The biosynthetic rates for both the transferrin receptor (TfR) and ferritin are regulated by iron. An iron-responsive element (IRE) in the 5' untranslated portion of the ferritin messenger RNA (mRNA) mediates iron-dependent control of its translation. In this report the 3' untranslated region of the mRNA for the human TfR was shown to be necessary and sufficient for iron-dependent control of mRNA levels. Deletion studies identified a 678-nucleotide fragment of the TfR complementary DNA that is critical for this iron regulation. Five potential stem-loops that resemble the ferritin IRE are contained within the region critical for TfR regulation. Each of two of the five TfR elements was independently inserted into the 5' untranslated region of an indicator gene transcript. In this location they conferred iron regulation of translation. Thus, an mRNA element has been implicated in the mediation of distinct regulatory phenomena dependent on the context of the element within the transcript.

Regulated translation of messenger RNA offers an important mechanism for the control of gene expression. The biosynthesis of the intracellular iron storage protein ferritin is translationally regulated by iron. A cis-acting element that is both necessary and sufficient for this translational regulation is present within the 5' nontranslated leader region of the human ferritin H-chain messenger RNA. In this report the iron-responsive element (IRE) was identified by deletional analysis. Moreover, a synthetic oligodeoxynucleotide was shown to be able to transfer iron regulation to a construct that would otherwise not be able to respond to iron. The IRE has been highly conserved and predates the evolutionary segregation between amphibians, birds, and man. The IRE may prove to be useful for the design of translationally regulated expression systems.

The human ferritin H chain messenger RNA contains a specific iron-responsive element (IRE) in its 5' untranslated region, which mediates regulation by iron of ferritin translation. An RNA gel retardation assay was used to demonstrate the affinity of a specific cytosolic binding protein for the IRE. A single-base deletion in the IRE eliminated both the interaction of the cytoplasmic protein with the IRE and translational regulation. Thus, the regulatory potential of the IRE correlates with its capacity to specifically interact with proteins. Titration curves of binding activity after treatment of cells with an iron chelator suggest that the factor acts as a repressor of ferritin translation.

Ferritin plays a key role in determining the intracellular fate of iron and is highly regulated by the iron status of the cell. We have identified a cis-acting element in the transcribed but nontranslated 5' leader sequence of human ferritin heavy-chain mRNA. In transiently transfected murine fibroblasts, the presence of a 157-nucleotide region of the 5' leader sequence was found to be necessary for iron-dependent regulation of ferritin biosynthesis. Further, this 5' leader region is sufficient to transfer iron-mediated translational control to the expression of a heterologous gene product, chloramphenicol acetyltransferase.Ferritin is the major intracellular repository of iron and exists as a heteropolymer of 24 subunits containing both heavy and light chains (1). It serves to sequester and thereby detoxify iron not otherwise utilized for cellular metabolism (2). A more active role of ferritin has been suggested (3), in which iron sequestration may limit and regulate iron availability for cellular metabolism. Ferritin's key role is highlighted by the following observations. (i) All eukaryotic cells investigated thus far have been shown to express ferritin (4). (ii) Cellular proliferation is dependent on the availability of iron, and ferritin levels are closely regulated during normal proliferation and have been found to be abnormal in various malignancies (5, 6). (iii) Ferritin in the cells lining the duodenal mucosa is likely to regulate intestinal iron absorption. Furthermore, since one ofthe most common autosomal recessive disorders, hereditary hemochromatosis, is characterized by pathologically increased absorption of dietary iron, abnormal regulation of ferritin expression or abnormal ferritin function could play a major role in this disorder (7). Thus, the regulation of ferritin levels by iron is the means by which ferritin can both influence and respond to iron homeostasis.It is therefore important to elucidate the mechanisms by which ferritin gene expression is regulated. Munro and his colleagues (8, 9) suggested a translational regulation of ferritin expression in rat liver, based on two observations: (i) actinomycin D does not inhibit the stimulatory effect of iron on ferritin expression and (ii) cytoplasmic ferritin mRNA shifts from an inactive messenger-ribonucleoprotein pool to translationally active polysomes upon iron induction. Similar results were obtained for reticulocytes and liver of bullfrog tadpoles (10, 11). In further support of translational control of ferritin expression, cytoplasmic mRNA levels were shown to remain constant over a wide range of biosynthetic rates induced by iron (9). The recent cloning of the human ferritin heavy-chain cDNA and gene (12)(13)(14)(15) has allowed us to directly address this question of the mechanism of biosynthetic regulation at a molecular level.Previous work showed that a 7.2-kilobase (kb) HindIII genomic DNA fragment (pUCM11) is actively transcribed in stably transformed and transiently transfected murine B6 fibroblasts (13), giving rise to a...

No abstract

At the 5' end of the rubella virus genomic RNA, there are sequences that can form a potentially stable stem-loop (SL) structure. The complementary negative-strand equivalent of the 5'-end SL structure of positive-strand rubella virus RNA [5' (+) SL structure] is thought to serve as a promoter for the initiation of positive-strand synthesis. We screened the negative-strand equivalent of the 5' (+) SL structure (64 nucleotides) and the adjacent region of the negative-strand RNA for their ability to bind to host cell proteins. Specific binding to the 64-nucleotide-long potential SL structure of three cytosolic proteins with relative molecular masses of 97, 79, and 56 kDa was observed by UV-induced covalent cross-linking. There was a significant increase in the binding of the 97-kDa protein from cells upon infection with rubella virus. Altering the SL structure by deleting sequences in either one of the two potential loops abolished the binding interaction. The 56-kDa protein also appeared to bind specifically to an SL derived from the 3' end of positive-strand RNA. The 3'-terminal structure of rubella virus negative-strand RNA shared the same protein-binding activity with similar structures in alphaviruses, such as Sindbis virus and eastern equine encephalitis virus. A possible role for the host proteins in the replication of rubella virus and alphaviruses is discussed.

No abstract

The number of microscopically visible lesions produced on the membrane for a given degree of lysis on normal cells as on PNH cells. Since complement lesions were not formed until C8 or C9 was incorporated into the complement sequence, the results suggest that increased lysis of red cells in PNH is due at least in part to more efficient penetration of the PNH membrane by the terminal lytic sequence of complement. Furthermore, the efficiency of the terminal lytic sequence in the lysis of PNH cells when complement was activated by the alternative pathway and the classical pathway was analyzed. There was no significant difference (p less than 0.01) in the number of lesion present at an equivalent degree of lysis when initiated by antibody, cobra venom factor, or acidification. Thus, the efficiency of the terminal lytic sequence does not vary with different modes of activation.