Endothelial TNF-α induction by Hsp60 secreted from THP-1 monocytes exposed to hyperglycaemic conditions

Abstract

A non-resolving inflammation of the endothelium is recognised to be an important process leading to atherosclerosis. In diabetes, this process is thought to account for a significant number of cardiovascular disease-associated death and disability. However, the molecular mechanisms by which diabetes contributes to endothelial inflammation remain to be established. Whilst there is some evidence linking hyperglycaemia-induced reactive oxygen species (ROS) formation by the mitochondrial electron-transport chain to oxidative stress, cellular injury and apoptosis in the endothelium, a clear link to endothelium inflammation has not yet been established. The mitochondrial molecular stress protein Hsp60 is known to be secreted from mammalian cells and is capable of activating pro-inflammatory mediators on target cells expressing Toll-like receptors (TLRs). Hsp60 is also known to be elevated in serum of diabetes patients and has been shown to be upregulated by hyperglycaemic growth conditions in cultured human HeLa cells. This study shows that Hsp60 induced in human acute monocyte leukaemia cell line (THP-1) cells grown under hyperglycaemic conditions (25 mM glucose) was able to be secreted into growth media. Furthermore, the secretion of Hsp60 from THP-1 cells was able to be inhibited by 5,5-(N-N-dimethyl)-amiloride hydrochloride (DMA), an exosomal inhibitor. Interestingly, the conditioned media obtained from THP-1 cells grown in the presence of 25 mM glucose was able to induce the secretion of TNF-α in human vascular endothelium cell line (HUVEC). When conditioned media was immuno-depleted of Hsp60, there was a significant reduction in the release of TNF-α from the HUVEC cells. This suggests that a potential link may exist between hyperglycaemia-induced expression of Hsp60 in monocyte cells and vascular inflammation. Circulating levels of Hsp60 due to mitochondrial stress in diabetes patients could therefore be an important modulator of inflammation in endothelial cells and thus contribute to the increased incidences of atherosclerosis in diabetes mellitus.

Keywords: Endothelial cells; Hsp60; Hyperglycaemia; Inflammation; THP-1 monocytes; TNF-α.

Fig. 1

Mitochondrial dehydrogenase activity of THP-1 cells grown for 5 days in the presence of 10 and 25 mM glucose expressed as a percentage of the 5 mM glucose levels. N = 3 independent experiments. The error bars represent ±S.E.M. (*) represents a statistically significant value (p < 0.05)

Fig. 2

I. Western blot of Hsp60 and actin expression from THP-1 cells grown for 5 days in the presence of 5 and 25 mM glucose. Lanes A and B cells grown in 25 and 5 mM glucose, respectively. Lane C is a heat-shocked positive control. The Western blot is a representative of N = 3 independent experiments. II. Expression of Hp60 relative to actin. Cells exposed to 25 mM glucose (lane A) had a 3.0-fold induction compared with cells grown in 5 mM glucose (lane B). The heat-shocked cells (lane C) which acted as a positive control had a 3.17-fold induction compared to cells grown in 5 mM glucose (lane B). The error bars represent ±S.E.M. (*) represents a statistically significant value (p < 0.05)

Fig. 3

Hsp60 detected in glucose-conditioned media (GCM) by anti-Hsp60 ELISA. The GCM (25 mM) contained 15 ± 3.5 ng/ml/10⁶ cells of Hsp60 protein (lane A). In the presence of DMA (at 5 or 10 nM), there was an 83–88% decrease in Hsp60 detected compared to levels seen in GCM (25 mM) (lanes B and C, respectively). The data is representative of N = 3 independent experiments. The error bars represent ±S.E.M. (**) represents a statistically significant value (p < 0.005)

Fig. 4

LDH assays of THP-1 cells grown in the presence of 5 and 25 mM glucose which were exposed to DMA (5 or 10 nM). The percentage of LDH content (relative to maximum control samples) was interpreted as the percentage of lysed cells on the assumption that every dead cell released equal amounts of LDH. The data is representative of N = 3 independent experiments. The error bars represent ±S.E.M.

Fig. 5

TNF-α secretion from HUVEC cells as measured by anti-TNF-α ELISA after incubation with GCM from THP-1 cells. When HUVEC cells were exposed to GCM (5 mM), no TNF-α was detected (A); when HUVEC were exposed to GCM (25 mM), 8 ± 2 pg/ml of TNF-α was detected (B); when cells were exposed to GCM (25 mM) in the presence of 5 and 10 nM DMA, there was a significant reduction in the amount of TNF-a secreted (p < 0.05) (C and D, respectively); when HUVEC cells were exposed to Hsp60 immuno-depleted GCM (25 mM), there was a significant reduction in the amount of TNF-α secreted (p < 0.05) (E); responsiveness of the HUVEC cells to IL-1α is shown in (H); DMA had no effect on the TNF-α responsiveness of HUVEC cells to stimulation by IL-1α (F and G). The data is representative of N = 3 independent experiments. Error bars represent ±S.E.M. (*) represents a statistically significant value (p < 0.05)