Transferrin Receptor 1: A Ferritin Receptor as Well?

Abstract

Li L, Fang CJ, Ryan JC, et al. (Department of Medicine, Veterans Administration Medical Center, San Francisco, CA; others). Binding and uptake of H-ferritin are mediated by human transferrin receptor-1. Proc Natl Acad Sci U S A 2010;107:3505–3510. The essential element iron is required for a number of cellular and metabolic functions besides its role as an oxygen carrier in hemoglobin. An excess or deficiency of iron, however, leads to a number of clinical complications, including iron overload or anemia, respectively. The genetic iron overload disorder hereditary hemochromatosis is characterized by excessive iron absorption and accumulation of iron in various parenchymal tissues and organs. In the body, absorbed iron is circulated bound to transferrin (Tf), which is internalized by the cell-surface receptor Tf receptor 1 (TfR1). This uptake of iron through the Tf–TfR1 pathway is the predominant route for iron entry into cells. The TfR1 gene was cloned and sequenced in 1984 (Nature 1984;311:675–678) and is among the most well-studied molecules in cell biology. The hemochromatosis protein HFE can bind to TfR1 and competes with Tf for binding (J Mol Biol 1999;294:239–245). It was recently shown that this interaction is important for the regulation of iron homeostasis (Cell Metabolism 2008;7:205–214). Ferritin is a multisubunit, spherical protein that stores iron within cells. It is composed of 24 subunits and comprises heavy-chain ferritin (HFt) and light-chain ferritin (LFt). The proportions of HFt and LFt vary across tissues and have different functions, with HFt having ferroxidase activity and being responsible for iron uptake into ferritin and LFt being involved in storage of iron within the protein as ferrihydrite. Ferritin is also present in serum, but its role here is less clear. Under normal conditions, the levels of circulating ferritin reflect iron stores and this feature has been utilized to assess iron status in individuals. However, ferritin is also an acute phase reactant, and other factors including inflammation can affect the levels of circulating ferritin. Circulating ferritin has itself been shown to act as a proinflammatory cytokine (Hepatology 2009;49:887–900). In humans, serum ferritin consists mostly of LFt, and there is regulated secretion of a glycosylated form from hepatocytes (Blood 2004;103:2369–2376). It is widely accepted that ferritin receptors exist on many cell types and that uptake of ferritin is another route for delivery of iron into cells. Until recently, the identity of these receptors remained obscure. Tim-2 was identified as a receptor for HFt on mouse B-cells, responsible for uptake of HFt into endosomes (J Exp Med 2005;202:955–965). Recently, Scara5 was identified as a receptor for LFt in mice, responsible for the uptake of ferritin-bound iron into cells (Dev Cell 2009;16:35–46). This study by Li et al (Proc Natl Acad Sci U S A 2010;107:3505–3510) used expression cloning to identify receptors for HFt in human cells. They identified the ubiquitous TfR1 as a receptor for HFt, which is capable of internalizing the ligand and responsible for its uptake into endosomes and lysosomes. The authors go on to demonstrate that this interaction is partially inhibited by diferric-Tf (Fe-Tf), but independent of HFE, and that TfR1 accounts for most of the HFt binding in all the cells studied. The investigators screened several human cell lines for the ability to bind human HFt in a saturable manner and positively identified the human B cell line 721.221; the mouse pro–B-cell line Ba/F3 on the other hand was selected as a cell line not demonstrating saturable binding of HFt. A retroviral cDNA library of the 721.221 cells was then used to transfect Ba/F3 cells, which were subsequently screened by cell sorting for binding to HFt. Over 100 clones were selected and 8 of these were sequenced. All proved to code for TfR1. That the binding was due to human TfR1 was shown in 1 of the cloned Ba/F3 cell lines as well as in cells transfected with the human TfR1 (hTfR1) plasmid. Saturable binding of HFt was demonstrated in cells expressing hTfR1 and could be inhibited by unconjugated HFt or by a TfR1-blocking antibody. The authors then used a TfR1-deficient Chinese Hamster Ovary cell line (TRVb) to demonstrate that stable expression of hTfR1 in these cells (TRVb-1 cells) was necessary and sufficient to bind and internalize HFt. In vitro binding studies using recombinant proteins also showed that soluble HFE does not inhibit significant HFt binding by soluble TfR1, and that TfR1 does not bind LFt. The latter was confirmed in intact MOLT-4 T cells. Surprisingly, in these cells HFt in molar excess only reduced Fe–Tf binding by 50%, whereas Fe–Tf in molar excess reduced HFt binding by 70%. Using internalization assays, confocal microscopy, and colocalization with a lysosomal marker, the investigators show that internalized HFt is endocytosed into endosomal structures and eventually into lysosomes. Examination of HFt binding in a number of human cell lines of various tissue origins, activated lymphocytes and reticulocytes, and blocking of the binding with a specific anti-TfR1 antibody showed that virtually all of the binding of HFt in these cells can be ascribed to TfR1. TfR1 is a receptor known primarily for the provision of iron to cells through the binding and internalization of iron-bound Tf. Several earlier studies have implicated TfR1 as a receptor for New World hemorrhagic fever arenaviruses (Nature 2007;446:92–96). This study by Li et al is the first to implicate TfR1 as a receptor for HFt in human cells. Interestingly, this does not seem to apply to mouse TfR1, which does not seem to bind mouse HFt despite a sequence conservation of 86% between species. The mutual and possibly combined binding of 2 iron carrier molecules, HFt and Tf, by 1 receptor suggests that iron intake occurs via a coordinated pathway. Ferritin is composed of a shell of 24 subunits of HFt and LFt chains with each molecule able to store and transport 4500 atoms of iron. Ferritin found in serum contains little if any iron and is thought to be a glycosylated form of LFt. Ferritin is regulated by iron levels both at the transcriptional level and post-transcriptionally through stabilization of mRNA via iron-responsive elements. Ferritin molecules consisting of HFt only or LFt only do not occur in vivo. Unlike LFt, which is principally involved in iron storage (J Gastroenterol Hepatol 1993;8:21–27), HFt has been shown to have a role as a proinflammatory cytokine (Hepatology 2009;49:887–900) and in immune function (J Autoimmun 2008;30:84–89). This study, a search for receptors of human HFt, is a follow-up of studies that identified Scara5 as a receptor for LFt in mice (Dev Cell 2009;16:35–46) and Tim-2 as a receptor for HFt also in mice (J Exp Med 2005;202:955–965). This is an interesting study that would have benefitted from the provision of details of the human and mouse cells that were screened for saturable binding of HFt. TfR1 is the receptor and mediator of endocytosis for Fe–Tf, a process that is competitively regulated by the binding of HFE. HFE binds TfR1 at the cell surface and competes with Fe–Tf for binding to TfR1. The capacity of HFE to compete with HFt binding to TfR1 was investigated and reported to have little to no competitive effect, suggesting that HFE is not likely to play a significant role in the regulation of HFt uptake by cells. However, the methods used to assess the effects of HFE and Fe–Tf on the binding of HFt to TfR1 were different. The effect of HFE was examined using soluble proteins and pull-down assays, whereas competition between Fe–Tf and HFt was assessed using labeled proteins on cells. Giannetti et al have previously demonstrated that HFE competition with Fe–Tf for binding to TfR1 is poor in solution but significant when membrane bound (J Biol Chem 2004;279:25866–25875). As such, an effect of HFE on HFt binding and subsequent endocytosis by TfR1 when membrane bound cannot be entirely discounted. It is of interest that murine TfR1 did not bind either human HFt or murine HFt, given the conserved nature of these proteins and of other iron regulatory pathways. It is possible that Tim-2, a recently identified murine HFt binding protein, may compensate for this (J Exp Med 2005;202:955–965). Given the suggested difference in the receptor responsible for HFt binding, it is surprising that although the screening was reportedly performed in human cell lines, the in vitro specificity and competitive assay experiments were performed in non-human cell lines. It is not reported if either the mouse cell line Ba/F3 or Chinese Hamster Ovary-TRVb cells express Tim-2. The ratio of HFt to LFt within ferritin varies between tissues and in response to inflammation (Blood Rev 2009;23:95–104). The identification of a human receptor for HFt now offers the potential to investigate the implications of this regulation. A comparison of the effects of varying ratios of HFt and LFt chains and iron-loaded ferritin on binding to TfR1 and subsequent internalization would be interesting. This may provide insight into the physiologic relevance of TfR1 binding to HFt and a potential regulated mechanism for the uptake of ferritin-bound iron into cells and maintenance of serum ferritin levels. In conclusion, this paper has provided novel insights into 2 of the most recognized and well-studied players in iron biology, highlighting the fact that there is more to learn about even the most well-established proteins.