Background: Recently, the research on hybrid artificial organs such as heart, liver, and kidney has been reported to yield new possibilities for clinical use. However, there have been few reports on the practical use of a hybrid artificial lung, primarily due to difficulties in the concept for its application and techniques for cell attachment. In this study, we investigated the possibility for developing a novel strategy: a hybrid artificial lung with constitutive nitric oxide synthase (NOS) and interleukin (IL)-10 gene-transfected endothelial cells to attenuate inflammatory reactions induced by cardiopulmonary bypass.
Methods and results: First, we performed an in vitro study to confirm the efficacy of our gene transfection into endothelial cells. Constitutive nitric oxide synthase and IL-10 cDNA were transfected into endothelial cells according to the hemagglutinating virus of Japan-liposome method. Levels of nitric oxide released from the endothelial constitutive NOS (ecNOS)-transfected endothelial cells were significantly higher than those of control cells (24 hours after the stimulation by lipopolysaccharide: 284.5 +/- 54.0 versus 95.7 +/- 27.9 mumol/L, P = 0.0001). On the other hand, IL-8 levels in the transfected endothelial cells were significantly lower than those in the control group (48 hours after stimulation by tumor necrosis factor-alpha: 3.1 +/- 2.4 versus 62.1 +/- 1.3 ng/mL, P = 0.0002), and IL-10 was detected in the transfected group but not in the control group. Next, we performed an in vivo study to evaluate the possibility of developing a hybrid artificial lung. One-hour partial cardiopulmonary bypass with this lung was established in rats undergoing femorofemoral bypass. Artificial lungs with no cells (group C; n = 5), that were coated with untreated endothelial cells (group E; n = 5), or that were coated with ecNOS and IL-10 gene-transfected endothelial cells (group T; n = 3) were used. At 45 minutes after the initiation of cardiopulmonary bypass, group T showed higher nitric oxide levels than groups C and E (T versus C versus E: 75.2 +/- 6.8 versus 67.2 +/- 4.3 versus 68.6 +/- 5.2 mumol/L, P = NS). The serum IL-8 levels just after cardiopulmonary bypass in group T were significantly lower than those in group C (1728 +/- 282 versus 4275 +/- 145 pg/mL, P = 0.0151). The Pao2 levels in group T just after weaning from cardiopulmonary bypass were significantly higher than those in group C (271.3 +/- 41.7 versus 136.6 +/- 12.3 mm Hg, P = 0.0362).
Conclusions: These results demonstrate that a hybrid artificial lung with IL-10 and ecNOS gene-transfected endothelial cells inhibited IL-8 release and increased nitric oxide production. This suggests the possibility of developing a hybrid artificial lung capable of preserving native lung function by attenuating cardiopulmonary bypass-induced inflammatory reactions via inhibition of IL-8 release and enhanced nitric oxide production.