[Role of tumor necrosis factor α in endothelial-mesenchymal transition in vitro]

To observe the role of tumor necrosis factor α (TNF-α) in endothelial-mesenchymal transition (EnMT), and to explore the mechanism of fibrosis disease.

Human umbilical vein endothelial cells (HUVEC) from umbilical cord of healthy fetus were isolated by enzymatic digestion and identified by immunofluorescence assay. The third to fifth generations of cultured HUVEC in logarithmic phase were harvested and seeded in 12-well plates and 6-well plates, and they were divided into control group (ordinary culture without any stimulation), 5, 10, 25, 50, and 100 ng/mL TNF-α groups (5, 10, 25, 50, 100 ng/mL of TNF-α was respectively added into the nutrient solution) according to the random number table, with three samples in each group. After being cultured for 72 hours, the cell morphology was observed under inverted phase-contrast microscope; the expression levels of coagulation factor VIII and α smooth muscle actin (α-SMA) were detected by immunofluorescence assay, and the ratios of numbers (absorbance values) of cells with expression of both factors were calculated. The mRNA expression levels of cadherin, α-SMA, and type I collagen were detected by RT-PCR (denoted as gray value ratio). Data were processed with one-way analysis of variance and LSD test.

(1) The shape of primary HUVEC was round, short-spindle, or flat, and cells grew vigorously in cobblestone appearance after passages. After being subcultured for 1, 2, 3, 4, 5 passage (s), the positive rate of coagulation factor VIII of HUVEC was respectively (85.5 ± 1.8)%, (88.1 ± 5.0)%, (93.6 ± 3.7)%, (92.9 ± 4.8)%, (89.5 ± 1.1)%, and they were significantly higher than that of primary HUVEC [(81.4 ± 3.8)%, with F values all equal to 7.481, P values all below 0.05]. (2) As compared with that in control group, the appearance of cells in 5, 10, 25, 50, and 100 ng/mL TNF-α groups was gradually transformed from round, short-spindle, or flat shape to long-spindle shape with reduced intercellular junction and larger intercellular gap along with the increase in the concentration of TNF-α. (3) The ratios of numbers and the absorbance values of coagulation factor VIII and α-SMA double positive cells in control group (0.055 ± 0.015, 0.078 ± 0.017) were significantly lower than those in 5, 10, 25, 50, and 100 ng/mL TNF-α groups (0.257 ± 0.106, 0.280 ± 0.129, 0.505 ± 0.059, 0.817 ± 0.035, 0.929 ± 0.101 and 0.437 ± 0.040, 0.456 ± 0.097, 0.496 ± 0.082, 0.787 ± 0.131, 0.885 ± 0.087, with F value respectively 45.009, 50.099, P values all below 0.01). (4) The expression levels of cadherin mRNA in 5, 10, 25, 50, and 100 ng/mL TNF-α groups were 0.70 ± 0.05, 0.63 ± 0.06, 0.60 ± 0.10, 0.45 ± 0.16, and 0.26 ± 0.14, and it was significantly lower in the latter four groups than in control group (0.83 ± 0.03, with F values all equal to 11.593, P < 0.05 or P < 0.01). The mRNA expression levels of α-SMA and collagen I in 5, 10, 25, 50, and 100 ng/mL TNF-α groups were 0.45 ± 0.10, 0.51 ± 0.16, 0.49 ± 0.12, 0.60 ± 0.09, 0.76 ± 0.03 and 0.38 ± 0.18, 0.45 ± 0.15, 0.52 ± 0.12, 0.66 ± 0.17, 0.76 ± 0.20, and they were significantly higher in the latter three groups than in control group (0.37 ± 0.14, 0.31 ± 0.12, with F value respectively 7.839, 2.898, P < 0.05 or P < 0.01).

TNF-α can obviously promote EnMT in a dose-dependent manner. EnMT may be another significant source of myofibroblasts that contributes to fibrotic tissue in scar formation.