Fragments of human genomic DNA corresponding to the promoter region of the £ene for the transferrin receptor have been cloned upstream or the bacterial gene for chloramphenicol acetyltransferase ana these constructs used to assess promoter activity following transfection into a human rhabdomyosarcoma cell line. Progressive 5' deletions as well as internal linker-substitution constructs support a critical role in gene expression of a sequence element approximately 70 bp upstream of the mRNA start site. In this region, the receptor gene was found to contain llbp that are identical to a segment of the enhancers of polyoma virus and adenovirus. A fragment encompassing this element was shown to increase gene expression when the fragment was placed in either orientation upstream of the remainder of the transfernn receptor promoter but the same fragment did not activate an enhancer-less SV40 promoter. Removal from within the receptor promoter of three potential binding sites for the transcription factor Spl did not decrease the promoter's activity.

Genomic DNA fragments corresponding to the promoter region of the human transferrin receptor were linked to either the full-length receptor cDNA or to the bacterial enzyme chloramphenicol acetyltransferase. These constructs were transfected into mouse and human cells, respectively. Gene expression was monitored 40-48 hours after transfection. Bal31 exonuclease was employed to produce 5' to 3' deletions of the promoter region. Deletion of DNA between -86 and -70 upstream of the receptor's mRNA start site resulted in a greater than 80% reduction in apparent promoter activity. DNA sequencing of the 150 bp upstream of the start site revealed that the promoter region contained several sequence elements more than 90% homologous to the consensus sequence for binding of the transcription factor Sp1. In addition, an 11 bp sequence identical to a segment of the enhancers of polyoma virus and adenovirus was located between -80 and -70. Internal deletions confirmed that this enhancer homologue was critical for full promoter activity. A 66 bp fragment encompassing the -80/-70 element augmented gene expression when the fragment was placed in either orientation upstream of the remainder of the transferrin receptor promoter.

Iron regulation of the human transferrin receptor gene was examined in murine cells transformed with chimeric constructs containing the human transferrin receptor gene's promoter and either the structural gene for bacterial chloramphenicol acetyltransferase or the human transferrin receptor cDNA. The activity of the transferrin receptor gene's promoter with the heterologous indicator gene was found to be approximately equal to 3-fold higher in cells treated with the iron chelator desferrioxamine than in cells treated with the iron source, hemin. A higher degree of iron regulation was seen in the expression of the human transferrin receptor cDNA driven by its own promoter. The receptor cDNA under the control of the simian virus 40 early promoter was also iron-regulated. Several human transferrin receptor transcripts differing in their 3' end were produced in the murine cells regardless of the promoter used, with the shorter transcripts being relatively unregulated by iron. Deletion of cDNA corresponding to most of the 3' untranslated portion of the mRNA for the receptor ablated the iron regulation. We conclude that at least two genetic elements exist for the regulation of the transferrin receptor gene by iron. One has its locus in the DNA upstream of the transferrin receptor gene's transcription start site, and the other is dependent upon the integrity of the sequences in the 3' end of the gene.

Fragments of human genomic DNA corresponding to the promoter region of the gene for the transferrin receptor have been cloned upstream of the bacterial gene for chloramphenicol acetyltransferase and these constructs used to assess promoter activity following transfection into a human rhabdomyosarcoma cell line. Progressive 5' deletions as well as internal linker-substitution constructs support a critical role in gene expression of a sequence element approximately 70 bp upstream of the mRNA start site. In this region, the receptor gene was found to contain 11bp that are identical to a segment of the enhancers of polyoma virus and adenovirus. A fragment encompassing this element was shown to increase gene expression when the fragment was placed in either orientation upstream of the remainder of the transferrin receptor promoter but the same fragment did not activate an enhancer-less SV40 promoter. Removal from within the receptor promoter of three potential binding sites for the transcription factor Sp1 did not decrease the promoter's activity.

A wide variety of substances that would otherwise be excluded from cells gain entry via receptor-mediated endocytosis [1, 2], Binding of ligands to their specific plasma membrane receptors is followed by internalization involving specialized regions of the membrane termed coated pits [2, 3]. While the pathways that are traversed subsequently by diverse ligands exhibit striking similarities, there are distinctive features that separate endocytic systems into several groups. The endocytosis of low density lipoproteins (LDL) by fibroblasts [4] and of asialoglycopro-teins (ASGP) by hepatocytes [5] are representatives of systems in which ligands are catabolized in lysosomes and receptors are reutilized. Constant numbers of receptors are maintained during extended continuous endocytosis of ligand molecules. In other systems, such as those mediating uptake of insulin [6] or epidermal growth factor (EGF) [7], both the ligand and its receptor are degraded in lysosomes. These systems exhibit “down-regulation” in that ligands enhance receptor degradation. As a result of a reduction in the number of receptors, the target cells exhibit lowered responsiveness to the continued presence of these ligands.

Treatment of K562 cells with desferrioxamine, a permeable iron chelator, led to an increase in the number of transferrin receptors. Increasing intracellular iron levels by treatment of cells with either human diferric transferrin or hemin lowered the level of the transferrin receptors. By using a cDNA clone of the human transferrin receptor, we showed that the changes in the levels of the receptor by iron were accompanied by alterations in the levels of the mRNA for the receptor. The rapidity of these changes indicated that the mRNA had a very short half-life. By using an in vitro transcriptional assay with isolated nuclei, we obtained evidence that this regulation occurred at the transcriptional level.

We have examined the mechanism by which hemin regulates the expression of the human transferrin receptor. Previous work led to the suggestion that the regulatory signal is provided by heme (Ward J. H., Jordan, I., Kushner, J. P., and Kaplan, J. (1984) J. Biol. Chem. 259, 13235-13240). We demonstrated that hemin regulates the expression of the receptor via alterations in the rate of receptor biosynthesis. However, this effect can be completely abolished by addition of desferrioxamine, an intracellular iron chelator. Competition curves demonstrate that desferrioxamine and hemin affect the same intracellular iron pool. Since the chelator cannot remove iron from heme, we propose that hemin acts simply by delivering iron to a chelatable iron pool and that levels of chelatable iron provide the regulatory signal for expression of the transferrin receptor gene.

Treatment of K562 cells with the iron chelator desferrioxamine results in the gradual increase in total cell receptors for transferrin. Receptor number rises 2.5-4.5-fold over 24 h and remains at the elevated level if the chelator is continuously present. Preincubation of the chelator with ferric chloride abolishes the effect. The drug has no effect on the 7-h half-life of the receptor. The increased number of receptors can be accounted for by a specific increase in the rate of receptor biosynthesis which reaches 3-4 times that seen in untreated cells by 6 h after the addition of the chelator. Isolation of mRNA from treated cells reveals that, after 8 h in the presence of desferrioxamine, there is a 3-fold increase in the specific translation of transferrin receptor over untreated cells. Total protein synthesis is not changed under these conditions.