Elsevier

Clinical Immunology

Volume 129, Issue 2, November 2008, Pages 195-201
Clinical Immunology

Rapid Communication
Tissue- and age-specific changes in gene expression during disease induction and progression in NOD mice

https://doi.org/10.1016/j.clim.2008.07.028Get rights and content

Abstract

Whole genome oligo-microarrays were used to characterize age-dependent and tissue-specific changes in gene expression in pancreatic lymph nodes, spleen, and peripheral blood cells, obtained from up to 8 individual NOD mice at 6 different time points (1.5 to 20 weeks of age), compared to NOD.B10 tissue controls. “Milestone Genes” are genes whose expression was significantly changed (∼ 3 fold) as the result of splicing or changes in transcript level. Milestone Genes were identified among genes within type one diabetes (T1D) susceptibility regions (Idd). Milestone Genes showing uniform patterns of changes in expression at various time points were identified, but the patterns of distribution and kinetics of expression were unique to each tissue. Potential T1D candidate genes were identified among Milestone Genes within Idd regions and/or hierarchical clusters. These studies identified tissue- and age-specific changes in gene expression that may play an important role in the inductive or destructive events of T1D.

Introduction

A strong association with the major histocompatibility complex (MHC) has been reported for all autoimmune diseases studied to date [1]. For many years a leading hypothesis for this association between the MHC and autoimmune disease has been that disease associated MHC alleles bind autoantigens and present them to autoreactive T lymphocytes [2], [3], [4]. However, a major problem for any proposed relationship between the MHC and autoimmune disease is the concordance rate for autoimmune diseases seen among monozygotic siblings, frequently below 50% [5], [6], [7], [8].

Studies reported below examine the hypothesis that genes encoded as antigen-presenting molecules in the MHC allow “autoantigen” recognition, and subsequent progression of autoimmunity mainly through stochastic events driven by inflammation. Results presented in this manuscript examine the following scenario: exposure to a common environmental antigen(s) can be recognized by genetically at risk individuals expressing disease associated MHC molecules in either a disease-provoking manner (recognition and presentation of a disease-provoking antigenic epitope), or conversely, by recognition and presentation of a non-disease-provoking epitope. This hypothesis would allow an autoimmune disease concordance rate of 50% for monozygotic twins encountering a single environmental antigen that expresses two equally dominant epitopes, one of which is disease-provoking. The larger the number of non-disease-provoking antigenic epitopes recognized on the inductive antigen(s), the less disease concordance, the more dominant the disease-provoking epitope, the greater the rate of disease concordance among monozygotic twins.

This hypothesis can be examined in an animal model of spontaneous autoimmune diabetes, through the use of MHC congenic mice, where imposition of the “non-permissive” H-2b haplotype onto NOD mice (NOD.B10) efficiently silences the NOD autoimmune disease [9]. In this hypothesis, NOD.B10 mice serve as surrogate to the non-disease expressing twin because they cannot present the disease-provoking epitope of the causative antigen in a disease inductive fashion. By only allowing the disease related I-Ag7 MHC bearing NOD mice to recognize and respond to the putative disease-provoking antigenic epitope, changes in gene expression seen in multiple tissues at multiple time points in these otherwise isogenic mice should allow identification of the genes whose expression (mRNA isoform or transcript level) is changed as a part of disease processes.

Here we report the initial results of a gene expression analysis examining three tissues [pancreatic lymph nodes (PLN), spleen (SPL) and peripheral blood cells (PBC)] at 6 different ages: 10 days and 4 weeks (disease initiation; peri-insulitis), and 8, 12, 16 and 20 weeks (disease onset; destructive insulitis) comparing tissues from NOD to NOD.B10 mice, using 41K Agilent mouse whole genome oligo arrays (Fig. 1A and Table 1). These data were used to develop a “road map” of the “expressed genotype” of NOD mice in an attempt to identify tissue-specific and age-dependent changes in gene expression that might identify type 1 diabetes (T1D) candidate genes (or gene subsets) for disease induction or progression within previously identified high-resolution congenic Idd regions [10], [11], [12], [13] and/or in the hierarchical clusters of Milestone Genes.

Section snippets

Research design and methods

(The details are given in the Supplementary Material)

Mice

NOD/LtJ (NOD), NOD.B10Sn-H2b/J (NOD.B10), NOD.B10 Idd9.2 (NOD Idd9.2) female mice of multiple ages were used for this study.

Milestone Genes (genes whose expression changes significantly) are tissue-specific and age-specific

The overall results of 41K whole genome expression profile for PLN, SPL and PBC are summarized in Supplementary Material Fig. S1. Genes whose apparent expression was significantly changed were selected as Milestone Genes (≥ 3 fold and false discovery rate: Q-value< 0.01; see Research design and methods), expressed in Fig. 1B and Supplementary Material Table S1 as the number of genes whose expression was significantly changed. When the serum glucose level was used as an independent variable, and

Discussion

Here we report for the first time, a temporal analysis of gene expression in three separate tissues from NOD mice, PLN, SPL and PBCs, compared to gene expression in tissue matched samples from congenic NOD.B10 mice, beginning from a newborn setting (10 days old) to 20 weeks of age. The purpose of this analysis was to look for tissue-specific and time-dependent changes in gene expression during disease progression in NOD mice in order to attempt to identify tissues in which one could begin to

Acknowledgments

We wish to thank Claire Holness and Demi Dang for invaluable technical assistance and Linda S Wicker for providing valuable information for Idd regions and Pearl Chang for FACS analysis. This work was funded by NIH grant U19- DK 61934 and U19- AI050864.

References (16)

There are more references available in the full text version of this article.

Cited by (48)

  • Deciphering early development of complex diseases by progressive module network

    2014, Methods
    Citation Excerpt :

    Of course, these T1DM pathway genes actually show the rewired interactions in different tissues and time points (Fig. 4). Especially, this module has more appearances (i.e. the appearance frequency is 0.8) after the 3rd time point (8 week) when T1DM symptom is diagnosed [21], so that such a disease-responsive module possibly represents a later event of development to the advanced T1DM and is the result of the pre-disease module occurrence. In addition, for all tissues-common modules (the module has been observed in each tissue at least one time), the appearance frequencies after the 3rd time point (8 week) have been calculated, in which the average frequency is about 0.65 (seeing Fig. S7).

  • Molecular pathwayalterations in CD4 T-cells of nonobese diabetic (NOD) mice in the preinsulitis phaseof autoimmune diabetes

    2014, Results in Immunology
    Citation Excerpt :

    The majority of the CD4 T-cell NOD altered genes were repressed, similar to the results of the spleen leukocyte study [25] and supported by many other studies in both mice and humans [36–42]. Kodama et al. [36] reported global repression of genes in various tissues of NOD mice (including spleen cells) in comparison to NOD.B10 controls, a strain in which the NOD MHC haplotype is replaced with that from the nondiabetic B10 mice. Liston et al. [37] also found a global dampening in gene expression in NOD thymocytes of the genes involved in T cell negative selection.

View all citing articles on Scopus
View full text