Pancreatic lymph nodes are required for priming of beta cell reactive T cells in NOD mice

Abstract

Nonobese diabetic (NOD) mice develop spontaneous autoimmune diabetes that results from the destruction of insulin secreting beta cells by diabetogenic T cells. The time and location of the encounter of autoantigen(s) by naive autoreactive T cells in normal NOD mice are still elusive. To address these issues, we analyzed diabetes development in mice whose spleen or pancreatic lymph nodes (panLNs) had been removed. Excision of panLNs (panLNx) at 3 wk protected mice against insulin autoantibodies (IAAs), insulitis, and diabetes development almost completely, but had no effect when performed at 10 wk. The protection afforded by panLNx at weaning was not due to modifications of the immune system, the absence of autoreactive T cells, or the increase in the potency of regulatory T cells. That panLNs are dispensable during adult life was confirmed by the capacity of 10-wk-old panLNx irradiated recipients to develop diabetes upon transfer of diabetogenic T cells. In contrast, splenectomy had no effect at any age. Partial excision of mesenteric LN at 3 wk did not prevent accelerated diabetes by cyclophosphamide as panLNx did. Thus, in normal NOD mice, autoreactive T cell initial priming occurs in LNs draining the target organ of the disease from 3 wk of age.

Figure 1.

Incidence of spontaneous diabetes in female NOD mice Sx (A) or panLNx (B) at different ages.

Figure 2.

Histopathological analysis of islets in the pancreas of NOD mice after panLNx at 3 wk of age. (A) Severity of insulitis was determined at various times after surgery. Sham-operated animals served as controls. (B) Insulitis in mice panLNx at 3 wk, injected with CY at 8 and 10 wk and killed at 12 wk of age.

Figure 3.

Proliferative responses of T cells to anti-CD3 and KLH after panLNx at 3 wk. T cells were tested 3 mo after surgery. Anti-CD3 response was performed on antibody coated dishes (top). T cells were primed in vivo 5 d before restimulation in vitro by KLH-pulsed APC (bottom).

Figure 4.

CY injection failed to induce spontaneous diabetes in mice panLNx at 3 wk, whereas it accelerated diabetes in mice MLNx at 3 wk. CY was administered 5 and 7 wk after excision of panLN or MLN. Incidence of the disease was followed for 28 d.

Figure 5.

panLNs are dispensable for the adoptive transfer of diabetes. NOD mice were panLNx before irradiation and injection of splenocytes from diabetic mice (▪, normal recipients; ♦, Sham-operated recipients; ○, panLNx recipients).

Figure 6.

Levels of insulin autoantibodies are greatly reduced in mice LNx at weaning. Serum samples were collected at 8 or 12 wk of age. +/n, number of positive samples/total samples.

Figure 7.

Surface expression of CD44 and CD62L on CD4 T cells from panLN of normal NOD (A) and BDC2.5/NOD mice (B). In BDC2.5/NOD mice a subset with an intermediate phenotype CD44⁺CD62L^int is highly increased compared with NOD mice, showing the transition step between naive CD44^loCD62L⁺ and activated/memory CD44^hiCD62L⁻ cells. Cells were successively gated on side vs. forward scatter, to select alive cells, and on CD4 expression.