The participation of tumor necrosis factor in the pathogenesis of lung allograft rejection in the rat.

Tumor necrosis factor alpha and beta are polypeptide cytokines with a wide range of metabolic, immunologic, and inflammatory activities. TNF is known to participate in immune mediated injury of native lungs, but a role for TNF in mediating lung allograft rejection (AR) has not been established. In experiments reported here, we assessed the role of TNF in mediating lung AR in a rat model of lung transplantation (BN-->Lew) (RT1n-->RT1l). This model shows florid AR with all grafts completely destroyed by day 6 posttransplant. Graft pathology is characterized by massive lymphocytic infiltrates and hemorrhagic necrosis. Initially, 5 lung allograft recipients in each group were sacrificed on days 1 to 6 posttransplant. Allografts were removed, mRNA isolated, and Northern blotting or RT-PCR performed with blots probed with cDNAs or oligos specific for rat TNF-alpha cyclophylin and gamma-actin. Data were compared with syngeneic transplants (Lew-->Lew) and with normal controls. In addition, frozen lung allograft tissue was examined by indirect immunofluorescence, using antibodies specific for TNF. TNF-alpha mRNA levels were detectable on day 2 posttransplant, and peaked on days 6-7 posttransplant. IF studies showed TNF protein expression in mononuclear cells of rejecting allografts on day 3, peaking on day 6. Both TNF-alpha mRNA and protein levels correlated with maximal AR and hemorrhagic necrosis of grafts. Minimal TNF-alpha mRNA or protein was detected in syngeneic grafts or in contralateral native lungs. We then examined the ability of a rabbit polyclonal anti-TNF-alpha (7000 U/day) and anti-TNF-beta (5000 U/day) with 30% crossreactivity with rat TNF to modify the AR response. For each group, 4-5 left lung transplants were performed as described, and animals treated with anti-TNF-alpha, anti-TNF-beta, (anti-TNF-alpha+anti-TNF-beta) or with preimmune rabbit sera. All animals were sacrificed on day 6 posttransplant. Several pathological categories of inflammation were examined and scored (0-4), with a score of 0 = 0% involvement; 1 = 1-25% involvement; 2 = 26-50% involvement; 3 = 51-75% involvement; and 4 = 76-100% involvement. The mean and SD scores were obtained for all animals in the treatment categories mentioned above, and compared with preimmune-treated controls. Briefly, no differences in perivascular, peribronchial, or peribronchiolar cell infiltrates or edema were seen in treatment groups compared with controls.(ABSTRACT TRUNCATED AT 400 WORDS)