We have been able to collect pancreatic and blood samples from several hundred living subjects as well as brain-dead organ donors with prediabetes (impaired glucose tolerance) or type 2 diabetes. Human pancreatic samples from non-diabetic subjects have also been collected and exposed in vitro to experimental conditions mimicking the diabetic milieu driving pancreatic beta cell failure in vivo. We have processed those samples and carried out some of the most extensive and in-depth analyses of gene and protein expression. In brief, we have generated more than 20,000 multi-omics data (learn more here), including quantitative peptidomics, lipidomics and metabolomics, as well as genomics, epigenetics and gene expression data.

By integrating these omics data with the clinical traits of the human donors, we have been able to identify molecular signatures associated with the progression of hyperglycemia and deterioration of pancreatic beta cell function.

More specifically, we have identified for the first time (i) a set of genes which are already dysregulated in prediabetic subjects and (ii) some lipids circulating in the blood stream which are associated with gene dysregulation and pancreatic beta cell failure in type 2 diabetes. Learn more here.

Our studies have resulted in the identification of novel candidate biomarkers specifically related to pancreatic beta cell function and viability in type 2 diabetes. These biomarkers could be valuable tools to better identify subjects with the highest risk of developing diabetes, or to develop more effective diabetes treatments. Learn more here.

In type 2 diabetes, muscle, fat, liver and pancreatic beta cells communicate with each other via circulating molecules (including lipids and peptides) to adapt to metabolic dysregulation.

Mouse preclinical models have been used to further investigate such inter-organ communication. More specifically, we aimed to identify genes deregulated in pancreatic beta cells and insulin target tissues, which could link blood biomarkers of type 2 diabetes susceptibility with alterations of insulin secretion and/or insulin action.

Three strains of mice have been used and fed with two different diets: regular chow or “high-fat high-sucrose” diet, in order to mimic both human genetic heterogeneity and imbalanced dietary intake frequently observed in the human population. We have performed multi-omics analysis (lipidomics and transcriptomics, learn more here) in liver, fat, muscle, pancreas and blood samples.

After data integration and analysis, we identified a set of genes and a number of lipids associated with the development of prediabetes: read more about the gene Elovl2 and the class of lipids “dihydroceramide” here. Importantly, our findings in mice models have been translated to humans. Mouse preclinical studies are therefore complementary to clinical studies in order to develop better treatment options for type 2 diabetes. Learn more here.

For the first time, we have identified a set of genes which are already dysregulated in prediabetic subjects. We also found that progressive dysregulation of islet gene expression concomitantly with continuous elevation of blood glucose levels is a broadly divergent disharmonic process which does not resemble a linear trajectory of mature beta cells toward a precursor or trans-differentiation stage. Furthermore, by integrating information on the clinical state of the donors with data on gene expression in the islets and lipids in their blood, we have been able to define the relative importance of molecular variables positively or negatively associated with the continuous elevation of blood glucose levels beyond thresholds for clinical classification of donors and with a direct or indirect causal relationship with pancreatic beta cell failure in type 2 diabetes.

The impact of metabolic stress on gene expression in pancreatic islet cells has also been assessed by exposing isolated human islets in vitro to conditions mimicking the diabetogenic milieu in vivo. Interestingly, it was shown that upon removal of the metabolic stress islet beta cells can recover from the damage induced by their prolonged exposure to highly saturated fatty acids or high glucose alone, while their dysfunction was irreversible upon prolonged exposure to both fatty acids and high glucose.

Studies in human islets from organ donors further suggested that hypermethylation of the gene PNLIPRP1 might have a role in the development of pancreatic cancer – a severe condition frequently associated with hyperglycemia and type 2 diabetes.

The C-peptide is a portion of the proinsulin molecule whose plasma levels can be used to reliably measure insulin secretion. We have developed a novel method to quantify the C-peptide, which may enable better monitoring of the progression to diabetes and/or testing of drugs that can prevent it.

By integrating multi-omics data from mice tissue samples, we obtained a map of gene modules in islets and insulin target tissues, and their interaction in the development of prediabetes (correlation with plasma lipids and phenotypic traits). Based on that, we identified a set of genes and lipids able to explain a phenotypic trait. For instance, dihydroceramides are now considered as potential biomarkers of type 2 diabetes susceptibility both in mice and humans. Also, the role of the Elovl2 gene, which is required for the production of the fatty acid DHA, is further clarified in relation to glucose-stimulated insulin secretion and susceptibility to glucolipotoxicity-induced apoptosis, in both mouse and human pancreatic beta cells.