* Department of Physiology, Dartmouth Medical School, Lebanon, New Hampshire;
Department of Basic Medical Science, Shock Trauma Research Center, University of Missouri Kansas City; and
Department of Immunology, University of Maryland Baltimore
1 Correspondence: Department of Physiology, HB7700, Dartmouth Medical School, 1 Medical Center Drive, Lebanon, NH 03756. E-mail: Lehn{at}Dartmouth.edu
ABSTRACT
The hemoglobin scavenger receptor (HbSR) CD163 is a monocyte/macrophage-specific glycoprotein that binds and facilitates uptake of haptoglobin-hemoglobin (Hp-Hb) complexes, which are rapidly formed in the circulation upon hemolysis of red blood cells. Hemolysis can be caused by a diverse range of infectious agents and provides pathogens a source of iron to enhance their survival and replication. Previous work demonstrated that lipopolysaccharide (LPS) activates monocytes to cleave cell-bound HbSR into a soluble mediator that retains the capacity to bind Hp-Hb complexes. We report that blocking LPS activation of Toll-like receptor 4 prevents LPS-mediated shedding of CD163. Furthermore, activation of two other cell surface Toll-like receptors (TLR), TLR2 and TLR5, induces shedding of the HbSR from human monocytes. In contrast, treatment of monocytes with intracellular TLR3, TLR7, and TLR9 agonists failed to cause HbSR shedding, suggesting that this shedding event is selective to cell surface TLR activation. These data demonstrate that the soluble HbSR is released from monocytic cells in response to TLR signaling as an acute innate immune response to extracellular pathogen infections.
Key Words: monocyte • macrophage • innate immunity • inflammation
INTRODUCTION
Recent studies have shown that CD163, a cell surface glycoprotein, expressed exclusively by monocytes and macrophages, can bind and internalize haptoglobin-hemoglobin (Hp-Hb) complexes [1
, 2
]. This implicates monocytes and macrophages as key physiologic regulators of free hemoglobin clearance and CD163 as a primary hemoglobin scavenger receptor (HbSR) [1
]. Hemoglobin clearance is important for host survival for at least two reasons. First, hemoglobin internalization prevents excessive toxicity by removing redox-reactive iron from the circulation and allows iron and the biologically important heme molecule to be recycled. Second, clearance removes hemoglobin as a potential source of iron, a normally limiting nutrient for pathogen growth. Host mediators of immune functions are also known to regulate surface expression of CD163. Anti-inflammatory mediators such as glucocorticoids and interleukin-10 markedly induce CD163 expression, and proinflammatory cytokines such as tumor necrosis factor-
(TNF-) and interferon-
(IFN-) suppress CD163 expression [3
, 4
].
CD163 has also been identified as a soluble protein in cell culture supernatants and in human plasma [5 , 6 ]. Sepsis, as well as certain Gram-negative and Gram-positive infections, has been correlated with increased soluble CD163 levels [6 7 8 ]. Soluble CD163 appears to be generated primarily via shedding of the extracellular domain of the cell surface glycoprotein. CD163 shedding has been demonstrated upon treatment of human monocytes with phorbol 12-myristate 13-acetate (PMA), lipopolysaccharide (LPS), or Fc receptor cross-linking ex vivo [9 , 10 ]. Furthermore, in vivo studies have shown that human experimental endotoxemia is associated with reduced monocyte surface CD163 and increased soluble CD163 in plasma [11 ].
LPS and other pathogen-associated molecular patterns (PAMPs) are molecules containing specific structures, conserved among microbial species and known to activate pattern recognition receptors, including Toll-like receptors (TLRs), on host immune cells [12 ]. TLRs enable innate immune cells to be activated by a diverse range of microbes without the need to express specific receptors for recognition of each new pathogen encountered. TLR2, TLR4, and TLR5 are expressed on the cell surface of monocytes, where they can be activated by PAMPs expressed on the surface of pathogens [12 ]. In contrast, TLR3, TLR7, and TLR9 are expressed within intracellular compartments of certain cells involved in host defense, where they are activated by specific nucleic acid sequences/structures derived from viruses or bacteria [12 ]. Although LPS activation of monocytes via TLR4 is believed to be responsible for CD163 shedding in response to Gram-negative infections and LPS administration [11 ], this assumption has not been tested rigorously using pure LPS. Moreover, the microbial products and receptors that underlie CD163 shedding during Gram-positive infections have yet to be elucidated.
In this report, we show that the HbSR is cleaved from the surface of human monocytes in response to PAMPs that activate cell surface TLRs but not in response to intracellular TLR agonists. Our findings define a dichotomy between intracellular and extracellular TLRs, which may provide new insights into the importance of CD163 beyond its role in the clearance of Hp-Hb complexes. As soluble CD163 has been shown to bind Hp-Hb complexes [13 ], this cleavage product of the cell-surface HbSR may be a physiologically relevant component of the innate immune response. Considered in this context, it will be important to determine how different macrophage activation signals, through effects on soluble and cell surface CD163, influence the availability of hemoglobin iron for use by pathogens that are dependent on iron for growth.
MATERIALS AND METHODS
Soluble CD163 and IFN- enzyme-linked immunosorbent assays (ELISAs)
Soluble CD163 was detected in cell culture supernatants using a previously described ELISA [5
]. Briefly, monoclonal antibody (mAb) Mac 2-158 [3
] (R&D Systems, Minneapolis, MN) was used for CD163 capture, and biotinylated R20 (Maine Biotechnology Services, Portland) was used as the detection antibody. A standard curve was generated for each plate based on the titration of a reference plasma standard, previously determined to have 18.5 ± 0.19 µg/mL soluble CD163 [11
]. IFN- concentrations were determined using an IFN- sandwich ELISA (R&D Systems) following the manufacturer’s protocol.
In vitro culture of mononuclear cells
Monocytes were purified from mononuclear cell fractions from leukopheresis donors as described by Mentzer et al. [14
]. Peripheral blood mononuclear cells (PBMCs) were isolated from heparinized whole blood by using Ficoll-Hypaque (d=1.077) [15
]. Purified monocytes or PBMCs were plated at a density of 2–2.5 x 106 cells/mL in 96-well plates. The cells were cultured in 5% CO2 at 37°C in complete medium consisting of HEPES-buffered RPMI 1640 (Hazelton Biologicals, Lenexa, KS), 20 µg/ml gentamicin sulfate (Elkins-Sinn, Cherry Hill, NJ), and 10% fetal bovine serum (Hyclone Laboratories, Logan, UT) for 
12 h prior to treatment.
CD163 shedding assays
Monocytes or PBMCs were cultured as described above and subsequently treated for 1–48 h as indicated with medium alone, the indicated concentrations of phenol-water-extracted (protein-free) Escherichia coli K235 LPS (TLR4 agonist) [16
, 17
], 10 ng/ml S-[2,3-bis(palmitoyloxy)-(2-RS)-propyl]-N-palmitoyl-(R)-Cys-(S)-Ser-Lys4-OH, trihydrochloride (Pam3Cys; EMC Microcollections GmbH, Tübingen, Germany; TLR2/1 agonist), 100 µg/mL zymosan (Sigma-Aldrich, St. Louis, MO; TLR2/6 agonist), 20 µg/ml polyinosinic:polycytidylic acid [poly(I:C); TLR3 agonist, Invivogen, San Diego, CA], 200 ng/mL Salmonella typhimurium flagellin (TLR5 agonist, Invivogen), 10 µg/mL imiquimod (TLR7 agonist, Invivogen), 1 µg/mL E. coli DNA (TLR9 agonist, Invivogen), 2 µM 2006 CpG DNA oligo (5'-TCGTCGTTTTCGTTTTGTCGT-3', TIB MolBio, Germany; TLR9 agonist), or 2 µM 2006-Control GpC DNA oligo (5'-TGCTGCTTTTGCTTTTGTGCT-3', TIB MolBio). S. typhimurium flagellin and protein-free LPS were digested as described by Bambou et al. [18
] with slight modifications. Briefly, S. typhimurium flagellin was added to pepsin (0.15 mg/ml, pH 1.8, Sigma-Aldrich) for 4 h at 37°C. After adjusting the pH to 7.8, trypsin (0.15 mg/ml, Sigma-Aldrich) was added to the solution for 12 h at 37°C. Digested flagellin was boiled for 15 min to inactivate the proteases.
After treatment, cells were spun at 450 g for 1 min. Supernatants were collected and analyzed immediately or frozen at –80°C for analysis of soluble CD163 and/or IFN- levels. Cells were resuspended and stained for CD163 expression with purified mouse immunoglobulin G1 (IgG1) mAb Mac 2-158 (anti-CD163, R&D Systems) or an IgG1 isotype control (Caltag Laboratories, Burlingame, CA), followed by a goat anti-mouse F(ab')2 fluorescein isothiocyanate secondary antibody. Cells were stained for CD86 with purified mouse IgG1 monoclonal anti-CD86 phycoerythrin (PE)-conjugated antibody (Ancell Corporation, Bayport, MN) or an IgG1 isotype control PE-conjugated antibody (PharMingen, San Jose, CA). Cells were then analyzed by flow cytometry on a FACScan flow cytometer, and monocytes were gated on forward/side-scatter (FSC/SSC). Mean fluorescent intensity (MFI) was determined by the geometric mean of the fluorescence of gated monocytes.
Inhibition of LPS-mediated CD163 shedding
To inhibit TLR4-mediated signaling, TLR agonists were preincubated with 15 µg/mL polymyxin B sulfate (PMX; Gibco, Grand Island, NY) or 100 ng/mL diphosphoryl lipid A (100-fold higher than the LPS concentration used) from Rhodobacter sphaeroides (RsDPLA) [19
] for 30 min at room temperature prior to addition to cells. The inhibitors remained in culture throughout the duration of the experiments.
Limulus amoebocyte lysate (LAL) assay for detection of LPS contamination
The QCL-1000® Chromogenic LAL Endpoint assay (Cambrex, Cottonwood, AZ) was used to detect endotoxin contamination of all TLR agonists used in the above-mentioned experiments following the manufacturer’s protocol. Briefly, protein-free, phenol-water-extracted E. coli K235 LPS was used to create a linear standard curve from 10 to 100 pg/ml. TLR agonists were serially diluted to determine a concentration of agonist within the standard curve of the assay. Subsequent LAL reactions were performed at multiple concentrations of agonist within the linear range of the standard curve to rule out product inhibition of the LAL reaction. Values from greater than or equal to three LAL reactions per agonist were averaged and adjusted to concentrations of agonist used in CD163 shedding assays.
Statistical analysis
Two-way ANOVA was performed on each group followed by a Bonferroni post-test analysis. A P value <0.05 was considered statistically significant. Experimental values were expressed as means ± SD and are representative of greater than or equal to three separate experiments from greater than or equal to three donors. Spearman correlation analysis was used to assess the relation between the level of surface CD163 on monocytes prior to addition of TLR agonists and cell culture supernatant levels of soluble CD163 following extracellular TLR activation. Correlation coefficients with a probability less than 0.05 were considered statistically significant. All statistics and graphs were performed using GraphPad Prism Version 4.00 for Windows (GraphPad Software, San Diego, CA).
RESULTS
Phenol-water-extracted LPS induces shedding of CD163
Previous reports have demonstrated that TLR2 and TLR4 are activated when crude LPS preparations are used to treat cells. However, repurification of LPS preparations using a modified phenol re-extraction protocol eliminated signaling through TLR2, resulting in a LPS preparation that signals only through TLR4 [17
]. To determine whether the previously observed LPS-induced shedding of CD163 [11
] was a result of TLR4 activation (or a contaminating bacterial component that signals through a different TLR), human monocytes were treated with varying concentrations of phenol-water-extracted LPS (protein-free). Treatment of monocytes with phenol-water-extracted LPS preparations resulted in decreased surface expression of CD163 and increased cell culture supernatant levels of soluble CD163 within 1 h of treatment (Fig. 1
). The acute shedding of CD163 occurred with as little as 10 pg/mL protein-free LPS and plateaued at 100 pg/mL protein-free LPS, indicating that TLR4 signaling alone was able to induce CD163 shedding.
 View larger version (10K): [in a new window] |
Figure 1. Protein-free LPS induces shedding of human monocyte CD163. Monocytes were treated with a dose titration of phenol-water-extracted, pure LPS for 1 h. (A) Supernatants from these cultures were collected and analyzed by a sandwich ELISA for soluble CD163. (B) Surface expression of CD163 was measured on purified monocytes following the 1-h treatment with LPS. Data shown are from a single donor and are one of five representative experiments using three different donors, and values are mean ± SD of triplicates. ***, P < 0.001, versus control.
|
 View larger version (24K): [in a new window] |
Figure 2. PMX and RsDPLA block shedding of human monocyte CD163 by protein-free LPS. Monocytes were treated for 1 h with 1 ng/mL protein-free LPS, which had been preincubated with or without 15 µg/mL PMX or 100 ng/mL RsDPLA for 30 min at room temperature. (A) Supernatants were analyzed by a sandwich ELISA for the release of soluble CD163. (B) Surface expression of CD163 was determined by flow cytometry, and the MFI of monocytes gated by FSC/SSC was recorded. Data shown are from a single donor and are one of five representative experiments using five different donors, and values are mean ± SD of triplicates. ***, P < 0.001, versus controls.
|
Contamination of TLR agonists with LPS
Historically, many researchers have not been concerned with picogram/milliliter levels of LPS contamination of reagents, as experimental readouts were not stringent enough to detect changes as a result of such low levels of LPS. However, Figure 1
shows the high sensitivity of CD163 shedding when monocytes are treated with as little as 10 pg/mL LPS. Because of the extreme sensitivity of monocytes to LPS, we tested all TLR agonists for LPS contamination as indicated in Materials and Methods. Table 1
demonstrates that LPS contamination of the TLR agonists used in subsequent experiments was below 1 ng/mL, a level that could be blocked by PMX or RsDPLA (Fig. 2)
.
|
View this table: [in a new window] |
Table 1. Contamination of TLR Agonists with LPS Determined by a LAL Assay
|
B [24
], and NF-B activation is common to all TLR signaling cascades [24
], leading to the initial prediction that stimulation via any TLR would lead to CD163 shedding. To test this hypothesis, human monocytes were treated with a panel of TLR agonists for 1 h and assessed for surface CD163 expression, and supernatants were collected and assayed for soluble CD163. Figure 3
demonstrates that the TLR agonists zymosan (TLR2/6), Pam3Cys (TLR2/1), and S. typhimurium flagellin (TLR5) decreased CD163 surface expression and concomitantly increased soluble CD163 levels in cell culture supernatants (Fig. 3)
. PMX or RsDPLA did not block the effect of these ligands, suggesting that contaminating LPS is unlikely to be the cause for CD163 shedding in these treatment groups (Fig. 3)
. Furthermore, protease digestion of S. typhimurium flagellin completely eliminated the effect of this TLR5 ligand on CD163 shedding in the presence of PMX or RsDPLA and had no effect on LPS-mediated shedding of CD163 (Fig. 3)
. Thus, selective activation of extracellular TLRs on human monocytes resulted in shedding of surface CD163.
 View larger version (34K): [in a new window] |
Figure 3. Cell surface TLR agonists activate shedding of human monocyte CD163 within 1 h. Purified human monocytes were treated for 1 h with TLR agonists: 1 ng/mL protein-free LPS, 100 µg/mL zymosan A, 10 ng/mL Pam3Cys, 200 ng/mL S. typhimurium flagellin, 200 ng/mL protease-digested S. typhimurium flagellin, and 1 ng/mL protease-digested LPS, with or without PMX or RsDPLA pretreatment. (A) Soluble CD163 release into culture supernatants was assessed by a sandwich ELISA. (B) Surface expression of CD163 was determined as described in Materials and Methods, and the MFI of monocytes gated by FSC/SSC was recorded. Data shown are from a single donor and are one of five representative experiments using five different donors, and values are mean ± SD of triplicates. *, P < 0.001, versus controls. (C) CD163-specific, surface MFI in untreated control samples was plotted against the amount of soluble CD163 detected in the cell culture supernatant following extracellular TLR-mediated shedding. Data shown include eight experiments testing a panel of extracellular TLR agonists for CD163 shedding using purified monocytes from five different donors.
|
 View larger version (32K): [in a new window] |
Figure 4. Intracellular TLR agonists do not activate the acute shedding of human monocyte CD163. (A) Purified human monocytes were treated for 1 h with TLR agonists: 1 ng/mL protein-free LPS, 20 µg/mL poly(I:C), 10 µg/mL imiquimod, 1 µg/mL E. coli DNA, and 2 µM CpG oligo, with or without PMX or RsDPLA pretreatment. Supernatants from these cultures were collected and analyzed by a sandwich ELISA for soluble CD163. (B) Monocyte surface expression of CD163 was determined by flow cytometry, and the MFI of monocytes gated by FSC/SSC was recorded. Data shown are from a single donor and are one of five representative experiments using five different donors, and values are mean ± SD of triplicates. ***, P < 0.001, or *, P < 0.01, versus controls.
|
PBMC activation of TLR3, TLR7, and TLR9 initiates cytokine production but not CD163 shedding
Although we found that a 1-h treatment of monocytes with intracellular TLR agonists failed to induce CD163 shedding, it is notable that several reported effects of TLR3, TLR7, and TLR9 activation were optimal following longer exposure to agonists of these receptors. To eliminate the possibility that intracellular TLR agonists need more than the previously tested 1 h for internalization, trafficking to the appropriate compartment, and signaling, PBMCs were treated for 12–48 h before assaying for CD163 shedding. Activation of TLR3 and TLR7 was confirmed by measuring IFN- production at 12 h, as previously reported [25
, 26
]. We chose to assay for IFN- production, as 1 ng/mL LPS did not stimulate detectable levels of IFN-, decreasing the likelihood that LPS contamination could account for the observed effect (Fig. 5A
). As CpG oligo 2006 was previously reported to be insufficient for production of IFN- [27
], up-regulation of CD86 on human monocytes was measured as a positive control for TLR9 activation after 48 h of treatment.
 View larger version (35K): [in a new window] |
Figure 5. Activation of TLR3 induces IFN- from PBMCs at 12 h. PBMCs were treated for 12 h with 1 ng/mL LPS or 20 µg/mL poly(I:C), which had been pretreated with or without 15 µg/mL PMX or 100 ng/mL RsDPLA. IFN- (A) and soluble CD163 levels (B) were analyzed in supernatants collected from PBMC cultures and quantified by sandwich ELISA. (C) Flow cytometry was used to determine surface expression of CD163, and the MFI of monocytes gated by FSC/SSC was recorded. Data shown are from a single donor and are one of three representative experiments using three different donors, and values represent means of duplicates. Values below the detection limit of the assay (12.5 pg/mL) are labeled as BD. ***, P < 0.001, or *, P < 0.05, versus controls.
|
production from PBMCs, an effect that was not blocked by PMX or RsDPLA. This indicates that production of IFN- cannot be attributed to contaminating LPS and suggests that it is mediated by poly(I:C) activation of TLR3. Although cell culture supernatant levels of soluble CD163 were increased, and surface expression of CD163 was decreased by 12-h treatment with poly(I:C) alone, these parameters were unchanged by poly(I:C) in the presence of PMX or RsDPLA (Fig. 5B
and 5C)
. Furthermore, LPS-free poly(I:C) obtained from Amersham Biosciences (UK) did not result in a decrease in surface expression of CD163 or an increase in soluble CD163 levels in cell culture supernatants, and it did induce IFN- production after 12 h of treatment (data not shown). This demonstrates that under conditions in which activation of TLR3 stimulates IFN- production from PBMCs, CD163 is not shed from monocytes.
To determine if activation of the intracellular TLR7 elicited similar responses, PBMCs were treated with imiquimod for 12 h. As with poly(I:C), imiquimod activated IFN- production from PBMCs (Fig. 6A
), and culture supernatants and cell surface levels of CD163 were unchanged (Fig. 6B
and 6C)
. Therefore, activation of TLR7 stimulates IFN- production from PBMCs without causing CD163 shedding from monocytes.
 View larger version (21K): [in a new window] |
Figure 6. Activation of TLR7 induces IFN- from PBMCs, which were treated for 12 h with 10 µg/mL imiquimod, 10 pg/mL LPS, or 1 ng/mL LPS. IFN- (A) and soluble CD163 levels (B) were measured in PBMC culture supernatants by sandwich ELISA. (C) Surface expression of CD163 was determined by flow cytometric analysis. Data shown are from a single donor and are one of three representative experiments using three different donors, and values are the mean of duplicates. Values below the detection limit of the assay (12.5 pg/mL) are labeled as BD. ***, P < 0.001, versus controls.
|
 View larger version (21K): [in a new window] |
Figure 7. Activation of TLR9 induces CD86 expression on monocytes at 48 h. PBMCs were cultured with 2 µM CpG oligo 2006, 2 µM GpC oligo control, 1 µg/mL E. coli DNA, 10 pg/mL LPS, and 1 ng/mL LPS for 48 h. (A) Surface expression of CD86 was analyzed by flow cytometry, and the MFI of monocytes gated by FSC/SSC was recorded. (B) Cell culture supernatants were harvested, and soluble CD163 levels were measured by sandwich ELISA. (C) Surface expression of CD163 was determined by flow cytometric analysis. Data shown are from a single donor and are one of three representative experiments using three different donors, and values are mean ± SD of triplicates. ***, P < 0.001, or *, P < 0.05, versus controls.
|
protease inhibitor-0 (TAPI-0) inhibited LPS-mediated CD163 shedding and concluded that a matrix metalloproteinase (MMP) was involved in CD163 shedding. However, MMP involvement in CD163 shedding following activation of TLR2 or TLR5 has not been addressed. To test this possibility, purified monocytes were treated with cell surface TLR agonists in the presence or absence of TAPI-0. Pretreatment of monocytes with TAPI-0 blocked the increase in cell culture supernatant levels of soluble CD163 associated with activation of TLR2, TLR4, and TLR5 (Fig. 8
). These data implicate a metalloproteinase in the acute cleavage of CD163 from monocytes upon activation of cell surface TLR. As TLR agonists are known to activate a disintegrin and metalloproteinase 17/TNF--converting enzyme, and TIMP-3 and TAPI-0 are able to inhibit this enzyme [11
, 32
, 33
], it is possible that this enzyme is involved in TLR-mediated CD163 shedding. However, we cannot rule out the possibility that another metalloproteinase plays a role, as TIMP-3 and TAPI-0 can inhibit numerous MMPs.
 View larger version (35K): [in a new window] |
Figure 8. Cell surface TLRs activate a metalloproteinase that triggers CD163 shedding from human monocytes. Purified monocytes were pretreated with 10 µM TAPI-0 for 30 min and then stimulated with the following TLR ligands: 1 ng/mL protein-free LPS, 100 µg/mL zymosan A, 10 ng/mL Pam3Cys, and 200 ng/mL S. typhimurium flagellin. After 1 h of activation, cell culture supernatants were harvested, and soluble CD163 levels were measured by sandwich ELISA. Data shown are from a single donor and are one of three representative experiments using three different donors, and values are mean ± SD of triplicates. ***, P < 0.001, comparing treatment groups with and without 10 µM TAPI-0.
|
Human monocytes and macrophages help maintain homeostasis and defend the body from invasion by pathogens [34 ]. CD163, a monocyte/macrophage-specific marker belonging to the scavenger receptor cysteine-rich superfamily of proteins, is poised at the interface of these seemingly dichotomous monocyte/macrophage functions [1 , 11 ]. Previous publications have reported elevated levels of soluble CD163 in sepsis, pneumonia, and leishmaniasis [6 7 8 ]. LPS from Gram-negative bacteria activates shedding of cell surface CD163 and causes a rise in soluble CD163 in plasma, but the PAMPs involved in activating CD163 shedding during Gram-positive infections have not been identified. This report clearly shows that PAMPs derived from Gram-positive and Gram-negative bacteria, in addition to yeast cell wall structures, are able to activate monocyte shedding of CD163. Purified LPS without contaminating bacterial lipoprotein caused CD163 shedding at strikingly low picogram/milliliter concentrations. Furthermore, sequestration of LPS with PMX or inhibition of the interaction of LPS with the TLR4 receptor complex by the competitive inhibitor RsDPLA completely inhibited LPS-mediated CD163 shedding. The synthetic lipopeptide Pam3Cys and the particulate yeast cell wall product zymosan A, known to activate TLR2/1 and TLR2/6, respectively, also caused CD163 shedding, independent of LPS contamination. The TLR5 agonist, flagellin from S. typhimurium, stimulated monocytes to shed CD163 in the presence or absence of LPS inhibitors. Collectively, these data identify cell surface TLR agonists from Gram-positive bacteria, Gram-negative bacteria, and yeast as efficient mediators of CD163 shedding. These data suggest that the soluble form of the HbSR may be an important mediator that is released during the innate immune response to infection by a diverse range of pathogens. Although a significant fraction of surface CD163 is shed and released as a soluble protein, our data do not exclude the possibility that a minor fraction of this protein is internalized upon extracellular TLR activation. Regardless of whether surface CD163 molecules are internalized simultaneous with CD163 shedding, release of surface CD163 as a soluble protein may still contribute to immune defense by sequestering hemoglobin iron that would otherwise be available for pathogen growth.
Intracellular TLRs can be activated by products that mimic viral nucleic acid structures and initiate antiviral responses from activated cells [12
]. When monocytes were treated with TLR3, TLR7, or TLR9 agonists, CD163 shedding was not activated during 12–48 h of treatment. However, poly(I:C) and imiquimod were able to stimulate PBMCs to produce IFN-, which is consistent with other reports [25
, 26
]. CpG oligos up-regulated CD86 on monocytes, an effect that is likely mediated by cytokines produced from CpG-activated plasmacytoid dendritic cells [28
]. These findings show that the intracellular TLR agonists used in our experiments are able to signal and induce proinflammatory sequelae without activating CD163 shedding from monocytes.
Collectively, the data reported herein support the hypothesis that CD163 is cleaved from the surface of monocytes by pathogens at the same time that hemolysis and Hp-Hb complex formation is likely to occur. These consequences of an infection implicate additional functions for the HbSR beyond binding and internalizing Hp-Hb complexes, as this receptor is cleaved from the surface of monocytes during infections at a time when Hp-Hb complexes are forming, and complex clearance is imperative. For example, soluble CD163 might inhibit hemoglobin use as a source of iron by pathogens, which would explain why this protein is so actively regulated by PAMPs.
Iron is a component of many proteins involved in a diverse range of cellular processes. Eukaryotes and prokaryotes are dependent on iron-containing proteins to help orchestrate the essential biochemical processes of respiration, proliferation, and signal transduction, which maintain homeostasis and/or promote growth [35 36 37 ]. Thus, acquisition of iron as a cofactor for basic cellular processes is essential for maintaining the health of the host, and the availability of host iron can influence pathogen virulence [35 36 37 ]. For this reason, mammals have evolved numerous methods to maintain adequate levels of iron during health while restricting iron availability during infection to limit pathogen growth.
Some human pathogens bypass the need to compete with their host for use of free iron by seeking alternative sources of iron. The abundance of hemoglobin, which contains 80% of the body’s iron, makes it an attractive source of iron for pathogens, despite its containment in red blood cells (RBC) within the circulatory system. Pathogens can gain access to hemoglobin within the circulatory system via transcytosis across epithelial layers [38 , 39 ], infection of open wounds [40 ], or expansion of infectious foci into sepsis [41 ]. In addition, certain Gram-positive [42 43 44 ] and Gram-negative bacteria [45 46 47 ], along with numerous Candida species [48 ], release hemolysins during infection. These hemolysins lyse RBC and release free hemoglobin into the circulation and interstitium. Although hemoglobin is quickly bound to the serum protein haptoglobin [1 ], formation of this complex does not always inhibit the ability of pathogens to use hemoglobin as a source of iron [49 50 51 52 ]. In fact, some pathogens are able to take up or use iron contained in Hp-Hb complexes as well or even better than hemoglobin alone [49 50 51 , 53 ]. Furthermore, heme iron was recently shown to be preferentially used by Staphylococcus aureus when compared directly with the use of transferrin iron by this bacteria [54 ]. These reports indicate that pathogens can actively use hemoglobin as a source of iron and highlight the potential benefit of sequestering host hemoglobin during an infection.
Our data highlight the potential importance of CD163 shedding as an innate immune response to pathogen stimulation of extracellular TLRs. As hemoglobin sequestration could be one of many mechanisms used by the innate immune response to limit pathogen access to host iron, it is intriguing to hypothesize that binding of soluble CD163 to Hp-Hb complexes may limit pathogen access to free hemoglobin. Future experiments are needed to clarify the role of soluble CD163 in limiting pathogen growth in media containing hemoglobin as a sole source of iron. If our hypothesis is supported by such studies, soluble CD163 therapy could be used to limit pathogen growth in conditions of pathogen-associated hemolysis and sepsis. The broad, clinical application of such therapy merits further investigation of the function of this protein.
ACKNOWLEDGEMENTS
This work was supported by the Englert Cell Analysis Laboratory, a shared resource of the Norris Cotton Cancer Center, which is supported by the Rippel Foundation, and a National Cancer Institute Cancer Center Grant (CA23108). These studies were also funded by the National Institutes of Health, the Molecular Pathogenesis Training Grant (AI07519), the Immunology Training Grant (AI07363), AI051547 (to P. M. G.), AI51877 (to C. Wira, Dartmouth Medical School, Lebanon, NH), GM50870 (to N. Q.), and AI-18797 (to S. N. V.). We thank Vivianne Tawfik and Michael Lacroix-Fralish for their expertise in statistics and Allan Munck and Mark Yeager for helpful discussions.
Received December 22, 2005; revised February 27, 2006; accepted March 18, 2006.
REFERENCES
R triggers shedding of the hemoglobin-haptoglobin scavenger receptor CD163 J. Leukoc. Biol. 76,271-277-primed, C3H/HeJ macrophages to a fully tumoricidal state J. Immunol. 139,3697-3702[Abstract]B-/NF-B and ERK1/2 pathways and adhesion J. Biol. Chem. 280,14264-14271 converting enzyme (TACE) is inhibited by TIMP-3 FEBS Lett. 435,39-44[CrossRef][Medline]Related Article
This article has been cited by other articles:
 |
|
 |
|
  H. P. Jia, D. C. Look, P. Tan, L. Shi, M. Hickey, L. Gakhar, M. C. Chappell, C. Wohlford-Lenane, and P. B. McCray Jr Ectodomain shedding of angiotensin converting enzyme 2 in human airway epithelia Am J Physiol Lung Cell Mol Physiol, July 1, 2009; 297(1): L84 - L96. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Maniecki, M. M. Lauridsen, T. B. Knudsen, J. Stentoft, B. K. Moller, S. K. Moestrup, and H. J. Moller A Macrophage Activation Switch (MAcS)-Index for Assessment of Monocyte/Macrophage Activation Blood (ASH Annual Meeting Abstracts), November 16, 2008; 112(11): 3550 - 3550. [Abstract]
|
|
|
|
 |
|
 A. P. Levy, K. R. Purushothaman, N. S. Levy, M. Purushothaman, M. Strauss, R. Asleh, S. Marsh, O. Cohen, S. K. Moestrup, H. J. Moller, et al. Downregulation of the Hemoglobin Scavenger Receptor in Individuals With Diabetes and the Hp 2-2 Genotype: Implications for the Response to Intraplaque Hemorrhage and Plaque Vulnerability Circ. Res., July 6, 2007; 101(1): 106 - 110. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Schaer, F. Vallelian, A. Imhof, G. Schoedon, and D. J. Schaer CD163-expressing monocytes constitute an endotoxin-sensitive Hb clearance compartment within the vascular system J. Leukoc. Biol., July 1, 2007; 82(1): 106 - 110. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. K. Weaver, P. A. Pioli, K. Wardwell, S. N. Vogel, and P. M. Guyre Up-regulation of human monocyte CD163 upon activation of cell-surface Toll-like receptors J. Leukoc. Biol., March 1, 2007; 81(3): 663 - 671. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
