A minor subset of Batf3-dependent antigen-presenting cells in islets of Langerhans is essential for the development of autoimmune diabetes

Abstract

Autoimmune diabetes is characterized by inflammatory infiltration; however, the initiating events are poorly understood. We found that the islets of Langerhans in young nonobese diabetic (NOD) mice contained two antigen-presenting cell (APC) populations: a major macrophage and a minor CD103(+) dendritic cell (DC) population. By 4 weeks of age, CD4(+) T cells entered islets coincident with an increase in CD103(+) DCs. In order to examine the role of the CD103(+) DCs in diabetes, we examined Batf3-deficient NOD mice that lacked the CD103(+) DCs in islets and pancreatic lymph nodes. This led to a lack of autoreactive T cells in islets and, importantly, no incidence of diabetes. Additional examination revealed that presentation of major histocompatibility complex (MHC) class I epitopes in the pancreatic lymph nodes was absent with a partial impairment of MHC class II presentation. Altogether, this study reveals that CD103(+) DCs are essential for autoimmune diabetes development.

Figure 1. Absence of diabetes or immunological infiltrates in NOD.Batf3^−/− mice.

(A) Diabetes incidence in female NOD (n=24), NOD.Batf3^−/− (n=24), and NOD.Batf3^+/− (n=8). Hematoxylin and eosin staining of (B) a 6 week female NOD islet, (C) a 6 week female NOD.Batf3^−/− islet, and (D) a 52 week female NOD.Batf3^−/− islet. Scale bars represent 100 μm. Flow cytometric profiles of CD4⁺ and CD8⁺ T cells from NOD and NOD.Batf3^−/− (E) thymocytes and (F) splenocytes, respectively. Percentage of CD103⁺ and CD11b^lo to negative cells of (G) pLN and (H) mesenteric lymph nodes (gated on CD45⁺, CD11c⁺, MHC-II⁺). (I) Percentage of DEC205⁺ (CD205) and CD8α⁺ cells of spleen (gated on CD45⁺, CD11c⁺, MHC-II⁺ cells). All flow cytometry plots are representative of two or more independent experiments. See also Figure S1.

Figure 2. Two sets of myeloid cells are identified in NOD mice islets

(A) Gating strategy for dispersed islets. (B) F4/80⁺ and CD103⁺ staining of myeloid cells from 4 week and 8 week NOD (left) and NOD.Batf3^−/− (right) islets as gated in panel (A). (C) CD11b⁺ and (D) MHC class II⁺ flow cytometry of F4/80⁺ (Red), CD103⁺ (Blue), or unstained (Green) islet myeloid cells as gated in panels (A). Representative flow cytometry plots and cumulative data from two or more independent experiments with each experiment pooling two or more mouse islets per sample.(E-H) Quantitative RT-PCR (qPCR) was performed on sorted populations of islet cells. Islets represent large and granular CD45- cells. F4/80⁺ and CD103⁺ were gated as shown in Figure S2. All qPCR is represented relative to the expression of 18s rRNA (ΔCt). All qPCRs represent data from 2 biological replicates of 10-15 pooled female NOD mouse islets each performed in duplicate (error bars, SD). All mice used for qPCR were 10-12 weeks old. See also Figure S2.

Figure 3. Lymphocyte infiltration is absent in NOD.Batf3^−/− mice islets

Flow cytometric analysis of (A) CD103⁺ myeloid cells gated as in Figure 2A and(B) CD3ε⁺ lymphocytes (gated on CD45⁺ cells) from 12 week NOD, NOD.Batf3^−/−, NOD.B16A, NOD.H4, and NOD.Rag1^−/− islets. (C) Representative graph of data from (A) including time course of NOD from 3 to 12 weeks of age. (D) Representative graph of data from (B) including NOD time course from 3 to 12 weeks of age. Representative flow cytometry plots and cumulative data from three or more independent experiments with each experiment pooling two or more mouse islets per sample (error bars, SD). (E) Flow cytometry of islet lymphocyte and myeloid cells from NOD, NOD.Batf3^−/−, and NOD.Batf3^−/− that received either NOD.Rag1^−/− or NOD.Batf3^−/−× Rag1^−/− bone marrow 23 weeks prior to being harvested. See also Figure S3.

Figure 4. Gene expression analysis reveals a quiescent state in NOD.Batf3^−/− mice islets

(A-B) Microarray analysis of the islets of NOD, NOD.Batf3^−/− and NOD.Rag1^−/− mice. Scatter-plots of the normalized probe intensity of all annotated microarray signals are shown for whole islet preparations. Numbers labeled in red are those that are at least 2-fold different at a 99% confidence interval by moderated t test. Data are plotted at a log₂ scale. (A) 6 week and (B) 8 week comparisons of NOD vs. NOD.Rag1^−/− (left), NOD vs. NOD.Batf3^−/− (middle), and NOD.Batf3^−/− vs. NOD.Rag1^−/− (right). Several statistically significantly different genes are highlighted in green. (C) Gene Ontology (GO) analysis of the differentially expressed genes between NOD and either NOD.Rag1^−/− or NOD.Batf3^−/− at 6 or 8 weeks of age. Corrected hypergeometric p values were calculated using the genes selected in the scatter plots shown in panels A-B. Each analysis represents the mean of 3-6 independent biological replicates.

Figure 5. Real time PCR shows progression of inflammation in NOD but not NOD.Batf3^−/− mice islets

(A) Taqman qPCR quantification of the indicated genes. Bars represent the mean of the ΔCt value of the gene of interest normalized to an Actb control. All values were adjusted by multiplying by a factor of 10⁵ to facilitate visualization. Bars represent the mean ⁺/− S.D. of 6 biological replicates performed and tested in duplicate. P values were calculated using Mann-Whitney U test. (B-D) The indicated populations were sorted from (B) NOD.Rag1^−/− and NOD.Batf3^−/− as shown previously, (C-D) NOD mouse islets as in Figure S4A, and (D) CD11c⁺ MHC-II⁺ cells from mediastinal lymph nodes. Taqman or SYBR Green I qPCR was used to quantify the indicated genes. Bars represent the mean⁺/− S.D. of 2 biological replicates performed and tested in duplicate. See also Figure S4.

Figure 6. TCR transgenic T cells proliferate more effectively in the pancreatic lymph nodes of NOD than NOD.Batf3^−/− mice

CFSE dilution of transferred (A) BDC2.5 NOD CD4⁺ T cells and (B) 8.3 NOD CD8⁺ T cells isolated from the pancreatic lymph nodes and inguinal lymph nodes from recipient 8-12 week NOD or NOD.Batf3^−/− mice. Cells were assayed 3 days post transfer. Dots represent individual mice. P values were calculated using Mann-Whitney U test (*** P < 0.005).

Figure 7. TCR transgenic T cell entrance into NOD or NOD.Batf3^−/− islets and splenocyte transfer into recipients

(A) Percentage entering islets and (B) CFSE dilution of CD45.1⁺ and CD45.2⁺ BDC2.5 NOD CD4⁺ T cells (both at day 3 post transfer). (D) Percentage entering islets and (E) CFSE dilution of CD45.1⁺ and CD45.2⁺ 8F10 NOD CD4⁺ T cells transferred into 6-12 week NOD or NOD.Batf3^−/− mice (both at day 7 post transfer). Dispersed islet cell presentation to (C) BDC2.5 hybridoma and (F) 8F10 hybridoma as measured by CTLL-2 assay for IL-2. Each dot represents an independent T cell assay. Representative flow cytometry plots and cumulative data from two or more independent experiments. P values were calculated using Mann-Whitney U test (** P<0.005). (G) Diabetic NOD splenic T cells were isolated using CD90.2 magnetic beads and then 10⁷ cells were transferred to the indicated irradiated (600 cGy) recipients. (H) 10-12 week old NOD or NOD.Batf3^−/− splenocytes (10⁷) were transferred into NOD.Rag1^−/− recipients. Recipient mice in (G) and (H) were then monitored for diabetes. Diabetes incidence was monitored as indicated in the main methods section.