Interleukin-6 and tumor necrosis factor-alpha levels of endothelial cells in different hypoxia modes: in vitro experiment

Abstract

To investigate the damage of different patterns of intermittent hypoxia (IH) and continuous hypoxia (CH) on endothelial cells. Human umbilical vein endothelial cells of the line ECV304 were cultured in a program-controlled gas delivery system newly developed and divided into 8 groups to undergo different IH/reoxygenation (ROX) cycles so as to simulate the patterns of hypoxic episode seen in recurrent apnea and chronic obstructive pulmonary disease: intermittent normoxia (IN) group (exposed to 21% O2 15 s/21% O2 225 s for 60 cycles), IH group (exposed to 1.5% O2 15 s/21% O2 225 s for 30 or 60 cycles), IH hypercapnia group (exposed to 1.5% O2 and 20% CO2 15 s/21% O2 and 5% CO2 225 s, for 60 cycles), continuous hypoxia (CH group, exposed to 1.5% or 10% O2 for 15, 30 or 60 min), CH hypercapnia group (exposed to 10% O2 and 10% CO2 for 15, 30 or 60 min), CH added to IH group (exposed to 1.5% O2 15 s/10% O2 225 s for 60 cycles), different intermittent hypoxia extent group (exposed to 1.5% or 10% O2 15 s/21% O2 225 s for 60 cycles), different intermittent hypoxia frequency group (exposed to 1.5% O2 15 s/21% O2 225 s 315 s, 495 s or 105 s for 60 cycles), and different intermittent hypoxia duration group (exposed to 1.5% O2 15 s or 30 s/21% O2 225 s for 60 cycles). Then ELISA was conducted to examine the levels of interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNFalpha) Bicinchoninic acid method was used to standardize the cell total protein level. The levels of IL-6 and TNFalpha levels in the IH group were (770.40 +/- 21.60) and (126.93 +/- 2.58) pg.ml(-1).100 mg protein(-1) respectively, both significantly higher than those in the IN group [(374.06 +/- 38.10) and (31.96 +/- 13.64) pg.ml(-1).100 mg protein(-1) respectively, U = 0.000, P = 0.002], but significantly lower than those in the IH hypercapnia group [(829.27 +/- 7.16) and (78.77 +/- 4.00) pg.ml(-1).100 mg protein(-1) respectively, U = 0.000, P = 0.002]. The IL-6 levels of the CH hypercapnia 15, 30, and 60 min subgroups were all significantly higher than those of the corresponding CH subgroup (U = 0.000, P = 0.002). The IL-6 and TNFalpha levels of the CH added to IH group were (536.74 +/- 14.97) and (51.10 +/- 6.80) pg.ml(-1).100 mg protein(-1) respectively, both were significantly higher than those of the IN group, but significantly lower than those of the IH group (chi(2) = 23.4, P < 0.05). The levels of IL-6 and TNFalpha increased along with the increase of the IH degree (chi(2) = 23.4, P < 0.05). The level changes of IL-6 and TNFalpha of the groups with different intermittent hypoxia frequency and with different intermittent hypoxia duration were complicated. IH and CH significantly damage the endothelial cells dose-dependently, especially combined with hypercapnia. In IH/ROX, the inflammatory damage comes from ROX phase but not IH phase. Hypoxia duration and hypoxia frequency are also important parameters in the activation of inflammation.