How does iron enter enterocytes? Ablating SLC11A2, the gene for the divalent metal ion transporter DMT1, supports evidence from the Belgrade rat and mk mouse models establishing DMT1 as the primary mechanism serving apical uptake of nonheme iron. DMT1 harnesses the energy from the proton electrochemical potential gradient to drive active transport of Fe2+ (and perhaps Mn2+ and other metal ions) into enterocytes. Fe(III) must first be reduced by ascorbic acid and surface ferrireductases. Among these is duodenal cytochrome B (DcytB), but lack of an obvious phenotype in DcytB (Cybrd1) knockout mice suggests ferrireductase redundancy. Our understanding of heme absorption has lagged, but the time is ripe for gains.
- duodenal cytochrome B
- divalent metal ion transporter
- iron absorption
iron deficiency is the most prevalent micronutrient deficiency worldwide. Meanwhile, iron overload associated with conditions such as hereditary hemochromatosis, thalassemia, or sickle cell disease poses a serious threat to many other individuals. The past decade has witnessed tremendous progress in our understanding of the intestinal absorption of iron and its contribution to iron homeostasis. Second in the series “Iron Imports,” this article reviews the topic of iron uptake at the intestinal brush border. Although dietary iron in omnivores comprises both heme iron and nonheme iron, we focus here on the latter form because the major recent advances in the field have been in the molecular identification and characterization of the machinery responsible for absorption of nonheme iron. Apical uptake of iron is mediated by the divalent metal ion transporter DMT1. We describe its functional properties and review lessons from DMT1 mutant animal models and the first disease-causing DMT1 mutation to be identified in the human population.
The literature in this field contains hypotheses or issues that some researchers may avoid and others that are clearly controversial. Scientific rigor demands that investigators not ignore inconvenient observations or hypotheses; rather, we should develop tests that distinguish apparent alternatives (and sometimes both alternatives will survive or be found wanting) or look for technical difficulties that may explain why an observation is misleading. This viewpoint has pushed us to discuss a number of avoided or controversial claims.
PRESENTATION OF IRON AT THE BRUSH-BORDER MEMBRANE
Iron absorption occurs mainly in the duodenum and proximal jejunum, but iron uptake by enterocytes depends also on several upstream factors. These include the chemical form of dietary iron, organic acids, and gastric acid secretion. For those who consume meat, heme is a major dietary source of bioavailable iron, so its absorption is discussed later. Those who do not consume meat rely on nonheme dietary iron. This form is often loosely referred to as “inorganic” iron even though most nonheme Fe(II) or Fe(III) in food is complexed with organic acids (e.g., citrate) or peptides (e.g., ferritin and albumin) and is not limited to Fe(II) salts. Bioavailability of nonheme iron may be determined in large part by the solubility of such complexes and the affinity with which they bind iron. The part played by gastric acid secretion in iron absorption is almost certainly to promote solubility of iron complexes. Phytic acid (abundant in cereals) is a major inhibitor of iron absorption because of the poor solubility of its iron chelate at any pH. Iron fortification of foods is likely to be most successful with highly soluble (low molecular weight), high-affinity complexes such as Na-Fe-EDTA (the stability of which makes it suitable for long shelf-life foods).
Vitamin C (l-ascorbic acid) is important in iron absorption, as established by the early work of Conrad and others. Ascorbic acid, either derived from the diet or from gastric or biliary secretions, can efficiently reduce ferric ion in a low-pH environment. Even at higher pH, ascorbate forms soluble Fe(II) or Fe(III) complexes that promote iron absorption. Iron deficiency leads to an induction of the gene encoding the intestinal vitamin C transporter SVCT1 in the rat (2) and to increased enterocyte ascorbate levels in the mouse. One should not surmise that these changes represent vitamin C repletion in these animals because rats and mice (unlike humans) can synthesize this vitamin in the liver. More likely, the significance of these observations is that increased ascorbate in enterocytes could serve as an electron donor to duodenal cytochrome B (DcytB), a candidate ferrireductase at this apical membrane (18).
APICAL MEMBRANE FERRIREDUCTASES
Ferric iron not reduced by ascorbate before reaching the intestinal brush border must be reduced by surface ferrireductases to be transported via DMT1 into the enterocyte. DcytB, the first candidate for this role (Fig. 1), was isolated using a subtractive cloning strategy from the iron-deficient rat (18). So it is not surprising that DcytB, which is localized alongside DMT1 on the apical membrane, is upregulated under conditions of iron deprivation (2); however, targeted disruption of the Cybrd1 gene coding for the murine homolog of DcytB suggests that DcytB is not necessary for intestinal iron absorption in mice fed a normal-Fe diet (13). The study just cited had difficulty detecting a statistically significant effect of the Cybrd1 knockout on liver nonheme iron in mice fed a low-iron diet but did find such a decrease in 12-wk-old female mice fed an iron-deficient diet for 8 wk. This difficulty in detecting a phenotype led the authors to focus on DcytB not being necessary for dietary iron absorption in mice. We remind the reader that humans are a scorbutic (or vitamin C challenged) species that may rely more on surface ferrireductases than mice do. Crossing the Cybrd1 knockout mouse with one unable to synthesize vitamin C or one with a severe genetic iron deficiency may allow us to speculate further as to the role of DcytB in mice and humans.
The lack of a severe iron-deficient phenotype in the Cybrd1/DcytB knockout may suggest that DcytB is only one of several intestinal ferrireductases. Ferrireductase activity is found in many tissues and critical for iron utilization in the transferrin cycle, yet investigators have not been able to show that DcytB participates in the transferrin cycle. Now the absence of an erythroid phenotype when Cybrd1 is ablated encourages us to predict that additional ferrireductase gene(s) is (are) forthcoming.
DIVALENT METAL ION TRANSPORTER DMT1
DMT1 is a ferrous ion (Fe2+) transporter (9, 12, 15) that is energized by the H+ electrochemical potential gradient (Fig. 1). Ferric ion (Fe3+) is excluded (12). Its primary location in the gut is on the brush-border membrane of mature villous enterocytes of the proximal duodenum (see Ref. 15 for citations; most recent data in Ref. 14), where the expression of DMT1 is tightly regulated by body iron status. This same transporter is also responsible for the recovery of iron from recycling endosomes during transferrin receptor (TfR)-associated cellular uptake in erythroid (red blood cell) precursor cells and many other cell types.
There are at least five mutants with defective DMT1 expression; these comprise four animal models and the first disease-causing mutation to be identified in the human population. Studies of the intestinal (and broader) phenotypes of these mutants teach us much about the physiological role of DMT1 in apical iron uptake. The Belgrade (b) rat (9) and microcytic (mk) mouse have been studied for many years and provided the first insights into DMT1 function in the physiological setting. Both mutants have a Gly185→Arg substitution (this is the position in the NH2-terminal isoform initiated from exon 2 as “exon 1B” is untranslated; for the 1A exon variant, it corresponds to Gly216 in the rat and Gly215 in the mouse) (6, 7). Thus one expects extensive similarities between the two models, with differences reflecting the different species in which DMT1 functions or malfunctions. Here, we cover first the similarities as well as the lessons from the other mutants; next, we comment on differences.
The b and mk animals exhibit a severe hypochromic, microcytic anemia with diminished erythrocyte survival. The resemblance to iron deficiency teaches us that DMT1, found largely on the apical surface of enterocytes, is not allowing iron to enter. Although bypassing intestinal uptake by parenteral injection or careful iron supplementation improves red cell production, it does not eliminate the iron-deficient appearance. This part of the phenotype places emphasis on a role of DMT1 during erythroid iron acquisition. Exactly how critical DMT1 is depends on the effect of the mutation. Therein lies controversy (see Ref. 9): some reports indicate that considerable transport activity survives, whereas others show that activity is very low or undetectable. Similarly, some groups (often overlapping with those that find considerable transport remaining) observe that DMT1 localizes differently and turns over more readily, whereas others see much less of such a phenomenon. The latest report (14) indicates that transport activity is undetectable in brush-border membrane vesicles from homozygous Belgrade rats and that DMT1 is increased and, to a large degree, properly localized in the same preparations. Because the observations supporting poor localization and higher turnover are with DMT1 in, and DMT1 constructs from, the microcytic mouse, most recently in Ref. 23, we speculate that some of the discrepancy in these results represents a species-related difference. If so, exchange of constructs, cell lines, and specific antibodies could resolve the issues. The end effect, regardless, is that little functional DMT1 actually takes up iron at the apical surface of mutant enterocytes. From this body of work, we learn that DMT1 is the major intestinal nonheme iron transporter (Fig. 1), and we retain a suspicion that one or more alternative pathways exist that (barely) permit the rodent mutants to survive.
A very similar phenotype results from a nonsense mutation in DMT1 in a zebrafish mutant called chardonnay (3). That chardonnay should have no DMT1 transport activity indicates that intestinal DMT1 is critical for the iron demand of the erythron but that alternative (as yet poorly defined) pathways can handle demands of iron homeostasis in most other tissues.
In the first example of a disease-causing mutation in human DMT1, a homozygous G→C substitution at nucleotide 1285 (or 1372 in the 1A isoform) led to the amino acid substitution E399D (E428D for 1A), but this change has little effect on DMT1 function per se (20). Similar to the prior mutations, however, the patient's erythroid phenotype was apparently iron deficient, but, even more remarkably, the patient exhibited signs of iron overload. The mutation also affected a splice donor site and provoked an increase in aberrant splicing in hematopoietic tissue involving skipping of exon 12. The predicted mRNA from this event produces a nonfunctional protein (20) so that the patient's erythroid cells should have much less DMT1 than do normal cells. One might explain the phenotype by assuming that the gut iron uptake regulatory system has sensed the iron-deficient erythron and upregulated DMT1 (and other parts of the transport pathway) so that increased transserosal iron diverts from its usual major target of red blood cell precursors to other tissues. Other possibilities deserve consideration, e.g., alterations in red blood cell turnover may also contribute to the patient's unusual phenotype.
Gunshin et al. (11) ablated the SLC11A2 gene coding for DMT1 in mice and made a series of mouse strains in which the gene was specifically removed in selected tissues. Their technically elegant work confirms and further solidifies the implications researchers derived from the earlier DMT1 deficiencies. The sites where DMT1 is most critical for iron homeostasis are the intestinal tract (where it is clearly the major transporter for iron entry) and the red blood cell precursor (where it is essential for full hemoglobinization). These investigators had difficulty raising the knockout mice past weaning and found that specific DMT1 gene disruption in the intestinal tract led to an anemia nearly as severe. When they made a double mutant by crossing the DMT1 knockout with mice in which the HFE gene (known to modulate DMT1 expression) was also ablated, they observed mildly improved survival. Although SLC11A2 is clearly a target for regulation by the HFE pathway, an increase in its activity cannot account for this improvement. Hence, one must infer that, although DMT1 is normally essential for iron uptake, HFE must ultimately regulate other parts of the iron transport pathway, possibly including an alternative means of apical iron uptake that can be sufficiently upregulated to allow the observed improvement. Whereas the existence of alternatives for uptake had been raised often before, this conclusion was debated in part because of controversy about how much transport activity survived the G185R mutations in the Belgrade rat and microcytic mouse (above). Although they reiterated the importance of DMT1, Gunshin et al. have fueled suspicions that alternatives exist. What remains is to identify the alternative(s), no easy task!
DMT1 AND THE ROLE OF PROTONS
The mucosal cell surface in the duodenum and proximal jejunum remains moderately acidic despite the progressive alkalinization of the luminal contents (17). This “acid microclimate” (which results from the activity of the Na+/H+ exchanger) is thought to provide the proton electrochemical potential gradient as the motive force for iron uptake (Fig. 1). Thus it was comforting when the first demonstration of the iron transport activity of DMT1 revealed an H+/Fe2+ cotransport mechanism that was stimulated at low pH (12).
Since then, at least one report has lessened the clarity of this coupling. Worthington et al. (25) found that Fe2+ transport was maximal at pH 6.75 in Caco-2 cells (as model enterocytes in which DMT1 is endogenously expressed) and in DMT1-transfected COS-7 cells. So, as these authors suggested, could the Fe2+-evoked current in the oocyte (12) reflect only a H+ current that was not coupled to Fe2+ transport? Countering this view, a recent study (16) in oocytes expressing DMT1 shows that the pH profile of 55Fe2+ uptake is no different from that of the Fe2+-evoked currents, and that both activities are maximally stimulated at low pH (at least to pH 5.2 without overstepping a pH optimum). Kinetic analyses showed that H+-Fe2+ cotransport is the favored activity, but DMT1 also mediated uncoupled Fe2+ or H+ fluxes in the absence of the cosubstrate (16). Uncoupled leaks of the driving ion (e.g., Na+ and H+), but not for the driven substrate, are commonplace among mammalian cotransporters. Fe2+ transport uncoupled from H+ is unlikely to be of much consequence at the apical membrane of the intestine (perhaps it is of consequence elsewhere) (16), but it does explain the residual Fe2+ transport activity at pH ≥ 7 observed by other groups reporting that DMT1-specific Fe2+ transport in Caco-2 cells is maximally stimulated at much lower pH (see Ref. 16 for citations). Why Worthington et al. (25), using the same cell system, obtained quite different results is not known, but possible reasons are suggested elsewhere (16). Results (M. D. Garrick, unpublished data) with an inducible DMT1 expression system confirm that transport occurs at pH 7.4 but still find a maximum around pH 5.5. (A much lower pH may denature the protein and inactivate the transport activity of DMT1.)
H+ coupling in DMT1 provides a thermodynamic driving force for Fe2+ transport, with half-maximal H+ concentration of ≈1 μM (i.e., pH 6.0) (16). Whereas duodenal mucosal cell surface pH is 6.7, the microclimate pH reaches 6.0 in human proximal jejunum in vivo (17). H+ binding also increases the affinity with which DMT1 binds Fe2+ (16). So, although duodenal Fe2+ concentrations may be higher than those of the jejunum, the properties of DMT1 ought to make it an effective scavenger of the available Fe2+ in the proximal jejunum. Inward currents in DMT1-expressing cells exceed those expected for strict stoichiometric H+/Fe2+ cotransport (19). Such slippage in cotransport systems is normally thought of only in terms of the energetic penalty incurred, but the excess protons transported by DMT1 may serve additional roles. These may include localized intracellular acidification, maintaining the chemical state of Fe2+ before it is handed off to an intracellular chaperone. Nelson proposes that variable H+/Fe2+ coupling can help manage varying levels of metal ions in the lumen of the gut by affecting the internal pH of the enterocyte and possibly the nutritional needs for the metal ions. Support for this hypothesis comes from the observation that an F227I mutation can decrease the H+-to-Fe2+ coupling ratio (19), suggesting that evolution can potentially select for optimal slippage. Nevertheless, more analysis is required to assess fully the roles of H+, pH, and slippage.
METAL ION SUBSTRATE PROFILE OF DMT1
Gunshin et al. (12) demonstrated that, in addition to Fe2+, a broad range of transition metals (Cd, Zn, Mn, Cu, Co, Ni, and Pb) can evoke inward currents in Xenopus oocytes expressing DMT1. Fe2+ transport was demonstrated in this study using a radiotracer assay. Nevertheless, like inhibition of uptake, evoked currents demonstrate DMT1 reactivity with these metals but do not demonstrate for each that it is actually transported. In fact, questions have been raised about most of the potential substrates except iron (10).
Radiotracer assays (in oocytes or in Caco-2 cells) have established that DMT1 is also capable of transporting Mn, Co, Zn, and Cd (16) (for earlier citations, see Refs. 10 and 15), but these observations require careful interpretation on more than one level. For example, DMT1 shows a strong preference for Fe2+ over Zn2+ (16), and Zn2+ is a very weak inhibitor, if at all, of Fe2+ uptake (10, 16). That Zn2+ is not a favored substrate for DMT1, together with the existence of alternative uptake pathways serving Zn2+, makes it unlikely that DMT1 is physiologically relevant to Zn absorption. On the other hand, Mn2+ may be a better substrate of DMT1, and the observation that intestinal Mn transport and reticulocyte Mn utilization also are impaired in the Belgrade rat (1, 14) points to a physiological role for DMT1 in Mn absorption. The question remains as to which other metal ions rely on DMT1 to enter enterocytes, but it is likely that Mn2+ does so and that Zn2+ does not.
DMT1 IS NOT LIMITED TO THE APICAL MEMBRANE
Although current models place DMT1 on the apical surface of enterocytes consistent with its function in entry of iron there (Fig. 1), several observers see it elsewhere in or near these cells. Their observations require verification and, if reliable, interpretation. Simovich et al. (22) detected DMT1 not only in microvilli but also in goblet cells and associated with the luminal mucin in the duodenum. They suggest that the extracellular DMT1 originates in the goblet cells and aids in iron acquisition in what they designate a Pac-Man model. These intriguing observations need confirmation, and the model needs to be tested. In a subsequent review in this series, evidence that supplying a bolus of iron to the enterocyte causes DMT1 to internalize from the apical surface on its way to hand iron off ultimately to apotransferrin should be covered. This model is intriguing, but it appears to redefine DMT1 as an iron-binding protein and introduces confusion about its membrane-transport role. Nevertheless, Sharp et al. (21) have also detected DMT1 departing from the brush border after exposure of Caco-2 cells exposed to iron, but they interpret this departure as regulatory rather than part of a handoff. Here, we have similar observations leading to two distinct hypotheses. Devising a test to distinguish the hypotheses (handoff vs. regulation) is in order.
Knöpfel et al. (14) also looked for DMT1 activity in both brush-border and basolateral membrane vesicles from enterocytes. As expected from the presence of DMT1 on the apical membrane, activity was present in brush-border preparations from control rats but undetectable in those from homozygous Belgrade rats. Surprisingly, basolateral vesicles also yielded comparable results. In these preparations, DMT1 was detected on immunoblots by isoform-specific antibodies, including one targeting the COOH terminus specific to the protein encoded by the mRNA containing a consensus iron-responsive element (IRE) in the 3′-untranslated region, often designated the +IRE form. Other antibodies targeted the unique COOH terminus encoded by the mRNA that lacks an IRE (the −IRE form) and the NH2 terminus encoded by exon 1A. Immunofluorescent localization confirmed that DMT1 isoforms had distinct distributions and that the −IRE form was particularly well represented on membranes distinct from the apical surface.
DMT1 at other locations could be involved in supplying the nutritional needs of enterocytes from ferric-transferrin by the transferrin cycle run in its usual direction, but other more interesting possibilities merit investigation. These include a role in delivering iron to apotransferrin, a downstream role in a handoff from DMT1 internalized from the apical surface after exposure to iron, or even a direct role in the basolateral exit of iron. This handoff possibility should be explored in a subsequent review in the series, but what is not clear is how to reconcile a role in Fe transcytosis with its transport activity in vitro. Even in the absence of a proton gradient, electrical and concentration gradients are likely to ensure that DMT1 serves only as an uptake system (16). If DMT1 were to play a direct role in iron exit, it is not redundant with that of IREG1/ferroportin1 (slc40a1) because selective intestinal knockout of slc40a1 causes a lethal microcytic, hypochromic anemia (4).
REGULATION OF APICAL IRON UPTAKE
DMT1 is dramatically upregulated in the intestine by dietary iron restriction or increased demand for iron and, despite high serum iron levels, is not appropriately downregulated in hereditary hemochromatosis, a defect associated with mutations in the HFE gene. The consensus IRE in the 3′-untranslated DMT1 mRNA of the predominant splice variant in intestine may underlie this iron-dependent regulation of DMT1. Binding of the IRE by an iron regulatory protein when intracellular iron is low could protect the mRNA from degradation. Although there is no doubt that the +IRE form is iron regulated, it is remarkable that investigators have known about the presence of the IRE for over eight years and yet have found little direct evidence that this regulation is mediated via the IRE itself.
It is now clear that the liver plays a major role in whole body iron homeostasis and that the hepatic hormone hepcidin is a key regulatory component (see other reviews in this series). Iron deficiency in the rat decreases hepcidin expression and stimulates intestinal iron absorption in parallel with increased duodenal expression of DMT1 and the basolateral efflux transporter IREG1/ferroportin1 (8). The prevailing view is that the effect of hepcidin on DMT1 is secondary to its action on IREG1/ferroportin1. Hepcidin binding by IREG1/ferroportin1 results in the internalization and degradation of IREG1/ferroportin1 protein, and consequent iron loading in the enterocyte could accelerate degradation of the +IRE DMT1 mRNA. DcytB levels also respond to hepcidin (8). Moreover, regulation of DMT1 expression at the protein level is observed in Caco-2 cells in response to iron loading (21). Further studies are required to explore the mechanisms by which apical iron uptake is regulated, and these are likely to include pathways that are independent of IREG1/ferroportin1.
ALTERNATIVE MECHANISMS FOR UPTAKE OF NONHEME IRON
Notwithstanding the dominant role of DMT1 in iron absorption, debate continues as to whether there exist alternative mechanisms for uptake of nonheme iron. Bacteria and yeast express many genes associated with iron transport; most of these are specific to certain chemical forms (salts, chelates, diferric transferrin, etc.). So it is plausible that uptake systems for specific iron complexes (among which are siderophores, ferritin, and lactoferrin) remain to be identified in humans.
One issue that needs experimental follow up is the role of mobilferrin, a protein that is either identical to the chaperone calreticulin or an unanchored form that has lost some COOH-terminal amino acid residues (see Ref. 22 for citations). Mobilferrin/calreticulin might be part of an uptake pathway for ferric ion (Fe3+), or it could participate along with DMT1 in Fe2+ uptake. The discussion above of what we learn from DMT1 mutants makes it unlikely that mobilferrin can substitute for DMT1 in Fe2+ uptake. Nevertheless, not substituting for DMT1 is not the same as playing no role. One cannot argue from the phenotype of calreticulin knockout mice because the knockout is lethal with a phenotype that points toward multiple roles for this chaperone (5). So how should one test the role of mobilferrin in iron metabolism? Investigators skilled at making tissue-specific ablation of genes in mice should do so for calreticulin in the gastrointestinal tract and ask what this does to iron homeostasis. Short of such a strong test, knockdown technology and studies of cell lines derived from embryonic stem cells from the calreticulin knockout mice are ways to learn whether mobilferrin/calreticulin plays a role in iron homeostasis.
ABSORPTION OF HEME IRON
Dietary heme iron has greater bioavailability than nonheme iron. Our understanding of iron absorption from heme has lagged behind that of nonheme iron absorption, and more research attention is clearly warranted. Heme (ferrous protoporphyrin IX) is liberated by the proteolytic digestion of hemoglobin and myoglobin in the lumen of the proximal small intestine. Heme is then taken up intact at the apical membrane (Fig. 1). Endocytosis of heme was originally demonstrated by tracking heme-associated peroxidase activity in the duodenum from animals fed hemoglobin or hemin chloride, and, around the same time, heme receptor activity was detected in the intestine (see Ref. 24 for citations). The stage is now set for molecular and genetic tools to enable identifying the intestinal heme receptor. Heme taken up into the endosome or lysosome is thought to be cleaved by heme oxygenase (to bilirubin and Fe2+), but how Fe2+ is mobilized from this compartment to the cytosol is not known. Whether DMT1 is among the candidates for this role (analogous to its role in transferrin-associated iron uptake in erythroid cells) could be tested by examining the intestinal handling of heme in the Belgrade rat and by determining whether one or more DMT1 isoforms are localized within the heme-containing endosomes/lysosomes.
HANDLING OF IRON BY THE ENTEROCYTE
Whether taken up by DMT1 at the brush border or mobilized from endosomes/lysosomes involved in heme uptake, Fe2+, we might speculate, meets the same cytosolic fate. Iron may be bound by ferritin, but the traditional view of ferritin as an iron storage protein may push us to look for other proteins that could chaperone the labile iron to the basolateral membrane for export. There are no obvious candidates, aside from Glass’ proposal that DMT1 may itself take part in transcytosis. The finding that mRNA for a lipocalin receptor increases in iron-deficient rat enterocytes (2) could mean that the search should widen to internal siderophores. It is thought that the ferroxidase hephaestin is an electron acceptor associated with the basolateral export of Fe2+ by IREG1/ferroportin1 before handing off Fe3+ to transferrin (Fig. 1). It is interesting to note that most of the hephaestin protein, however, localizes to perinuclear vesicles rather than the basolateral membrane. This may simply reflect a regulatory redistribution, but it seems worth exploring whether hephaestin could also be involved in shuttling iron to IREG1/ferroportin1.
NOTE ADDED IN PROOF
Three articles of particular relevance have appeared since the submission of our manuscript. Frazer et al. [Frazer DM, Wilkins SJ, Vulpe CD, and Anderson GJ. The role of duodenal cytochrome b in intestinal iron absorption remains unclear. Blood. In press.] present new evidence and discuss the conclusions made by Gunshin et al. in their study of the DcytB (Cybrd1) knockout mouse. Iolascon et al. report microcytic anemia and hepatic iron overload in a patient with novel compound heterozygous mutations in DMT1 [Iolascon A, d’Apolito M, Servedio V, Cimmino F. Piga A, and Camaschella C. Microcytic anemia and hepatic iron overload in a child with compound heterozygous mutations in DMT1. Blood. In press.]. Finally, Shayeghi et al. propose HCP1 as a candidate intestinal heme transporter/receptor [Shayeghi M, Latunde-Dada GO, Oakhill JS, Laftah AH, Takeuchi K, Halliday N, Khan Y, Warley A, McCann FE, Hider RC, Frazer DM, Anderson GJ, Vulpe CD, Simpson RJ and McKie AT. Identification of an intestinal heme transporter. Cell 122: 789–801, 2005].
Work in the authors' laboratories was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-59794, United States Department of Agriculture Grant 2001-35200-10723 (to M. D. Garrick), and by the Univ. of Cincinnati (to B. Mackenzie).
We thank Dr. L. M. Garrick (Depts. of Biochemistry and Medicine, State University of New York) for constructive comments on this review.
- Copyright © 2005 the American Physiological Society