Association of TNF-α with insulin resistance in type 2 diabetes mellitus

Tumour necrosis factor alpha (TNF-α) is an adipocytokine involved in systemic inflammation and stimulates the acute phase reaction1. TNF-α is primarily secreted by macrophages, and also by a broad variety of other cells including adipocytes23. TNF-α inhibits insulin transduction, and has an effect on glucose metabolism45. Disturbances in the TNF-α metabolism have been implicated in metabolic disorders, such as obesity and insulin resistance6, indicating that perturbations of TNF-α metabolism may affect the onset of type 2 diabetes mellitus and the progression of the disease. However, the role of TNF-α in the duration and progression of the disease is not clear. The aim of this study was to investigate the levels of TNF-α in type 2 diabetes mellitus (T2DM) and to analyze its association with beta cell function, insulin resistance and body mass index (BMI) and its role with respect to duration of diabetes.

Material & Methods

A total of 50 (male=28, female=22) diabetic subjects with mean age 51.2 ΁ 11.7 yr were selected from the outpatient Department of General Medicine, Narayana Medical College and Hospital, Nellore, Andhra Pradesh (Table I). The study was conducted over a four month period (April - July 2010). The study protocol was approved by the Institutional Ethics Committee. Informed written consent was obtained from all subjects. A detailed medical history was taken and physical examination was done on all subjects.

Table I Characteristics of study subjects and groups

Inclusion criteria: (i) Diabetic subjects of either sex between 35-70 yr. (ii) The subjects who used only sulphonylureas in the past and currently not on any oral hypoglycaemic agents for a minimum period of 6 months were only chosen. This is to avoid any effect of the drugs on insulin secretion by pancreatic beta cells.

Studies have shown that sulphonylureas only modestly affect inflammatory markers in patients with type 2 diabetes mellitus7–10. So far only metformin and the thiazolidinediones have been shown to be directly anti-inflammatory11.

Exclusion criteria: (i) Patients on insulin or any other drugs that would affect glucose homeostasis. (ii) Clinically significant hepatic, neurological, endocrinological, or other major systemic disease, including malignancy. (iii) Patients with complications of diabetes like diabetic neuropathy, nephropathy, retinopathy, etc., based on clinical and laboratory investigations. (iv) Patients with the habit of smoking and alcohol were also excluded from the study.

Anthropometric measurements: Height, waist and hip circumference were noted using a measuring tape (to the nearest 0.1 cm), with the subjects wearing light clothes and no shoes. Waist circumference was measured at the midpoint between lower border of the rib cage and the iliac crest . Hip circumference was measured at the level of trochanter. Waist hip ratio (WHR) was calculated as waist circumference divided by hip circumference. Weight was measured to the nearest 0.1 kg using a mechanical weighing machine. BMI, defined as mass in kilograms divided by the square of height in meters, was calculated.

Biochemical assays: Blood samples were obtained by venipuncture after overnight fasting. The serum was separated and stored at -20°C. Glucose was measured using glucose oxidase method by automated chemical analyzer, Humaster-300 (GmbH, Germany). Serum insulin was determined utilizing chemiluminescence immunoassay, method (Human GmbH, Germany). Serum TNF-α was quantified using sandwich ELISA kit from eBioscience, Bender Med Systems, Austria, which has an inter-assay coefficient of variation of 7.5-10.4 per cent and a lower limit of detection of 0.5 pg/ml.

Homeostatic Model Assessment (HOMA) method, which has been validated as a reliable measure of insulin sensitivity in vivo in humans12, was used to estimate insulin resistance (HOMA IR) and % beta cell functioning (HOMA β). The subjects were arbitrarily categorized into three groups based on duration of diabetes. Group 1 included subjects with diabetes of less than 5 yr duration, group 2 included diabetics of 6-10 yr duration and group 3 greater than 10 yr duration.

Statistical analysis: Statistical analysis was performed using SPSS software (Version 12.0). Distributions of continuous variables were tested for normality, and, if appropriate, the natural log transformations of skewed variables were used in subsequent analyses. All analysis was two tailed and P<0.05 was considered as statistically significant. Spearman's correlation was used to estimate the association between the variables.

Results & Discussion

We found a strong correlation between BMI and TNF-α (P =0.006), (Table II) which is in agreement with earlier studies1314. Bertin et al15 detected a correlation between TNF-α and BMI with indices of intra-abdominal fat tissue, but not with glycaemia or total amount of fatty mass in the body. Mishima et al14 found that serum TNF-α concentration in obese persons with type 2 diabetes mellitus depends on the degree of their insulin resistance but does not depend on BMI.

Table II Spearman's correlation analysis of data

The lack of correlation with WHR suggests that TNF-α expression is not tightly linked to differences in distribution of body fat. However, previous studies have indicated the importance of regional fat deposition as a determinant of increased risk for insulin resistance and T2DM16. Further clinical studies in larger groups will be necessary to address this issue. Skewed data of variables was log transformed.

There was a significant correlation between TNF-α and insulin resistance in group 2 whereas TNF-α was significantly correlated with BMI in group 1. A significant negative correlation of HOMA b with fasting blood sugar (FBS) and a positive correlation with insulin levels in groups 1 and 2 was observed (Table II).

The loss of correlation between HOMA β with fasting sugar and insulin in group 3 when compared to group 1 and 2 indicates the progressive loss of beta cell function as the disease progresses. Thus, a better percentage of beta cell functioning implies a relative early stage of the disease.

The significant correlation of TNF-α with HOMA β and insulin may indicate the compensatory overfunctioning of beta cells to nullify the insulin resistance produced by TNF-α in the peripheral tissues. Therefore, intervention of TNF-α, before the progressive loss of beta cell function, may yield promising results in the treatment of diabetes.

Miyazaki et al17, have concluded that TNF-α increased before the onset of diabetes and further increase was not associated with insulin resistance. But Bluher et al18 reported no role of TNF-α in the genesis of early stages of insulin resistance. They attributed the genesis of insulin resistance to non-esterified fatty acids in particular18. Demirbas et al19 showed that in patients with hypertension serum TNF-α concentration increased together with increase in concentrations of insulin, and HOMA IR. No correlations were found between insulin resistance and TNF-α.

Our limited data suggest that TNF-α may play a potentially important pathophysiological role in the development of insulin resistance, particularly in males and in people with high BMI. Longitudinal studies with larger sample size need to be carried out to address this issue.

Similarly, it will be critical to assess in the future the effect of neutralizing TNF-α activity on the reversibility of the inflammatory responses and on the progression of diabetes related complications.

In conclusion, the current observations support the hypotheses that TNF-α may be involved in the aetiology of insulin resistance in type 2 diabetes mellitus