Molecular Vision 2004; 10:598-607 <>
Received 2 April 2004 | Accepted 19 August 2004 | Published 26 August 2004

Immunolocalization and regulation of iron handling proteins ferritin and ferroportin in the retina

Paul Hahn,1 Tzvete Dentchev,1 Ying Qian,1 Tracey Rouault,2 Z. Leah Harris,3 Joshua L. Dunaief1

1F. M. Kirby Center for Molecular Ophthalmology, Scheie Eye Institute, Philadelphia, PA; 2Cell Biology and Metabolism Branch, National Institute of Child Health and Human Development, Bethesda, MD; 3Department of Anesthesiology and Critical Care Medicine, Division of Pediatric Anesthesiology and Critical Care Medicine, The Johns Hopkins University, Baltimore, MD

Correspondence to: Joshua L. Dunaief, 305 Stellar Chance Labs, 422 Curie Boulevard, Philadelphia, PA, 19104; Phone: (215) 898-5235; FAX: (215) 573-3918; email:


Purpose: CNS iron accumulation is associated with several neurodegenerative diseases, including age-related macular degeneration. Intracellular overload of free iron is prevented, in part, by the iron export protein, ferroportin, and the iron storage protein, ferritin. The purpose of this study was to assess retinal localization and regulation of ferroportin and ferritin.

Methods: Normal murine retinas were analyzed by immunohistochemistry to localize ferroportin, cytosolic ferritin, and mitochondrial ferritin, with double-labeling using cell-specific markers to identify cell types. Retinas deficient in the ferroxidases, ceruloplasmin and hephaestin, accumulate iron in their retinas and RPE, while retinas deficient in iron regulatory proteins (IRPs) lack the ability to regulate several proteins involved in iron metabolism; retinas from these knockout mice along with their age matched wild type littermates were also examined to study regulation of ferritin and ferroportin. To enable visualization of label in the retinal pigment epithelial cells, sections from pigmented mice were bleached with H2O2 prior to IHC, a novel use of this technique for study of the RPE.

Results: In normal retinas, cytosolic ferritins were found predominantly in rod bipolar cells and photoreceptors. Ferroportin was found in RPE and Müller cells. Iron accumulation in mice deficient in ceruloplasmin and hephaestin was associated with upregulation of ferritin and ferroportin. Mice deficient in IRPs showed upregulation of ferritin and ferroportin, likely because of their inability to repress translation.

Conclusions: Normal retinas contain ferritin and ferroportin, whose levels are regulated by iron-responsive, iron regulatory proteins. Ferroportin colocalizes with ceruloplasmin and hephaestin to RPE and Müller cells, supporting a potential cooperation between these ferroxidases and the iron exporter. Cytosolic ferritin accumulates in rod bipolar synaptic terminals, suggesting that ferritin may be involved in axonal iron transport. Mitochondrial ferritin increases with iron accumulation, suggesting a role in iron storage.


Iron is essential for survival but is also highly toxic due to its ability to generate free radicals via the Fenton reaction. Homeostatic mechanisms must thus delicately balance iron levels to prevent the deleterious consequences of either iron overload or deficiency. Cells regulate iron homeostasis in part by regulating the levels of iron import, storage, and export proteins in response to iron levels. The major pathway for iron import occurs when transferrin, the extracellular iron carrier protein, binds transferrin receptor on the cell surface and is endocytosed. Iron is then released from transferrin by the acidity of the low pH endosome [1] and is exported from the endosome by divalent metal transporter-1 (DMT-1) for use, storage, or export by the cell [2]. DMT-1 is also responsible for absorption of iron from the intestinal lumen into the intestinal endothelial cell [3].

Export of iron is achieved by iron transporters, such as ferroportin (also known as MTP-1 and Ireg-1). Ferroportin exports ferrous (2+) iron, which must be oxidized to its ferric (3+) form to be accepted by circulating transferrin [4,5]. Ferroportin is thus believed to cooperate with ferroxidases, ceruloplasmin (Cp) and hephaestin (Heph); exogenous Cp has been shown to augment iron export via ferroportin from oocytes [5], and co-immunoprecipitation studies on CNS astrocytes have demonstrated interaction between ferroportin and ceruloplasmin [6].

Storage of iron is achieved through sequestration by cytosolic ferritin, a multimeric heteropolymer comprised of 24 subunits of both heavy (H) and light (L) ferritin [7]. H-ferritin has ferroxidase activity essential for iron incorporation into the macromolecule. A single ferritin complex can accommodate up to 4500 iron atoms, making ferritin storage an efficient means of iron sequestration and detoxification.

Both ferroportin and H-/L-ferritin are post-transcriptionally regulated to increase with increasing cellular iron in order to maintain intracellular iron homeostasis. Iron regulatory proteins -1 and -2 (IRP-1 and -2) can bind to iron-responsive elements (IREs) in the 5' region of the mRNAs of ferroportin and H-/L-ferritin. As intracellular iron increases, IRPs lose the ability to bind IREs. IRP1 undergoes a conformational change upon insertion of an iron-sulfur cluster at the IRE binding site, resulting in occlusion of the IRE binding site. IRP2 undergoes iron-dependent degradation. Neither IRP binds to IREs in iron-replete cells, removing a steric obstruction to translation and resulting in an increase in ferritin levels [8,9]. Iron-dependent changes in ferroportin levels can act in both a similar and opposite manner as ferritin. In liver, iron depletion results in ferroportin downregulation, while iron depletion in duodenum results in increased ferroportin [10].

Another form of ferritin, mitochondrial ferritin (MtF), has recently been identified [11]. Most similar to H-ferritin, MtF has ferroxidase activity and high iron affinity suggesting that it may store iron in the mitochondria. Increased MtF has been detected in the mitochondria of iron-loaded sideroblasts from patients with sideroblastic anemia [12], but MtF does not have a recognizable IRE and has not been shown to be increased in response to other pathological conditions of iron overload.

Iron accumulation has been demonstrated in a variety of neurodegenerative diseases, including Alzheimer's disease, Parkinson's disease, and age-related macular degeneration [13-15], implicating iron in their pathogenesis and indicating a need to understand normal mechanisms of iron detoxification in the CNS. The retina is a particularly suitable model for CNS iron studies. The retina is constantly exposed to photo-oxidative stress and is thus especially vulnerable to damaging free radicals generated with iron excess. Diseases such as aceruloplasminemia and age-related macular degeneration as well as conditions such as siderosis bulbi and subretinal hemorrhage are associated with increased intraocular iron, which may contribute to the ensuing retinal degenerations [15-20]. Additionally, the eye is particularly iron-dependent. The extensive membrane biogenesis necessary to replenish continually shed photoreceptor outer segments requires iron as an essential cofactor [21]. These outer segments are phagocytosed by the retinal pigment epithelium (RPE), which must process the associated iron; further, as part of the blood-retinal barrier, the RPE functions to regulate flow of iron and other nutrients between the outer retina and the choroidal vasculature. Iron in a normal adult rat retina has been detected at highest levels in the choroid, RPE, and photoreceptor inner and outer segments [22].

The roles of iron handling proteins in the retina have begun to be investigated. Import of iron into the retina can occur by transferrin-receptor-mediated endocytosis in both retinal vascular endothelial cells and RPE, whose tight junctions form the basis of the blood-retinal barrier in the retinal and choroidal vasculature, respectively [23,24]. Iron in the retina may be bound to transferrin, which is expressed by RPE and neural retina, and can be endocytosed by cells throughout each retinal layer, which contain transferrin receptor [22]. Iron storage and export in the retina has been less extensively studied. Ferritin immunolocalized in the rat retina to the choroid, RPE, and inner segments [22]. Strong expression of murine MtF mRNA has been observed by RT-PCR analysis in testis and in lower amounts in brain, thymus, kidney, heart, and retina. Immunohistochemical localization of MtF has been successful in erythroid cells, in sperm, and in mitochondria-dense endocrine cells of the testis and pancreas [25]; retinal localization of MtF protein has not yet been published. Ferroportin, originally detected in the basolateral membrane of enterocytes, is also found in CNS neurons and cultured astrocytes [6,26] but has not been immunolocalized within retina.

The purpose of the current study was to systematically investigate the distribution of some of the major proteins involved in iron detoxification (ferroportin, H- and L-ferritin, and MtF) in the murine retina. We used an immunohistochemical approach in normal retinas, in iron overloaded retinas of ferroxidase-deficient, Cp-/-Heph-/Y mice, and in IRP deficient retinas with defective IRP-mediated iron regulation. The localization of these proteins sheds light on potential functions, both systemically and in the retina, and provides a baseline pattern from which to compare their levels and distribution in pathology.


Generation of mice and fixation of eyes

Retinas from wild type C57BL/6 and BALB/c mice (Jackson Laboratories, Bar Harbour, ME) 3-4 months old (n=2) and 6-7 months old (n=2) were studied for normal immunolocalization of proteins. To study the effects of iron accumulation, retinas from C57BL/6 mice with a mutation in Cp and/or Heph (Cp-/- and Cp-/-Heph-/Y) were studied along with their age-, strain-, and fixation matched, wild type littermates (n=4 for each genotype). Additionally, retinas from 9-12 month old C57BL/6 mice with mutations in IRP-1 and IRP-2 (IRP1+/-IRP2-/-, as IRP1-/-IRP2-/- mice die during embryogenesis) were studied along with their age-, strain-, and fixation matched, wild type littermates (n=2 for each genotype). Heph, an X-linked gene, is also called sla, as it was originally identified in sex-linked anemia mice [27].

Mice were reared with a 12 h light-dark cycle in a dedicated animal facility throughout their lives until sacrifice during daylight hours. Eyes from knockout mice and their age and strain matched wild type littermate controls were enucleated immediately after sacrifice and fixed overnight in 4% paraformaldehyde. In addition, wild type C57BL/6 and BALB/c eyes were enucleated and lightly fixed in 4% paraformaldehyde for 2 h to increase the sensitivity of immunohistochemistry for those sections.

Within each figure, all retinas were from age and strain matched mice and were fixed and immunostained simultaneously and identically. All results were verified by repeating the staining on retinas from independent sets of genotype matched mice. All procedures conformed to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and the procedures were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania.


Fixed globes were rinsed in PBS and prepared as eyecups, cryoprotected in 30% sucrose, and embedded in Tissue-Tek OCT (Sakura Finetek, USA, Inc., Torrance, CA). Immunohistochemistry was performed on cryosections 10 μm thick as published [28]. When comparisons were made among panels of individual figures, retinas in each panel were processed for immunohistochemistry identically and simultaneously. Primary antibodies were selected based on previous demonstrations of their specificity by Western analysis and immunohistochemistry. Rabbit anti-ferroportin [10], a gift from D. Haile (University of Texas Health Science Center, San Antonio, Texas), was diluted 1:20. Mouse anti-CRALBP [29], a gift from J. Saari (University of Washington, Seattle, Washington), was diluted 1:250. Mouse anti-PKC-alpha, which labels rod bipolar cells [30], was purchased from Pharmingen (San Diego, CA) and was diluted 1:500. Mouse anti-mitochondrial ATPase complex V (clone 7H10) was purchased from Molecular Probes (Eugene, OR) and was diluted 1:200. P. Santambrogio, S. Levy, and P. Arosio (IRCCS, Milan, Italy) generously provided the following antibodies: rabbit anti-light ferritin (F17, 1:2500, 6.2 μg/ml), rabbit anti-heavy ferritin [31] (Y17; 1:2500, 4.2 μg/ml), rabbit anti-mitochondrial ferritin (1:1000) [32]. Control sections were treated identically but with omission of primary antibody.

After observing that pigment within the RPE quenches immunofluorescence, some sections were pre-treated with 3% H2O2 for 13 h in order to bleach endogenous melanin. In the past, bleaching with H2O2 has been used prior to immunostaining of melanotic neoplasms [33,34]. We have applied a similar bleaching technique, which, to our knowledge, is its first reported use in the RPE. In the past, we have used H2O2 bleaching prior to Perl's staining for iron [15], with preservation of retinal morphology. In the current paper, comparison of anti-ferroportin labeling in bleached sections from pigmented mice to unbleached sections from albino mice suggests that H2O2 bleaching prior to retinal IHC is useful and valid. Like H2O2 bleached sections from pigmented, iron overloaded cp-/-heph-/Y mice, the RPE in sections from albino mice labels with anti-ferroportin (Figure 1).

Secondary antibodies (donkey anti-rabbit and anti-mouse) were labeled with Cy-3 (red; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) and/or Cy-2 (green). Nuclei were counterstained with DAPI (1.5 μg/ml)-supplemented Vectashield mounting medium (Vector Laboratories, Burlingame, CA). Sections were analyzed by fluorescence microscopy using identical exposure parameters across genotypes when comparisons were made. Confocal microscopy was performed with a Zeiss LSM 510 confocal microscope (Carl Zeiss, Inc., Oberkochen, Germany), and epifluorescence microscopy was performed with a Nikon TE-300 microscope (Nikon Inc., Kanagawa, Japan) and SpotRT Slider camera (Diagnostic Instruments, Inc., Sterling Heights, MI) with ImagePro Plus software, version 4.1 (Media Cybernetics, Silver Spring, MD). All images were photographed at 1000 μm from the optic nerve head.

Reverse transcriptase-polymerase chain reaction (RT-PCR) analysis

C57BL/6 mice were sacrificed and eyes promptly enucleated. Anterior segments were removed, and retinas were completely dissected away from the underlying RPE. Retinas were then flash frozen and stored at -80 °C. RPE was enzymatically detached from the remaining posterior segment by incubation in 0.25% trypsin at 37 °C, 5% CO2 for 20 min. Detached RPE was collected in an eppendorf tube, pelleted, flash frozen and stored at -80 °C.

RNA was extracted from primary murine retina or RPE using TRIZOL (Gibco, BRL Life Technologies, Rockville, MD) and used to generate first strand cDNA with a T7-(dT)24 oligomer primer and the SuperScript II reverse transcriptase (InVitrogen, Carlsbad, CA). Specific sequences were then amplified by PCR using a MJ research PTC-0200 thermocycler (MJ Research, Inc., Waltham, MA).

Primers specific for murine H-ferritin or L-ferritin were designed to span intron-exon boundaries within genomic DNA in order to amplify cDNA and minimize amplification of potentially contaminating genomic DNA. The H-ferritin forward primer was 5'-TTT GAG CCT GAG CCC TTT G-3', and the reverse primer was 5'-TCA AAG AGA TAT TCT GCC ATG C-3'. The L-ferritin forward primer was 5'-AGC GTC TCC TCG AGT TTC AG-3', and the reverse primer was 5'-AGG TTG GTC AGA TGG TTG C-3'.


Ferroportin is present in the RPE and Müller cell endfeet

Normal lightly fixed retinas (n=4) labeled for the iron transporter ferroportin (Figure 1) had positive Müller endfeet at the inner surface of the retina (Figure 1B), as confirmed by double labeling with cellular retinaldehyde binding protein (CRALBP; Figure 1C-D), a marker of Müller cells and RPE. Ferroportin was occasionally detectable within Müller processes, but levels were highest near the Müller endfeet. Ferroportin was also present in photoreceptor inner segments, in the outer plexiform layer (OPL), and in a punctate pattern throughout the inner plexiform layer (IPL). BALB/c (Figure 1) and C57BL/6 (not shown) retinas had an identical pattern of localization within Müller endfeet and occasionally within Müller processes. Ferroportin was present in normal RPE (Figure 1D; yellow label in RPE), with exclusion from the RPE's apical microvilli (Figure 1D; green only CRALBP label in RPE). Because the pigment in C57BL/6 eyes quenches fluorescent label in the RPE, albino BALB/c mice are shown in figures of normal mice to best represent retinal and RPE patterns of immunolocalization. In figures of knockout mice (below), C57BL/6 mice are shown, to match the genetic background of the knockout mice.

Ferroportin increases with iron accumulation

To examine the effects of iron accumulation on ferroportin levels, we examined Cp-/-Heph-/Y retinas, which have RPE and neural retinal iron overload by 6 months [35]. Six month old Cp-/-Heph-/Y retinas were compared with retinas from age, strain, and fixation matched wild type and Cp-/- littermates (n=4 for each genotype) immunolabeled for ferroportin. Cp-/- retinas (Figure 2B) had a modest increase in ferroportin label compared to wild type retinas (Figure 2A), and Cp-/-Heph-/Y retinas had substantially increased ferroportin (Figure 2C). Ferroportin immunolabel in the RPE was obscured by endogenous melanin of the pigmented mice, and bleaching of the RPE was required to optimally observe ferroportin levels within the RPE. In Cp-/-Heph-/Y bleached RPE (Figure 2F) ferroportin localized to basolateral and apical regions, and ferroportin levels were strikingly increased compared with wild type (Figure 2D) and Cp-/- (Figure 2E) mice. The ferroportin inner segment label was highly fixation dependent and was visible in lightly fixed sections (Figure 1) but not in these more heavily fixed sections (Figure 2). Sections within each panel of Figure 2 were fixed identically to allow semi-quantitative comparisons among panels in Figure 2 but not between the more heavily fixed retinas of Figure 2 and the lightly fixed retina in Figure 1.

Ferroportin increases with IRP-deficiency

To determine whether IRPs mediated the iron overload-induced increase in retinal ferroportin, we immunolabeled ferroportin in retinas from mice with IRP deficiency (n=2) and their wild type, age matched littermates (n=2). These mice had the genotype IRP1+/- IRP2-/- which results in neurodegenerative disease by 9-12 months that is much more severe than that of IRP2-/- animals [36,37]. IRP1 contributes to baseline iron homeostasis [38], and since loss of one IRP1 allele significantly worsens the neurodegenerative disease of IRP2-/- animals [36], we chose to examine ferritin and ferroportin expression in these severely affected animals. These IRP deficient retinas (Figure 3B) had increased ferroportin in the inner segments, Müller endfeet, and inner retina compared to their age, strain, and fixation matched wild type retinas (Figure 3A), suggesting that in the retina, ferroportin levels are regulated by IRPs.

H- and L-ferritin are present in rod bipolar cells

Normal lightly fixed BALB/c (Figure 4) and C57BL/6 (not shown) retinas (n=3) labeled for H- or L-ferritin both had positive processes in the photoreceptor inner segments, outer plexiform layer (OPL), inner nuclear layer (INL) cell bodies, and small spheres in the innermost inner plexiform layer (IPL) adjacent to the ganglion cell layer (GCL; Figure 4A). Specificity of ferritin label was confirmed by pre-incubation of antibody with a 5 M excess of purified H or L-ferritin (gift from P. Arosio), each of which eliminated label by the corresponding antibody but did not affect label with the unmatched antibody (not shown). Double labeling with PKCα, a marker for rod bipolar cells, and H-ferritin (Figure 4A-C) or L-ferritin (Figure 4D-F) demonstrated that many of the ferritin positive INL cells were rod bipolars, with ferritin accumulation in their axon terminals (punctate label in the IPL, Figure 4C,F). In the RPE, both ferritin subtypes had minimal immunofluorescence even in non-pigmented BALB/c RPE (Figure 4A,D). To confirm ferritin expression in retina and to determine whether ferritin is expressed in the RPE, RT-PCR analysis was performed on cDNA from freshly harvested murine retina and RPE cells using primers specific for either H- or L-ferritin. In two different experiments, one from a single 6 month old C57BL/6 mouse (Figure 4G) and another from 3 mice aged 3, 5, and 7 months (not shown), both H- and L-ferritin mRNAs were detected in both retina, consistent with immunohistochemical results, and RPE. The purity of isolated RPE was demonstrated by amplification of RPE65, an RPE specific gene, but not opsin, a retina specific gene.

Ferritin increases with IRP-deficiency

We have previously demonstrated that iron accumulation in Cp-/-Heph-/Y retinas results in an increase in both H- and L-ferritin levels (unpublished data). To determine whether these changes in ferritin levels might result from IRP regulation, we immunolabeled retinas from IRP deficient animals (n=2) and their age matched, wild type littermates (n=2) for L-ferritin. If retinal ferritin is IRP-regulated, these deficient retinas should have increased ferritin levels from relief of IRP-mediated inhibition of ferritin translation. Retinas from IRP deficient mice had increased L-ferritin label compared to their age, strain, and fixation matched wild type retinas. The L-ferritin label was increased in rod bipolar axon termini, outer plexiform processes, and photoreceptor inner segments of IRP deficient mice, suggesting that retinal ferritin levels are regulated by IRPs. As with ferroportin, the ferritin inner segment label was fixation dependent and was visible in lightly fixed sections (Figure 4) but difficult to visualize in more heavily fixed sections (Figure 5). Sections within each panel of Figure 5 were fixed identically to each other to allow semi-quantitative comparisons between the panels of Figure 5 only.

Mitochondrial ferritin is present in inner segments and increases with iron accumulation

Normal retinas had subtle mitochondrial ferritin (MtF) label in the mitochondria-rich inner segments of photoreceptors (Figure 6A) and diffusely throughout the inner retina. To confirm that this label corresponded to MtF and to examine the effect of iron accumulation on MtF levels, retinas with iron accumulation from deficiency of Cp and Cp/Heph (n=3 for each genotype) were also labeled for MtF. All genotypes had a diffuse pattern of MtF label throughout the inner retina (not shown). MtF label in the photoreceptor inner segments was increased in Cp-/- retinas (Figure 6B) compared with their age, strain, and fixation matched wild type retinas (Figure 6A), and MtF was further increased in Cp-/-Heph-/Y retinas (Figure 6C). To verify that anti-MtF labels mitochondria, retinas were co-labeled with anti-MtF and a mitochondria-specific antibody recognizing an ATPase in Complex V of the electron transport chain. Mitochondria are prominent in the inner segment ellipsoids, excluding the inner segment myoid, which lies between the ellipsoid and the photoreceptor nuclei [39]. As shown in a Cp-/- retina (Figure 6D-F), the mitochondria-specific ATPase and MtF co-localize specifically to ellipsoids and exclude the myoid (arrowheads, Figure 6D-F).


Iron toxicity has been implicated in the pathogenesis of several neurodegenerative diseases, but mechanisms which regulate normal iron homeostasis are incompletely understood. Cells can handle excess iron by decreasing import of iron or by increasing iron export by ferroportin, or by increasing iron storage by cytosolic ferritin. Another type of ferritin, mitochondrial ferritin, can bind iron and protect mitochondria from iron toxicity [32]. In this study, we used an immunohistochemical approach to shed light on the retinal functions and regulation of ferroportin, cytosolic ferritin, and mitochondrial ferritin.

In normal adult retinas, ferroportin was present in photoreceptor inner segments, the outer plexiform layer, Müller endfeet, and RPE. Iron overload in Cp-/-Heph-/Y retinas resulted in increased ferroportin levels in Müller endfeet and in the apical and basolateral surfaces of the RPE. This increased ferroportin is likely a result of iron-mediated inhibition of IRP function, as IRP deficient retinas also have increased ferroportin levels relative to their age matched wild type littermates. As ferroportin functions as an iron exporter [4,10], IRPs respond to iron overload in the retina by increasing levels of ferroportin, presumably to increase iron export. Ferroportin also immunolocalized to the outer plexiform layer, which contains synapses of photoreceptor axons and inner nuclear layer neuronal dendrites. Since ferroportin has recently been identified in association with brain synaptic vesicles [40], immuno-electron microscopy studies will be performed to determine whether ferroportin is present in synaptic vesicles in the retina.

Ferroportin transports ferrous iron [5], which must be oxidized to its ferric form to bind transferrin. Analogously, in yeast, the transporter protein Ftr1 acts in concert with ferroxidase Fet3 to efficiently transport iron across the yeast membrane [41]. A critical question in mammalian iron biology is whether ferroportin activity is similarly coupled to ferroxidases, Heph and/or Cp. Previous reports have demonstrated that addition of exogenous Cp can increase iron export from oocytes [4] and from Cp-/- mouse brain astrocytes in culture [6]. Further, ferroportin and Cp have been shown to co-immunoprecipitate in cultured astrocytes [6]. We have previously shown by immunohistochemistry, western analysis, and RT-PCR that Cp and Heph are present in both RPE and Müller cells, particularly their endfeet [35]. Our finding in this report that ferroportin is also present in Müller endfeet and RPE further supports a cooperative role for ferroportin and both ferroxidases in RPE iron export or in Müller cell iron export into the vitreous; Cp is also present in the vitreous [42] and in plasma of the choriocapillaris and may further facilitate iron export through Müller endfeet and RPE, respectively.

Prominent ferritin label was present in murine rod bipolar cells, particularly their cell bodies in the INL and their synaptic terminals in the innermost IPL. Retinal ferritin appears to be IRP-regulated, as IRP-deficient retinas have increased ferritin label relative to their age matched wild type controls. H- and L-ferritin label was present in inner segments of normal murine photoreceptors, consistent with a previous report studying rat retinas [22]. This report also detected high levels of iron in photoreceptor inner segments, and inner segment levels of ferritin appear sensitive to IRPs, suggesting that the inner segment may be an important site for iron sequestration. In contrast, neither H- nor L-ferritin was detected in photoreceptor outer segments, which are shed daily and phagocytosed by RPE. The inner segment label for both ferritin and ferroportin was present in retinas from both BALB/c (Figure 1 and Figure 4) and C57BL/6 (not shown) mice, but only when the retinas were lightly fixed. Perhaps surprisingly, little ferritin protein was detected in RPE, although we did detect H- and L-ferritin expression in RPE by RT-PCR. As the RPE comprises the outer blood-retinal barrier and thus may be subject to high iron flux, normal, non-iron overloaded RPE, which has robust ferroportin label, may have little need to store iron in ferritin, explaining low protein levels.

Mice deficient in Cp and Heph accumulate iron in their RPE and retinas by 6 months, and mice deficient in IRPs become neurologically symptomatic by 9-12 months; we thus studied 6 month old Cp/Heph-deficient mice and 9-12 month old IRP-deficient mice along with their age matched littermate wild type controls. Because immunohistochemistry is a semi-quantitative technique, we have taken many measures to standardize experimental conditions. Within each figure, retinas were age, strain, and fixation matched, and retinas from the individual panels in each figure were processed in parallel, beginning at the time of enucleation until the end of immunolabeling. While we did not observe any obvious differences in immunolabel of any studied proteins between normal mice at 3-4 months and 6-7 months, there could be subtle age-related changes in iron-handling proteins. Therefore, comparisons in label among different genotypes were always made with simultaneously and identically processed age matched littermate wild type controls. Differences in label intensity within each figure were striking and reproducible among replicate experiments. At the ages studied herein, despite evidence of abnormal iron metabolism, electroretinography on IRP deficient mice was normal. These results indicate that mice of these ages have normal summated rod, cone, and bipolar cell function, but do not preclude the possibility that further disruption of iron homeostasis in older mice might lead to abnormal ERGs.

In IRP2-deficient mice, axon bundles in the brain have increased ferritin label, and the first sign of pathology is axonal degeneration, suggesting that ferritin may be involved in iron trafficking from the cell body to the synaptic terminal [36]. In retina, the unmyelinated rod bipolar cell axons label for ferritin, particularly at their synaptic terminals, supporting the hypothesis that ferritins may be involved in iron trafficking to synaptic terminals.

Mitochondrial ferritin (MtF) is a third, recently described ferritin subtype [11]. We provide immunohistochemical identification of MtF in the mitochondria of photoreceptor inner segment ellipsoids. MtF is increased in conditions of iron excess induced by deficiency of Cp and further increased by deficiency of both Cp and Heph. While the MtF in Cp-/- inner segments colocalizes with ellipsoid mitochondria, the MtF in the Cp-/-Heph-/Y inner segments appear less specifically localized to ellipsoids and present more diffusely throughout both the myoid and ellipsoid. MtF is translated in the cytosol within the myoid, and a mitochondrial localization signal sequence targets the newly translated protein to the mitochondria [11]. Perhaps the dramatic iron overload in the Cp-/-Heph-/Y retina upregulates MtF translation to the point at which it exceeds the capacity for translocation into mitochondria, resulting in a more diffuse localization to the Cp-/-Heph-/Y inner segment myoid and ellipsoid. The increases in MtF within Cp-/- and Cp-/-Heph-/Y iron-overloaded inner segments are consistent with increases observed in sideroblastic anemia [12]. MtF, however, does not have a known iron-responsive element, and further studies will be needed to understand the role and regulation of MtF.

The retina has two blood supplies, each with a blood-brain barrier. Supplying the inner retina is the retinal vasculature, whose blood-brain barrier consists of tight junctions between retinal endothelial cells. Supplying the photoreceptors and RPE is the choroidal circulation, whose blood-brain barrier consists of the tight junctions between RPE cells. Import of iron into the retina can occur by typical transferrin receptor mediated endocytosis in both RPE and retinal endothelial cells [23,24]. Transferrin is expressed by both RPE and neural retina, including photoreceptors, and may deliver iron to transferrin receptor, which is present on cells within each retinal layer, including photoreceptor inner segments [22,43].

Export of iron in the retina has been less extensively studied, although the data presented herein suggests that at least one mechanism of iron export from the retina may be into the choroidal circulation through the RPE or into the vitreous through Müller cell endfeet at the inner limiting membrane. Together with the exclusion of ferroportin from the apical surface of normal RPE, these data suggest that at least one pathway for iron flux through the retina may be to normally enter the retina via the retinal vasculature and exit the retina through the basolateral RPE into the choroidal circulation or through the Müller cell endfeet and the ILM into the vitreous. Future studies will investigate the subcellular localization of iron handling proteins, including iron import, storage, and export proteins, and in vitro RPE culture systems will be used to better understand the normal flux of iron and its regulatory mechanisms, ultimately to further understand the effects of iron misregulation in neurodegenerative diseases including AMD.


We gratefully acknowledge the generous antibody gifts from P. Arosio, P. Santambrogio, S. Levy, D. Haile, and J. Saari.

This work was supported by NIH: R01EY015240, K08EY00417, MSTP T32GM7170, DK02464 and DK58086, a Career Development Award from Research to Prevent Blindness, International Retina Research Foundation, the Steinbach Foundation, and the Paul and Evanina Bell Mackall Foundation Trust.


1. Sipe DM, Murphy RF. Binding to cellular receptors results in increased iron release from transferrin at mildly acidic pH. J Biol Chem 1991; 266:8002-7.

2. Fleming MD, Romano MA, Su MA, Garrick LM, Garrick MD, Andrews NC. Nramp2 is mutated in the anemic Belgrade (b) rat: evidence of a role for Nramp2 in endosomal iron transport. Proc Natl Acad Sci U S A 1998; 95:1148-53.

3. Gunshin H, Mackenzie B, Berger UV, Gunshin Y, Romero MF, Boron WF, Nussberger S, Gollan JL, Hediger MA. Cloning and characterization of a mammalian proton-coupled metal-ion transporter. Nature 1997; 388:482-8.

4. Donovan A, Brownlie A, Zhou Y, Shepard J, Pratt SJ, Moynihan J, Paw BH, Drejer A, Barut B, Zapata A, Law TC, Brugnara C, Lux SE, Pinkus GS, Pinkus JL, Kingsley PD, Palis J, Fleming MD, Andrews NC, Zon LI. Positional cloning of zebrafish ferroportin1 identifies a conserved vertebrate iron exporter. Nature 2000; 403:776-81.

5. McKie AT, Marciani P, Rolfs A, Brennan K, Wehr K, Barrow D, Miret S, Bomford A, Peters TJ, Farzaneh F, Hediger MA, Hentze MW, Simpson RJ. A novel duodenal iron-regulated transporter, IREG1, implicated in the basolateral transfer of iron to the circulation. Mol Cell 2000; 5:299-309.

6. Jeong SY, David S. Glycosylphosphatidylinositol-anchored ceruloplasmin is required for iron efflux from cells in the central nervous system. J Biol Chem 2003; 278:27144-8.

7. Harrison PM, Arosio P. The ferritins: molecular properties, iron storage function and cellular regulation. Biochim Biophys Acta 1996; 1275:161-203.

8. Hentze MW, Kuhn LC. Molecular control of vertebrate iron metabolism: mRNA-based regulatory circuits operated by iron, nitric oxide, and oxidative stress. Proc Natl Acad Sci U S A 1996; 93:8175-82.

9. Rouault T, Klausner R. Regulation of iron metabolism in eukaryotes. Curr Top Cell Regul 1997; 35:1-19.

10. Abboud S, Haile DJ. A novel mammalian iron-regulated protein involved in intracellular iron metabolism. J Biol Chem 2000; 275:19906-12.

11. Levi S, Corsi B, Bosisio M, Invernizzi R, Volz A, Sanford D, Arosio P, Drysdale J. A human mitochondrial ferritin encoded by an intronless gene. J Biol Chem 2001; 276:24437-40.

12. Cazzola M, Invernizzi R, Bergamaschi G, Levi S, Corsi B, Travaglino E, Rolandi V, Biasiotto G, Drysdale J, Arosio P. Mitochondrial ferritin expression in erythroid cells from patients with sideroblastic anemia. Blood 2003; 101:1996-2000.

13. Thompson KJ, Shoham S, Connor JR. Iron and neurodegenerative disorders. Brain Res Bull 2001; 55:155-64.

14. Perry G, Sayre LM, Atwood CS, Castellani RJ, Cash AD, Rottkamp CA, Smith MA. The role of iron and copper in the aetiology of neurodegenerative disorders: therapeutic implications. CNS Drugs 2002; 16:339-52.

15. Hahn P, Milam AH, Dunaief JL. Maculas affected by age-related macular degeneration contain increased chelatable iron in the retinal pigment epithelium and Bruch's membrane. Arch Ophthalmol 2003; 121:1099-105.

16. Klomp LW, Farhangrazi ZS, Dugan LL, Gitlin JD. Ceruloplasmin gene expression in the murine central nervous system. J Clin Invest 1996; 98:207-15.

17. Morita H, Ikeda S, Yamamoto K, Morita S, Yoshida K, Nomoto S, Kato M, Yanagisawa N. Hereditary ceruloplasmin deficiency with hemosiderosis: a clinicopathological study of a Japanese family. Ann Neurol 1995; 37:646-56.

18. Doly M, Bonhomme B, Vennat JC. Experimental study of the retinal toxicity of hemoglobinic iron. Ophthalmic Res 1986; 18:21-7.

19. Vergara O, Ogden T, Ryan S. Posterior penetrating injury in the rabbit eye: effect of blood and ferrous ions. Exp Eye Res 1989; 49:1115-26.

20. Tawara A. Transformation and cytotoxicity of iron in siderosis bulbi. Invest Ophthalmol Vis Sci 1986; 27:226-36.

21. Shichi H. Microsomal electron transfer system of bovine retinal pigment epithelium. Exp Eye Res 1969; 8:60-8.

22. Yefimova MG, Jeanny JC, Guillonneau X, Keller N, Nguyen-Legros J, Sergeant C, Guillou F, Courtois Y. Iron, ferritin, transferrin, and transferrin receptor in the adult rat retina. Invest Ophthalmol Vis Sci 2000; 41:2343-51.

23. Hunt RC, Dewey A, Davis AA. Transferrin receptors on the surfaces of retinal pigment epithelial cells are associated with the cytoskeleton. J Cell Sci 1989; 92:655-66.

24. Burdo JR, Antonetti DA, Wolpert EB, Connor JR. Mechanisms and regulation of transferrin and iron transport in a model blood-brain barrier system. Neuroscience 2003; 121:883-90.

25. Drysdale J, Arosio P, Invernizzi R, Cazzola M, Volz A, Corsi B, Biasiotto G, Levi S. Mitochondrial ferritin: a new player in iron metabolism. Blood Cells Mol Dis 2002; 29:376-83.

26. Burdo JR, Menzies SL, Simpson IA, Garrick LM, Garrick MD, Dolan KG, Haile DJ, Beard JL, Connor JR. Distribution of divalent metal transporter 1 and metal transport protein 1 in the normal and Belgrade rat. J Neurosci Res 2001; 66:1198-207.

27. Vulpe CD, Kuo YM, Murphy TL, Cowley L, Askwith C, Libina N, Gitschier J, Anderson GJ. Hephaestin, a ceruloplasmin homologue implicated in intestinal iron transport, is defective in the sla mouse. Nat Genet 1999; 21:195-9.

28. Dunaief JL, Dentchev T, Ying GS, Milam AH. The role of apoptosis in age-related macular degeneration. Arch Ophthalmol 2002; 120:1435-42.

29. Bunt-Milam AH, Saari JC. Immunocytochemical localization of two retinoid-binding proteins in vertebrate retina. J Cell Biol 1983; 97:703-12.

30. Negishi K, Kato S, Teranishi T. Dopamine cells and rod bipolar cells contain protein kinase C-like immunoreactivity in some vertebrate retinas. Neurosci Lett 1988; 94:247-52.

31. Santambrogio P, Cozzi A, Levi S, Rovida E, Magni F, Albertini A, Arosio P. Functional and immunological analysis of recombinant mouse H- and L-ferritins from Escherichia coli. Protein Expr Purif 2000; 19:212-8.

32. Campanella A, Isaya G, O'Neill HA, Santambrogio P, Cozzi A, Arosio P, Levi S. The expression of human mitochondrial ferritin rescues respiratory function in frataxin-deficient yeast. Hum Mol Genet 2004; 13:[print version pending].

33. Orchard GE. Heavily pigmented melanocytic neoplasms: comparison of two melanin-bleaching techniques and subsequent immunohistochemical staining. Br J Biomed Sci 1999; 56:188-93.

34. Li LX, Crotty KA, Kril JJ, Palmer AA, McCarthy SW. Method of melanin bleaching in MIB1-Ki67 immunostaining of pigmented lesions: A quantitative evaluation in malignant melanomas. Histochem J 1999; 31:237-40.

35. Hahn P, Qian Y, Chen L, Beard J, Harris ZL, Dunaief JL. Disruption of ceruloplasmin and hephaestin in mice causes retinal iron overload and retinal degeneration with features of age-related macular degeneration. Proc Natl Acad Sci U S A. In press 2004.

36. LaVaute T, Smith S, Cooperman S, Iwai K, Land W, Meyron-Holtz E, Drake SK, Miller G, Abu-Asab M, Tsokos M, Switzer R 3rd, Grinberg A, Love P, Tresser N, Rouault TA. Targeted deletion of the gene encoding iron regulatory protein-2 causes misregulation of iron metabolism and neurodegenerative disease in mice. Nat Genet 2001; 27:209-14.

37. Smith SR, Cooperman S, Lavaute T, Tresser N, Ghosh M, Meyron-Holtz E, Land W, Ollivierre H, Jortner B, Switzer R 3rd, Messing A, Rouault TA. Severity of neurodegeneration correlates with compromise of iron metabolism in mice with iron regulatory protein deficiencies. Ann N Y Acad Sci 2004; 1012:65-83.

38. Meyron-Holtz EG, Ghosh MC, Iwai K, LaVaute T, Brazzolotto X, Berger UV, Land W, Ollivierre-Wilson H, Grinberg A, Love P, Rouault TA. Genetic ablations of iron regulatory proteins 1 and 2 reveal why iron regulatory protein 2 dominates iron homeostasis. EMBO J 2004; 23:386-95.

39. Albert DM, Jakobiec FA. Principles and practice of ophthalmology: basic sciences. Philadelphia: Saunders; 1994.

40. Wu LJ, Leenders AG, Cooperman S, Meyron-Holtz E, Smith S, Land W, Tsai RY, Berger UV, Sheng ZH, Rouault TA. Expression of the iron transporter ferroportin in synaptic vesicles and the blood-brain barrier. Brain Res 2004; 1001:108-17.

41. Aisen P, Enns C, Wessling-Resnick M. Chemistry and biology of eukaryotic iron metabolism. Int J Biochem Cell Biol 2001; 33:940-59.

42. Chen L, Dentchev T, Wong R, Hahn P, Wen R, Bennett J, Dunaief JL. Increased expression of ceruloplasmin in the retina following photic injury. Mol Vis 2003; 9:151-8 <>.

43. Davis AA, Hunt RC. Transferrin is made and bound by photoreceptor cells. J Cell Physiol 1993; 156:280-5.

Hahn, Mol Vis 2004; 10:598-607 <>
©2004 Molecular Vision <>
ISSN 1090-0535