Fig 1.
Characterisation of NODlow and NODhigh colonies.
(A) Kaplan-Meier survival curves showing diabetes-free survival up 30 weeks of age in 85 NODlow (black circles) and 45 NODhigh (white circles) female mice. Incidence curves were compared by the log-rank test. From [28] with permission; Copyright 2013, The American Association of Immunologists, Inc. (B) and (C) Percent of islets exhibiting no insulitis, peri- or intra-insulitis in six NODlow vs. eight NODhigh 6-week-old female mice (B), or in seven NODlow vs. five NODhigh 30-week-old female mice (C). Each bar represents analysis of three pancreatic sections from an individual mouse. (D) Examples of H&E stained pancreatic sections from 30-week-old NODlow females showing, on the left, two islets without insulitis and, on the right, one islet each with peri-insulitis (*) and intra-insulitis (**).
Fig 2.
T1D is induced in NODlow mice by anti-PD-L1 antibody, but not by cyclophosphamide.
(A) Normoglycaemic (pre-diabetic) NODlow (black circles) and NODhigh (white circles) female mice were treated at 16–18 weeks of age with cyclophosphamide, and T1D development was monitored daily up to 25 days post intra-peritoneal injection. (B) Representative FACS plots for CD4 (x-axis) vs. FoxP3 (y-axis) on gated CD4+ T cells from NODlow (left) and NODhigh (right) spleens. (C) Percentages (left) and absolute numbers (right) of FoxP3+ CD4 T cells from pancreatic lymph nodes of NODlow (full circles) and NODhigh (white circles) mice. Statistical analysis was performed using Student’s t test. (D) Normoglycaemic (pre-diabetic) NODlow females were treated with anti- (α-) PD-L1 (clone MIH5, grey circles) or PBS (black circles) at 12–14 weeks of age, and T1D development was monitored daily up to 25 days post intra-peritoneal injection. In (A) and (D), diabetes-free survival was compared between groups by Kaplan-Meier analysis and log rank test.
Fig 3.
Microbiological characterisation of female NODlow and NODhigh mice.
(A) Venn diagram depicting microorganisms consistently detected by routine health screens in sentinel mice from the NODlow and NODhigh colonies (also includes the results of screening for SFB). None of the other viruses, bacteria or parasites that are routinely tested under FELASA guidelines [30, 31] were detected in either colony. (B and C) Quantification by qPCR (normalized to EUB) of H. hepaticus (B) and SFB (C) in the feces of individual NODlow (black circles) and NODhigh (white circles) mice; horizontal bars represent means. Species-specific primers for the 16S RNA gene were used. (D) Metagenomic analysis of bacterial 16S rRNA genes from NODlow and NODhigh females at 5 weeks of age. Different bacterial clades are color-coded, and log2-fold mean differences between the detection frequencies in NODhigh and NODlow mice shown on the y-axis; positive values show over-representation in the NODhigh colony. The dotted horizontal lines represent tenfold differences in either direction; significant colony differences (p < 0.05 after correcting for multiple comparisons) of tenfold or greater are shown as large squares with their names given. (E) Weights of individual age-matched NODlow and NODhigh females (symbols as in B/C); means (horizontal lines) were compared by Student’s t test.
Fig 4.
Increased B-cell activation in mesenteric lymph nodes of NODhigh mice.
(A) Total cell counts in the indicated lymphoid organs of NODlow (closed circles) vs. NODhigh (open circles) mice: MLN = mesenteric lymph nodes, PLN = pancreatic lymph nodes, ILN = inguinal lymph nodes. Analysis by 2-way ANOVA with p values given for significant colony differences. (B) Representative flow cytometry dot plots of CD69high B cells isolated from mesenteric lymph nodes of NODlow (left) and NODhigh (right) females at six weeks of age. (C) Frequencies of CD69high cells in gated B220+ B cells from mesenteric lymph nodes (MLN) or spleens (SP) of NODlow (closed circles) vs. NODhigh (open circles) female mice. Individual mice, means and significant p values (p < 0.05, by 2-way ANOVA) are shown.
Fig 5.
Exposure to a diabetogenic environment after the age of weaning does not modify T1D development in NODlow mice.
(A) Schematic illustration of continuous co-housing of NODlow (grey outline) with NODhigh (black outline) mice from 3 weeks of age onwards. (B) Kaplan-Meier analysis of diabetes-free survival in females up to 30 weeks of age, comparing the original NODlow (black circles, n = 85) and NODhigh (white circles, n = 45) colonies (cf. Fig 1A) with NODlow (full grey triangles) and NODhigh females (empty grey triangles) that were co-housed from three weeks of age. Despite being in a shared environment, the animals retained the disease incidence curves of their colonies of origin. (C) Diabetes-free survival in females from the original colonies (symbols as in B) was compared with that of NODlow mice orally gavaged with fecal matter collected from 12-week-old female pre-diabetic NODhigh mice (closed grey diamonds) at three weeks of age and maintained in isolators. Fecal gavage did not raise T1D incidence over that in the NODlow parental colony (p > 0.05, log rank test), despite successful transmission of H. hepaticus (cf. S3 Fig).
Fig 6.
Increased T1D incidence and acquisition of NODhigh-like B-cell phenotypes in female offspring of co-housed NODlow mice.
(A) Scheme illustrating co-housing of NODlow with NODhigh mice from 3 weeks of age, followed by breeding and continued co-housing of offspring from their age of weaning. (B) Kaplan-Meier analysis, comparing diabetes-free survival in females up to 30 weeks of age between the original NODlow (black circles, n = 85) and NODhigh (white circles, n = 45) colonies (cf. Fig 1A), and between the offspring of co-housed NODlow (closed grey squares) and co-housed NODhigh (open grey squares) mice. Survival curves for the offspring of co-housed animals were compared to the appropriate colony of origin by log rank test. (C) Intestinal microbiota (by metagenomic analysis) at 5 weeks of age in the offspring of co-housed NODlow mice (filled purple circles), compared with those in the original NODlow (filled orange triangles) and NODhigh (open green squares) colonies. Principal Coordinates Analysis was performed, with the top two components displayed. Each group of mice clusters separately, with the offspring of co-housed NODlow animals intermediate between the two original colonies. See also S1 Fig. (D) CD69 expression in B cells from mesenteric lymph nodes (MLN), pancreatic lymph nodes (PLN), and spleens (SP) obtained from 6-week-old NODlow females (grey full squares) and NODhigh females (grey empty squares). Group means were compared by 2-way ANOVA with post-test for colony differences.