Role of RNA secondary structure of the iron-responsive element in translational regulation of ferritin synthesis.

Iron regulates synthesis of the iron storage protein ferritin at the translational level through interaction between a stem-loop structure, the iron-responsive element (IRE), located in the 5'-untranslated region (5'-UTR) of ferritin mRNAs, and a protein, the iron regulatory protein (IRP). The role of IRE secondary structure in translational regulation of ferritin synthesis was explored by introducing ferritin constructs containing mutations in the IRE into Rat-2 fibroblasts. Our in vivo studies demonstrate that size and sequence of the loop within the IRE and the distance and/or spatial relationship of this loop to the bulged nucleotide region closest to the loop must be preserved in order to observe iron-dependent translation of ferritin mRNA. In contrast, changes in nucleotide sequence of the upper stem can be introduced without affecting translational regulation in vivo, as long as a stem can be formed. Our in vivo results suggest that only a very small variation in the affinity of interaction of IRP with IRE can be tolerated in order to maintain iron-dependent regulation of translation.

Mutagenesis of the iron-regulatory element further defines a role for RNA secondary structure in the regulation of ferritin and transferrin receptor expression.

Within the 5'-untranslated region of ferritin mRNAs, there is a conserved region of 28 nucleotides (nt) (the iron regulatory element (IRE)) that binds a protein (the IRE-binding protein (IRE-BP)) involved in the iron regulation of ferritin mRNA translation. We have examined the role of RNA secondary structure on the interaction of the IRE with the IRE-BP. First, the rat light ferritin IRE possesses a structure similar to that of the bullfrog heavy ferritin IRE (Wang, Y.-H., Sczekan, S. R., and Theil, E. C. (1990) Nucleic Acids Res. 18, 4463-4468). This includes an extended stem, interrupted at various points by bulge nucleotides and a 6-nt single-stranded loop (CAGUGU) at its top. Computer predictions and mapping results suggest the presence of a 3-nt (UGC) bulge 5 bases 5' of the loop in the rat IRE. Second, disruption of the base pairing in the upper stem alters IRE secondary structure and reduces the affinity with which the IRE-BP binds the IRE. Third, increasing the size of the loop or the distance between the UGC bulge and the loop reduces the IRE/IRE-BP interaction. Our results indicate that several aspects of IRE secondary structure are important for its high affinity binding to the IRE-BP.

Ferritin gene expression in health and malignancy.

Intracellular iron can be stored in the protein shell of ferritin to protect the cell against the toxic action of the iron. In response to increased cell iron, more ferritin subunits are synthesized using translational and transcriptional mechanisms. Translational control involves a unique stem-loop structure in the 5' untranslated region of the subunit messengers. When iron level is low, a protein binds to this stem-loop structure and prevents translation. When intracellular iron level rises, the repressor protein is discharged and the large population of messengers begins to translate subunits. Similar stem-loop motifs occur in the 3' untranslated region of the transferrin receptor messenger where they regulate breakdown of the receptor mRNA. Finally, the presence of excess iron preferentially stimulates transcription of more ferritin message of one type (L-mRNA) which produces ferritin shells favoring iron storage. In this way, protection of the cell against iron excess is enhanced by coordinate changes in rate of synthesis of ferritin mRNA of the L-type, by release of ferritin mRNA stored in the cytoplasm, and by a reduction in the number of receptors for accepting iron into the cell. The application of these principles with reference to malignant cells is discussed.

Regulation of ferritin and heme oxygenase synthesis in rat fibroblasts by different forms of iron.

Synthesis of the iron-storage protein ferritin is thought to be regulated at the translational level by the cytosolic content of chelatable iron. This response to iron is regulated by the iron-modulated binding to ferritin mRNAs of a repressor protein, the iron regulatory element-binding protein. From measurements made in a cell-free system, regulation of the iron regulatory element-binding protein has been recently suggested to involve direct interaction with hemin. The following observations on the synthesis of ferritin and of heme oxygenase (HO), the heme-degrading enzyme, in rat fibroblasts or hepatoma cells lead us to conclude that chelatable iron is a direct physiological regulator of ferritin synthesis in intact cells: (i) the inhibitor of heme degradation, tin mesoporphyrin IX, reduces the ability of exogenous hemin to induce ferritin synthesis but enhances HO synthesis; (ii) the iron chelator desferal suppresses the ability of hemin to induce synthesis of ferritin but not of HO; (iii) the heme synthesis inhibitor succinylacetone does not block iron induction of ferritin synthesis; (iv) there is no apparent relationship between the ability of various metalloporphyrins to inactivate the iron regulatory element-binding protein in cell-free extracts and their capacity to induce ferritin synthesis in intact cells; (v) administered inorganic iron significantly induces the synthesis of ferritin but not of HO; (vi) addition of delta-aminolevulinic acid to stimulate heme synthesis represses the ability of inorganic iron to induce ferritin synthesis while activating HO synthesis. Taken together, our results demonstrate that (i) release of iron by HO plays an essential role in the induction of ferritin synthesis by heme and (ii) chelatable iron can regulate ferritin synthesis independently of heme formation.

Iron regulation of ferritin gene expression.

Ferritin is a ubiquitous iron-storage protein found in the cells of animals, plants, molds, and bacteria which it protects from toxic intracellular levels of iron. Ferritin stores iron within a hollow protein shell formed by subunits of two types, H and L. The 5' untranslated regions of the two subunit mRNAs contain an almost identical 28-nucleotide sequence which regulates translation by binding to a specific cell sap protein. When cell iron level is low, this repressor protein obstructs translation of stored ferritin mRNAs, whereas increased iron levels release this protein, thus permitting extensive ferritin subunit synthesis to respond rapidly. Similar motifs in the 3' untranslated region of transferrin receptor mRNA interact with this protein to regulate breakdown of the mRNA and thus change the receptor population. Finally, transcription of the H and L genes can be independently increased by iron and other factors. In the case of iron, synthesis of the L-mRNA is increased preferentially since ferritin shells with a preponderance of L-subunits store iron more efficiently. Thus regulation of ferritin synthesis at the translational and transcriptional levels and by transferrin receptor mRNA abundance at the level of breakdown provide a coordinated mechanism for protecting cells against the effects of excess iron.

Translational control during the acute phase response. Ferritin synthesis in response to interleukin-1.

Interleukin-1 (IL-1 beta) increases the synthesis of both heavy and light (L)-ferritin subunits when added to human hepatoma cells (HepG2) grown in culture. RNase protection and Northern blot analysis with L-ferritin probes revealed that no changes in L-ferritin mRNA levels occur after cytokine stimulation. However, the induction coincides with an increased association of the L-subunit mRNA with polyribosomes. Since the recruitment of stored ferritin mRNA onto polyribosomes is seen when iron enters the cell, the effect of IL-1 beta on iron uptake was tested and was found to be unaffected by the lymphokine. Neither transferrin receptor mRNA levels nor the number of receptors displayed on the cell surface was affected by IL-1 beta. However, the action of the cytokine on ferritin translation is inhibited by the action of the intracellular iron chelator deferoxamine. These data indicate that IL-1 beta induces ferritin gene expression by translational control of its mRNA. The pathway of induction is different from iron-dependent ferritin gene expression whereas regulation requires the background presence of cellular iron.

Determinants of the interaction between the iron-responsive element-binding protein and its binding site in rat L-ferritin mRNA.

Ferritin messenger RNA has been shown to be translationally inactivated by the binding of a cytosolic protein to a 28-nucleotide iron-responsive element (IRE) located in the 5'-untranslated region of the mRNA. This interaction has been studied using quantitative receptor-ligand binding methods with gel retardation and nitrocellulose filter binding assays for the separation of bound complex from free RNA. In competition assays the entire 5'-untranslated region and the isolated IRE bound identically. The specificity of the RNA binding was studied using IRE variants. Two IREs from transferrin receptor mRNA and several variants with single base substitutions in the stem or loop had similar affinities. RNAs which could not form a stem-loop structure bound 1000-fold less well. These studies demonstrate the importance of the RNA conformation and the relative insensitivity of binding to much of the primary sequence. Saturation assays with increasing concentrations of 32P-IRE resulted in a binding hyperbola characteristic of mass action binding to a single class of sites with a KD = 0.09 nM. At 37 degrees C the dissociation rate is 0.04 min-1 (t 1/2 = 17 min). This rate is fast enough to account for the shift of ferritin RNA from the ribonucleoprotein pool to polysomes after rats are injected with iron. Determination of the concentration of the repressor requires accounting for three interconverting pools: free active repressor, mRNA-bound protein, and inactive (low affinity) repressor. Rat liver cytosol has a concentration of free active repressor of about 1 pmol/mg protein. Protein bound to endogenous mRNA can be measured by pretreatment with micrococcal nuclease or by separation with DEAE-Sepharose chromatography; it is present at a level similar to that of the free active protein. Inclusion of high levels of thiol reductants in the binding incubations reduces the inactive or low affinity repressor, forming unstably activated protein which has the same KD as the endogenous active protein; this inactive or low affinity protein is 2-4 times more abundant. A mechanism for iron regulation is proposed which accounts for the kinetics, the multiple protein pools, and the characteristics of the protein in these pools.

Effect of dietary iron overload on lipid peroxidation, prostaglandin synthesis and lymphocyte proliferation in young and old rats.

The effect of dietary iron overload on lipid peroxidation (LP), prostaglandin (PG) synthesis and lymphocyte proliferation was examined in young and old F344 rats. Rats 4 and 19-22 mo old were fed AIN-76 diet for 11-12 wk supplemented with 2.5% carbonyl iron obtained from two sources (Type A and B). Animals supplemented with Type A iron showed reduced food intake and weight gain associated with marked increases in extrahepatic and hepatic iron concentration. Rats receiving Type B iron had food intakes and body weights similar to those of controls but exhibited small increases in tissue iron concentration. Old control rats compared to young had significantly higher conjugated dienes (CD) in hepatic microsomes. Feeding Type A iron diets induced a significantly higher level of CD in hepatic microsomes from old rats compared to young rats. Iron overloaded rats also showed highly correlated (r = 0.94) increases in the urinary excretion of thiobarbituric acid-reactive substances and PG metabolites indicating increased in vivo LP and PG synthesis. Mitogen-stimulated PGE2 synthesis in young rats was increased at 4 wk in association with enhanced T-cell proliferation stimulated by Concanavalin A. Lymphocyte proliferation was significantly lower in old than in young control or iron-treated rats. The lack of efficacy of Type B vs. Type A iron appears due to a larger particle size and lower bioavailability. In conclusion, iron overloading increases in vivo LP and PG metabolism. Furthermore, the mitogenic response to Concanavalin A in young rats is enhanced after 4 wk of iron overloading.

Translational regulation of ferritin synthesis by iron.

This review starts with a description of certain features of mammalian ferritins and their DNA and RNA structures relevant to translational control of ferritin synthesis. Although the amino acid sequences of the two ferritin subunits (H and L) diverge in about 50% of the coding region, their five alpha-helices and the exon sizes of their genes are compatible with the proposition that they diverged from a single ancestral gene. Of particular note is their long 5'-untranslated regions (5'UTRs) which include a 28-nucleotide sequence almost completely identical in the H- and L-subunits of a range of species. This motif near the cap region of the 5'-UTR, which forms a specific stem-loop structure, provides for regulation of the translation of H- and L-ferritin mRNAs. When intracellular levels of chelatable iron are not in excess, a large reserve of H- and L-mRNAs is present in the cell sap, restrained from translation by a protein with an Mr of about 90-100,000 which binds to the stem-loop structure. When excess iron floods the cytosol, this protein/RNA complex appears to dissociate and the 40S ribosome subunit is now able to initiate ferritin protein synthesis so that the dormant mRNAs become active and are transferred to the polyribosomes. The mechanism whereby the binding protein is regulated in response to iron is currently under investigation. The regulatory protein occurs in the cell sap and is present in several interchangeable forms which appear to differ in the redox state of specific sulphydryls within the protein. Under some circumstances, the abundance of these forms appears to be altered by intracellular iron status. It is unclear how iron influences binding of the regulatory protein to ferritin mRNA. Some investigators consider that iron binds in the form of heme to the regulatory protein, for which they offer in vitro evidence. We have examined the role of heme versus inorganic chelatable iron in the regulation of ferritin and heme oxygenase synthesis in rat fibroblasts and hepatoma cells. By manipulating the flow of iron between the intracellular chelatable iron and heme iron pools we have concluded that chelatable iron can act as a regulator of ferritin synthesis in a manner which is independent of heme formation. This conclusion does not exclude a role for heme in some specialized cell types.

Vitamin C status and nutrient interactions in a healthy elderly population.

Vitamin C status and interactions with other nutrients were studied in 677 healthy, noninstitutionalized elderly people aged 60-98 y. Although 6% of the males and 3% of the females showed marginal vitamin C status (plasma ascorbic acid 11 to less than 23 mumol/L), only one person had a plasma ascorbic acid (AA) level less than 11 mumol/L. At all levels of total vitamin C intake, mean plasma AA levels were higher in females than males. Vitamin C supplement use was associated with generally higher blood levels of vitamins B-6, B-12, and E and folate in both sexes and with higher levels of retinol in females. However, after both age and the total dietary intake of the specific nutrient being examined were controlled for, plasma AA levels were significantly correlated only with plasma levels of vitamin E and folate in females.

Induction of ferritin subunit synthesis by iron is regulated at both the transcriptional and translational levels.

Synthesis of the iron storage protein ferritin is induced in rat liver by iron administration, the ferritin L subunit being preferentially stimulated over the H subunit. To examine the basis for this differential regulation of the two subunits, the transcription rates of the L and H genes, the total cellular levels of L and H subunit mRNA, and the distribution of the mRNAs between a stored ribonucleoprotein form and the polysomes were examined in rat liver at several times after iron injection. Iron caused a rapid increase in transcription of the L subunit gene, followed by a rise in L mRNA levels, whereas H subunit gene transcription and H mRNA levels did not increase significantly. Differential transcriptional regulation of ferritin L and H mRNA levels by iron contrasted with the coordinate control of the two subunit mRNAs at the translational level. On giving iron, there was a rapid and synchronous shift of both mRNAs from the ribonucleoprotein fraction onto the polysomes, the same proportion of each mRNA being mobilized. Thus, regulation of ferritin subunit synthesis at these two levels allows both a rapid translational response and a specific transcriptional response to increased intracellular iron levels.

Glycoconjugates as noninvasive probes of intrahepatic metabolism: III. Application to galactose assimilation by the intact rat.

A tracer methodology has been developed for noninvasive assessment of intrahepatic metabolism of administered labeled sugars. In this procedure, we measure the output of the label from the liver in two glycoconjugates derived from hepatic UDP-glucose, namely, glucuronic acid formed through UDP-glucuronic acid and excreted in the urine following acetaminophen administration, and galactose formed through UDP-galactose and then secreted in the carbohydrate portion of glycoproteins in the plasma. Comparison of the distribution of label from various sugar precursors in these end-products can indicate exchanges between hepatic UDP-glucose, UDP-galactose, and UDP-glucuronic acid. In this study we apply the technique to explore whether the enzyme UDP-galactose-4-epimerase catalyzing the step UDP-galactose to UDP-glucose is nonequilibrium and therefore potentially has a regulatory role for utilization of free galactose. The specific activity in the two glycoconjugates was compared when either [1-3H]galactose or [U-14C]glucose was the infused precursor sugar. In rats under a variety of conditions (fasting, oral refeeding, intravenous administration of galactose), label from [1-3H]galactose accumulated in glycoprotein-bound galactose much more than in acetaminophen-bound glucuronic acid, in comparison to label from [U-14C]glucose, demonstrating limitation of the rate of transfer from UDP-galactose to UDP-glucose at the epimerase step. Accordingly, epimerase is suggested to have a regulatory role in the galactose assimilation pathway.(ABSTRACT TRUNCATED AT 250 WORDS)

Cytoplasmic protein binds in vitro to a highly conserved sequence in the 5' untranslated region of ferritin heavy- and light-subunit mRNAs.

The mRNAs for the heavy and light subunits of the iron-storage protein ferritin occur in cells largely as inactive ribonucleoprotein particles, which are recruited for translation when iron enters the cell. Cytoplasmic extracts from rat tissues and hepatoma cells were shown by an electrophoretic separation procedure to form RNA-protein complexes involving a highly conserved sequence in the 5' untranslated region of both ferritin heavy- and light-subunit mRNAs. The pattern of complex formation was affected by pretreatment of rats or cells with iron. Crosslinking by UV irradiation showed that the complexes contained an 87-kDa protein interacting with the conserved sequence of the ferritin mRNA. We propose that intracellular iron levels regulate ferritin synthesis by causing changes in specific protein binding to the conserved sequence in the ferritin heavy- and light-subunit mRNAs.

Iron regulates ferritin mRNA translation through a segment of its 5' untranslated region.

In previous studies, we showed that acute administration of iron to intact rats or to rat hepatoma cells in culture induces synthesis of the iron-storage protein ferritin by activating translation of inactive cytoplasmic ferritin mRNAs for both the heavy (H) and the light (L) subunits. In the course of activation, these ferritin mRNAs are recruited onto polysomes. To elucidate the structural features of these mRNAs involved in the translational response to iron, a chimera was constructed from the 5' and 3' untranslated regions (UTRs) of ferritin L subunit mRNA fused to the reading frame of the mRNA of bacterial chloramphenicol acetyltransferase (CAT). This chimera and deletion constructs derived from it were introduced into a rat hepatoma cell line by retrovirus-mediated gene transfer. The complete chimera showed increased CAT activity in response to iron enrichment of the medium, whereas deletion of the first 67 nucleotides of the 5' UTR, which contain a highly conserved sequence, caused loss of regulation by iron. Whereas cis-acting sequences located in the 5' flanking regions of many genes have been repeatedly implicated in modulating their transcriptional expression, we report here a specific regulatory translational sequence found within the 5' UTR of a eukaryotic mRNA.

Conservation of ferritin heavy subunit gene structure: implications for the regulation of ferritin gene expression.

Ferritin stores iron within a protein shell consisting of 24 subunits of two types, heavy (H) and light (L). According to Southern blotting, the rat genome contains four copies homologous to the H-subunit cDNA (H cDNA). To determine whether only one of these is expressed, H cDNAs isolated from rat liver and heart mRNAs were compared and found to share identical nucleotide sequences. Next, genomic clones for three of the four rat H-subunit loci were isolated. Two were classical processed pseudogenes, whereas the third contained an expressed gene. RNase intron mapping of this expressed gene generated the same exon protection pattern when total RNA from rat liver or heart was used, indicating that this gene accounts for most or all of the H-subunit mRNAs (H mRNAs) in these tissues. Comparison of the expressed rat H-subunit gene (H gene) structure with published sequences for other species displays considerable conservation. The coding sequence of the rat H gene predicts 95% similarity to the human amino acid sequence, thus being more highly conserved than the L-subunit sequence of these species. Near the cap region of the 5' untranslated region, the rat H mRNA displays a 28-nucleotide sequence that is almost totally conserved in the corresponding region of the human, bullfrog, and chicken H mRNA and is also faithfully represented in the rat and human L-subunit mRNAs (L mRNAs), thus making this sequence a prime candidate for involvement in the known translational regulation of both subunits by iron. In the 5' flanking region, partially conserved sequences common to H gene and L-subunit gene (L gene) of the rat may be involved in transcriptional regulation by iron, whereas those conserved only in the H gene of man and the rat imply that other factors may independently control H-subunit regulation.

Protein nutriture of a group of free-living elderly.

The adequacy of the protein intakes of elderly people without overt debilitating diseases was investigated on 691 free-living men and women divided into those aged 60-75 y and those greater than 75 y. In both age groups men and women had average protein intakes of 1.02-1.06 g/kg body weight, values well above the safe level of 0.75 g/kg recommended in a WHO/FAO/UNU report. Although plasma concentrations of albumin, prealbumin, and transferrin declined with age, these were not related to low intakes of protein by individual elderly people. Similarly, upper-arm muscle mass was not less in those elderly people at the lower end of the range of protein intakes. Thus in this population of overtly healthy elderly men and women, there was no evidence of protein deficiency in contrast to other surveys where elderly people with chronic diseases were included.

Glycoconjugates as noninvasive probes of intrahepatic metabolism: I. Kinetics of label incorporation with evidence of a common precursor UDP-glucose pool for secreted glycoconjugates.

Conditions are described under which hepatic metabolism of administered labeled sugars can be monitored in the intact rat by measuring output of label from the liver in two secreted glycoconjugates derived from intracellular UDP-glucose, namely glucuronic acid linked to an administered drug (acetaminophen), and galactose secreted in glycoproteins. Rats with indwelling intrajugular catheters were given constant infusions of acetaminophen at nontoxic doses, and acetaminophen-glucuronide was isolated from rat urine by HPLC. When an unprimed continuous infusion of [U-14C]glucose or [1-3H]galactose is begun during acetaminophen infusion, labeling of urinary acetaminophen-glucuronide attains a plateau with an apparent half-life of 2.1 h from labeled glucose as the 14C-donor, and 1.2 h from labeled galactose. The times required to attain steady-state are attributable either to formation or to hepatic secretion of the glucuronyl conjugate rather than to mixing in extracellular pools or to delays in renal excretion since the half-time of urinary excretion of intravenously injected acetaminophen-glucuronide is only 21 minutes. Consequently, the excreted glucuronide provides noninvasive evidence of the recent specific activity in the precursor pool of hepatic UDP-glucose feeding glucuronic acid. The second glycoconjugate used for comparison was galactose, isolated from the acute-phase plasma protein alpha 1-acid glycoprotein. This represents hepatic UDP-galactose labeling. Comparison of the labeling patterns of secreted galactose and glucuronic acid demonstrates that they are derived from the same intracellular pool of UDP-glucose. This is in contrast to the patterns of labeling of hepatic glycogen from (1-3H) and [U-14C]glucose which have been shown to diverge from those of secreted glucuronic acid.