Cytokine gene expression is a critical component of the lung allograft rejection (AR) response and tolerance development in rat models. In order to determine the specificity of cytokine gene expression for AR and tolerance, we examined cytokine (interleukin-2) (IL-2), (gamma-interferon) (gamma-IFN), IL-4, IL-10 and tumor necrosis factor-alpha (TNF-alpha) and control (cyclophilin) mRNA levels in two models of rat lung allograft rejection by RT-PCR (reverse transcriptase polymerase chain reaction), Southern blotting. The first model (WKY-->F344) develops a mild to moderate lymphocytic infiltrate on days 14-21 post-transplant (stage II-III AR), which spontaneously resolves by day 35 post-transplant with subsequent development of allograft tolerance (grafts surviving without evidence of AR for > 140 days). Conversely, F344-->WKY develops a similar lymphocytic infiltrate by day 14 post-transplant, but by day 21 post-transplant the graft shows severe AR (stage III-IV) and has haemorrhagic infarction with alveolar haemorrhage. METHODS RNA was extracted from allografts removed on days 3, 7, 14, 21, 35 and 42 post-transplant. Five animals for each group (WKY-->F344) and F344-->WKY) were examined at each time point, except that no animals in the F344-->WKY were examined on day 42. cDNA was synthesized from total extracted RNA and primers specific for rat TNF-alpha, rat IL-2, rat gamma-IFN, rat IL-4, rat-IL-10 and rat cyclophilin were used for gene-specific amplification. (TNF-alpha, gamma-IFN, IL-10, 20 cycles; IL-2, IL-4, 30 cycles; cyclophilin, 20 cycles). The cycles numbers chosen for comparison were found to be optimal during preliminary experiments and occurred during the exponential phase of amplification. PCR products were electrophoresed on a polyacrylamide gel and silver-stained. Gels were subsequently electrotransferred to nylon membranes which were probed with murine cDNAs specific for IL-2, gamma-IFN IL-4, IL-10 and TNF-gamma. RESULTS Cyclophilin gene expression was similar for both models at all time points tested; this also served as an internal standard for RT-PCR. In the WKY-->F344 tolerance model, TNF-alpha mRNA levels were not detectable on days 3 and 7 post-transplant, were at very low levels on day 14 and were undetectable on day 21 post-transplant. In marked contrast, the F344-->WKY rejection model showed TNF-alpha mRNA present on day 3 which increased markedly on day 7 and peaked on day 14 post-transplant. TNF-alpha mRNA levels decreased on days 21 and 35 post-transplant, a time when the lung was undergoing AR. The pattern of IL-2 and gamma-IFN mRNA expression was similar to that for TNF-alpha. However, IL-2 mRNA was clearly detectable in the WKY-->F344 tolerance model on day 7 and gamma-IFN was not present until day 14 post-transplant. The F344-->WKY rejection model showed very high levels of IL-2 and gamma-IFN on day 3 which peaked on day 14. The ratio of IL-2/IL-10 in the F344-->WKY rejection model was more than 5 times that seen in the WKY-->F344 tolerance model on day 3 (p < 0.0005). The ratio of IL-2/IL-4 was higher (1.5 times) in the F344-->WKY rejection model than in the WKY-->F344 tolerance model (p < 0.007) on day 3. On day 14 post-transplant, the IL-2/IL-10 ratio in the F344-->WKY rejection model was three times that of the tolerance model (p < 0.0015). The IL-2/IL-4 ratio was 3.5 times greater in the WKY-->F344 tolerance model than in the rejectin model (p < 0.003). This was due to equal expression of IL-2 and IL-4 in the rejection model, but poor IL-4 expression in the tolerance model. CONCLUSIONS 1) The WKY-->F344 tolerance model develops mild to moderate lymphocytic infiltrates on day 14 which is associated with low level IL-2, gamma-IFN and TNF-alpha gene expression. IL-10 and IL-4 are present at day 3; however, by day 14, IL-10 is the predominantly expressed Th2 cytokine and IL-4 is not expressed. The infiltrates ultimately resolve and the animals develop a functional tolerance to their grafts.4