Gene expression profiles of Beta-cell enriched tissue obtained by laser capture microdissection from subjects with type 2 diabetes

Abstract

Background: Changes in gene expression in pancreatic beta-cells from type 2 diabetes (T2D) should provide insights into their abnormal insulin secretion and turnover.

Methodology/principal findings: Frozen sections were obtained from cadaver pancreases of 10 control and 10 T2D human subjects. Beta-cell enriched samples were obtained by laser capture microdissection (LCM). RNA was extracted, amplified and subjected to microarray analysis. Further analysis was performed with DNA-Chip Analyzer (dChip) and Gene Set Enrichment Analysis (GSEA) software. There were changes in expression of genes linked to glucotoxicity. Evidence of oxidative stress was provided by upregulation of several metallothionein genes. There were few changes in the major genes associated with cell cycle, apoptosis or endoplasmic reticulum stress. There was differential expression of genes associated with pancreatic regeneration, most notably upregulation of members of the regenerating islet gene (REG) family and metalloproteinase 7 (MMP7). Some of the genes found in GWAS studies to be related to T2D were also found to be differentially expressed. IGF2BP2, TSPAN8, and HNF1B (TCF2) were upregulated while JAZF1 and SLC30A8 were downregulated.

Conclusions/significance: This study made possible by LCM has identified many novel changes in gene expression that enhance understanding of the pathogenesis of T2D.

Figure 1. Hierarchical clustering of differentially expressed genes.

Hierarchical clustering analysis of genes differentially expressed was performed using dChip software. The distances among high-dimensional expression profiles are represented as a dendrogram that arranges the clustered samples in terms of similarity to one another. The analysis showed a clear separation in two groups of samples from diabetic and control subjects. The analysis divides samples into groups with similar patterns in gene expression data (a p value of less than 0.05 was considered).

Figure 2. Principal Component Analysis.

Principal Component Analysis (PCA) reduces the number of variables and sort microarray experiments into groups. The analysis was performed using Rosetta Resolver software, version 7.2.2.0, with all the array data after normalization by dChip software. Three principal components were generated and plotted; each individual point identifies a single expression profile. The first principal component (PC1) that captures the maximum amount of variation between samples determined the clustering of samples into two groups: one group (12 samples, left) was run at the Joslin Diabetes Center Genomic Core facility, the other group (8 samples, right) was run at the Genomic Core facility of the Massachusetts General Hospital. Variations were also observed along the second principal component (PC2) and the third principal component (PC3). The purple and brown dots refer to samples from diabetic subjects run at the Joslin Diabetes Center and Massachusetts General Hospital, respectively; the blue and green dots refer to control subject samples run at the Joslin Diabetes Center and Massachusetts General Hospital.

Figure 3. Real-time PCR analysis of the regenerating islet genes Reg 1 alpha, Reg 1 beta, Reg 3 alpha, and Reg 3 gamma.

The assay was performed on samples from four T2D subjects (D) and four controls (C). Panels A–D show the relative gene expression of each single sample. In panels a–d data are reported as mean ± SE ratios of relative expression values from T2D samples (D) over values from control samples (C). The statistical significance was evaluated by the two-tailed Student's t-test using the dCt values (not performed with Reg 3 alpha and Reg 3 gamma since the expression was not detectable in two control samples each).

Figure 4. Real-time PCR analysis of Sox9 and matrix metalloproteinase 7 (MMP7) genes.

The assay was performed on samples from four T2D subjects (D) and four controls (C). Panels A–B show the relative gene expression of each single sample. In panels a–b data are reported as mean ± SE ratios of relative expression values from T2D samples (D) over values from control samples (C). The statistical significance was evaluated by the two-tailed Student's t-test using the dCt values.