To characterize the mechanisms determining tissue-specific ceruloplasmin gene expression during development, the rat ceruloplasmin gene was isolated in a series of overlapping phage clones. The 5'-flanking region was characterized and the transcription initiation site was identified by primer extension and RNase protection. Nucleotide sequence analysis of this region revealed a typical eukaryotic promotor structure, but no obvious homology with cis-acting elements previously characterized as determining tissue-specific gene expression. Transient expression of chimeric ceruloplasmin-reporter gene constructs containing up to 5200 base pairs (bp) of the 5'-flanking region revealed that sequences 732 bp upstream of the start nucleotide were sufficient to confer hepatocyte-specific expression. The region from -393 to -348 was determined by deletion analysis to contain a positive-acting element, and includes sequence partially homologous to the rat albumin D site. Mobility shift analysis revealed that this region specifically binds a heat-labile nuclear protein from rat liver and from newborn but not adult rat lung. Binding to this region was competed by oligonucleotides corresponding to the albumin D site, but not by oligonucleotides corresponding to binding sites for the hepatocyte transcription factors HNF-1, HNF-3, HNF-4, and C/EBP. These data indicate that ceruloplasmin gene expression is determined in part by a cis-acting region 393 bp upstream of the transcription start site, which binds a previously uncharacterized nuclear protein. The tissue distribution of this nuclear protein suggests that it plays a role in directing ceruloplasmin gene expression in lung and liver during development.

To determine the effect of inflammation on extrahepatic ceruloplasmin gene expression we examined the ceruloplasmin mRNA content of adult rat tissues after endotoxin injection. Within 8 h of a dose of endotoxin ceruloplasmin mRNA content increased in the liver as expected and was also detectable in the lung. The effect of endotoxin was tissue specific because ceruloplasmin mRNA was not consistently detected in other extrahepatic tissues. The kinetics of ceruloplasmin mRNA accumulation in lung and liver tissue were similar with a maximum seven- to ninefold increase in ceruloplasmin mRNA content in each tissue within 24 h. The relative rate of ceruloplasmin gene transcription was increased in both tissues within 3 h of endotoxin, suggesting similar mechanisms of regulation of ceruloplasmin gene expression during inflammation. One cellular site of ceruloplasmin production in the inflamed lung was found to be the alveolar macrophage, which expressed the ceruloplasmin gene and synthesized ceruloplasmin protein in response to endotoxin in vitro. Because of these findings we also examined the effects of hyperoxia on ceruloplasmin gene expression. Exposure of adult rats to 95% O2 resulted in a five- to sixfold induction of ceruloplasmin mRNA in lung tissue within 46 h, and this response was time dependent, reaching maximum values at 86 h. Hyperoxic induction of ceruloplasmin mRNA was specific to the lung and not the result of systemic inflammation because hepatic ceruloplasmin mRNA content remained constant. These data indicate that the lung is a prominent site of ceruloplasmin gene expression during inflammation and hyperoxia and suggest that this protein may play a previously unappreciated role in pulmonary injury or repair.

cDNA clones corresponding to rat ceruloplasmin were isolated from newborn rat lung and liver cDNA libraries and the nucleotide sequence was obtained. The derived amino acid sequence of rat ceruloplasmin is 93% homologous to the corresponding human sequence and contains a 19-amino acid leader peptide plus 1040 amino acids of mature protein. Southern blot analysis indicates that the ceruloplasmin gene exists as a single copy in the rat haploid genome. Using these cDNA clones in RNA blot analysis, a single 3.7-kilobase ceruloplasmin-specific transcript is detected in fetal rat liver and lung by day 15 of gestation. During fetal development the abundance of this transcript increases selectively in these two tissues and at birth is 60% of that found in the adult liver. Postnatally the temporal pattern of ceruloplasmin gene expression in lung and liver differs. Within the first 3 weeks postpartum ceruloplasmin mRNA content decreases in lung to undetectable levels, while that in the liver reaches adult levels. Primer extension reveals a single identical start site of ceruloplasmin gene transcription in lung and liver and biosynthetic studies indicate that each tissue synthesizes a ceruloplasmin protein which is qualitatively similar to that synthesized by adult liver. Ceruloplasmin mRNA is also detected in human fetal lung explant and a human lung adenocarcinoma cell line suggesting that a similar pattern of expression occurs in the developing human lung. These data indicate that lung is the predominant extrahepatic site of ceruloplasmin gene expression during fetal development and suggest that this protein may play a previously unappreciated role in lung development or pulmonary antioxidant defense.