Novel phenotypes of prediabetes?

Abstract

This article describes phenotypes observed in a prediabetic population (i.e. a population with increased risk for type 2 diabetes) from data collected at the University hospital of Tübingen. We discuss the impact of genetic variation on insulin secretion, in particular the effect on compensatory hypersecretion, and the incretin-resistant phenotype of carriers of the gene variant TCF7L2 is described. Imaging studies used to characterise subphenotypes of fat distribution, metabolically healthy obesity and metabolically unhealthy obesity are described. Also discussed are ectopic fat stores in liver and pancreas that determine the phenotype of metabolically healthy and unhealthy fatty liver and the recently recognised phenotype of fatty pancreas. The metabolic impact of perivascular adipose tissue and pancreatic fat is discussed. The role of hepatokines, particularly that of fetuin-A, in the crosstalk between these organs is described. Finally, the role of brain insulin resistance in the development of the different prediabetes phenotypes is discussed.

Keywords: Brain insulin resistance; Insulin; Liver fat; Phenotype; Prediabetes; Review; Secretion; Sensitivity.

Fig. 1

Insulin sensitivity and insulin secretion of participants of the Tübingen Family Study. Insulin sensitivity is estimated from the OGTT (Matsuda–deFronzo index). Insulin secretion is estimated from the OGTT (AUC for C-peptide/glucose). Green circles, NGT; yellow circles, IGT; red circles, type 2 diabetes. AU, arbitrary units

Fig. 2

(a) Associations between the genotypes of rs7903146 polymorphism in TCF7L2 with insulin secretion during a hyperglycaemic clamp in 73 German individuals. White circles, CC; black circles, CT and TT. AIR, acute insulin response. The p values are for comparison between the genotypes for the first and second phases of glucose-induced insulin secretion, first and second phases of GLP-1-induced insulin secretion and acute insulin secretory response to arginine; figure reproduced with permission from [20]. (b) Association between C-peptide levels at 30 min of the OGTT and glucose levels at 30 min during the OGTT by TCF7L2 SNP rs7903146. Regression lines are shown. Dotted line, CC; dashed line, CT; solid line TT genotype of TCF7L2 SNP rs7903146; figure reproduced with permission from [23]. To convert glucose values from from mg/dl to mmol/l, please multiply by 0.0555

Fig. 3

Subphenotypes of obesity. Whole-body MRI measurements are used to quantify fat compartments [–58]. (a) Yellow, subcutaneous adipose tissue; red, visceral adipose tissue. (b) pVAT, perivascular visceral adipose tissue

Fig. 4

(a) Association of insulin resistance with the amount of liver fat. Individuals were divided into seven groups: quartiles of liver fat in individuals without fatty liver (liver fat <5.56%, n=225) and tertiles of liver fat in individuals with fatty liver (liver fat ≥5.56%, n = 112). Each group was then divided by the median insulin sensitivity into an insulin-sensitive (IS, white circles) and an insulin-resistant (IR, black circles) subgroup. Diamonds indicate mean and the 95% CI. Within each of the seven groups, the subgroups did not differ in liver fat. However, insulin sensitivity was lower in each group; figure reproduced with permission from [58]. (b, c) Liver fat content (b) and insulin sensitivity (c) associated with the I148M variant of PNPLA3,; figure reproduced with permission from [60]. (d) Fatty-acid profiles of hepatic triacylglycerol stores are dependent on the I148M variant of PNPLA3. White bars, wild-type individuals; black bars, PNPLA3 ^I148M individuals; figure reproduced with permission from [61]. TAG, triacylglycerol; TAT, total adipose tissue; VAT, visceral adipose tissue

Fig. 5

The fatty pancreas. (a) Histochemistry of pancreas sections. (b, c) Association of pancreatic fat and insulin secretion in people with NGT (b) and IGT (c); figure reproduced with permission from [79]. (d) Association of increased fetuin-A from fatty liver with impaired insulin secretion; figure reproduced from [80]. (e) Cell-to-cell crosstalk between intrapancreatic fat cells, islets and macrophages

Fig. 6

Reduction of liver fat content during a lifestyle intervention is very much related to physical fitness at baseline (p = 0.004) (a) and to genetic factors (b, c). Genetic variation in transcription factor PPARδ: black bar, carriers of the risk allele; white bar, non-carriers of the risk allele (p = 0.001). Genetic variation in the receptor for adiponectin ADIPOR1: black bar, carriers of the risk allele; white bar, non-carriers of the risk allele (p = 0.004). Error bars are SEM. Figures plotted using data from [91] (a), [88] (b) and [87] (c)

Fig. 7

Change in aerobic capacity plotted against change in insulin sensitivity during lifestyle intervention. ISI, Insulin Sensitivity Index calculated from OGTT

Fig. 8

Insulin-sensitive regions in the human brain. (a) Hypothalamus. (b) Prefrontal area. Data from studies with intranasal insulin application. Figure modified with permission from [106]. CBF, cerebral blood flow; VAT, visceral adipose tissue

Fig. 9

(a) Diagrams illustrating the investigation of visceral adipose tissue with magnetic resonance imaging (MRI) and the measurement of cerebral insulin sensitivity using the technique of magnetoencephalography (MEG). (b) Brain insulin sensitivity before lifestyle intervention (as insulin-stimulated theta activity) and its association with the change in visceral adipose tissue (adjusted for baseline) during lifestyle intervention; r = −0.76; p = 0.001; figure reproduced with permission from [112]. VAT, visceral adipose tissue

Fig. 10

Concept of the pathogenesis of type 2 diabetes mellitus. Chronological development of disadvantageous organ crosstalk involved in the progression from NGT to the prediabetic situation and to type 2 diabetes