Identification of Ferritin Receptors: Their Role in Iron Homeostasis, Hepatic Injury, and Inflammation

Article Outline

• Comment
• Copyright

Li JY, Paragas N, Ned RM, et al. (Renal Division, College of Physicians and Surgeons, Columbia University, New York, New York). Scara5 is a ferritin receptor mediating non-transferrin iron delivery. Developmental Cell 2009;16:35–46.

Iron is an essential requirement for normal cellular physiology. However, excessive intestinal absorption of iron, as occurs in hereditary hemochromatosis, leads to markedly increased deposition of iron in parenchymal cells of various organs such as the heart, pancreas, parathyroid, pituitary glands, and especially the liver. This can result in cellular toxicity and tissue injury, and pathologies such as arthritis, diabetes, cardiomyopathy, bronzing of the skin, hepatic cirrhosis, and hepatocellular carcinoma (Annu Rev Med 1999;50:87–98). Cellular injury owing to excess iron is thought to be induced by the generation of oxyradicals and peroxidation of lipid membranes, which in the liver leads to damage of hepatocellular organelles such as lysosomes and mitochondria, contributing to hepatocyte necrosis and apoptosis, and ultimately to the development of hepatic fibrosis (Semin Liver Dis 2005;25:433–449).

Iron required for normal cellular function, enzymatic reactions, and heme biosynthesis is sourced from circulating holo-transferrin bound to cell surface transferrin receptors, which are internalized through endocytosis. The acidic milieu of the endosome causes the liberation of iron from transferrin, which then recycles to the cell surface as apotransferrin. The free iron is transported from the endosome into the cytosol via divalent metal transporter-1 (J Biol Chem 2000;275:22220–22228) where it is taken up by ferritin to be stored in a non-toxic but biologically available form. However, there is evidence that mammalian cells can also acquire iron via transferrin-independent pathways. Despite severe anemia, normal organogenesis occurs in both hypotransferrinemic mice (Blood 2000;96:1113–1118) and humans (Am J Hum Genet 1993;53:201–213). In addition, although specific developmental defects appear in the spinal cord, and lymphoid and erythroid tissues of transferrin receptor-1 (TfR1)-deficient (TfR1^−/−) embryos, most other organs seem to initiate development despite the phenotype being embryonic lethal (Nat Genet 1999;21:396–399).

In this study by Li et al (Dev Cell 2009;16:35–46), the authors have developed an in vivo assay to assess the mechanism of iron delivery in kidney organogenesis. They identified a novel receptor, scavenger receptor, member 5 (Scara5), as a receptor for L-ferritin which mediates the delivery of non–transferrin-bound iron (ie, iron bound to ferritin) to the developing kidney. The authors demonstrate that cell type-specific mechanisms of iron trafficking exist during organogenesis that utilize either transferrin or ferritin. Although this is a study of kidney organogenesis, the research has important implications for the regulation of iron homeostasis in liver embryology and development, in normal iron metabolism, and equally in the pathophysiology of the iron overload disease hereditary hemochromatosis.

The investigators generated murine chimeras composed of fluorescently tagged cells deficient in TfR1 and untagged wild-type TfR1^+/+ cells to identify the iron transport systems present in kidney development. They observed that in embryonic development TfR1^−/− cells populated many organs including kidney, liver, lung, midgut, bladder and gonads, and that within these organs, TfR1^−/− cells produced both mesenchymal and epithelial cell precursors. These data suggest that TfR1 expression is not an absolute requirement for the specific cell lineages which are essential for developmental processes involved in organogenesis. Their analysis of kidney organogenesis was aided by the limited expression of the second transferrin receptor (TfR2; a liver-specific molecule central to the regulation of iron homeostasis in hemochromatosis), which could potentially have complicated their investigation on pathways of iron trafficking. Their results showed that although the uteric buds of the kidney demonstrated scant evidence of TfR1^−/− cells, other renal cells in the mesenchyme, capsule, and stroma were significantly populated by TfR1^−/− cells, suggesting that iron delivery required for normal cellular development and tissue organogenesis must have occurred via a non–transferrin-dependent pathway. The authors of this study went on to show that rather than transferrin, TfR1^−/− renal capsule cells internalized ferritin, whereas TfR1^+/+ cells did not. To examine the possible role of ferritin, they isolated fluorescently labeled TfR1^−/− cells from kidney stroma/capsule by fluorescence-activated cell sorter and profiled their gene expression. They demonstrated that 156 genes were consistently up-regulated >2-fold in TfR1^−/− cells versus wild-type cells; however, only 1 gene was significantly elevated (ie, 7-fold in TfR1^−/− vs TfR1^+/+); this gene was Scara5. In situ hybridization was then used to demonstrate that Scara5 was localized to cells where ferritin was captured in both embryonic and adult kidney. Scara5 was also present in the lung, aorta, muscle, and gonadal epithelia.

Scara5 is a member of the scavenger receptor family—proteins that are expressed on the cell surface and are capable of binding a number of different ligands. This is the first study to document the binding of iron transport proteins by scavenger receptors. Transfection of Scara5 into Scara5^−/−/TfR-1^−/− cells, showed that the receptor was able to bind and subsequently internalize ferritin from the cell surface, providing another means of delivering iron to the cell. Finally, L-ferritin, the principal component of serum ferritin and liver-derived ferritin, was preferentially bound by Scara5, whereas H-ferritin was not. This study is the first to identify a potential L-ferritin specific receptor and has clearly shown that cellular iron delivery can occur via 2 independent endocytotic pathways.

Back to Article Outline

Comment 

Ferritin is an extremely important iron-binding molecule that stores iron in a biologically available form for vital cellular processes while protecting proteins, lipids, and DNA from the potential toxicity of this element (Blood Rev 2009;23:95–104). Ferritin is present in the circulation either via secretion, that is, serum ferritin, or through release from damaged cells as tissue-derived ferritin. Variations in serum ferritin levels are commonly observed in clinical practice, such as in hemochromatosis, where alterations in circulating ferritin accurately reflect changes in body iron status. Circulating ferritin is also markedly increased in a multitude of other conditions including malignancies, and in inflammatory and neurodegenerative diseases (Blood Rev 2009;23:95–104). In the liver, ferritin is elevated in conditions of hepatic inflammation including alcoholic liver disease, nonalcoholic steatohepatitis, conditions of hepatic necrosis, and hepatocellular carcinoma. Although it is generally believed that circulating ferritin levels simply reflect an acute phase response, the mechanism of ferritin release or secretion, and indeed the precise physiological function of ferritin in these conditions are not fully understood.

The ferritin molecule is comprised of an apoprotein shell of 24 subunits of H-chain (acidic), and/or L-chain (basic) ferritin, with the capacity to store (or transport) up to 4,500 atoms of iron per molecule. The ratio of these H- and L-ferritin subunit chains in the ferritin holo-molecule varies depending on the tissue where ferritin is found; for instance, H-subunit–rich tissue ferritins are found predominantly in the heart and kidney, whereas L-subunit–rich tissue ferritins are mostly evident in liver and spleen (Blood Rev 2009;23:95–104). Serum ferritin (which contains little if any iron) is composed almost entirely of L-ferritin subunits. Ferritin expression is regulated both at the mRNA level and post-transcriptionally after chronic iron exposure, with L-ferritin more tightly regulated by iron (J Gastroenterol Hepatol 1993;8:21–27). Whereas L-ferritin is principally involved with iron storage, H-ferritin has been shown to have potential immunomodulatory activity (reviewed in J Autoimmun 2008;30:84–89). A role for H-ferritin as a proinflammatory cytokine in hepatic stellate cell activation has recently been demonstrated (Hepatology 2009;49:887–900). In this study, H-ferritin induced an iron-independent signaling pathway resulting in up-regulation of nuclear factor κB-transcriptionally regulated genes involved in hepatic fibrogenesis, including interleukin-1β and RANTES, via an as yet unidentified receptor (Hepatology 2009;49:887–900).

In hepatic injury, the ferritin released by damaged hepatocytes or Kupffer cells has the potential to markedly elevate the local ferritin concentration and thus may have a direct paracrine effect on nearby liver cells. Ferritin uptake via endocytosis has previously been documented in a number of different cell types including hepatocytes (Hepatology 1988;8:719–721), hepatic stellate cells (J Clin Invest 1994;94:9–15), and lymphoid (J Inorg Biochem 1992;47:219–227), and erythroid precursors (Blood 1999;94:3205–3211), although the identity of the specific receptor responsible has remained elusive. In 2005, Chen et al, identified T-cell immunoglobulin domain and mucin domain 2 (Tim-2), as a receptor for H-ferritin endocytosis in B cells (J Exp Med 2005;202:955–965); more recently, others have shown Tim-2 is the primary mechanism for iron acquisition by oligodendrocytes (J Neurochem 2008;107:1495–1505). This study by Li et al is the first to demonstrate an L-ferritin receptor that has the potential to internalize significant quantities of L-ferritin–bound iron. Although not investigated, the study has important implications for trafficking and delivery of iron to the developing liver and, importantly, for maintaining normal iron homeostasis in adult liver.

Unfortunately, neither study identifying Tim-2 or Scara5 assessed ferritin binding affinities, but showed ferritin binding could be competitively displaced; thus, the possibility exists that these molecules may simply be ferritin scavenger receptors. Previous studies have identified very high affinity binding for ferritin in human, rat, and pig liver (Hepatology 1988;8:719–721; J Inorg Biochem 1992;47:219–227) with association constants (K_a) ranging between 10⁻⁸ and 10⁻⁹ mol/l, although these analyses were conducted on whole liver homogenates (thus mixed hepatic cell populations), with ferritin binding not specific for H- or L-ferritin. Studies on highly purified cultures of rat hepatic stellate cells demonstrate the existence of a highly specific, very high affinity (K_a = 5 × 10⁻¹⁰ mol/l) H-ferritin–dependent binding site (J Clin Invest 1994;94:9–15), which may be responsible for transducing the proinflammatory effects of ferritin on nuclear factor κB-regulated fibrogenesis genes in these cells (Hepatology 2009;49:887–900). The identity of this hepatic stellate cell ferritin receptor remains to be elucidated. It is apparent that additional ferritin receptors may exist and have specific roles on different cell populations. Neither the Tim-2 study (J Exp Med 2005;202:955–965) nor the Scara5 paper by Li et al assessed the potential signaling pathways that may be induced by ferritin binding. A comprehensive analysis of the role of ferritin as a signaling molecule via Tim-2, Scara5, or via as yet unidentified receptors, will be of great interest and may lead to a better understanding of the precise role of circulating ferritin in inflammation associated with chronic liver disease.