Analysis of differential survival of syngeneic islets transplanted into hyperglycemic C57BL/6J versus C57BL/KsJ mice

C57BL/KsJ (BKs) male mice were more sensitive to diabetes induction by administration of multiple low-doses of streptozotocin (Sz) than were C57BL/6J (B6) male mice. Analysis of islet size and insulin content of the two parental strains did not indicate that differences in drug sensitivity could be attributed to an effect of genetic background on islet size or insulin content. 50 BKs islets implanted into the spleens of BKs male mice made diabetic by Sz were eliminated within 12 days posttransplantation, whereas an equal number of B6 islets implanted into the spleens of diabetic B6 recipients were retained, even though the numbers of islets implanted were insufficient to effect remission from hyperglycemia. In contrast to the rapid loss of islets implanted into spleens of hyperglycemic BKs recipients, BKs islets implanted into spleens of normoglycemic recipients were not eliminated, thus suggesting that the basis for the differential survival between the B6 and BKs strains reflected their ability to survive hyperglycemic stress rather than a differential ability to replicate. Since BKs beta cells have been shown to respond to hyperglycemia by expression of an endogenous retroviral gene that cannot be expressed by B6 beta cells, the possibility that this differential survival represents a strain difference in autoreactivity against islet cells is raised.     

Effect of genetic background on the therapeutic effects of dehydroepiandrosterone (DHEA) in diabetes-obesity mutants and in aged normal mice

Dehydroepiandrosterone (DHEA) was fed at 0.1-0.4% in the diet to genetically diabetic (db/db) or obese (ob/ob) C57BL/KsJ (BL/Ks) or C57BL/6J (BL/6) mice. Treatment of BL/Ks-db/db or ob/ob mice with 0.4% DHEA prevented hyperglycemia, islet atrophy, and severe diabetes associated with this inbred background, but did not affect weight gain and food consumption. Homozygous obese (ob) or diabetes (db) mice on the BL/6 background were more sensitive to DHEA, and the mild, transient hyperglycemia associated with ob or db gene expression on the BL/6 inbred background could be prevented by 0.1% DHEA. Both body weight and food consumption were decreased in BL/6 mutants maintained on 0.1% DHEA whereas this effect was not seen in BL/Ks mutants fed up to 0.4% DHEA. Early therapy with 0.4% DHEA, initiated at 2 wk of age, prevented the development of most diabetes symptoms and decreased the rate of weight gain in pups of all genotypes. In addition to therapeutic effects on both obese mutants, DHEA effected significant changes in an aging study using normal BL/6 female mice. Four weeks of DHEA treatment initiated at 2 yr of age improved glucose tolerance and at the same time reduced plasma insulin to a "younger" level. This suggests that DHEA may act in insulin-resistant mutant mice and in aging normal mice to increase the sensitivity to insulin.     

Immunoreactive neurotensin in the pancreas of genetically obese and diabetic mice. A longitudinal study

Immunoreactive neurotensin (IR-NT) content in 2 N acetic acid extracts of pancreas was measured in genetically diabetic (C57BL/KsJ db/db and ob/ob) and obese (C57BL/6J ob/ob and db/db) mice and normal littermate controls from 5 to 24 wk of age to determine the relationship of any changes to the development of metabolic abnormalities. Pancreatic IR-NT in obese mice showed no consistent change compared with lean littermate controls. In contrast, diabetic mice demonstrated an increase in pancreatic IR-NT that occurred at 6-8 wk of age, and maximal about the time of islet B-cell failure (8-10 wk), and persisted over the study period. Pancreatic IR-NT eluted in two peaks on reverse phase high-pressure liquid chromatography, one of which exhibited a retention time similar to that of synthetic NT. These findings suggest that pancreatic IR-NT concentration is regulated by insulin, with elevated levels occurring in association with insulin deficiency and its metabolic consequences but not with insulin resistance. Taken together with the previous demonstration that NT influences pancreatic islet hormone secretion, the present findings support a possible role of endogenous NT in islet hormone regulation.     

Temporal relationship of tissue somatostatin-like immunoreactivity to metabolic changes in genetically obese and diabetic mice

Somatostatin-like immunoreactivity (SRIF-LI) content in 2 N acetic acid extracts of hypothalamus, gastric antrum, and pancreas was measured in genetically obese (C57BL/6J ob/ob and db/db) and diabetic (C57BL/KsJ db/db and ob/ob) mice and normal littermate controls from 5 to 24 wk to determine the relationship of previously reported changes to the development of metabolic abnormalities. Hypothalamic SRIF-L concentration was similar in control, diabetic, and obese mice at all ages and increased progressively with age in all groups. Gastric antrum SRIF-LI was similar in all groups of mice at all ages. Obese mice gained weight progressively and showed moderate hyperglycemia and marked hyperinsulinemia from 5 wk of age. Pancreatic SRIF-LI content in obese (C57BL/6J) animals was similar to that in lean littermate controls, but pancreatic SRIF-LI concentration (expressed by weight or protein content) was decreased until 8 (6J ob/ob) and 10 (6J db/db) wk. Diabetic (C57BL/KsJ) mice showed a similar metabolic pattern until 10 wk with no change in pancreatic SRIF-LI content or concentration. Thereafter a progressive fall in serum insulin and a marked rise in serum glucose was associated with increasing pancreatic SRIF-LI content and concentration. These studies suggest that the genetically hyperphagic syndromes are unassociated with any change in hypothalamic or gastric SRIF-LI; that pancreatic SRIF-LI increases occur in response to, rather than as the cause of, relative hypoinsulinemia; and that the genetic background of the mice (KsJ or 6J) rather than the mutant gene (db or ob) determines the defect in carbohydrate metabolism and the pancreatic SRIF-LI response.     

Altered beta-endorphin, Met- and Leu-enkephalins, and enkephalin-containing peptides in pancreas and pituitary of genetically obese diabetic (db/db) mice during development of diabetic syndrome

We have recently shown that in addition to beta-endorphin the opioid peptides Met- and Leu-enkephalin and their apparent precursors are localized in islet endocrine cells of the rat pancreas. To begin evaluating a possible role for these pancreatic opiates in the pathophysiology of genetic diabetes in rodents, immunoreactive beta-endorphin and Met- and Leu-enkephalins were measured in acetic acid extracts of pancreas and pituitary of C57BL/KsJ db/db mice and their lean littermates. Groups of animals were studied during three phases of development of the diabetic syndrome in the mutant mice: at 4 (hyperinsulinemic and prediabetic); 6, 9, and 12 (frankly obese and diabetic); and 30 (hypoinsulinemic) wk of age. Elevations or decreases (P less than .05) were found in db/db mice (vs. lean littermates) as follows: pituitary content of Met-enkephalin was twofold higher at all ages studied; pituitary free Leu-enkephalin was lower at 4 wk and reversed to higher at 6-30 wk; pancreatic beta-endorphin was 30% lower at 4 wk and reversed to threefold higher at 6-12 wk; Met- and Leu-enkephalin-containing larger peptides were elevated at one or more points between 6 and 12 wk in both the pancreas and the pituitary. Thus, the onset of overt obesity between 4 and 6 wk of age was accompanied by a marked rise in both pancreatic beta-endorphin and pituitary Leu-enkephalin; similar elevations in these parameters have been reported previously in C57BL/6J ob/ob mice at approximately 12 wk of age.(ABSTRACT TRUNCATED AT 250 WORDS)     

Molecular mimicry between insulin and retroviral antigen p73. Development of cross-reactive autoantibodies in sera of NOD and C57BL/KsJ db/db mice

Enzyme-linked immunosorbent assay (ELISA) was used to study temporal development of murine autoantibodies against insulin and both type C and intracisternal type A retroviral antigens. The nonobese diabetic (NOD) mouse, a model for autoimmune, insulin-dependent diabetes, was compared with a related, but diabetes-resistant, strain, nonobese normal (NON). Similarly, C57BL/KsJ db/db mice (insulin-resistant model of insulin-dependent diabetes and obesity) were compared with diabetes-resistant C57BL/6 db/db mice. NOD mice developed much higher autoantibody titers than did NON mice. Whereas type C autoantibodies in NOD developed to peak titer shortly after mice were weaned, autoantibodies against insulin and p73 (group-specific antigen of the intracisternal type A particle) did not develop until shortly before, or concomitant with, the development of hyperglycemia. Two NOD mice not developing hyperglycemia during the 40-wk study period were distinguished from the mice developing diabetes by a delayed onset of insulin (but not p73) autoantibodies. Our findings suggest that in NOD mice, the appearance of insulin and p73 autoantibodies signifies that extensive underlying necrosis of beta-cells occurred. C57BL/KsJ db/db mice (with extensive beta-cell necrosis and early hyperglycemia) developed much higher autoantibody titers to insulin and p73 than did the diabetes-resistant C57BL/6 db/db mice. However, the presence of autoantibodies in normoglycemic C57BL/KsJ +/db controls demonstrated that elevated autoantibody titers alone were insufficient to produce diabetes in this model. Absorption studies indicated that autoantibodies against p73 recognized a common epitope on insulin and IgE-binding factor. The potential significance of this molecular mimicry is discussed.(ABSTRACT TRUNCATED AT 250 WORDS)     

Ciglitazone, a new hypoglycemic agent. I. Studies in ob/ob and db/db mice, diabetic Chinese hamsters, and normal and streptozotocin-diabetic rats

Ciglitazone, 5-[4-(1-methylcyclohexylmethoxy) benzyl]-thiazolidine-2,4-dione, is a new hypoglycemic agent orally active in the obese-hyperglycemic animal models. In C57BL/6J-ob/ob mice, treatment with 100 mg/kg ciglitazone for 2 days elicited a drastic fall in blood glucose. It also decreased plasma insulin, triglycerides, and free fatty acids and food intake without affecting the body weight. Its hypoglycemic activity was independent of its effect on food intake. Regranulation of islet beta-cells and increased pancreatic insulin content were observed in ob/ob mice treated for 41-44 days with 100 mg/kg/day ciglitazone. Ciglitazone showed no effect on food intake, blood glucose, or pancreatic islet beta-cells in a group of lean C57BL/6J-+/? mice concomitantly treated at a dose of 150 mg/kg/day. In C57BL/KsJ-db/db mice, ciglitazone decreased blood glucose and food intake. The untreated db/db mice lost weight despite hyperphagia, whereas the ciglitazone-treated db/db mice gained weight. In the spontaneously diabetic Chinese hamsters, ciglitazone showed no significant effect on food intake, body weight, blood glucose, or insulin content in either plasma or pancreas, but it lowered plasma lipids. In normal rats, ciglitazone failed to affect fasting blood glucose but improved glucose tolerance without increasing insulin secretory response to glucose. In streptozotocin-diabetic rats, it showed no effect on blood glucose or glycemic response to exogenous insulin. The hypoglycemic activity of ciglitazone was specific for obese-hyperglycemic and insulin-resistant animals.     

Alteration of islet cell populations in spontaneously diabetic mice.

Endocrine-cell populations in the islets of Langerhans of mutant mice with a severe hypoinsulinemic diabetes (ob/ob or db/db on the C57BL/KsJ background) or with a mild hyperinsulinemic diabetes (ob/ob or db/db on the C57BL/6J background) were studied quantitatively by immunofluorescence and morphometry. In severely diabetic mice, islets presented a reduced proportion of insulin containing cells but increased glucagon-, somatostatin-, and pancreatic polypeptide (PP)-containing cells, as compared with islets of control (+/+) mice. An inverse change was observed in islets of mildly diabetic mice: islets were hypertrophic and composed mostly of insulin-containing cells, with decreased proportions of glucagon-, somatostatin-, and PP-containing cells. In both types of diabetic syndromes, the changes in cell populations induced a qualitative alteration of cellular interrelationships in the affected islets.     

Reduced sensitivity of inducible nitric oxide synthase-deficient mice to multiple low-dose streptozotocin-induced diabetes

Nitric oxide (NO), synthesized by the inducible isoform of nitric oxide synthase (iNOS), has been proposed as a mediator of immune-induced beta-cell destruction in type 1 diabetes. To evaluate the role of iNOS for beta-cell dysfunction and death, we investigated the sensitivity of beta-cells from mice genetically deficient in this enzyme (iNOS-/-, background C57BL/6x129SvEv, H-2b) both to interleukin (IL)-1beta-induced beta-cell dysfunction in vitro and to multiple low-dose streptozotocin (MLDS)-induced diabetes in vivo. Exposure of islets isolated from C57BL/6 mice to IL-1beta for 24 h in vitro resulted in an induction of iNOS mRNA expression, an increase in nitrite formation, and a decrease in insulin release and proinsulin biosynthesis as compared with untreated C57BL/6 islets. IL-1beta failed to induce iNOS mRNA expression and increase nitrite formation by islets isolated from iNOS knockout mice (iNOS-/-), and no impairment in islet function was observed. The iNOS-/- mice showed a reduced incidence of hyperglycemia after treatment with MLDS as compared with wild-type C57BL/6 (H-2b) and 129 SvEv (H-2b) mice. On day 21 after the first streptozotocin (STZ) injection, 75% of the C57BL/6 mice and 100% of the 129SvEv mice had blood glucose levels >11 mmol/l, whereas the corresponding number for iNOS-/- mice was only 23%. This protection was not due to a delay in the onset of hyperglycemia, since no increase in number of hyperglycemic iNOS-/- mice was observed when the animals were followed up to 42 days. Moreover, islets isolated from iNOS-/- mice were susceptible to the in vitro deleterious effects of STZ. In conclusion, the present study provides evidence that iNOS may contribute to beta-cell damage after exposure to IL-1beta in vitro and treatment with MLDS in vivo.     

Hyperexpression of Foxp3 and IDO during acute rejection of islet allografts

BACKGROUND: We investigated the hypothesis that Foxp3+ cells are an integral component of antiallograft immunity but are dominated by pathogenic effectors. METHODS: Wild-type H-2b C57BL/6 (B6) mice or B6 mice with a targeted disruption of c-Rel gene (c-Rel-/-) were used as recipients of islet grafts from allogeneic DBA/2 (H-2d) mice or syngeneic B6 mice. We developed kinetic quantitative polymerase chain reaction assays and measured intragraft expression of mRNA for Foxp3, IDO, cytolytic molecules, proinflammatory cytokines, and chemokines/receptors. RESULTS: Intraislet levels of mRNA for Foxp3, IDO, CD3, CD25, tumor necrosis factor-alpha, RANTES, IP-10, and CXCR3 were highest in DBA/2 islet allografts from WT B6 recipients compared to DBA/2 islet allografts from c-Rel-/- B6 recipients or syngeneic B6 islet grafts from WT B6 mice. The ratio of granzyme B or IFN-gamma to Foxp3 was higher with the DBA/2 islet allografts from the WT B6 recipients compared to DBA/2 islet allografts from c-Rel-/- B6 recipients or B6 islet grafts from WT B6 recipients. CONCLUSIONS: Foxp3+ cells are an integral component of acute rejection of allografts but may be dominated by pathogenic effectors.     

Possibility of graft-vs-leukemia determinants independent of the major histocompatibility complex in allogeneic marrow transplantation

The role of major histocompatibility (MHC) versus non-MHC determinants in the antileukemic effect exerted by engrafted normal marrow (graft-vs-leukemia, GvL) was studied in Rauscher leukemic SJL/J mice. The marrow donor strains included normal syngeneic SJL/J (H-2s), allogeneic C57BL/10 and 129/J (H-2b), congenic B10.S (H-2s, but otherwise genetically identical to the C57BL/10), and also F1 hybrid mice of the SJL/J and B10.S or C57BL/10 strains. Prior to transplant the recipients were exposed to a dose of total body irradiation that was large, but lower than that required to eliminate all hematopoietic precursors, such that GvL activity of the donor marrow would be necessary to avoid leukemic relapse. Total relapse within 60 days was observed when the syngeneic SJL/J donors were used. Transplantation either of the H-2b C57BL/10 or the H-2s B10.S marrow resulted in approximately 50% unrelapsed survival at 4 months. In contrast, only 26% unrelapsed survival was obtained with H-2b 129/J marrow. Marrow from (SJL/J X B10.S)F1 hybrids yielded a survival curve that was intermediate between those for the two parental strains; a similar but somewhat improved pattern was seen with (SJL/J X C57BL/10)F1-hybrid donors. The results suggest that although MHC genetic differences between the donor and recipient may produce a GvL effect in marrow transplantation therapy, other non-MHC determinants may also be capable of exerting an independent GvL effect of at least equivalent strength.     

Qa-1-associated antigens. IV. Evidence for additional Qa-1 polymorphism defined biochemically and by cytotoxic T lymphocyte recognition

Qa-1-specific, H-2-unrestricted cytotoxic T lymphocytes (CTLs) generated from reciprocally immunized B10.BR (H-2k, Qa-1a) and CBA (H-2k, Qa-1b) mice, and immunoprecipitation of cell surface Qa-1 were used to examined Qa-1 region determinant expression by H-2r and H-2f mice. The H-2b, Qa/Tla congenic mice B6-Tlaa and C57BL/6J (B6) were used as prototype Qa-1a and Qa-1b strains, respectively. Cells from H-2r strains expressed determinants recognized by B10.BR anti-CBA CTLs. Reciprocal cold target inhibition of cytolysis demonstrated that some, but not all, Qa-1b-associated determinants recognized on B6 were expressed by H-2r, Qa-1c cells. In contrast, cells from H-2f strains expressed Qa-1 determinants recognized by both B10.lBR anti-CBA and by CBA anti-B10.BR CTLs. Cold target inhibitions indicated that the H-2f, Qa-1d cells expressed some, but not all, Qa-1a-associated antigens recognized on B6-Tla and did not express the same Qa-1b-associated antigens as B6. Furthermore, H-2f strains expressed cell surface antigens immunoprecipitable by antisera specific for both Qa-1a and Qa-1b-encoded determinants; these molecules were of the same molecular weight as those immunoprecipitated from B6 and B6-Tlaa. These data suggest that CTLs and antisera define the same, closely related Qa-1 determinants encoded by a polymorphic locus.     

In mice, proteinuria and renal inflammatory responses to albumin overload are strain-dependent.

BACKGROUND: The availability of genetically modified mice has increased the need for relevant mouse models of renal disease, but widely used C57BL/6 mice often show resistance to proteinuria. 129/Sv mice are considered more sensitive to certain renal models. Albumin overload, an important model of proteinuric disease, induces marked proteinuria in rats but barely in C57BL/6 mice. We hypothesized that albumin overload would induce more proteinuria in 129S2/Sv than C57BL/6J mice. METHODS: Male and female C57BL/6J and 129S2/Sv mice received bovine serum albumin (BSA) for 11 days. Control groups received saline injections. Injected BSA was immunohistochemically localized to study intrarenal handling of overloaded protein. Renal macrophage infiltration (F4/80 immuno-staining) and glomerular ultrastructure (electron microscopy) were assessed. RESULTS: The BSA-treated groups were similarly hyperproteinemic at Day 11 (D11). Proteinuria differed widely. In C57BL/6J mice, it remained unchanged in females but significantly, though mildly, increased in males (from 3+/-1 to 8+/-2 mg/day, P < 0.05). In 129S2/Sv, proteinuria was marked in both males and females (4+/-1 to 59+/-14, and 0.6+/-0.2 to 29+/-9 mg/day, respectively, both P < 0.01). Proteinuria was accompanied by tubulo-interstitial macrophage infiltration in 129S2/Sv mice. Injected BSA was visualized within glomeruli in both strains and in the urinary space and tubules of 129S2/Sv but not C57BL/6J mice, indicating much greater glomerular leakage in the former. No glomerular macrophages or ultra-structural differences were detected. CONCLUSION: There are major strain differences in the proteinuria and renal inflammatory response of mice to albumin overload, which are not due to structural variation in the filtration barrier but possibly to functional differences in glomerular protein permeability.     

Gene expression profiles of nondiabetic and diabetic obese mice suggest a role of hepatic lipogenic capacity in diabetes susceptibility

Obesity is a strong risk factor for the development of type 2 diabetes. We have previously reported that in adipose tissue of obese (ob/ob) mice, the expression of adipogenic genes is decreased. When made genetically obese, the BTBR mouse strain is diabetes susceptible and the C57BL/6J (B6) strain is diabetes resistant. We used DNA microarrays and RT-PCR to compare the gene expression in BTBR-ob/ob versus B6-ob/ob mice in adipose tissue, liver, skeletal muscle, and pancreatic islets. Our results show: 1) there is an increased expression of genes involved in inflammation in adipose tissue of diabetic mice; 2) lipogenic gene expression was lower in adipose tissue of diabetes-susceptible mice, and it continued to decrease with the development of diabetes, compared with diabetes-resistant obese mice; 3) hepatic expression of lipogenic enzymes was increased and the hepatic triglyceride content was greatly elevated in diabetes-resistant obese mice; 4) hepatic expression of gluconeogenic genes was suppressed at the prediabetic stage but not at the onset of diabetes; and 5) genes normally not expressed in skeletal muscle and pancreatic islets were expressed in these tissues in the diabetic mice. We propose that increased hepatic lipogenic capacity protects the B6-ob/ob mice from the development of type 2 diabetes.     

Transplantation of pancreatic islets into cleared mammary fat pads

Hyperplastic islets of Langerhans from 129/J db3J/db3J mice were used to test the efficacy of the cleared mammary fat pad as an ectopic transplantation site. Eight to 10 islets were inoculated into a cleared mammary fat pad of syngeneic 129/J or allogeneic BALB/cByJ female mice following diabetes induction using streptozotocin. Islets transplanted into syngeneic hosts reversed the weight loss and hyperglycemia characteristic of nontransplanted streptozotocin-diabetic mice. In contrast to the degenerate islets found in the host pancreas, the ectopic islets were filled with well-granulated beta cells and they incorporated 3H-thymidine. Islets transplanted into allogeneic BALB/cByJ diabetic hosts produced weight gain and remission of hyperglycemia in four of six recipients. Extirpation at the 9th week of the cleared fat pad containing the islet graft in one of the cured mice returned the animal to a hyperglycemic state within 1 week. Although some of the grafted islets in allogeneic hosts appeared to be completely intact at 8 weeks, others appeared to be undergoing rejection by the host. Nevertheless, the maintenance of viable and functional islets in the cleared fat pads in H-2-histoincompatible recipients for 8 weeks without immunosuppression of the host or prior culture of the islets makes this model attractive for further study as a potentially immunoprivileged site.     

New mouse model to study islet transplantation in insulin-dependent diabetes mellitus

BACKGROUND: Islet transplantation studies with diabetic rodents frequently use treatment with diabetogens such as alloxan or streptozotocin to render hosts hyperglycemic. These chemicals produce unwanted toxic side effects, which complicate interpretations of damage produced by hyperglycemia versus direct toxin-induced damage. A mouse that spontaneously developed insulin-sensitive diabetes without beta-cell autoimmunity would provide an excellent vehicle for testing beta-cell replacement protocols. The Ins2Akita mutation disrupts normal insulin processing and causes a failure in secretion of mature insulins, which results in the early development of hyperglycemia. This report examines the insulin sensitivity of mice that carry Ins2Akita and their responsiveness to engraftment with syngeneic pancreatic islets. METHODS: Ten-week-old C57BL/6J-Ins2Akita/+ males were given 1 unit of insulin to determine insulin sensitivity. Also, 10-week-old, hyperglycemic B6-Ins2Akita/+ received either 400 islets isolated from syngeneic C57BL/6J males (n=7) or from allogeneic BALB/cJ males (n=5) under the renal capsule. These mice were followed for 8 weeks after engraftment or until remission of euglycemia. Nephrectomy of the graft-containing kidney was performed on mice that remained euglycemic. These mice were then followed for 2 weeks for return of hyperglycemia. RESULTS: B6-Ins2Akita/+ mice are insulin responsive. Insulin treatment of hyperglycemic B6-Ins2Akita/+ males significantly lowered blood glucose values within 1 hr. In addition, B6-Ins2Akita/+ recipients of syngeneic islet grafts reversed their diabetic state in less than 72 hr. These islet-engrafted mice remained normoglycemic until removal of the graft-containing kidney. Removal of the graft resulted in a return to hyperglycemia. Mice that received allogeneic grafts efficiently rejected the graft. CONCLUSIONS: Our data support the hypothesis that B6-Ins2Akita/+ mice are insulin sensitive and provide an excellent model for islet transplantation studies. In addition, the reduced beta-cell mass and the absence of beta-cell autoimmunity, coupled to the fact that these mice also reject allografts, suggest that these mice may be useful for a variety of other applications, including testing functionality of human islets prepared for transplantation and perhaps also for exploring beta-cell restorative therapy using pancreatic islet stem cells.     

Increased hemoglobin AIc in diabetic mice.

The minor hemoglobins AIa, AIb, and AIc were studied in mice with either genetic or chemically induced diabetes. Hemoglobin AIc was elevated approximately twofold in all the phenotypically diabetic mice studied (C57BL/KsJ-db/db, C57BL/KsJ-ob/ob, C57BL/6J-db/db, and alloxan- and streptozotocin-treated mice). Elevation of the hemoglobin AIc in C57BL/6J-db/db mice was of short duration, reflecting the transitory diabetes characteristic of these mice. The degree of increase of hemoglobin AIc levels was unrelated to severity of hyperglycemia, duration of diabetes, age of mouse, or body weight. It is not known what factor(s) dictates the steady-state concentration of hemoglobin AIc.