Abstract
Background and aims: Beta-cell dysfunction is a key feature of glycaemic deterioration in type 2 diabetes. While associations of circulating proteins with type 2 diabetes are well explored, their role in beta cell dysfunction is unclear, especially in longitudinal terms. Here, we investigated significant associations between temporal changes in beta cell glucose sensitivity (GS) and changes in several proteins over the same period, in type 2 diabetes of recent diagnosis.
Materials and methods: In the IMI DIRECT cohort of white North European type 2 diabetes patients of recent diagnosis (N=552), several circulating plasma proteins (n=141) were measured at months 0 and 36 using two Olink targeted panels (Metabolomics and Cardiovascular II). Protein values distributions were normalized through the absPQN procedure and adjusted for plate, centre, seasonality, time of day, and sample age via linear mixed-effect modelling. GS was estimated from mixed-meal tests at months 0, 18, and 36. Trajectories of both GS and proteins values were modelled as linear functions of time, considering the slopes as indices of progression. The associations between GS and proteins, bot as slopes, were investigated using linear regression models. Each model had the slope of a specific protein as independent variable, and was adjusted for protein, GS, HbA1c, age, and BMI, all at baseline, for sex, changes in medications with reported effects on glycaemia, and the principal components in the slopes and baselines of proteins.
Results: Linear regression analysis identified twenty-one associations as statistically significant (FDR-adjusted p<0.05), with standardized protein coefficients between -0.239 and 0.183 (Figure). The most significant and strongest associations were found for Cadherin 2 (CDH2) and EMR2 (ADGRE2). CDH2 is a transmembrane adhesion protein mediating cell-cell adhesion in many tissues, while EMR2 is a class-B GPCR, known to be expressed both in myeloid cells and pancreatic islets. The other detected proteins are generally related to inflammatory and immune system processes, metabolic pathways, and cell-matrix organization. Some of the identified proteins (e.g., CDH2, FBP1, and KITGL) had already been linked to beta cell function or, more generally, to type 2 diabetes.
Conclusion: We identified a set of proteins with changes over 3 years associated to beta cell function variation over the same period in type 2 diabetes of recent diagnosis. Further investigations may demonstrate the utility of these proteins as biomarkers of progressive beta cell dysfunction, or point to a mechanistic role of some of these proteins in its evolution.
Materials and methods: In the IMI DIRECT cohort of white North European type 2 diabetes patients of recent diagnosis (N=552), several circulating plasma proteins (n=141) were measured at months 0 and 36 using two Olink targeted panels (Metabolomics and Cardiovascular II). Protein values distributions were normalized through the absPQN procedure and adjusted for plate, centre, seasonality, time of day, and sample age via linear mixed-effect modelling. GS was estimated from mixed-meal tests at months 0, 18, and 36. Trajectories of both GS and proteins values were modelled as linear functions of time, considering the slopes as indices of progression. The associations between GS and proteins, bot as slopes, were investigated using linear regression models. Each model had the slope of a specific protein as independent variable, and was adjusted for protein, GS, HbA1c, age, and BMI, all at baseline, for sex, changes in medications with reported effects on glycaemia, and the principal components in the slopes and baselines of proteins.
Results: Linear regression analysis identified twenty-one associations as statistically significant (FDR-adjusted p<0.05), with standardized protein coefficients between -0.239 and 0.183 (Figure). The most significant and strongest associations were found for Cadherin 2 (CDH2) and EMR2 (ADGRE2). CDH2 is a transmembrane adhesion protein mediating cell-cell adhesion in many tissues, while EMR2 is a class-B GPCR, known to be expressed both in myeloid cells and pancreatic islets. The other detected proteins are generally related to inflammatory and immune system processes, metabolic pathways, and cell-matrix organization. Some of the identified proteins (e.g., CDH2, FBP1, and KITGL) had already been linked to beta cell function or, more generally, to type 2 diabetes.
Conclusion: We identified a set of proteins with changes over 3 years associated to beta cell function variation over the same period in type 2 diabetes of recent diagnosis. Further investigations may demonstrate the utility of these proteins as biomarkers of progressive beta cell dysfunction, or point to a mechanistic role of some of these proteins in its evolution.
| Original language | English |
|---|---|
| Article number | 259 |
| Pages (from-to) | S137-S138 |
| Number of pages | 2 |
| Journal | Diabetologia |
| Volume | 66 |
| Issue number | SUPPL 1 |
| Publication status | Published - 4 Sept 2023 |
| Event | 59th EASD Annual Meeting of the European Association for the Study of Diabetes - Hamburg, Germany Duration: 2 Oct 2023 → 6 Oct 2023 Conference number: 59 https://www.easd.org/annual-meeting/easd-2023.html |
