Published online before print August 1, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Biochemistry and Molecular Biology,
Department of Veterinary Sciences, Penn State University, University Park, PA 16802
Department of Biology, Amherst College, Amherst, MA 01002
1 Correspondence: Department of Biochemistry and Molecular Biology, 428 South Frear Laboratory, Penn State University, University Park, PA 16802. E-mail: ur3{at}psu.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: macrophages • annexin V • CD14 • Mac-1 • caspases
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Loss of phospholipid asymmetry and exposure of PS on the apoptotic cell surface is not only necessary, but also sufficient, to engage a number of macrophage receptors implicated in recognition of apoptotic cells. Elevating cytosolic Ca2+ levels using ionophore activates the scramblase and brings PS to the surface of thymocytes in a matter of minutes [7 ]. Phagocytosis of these targets is inhibited by PS vesicles, annexin V, the tetrapeptide RGDS, N-acetylglucosamine, and monoclonal antibody 61D3 [8 , 9 ], suggesting that a PS receptor, an integrin, a lectin-like receptor and CD14 on the macrophage surface are all engaged by these targets. Because such a short treatment with ionophore does not leave time for gene expression or protein synthesis to generate new ligands on the target cell surface, abolition of phospholipid asymmetry may induce conformational changes in or rearrange pre-existing membrane proteins, exposing new recognition epitopes [9 ].
Unexpectedly, recent results suggest that PS expression is required not just on the target cell surface, but also on the macrophage that engulfs the apoptotic cell. Unlike virtually all normal, healthy cells, macrophages appear to constitutively exhibit at least a partial loss of lipid asymmetry [10 ] and associated PS expression [8 , 11 ]. When macrophages are pretreated with annexin V to mask this surface PS, inhibition of phagocytosis is just as efficient as when apoptotic thymocytes are pretreated [8 , 11 ]. On the other hand, monocytes, the precursors of macrophages, do not express PS on their surface: monocytes drawn from the circulation maintain an asymmetric distribution of phospholipids [12 ] and do not bind annexin V [13 ]. As might be expected, freshly isolated human monocytes are unable to engulf PS-expressing apoptotic cells [14 , 15 ], whereas in vitro maturation to macrophages enables phagocytosis of erythrocytes with PS on their surface [16 ] and apoptotic target cells [5 , 14 ]. In the present study, we have used a human monocytic cell line, as well as primary human monocytes, to explore in more detail the relationship between the acquisition of expression of PS on the phagocyte cell surface and the phagocytosis of apoptotic thymocytes. Several human monocytic cell lines can be induced to differentiate with PMA, and at least one line (THP-1) has been reported to acquire the ability to phagocytose apoptotic cells in a manner inhibitable by PS vesicles [14 , 15 ], just as primary macrophages. However, whether differentiation induces the expression of PS on the surface of such cells, and if so, whether that PS expression is required for phagocytosis of apoptotic cells, has not been examined.
It was recently reported that the differentiation of U937 cells to macrophages induced by treatment with phorbol esters is accompanied by changes normally associated with apoptosis, including nuclear fragmentation and chromatin condensation, DNA fragmentation, mitochondrial cytochrome C release, and activation of caspase-3 [17 ]. All of the events could be blocked by the broad-spectrum caspase inhibitor zVAD-fmk, leading to the suggestion that in these cells the processes of terminal differentiation and apoptosis are related. U937 cells therefore provide an opportunity to investigate not only PS expression on differentiating monocytic cells and its relationship to the ability to phagocytose apoptotic cells, but also its relationship to the mechanisms that lead to PS expression on apoptotic cells. Here, we demonstrate that differentiation of U937 cells, just like their primary monocyte counterparts, is associated with expression of PS on the cell surface, even in the presence of caspase inhibitors. More important, this surface PS is required for phagocytosis of apoptotic thymocytes, just as for primary macrophages.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
dihydroxyvitaminD3 (VitD3) from EMD Biosciences (Calbiochem), U937 monocytic cells from American Type Culture Collection, and fluoresceinated latex beads (1 µm in diameter) from Polysciences. Phycoerythrin (PE)-conjugated annexin V, PE-conjugated anti-CD11b (clone ICRF44), PE-conjugated anti-CD14 (clone M5E2), and PE-conjugated murine IgG2a and IgG1 isotype controls (clones MOPC-21 and G155-178) were purchased from PharMingen, FITC-conjugated annexin V from Molecular Probes, and Ficoll-PaqueTM from Amersham-Pharmacia Biotech (Piscataway, NJ). Unlabeled annexin V was purified as described previously in detail [6
, 18
].
Monocytes and macrophages
U937 cells were cultured in RPMI 1640 medium supplemented with 2 mM glutamine and 10% FBS at 37°C in 5% CO2. Apoptosis was induced in log phase U937 cells by culturing for 3 h with 4 µg/ml of CAM. Log phase U937 cells were induced to differentiate by culturing for 24-72 h with 10 ng/ml of PMA or by culturing for 72-168 h with either 100 nM VitD3 or 1 µM t-RA alone, or in combination. In some experiments, U937 cells were preincubated for 30 min in medium containing 10 µM zVAD-fmk, then PMA or CAM added directly to the cultures as above. Human monocyte-derived macrophages (HMDM) were prepared from fresh human venous blood obtained from volunteers according to institutional guidelines [19
]. Blood was collected into ice-cold PBS (7.4 mM Na2HPO4, 2.6 mM NaH2PO4, 137 mM NaCl, 10 mM KCl) containing 10 U/ml of heparin, and centrifuged at 278 x g for 5 min. The buffy coat and top 10% of erythrocytes were removed and diluted 1:2 with PBS. Mononuclear cells in the suspension were separated by centrifugation on Ficoll-Paque, according to the manufacturer’s instructions. After culturing at 37°C for 2 h (0-day cultures) or overnight in RPMI 1640 medium containing 10% FBS, medium and nonadherent cells were removed by aspiration. Fresh medium was added to long-term cultures and changed approximately every 3 days. Monocytes in 0-day cultures were identified by flow cytometry based on their characteristic high forward angle light scatter (FS) and intermediate side-angle light scatter (SS) [20
]. In long-term cultures, HMDM were identified by their high FS and SS characteristics and were easily distinguished from monocytes.
Thymocytes
Male CBA/J mice, 4-8 wk of age, were maintained on food and water ad libitum in accordance with the guidelines of the Institutional Animal Care and Use Committee. Thymuses were removed and dissociated in PBS. After collecting cells by centrifugation and resuspending in 17 mM Tris, 140 mM NH4Cl, pH 7.2, to lyse erythrocytes, thymocytes were washed and resuspended at 106 cells/ml in RPMI 1640 medium containing 10% FBS. Apoptosis was induced by incubation at 37°C in 5% CO2 for 6 h in the presence of 10-6 M dexamethasone.
61D3 mAb
61D3 hybridoma cells were a gift from Dr. Donald Capra, Oklahoma Medical Research Foundation, Oklahoma City, OK. 61D3 cells were grown to 
107 cells/ml in medium supplemented with IgG-deficient FBS. After removing the cells by centrifugation, 61D3 mAb was purified from the supernatant by Protein-A chromatography, according to the manufacturer’s instructions (BioRad).
Annexin V and antibody staining
Adherent monocytes, HMDM, and U937 cells were collected by scraping them into buffer. 1 x 106 cells were incubated with 5 µl of PE-annexin V or FITC-annexin V in 100 µL of annexin V buffer (10 mM HEPES/NaOH, pH 7.4, 140 mM NaCl, 2.5 mM CaCl2), according to the manufacturer’s instructions. After 15 min on ice, 400 µl of ice-cold annexin V buffer and, in some experiments, 5 µl of a 1 mg/ml solution of PI was added, and the cells analyzed by flow cytometry. As a negative control for all annexin V staining procedures, cells were washed extensively in Ca2+-free PBS and stained with annexin V in the absence of Ca2+. For antibody staining, cells were resuspended at 107 cells/ml in ice-cold staining buffer (PBS containing 1% FBS). For direct immunofluorescence, 100 µl of cells was added to the volume of mAb or isotype control antibody specified by the manufacturer. For indirect immunofluorescence, 1 µg of either 61D3 mAb or isotype control (clone MOPC 141) was added to 100 µl of cells. After 25 min on ice, the cells were washed and 1 µg of FITC-conjugated anti-murine IgG added. All samples labeled with antibody were washed in staining buffer after 25 min on ice and cells resuspended in 500 µl of staining buffer for flow cytometric analysis.
Flow cytometry
Approximately 10,000 cells/sample were analyzed using an EPICS-XL-MCL flow cytometer (Coulter Electronics) fitted with a single 15 mW argon ion laser providing excitation at 488 nm. Cells stained with FITC were monitored through a 525-nm bandpass filter, and cells stained with PE or macrophages phagocytosing fluorescent latex beads were monitored through a 575-nm bandpass filter. As presented in Results, late apoptotic and dead U937 cells had the same decreased FS and increased SS characteristics, and both were gated out of all staining profiles. For both primary monocytes and HMDM, PI-positive cells were gated out of all profiles.
DNA fragmentation analysis
PMA-treated U937 cells were scraped into overlying medium. 4-5 x 105 of these cells or untreated U937 cells were collected by centrifugation, the supernatant removed and genomic DNA extracted and fractionated into high-molecular-weight (HMW) and low-molecular-weight (LMW) fractions using the Suicide-Track DNA Ladder Isolation kit, according to the manufacturer’s instructions. The entire content of each LMW sample was electrophoresed in a single lane of a 1.5% agarose TAE gel at 4-5 V/cm. The gel was then stained with 0.5 µg/ml of ethidium bromide for 30 min, destained in 1 mM MgSO4, and DNA was visualized by UV light.
Phagocytosis assays
Nonadherent U937 cells were plated onto glass 16-well microchamber slides preincubated for 5 min at room temperature with 100 µl/well of 0.1 mg/ml of poly-L-lysine prepared in sterile H2O and then washed with sterile H2O. 3 x 105 cells/well were incubated in annexin V buffer for 1-2 h prior to phagocytosis assays to permit cell adherence. Primary mononuclear cells were plated directly onto 18-mm coverslips. After 2 h incubation at 37°C, nonadherent cells were removed by aspiration, and adherent cells were washed once with ice-cold annexin V buffer. Adherent HMDM and U937 cells were released from culture dishes by scraping, and 3 x 105 cells in annexin V buffer were pipetted onto 18-mm bicarbonate-treated glass coverslips kept in 30 mm petri dishes. For all assays, 106 thymocytes in 150 µl of annexin V buffer were overlaid onto 3 x 105 macrophages. In some experiments, either the target cells or macrophages were first incubated in 150 µl of annexin V buffer containing 10 µM annexin V for 15 min at room temperature and then washed twice with annexin V buffer before targets overlaid onto macrophages. In other experiments, macrophages were pretreated with either 5 µg of Leu-M3 anti-CD14 mAb or 5 µg of 61D3 anti-CD14 mAb in 150 µl of annexin V buffer for 15 min at room temperature before being overlaid with target cells. After 30 min at 37°C in 5% CO2 for phagocytosis by HMDM, or 1 h for phagocytosis by U937 cells, coverslips were washed vigorously in annexin V buffer and fixed in 1.8% formaldehyde for 15 min before staining with Diff-Quik. The number of cells phagocytosed was enumerated as described in detail [5
]. Results are presented as the mean +/- standard deviation of triplicate coverslips. For phagocytosis of fluorescent latex beads, beads were washed twice in RPMI 1640 medium and resuspended in annexin V buffer at a 10,000-fold dilution, then 100 µl was added to 3 x 105 macrophages per well of a 24-well tissue culture plate that either had or had not been pretreated for 10 min with 200 µl of either 1 µM annexin V or BSA at room temperature. After 30 min at 37°C in 5% CO2, macrophages were washed and released from the dishes with 0.5 mM EDTA in PBS and analyzed by flow cytometry.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (25K): [in a new window] |
Figure 1. Light scatter characteristics and annexin V staining of U937 cells induced to undergo apoptosis by treatment with CAM. (A) Flow cytometric light scatter profile of U937 cells after 2 h treatment with CAM. Region 1 (R1) designates shrunken, late-phase apoptotic cells identified by their reduced forward angle light scatter and increased side-angle light scatter. PI-positive cells were not gated out of the profile. (B) Percent U937 cells in R1 as a function of time of treatment with CAM. (C) Untreated U937 cells (upper panel) or U937 cells treated with CAM for 3 h (middle panel) or with CAM plus zVAD-fmk for 3 h (lower panel) were stained with annexin V in the absence (thin line) or presence (bold line) of Ca2+ and analyzed by flow cytometery. The R1 subpopulation was gated out of the profile. Data are representative of four different experiments with similar results.
|
 View larger version (86K): [in a new window] |
Figure 2. Analysis of low-molecular-weight DNA from U937 cells induced to undergo apoptosis by treatment with CAM or induced to differentiate by treatment with PMA. Low-molecular-weight DNA was isolated from untreated U937 cells, U937 cells treated with CAM plus or minus zVAD-fmk for 3 h, or U937 cells treated with PMA plus or minus zVAD-fmk for 1 day, and was electrophoresed through a 1.5% agarose gel and stained with ethidium bromide. M: 100 bp ladder of marker DNA. Data are representative of two different experiments with similar results.
|
 View larger version (21K): [in a new window] |
Figure 3. Mac-1 expression on U937 cells treated with PMA to differentiate. (A) Untreated U937 cells or (B) U937 cells treated with PMA for 1 day or (C) two days were stained with either isotype control (thin line) or anti-CD11b mAb (bold line) and analyzed by flow cytometry. In panel D, cells treated with PMA alone (thin line) or PMA plus zVAD-fmk (bold line) for 1 day were stained with anti-CD11b mAb. The data presented are representative of three different experiments with similar results.
|
Collectively, these results indicate that although inhibition of caspase activity in differentiating U937 cells may block events characteristic of apoptosis, it has no effect on the morphological changes, or on the regulation of expression of cell surface molecules such as Mac-1.
Binding of annexin V was used to investigate whether the differentiation of U937 cells entails changes in expression of PS on the cell surface. Because apoptotic cells in the population will express PS irrespective of their differentiation status, the presence of these cells was monitored by measurement of light scatter characteristics as described for Fig. 1
. As shown in the insets to Fig. 4
, the low level of apoptotic cells in a normal population (3.9%) is increased somewhat by PMA treatment (5.9%), and this increase is reduced by treatment with zVAD-fmk (5.1%). As expected, these cells are all annexin-positive (data not shown). When these cells were gated out, the remaining viable cells nearly all displayed a 2.5- to threefold increase in median fluorescence after 2 days of treatment with PMA (Fig. 4B)
relative to untreated cells (Fig. 4A)
. By comparison, the median increase in fluorescence was only 1.5-fold after one day of PMA treatment (not shown). Unlike the case of PS expression on U937 cells treated with CAM to induce apoptosis, zVAD-fmk did not prevent the expression of PS on cells treated with PMA (Fig. 4C)
.
 View larger version (17K): [in a new window] |
Figure 4. Annexin V staining of U937 cells treated with PMA to differentiate. (A) Untreated U937 cells or (B) U937 cells treated with PMA for 2 days were stained with annexin V in the absence (thin line) or presence (bold line) of Ca2+ and analyzed by flow cytometry. In panel C, cells treated with PMA alone (thin line) or PMA plus zVAD-fmk (bold line) for 2 days were stained with annexin V in the presence of Ca2+ and analyzed by flow cytometry. Light scatter profiles are included as insets. The R1 subpopulation (see Figure 1A
) was gated out of the annexin V staining profile. The data presented are representative of three different experiments with similar results.
|
 View larger version (16K): [in a new window] |
Figure 5. Phagocytosis of apoptotic thymocytes by U937 cells treated with PMA to differentiate following pretreatment with annexin V or cotreatment with zVAD-fmk. Untreated (nonapoptotic) murine thymocytes or thymocytes treated with dexamethasone to undergo apoptosis were presented to monolayer cultures of U937 cells that were treated (A and C) for 1 day or (B) 2 days with PMA. (A and B) Either thymocytes (T) or U937 cells (U) were pretreated with 10 µM annexin V (AV) for 15 min. Following pretreatment, cells were washed twice to remove excess annexin V before thymocytes were presented to macrophages. (C) U937 cells were cotreated with PMA and zVAD-fmk. The data presented are representative of two different experiments with similar results.
|
Differentiated U937 cells utilize CD14 for phagocytosis of apoptotic thymocytes
Phagocytosis of apoptotic cells requires protein receptors, as well as the expression of PS on macrophages. One such protein is the LPS receptor, CD14, as indicated by the ability of the anti-CD14 mAb 61D3 to inhibit phagocytosis of apoptotic lymphocytes [5
, 22
23
24
]. The expression of CD14 on U937 cells was monitored using four different monoclonal antibodies to CD14 (see Materials and Methods), all yielding similar results (not shown). Expression on undifferentiated U937 cells is low and not significantly increased until two days post-treatment with PMA. Nonetheless, phagocytosis was inhibited by 61D3 mAb, an anti-CD14 antibody shown to inhibit specifically the phagocytosis of apoptotic cells by macrophages [5
, 25
26
27
], after only one day of PMA treatment (not shown).
PS is also expressed on U937 cells induced to differentiate by Vitamin D3 and retinoic acid
To determine whether surface expression of PS is a general feature of differentiation, rather than a specific effect of treatment with PMA, the effects of other inducers on PS expression were examined. VitD3 and t-RA are lipophilic agents that induce growth arrest and differentiation of a variety of cell types. When U937 cells are treated with each of these structurally disparate molecules, they respond in different ways. Cells treated with VitD3 for 3 days have been reported to become adherent, form aggregates, and express CD14 [28
]. On the other hand, cells exposed to t-RA for 3 days do not display these characteristics, but instead express the low affinity IgE receptor CD23, the 6 integrin CD49f, and the biliary glycoprotein CD66a [28
29
30
]. Treatment with either agent increases the expression of Mac-1 [31
], and importantly, cotreating U937 cells with both VitD3 and t-RA potentiates the effects of either agent alone [32
].
When U937 cells were treated with either VitD3 or t-RA alone, neither increased adherence, nor was aggregation observed. However, when the cells were cotreated with VitD3 and t-RA, they became more adherent (although not as tenaciously as when treated with PMA) and formed aggregates. VitD3 alone induced increased expression of CD14, whereas t-RA alone did not (not shown). As expected, either agent alone induced increased expression of Mac-1, an increase that was detectable as early as 3 days of treatment (not shown) but maximal at 7 days, while cotreatment with both agents produced dramatically increased levels at 7 days (Fig. 6A ). To measure PS expression cells were stained with annexin V. After gating out apoptotic and dead cells by both light scatter and PI staining, the remaining viable cells showed little increase in annexin binding after 3 days of treatment with either VitD3 or t-RA, alone or in combination (not shown). After 7 days, treatment with either VitD3 or t-RA alone still produced little discernible increase, whereas dual treatment with VitD3 and t-RA produced a 1.5-fold increase in median fluorescence (Fig. 6B) , similar to that seen after one day of treatment with PMA.
 View larger version (19K): [in a new window] |
Figure 6. Mac-1 expression and annexin V staining of U937 cells treated with VitD3 and /or t-RA to differentiate. (A) Untreated U937 cells or U937 cells treated with VitD3 or t-RA alone, or in combination, were stained with either isotype control (thin line) or anti-CD116 mAb (bold line) and analyzed by flow cytometry. (B) Untreated U937 cells or U937 cells treated with VitD3 or t-RA alone, or in combination, were stained with annexin V in the absence (thin line) or presence (bold line) of Ca2+ and analyzed by flow cytometry.
|
 View larger version (23K): [in a new window] |
Figure 7. Phagocytosis of apoptotic thymocytes by U937 cells treated with VitD3 and/or t-RA to differentiate following pretreatment with annexin V. Untreated (nonapoptotic) murine thymocytes or thymocytes treated with dexamethasone to undergo apoptosis were presented to monolayer cultures of U937 cells that were treated for 7 days with VitD3 and/or t-RA. In one experiment, (A) annexin V was present during phagocytosis assays. In a separate experiment, (B) U937 cells were pretreated with annexin V (and washed) prior to phagocytosis assays.
|
 View larger version (15K): [in a new window] |
Figure 8. Annexin V staining of primary human monocytes and differentiated, monocyte-derived macrophages. Monolayer cultures of freshly isolated human monocytes cultured for (A) 2 h or (B) 9 days, to allow differentiation into monocyte-derived macrophages, were stained with annexin V in the absence (thin line) or presence (bold line) of Ca2+ and analyzed by flow cytometry. The staining of (A) only monocytes or (B) only macrophages, gated by their light scatter profiles and lack of staining with PI, are presented. The data presented are representative of two different experiments with similar results.
|
 View larger version (11K): [in a new window] |
Figure 9. Phagocytosis of apoptotic thymocytes by primary human monocytes and differentiated, monocyte-derived macrophages following pretreatment with annexin V. Untreated (nonapoptotic) murine thymocytes or thymocytes treated with dexamethasone to undergo apoptosis were presented to monolayer cultures of freshly isolated human monocytes (day 0) or differentiated, monocyte-derived macrophages (day 11). Macrophages (M) were pretreated with 10 µM annexin V (AV) for 15 min and washed twice to remove excess annexin V before targets were presented to them. The data presented are representative of two different experiments with similar results.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results presented here show that, like circulating monocytes [12 ], untreated, nonadherent U937 cells do not express PS on their surface. However, upon induction of differentiation by PMA or VitD3 and t-RA, U937 cells do display PS on their surface, like terminally differentiated macrophages [8 ]. Since surface expression of PS is a universal feature of cells undergoing apoptosis [33 ], the report that PMA treatment of U937 cells induces other events characteristic of apoptosis, including chromatin condensation and DNA degradation [17 ], raised the possibility that PS expression on U937 cells was part of an apoptotic program. The observation that the events characteristic of apoptosis were sensitive to inhibition by zVAD-fmk further suggested that differentiation to the macrophage phenotype might depend on caspase activation as apoptosis does. As shown here, zVAD-fmk is an efficient inhibitor of all aspects of apoptosis induced by CAM, including PS exposure. Similarly, when cells were treated with PMA, the DNA degradation observed was efficiently blocked by zVAD-fmk, as previously reported [17 ]. The amount of DNA degraded was low, however, and consistent with the possibility that it originates from a small subpopulation of cells undergoing apoptosis within the larger, differentiating population.
In marked contrast, however, zVAD-fmk does not prevent alterations characteristic of differentiation to macrophages. The morphological changes seen when U937 cells are treated with PMA were not prevented by zVAD-fmk. The expression of Mac-1 was similarly insensitive to this inhibitor, indicating that inhibiting caspases has no effect on differentiation. Therefore, the finding that zVAD-fmk did not block PMA induction of PS expression, unlike the case when PS expression was induced by CAM, suggests that the former occurs as part of the differentiation, rather than apoptosis, program. In addition, PS exposure on differentiating cells was much less extensive than on the same cells undergoing apoptosis and occurred on a much different timescale. Together, these results argue that surface expression of PS is a functional marker for monocyte-macrophage differentiation in U937 cells.
While this manuscript was in revision, Sordet et al. [34 ] reported that several caspases are activated when U937 cells treated with PMA or human primary monocytes differentiate to macrophages. However, despite caspase activation, a variety of assays indicated that the bulk population was not undergoing apoptosis and no more than 4% of the cells were morphologically apoptotic, in good agreement with our results. When cells were treated with a 10-fold higher concentration (100 µM) of zVAD-fmk than was used in our study, nonapoptotic cell death and lysis occurred, obscuring whether this higher concentration could prevent differentiation. In contrast to our results, the investigators reported that after one day of treatment with PMA, U937 cells were not labeled by annexin V, based on the lack of a distinct subpopulation with increased staining. However, the subtle 1.5-fold increase in staining of the population as a whole on day one that we report might well go unnoticed. Together, these studies demonstrate that such events as caspase activation and PS exposure normally associated with apoptosis can also be hallmarks of differentiation, independent of apoptosis.
It should be noted that PS exposure has also been described as part of the activation or differentiation of other cell types. In addition to the classic case of activated platelets [35 ], it has been known for some years that activation of leukocytes can lead to transient expression of PS on the cell surface [12 , 36 37 38 ]. In addition, studies with whole animals suggest that the same is true for megakaryocytes prior to their fragmentation into platelets [33 ], and both whole animal and in vitro studies indicate that myoblasts also expose PS prior to fusion to myotubes [33 , 39 ]. As in the case of differentiating U937 cells, exposure of PS on fusing myoblasts does not depend on caspases [39 ], arguing that neither of these two cases is a mistaken instance of apoptosis, or is a simple adaptation of the apoptotic pathway to new purposes. The functional consequences of PS exposure differ in these several cases. On activated platelets, PS provides a surface for assembling protease complexes [40 ]; on myoblasts, and perhaps megakaryocytes, it is required for membrane fusion; on activated leukocytes, it may function in cell signaling; and as shown here, on differentiating monocytes (and their apoptotic targets), PS exposure is required for phagocytosis. It will be interesting to learn whether there is a common mechanistic thread that unites these disparate functions.
In several of the cases mentioned above, PS exposure is either transient or reversible, but the PS expression on macrophages is clearly not transient in nature, and is persistent once induced. This is also the case for primary peritoneal macrophages and the J774 macrophage cell line that constitutively express PS, which is required for phagocytosis of apoptotic cells [8 ]. Little is known of the mechanisms responsible for expression of PS on macrophages, which are activated upon differentiation. Presumably, exposure of PS on differentiating U937 cells requires down-regulation of the aminophospholipid translocase activity that normally restricts PS to the internal leaflet, and up-regulation of a scramblase activity to bring PS to the surface, as is the case in activated platelets and apoptotic cells [2 , 3 ]. Testing this presumption may be difficult, given that the low level of PS exposure observed here would not require large-scale or global regulation of these activities. Another protein, the ABC1 or ced-7 ABC ATPase, has been implicated in PS externalization [11 ], and this protein is up-regulated during monocyte/macrophage differentiation [41 ]. Whether up-regulation of ABC1 expression contributes to the appearance of PS on U937 cells is currently under investigation.
A second primary goal of these studies was to establish the functionality of PS expression, that is, its requirement for phagocytosis of apoptotic cells. Expression of PS on differentiating U937 cells after one day of PMA treatment or 7 days of treatment with VitD3/t-RA was accompanied by the development of their ability to phagocytose apoptotic cells. Development of this capacity is specific for apoptotic targets, as undifferentiated U937 cells are already capable of phagocytosis per se. PS expression is specifically required for phagocytosis of apoptotic targets, since using annexin V to mask PS on the surface of differentiated U937 cells and primary monocyte-derived macrophages compromises their ability to phagocytose apoptotic targets, but has no effect on phagocytosis of latex beads, as previously shown with other macrophages [8 ]. Importantly, both PS expression and the ability to phagocytose apoptotic cells were acquired by U937 cells using two different regimes to induce differentiation, indicating that these traits are not a peculiarity of induction by PMA and suggesting that their acquisition is a general feature of monocyte differentiation independent of the mode of induction. However, some variability was seen in the extent to which annexin inhibited phagocytosis. For U937 cells treated for one day with PMA and for primary monocytes differentiated to macrophages, inhibition was nearly quantitative. However, for U937 cells treated for two days with PMA or treated with VitD3 and t-RA, inhibition was only partial. These results suggest that there may be mechanisms for recognizing and phagocytosing apoptotic cells that operate independently of PS expression on macrophages and are variably expressed depending on the mode of induction or extent of differentiation.
CD14 has previously been shown to be necessary for macrophages to phagocytose apoptotic lymphocytes [5 , 25 26 27 ]. In the present study, treating U937 cells for one day with PMA did not induce a noticeable increase in CD14 expression as detected by immunofluorescence and flow cytometry. Still, mAb 61D3, specific for CD14, effectively inhibited phagocytosis of apoptotic cells by U937 cells treated with PMA for one day, suggesting that low level expression of CD14 on U937 cells is sufficient for phagocytic recognition. Interestingly, resonance energy transfer has been used to show that Mac-1 interacts with CD14 in the presence of LPS binding protein or serum upon LPS activation of neutrophils [42 ]. Furthermore, it has been suggested that ß2 integrins, such as Mac-1, may serve as "promiscuous signal transduction devices" for GPI-linked receptors such as CD14 [43 ]. The increase in Mac-1 expression observed here may thus facilitate the successful involvement of a small number of cell surface CD14 molecules in the uptake of apoptotic cells. PS expression may facilitate such cooperation among macrophage receptors, for example by allowing receptors to rearrange on the cell surface, as has been proposed for ligands on apoptotic target cells [9 ].
As shown in the present study, the ability to phagocytose apoptotic cells also requires the development of low levels of PS on the macrophage surface. This feature of monocyte differentiation is a gateway that allows macrophages to achieve an alternative phenotype required for the resolution of inflammatory episodes [44
]. When monocytes are recruited to sites of infection or tissue damage and differentiate into macrophages, they promote inflammation via secretion of proinflammatory cytokines. However, resolution of inflammation requires clearance of the apoptotic leukocytes that gave rise to the inflammatory response, as well as damaged apoptotic cells that were resident in tissues. As macrophages engage in phagocytosis of unlysed apoptotic cells, they actively suppress inflammatory mediator production. For example, when monocyte-derived macrophages phagocytose apoptotic neutrophils in vitro, the production of interleukin (IL)-1ß, IL-8, granulocyte macrophage colony-stimulating factor, and tumor necrosis factor-, as well as leukotriene C-4 and thromboxane B2, is inhibited. In contrast, production of transforming growth factor-ß1, prostaglandin E2, and platelet-activating factor increases, inhibiting proinflammatory cytokine production [45
]. These effects can also occur in vivo. When apoptotic cells are instilled into LPS-stimulated lungs of mice, reduced proinflammatory chemokine levels are seen in bronchoalveolar lavage fluid, and total inflammatory cell counts in the fluid are markedly reduced [46
]. There is also evidence that clearance of apoptotic inflammatory cells is impaired in persistent inflammatory responses such as that seen in cystic fibrosis airways [47
]. Thus, acquisition of PS by differentiating monocytes and transition to an anti-inflammatory phenotype has profound pathophysiological consequences.
Received September 5, 2002; revised July 10, 2003; accepted July 10, 2003.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  L. E. Munoz, S. Franz, F. Pausch, B. Furnrohr, A. Sheriff, B. Vogt, P. M. Kern, W. Baum, C. Stach, D. von Laer, et al. The influence on the immunomodulatory effects of dying and dead cells of Annexin V J. Leukoc. Biol., January 1, 2007; 81(1): 6 - 14. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Boullier, Y. Li, O. Quehenberger, W. Palinski, I. Tabas, J. L. Witztum, and Y. I. Miller Minimally Oxidized LDL Offsets the Apoptotic Effects of Extensively Oxidized LDL and Free Cholesterol in Macrophages Arterioscler Thromb Vasc Biol, May 1, 2006; 26(5): 1169 - 1176. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Tunbridge, T. M. Stevanin, M. Lee, H. M. Marriott, J. W. B. Moir, R. C. Read, and D. H. Dockrell Inhibition of Macrophage Apoptosis by Neisseria meningitidis Requires Nitric Oxide Detoxification Mechanisms Infect. Immun., January 1, 2006; 74(1): 729 - 733. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. R. Boettner, C. D. Huston, J. A. Sullivan, and W. A. Petri Jr. Entamoeba histolytica and Entamoeba dispar Utilize Externalized Phosphatidylserine for Recognition and Phagocytosis of Erythrocytes Infect. Immun., June 1, 2005; 73(6): 3422 - 3430. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Shi, P. G. Gillespie, and A. L. Nuttall Na+ influx triggers bleb formation on inner hair cells Am J Physiol Cell Physiol, June 1, 2005; 288(6): C1332 - C1341. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Dalgaard, K. J. Beckstrom, F. L. Jahnsen, and J. E. Brinchmann Differential capability for phagocytosis of apoptotic and necrotic leukemia cells by human peripheral blood dendritic cell subsets J. Leukoc. Biol., May 1, 2005; 77(5): 689 - 698. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. M. Jackson and J. D. Capra IgH V-Region Sequence Does Not Predict the Survival Fate of Human Germinal Center B Cells J. Immunol., March 1, 2005; 174(5): 2805 - 2813. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Ma, T. Greenwell-Wild, K. Lei, W. Jin, J. Swisher, N. Hardegen, C. T. Wild, and S. M. Wahl Secretory Leukocyte Protease Inhibitor Binds to Annexin II, a Cofactor for Macrophage HIV-1 Infection J. Exp. Med., November 15, 2004; 200(10): 1337 - 1346. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Bondanza, V. S. Zimmermann, P. Rovere-Querini, J. Turnay, I. E. Dumitriu, C. M. Stach, R. E. Voll, U. S. Gaipl, W. Bertling, E. Poschl, et al. Inhibition of Phosphatidylserine Recognition Heightens the Immunogenicity of Irradiated Lymphoma Cells In Vivo J. Exp. Med., November 1, 2004; 200(9): 1157 - 1165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Fan, S. Krahling, D. Smith, P. Williamson, and R. A. Schlegel Macrophage Surface Expression of Annexins I and II in the Phagocytosis of Apoptotic Lymphocytes Mol. Biol. Cell, June 1, 2004; 15(6): 2863 - 2872. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
