NEW YORK (GenomeWeb) – Researchers at Lund University in Sweden reported today that they have identified patterns of age-related epigenetic changes associated with increased risk of type 2 diabetes.
Published in Nature Communications, the study identified areas of differential methylation in pancreatic tissue, showed it was linked to altered gene expression and insulin secretion in pancreatic beta cells in vitro, and is at least partially present in blood, suggesting the possibility of developing a non-invasive test to predict the development of diabetes as people age.
Predicting and detecting incipient type 2 diabetes (T2D) is already possible using tests that measure blood glucose over time. However, the Lund authors argued that the predictive ability of these current methods is relatively modest.
Their new study was based on a simple observation that as people age, the incidence of impairment of the pancreatic islets increases, as does the development type 2 diabetes.
According to the authors, as they age, humans become more and more likely to develop glucose tolerance. With the inception of this tolerance the body initially works to sustain normal blood sugar levels by adapting to increase the amount of insulin secreted from the pancreatic beta cells. However, in some people, aging β-cells eventually fail to meet the increasing demands for insulin production, which manifests in hyperglycemia and T2D.
"Why aging beta cells fail is unclear … [but] one possible factor may be that aging causes epigenetic changes that affect gene expression and thereby potentially insulin secretion in pancreatic islets," the study authors wrote.
To investigate this, the Lund team set out looking for age-related epigenetic changes that appear to affect human islet function, and to see whether the same epigenetic biomarkers detected in circulation might also predict the development of T2D.
First, the researchers analyzed genome-wide DNA methylation using the Illumina Infinium HumanMethylation450 BeadChip in samples from the pancreatic islets of 87 non-diabetic donors, aged 26 to 74 years. They initially found that higher age was associated with increased DNA methylation of 241 total sites distributed in 154 genes and intergenic regions in this tissue.
A literature search then showed that 30 (19 percent) of the methylated genes were in some way associated with diabetes, β-cell function, or mitochondrial function, including the genes CCND2, CILP2, PBX4, SH2B3, SLC6A4, TCF7 and KLF14, which are each known to have polymorphisms associated with diabetes risk, the authors wrote.
The researchers then tested the effect of methylation of four of these genes — FHL2, and ZNF518B, GNPNAT1 and HLTF — on insulin secretion in pancreatic β-cells in vitro, silencing their expression inclonal rat β-cells. This appeared to reduce the expression of corresponding mRNA by 65 percent to 75 percent and also looked to have functional effects on clonal β-cells, the authors wrote.
Then the team turned to the blood, looking to see if the age-related methylation changes they identified in pancreatic islets might also be reflected in blood cells, which would offer a non-invasive measurement option.
Encouragingly, at least some of the changes the group saw in pancreatic islets — almost 60 percent — were also present in blood. In addition, the authors wrote, some of these blood-based epigenetic markers correlated with current and future insulin levels measured in vivo and were associated with lower risk of future T2D.
This final analysis was based on the examination of samples from several other studies. Initially, the researchers compared their own results with data from a previously published study examining the impact of age on DNA methylation in leucocytes in 421 individuals aged 14 to 94 years.
According to the authors, more than half of the 241 methylation sites that were significantly associated with age in their own study of pancreatic islets were similarly associated in leucocytes in the this separate cohort. The sites covered 83 unique genes, including KLF14, FHL2, FAM123C and ZNF518B.
Blood samples from this first cohort were not available, so the researchers used pyrosequencing to analyze DNA methylation of those same four genes — KLF14, FHL2, FAM123C and ZNF518B — in blood samples taken from another cohort of 112 individuals from a study called the Danish Family Study.
According to the authors, it was apparent that with increased age, there was also increased methylation of KLF14, FHL2, FAM123C and ZNF518B at both baseline and at a ten-year follow up. In addition, the degree of DNA methylation increased for all four genes when they compared the two time points of blood samples.
Looking at data from a second study led by researchers from Leiden University Medical Centre, the Lund team was then able to compare methylation in matched pancreatic tissue and blood from the same subjects, again focusing on KLF14, FHL2, FAM123C and ZNF518B.
According to the authors, the methylation data for these two tissues correlated significantly in the studied individuals.
Taken together, data from the two studies "support that age-related methylation changes found in blood in many cases are similar to methylation changes in pancreatic islets and potentially can be used as biomarkers for alterations in primary tissues for diabetes," they wrote.
Finally, to see if epigenetic markers also associated with risk of future diabetes, the investigators used pyrosequencing to analyze methylation in blood from 299 subjects in a third study, Finland's Botnia study, which followed healthy participants prospectively to detect progression to T2D.
According to the Lund authors, higher methylation was indeed associated with smaller hazard ratios, and a lower risk of future T2D in the Botnia cohort.
Interestingly, the authors wrote, while previous studies have found differential methylation patterns that distinguish subjects with T2D from controls, these associations have largely linked the disease with low methylation levels, in contrast to the present study, which found increased methylation associated with aging, and in turn, the development of diabetes.
Moreover, there was also only a minor overlap between the genes identified in the new study and what has been seen in previous islet case-control cohorts, the authors wrote, suggesting that different biological pathways may underlie age-related or risk-related from actual disease-related methylation changes.
Because of the relatively small numbers of subjects, and limitations in the group's ability to analyze matched pancreatic tissue and blood samples, the results will require further follow up to confirm.
The Lund researchers reported that they hope to move forward from the current study by performing epigenetic whole-genome sequencing in a larger population group in order to try to develop a stronger predictive signature.