Published online before print December 20, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Medical Clinic, Uster Hospital, Switzerland;
Medical Clinic B Research Unit and
Division of Hematology, Department of Medicine, University Hospital, Zürich, Switzerland; and
Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, Washington
1 Correspondence: Medical Clinic B Research Unit, Department of Medicine, University Hospital of Zürich, CH-8091 Zürich, Switzerland. E-mail: klinsog{at}usz.unizh.ch
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: monocyte/macrophage • hemoglobin-haptoglobin complex
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and endotoxin suppress CD163 expression [6
, 8
, 9
, 11
, 12
]. We have previously reported that CD163 is expressed on blast cells from patients with acute myeloid leukemia (AML) of monocytic lineage [13 ]. In the present study, which investigates functional aspects of CD163 expression, we confirm that CD163 expression is restricted to leukemic blasts from AML types M4/M5, and we show that glucocorticoid treatment further induces CD163 expression on these blasts. Furthermore, we demonstrate for the first time that CD163 expressed on AML types M4/M5 leukemic blasts is capable of scavenging Hb–Hp complexes and is therefore functionally comparable with CD163 on normal macrophages. Lineage-targeted therapies for malignant hematopoietic diseases are an attractive concept. Our findings demonstrating functional expression of CD163 on leukemic blasts of AML types M4 and M5 therefore implicate CD163 as a putative target for therapy in a subset of patients with AML.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Induction of CD163 expression on leukemic blast cells
Primary leukemic blast cells from patients with AML were isolated by Ficoll Hypaque (Pharmacia, Uppsala, Sweden) density gradient centrifugation and washed twice with phosphate-buffered saline (PBS; Gibco, Invitrogen AG, Basel, Switzerland). Cells were cultured in Iscove’s modified Dulbecco’s medium (Gibco), supplemented with 20% human pool serum (off the clot, PAA, Linz, Austria) at 37ºC, 95% humidity, 5% CO2, in a sterile atmosphere (Forma Scientific SteriCult tissue-culture incubator). For induction, 
106/mL cells were cultured for 36 h in the presence or absence of 2.5 x 10–7 M dexamethasone (Sigma, Fluka, Buchs SG, Switzerland) [9
]. Cells were then harvested and prepared for fluorescein-activated cell sorter (FACS) analysis. Purified human CD34+ hematopoietic stem cells were cultured in the appropriate media (cells and media from Clonetics, UK) and analyzed by FACS.
FACS analysis
Peripheral blood mononuclear cells or bone marrow samples were separated by Ficoll Hypaque density gradient centrifugation and washed twice with PBS. Immunophenotypic cell-surface markers (clusters of differentiation) were analyzed by flow cytometry using immunofluorescent double-labeling. For surface marker detection, washed cells were incubated with appropriately diluted antibodies. For intracellular marker detection, cells were first fixed with paraformaldehyde, washed once with PBS, permeabilized (Fix and Perm permeabilization kit, Caltag Laboratories, Burlingame, CA), and then incubated with appropriately diluted, directly labeled antibodies. A set of monoclonal antibodies (mAb) was used routinely against the following antigens: CD13, CD14, CD15, CD33, CDw65, CD117, CD34, CD45, myeloperoxidase, and terminal deoxynucleotidyl transferase (Dako Diagnostics Zug, Switzerland; Caltag Laboratories, South San Francisco, CA; Becton Dickinson, Mountain View, CA; and Serotec, Oxford, UK). To detect cell-surface expression of CD163, cells were labeled with fluorescein isothiocyanate-labeled anti-CD163 (clone 5C6-FAT; BMA Biomedicals AG, Augst, Switzerland) or the appropriate nonbinding isotype control antibody (BD Biosciences, Heidelberg, Germany) at a concentration of 5 µg/mL for 15 min.
After washing twice, cells were resuspended in 500 µL PBS and analyzed on a FACSCalibur flow cytometer using Lysis software (Becton Dickinson, Allschwil, Switzerland). Leukemic blast cells were gated according to their forward- and side-scatter pattern; at least 10,000 gated events were acquired for analysis. For primary analysis, gates were set on the CD45+/CD14+ (to select for monocytes/macrophages), CD3+/19+ (to select for T and B lymphocytes), and CD34+ (to select for hematopoietic stem cells) cell populations. The fluorescence intensity of corresponding samples labeled with the isotype control antibody was used to set the individual threshold of antigen positivity.
Hematopoietic stem cells were generated and harvested from healthy donors in our blood transfusion unit according to a standardized protocol [15 ]. Aliquots of samples were analyzed for CD163, CD34, CD33, CD45, and CD14 expression by FACS analysis, with gating on CD45+/14+ and CD34+/CD163+ cell populations.
Uptake of Hb–Hp complexes by CD163-expressing leukemic blast cells
Purified Hb and Hp (Sigma) were fluorescently labeled using the Alexa-488 protein-labeling kit (Molecular Probes, Eugene, OR; Invitrogen AG). Parallel experiments were performed with the Hb or the Hp moiety serving as the fluorescent indicator to verify specificity of the receptor–ligand interaction. For quantitative determination of Hb–Hp uptake, leukemic blast cells were incubated with the respective ligands at a concentration of 2 µg/mL for 2 h. As the ligand–receptor complex dissociates immediately at calcium ion concentrations below 0.2 mM [16
], cells were washed three times in ice-cold, Ca2+-free PBS containing EDTA to remove noningested Hb–Hp. Uptake of fluorescent Hb–Hp was then quantified by flow cytometry. Competitive inhibition of Hb–Hp uptake was demonstrated by incubation with a 100-fold excess of unlabeled Hb–Hp or a polyclonal goat anti-human CD163 antibody (20 µg/mL) [2
]. For fluorescence microscopy, cells were prepared as follows: After incubation with Alexa-488-labeled Hb–Hp complexes, cells were washed three times in Ca2+-free PBS to remove noninternalized ligand and fixed with 4.5% buffered paraformaldehyde for 15 min. After washing with PBS, cells were incubated for 30 min with Alexa-594-labeled phalloidin for staining of actin filaments and 4',6-diamidino-2-phenylindole (DAPI; nuclear staining; both Molecular Probes). Cells were spun onto glass slides at 900 rpm for 5 min in a Shandon cytocentrifuge (Shandon Inc., Pittsburgh, PA). Samples were analyzed as 0.2 µm thin optical sections with a confocal laser-scanning microscope (CLSM; Leica, Heidelberg, Germany). Analysis of the sections was made using Imaris 3.0 software (Bitplane AG, Zürich, Switzerland).
Determination of soluble CD163 (sCD163) by enzyme-linked immunosorbent assay (ELISA)
The concentration of sCD163 in serum from patients with AML was measured by ELISA (S-1015; BMA Biomedicals AG). A normal range of 0.7 ± 0.6 mg/L was established by analyzing serum samples from 19 healthy volunteers. AML patient serum samples taken at the time of diagnosis were analyzed in duplicate at two different dilutions. Serum ferritin and C reactive protein (CRP) were measured by an Access immunoanalyzer (Opera and Advia Centauer, respectively, Bayer Diagnostics, Germany).
Statistical analysis
Correlations were calculated by nonparametric Spearman rank test using SPSS Version 10.01 (SPSS, San Diego, CA); P < 0.05 was considered significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (16K): [in a new window] |
Figure 1. Constitutive cell-surface expression of CD163 on leukemic blast cells. Surface CD163 was analyzed in 153 samples of patients with AML by flow cytometry. Data are presented as median and range of the respective AML types.
|
Glucocorticoid induction of CD163 expression in AML types M4 and M5 blast cells
Bone marrow and peripheral leukemic blast cells from 24 patients with AML and stem cell preparations from two healthy donors were cultured with and without 2.5 x 10–7 M dexamethasone, a concentration known to induce maximal CD163 expression on normal monocytes [9
]. As shown in Figure 2A
, dexamethasone failed to induce CD163 expression on primary blast cells from patients with non-M4/M5 AML (n=4). In contrast, dexamethasone-inducible CD163 expression was found in leukemic blasts from 20 patients with AML types M4 and M5 (Fig. 2B
2C
2D)
. The extent of CD163 induction inversely correlated with baseline expression; that is, maximum induction was observed in leukemic blasts from patients with low (<10%), constitutive CD163 expression (18.3-fold induction above baseline; range 1.3- to 44-fold; Fig. 2C
, n=7). By comparison, glucocorticoid-induced enhancement of CD163 expression was more moderate in patients with high (>10%) baseline CD163 expression (1.3-fold; range 0.9- to 2.5-fold; Fig. 2D
, n=9). Furthermore, CD163 expression was also enhanced on blasts of myelomonocytic type AML (M4), which have a low constitutive CD163 expression (Fig. 2B
, n=4).
 View larger version (15K): [in a new window] |
Figure 2. Dexamethasone-induced CD163 expression. Leukemic blast cells were incubated in the presence or absence of dexamethasone (Dexa) as described in Materials and Methods. Data are shown as the percentage of CD163-positive blasts at baseline and after induction with dexamethasone (n=26). (A) AML types M0 and M1 blasts and CD34+ progenitor cells from healthy donors could not be induced to express CD163 by dexamethasone treatment (n=6). (B) AML type M4 blasts were induced to express CD163 by 1.8- to 26.5-fold (n=4). (C) Leukemic blasts of AML type M5 with low baseline CD163 expression (<10% CD163-positive cells) could be induced to express CD163 by a mean of 18.3-fold (n=7). (D) AML type M5 blasts with high baseline CD163 expression (>10% CD163-positive cells) could be induced to express CD163 by a mean of 1.3-fold (n=9).
|
 View larger version (40K): [in a new window] |
Figure 3. FACS analysis of inducible CD163 expression. FACS analysis of CD163 and CD34 expression on cells obtained from patients with (A) AML type M5, (B) AML type M0, and (C) purified CD34+ cells of a healthy subject. Cells were treated in the presence or absence of dexamethasone as described in Materials and Methods. (A) The AML type M5 CD34+ cell population, which did not express CD163 at baseline (black dots), was induced to express CD163 upon dexamethasone treatment. The majority of the CD34-negative population (gray dots), which consisted mostly of monocytes (CD45+/CD14+ population, box plot not shown), expressed CD163 at baseline and could further be induced to express CD163 after dexamethasone treatment. (B) AML type M0 CD34+ cells (black dots) remained negative for CD163 expression, and the CD45+/CD14+ cell population (gray dots) was induced to express CD163. (C) CD34+ cells from a healthy donor remained negative for CD163 expression, even after dexamethasone treatment.
|
 View larger version (24K): [in a new window] |
Figure 4. Uptake of Hb–Hp complexes by functional CD163 on leukemic blast cells. Blasts were incubated with fluorescent-labeled Hb–Hp complexes and analyzed by FACS as described in Materials and Methods. (A) Representative FACS of CD163-mediated uptake of fluorescently labeled Hb–Hp complexes by AML type M5 leukemic blast cells, which had high constitutive CD163 expression. (B) Representative FACS of Hb–Hp uptake by AML type M5 leukemic blasts before (shaded peak) and after (solid peak) induction of CD163 with dexamethasone. (C) FACS analysis of CD163-mediated uptake of Hb–Hp complexes by leukemic blasts. For competition assays, unlabeled reactants were added 100-fold in excess of the labeled ligand. f, Labeled part of the ligand. Data are mean ± SD; n = 3.
|
 View larger version (86K): [in a new window] |
Figure 5. Dexamethasone-induced Hb–Hp uptake by AML type M5 leukemic blast cells. (A) Triple-fluorescent staining of Hb–Hp uptake by AML type M5 blast cells upon treatment with dexamethasone. Green, Alexa-488-labeled Hb–Hp complexes; red, Alexa-594 phalloidin (cytoskeletal stain); blue, DAPI counterstain of nucleus. (B) Uptake of Hb–Hp complexes by the same cell type without dexamethasone treatment. Original magnification, 400x. (C) CLSM of Hb–Hp uptake by AML type M5 blast cells after dexamethasone treatment as in A.
|
As shown in Figure 6 , the mean surface expression of CD163 did not correlate with the amount of sCD163 in the corresponding serum of AML patients (P=0.977). Furthermore, AML type did not correlate with sCD163 (P=0.3) but correlated significantly with surface CD163 expression (P<0.0001), as described for the data presented in Figure 1 . CRP and ferritin were measured to control for systemic inflammation and macrophage activation; however, neither parameter correlated significantly with sCD163 (P=0.165 and P=0.521, respectively).
 View larger version (18K): [in a new window] |
Figure 6. sCD163 in relation to surface CD163 expression in AML patients. sCD163 concentration in serum was measured simultaneously with surface markers on the day of AML diagnosis. AML types M1 (n=9), M2 (n=10), M3 (n=4), M4 (n=9), and M5 (n=7) were analyzed. CD163 surface expression correlated to AML type (P>0.001), but sCD163 did not (P=0.97). Data are mean ± SD.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Various studies have identified the synthetic glucocorticoid dexamethasone to be among the strongest transcriptional inducers of monocyte CD163 expression in vitro and in vivo [9 , 23 , 24 ], an effect that is mediated by multiple glucocorticoid responsive elements in the promoter of the gene [25 ]. To further characterize functional CD163 expression on leukemic blasts, we therefore investigated its regulation by dexamethasone in vitro. In all cases tested, CD163 expression could only be enhanced on leukemic blast cells of AML types M4/M5 (Fig. 2B 2C 2D) . This suggests that regulation of CD163 expression is preserved in malignant cells of the monocytic lineage. In contrast, CD34+ stem cells from healthy donors do not express CD163 constitutively (Figs. 2A and 3C) , and expression cannot be induced by treatment with glucocorticoids. This observation suggests that expression and regulation of CD163 differ between normal progenitor cells and leukemic blasts of AML types M4/M5.
This finding is further substantiated by an interesting observation in one of our patients with AML type M5. At the time of initial diagnosis, 35% of the CD34+ blasts expressed CD163. After remission induction chemotherapy, 30% of the cells harvested for peripheral stem cell transplantation was CD34+; however, only 1% of these expressed CD163. Dexamethasone treatment induced only 3% of the harvested cells to express CD163. The CD45+/14+ monocytes (35% of the harvested cells) did not express CD163 at baseline (0%), but 18% of this subpopulation was induced to do so by dexamethasone. These data indicate that AML types M4 and M5 leukemic blasts retain their monocytic/monoblastic phenotype by expressing CD163 and suggest that these blasts differ from nonleukemoid CD34+ progenitor cells from healthy donors and patients in remission and leukemic blasts of other AML types. Although it appears that CD163 is aberrantly expressed at an early stage in CD34+ AML blasts of myelo-/monocytic origin, it is unknown whether this expression has a role in the pathogenesis of the disease, and further studies have to address this question.
CD163 belongs to the cystein-rich scavenger receptor family, and Hb–Hp complexes have been identified as its physiological ligand [2 , 9 ]. Therefore, we investigated whether the functional properties of CD163 are conserved on AML blast cells. As shown in Figures 4 and 5 , AML type M5 leukemic blasts, which express CD163, are indeed capable of internalizing fluorescently labeled Hb–Hp complexes to an endosomal compartment. Binding and internalization of fluorescent Hb–Hp complexes were competitively inhibited by excess, unlabeled ligand and were blocked by a polyclonal antibody directed against CD163, indicating that Hb–Hp uptake by leukemic blasts is functionally mediated by CD163. As expected, no significant interaction and internalization of Hb–Hp complexes were detected with CD163-negative blasts or with AML-derived CD163-negative cell lines such as THP-1 or U-937 (data not shown). Clearly, characterization of the lineage-restricted expression of CD163 on a subset of AML blasts but not on other hematopoietic progenitor cells identifies this receptor as a potential disease-specific drug target. It is therefore of high interest to note the preservation of high endocytotic activity of CD163 expressed on leukemic cells, providing a means to specifically deliver cytotoxic drugs. A limitation of this approach encompasses the extent of CD163 expression on CD163+ leukemias. However, although basal expression of CD163 is sometimes rather low, binding and internalization of a putative CD163-directed drug could be enhanced considerably by glucocorticoid induction of CD163 expression. In preliminary studies, we found efficient internalization of the anti-CD163 mAb 5C6-FAT in CD163-expressing leukemic blast cells (unpublished observation).
The clinical success of the CD33-targeting, antileukemia immunoconjugate, gemtuzumab ozogamicin (Mylotarg®), can serve as a paradigm, demonstrating that targeted therapy with cytotoxic drugs is feasible, even if the targeted receptor is displayed on a restricted number of mature cells [26
]. Of concern in the case of CD163 is that normal monocytes express CD163 on their cell surface, potentially interfering with therapeutic targeting of CD163 on AML blast cells. In healthy individuals, 20% of the blood monocytes constitutively expresses CD163. In this study, we found that roughly 50% of normal blood monocytes of healthy stem cell donors, which received granulocyte-colony stimulating factor to trigger stem cell proliferation, expressed CD163. In acute M4/M4 type leukemias, there is however only a small proportion of normal monocytes compared with the bulk of leukemic blast cells, suggesting that the CD163-expressing blast cells would be the primary targets, even when only a part of the blasts expresses the receptor. Specifically reducing cellular mass would be of advantage in some clinical situations, for example, hypercellular leukemias, as is often the case in AML types M4/M5. A combination with chemotherapy would be required to treat the remaining blasts not expressing CD163.
Targeting CD163-expressing cells is expected to be limited to monocytes/macrophages and spare normal CD34+ stem cells in the bone marrow so that the regenerative supply of a substantial proportion of normal monocytes should be preserved. Another normal cell type potentially affected by anti-CD163-targeted therapy is the tissue macrophage. However, only a distinct population of CD68+ tissue macrophages is positive for CD163, and macrophages in germ centers of lymphatic tissues are negative for CD163 (unpublished observation). Therapeutic targeting of CD163 in AML types M4/M5 could therefore be expected to leave a majority of normal CD163– blood monocytes and distinct populations of tissue macrophages unaffected, which may offer some benefit for control of infections during induction therapy. Moreover, T and B lymphocytes and granulocytes, which importantly contribute to the control of infections, are not affected by CD163 targeting therapy. Certainly, further preclinical studies are needed to investigate potential toxic side-effects of targeting CD163 in normal monocytes and leukemic blast cells.
sCD163 levels well above the range found in healthy individuals have been reported in patients with a broad spectrum of hematological disease [21 ]. sCD163 may therefore serve as a sink for therapeutic anti-CD163 antibodies, potentially interfering with efficient delivery of a biologic drug to the target cell and increasing off-target side-effects. We thus determined surface expression and sCD163 serum concentration at the time of diagnosis in 39 comprehensively characterized patients with AML. There was no increase in sCD163 concentration in patients whose leukemic blasts displayed high surface CD163 expression (Fig. 6) , indicating that sCD163 is not constitutively released from the leukemic blasts. Our data therefore suggest that sCD163 may not compete with therapeutic targeting of surface CD163 on leukemic blasts. However, as excessive levels of sCD163 can be found in patients with macrophage activation syndromes [20 , 27 ] and also to some extent, in patients with sepsis, we cannot exclude the possibility that considerable amounts of sCD163 could be released from malignant cells during inflammation or infection, and further studies need to explore this possibility.
In conclusion, our study confirms a lineage-restricted CD163 expression on leukemic blasts with preservation of regulation and function. This renders CD163 a potential, novel, therapeutic target for a subset of patients with AML. Further preclinical studies investigating drug-mediated cell death with conjugates using CD163 to select target cells are thus warranted.
Received June 9, 2005; revised August 17, 2005; accepted September 28, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
}R triggers shedding of the hemoglobin-haptoglobin scavenger receptor CD163 J. Leukoc. Biol. 76,271-277This article has been cited by other articles:
 |
|
 |
|
  A. Kapralov, I. I. Vlasova, W. Feng, A. Maeda, K. Walson, V. A. Tyurin, Z. Huang, R. K. Aneja, J. Carcillo, H. Bayir, et al. Peroxidase Activity of Hemoglobin{middle dot}Haptoglobin Complexes: COVALENT AGGREGATION AND OXIDATIVE STRESS IN PLASMA AND MACROPHAGES J. Biol. Chem., October 30, 2009; 284(44): 30395 - 30407. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
 |
|
 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]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
