Alterations of leptin in the course of inflammation and severe sepsis

Abstract

Background: The adipokine leptin regulates energy expenditure, vascular function, bone and cartilage growth as well as the immune system and systemic inflammatory response. Several activating effects towards T cells, monocytes, endothelium cells and cytokine production have been reported suggesting a protective role of leptin in the setting of an acute systemic inflammation. However, the pathophysiological role of leptin during severe sepsis is currently not elucidated in detail. This study aims to investigate leptin expression in cultured human adipocytes within an inflammatory model and in patients suffering from severe sepsis and evaluates treatment effects of drotrecogin alpha (activated) (DAA), the recombinant form of human activated protein C.

Methods: In an in-vitro inflammatory model of adipocyte cell-culture the effect of DAA on leptin mRNA expression was evaluated. Synthesis of mRNA was measured by quantitative polymerase chain reaction (qPCR). Additionally, supernatants of these adipocytes as well as serum levels of adiponectin were measured in blood of 104 severe septic patients by ELISA-method. 26 patients were treated with DAA (DAA+), 78 patients were not treated with DAA (DAA-).

Results: Stimulation of human adipocytes with TNF alpha over 6 and 24 hours resulted in a significant decrease by 46% and 59% of leptin mRNA transcripts compared to un-stimulated controls (p < 0.05). Leptin levels of supernatants of adipocyte culture decreased by 25% and 23% (p < 0.05) after incubation with TNF alpha after 6 and 24 hours. Incubation with DAA at 50 ng/ml DAA and 5 μg/ml doubled mRNA expression significantly at 24 hours (p < 0.05) but not at 6 hours. From day 1 to day 3 of sepsis, leptin levels increased in DAA+ compared to DAA- patients (p<0.10).

Conclusions: Leptin appears to be involved in the pathogenesis of a systemic inflammatory response during sepsis. Administration of DAA significantly increased leptin expression. The specific mechanism or even benefit of DAA towards leptin needs further ongoing research.

Trial registration: ClinicalTrials.gov NCT00222222.

Figure 1

Confluent cultures of undifferentiated preadipocytes (a) and mature adipocytes after differentiation for 15 days (b,c). The high amount of fat droplets, visible as light globules (b,c), demonstrates the cells being differentiated to mature adipocytes. Cells were photographed with 100x (a,b) and 400x (c) magnification.

Figure 2

Leptin mRNA expression was significantly decreased after 6 and 24 hours of stimulation with 1 ng/ml TNF alpha (p < 0.05) in human adipocytes. Drotrecogin alpha (activated) (DAA) dose-dependently increased leptin mRNA levels after 24 hours from DAA 50 ng/ml to DAA 5 μg/ml (p < 0.05), but at 6 hours no significant change could be detected. Significances (n.s. = not significant) of t-tests with Holm-Bonferroni adjustments for multiple comparisons are indicated. Only the relevant in advance specified group comparisons were analysed. Data are presented as mean with standard error of mean (SEM).

Figure 3

Leptin levels out of supernatants of the adipocytes cell culture measured by ELISA showed decreased levels of leptin after incubation with TNF alpha for 6 and 24 hours (p < 0.05). Increasing levels of leptin were observed after 6 hours of incubation according to the applied dosages of DAA (50 ng/ml and 5 μg/ml). Significances (n.s. = not significant) of t-tests with Holm-Bonferroni adjustments for multiple comparisons are indicated. Only the relevant in advance specified group comparisons were analysed. Data are presented as mean with standard error of mean (SEM).

Figure 4

Treating patients with DAA resulted in an increase in leptin levels after 96 hours of DAA infusion compared to patients without treatment of DAA from day 1 to day 3 and 5 (p < 0.05) (leptin concentrations, Day 1: DAA+, mean = 18991 pg/ml, ±SEM = 6009 pg/ml. Day 3: DAA+, mean = 57315 pg/ml, ± SEM = 17075 pg/ml. Day 5: DAA+, mean = 54980 pg/ml, ±SEM = 17660 pg/ml). Significances of t-tests with Holm-Bonferroni adjustments for multiple comparisons are indicated, otherwise no significance could be reached. Data are presented as mean with standard error of mean (SEM). * p < 0.10, ** p < 0.05.