Insulitis† Marker Distance* diabetes P Insulitis† P None — 0 24/24 NA D1Rat1 9.1 2.78 0.0017 T-cells 5 ϫ 106 0 0/6 0/6 D3Rat34 42.8 2.25 0.0056 2 ϫ 106 0 0/3 0/3 D4Rat29 37.4 2.60 0.0025 2 ϫ 105 0 0/3 0/3 D6Mit1 30.1 1.63 0.0232 5 ϫ 106 7 2/3 1/1 D6Rat94 80.5 2.04 0.0091 5 ϫ 106 14 3/3 NA D6Rat19 52.2 1.84 0.0143 1.62 0.0232 CD4ϩ T-cells 2 ϫ 106 0 0/3 0/3 D9Rat23 40.0 2.15 0.0070 1.62 0.0241 2 ϫ 105 0 1/3 2/2 D16Rat53 28.3 2.06 0.0087 2.81 0.0015 CD4ϩ CD45RCϪ 5 ϫ 106 0 0/5 0/5 ϫ 5 P values are of linkage as assessed by independently calculated 2 10 0 0/3 0/3 ϫ ϩ ϩ ϫ 6 ANOVA (22, 23). *The distances (cM) are as reported for (SHR BN) CD4 CD45RC 5 10 0 2/2 NA cross at the Whitehead Institute site. †LOD scores. CD8ϩ T-cells 2 ϫ 106 0 2/3 1/1

After irradiation, TX-R BBDR rats received an intravenous injection trait with variable penetrance. However, none of the nine of the indicated T-cell subset isolated from 3-month-old, unmanipu- ϫ lated BBDR rats. *The day of irradiation is considered as day 0; F1(BBDR congenic ACI) animals became diabetic. Al- ϫ †insulitis in nondiabetic animals. though the number of these F1(BBDR congenic ACI) is low, the lack of diabetics among them is consistent with TX-R–induced diabetes is observed only in BBDP- the interpretation that the ACI background carries factors related strains. We next examined whether susceptibility of resistance to TX-R–induced diabetes. Importantly, we to TX-R–induced diabetes is genetically determined by have previously provided evidence for the presence of exposing nonlymphopenic, BBDP-related strains as well factors of resistance to spontaneous diabetes in the ACI as other BB-unrelated strains to TX-R. Several inbred genetic background (18). strains of nonlymphopenic, RT1u rats, including BBDR In a second step, we performed a segregation analysis of animals, were derived from BBDP animals at various times diabetes and insulitis in a cohort of 102 nonlymphopenic ϫ and for different purposes. These strains exhibit ϳ15% of F2(WF BBDP) rats. We selected these parental strains u genetic polymorphism across the whole genome with for several reasons. Both strains carry the RT1 haplotype. BBDP rats (www.genome.wi.mit.edu/rat/public). The two The reported degree of genetic polymorphism between BBDP-related strains that were studied were nonlyp BB/W WF and BBDP rats, ϳ35%, is relatively high (www. and DR.Lyp/ϩ rats (9). Two other groups of strains genome.wi.mit.edu/rat/public). The highest incidence of ϫ unrelated to BBDP rats were tested for susceptibility to TX-R–induced diabetes among F1(WF BBDP-related) TX-R-induced diabetes. One group, comprised of WF, animals was observed in the progeny of WF ϫ BBDP WAG, PVG.RT1u, and ACI.1u/u rats carry the same MHC crosses. Animals were exposed to TX-R at 4 weeks and RT1u haplotype as BBDP and BBDR rats. The other group followed prospectively for diabetes up to 10 weeks after was composed of animals carrying non–u RT1 haplotypes, irradiation. Inheritance of the T-lymphopenia showed l n a namely LEW (RT1 ), BN (RT1 ), ACI (RT1 ), and PVG Mendelian segregation because 25.2% of the F2 animals (RT1c) rats. All of the rats were followed for up to 3 were lymphopenic. months after TX-R. A genome-wide scan of F2 animals was performed using After TX-R, 100% of nonlymphopenic, BBDP-related 117 markers polymorphic between BBDP and WF. Regions animals developed diabetes with comparable kinetics (Ta- on chromosome 5 (between D5Rat108 and D5Rat50, ϳ30 ble 1). In contrast, none of the BBDP-unrelated animals cM), chromosome 10 (top to D10Rat85, ϳ40 cM), and became diabetic, except for one PVG.RT1u rat (one of six). chromosome 14 (top to D14Rat24, ϳ28 cM) were not None of the nondiabetic rats had insulitis at the time they analyzed because of the lack of polymorphic markers were killed (data not shown). These results demonstrate distinguishing the two parental strains at these sites. that nonlymphopenic animals that are genetically related Additional markers were tested in areas that showed some to the BBDP strain are uniquely susceptible to TX-R– evidence for linkage. induced diabetes. At this stage and in the absence of Diabetes and insulitis developed in 43.1 and 81% of the nonlymphopenic, BBDP-related rats congenic for a non–u animals, respectively. Both sexes were equally repre- RT1 haplotype, the role of MHC genes in susceptibility to sented in each of the phenotypic categories (data not TX-R–induced diabetes remains unknown. shown). All of the animals were BB/WF or WF/WF in the Genetic linkage analysis of TX-R–induced diabetes. Lyp region on chromosome 4 (D4Got54) because lym- We decided to characterize further the genetic basis of the phopenic animals were excluded from the study. The susceptibility of nonlymphopenic, BBDP-related rats to microsatellite markers that showed significant linkage to TX-R–induced diabetes. In a first step, we determined the diabetes and/or insulitis with LOD scores Ͼ1.5 are shown in Table 5. Linkages to diabetes and insulitis were ob- incidence of diabetes in F1 animals resulting from a cross between BBDP-related and either WF- or RT1-congenic served on chromosomes 1, 3, 4, 6, 9, and 16. ACI rats (Table 1). Diabetes was observed in a large ϫ DISCUSSION proportion of F1(WF BBDP-related) and F1(BBDP- related ϫ WF) animals, demonstrating that in these This study describes a novel, autoimmune, type 1 diabetic crosses, TX-R–induced diabetes is inherited as a dominant syndrome that can be rapidly induced in BBDP-related

2978 DIABETES, VOL. 51, OCTOBER 2002 S. RAMANATHAN AND ASSOCIATES strains with a very high incidence. Many features of this for exposure to SPF conditions. However, it has been diabetic syndrome distinguish it from those induced in shown in models of autoimmune thyroiditis and diabetes RT1u strains through viral infection, administration of poly induced by thymectomy and multiple low doses of TX-R I:C alone or in combination with anti-RT6 antibody, or in that the contains regulatory T-cells that prevent PVG rats by thymectomy and multiple low doses of TX-R autoimmunity (28). Furthermore, in the case of autoim- (15,26). Susceptibility to diabetes induced by KRV and mune thyroiditis, the development of these regulatory poly I:C is widely distributed in nonlymphopenic, RT1u- T-cells required the presence of a functional gland expressing strains, whereas TX-R–induced diabetes seems (29). to be restricted to BB-related strains (11,12). The latter One of the intriguing aspects of spontaneous and exper- syndrome may therefore prove helpful in identifying dia- imentally induced diabetes in BBDP-related strains is that betes susceptibility factors that are peculiar to the BB the various manipulations that prevent or precipitate genetic background and possibly contribute to both spon- diabetes have to be performed before 4 weeks (30,31). taneous and experimentally induced diseases. Specifically, reconstitution of diabetes-prone BBDP rats The diabetic syndrome induced by thymectomy and with normal T-cells protects the recipients from diabetes, multiple low doses of TX-R occurs in PVG animals, inde- provided that the protective T-cells are injected in the first pendent of their RT1 haplotype, but also in WAG rats, two 4 weeks of life (30). Thymectomy of BBDP rats prevents strains shown here to be resistant to TX-R–induced diabe- diabetes when performed in Յ4-week-old animals but tes (Table 1). Adoptive transfer of diabetes induced by does not affect the time course and incidence of the thymectomy and multiple low doses of TX-R to irradiated, disease when delayed beyond that age (31; S. Ramanathan syngeneic recipients by splenic T-cells was unsuccessful and P. Poussier, unpublished observations). There is evi- despite preactivation of donor T-cells by ConA in vitro dence in the NOD mouse that activation of diabetogenic (16). Furthermore, only a small proportion of the T-cell T-cells by their specific ␤-cell antigens occurs in pancre- recipients developed lesions of insulitis. In contrast, all atic lymph nodes around the age of 2 weeks (32). Whether recipients of splenic T-cells freshly isolated from TX-R– this early and potentially deleterious presentation of islet induced diabetic donors developed diabetes in a few cell antigens is a physiological phenomenon remains an weeks. The time constraints for successful induction of open question. The present study and others have demon- diabetes through thymectomy and multiple low doses of strated that potentially diabetogenic T-cells are present in TX-R seem less stringent than those required for TX-R– the pool of recirculating T-cells of unmanipulated BBDR induced diabetes. Specifically, induction of the former rats (17,26). Presentation of self-antigens to their specific diabetic syndrome requires that thymectomy be per- T-cells, if it occurs in unmanipulated BBDR rats, must formed between 3 and 6 weeks and irradiation be initiated result in T-cell tolerance or ignorance because these 2 weeks later, whereas in the case of the latter diabetic animals remain diabetes free. Here we provide evidence, syndrome, thymectomy has to be performed in animals although indirect, that either presentation of islet cell that are Յ4 weeks old and irradiation cannot be postponed antigens is restricted to the first 4 weeks of life or, in the beyond 1 week after thymectomy (Table 2). It remains, event that this self-presentation persists beyond 4 weeks, however, that in both diabetic syndromes, the experimen- it can no longer prime a deleterious autoimmune response tal procedure seems to affect the development and/or in BBDP-related strains. function of regulatory T-cells with the CD45RCϪ CD4ϩ Assuming that potentially deleterious presentation of TcR␣␤ϩ membrane phenotype because reconstitution of ␤-cell antigens persists throughout the life of BBDR rats, this subset immediately after irradiation prevents the our results suggest that thymectomy followed by TX-R has development of the disease (26) (Table 4). a differential effect on the homeostasis and/or repertoire of It has been previously reported that type 1 diabetes can peripheral T-cells in young and adult animals. CD45RCϪ be induced in BBDR rats by a single sublethal dose of TX-R CD4ϩ T-cells shown here and in another model of diabetes (10). In our hands, this protocol remained unsuccessful, to prevent autoimmunity account for a low proportion of independent of the dose of irradiation and age of the peripheral T-cells in young animals (33). The proportion of animals (Table 2 and data not shown). We believe that this this regulatory T-cell subset increases as the contribution discrepancy is related to environmental factors. The of recent thymic emigrants to the pool of recirculating Worcester colony of BBDR rats, the source of our animals, T-cells decreases with age. It is not implausible that the became VAF through cesarean rederivation in 1990. Be- differential effect of thymectomy and TX-R on diabetes fore that transfer, the animals were kept in SPF conditions susceptibility in young and adult animals is related to (27). Changes in the susceptibility of BBDR rats to exper- age-related changes in the repertoire of peripheral T-cells. imentally induced diabetic syndromes were associated However, the peripheral T-lymphopenia is so severe im- with the breeding of these animals in VAF conditions. For mediately after irradiation that we could not detect reli- example, treatment of BBDR rats with a depleting mAb able differences in the proportions of naı¨ve and memory specific for RT6, an ADP-ribosyltransferase expressed on T-cells between young and adult rats. the surface of most mature rat T-cells, was diabetogenic in Genetic analysis of TX-R–induced diabetes showed SPF conditions but not in a VAF environment (27). Simi- weak linkage of the disease to regions on Ch 1, 3, 4, 6, 9, larly, induction of diabetes in BBDR rats by a single and 16 (Table 5). This absence of robust linkages suggests sublethal dose of TX-R was possible at a time when these that the development of insulitis and clinical disease is animals were maintained in SPF conditions but remained under the regulation of multiple genes and that most of unsuccessful in our VAF facility (10) (Table 2). It remains these non-MHC loci make only incremental contribution to unclear at this stage how thymectomy can be substituted the diabetogenic process. Our analysis relied on complex

DIABETES, VOL. 51, OCTOBER 2002 2979 THYMECTOMY AND RADIATION-INDUCED TYPE 1 DIABETES phenotypes, diabetes and insulitis, that most likely result 12. Ellerman KE, Like AA: Susceptibility to diabetes is widely distributed in from the action of many genes involved in complex normal class IIu haplotype rats. Diabetologia 43:890–898, 2000 13. Sobel DO, Newsome J, Ewel CH, Bellanti JA, Abbassi V, Creswell K, Blair pathways, and, furthermore, it was performed in a limited O: Poly I:C induces development of diabetes mellitus in BB rat. Diabetes number of F2 animals. It is therefore not surprising that the 41:515–520, 1992 influence of individual loci proved difficult to assess. 14. Whalen BJ, Doukas J, Mordes JP, Rossini AA, Greiner DL: Induction of Similar difficulties in genome-wide mapping of diabetes insulin-dependent diabetes mellitus in PVG.RT1u rats. Transplant Proc susceptibility loci have also been observed in other human 29:1684–1685, 1997 and murine diabetic syndromes for similar reasons (34– 15. Penhale WJ, Stumbles PA, Huxtable CR, Sutherland RJ, Pethick DW: Induction of diabetes in PVG/c strain rats by manipulation of the immune 36). One way to overcome these difficulties is to identify system. Autoimmunity 7:169–179, 1990 preclinical, disease-related phenotypes and determine the 16. Stumbles PA, Penhale WJ: IDDM in rats induced by thymectomy and genetic loci that govern them. This approach has helped in irradiation. Diabetes 42:571–578, 1993 identifying Idd loci that control various stages of insulitis 17. Ramanathan S, Poussier P: reconstitution of BB/W rats after the and progression to diabetes in the NOD model (37). An initiation of insulitis precipitates the onset of diabetes. J Immunol 162:5134–5142, 1999 additional factor that complicates the search for diabetes- 18. Colle E, Fuks A, Poussier P, Edouard P, Guttman RD: Polygenic nature of susceptibility loci in rats is that several inbred strains of spontaneous diabetes in the rat: permissive MHC haplotype and presence ϩ RT1u animals seem to carry susceptibility alleles to of the lymphopenic trait of the BB rat are not sufficient to produce experimentally induced diabetes (12). Outside the MHC susceptibility. Diabetes 41:1617–1623, 1992 class II and Lyp loci, 4–6 Iddm loci have been linked to 19. Metroz-Dayer M, Mouland A, Brideau C, Duhamel D, Poussier P: Adoptive transfer of diabetes in BB rats induced by CD4 T lymphocytes. Diabetes spontaneous diabetes (38–41) and up to three to experi- 39:928–932, 1990 mentally induced diabetic syndromes (23,42). In our study, 20. Mason DW: Subsets of T cells in the rat mediating lethal graft-versus-host we did not observe linkage to any of these previously disease. Transplantation 32:322–326, 1981 reported Iddm loci, which suggests that variants at non- 21. Bihoreau MT, Gauguier D, Kato N, Hyne G, Lindpaintner K, Rapp JP, James MHC loci conferring risk to type 1 diabetic syndromes in MR, Lathrop GM: A linkage map of the rat genome derived from three F2 crosses. Genome Res 7:434–440, 1997 rats are not all identical. It is possible, however, that many 22. Churchill GA, Doerge RW: Empirical threshold values for quantitative trait of these genes function in conserved pathways central to mapping. Genetics 138:963–971, 1994 diabetes pathophysiology. 23. Martin AM, Blankenhorn EP, Maxson MN, Zhao M, Leif J, Mordes JP, Greiner DL: Non-major histocompatibility complex-linked diabetes sus- ceptibility loci on chromosomes 4 and 13 in a backcross of the DP-BB/Wor ACKNOWLEDGMENTS rat to the WF rat. Diabetes 48:50–58, 1999 This work was supported by a grant from the Canadian 24. Lander E, Kruglyak L: Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet 11:241–247, 1995 Institutes of Health Research. L.M. was supported by a 25. Rossini AA, Mordes JP, Greiner DL, Nakano K, Appel MC, Handler ES: postdoctoral fellowship from the government of Iran. cell transfusion in the Bio-Breeding/Worcester rat: prevention of We thank C. Cantin (Cytometrics) and Gisele Knowles diabetes, major histocompatibility complex restriction, and long-term (Sunnybrook and Women’s College Health Sciences Cen- persistence of transfused cells. J Clin Invest 77:1399–1401, 1986 ter) for expert assistance with flow cytometry. 26. Fowell D, Mason D: Evidence that the T cell repertoire of normal rats contain cells with the potential to cause diabetes: characterization of the CD4ϩ T cell subset that inhibit this autoimmune potential. J Exp Med REFERENCES 177:627–636, 1993 27. Thomas VA, Woda BA, Handler ES, Greiner DL, Mordes JP, Rossini AA: 1. Crisa L, Mordes JP, Rossini AA: Autoimmune diabetes mellitus in the BB Altered expression of diabetes in BB/Wor rats by exposure to viral rat. Diabetes Metab Rev 8:4–37, 1992 pathogens. Diabetes 40:255–258, 1991 2. Jacob HJ, Pettersson A, Wilson D, Mao Y, Lernmark A, Lander ES: Genetic 28. Saoudi A, Seddon B, Fowell D, Mason D: The thymus contains a high dissection of autoimmune type I diabetes in the BB rat. Nat Genet 2:56–60, frequency of cells that prevent autoimmune diabetes on transfer into 1992 (published erratum appears in Nat Genet 7:215, 1994) prediabetic recipients. J Exp Med 184:2393–2398, 1996 3. Colle E, Guttman RD, Seemayer TA: Spontaneous diabetes mellitus 29. Seddon B, Mason D: Peripheral autoantigen induces regulatory T cells that syndrome in the rat. 1. Association with the major histocompatibility prevent autoimmunity. J Exp Med 189:877–882, 1999 complex of the rat. J Exp Med 154:1237, 1981 30. Burstein D, Mordes JP, Greiner DL, Stein D, Nakamura N, Handler ES, 4. Zadeh HH, Greiner DL, Wu DY, Tausche F, Goldschneider I: Abnormalities Rossini AA: Prevention of diabetes in BB/Wor rat by single transfusion of in the export and fate of recent thymic emigrants in diabetes-prone BB/W rats. Autoimmunity 24:35–46, 1996 spleen cells: parameters that affect degree of protection. Diabetes 38:24– 5. Ramanathan S, Norwich K, Poussier P: Antigen activation rescues recent 30, 1989 thymic emigrants from programmed cell death in the BB rat. J Immunol 31. Like AA, Kislauskis E, Williams RR, Rossini AA: Neonatal thymectomy 160:5757–5764, 1998 prevents spontaneous diabetes mellitus in the BB/W rat. Science 216:644– 6. Like AA, Guberski DL, Butler L: Diabetic BioBreeding/Worcester (BB/Wor) 646, 1982 rats need not be lymphopenic. J Immunol 136:3254–3258, 1986 32. Hoglund P, Mintern J, Waltzinger C, Heath W, Benoist C, Mathis D: 7. Guberski DL, Thomas VA, Shek WR, Like AA, Handler ES, Rossini AA, Initiation of autoimmune diabetes by developmentally regulated presenta- Wallace JE, Welsh RM: Induction of type I diabetes by Kilham’s rat virus in tion of islet cell antigens in the pancreatic lymph nodes. J Exp Med diabetes-resistant BB/Wor rats. Science 254:1010–1013, 1991 189:331–339, 1999 8. Like AA, Guberski DL, Butler L: Influence of environmental viral agents on 33. Yang CP, Bell EB: Functional maturation of recent thymic emigrants in the frequency and tempo of diabetes mellitus in BB/Wor rats. Diabetes periphery: development of alloreactivity correlates with the cyclic expres- 40:259–262, 1991 sion of CD45RC isoforms. Eur J Immunol 22:2261–2269, 1992 9. Markholst H, Eastman S, Wilson D, Andreason BE, Lernmark A: Diabetes 34. Cordell HJ, Todd JA: Multifactorial inheritance in type 1 diabetes. Trends segregates as a single locus in crosses between inbred BB rats prone or Genet 11:499–504, 1995 resistant to diabetes. J Exp Med 174:297–300, 1991 35. Concannon P, Gogolin-Ewens KJ, Hinds DA, Wapelhorst B, Morrison VA, 10. Handler ES, Mordes JP, McKeever U, Nakamura N, Bernhard J, Greiner Stirling B, Mitra M, Farmer J, Williams SR, Cox NJ, Bell GI, Risch N, DL, Rossini AA: Effects of irradiation on diabetes in the BB/Wor rat. Spielman RS: A second-generation screen of the human genome for Autoimmunity 4:21–30, 1989 susceptibility to insulin-dependent diabetes mellitus. Nat Genet 19:292– 11. Ellerman KE, Richards CA, Guberski DL, Shek WR, Like AA: Kilham rat 296, 1998 triggers T-cell–dependent autoimmune diabetes in multiple strains of rat. 36. Mein CA, Esposito L, Dunn MG, Johnson GC, Timms AE, Goy JV, Smith Diabetes 45:557–562, 1996 AN, Sebag-Montefiore L, Merriman ME, Wilson AJ, Pritchard LE, Cucca F,

2980 DIABETES, VOL. 51, OCTOBER 2002 S. RAMANATHAN AND ASSOCIATES

Barnett AH, Bain SC, Todd JA: A search for type 1 diabetes susceptibility 40. Kloting I, van den Brandt J, Kuttler B: Genes of SHR rats protect genes in families from the United Kingdom. Nat Genet 19:297–300, 1998 spontaneously diabetic BB/OK rats from diabetes: lessons from congenic 37. Fox CJ, Paterson AD, Mortin-Toth SM, Danska JS: Two genetic loci BB.SHR rat strains. Biochem Biophys Res Commun 283:399–405, 2001 regulate T cell-dependent islet inflammation and drive autoimmune diabe- 41. Jackerott M, Hornum L, Andreasen BE, Markholst H: Segregation of tes pathogenesis. Am J Hum Genet 67:67–81, 2000 autoimmune type 1 diabetes in a cross between diabetic BB and brown 38. Kloting I, Kovacs P: Genes of the immune system cosegregate with the age Norway rats. J Autoimmun 10:35–41, 1997 at onset of diabetes in the BB/OK rat. Biochem Biophys Res Commun 42. Martin AM, Maxson MN, Leif J, Mordes JP, Greiner DL, Blankenhorn EP: 242:461–463, 1998 Diabetes-prone and diabetes-resistant BB rats share a common major 39. Kloting I, Vogt L, Serikawa T: Locus on chromosome 18 cosegregates with diabetes susceptibility locus, iddm4: additional evidence for a “universal diabetes in the BB/OK rat subline. Diabetes Metab 21:338–344, 1995 autoimmunity locus” on rat chromosome 4. Diabetes 48:2138–2144, 1999

DIABETES, VOL. 51, OCTOBER 2002 2981