Danish Medical Bulletin    |    Contact Front page  

 
Genetic aspects of type 2 diabetes and related traits ¿ The past, present and future

Niels Grarup  


  Download file   (2901 KB)

Accepted by: Faculty of Health Sciences Aarhus University
Defended on: May 27, 2009
Official opponents: Sten Lund , Marju Orho-Melander , Birger Thorsteinsson
Tutors: Ole Schmitz , Oluf Borbye Pedersen , Torben Hansen

Published in the PhD Database: June 2, 2009


English abstract
The PhD thesis originates from Aarhus University, Steno Diabetes Center and Hagedorn Research Institute and is based on four published original papers.
Type 2 diabetes (T2D) is a major global health problem and occurs with increasing prevalence. T2D is primarily characterized by insulin resistance, a relative deficient insulin secretion and in most cases obesity with associated subclinical inflammation. This common metabolic disorder is influenced by both lifestyle and genetic factors. This PhD thesis focuses on the genetic background of T2D and related metabolic phenotypes.
Until spring 2007 the progress in finding genes responsible for the genetic predisposition to T2D and related traits was sparse. Studies were primarily done by investigating biological candidate genes thought to be involved in disease pathogenesis. Only a few candidate genes associated with T2D have since been widely replicated in large-scale studies.
We investigated the common amino-acid changing K121Q variant in ENPP1 in relation to T2D and insulin resistance (Study I). ENPP1 encodes a protein involved in insulin signalling in the target tissues. In a case-control study of ~6,000 individuals we found no association with T2D. However, when combining all published data in the literature in a meta-analysis we demonstrated a modest impact on risk of T2D. A subsequent, updated meta-analysis confirmed an 8% increase in risk of T2D.
PPARD encodes a multi-potent regulator of glucose and lipid metabolism and insulin sensitivity expressed both in the liver and in skeletal muscles. We investigated 12 variants in PPARD and found no robust impact on T2D, serum lipid traits or insulin sensitivity (Study II). In the thesis a meta-analysis is included combining own and published data indicating an 8% increase in risk of T2D per risk-allele of rs2016520.
As well, variation in SREBF1 may significantly impact risk of T2D as demonstrated in a meta-analysis of Danish and published data presented in study III. Also, this impact is modest. Furthermore, in the population-based Inter99 cohort of ~6,000 individuals and the ADDITION Denmark screening sample of ~9,000 participants we found associations of SREBF1 variation with quantitative levels of glycemia (Study III).
Since the development of genome-wide association studies many gene variants have been convincingly shown to influence risk of T2D and related traits. We investigated one of the lipid-associated variants in hepatic lipase, LIPC, in two large cohorts and found a highly statistically significant 0.04-0.06 mmol/l increase in high-density lipoprotein (HDL) cholesterol levels in studies of ~16,000 Danes. In addition, we showed that this effect may be modulated by physical activity showing a higher increase in HDL-cholesterol in vigorously physically active individuals (Study IV).
Despite the high number of validated disease-associated common variants, the explained proportion of the genetic contribution and variation in these traits is modest. In the Inter99 cohort the validated variants explain from 1 to 3.5% of variation in metabolic phenotypes. Similarly, validated T2D variants explain below 10% of the genetic contribution to risk of disease which is underlined by the poor ability of genetic variants to predict T2D. Many more common, low-impact variants probably exist and variants with a frequency below 5% have not been thoroughly investigated. Simulation analyses presented in the PhD thesis show that discovery of a high number of both common, low-impact variants and variants with lower frequencies and higher impacts is necessary to develop a useful algorithm for prediction of T2D based on genetic information.
In summary, the studies performed in the PhD project contribute to the ongoing exploration of the molecular genetics of T2D and related phenotypes and may provide further basis for the genotype-based improvement of diagnosis, and prevention and treatment of this common disease.



Danish abstract
Ph.d.-afhandlingen udgår fra Aarhus Universitet, Steno Diabetes Center og Hagedorn Research Institute og bygger på 4 publicerede originalartikler.
Type 2 diabetes (T2D) er et alvorligt sundhedsproblem i kraftig vækst. T2D er karakteriseret ved nedsat insulinfølsomhed, relativ nedsat insulinsekretion samt som regel fedme som er associeret med subklinisk inflammation. Denne hyppige metaboliske sygdom påvirkes af livsstil, miljø og genetiske faktorer. Ph.d.-afhandlingen fokuserer på den genetiske baggrund for T2D og relaterede metaboliske fænotyper.
Indtil foråret 2007 var fremgangen i identifikationen af den genetiske sårbarhed for T2D og relaterede fænotyper beskeden. Genetiske studier blev primært foretaget ved at undersøge kandidatgener involveret i sygdomspatogenesen. Kun enkelte kandidatgener associeret med T2D er siden blevet repliceret i større studier.
Vi undersøgte en hyppig aminosyre-ændrende K121Q variant i ENPP1 i relation til T2D og insulinfølsomhed (Studie I). ENPP1 koder for et protein involveret i insulin signalering i målvævene. I et case-control studie med ca. 6000 deltagere fandt vi ingen association med T2D. I en kombineret meta-analyse af egne og alle publicerede data viste vi dog en moderat effekt på risikoen for T2D. En efterfølgende opdateret meta-analyse har siden konfirmeret en 8 % øget risiko for T2D pr. risiko allel.
PPARD koder for et protein der regulerer glukose- og lipidmetabolismen samt insulinfølsomhed. Vi undersøgte 12 varianter i PPARD og fandt ingen robust association med T2D, lipidniveauer i serum eller insulinfølsomhed (Studie II). I afhandlingen præsenteres en meta-analyse af egne samt publicerede data i litteraturen. Denne meta-analyse antyder en 8 % øget risko for T2D pr. risiko-allel.
Desuden er det muligt at variation i SREBF1 påvirker risikoen for T2D. Dette blev demonstreret i en meta-analyse af danske og publicerede data præsenteret i studie III. Også denne risikoøgning er moderat. I ~6000 deltagere i den populationsbaserede Inter99 kohorte og i ~9000 personer i den danske ADDITION screeningskohorten fandtes associationer af variation i SREBF1 med ændrede kvantitative mål for glykæmi (Studie III).
Siden udviklingen af helgenoms associationsstudier er mange flere varianter overbevisende vist at have en effekt på risikoen for T2D og relaterede træk. I en undersøgelse en af de lipid-associerede genvarianter i hepatisk lipase, LIPC, i to større kohorter fandt vi en højsignifikant 0.04-0.06 mmol/l øgning i high-density lipoprotein (HDL) kolesterol i studier af ~16000 danskere (Studie IV). Vi viste desuden at denne effekt muligvis bliver moduleret af niveauet af fysisk aktivitet idet meget fysisk aktive individer havde en større øgning i HDL-kolesterol.
På trods af det store antal validerede hyppige genvarianter forklarer disse varianter kun en lille del af variationen i disse træk. I Inter99 forklarer de validerede varianter mellem 1 og 3,5 % af variationen i metboliske fænotypetræk. Ligeledes forklarer de validerede T2D varianter under 10 % af den genetiske komponent til T2D hvilket også understreges af et ringe potentiale for at forudsige T2D. Der findes mange årsager til den store uforklarede andel af genetikken bag komplekse sygdomme som T2D. Der formentlig mange flere hyppige varianter med lav effekt og varianter med en frekvens under 5 % er endnu ikke blevet undersøgt systematisk. Simulationsanalyser præsenteret i ph.d.-afhandlingen viser at opdagelse af et stort antal både hyppige varianter med lav effekt og sjældne varianter med højere effekt er nødvendig for kunne udvikle en brugbar algoritme til forudsigelse af T2D baseret på genetisk information.
Sammenfattende bidrager studierne foretaget i ph.d.-projektet til en større forståelse for den genetiske komponent der ligger til baggrund for T2D og relaterede træk, og de kan på sigt være med til at øge muligheden for genotype-baseret forbedring af diagnosen samt forebyggelse og behandling af denne udb