In the present study, we used high-density oligonucleotide arrays to elucidate global gene expression patterns in non-atherosclerotic, non-calcified, normal-appearing arterial tissue from patients with type 2 diabetes. Previously, we used these data in a study aimed at identifying new molecular markers of arterial disease in type 2 diabetes. Examinations were done at the single gene level with microarrays, and we were able to validate several genes with q-RT-PCR including FBLN1 as significantly upregulated in the vascular wall both at the mRNA and protein level . In the present bioinformatic study, we use our mRNA expression data to explore biological pathways and networks that are dysregulated in the arterial wall in diabetes. Although our data indicate that differences in gene expression between arterial tissue from patients with diabetes and non-diabetic subjects are modest at the single gene level , we demonstrate that clusters of highly interconnected genes are significantly dysregulated at the transcriptional level in arterial tissue in diabetes. Providing credibility to our results, a number of gene products of these pathways has previously been found dysregulated in the vessel wall in diabetic vasculopathy, as discussed below.
The results from GSEA demonstrate significant upregulation of gene sets and pathways related to cytokine-growth factor- and hormonal actions in arterial tissue from diabetic individuals. This include the gene sets caries pulp up, cell adhesion molecules, monocyte dend DN, and cytokine-cytokine receptor interaction with caries pulp up as the most significantly upregulated gene set. Interestingly, there is an increasing evidence supporting an association between periodontitis and diabetic complications [29, 30] suggesting that genes related to periodontitis could play a role for the development of diabetic vasculopathy. Because the tissue in our study is without atherosclerosis and cellular infiltration, it seems that the vascular smooth muscle cells themselves may express an inflammatory phenotype. In agreement, previous observations showed that vascular smooth muscle cells produce a range of peptide factors that have been suggested to play a role in diabetic vasculopathy . In addition, this finding is in line with previous observations that vascular cells in diabetes may display increased inflammatory capacity, at least partly related to effects of toll-like receptors [31, 32].
Over the past decades there has been much debate regarding the relative importance of hyperinsulinism and insulin resistance in vascular cells from diabetic patients [33–36]. Interestingly, the insulin 2F-pathway, representing genes upregulated 2-fold by insulin stimulation of skeletal muscle in healthy subjects , was significantly downregulated in our study, supporting the idea that at least some effects of insulin in vascular tissue are influenced by insulin resistance and not by hyperinsulinism. This is in agreement with several human and experimental observations [37–39].
Surprisingly, a triglyceride pathway was the most significantly upregulated pathway using GO-Elite. Several organs, such as muscle and liver, are known to accumulate triglycerides in type 2 diabetes ; however, no information concerning arterial tissue is currently available. Our data are compatible with the hypothesis that accumulation of triglycerides could also take place in the arterial wall.
Since the apoptotic process has been found to be regulated in vascular smooth muscle cells from arterial tissue from individuals with type 2 diabetes , it is of interest that our data demonstrate an increased expression of the apoptotic pathway using GO-Elite. The apoptotic pathway may be regulated in arterial tissue in diabetes; however, further studies are warranted to assess the precise implication of apoptosis in the pathogenesis of diabetic arteriopathy. Of interest, also a pathway defined as genes downregulated in glomeruli from patients with diabetic nephropathy was downregulated in the arterial wall in diabetes . This observation may be an indication of a common set of dysregulated genes present in both micro- and macro-angiopathy in diabetes.
Although many detailed unique processes are suggested to be implicated in the development of arterial disease in diabetes, an understanding of the disease mechanisms as an integrated whole may be warranted. Network analysis provides a broad insight into biology in the context of known functional interrelationships among proteins. Our network analysis demonstrates a statistically significant cluster of genes that are dysregulated in the arterial wall in diabetes.
Looking for important hubs in the network, it is interesting to note that the insulin receptor (INSR) appears, although it is not itself regulated. It seems that many genes, interacting with the receptor, are indeed regulated in the arterial wall in diabetes, which is in accordance with the idea that dysfunctional effects of insulin may play a role in diabetic arterial disease (35-37). Another relevant hub is SMAD4, which seems to have an important place in the network, interacting both with intracellularly expressed genes like RRAS2 and PARD3, as well as an extracellular matrix gene, fibulin-1 (FBLN1). SMAD4 is an important intracellular signaling molecule in the TGF-beta system which is in agreement with other observations showing that this system is involved in the development of matrix accumulations in diabetic complications [42, 43]. A recent transcriptome analysis of human diabetic kidney disease pointed towards CDC42 signaling as an important dysregulated pathway , which is in accordance with the presence of CDC42 as an important hub in the network we present here. MMP2 is an important hub in the network with connections to other relevant genes including MMP14, DCN and ITGAV. This part of the network is in agreement with other previously reported data showing that MMP2 is dysregulated in the arterial wall in diabetes  and seems to indicate that matrix remodeling may be an important feature of the non-atherosclerotic arterial disease seen in diabetes.