Published online before print October 18, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Medicine, University Hospital, Zurich, Switzerland
1 Correspondence: Medical Clinic Research Unit, Department of Medicine, University Hospital, CH-8091 Zurich, Switzerland. E-mail: dominik.schaer{at}usz.ch
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The profound suppression of HCP-1 expression by inflammatory macrophage activation parallels the regulation of the iron exporter ferroportin. In contrast, dexamethasone enhanced HCP-1 expression significantly. Given the spatial relationship, we propose that the Hb scavenger receptor CD163 and HCP-1 constitute a linked pathway for Hb catabolism and heme-iron recycling in human macrophages.
Key Words: iron metabolism • inflammation • anemia of chronic disease
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
So far, endocytosis by the monocyte/macrophage-specific scavenger receptor CD163 comprises the only known pathway for uptake of cell-free Hb as well as for Hb bound to the plasma protein haptoglobin (Hp; Hb-Hp complexes) [3 4 5 ]. We have shown recently that the high capacity of the CD163 pathway for Hb uptake is related to the receptor’s capacity to undergo ligand-independent (constitutive) endocytosis and subsequent recycling to the cell surface [6 ]. This study revealed a response in macrophages to Hb exposure, which is dominated by induction of the heme breakdown enzyme heme oxygenase (HO-1). The effector proteins involved in heme and heme-iron handling after CD163-mediated Hb endocytosis have not been investigated yet. However, it seems prudent to speculate that specific transporters are involved in the transit of Hb-derived heme and heme-iron from the endosomal/lysosomal compartment to the cytoplasm. Once in the cytoplasm, heme is catabolized further by HO-1, and the released heme-iron is then bound to the storage protein ferritin or exported to plasma via the iron exporter ferroportin.
Heme carrier protein (HCP-1) was identified recently as a luminal transporter in duodenal enterocytes, which transports folate, and with a lower affinity, heme [7 8 9 ] HCP-1 is the first mammalian cellular heme importer to be characterized and belongs to the major facilitator superfamily. As many functional and regulatory facets of the molecular machinery for heme catabolism and iron handling are shared by enterocytes and macrophages, it is reasonable to speculate that HCP-1 could also play a pivotal role in the turnover of Hb-derived heme by macrophages [10 ]. The existence of an intracellular pool of HCP-1, which was discovered in enterocytes, might provide clues regarding its role in the export of Hb-derived heme from endosomes and/or lysosomes [7 ].
Dysregulation of macrophage iron metabolism is a major factor underlying anemia in chronic disease states. This common form of anemia is, in part, caused by inflammatory-mediated sequestration of iron within macrophages, which leads to iron-restricted erythropoesis [11 ]. Inflammation induces transcriptional and post-transcriptional changes that modify expression of several iron-handling proteins in the macrophage, including the transferrin receptor, the divalent metal transporter 1 (DMT-1), and the iron exporter ferroportin [12 13 14 15 16 17 18 19 20 21 ]. Inflammation-related disturbances in macrophage heme-iron recycling, through altered expression of proteins involved in Hb or heme catabolism, could prove to be of similar significance eventually.
In this study, we investigated the expression of HCP-1 in human macrophages and asked whether the subcellular location of HCP-1 within the Hb endocytic pathway could lend evidence to an active role for this protein in Hb-derived heme processing. In addition, we reveal that HCP-1 is a novel target for inflammatory and anti-inflammatory stimulation of macrophages.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Human blood-derived macrophages were prepared from buffy coats of healthy donors purchased from the Swiss Red Cross Blood Bank (Zurich), as described previously [22 ], and cultured in IMDM (Gibco Europe), supplemented with 10% heat-inactivated, pooled human serum (PAA Laboratories GmbH, Linz, Austria).
Establishment of C-terminal GFP fusion proteins expressed in HEK293 cells
The following primers were used to amplify the full-length HCP-1 fragments for generation of HCP-1A-emerald green fluorescent protein (EmGFP) and HCP-1B-EmGFP fusion proteins: forward: 5'-ACG CAC ATG GAG GGG AG-3'; reverse: 5'-GGG GCT CTG GGG AAA CTG CTG GAA C-3'. For the 1377-bp, high-abundance variant (HCP-1A), we used reverse-transcribed cDNA as template, obtained from mRNA, isolated from human blood-derived macrophages. A commercial cDNA clone (Geneservice Ltd., Cambridge, UK) was used for the 1293-bp, low-abundance variant (HCP-1B). Amplicons were generated using AccuPrime PFX DNA polymerase (Invitrogen). C-terminal EmGFP fusion proteins were generated using the pcDNA 6.2/C-EmGFP-GW/TOPO plasmid vector (Invitrogen). Insert orientation and integrity were verified by sequencing (Microsynth, Switzerland). To obtain stable transfectants, expression constructs were linearized with PSP 1406 I (Roche Diagnostics, Rotkreuz, Switzerland) and transfected into CD163-expressing HEK293 cells using Lipofectamine 2000 (Invitrogen). Uniform populations of CD163/HCP-1A-EmGFP and CD163/HCP-1B-EmGFP double-positive HEK293 cells were then obtained by FACS; the resulting cells were passaged at least five times before performing experiments.
The pEGFPNI fusion constructs containing the full-length sequence of human divalent metal transporter DMT-1A [iron-responsive element (IRE) form] or DMT-1B [non-IRE (n-IRE) form], respectively, were a kind gift from Mitsuaki Tabuchi, Yamaguchi University (Japan) [23 ]. Subconfluent CD163+-HEK293 cells were transfected with linearized vector using Lipofectamine 2000, and selection was performed using 1 mg/mL geneticin and 12 µg/mL blasticidin to obtain stable, double-transfected cell lines. After FACS, uniform DMT-1A-enhanced GFP (EGFP)-expressing and DMT-1B-EGFP-expressing CD163+-HEK293 cell lines were obtained.
Immunofluorescence and subcellular localization of HCP-1-EmGFP variants and DMT-1A/B-EGFP
For immunofluorescence microscopy, transfected HEK293 cells were cultured on baked (220°C, 4 h), endotoxin-free, sterile, round, 12-mm glass coverslips (Hecht-Assistent, Germany), pretreated with 6.25 µg/cm2 poly-D-Lysine Hydro-bromide (BD Biosciences, San Jose, CA, USA) in 24-cluster wells. For colocalization studies, cells were incubated with fluorescent Hb-Hp complexes or transferrin and washed three times with PBS, pH 7.4 (Sigma Chemical Co., St. Louis, MO, USA), before fixation. Hp phenotype 1-1 (Sigma Chemical Co.) was labeled using the Alexa594 protein-labeling kit (Molecular Probes, Eugene, OR, USA), and Hb-Hp complexes were generated by combining Hb and fluorescent Hp at a 1:1 molar ratio 10 min before experimentation. Fluorescent Alexa594-labeled transferrin (Molecular Probes) was used at a concentration of 20 µg/mL. All uptake assays were performed in cell culture medium containing 10% FCS. Fixation was performed with 2.5% paraformaldehyde in PBS, pH 7.4, for 15 min, and cells were subsequently washed twice with PBS before permeabilization for 5 min with 0.1% Triton X-100 (Sigma Chemical Co.) in PBS at room temperature. After washing, nonspecific binding sites were blocked for 1 h at room temperature with 10% goat serum in PBS, supplemented with 1% BSA (Sigma Chemical Co.). Staining of lysosomal-associated membrane protein 1 was performed by incubation with mouse anti-lysosome-associated membrane protein 1 (LAMP-1) antibody (RDI Research Diagnostics, Concord, MA, USA) for 1 h at room temperature in a dilution of 1:800 in PBS, supplemented with 1% goat serum and 0.1% BSA. Alexa Fluor 568 goat anti-mouse IgG (Molecular Probes) was used as a secondary antibody, and nuclear counterstain was performed using 10 µg/mL 4,6-diamidino-2-phenylindole (DAPI; Sigma Chemical Co.). After three washes in PBS, coverslips were mounted with ProLong Gold antifade reagent-mounting medium (Molecular Probes).
Samples were analyzed as 0.2 µm-thick optical sections using a Leica SP2 AOBS UV confocal laser-scanning microscopy system (Leica, Heidelberg, Germany) with an original optical magnification of 630x. Images were imported into Adobe Photoshop software (Adobe Systems, San Jose, CA, USA). The quantitative degree of colocalization was calculated using the colocalization module in Axiovision software, Version 4.6.3. (Carl Zeiss, Feldbach, Switzerland). A colocalization coefficient was calculated for the red channel fluorescence, which indicates the ratio of red channel pixels (i.e., Hb-Hp) colocalized with green channel pixels (i.e., DMT-1-EGFP) to the total number of red fluorescence channel pixels.
RNA isolation and quantitative real-time RT-PCR
Total cellular RNA from human blood-derived macrophages was isolated using the RNeasy mini kit (Qiagen, Basel, Switzerland), according to the manufacturer’s instructions, including a DNase I digestion step (Qiagen). RNA was quantified spectrophotometrically using a Nanodrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA), and equal amounts of total RNA (400 ng total DNA-free RNA of each cellular preparation) were reverse-transcribed into cDNA using StrataScript First-Strand synthesis system (Stratagene, Rotkreuz, Switzerland), according to the manufacturer’s instructions. Duplicates of cDNA samples were amplified by RT-PCR using the LightCyclerTM real-time PCR system with FastStart DNA Master Plus SYBR Green I kit (both purchased from Roche Diagnostics) and sequence-specific primer pairs for HCP-1 (forward 5'-AAC TCA CTC TAC CCA GCC ACT C-3', reverse 5'-ATC AGC CTT TTC CAG CAT CCC-3'), CD163 (forward 5'-ACA TAG ATC ATG CAT CTG TCA TTT G-3', reverse 5'-CAT TCT CCT TGG AAT CTC ACT TCT A-3'), ferroportin (forward 5'-CAG GAG AAG ACA GAA GCA AAC-3', reverse 5'-TAA AGC CAC AGC CGA TGA C-3'), DMT-1 (forward 5'-AAC GAG CAG GTG GTT GAA G-3', reverse 5'-AAG CAG AGT GGG GAT GAT GG-3'), and GAPDH (forward 5'-AAC AGC GAC ACC CAC TCC TC-3', reverse 5'-GGA GGG GAG ATT CAG TGT GGT-3'). Primers were designed using Primer Selection software SeqWeb, Version 2.1 (Accelrys Software Inc., Cambridge, UK). Melting curve analysis was performed simultaneously in each PCR experiment to detect primer-dimer formation and the specificity of each amplicon. Temperature cycling profiles were as follows: 10 min at 95°C, 45 cycles of 15 s at 95°C, 10 s at 67–55°C, and 13 s at 72°C. For ferroportin primers, the annealing temperature was 60°C, and extension time was 7 s, which resulted in a clear and single peak in the melting curve analysis. Real-time PCR results were analyzed with Light Cycler3 Analysis software, Version 3.5 (Roche Diagnostics). Relative mRNA levels were quantified with the use of external standard curves that were included in each assay. Expression levels of HCP-1, CD163, ferroportin, and DMT-1 were normalized to GAPDH levels in each experimental sample.
Changes in mRNA expression were analyzed for significance by ANOVA and Bonferroni-corrected post-test where appropriate.
SDS-PAGE and Western blot analysis
Total cellular protein was extracted from subconfluent human blood-derived macrophages, grown in six-well plates, using 300 µL/well CelLytic-M reagent (Sigma Chemical Co.), supplemented with Complete Mini Protease Inhibitor (Roche Diagnostics). After three freeze-thaw cycles and sonication with a Branson Sonifier 250, cellular debris was removed by centrifugation at 16,000 g for 15 min. Protein concentrations of each sample were determined using a Protein Bradford assay (Bio-Rad, Hercules, CA, USA). A reducing SDS-PAGE was loaded with 40 µg total protein/well for subsequent Western blot analysis of HCP-1 and ferroportin and a nonreducing SDS-PAGE with 20 µg total protein/well for CD163 using a Criterion Precast Tris-HCl gel (Bio-Rad). Proteins were transferred to polyvinylidene difluoride membranes (Bio-Rad), which were blocked by incubating for 1 h at room temperature in 10% goat serum (Gibco Europe) and 1% BSA in PBS on an orbital shaker. Primary antibodies directed against HCP-1 (rabbit anti-human HCP-1, Abcam Inc., Cambridge, MA, USA; 1:500), ferroportin (rabbit anti-mouse ferroportin, not distinguishing between mouse and human isoforms, Alpha Diagnostic International, San Antonio, TX, USA; 1:1000), and CD163 (mouse anti-human CD163, clone 5C6-FAT, BMA, Augst, Switzerland; 1:1000) were used for overnight incubation at 4°C, diluted in PBS, supplemented with 1% goat serum, 0.1% BSA, and 0.1% Tween 20. Membranes were washed in PBS with 0.1% Tween 20 (Merck, Dietikon, Switzerland), and appropriate HRP-conjugated secondary antibodies (Amersham Biosciences, Piscataway, NJ, USA) were used in a dilution of 1:10,000. Blots were developed with ECL Plus Western blotting detection reagent (Amersham Biosciences) and analyzed on a Chemi Doc XRS system with Quantity One 1-D Analysis software, Version 4.5.0 (Bio-Rad).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (28K): [in a new window] |
Figure 1. Two HCP-1 splice variants with different C-terminal membrane topology are expressed in human macrophages. (A) The full-length mRNA transcript of HCP-1 amplified from human monocyte-derived macrophages (Lane B) was compared with the amplification product from a plasmid encoding the shorter of the two predicted splice variants (Lane A). The longer variant appears to be the primary transcript expressed in macrophages. (B) Amplification of macrophage cDNA using intron-spanning primers revealed PCR products representing both predicted splice variants. The larger variant constitutes the predominantly expressed HCP-1 transcript. Schematic presentation of the predicted membrane topology of the two variants shows that the lack of one transmembrane domain in the shorter variant causes loss of one transmembrane domain (red) and reversal of the C-terminal protein orientation (blue).
|
 View larger version (26K): [in a new window] |
Figure 4. HCP-1, CD163, DMT-1, and ferroportin are regulated by inflammatory macrophage stimulation. Human monocyte-derived macrophages were treated with or without endotoxin (black bars, 10 ng/mL), IFN- (open bars; 100 IU/mL), or both (shaded bars) for 12 h (A) or 48 h (B). HCP-1, CD163, DMT-1, and ferroportin mRNAs were quantified by quantitative RT-PCR. Results are presented as the relative quantity of specific mRNA in relation to untreated cells. mRNA were normalized to GAPDH. Means ± SD of at least six independent experiments per treatment and time-point are given. *, Significant changes of gene expression (P<0.05). (C) Western blot of HCP-1 protein levels in human monocyte-derived macrophages. Cells were treated with or without LPS for 24–120 h.
|
50% of the enocytosed Hb-Hp colocalized with HCP-1A-EmGFP (Fig. 2A)
. The intracellular HCP-1A-EmGFP-positive compartment colocalized further with fluorescent transferrin, a well-established marker of the early endosomal compartment (Fig. 2B)
. Accordingly, inhibition of receptor recycling between early endosomes and the cell surface by monensin resulted in accumulation of HCP-1A-EmGFP and transferrin in coarse, intracellular vesicles. The granular intracellular HCP-1A-EmGFP did not colocalize within LAMP-1-positive lysosomes (not shown).
 View larger version (33K): [in a new window] |
Figure 2. HCP1-EmGFP colocalizes with the Hb-Hp endocytic pathway in the early endosomal compartment. (A) HEK293 cells were transfected with HCP-1A-EmGFP and CD163. Double-transfected cells were incubated with fluorescent Hb-HpAlexa594 for 4 min. HCP-1A-EmGFP is expressed at the cell surface and colocalizes with endocytosed Hb-Hp complexes (yellow in the merged image). The colocalization coefficient was calculated for the red channel fluorescence [coloc. coeff, coloc/(coloc+Alexa594)] and is given in the right upper corner of each image. The false color image exemplifies the assignment of pixel to green or red channel fluorescence after background subtraction. Hpf, Fluorescent Hp. (B) HCP-1A-EmGFP also colocalizes with Alexa594-labeled transferrin, which is a marker for the early endosomal compartment (yellow in the merged image). The colocalization is more evident after transferrin incubation in the presence of monensin, which inhibits recycling of the transferrin receptor to the cell surface and thus, causes accumulation of HCP-1A and transferrin in coarse perinuclear vesicles. (C) HCP-1B does not colocalize with transferrin but shows a diffuse distribution within the cytoplasm. Original optical magnification, 630x. In all images, DAPI-stained nuclei are blue.
|
Hb-Hp traffics through the HCP-1/DMT-1B-positive endosomal compartment to DMT-1A-positive lysosomes
We coexpressed CD163 with EGFP fusion constructs of the two major splice variants of DMT-1, which are targeted to early endosomes (DMT-1B) and late endosomes/lysosomes (DMT-1A) [23
]. Both were expressed in human macrophages, as demonstrated by splice variant-specific RT-PCR (Fig. 3 A
). At an early time-point after endocytosis (2–4 min after the start of incubation), a large proportion of fluorescent Hb-Hp was detected in small, peripheral vesicles that were positive for the early endosomal DMT-1B and transferrin (Fig. 3B
and 3C)
. In contrast, incubation for 30 min after endocytosis revealed fluorescent Hb-Hp in larger vesicles that were negative for the DMT-1B and transferrin but strongly positive for lysosomal DMT-1A and LAMP-1 (Fig. 3B
and 3C)
.
 View larger version (43K): [in a new window] |
Figure 3. Convergence of cellular Hb and iron uptake pathways: Hb-Hp traffics trough the DMT-1B-positive endosomal compartment into the DMT-1A-positive lysosomes. (A) Using splice variant-specific primer pairs, both splice variants of DMT-1 were detected in human monocyte-derived macrophages (PCR 1A/B is not variant-specific; PCR 1A amplifies only variant DMT-1A; PCR 1B only DMT-1B). (B) CD163-expressing HEK293 cells were transfected with plasmids encoding EGFP fusion proteins of the two variants of DMT-1. DMT-1B localizes to early endosomes and DMT-1A to lysosomes. After 4 min of incubation, fluorescent Hb-Hp colocalizes exclusively with the endosomal isoform DMT-1B (yellow shows colocalization). When cells were incubated in the absence of fluorescent ligand for 30 min (after initial incubation with fluorescent Hb-Hp for 15 min), the Hb-Hp complexes were detected in the DMT-1A-positive lysosomal compartment. The mutually exclusive localization of the two splice variants of DMT-1 into endosomes and lysosomes was confirmed by colocalization of the two DMT-1 variants with the endosomal marker transferrin (Tf; DMT-1B) and the lysosomal protein LAMP-1 (DMT-1A), respectively (C). No colocalization was observed of the DMT-1A variant with fluorescent transferrin or the DMT-1B variant with LAMP-1, respectively. Original optical magnification, 630x. In all images, DAPI-stained nuclei are blue. Colocalization coefficients for the red channel fluorescence are indicated in the right upper corner of each image.
|
. After 12 and 48 h, we observed down-regulation of HCP-1 and ferroportin mRNA by LPS. Combined treatment with LPS/IFN- produced a greater effect (Fig. 4A
and 4B
), and the effect of IFN- alone was weaker. CD163 mRNA was also suppressed by IFN- at 12 and 48 h. However, in contrast to HCP-1 and ferroportin, CD163 mRNA was induced after 48 h treatment with LPS or LPS/IFN-. In accordance with earlier reports, we observed a slight induction of DMT-1 mRNA after 12 h treatment with all stimuli. However, this response was transient, and mRNA levels returned to basal after 36–48 h. The suppression of HCP-1 protein expression by LPS was confirmed by Western blot (Fig. 4C)
. Stimulation of macrophages with specific agonists of TLR1–9 suppressed HCP-1 expression. Incubation with TLR1, -2, -4, -5, -8, and -9 agonists for 36 h decreased HCP-1 expression by up to 90% to levels that were barely detectable. Overall, the effect of TLR agonists on HCP-1 expression was similar to that observed on ferroportin expression (Fig. 5 ).
 View larger version (29K): [in a new window] |
Figure 5. HCP-1, CD163, DMT-1, and ferroportin mRNA levels following treatment with TLR agonists. Human monocyte-derived macrophages were incubated with or without the indicated TLR agonists for 12 h or 36 h, respectively, and mRNA expression levels were measured by real-time PCR. mRNA levels were normalized to GAPDH. Means ± SD of at least six independent experiments per treatment and time-point are given. *, Significant changes of gene expression (P<0.05). Specificity of TLR ligands: ODN, Oligodeoxynucleotide; CpG oligonucleotides—TLR9; ssRNA—TLR8; Imiquimod—TLR7; FSL-1, fibroblast-stimulating lipopeptide 1—TLR6/2: flagellin—TLR5; LPS, endotoxin—TLR4; poly (I:C), polyinosinic:polycytidylic acid; dsRNA—TLR3; HKLM, heat-killed Listeria monocytogenes—TLR2; Pam3CSK4, N-palmitoyl-S-[2,3-bis(palmitoloxy)-(2RS)-propyl]-Cys-Ser-Lys(4); synthetic lipopeptide—TLR1/2.
|
 View larger version (27K): [in a new window] |
Figure 6. HCP-1, CD163, DMT-1, and ferroportin mRNA expression is modified by GC treatment. (A) Human monocyte-derived macrophages were incubated for 12 h, with or without dexamethasone (Dex, Dx; 2.5x10–7 M), heme (h), or Hb. HCP-1, CD163, DMT-1, and ferroportin mRNAs were measured using quantitative RT-PCR. Results are presented as the relative quantity of specific mRNA compared with untreated cells. mRNA abundance was corrected for differences in GAPDH abundances. Means ± SD of at least six independent experiments per treatment and time-point are given. *, Significant changes of gene expression (P<0.05). (B) Incubation of human macrophages with dexamethasone for 24 h resulted in a marked increase of CD163, HCP-1, and ferroportin protein when analyzed by Western blot (C, nontreated samples; Dex, dexamethasone treatment at 2.5x10–7 M for 24 h).
|
 View larger version (23K): [in a new window] |
Figure 7. Effects of endotoxin and GC treatment on HCP-1, CD163, and ferroportin mRNA expression. Human monocyte-derived macrophages were incubated with LPS at 10 ng/mL or 100 ng/mL in the presence (Dex) or absence (C) of 2.5 x 10–7 M dexamethasone for 24 h. mRNA levels were normalized to GAPDH. Means ± SD of three independent experiments are shown. *, Significant changes of gene expression (P<0.05).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
After demonstrating that HCP-1 is expressed during the in vitro differentiation of PBMC-derived macrophages, we aimed to elucidate the spatial relationship between HCP-1 and the Hb-endocytic pathway. Our assumption that HCP-1 could constitute an intracellular heme carrier (which transports heme out of the Hb-containing endosome) was based on the work of Shayeghi et al. [7] that described localization of HCP-1 in—not further characterized—intracellular vesicles in duodenal enterocytes. In addition, HCP-1 harbors a dileucine-based internalization motif within the second intracytoplasmic loop. As the available antibodies were not suitable for specific investigation of the subcellular HCP-1 localization in macrophages, we expressed EGFP fusion proteins of the two main HCP-1 splice variants found in human macrophages. A significant fraction of the longer and more abundant splice variant was detected in the early endosomal compartment, where it colocalizes with the early endosomal marker transferrin and also with endocytosed Hb-Hp in CD163-positive cells. This suggests that HCP-1 transports Hb-derived heme out of the endosome into the cytoplasm, where it is targeted to the microsomal enzyme HO-1 for degradation into carbon monoxide, bilirubin, and iron. The low-abundance splice variant of HCP-1 (HCP-1B), which lacks one transmembrane domain and is thus predicted to display altered membrane topology within its C-terminus, is not detected within vesicular organelles but is distributed diffusely throughout the cytoplasm. This unexpected finding underscores the critical role of the C-terminal amino acid sequence for the subcellular localization of HCP-1. It remains to be seen whether HCP-1B has a specific functional role or whether it represents a nonfunctional splicing "accident."
Along with heme, free iron is another component of Hb, which can be released into the endocytic pathway. A specific transporter is required to transfer free iron to the cytoplasm, where it is bound to the iron-storage protein ferritin or exported by the iron exporter ferroportin. We analyzed the spatial relationship of the Hb-endocytic pathway with the endosomal/lysosomal iron exporter DMT-1. We found the two major DMT-1 splice variants, which display distinct, subcellular localizations to be expressed in macrophages. DMT-1A, which contains an IRE within its nontranslated mRNA, is directed to lysosomes, and DMT-1B locates to early endosomes. As shown here, Hb passes the DMT-1B/transferrin-positive endosomal compartment on its route from the cell surface to the DMT-1A/LAMP-1-positive lysosomes. The functional role of DMT-1 in heme-iron handling after Hb endocytosis is unknown. However, under specific conditions such as after oxidation of Hb by hydrogen peroxide or at low pH in the presence of proteases, nonenzymatic heme breakdown and free iron release have been reported [30 31 ]. It might thus be speculated that some nonenzymatic heme decomposition can also occur within the lysosme of macrophages. DMT-1A might therefore have a role in the export of this nonenzymatically released heme-iron from the lysosomal lumen into the cytoplasm.
The sequential colocalization of endocytosed Hb with different heme- and iron-transport proteins suggests the organization of molecular pathways involved in the macrophage heme-iron recycling from Hb, as proposed in Figure 8 . Dysregulation of macrophage iron recycling is a major pathogenic mechanism of anemia associated with chronic disease, one of the most frequent causes of anemia under conditions of chronic immune activation [11 ]. Sequestration of iron within macrophages of the reticuloendothelial system is a major pathogenic mechanism, assumed to be a consequence of enhanced iron acquisition and suppressed iron release. Known pathways of increased iron uptake include enhanced erythrophagocytic activity, increased uptake of transferrin iron, and increased expression of DMT-1. In contrast, transcriptional and post-translational suppression of ferroportin expression is the major mechanism of impaired iron release [32 ].
 View larger version (53K): [in a new window] |
Figure 8. Proposed model of the Hb/heme-iron pathway in macrophages. Hb released from lysed erythrocytes dimerizes and is subsequently endocytosed by the Hb scavenger receptor CD163. Upon acidification of the endosomal lumen, Hb is released from CD163. The receptor recycles back to the cell surface to undergo further cycles of endocytosis. Heme is released from globin in the acidic environment and is exported to the cytoplasm by the heme transporter HCP-1. The remaining intact Hb as well as (heme-free) globin is transported to the lysosome, where globin is degraded. In the oxidant, acidic, and proteolytic environment of the lysosome, some iron (Fe) is supposed to be released from heme by HO-1-independent heme degradation and subsequently, transported to the cytoplasm by the iron transporter DMT-1A. Within the cytoplasm, iron is bound by the storage protein ferritin or exported to the plasma by ferroportin. The identity of the lysosomal heme exporter remains unknown.
|
suppressed HCP-1 mRNA to barely detectable levels, supporting the hypothesis that inflammation disrupts iron homeostasis in macrophages also via down-regulation of HCP-1. Furthermore, activation of TLR 1–9 altered the expression of several genes involved in Hb-iron recycling, the most profound effect being on HCP-1 and ferroportin mRNA after treatment with TLR1, -2, -4, -5, -8, and -9 agonists. This uniform response to different inflammatory stimuli may explain why dysregulation of iron homeostasis and subsequent anemia of chronic disease represent such a ubiquitous phenomenon in human disease. Further studies are needed to define the consequences of HCP-1 down-regulation by inflammatory macrophage activation. However, one might speculate that suppression of HCP-1 could impair the transport of heme from extracellular sites or from Hb-heme, which is endocytosed by CD163, to the microsomal heme oxygenase, which finally releases free iron for subsequent export by ferroportin. This effect of inflammation on heme catabolism may aggravate the overall impairment of iron recycling from macrophages, which is caused by sequestration of iron within macrophages as a consequence of ferroportin down-regulation and increased iron storage/immobilization by ferritin. GC are currently the most widely used anti-inflammatory agents in clinical practice. Increased expression of the Hb scavenger receptor CD163 has been observed following GC treatment in vitro and in vivo. We investigated whether other members of the macrophage Hb-iron recycling pathway studied here are also targets of GC activity and whether GC could mitigate the inflammation-induced dysregulation of iron homeostasis. HCP-1 and CD163 expression increased more than tenfold upon macrophage treatment with GC, suggesting an essential role for GC in the regulation of Hb/heme-iron recycling. That GC treatment might allay anemia associated with chronic disease—not only through the multiple anti-inflammatory activities of GC but also through more specific modulation of the Hb-iron recycling pathway—may point to a novel activity on the part of this class of drugs. This hypothesis is supported further by the finding that dexamethasone treatment alleviated the endotoxin-mediated suppression of macrophage HCP-1 and even reversed that of CD163. In our experiments, Hb and heme did not increase expression of ferroportin significantly, a finding that is somewhat counterintuitive and contradictive to experimental findings published in other work [33 ]. However, this might well be related to specific experimental settings and to the fact that the expression of ferroportin is extremely sensitive to suppression by inflammatory macrophage activation. Heme and Hb have been accounted to exert inflammatory activities that are mediated partially by the TLR4 ligand activity of heme [34 ]. The expression level of ferroportin might thus represent a balance of positive signals, mainly imparted by the ferroportin-inducing activity of iron, and negative signals on gene expression. Similar mechanisms might also be responsible for the observation that HCP-1 is suppressed somewhat by Hb and heme.
Further investigation of the Hb-heme-iron recycling pathway in macrophages may reveal novel, therapeutic strategies aimed at relieving the detrimental consequences of anemia associated with chronic disease.
Received April 16, 2007; revised August 20, 2007; accepted September 14, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and tumor necrosis factor 
induce hypoferraemia, ferritin production and a decrease in circulating transferrin receptors in cancer patients Eur. J. Clin. Invest. 28,520-527[CrossRef][Medline] J. Clin. Invest. 91,969-976[Medline]This article has been cited by other articles:
 |
|
 |
|
  P. W. Buehler, B. Abraham, F. Vallelian, C. Linnemayr, C. P. Pereira, J. F. Cipollo, Y. Jia, M. Mikolajczyk, F. S. Boretti, G. Schoedon, et al. Haptoglobin preserves the CD163 hemoglobin scavenger pathway by shielding hemoglobin from peroxidative modification Blood, March 12, 2009; 113(11): 2578 - 2586. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Zhao, S. H. Min, Y. Wang, E. Campanella, P. S. Low, and I. D. Goldman A Role for the Proton-coupled Folate Transporter (PCFT-SLC46A1) in Folate Receptor-mediated Endocytosis J. Biol. Chem., February 13, 2009; 284(7): 4267 - 4274. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
