Specific involvement of caspases in the differentiation of monocytes into macrophages

A family of cysteine proteases known as caspases plays a central role in many forms of apoptosis.1 These enzymes are synthesized as inactive zymogens that must be cleaved after conserved aspartate residues to be activated. Two main pathways were shown to trigger caspase activation in cells undergoing apoptosis.2 Schematically, the intrinsic pathway involves the disruption of the outer mitochondrial membrane barrier function, thus permitting the release of proapoptotic molecules from the mitochondria to the cytosol. One of these molecules is cytochrome c, which, once in the cytosol, oligomerizes the adaptor molecule Apaf-1 to recruit and activate the initiator caspase-9. In turn, caspase-9 cleaves and activates downstream effector enzymes such as caspase-3. The extrinsic pathway to cell death involves plasma membrane death receptors. In response to their engagement, these receptors trimerize and recruit the adaptor molecule Fas-associated death domain protein (FADD), which, in turn, interacts with and activates an initiator enzyme, usually caspase-8. This enzyme, either directly or through the previously described mitochondrial pathway, activates downstream effector enzymes including caspase-3.3 In both pathways, effector caspases trigger the limited proteolytic cleavage of intracellular structural and regulatory proteins, thus leading to membrane blebbing, chromatin condensation, and nuclear DNA fragmentation, which characterize apoptosis.

Most of the studies concerning caspases have highlighted the relationship between their activation and the occurrence of cell death by apoptosis. However, caspase-1 was initially identified as the protease responsible for the maturation of the multifunctional cytokine interleukin-1β (IL-1β)4 and was subsequently demonstrated to trigger the maturation of IL-18,5 while caspase-3 activation could mediate the maturation of IL-16 in nonapoptotic T cells.6 In addition, caspases were involved in T-cell activation7,8 and various differentiation pathways.9-14 Erythropoiesis was proposed to be regulated by a negative feedback loop in which mature erythroblasts expressing death-receptor ligands inhibit the differentiation of immature erythroblasts through engagement of their death receptors and caspase-mediated degradation of the transcription factor GATA-1.9,10 We have recently identified a distinct process in which a transient activation of caspases that does not lead to GATA-1 cleavage is required for erythroid differentiation.11 Caspase activation was also demonstrated to be required for the terminal differentiation of lens epithelial cells12,13 and keratinocytes.14 

The present study identifies an activation of caspase-3 and caspase-9 in human peripheral blood monocytes induced to differentiate into macrophages in response to macrophage colony-stimulating factor (M-CSF). Caspase activation was not related to apoptosis; nor it was observed in monocytes exposed to IL-4 and granulocyte-macrophage colony-stimulating factor (GM-CSF) for inducing their differentiation into dendritic cells. By using the U937 human monocytic cell line exposed to phorbol ester as a model system, we show that caspase activation actively contributes to the differentiation of monocytes into macrophages.

Etoposide (VP16), cycloheximide (CHX), and 12-O-tetradecanoylphorbol 13-acetate (TPA) were obtained from Sigma-Aldrich Laboratories (St Quentin Fallavier, France) and the caspase peptide inhibitor benzyloxycarbonyl-Val-Ala-_DL-Asp-fluoromethylketone (z-VAD-fmk) from Bachem (Voisins-le-Bretonneux, France). Stock solutions were prepared by diluting these reagents in dimethyl sulfoxide (DMSO) and stored at −20°C. Further dilutions were made in culture medium just before use. The final concentration of DMSO in culture medium never exceed 0.3% (vol/vol), which was nontoxic to the cells. Human recombinant tumor necrosis factor-α (rhTNF-α) was obtained from PeproTech-Tebu (Le Perray-en-Yvelines, France). M-CSF, GM-CSF, and IL-4 were obtained from R&D Systems (Abington, United Kingdom). All other chemicals were of reagent grade and purchased from local sources.

The tested mouse monoclonal antibodies (mAbs) include antihuman HSC70 from Santa Cruz Biotechnology (Santa Cruz, CA), CD11b–fluorescein isothiocyanate (FITC) conjugated from Immunotech (Marseille, France), cytochrome c (clone B75-1) from PharMingen (San Diego, CA), both agonistic (clone CH11) and antagonistic (clone ZB4) anti-Fas receptor from Biovalley (Rockville, MD), and antagonistic anti–TNF receptor-1 (anti–TNF-R1; clone 16803) and anti–TNF-α (clone 1825.12) from R&D Systems. The tested rabbit polyclonal antibodies (pAbs) include antihuman caspase-3 and -9 active fragments (Cell Signaling, Beverly, MA), an antibody that recognizes both caspase-3 full-length and active fragments (PharMingen, San Diego, CA), and an anti–poly(ADP-ribose)polymerase (anti-PARP antibody; Boehringer-Mannheim, Mannheim, Germany). We also used the polyclonal goat antibody raised against the human p23 cleavage fragment of acinus (clone K-14; Santa Cruz Biotechnology).

Human peripheral blood monocytes were obtained from healthy donors with informed consent and purified using a monocyte isolation kit with a light-scattering (LS) column according to the manufacturer's instructions (Miltenyi Biotec, Paris, France) and then incubated (2.5 × 10⁵/mL) with 100 ng/mL M-CSF or with a combination of GM-CSF (100 ng/mL), IL-4 (10 ng/mL), and/or β-mercaptoethanol (50 μM) for up to 7 days.15The human leukemic cell line U937 (CRL-1593.2, mycoplasma free and virus free; American Type Culture Collection [ATCC], Rockville, MD) and a derivative cell clone containing the full-length humanbcl-2 cDNA (U937/Bcl-2, kindly provided by J. Bréard, INSERM U461, Chatenay-Malabry, France) were grown in suspension in RPMI 1640 medium with glutamax-I (Gibco, Life Technologies, Cergy Pontoise, France) supplemented with 10% (vol/vol) fetal bovine serum (BioWhittaker, Fontenay-sous-bois, France) in an atmosphere of 95% air and 5% CO₂ at 37°C. Cell viability was determined by using a trypan blue exclusion assay. To ensure exponential growth, cells were resuspended at a density of 0.5 × 10⁶/mL in fresh medium 24 hours before each treatment. To induce differentiation, cells were cultured in the presence of 20 nM TPA for up to 72 hours. After treatment, adherent cells were harvested using a cell scraper. Cellular differentiation was assessed by May-Grünwald Giemsa staining and CD1a, CD11b, and CD14 plasma membrane expression analysis as described.16 

The FLAG-peptide epitope tagged baculovirus p35 cDNA (kindly provided by J. C. Ameissen, INSERM EPI 99.22, Paris, France) was inserted in a pTarget vector (Promega, Charbonnière, France). Stable transfection was performed by electroporation of 1.0 × 10⁶ U937 cells with 10 μg pTarget containing or not containing p35 cDNA, and selection was done by culturing the cells for 4 weeks in the presence of 1 mg/mL geneticin (G418) to generate mixed cell populations in which expression of p35 was detected by immunoblotting using an anti-FLAG antibody (Boehringer-Mannheim).

Cellular DNA was extracted by a previously described salting-out procedure,17 and electrophoresis was performed in 1.8% agarose gel in Tris (tris(hydroxymethyl)aminomethane)–borate-EDTA (ethylenediaminetetraacetic acid) buffer (pH 8.0) at 20 V for 14 hours. After electrophoresis, DNA was visualized by ethidium bromide staining.

Whole-cell lysates were prepared by lysing the cells in boiling buffer (1% SDS, 1 mM sodium vanadate, 10 mM Tris [pH 7.4]) in the presence of protease inhibitors (0.1 mM phenylmethylsulfonyl fluoride [PMSF], 2.5 μg/mL pepstatin, 10 μg/mL aprotinin, 5 μg/mL leupeptin). The viscosity of the samples was reduced by several passages through a 26-gauge needle. Mitochondrial (Mit) and cytosolic (S100) fractions for cytochrome c release studies were prepared as previously described.18 Protein concentration was measured using the Bio-Rad DC protein assay kit (Ivry sur Seine, France). Thirty microgram proteins were incubated in loading buffer (125 mM Tris-HCl [pH 6.8], 10% β-mercaptoethanol, 4.6% SDS, 20% glycerol, and 0.003% bromophenol blue), separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), and electroblotted to polyvinylidene fluoride (PVDF) membrane (Bio-Rad). After blocking nonspecific binding sites overnight by 5% nonfat milk in TPBS (phosphate-buffered saline [PBS], Tween 20 0.1%), the membrane was incubated for 2 hours at room temperature with the primary Ab. After 2 washes in TPBS, membrane was incubated with horseradish peroxidase–conjugated goat antimouse or antirabbit (Jackson ImmunoResearch Laboratories, West Grove, PA) or swine antigoat (Caltag Laboratories, Burlingame, CA) Abs for 30 minutes at room temperature and then washed twice in TPBS. Immunoblot was revealed using enhanced chemiluminescence detection kit (Amersham, Les Ulis, France) by autoradiography.

Cells were incubated in lysing buffer (150 mM NaCl, 50 mM Tris-HCl [pH 8.0], 0.1% SDS, 1% Nonidet P-40 [NP-40], 0.5% sodium deoxycholate) for 30 minutes at 4°C and centrifuged (10 000g, 20 minutes, 4°C). Fifty microgram proteins of the resulting supernatant was incubated in buffer assay (100 mM HEPES [N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid] [pH 7.0], 1 mM ETDA, 0.1% CHAPS [3-[(3-cholamidopropyl)dimethylammonio]-1-propane sulfate], 10% glycerol, 20 mM dithiothreitol) in the presence of 100 μM fluorogenic peptide substrate (Ac-DEVD-AMC [7-amino-4-methylcoumarin], z-IETD-AFC [7-amino-4-triFluoro-methylcoumarin], z-VDVAD-AFC, and Ac-LEHD-AFC; France Biochem, Meudon, France). AMC and AFC released from the substrate were excited at 380 and 400 nm to measure emission at 460 and 505 nm, respectively. Fluorescence was monitored continuously at 37°C for 30 minutes in a dual luminescence fluorimeter (MicroTek OS; Bio-Tek Kontron Instruments, Winooski, VT). Enzyme activities were determined as initial velocities expressed as relative intensity per minute per milligram.

Immunofluorescence analysis was performed as previously described.19 Briefly, cells were fixed in 2% paraformaldehyde (5 minutes, 4°C), washed twice, saturated in PBS, 0.1% saponin, 1% bovine serum albumin, and incubated for 2 hours at room temperature in the presence of primary antibody diluted in PBS, 0.1% saponin, and 0.5% bovine serum albumin. After washing, cells were incubated for 30 minutes with 488-alexa goat antirabbit Ab and/or 568-alexa goat antirabbit Ab (Molecular Probes, Eugene, OR). Nuclei were stained with Hoechst 33342 (Sigma-Aldrich). Analyses were made by using a fluorescence (Nikon, Champigny, France) or a confocal (Leica, Bron, France) microscope. The caspase-3 and -9 active fragments were also detected by flow cytometry using a Becton Dickinson FACScan cytometer (Le Pont de Claix, France).

Mitochondrial membrane depolarization was analyzed using the DePsipher kit (R&D Systems) according to the manufacturer's instructions. Briefly, 10⁶ cells were incubated for 20 minutes at 37°C in the presence of 5 μg/mL DePsipher solution and then washed twice in PBS. Dihydroethidine was used as a substrate for measuring radical oxygen species (hydrogen peroxide) production. Expression of phosphatidylserine in the external layer of the plasma membrane was determined by studying the fixation of annexin V–FITC in the presence of Ca⁺⁺ in cells that remain nonpermeant to propidium iodide (TACS annexin V–FITC assay, R&D Systems). All these analyses were performed by the use of a FACScan cytometer (Becton Dickinson).

The differentiation of human peripheral blood monocytes into macrophages is associated with caspase activation. Human monocytes were purified from healthy donor peripheral blood and exposed to 100 ng/mL M-CSF to trigger their differentiation into macrophages.15Using a flow cytometry assay, we detected the appearance of caspase-3 active fragments in the cells 24 hours after the beginning of culture in the presence of M-CSF (Figure 1A). The appearance of these active fragments in the cytoplasm was not associated with apoptotic features; for example, in the cells labeled with the antiactive caspase-3 antibody, Hoechst-stained nuclear chromatin did not demonstrate a characteristic apoptotic condensation and the nucleus was not fragmented (Figure 1D). Interestingly, when human monocytes were induced to differentiate into dendritic cells by exposure to 10 ng/mL IL-4 and 100 ng/mL GM-CSF, caspase-3 active fragments could not be identified (Figure 1B-D). Similar results were obtained by using an anticaspase-9 active fragments antibody (Figure 1C-D). These results suggested that the differentiation of monocytes into macrophages could specifically involve the activation of several caspases.

Caspases are activated during TPA-induced U937 cell differentiation. To confirm these observations, we used the U937 human monocytic cells. Exposure of these cells to 20 nM 12-O-tetradecanoylphorbol 13-acetate (TPA) induces their differentiation into macrophages.20 21 The differentiated phenotype includes adhesion of these initially floating cells to the culture flask (Figure2A) and increased expression of the glycoprotein CD11b on their plasma membrane (Figure 2B). To investigate whether caspases were activated during this differentiation process, we first studied the ability of cell lysates to cleave DEVD-AMC, a fluorigenic peptide that mimics the target site of caspase-3 and closely related caspases. Acquisition of the differentiated phenotype was associated with an increase in the DEVD-AMC cleavage activity that appeared between 6 and 12 hours after the beginning of cell treatment, reached a maximum between 12 and 24 hours, and then decreased with time (Figure 2C). The ability of lysates from cells undergoing differentiation to cleave 3 other caspase substrates, namely VDVAD-AFC (caspase-2), LEDH-AFC (caspase-9), and IETD-AFC (caspase-8), was observed to increase along the differentiation process with similar kinetics (Figure 2C). Caspase-3 activation was further confirmed by flow cytometry analysis using an antibody that specifically recognizes active caspase-3 fragments (Figure 2D). The mean fluorescence index increased from 6.0 ± 1.1 at day 0 to 13.9 ± 4.3 at day 2 (mean ± SD of 3 experiments). Lastly, Western blotting identified the 21-, 19-, and 17-kDa cleavage fragments of the protein in U937 cells exposed for 24 hours to 20 nM TPA (Figure 2E).

TPA-induced caspase activation is not associated with apoptosis. To determine whether caspase activation was associated with apoptosis, we performed immunofluorescence microscopic studies. The active fragments of caspase-3 and caspase-9 were detected in the cytoplasm of U937 cells exposed for 24 hours to TPA. In labeled cells, Hoechst 33342 staining of the nuclear chromatin did not show the chromatin condensation or the nuclear fragmentation observed in U937 cells undergoing typical apoptosis under etoposide exposure (Figure 3A-B). A careful analysis of TPA-treated cells demonstrating morphologic apoptotic features indicated that their percentage always remained lower than 4% (Figure 3C, upper numbers). Agarose gel electrophoresis of DNA did not detect oligonucleosomal DNA fragmentation at any step of the differentiation process (Figure 3C), Western blot analysis of the poly(ADP-ribose)polymerase (PARP) nuclear enzyme did not identify any cleavage of the 116-kDa parental protein in an 85-kDa (amino-terminal) fragment (Figure 3C), and cells remained unlabeled with annexin V (Figure 3D) while these 3 events were associated with etoposide-induced U937 cell apoptosis (Figure 3C-D). However, both etoposide-induced apoptosis and TPA-mediated differentiation were associated with the appearance of the 23-kDa active fragment of the protein acinus, a caspase-3–mediated cleavage product of the 98-kDa mature protein (Figure 3C).22 

Differentiation-associated caspase activation involves the mitochondria. In a search for the molecular pathway leading to caspase activation during TPA-induced U937 cell differentiation, we first investigated the role of the mitochondria.23 Exposure of U937 cells to TPA induced a slight decrease in Δψ_m(Figure 4A) and provoked the release of cytochrome c in the cytosol 24 hours after the beginning of TPA treatment (Figure 4B). Bcl-2 overexpression prevented a TPA-induced decrease in Δψ_m (Figure 4A), cytosolic release of cytochrome c (Figure 4B), and caspase-3 cleavage in its active fragments, as determined by flow cytometry (Figure 4C) and immunoblotting (Figure 4D). These results suggested that the pathway leading to caspase activation along the differentiation process involved the mitochondrial release of cytochrome c. We also tested the role of tumor necrosis factor-α (TNF-α) whose secretion was associated with the differentiation process.24 Antibodies targeting either TNF-α or its death domain–containing receptor TNF-R1 were used in conditions that completely inhibit TNF-α–induced caspase activation in U937 cells (Figure 4E). Neither the anti–TNF-α antibody, an anti–TNF-R1 antibody (not shown), nor a combination of both could prevent the appearance of a DEVD-AMC cleavage activity (Figure 4E)—nor CD11b expression increase (not shown) in U397 cells exposed to TPA. Similarly, the blocking anti-Fas antibody ZB4 did not influence caspase-3 activation (Figure 4E) or prevent CD11b up-regulation induced by TPA exposure (not shown). These results indicate that neither TNF-α nor Fas-ligand was involved in the caspase-3 activation identified in U937 cells undergoing TPA-induced differentiation.

Overexpression of Bcl-2, expression of the baculovirus p35, and z-VAD-fmk treatment prevent both caspase activation and TPA-induced U937 cell differentiation. Overexpression of Bcl-2 decreased the percentage of U937 cells that became adherent to the culture flask (Figure 5A) as well as the number of those who expressed CD11b (Figure 5B) in response to TPA exposure. Similarly, expression of the baculovirus caspase inhibitory protein p3525 in U937 cells prevented TPA-induced DEVDase activity (Figure 6A), cell adhesion to the culture flasks (Figure 6B), and CD11b expression increase (Figure 6C). Lastly, the broad-spectrum caspase inhibitor z-VAD-fmk26completely suppressed DEVD-AMC peptide cleavage activity (Figure 6D) as well as other caspase activities (not shown) and the appearance of the 23-kDa acinus fragment (not shown) in TPA-treated U937 cells. The z-VAD-fmk prevented the adhesion of TPA-treated U937 cells in a dose-dependent manner (Figure 6E) and blocked the appearance of CD11b on their plasma membrane (Figure 6F).

The caspase inhibitor z-VAD-fmk induces a switch from differentiation to death. The z-VAD-fmk did not demonstrate any significant toxicity toward U937 cells in the absence of TPA. In contrast, z-VAD-fmk–induced inhibition of TPA-mediated U937 cell differentiation was related to a dose- and time-dependent cell death. This death was characterized by plasma membrane disruption, as assessed by trypan blue staining (Figure 7A-B) and permeability to propidium iodide (Figure 7C), whereas PARP remained uncleaved (Figure 7D). Using dihydroethidine as a substrate for measuring hydrogen peroxide in a flow cytometry assay, we observed that the production of radical oxygen species (ROS) associated with the TPA-induced differentiation process was significantly increased in cells cotreated with z-VAD-fmk (Figure 7E). Similar results were obtained by adding z-VAD-fmk (100 μM) to human monocytes exposed to M-CSF, which slightly increased ROS production (Figure8A) and induced cell death attested by trypan blue staining (Figure 8B). In both TPA-treated U937 cells and M-CSF–treated monocytes, inhibition of ROS production by addition of 25 mM N-acetylcysteine prevented both cell death induced by 100 μM z-VAD-fmk, as attested by trypan blue exclusion, and the differentiation process (not shown).

Several proteins that influence cell viability are also key determinants in cell proliferation and differentiation.27This has been extensively demonstrated with the Bcl-2 family of proteins. These proteins, which modulate activation of the caspase cascade in response to various apoptotic stimuli,28 also interfere with cell cycle progression29 and cell differentiation30; for example, mcl-1 was identified as an early-induced gene in ML-1 human myleoblastic leukemia cells undergoing phorbol ester–induced differentiation,30interacts with proliferating cell nuclear antigen,31 and negatively interferes with cell death pathways.27 The present study demonstrates that activation of caspases, a central event in many cell death processes, also occurs during the differentiation of monocytes into macrophages, in the absence of apoptosis. This activation appears to be specific because it is not observed during the differentiation of monocytes into dendritic cells. The caspase activation pathway involves the mitochondrial release of cytochrome c and is delayed by overexpression of Bcl-2 and the caspase inhibitory protein p35. In addition, high doses of the broad caspase inhibitor z-VAD-fmk induces a switch from cell differentiation to cell death. Altogether, these results argue for an active role of caspases in the maturation of macrophages.

Besides their extensively studied role in apoptosis, caspases have been involved in cytokine maturation and cell cycle regulation.32 In addition, these enzymes have been shown to play an active role in the differentiation of erythroblasts,9-11 lens epithelial cells,12,13 and keratinocytes.14 All these cells lose their nucleus and other organelles upon terminal differentiation.32 These events were regarded as a caspase-mediated incomplete apoptotic process because they can be blocked by caspase inhibitors and Bcl-2 overexpression. We show here that a limited activation of caspases can be detected in monocytes undergoing differentiation into macrophages, a process that does not end with enucleation. A role for caspases has been suspected in other differentiation processes; for example, overexpression of the CrmA viral caspase inhibitor abrogates osteosarcoma cell differentiation induced by a synthetic triterpenoid.33 Coexpression of the various homologs of caspases, redundancies in the pathways, and premature death of animals could have prevented identification of the role of caspases in some terminal differentiation pathways such as that leading to macrophages when studying the phenotype of caspase gene knockout mice.34 35 

TPA ability to trigger the differentiation of U937 monocytic cells into macrophages was demonstrated more than 15 years ago.20However, not all the U937 cell clones respond equally to phorbol esters. In our hands, exposure of U937 cells (CRL-1593.2 from ATCC) to 20 nM TPA induces their strong adhesion to the culture flask and increases CD11b expression on their plasma membrane16,21without inducing significant cell death. Other groups have described an apoptotic response of U937 cells to TPA36 that involves TNF-α as an autocrine apoptosis-stimulating factor.24,36Our observations indicate that caspase activation identified in U937 cells undergoing differentiation does not depend on TNF-α. Our observations also exclude a role for a Fas-L–Fas interaction in the differentiation-associated activation of caspases, which is in accordance with the previously described resistance to Fas that accompanies the differentiation of monocytes into macrophages.37 

Differentiation-related caspase activation may be tightly regulated to prevent cells undergoing differentiation from dying by apoptosis. A coordinated regulation of Bcl-2 and related proteins has been identified in differentiating cells38 and was proposed to control caspase activation associated with lens fiber maturation.13,39 TPA-induced U937 cell differentiation involves a rapid and transient induction of Mcl-1 that precedes caspase-3 activation, whereas the constitutive expression of Bcl-2 decreases belatedly.27,40 The specific depletion of Mcl-1 has been recently shown to induce apoptosis of TPA-treated U937 cells,41 suggesting that Mcl-1 transient induction could prevent excessive caspase activation during the first hours of the differentiation process. Then, the time-dependent decrease in Bcl-2 expression associated with U937 cell differentiation could facilitate the mitochondrial release of cytochrome c. Bcl-2 overexpression inhibits caspase-3 activation and delays the differentiation process, in accordance with the recent observation that Bcl-2 overexpression also prevents bleomycin-induced U937 cell differentiation.42 

Little is known about the proteins cleaved by caspases when activated during the differentiation process. In many examples of apoptosis, one of the caspase target proteins is the nuclear enzyme PARP, a substrate for caspase-3 and related enzymes. Similarly to what is observed in apoptosis, PARP is cleaved from the 116-kDa form to an 85-kDa fragment in lens epithelial cells and erythroblasts undergoing differentiation.11-13 The protein acinus is another caspase-3 target whose cleavage generates an active fragment that contributes to nuclear chromatin condensation associated with cell death.22 Such a cleavage was identified in erythroblasts undergoing maturation.11 We show here that, in U937 cells undergoing differentiation, caspase-3 activation is associated with the appearance of an acinus active fragment similar to that observed in apoptotic cells, whereas PARP remains uncleaved. In human erythroblasts, caspase-3 activation leads to GATA-1 cleavage when induced by death receptor engagement9 while GATA-1 remains uncleaved when caspase-3 is activated along the differentiation process.11 These observations suggest that caspase-mediated cleavage of intracellular target proteins strongly depends on the cellular context, including the differentiation status.

Caspase activation was not identified in human monocytes induced to differentiate into dendritic cells, indicating that caspase activation may not be a general event in cell differentiation. How these enzymes contribute to specific lineage differentiation remains a matter of speculation. First, caspase-mediated proteolytic cleavage of one or several transcription factors could play a role in the lineage-specific program of gene expression.43 Secondly, caspases could influence some of the epigenetic factors that modulate the pattern of gene expression along the differentiation pathways44; for example, the active fragment of acinus could modify the nuclear chromatin structure in a manner that facilitates macrophage lineage specification.22 Third, the cleavage of caspase substrates could generate protective signals that prevent cells undergoing differentiation from dying.45 Low levels of caspase activity, such as those identified in differentiating cells, have already been associated with such a protective effect; for example, the partial cleavage of RasGAP has been shown to generate fragments that prevent cell death.46 One of these protective effects could be to counterbalance ROS production associated with the differentiation of monocytes into macrophages. Although we cannot rule out a nonspecific effect of z-VAD-fmk, the ability of this peptide to increase the production of ROS in differentiating cells suggests that caspase activation could generate a signal that negatively regulates the redox metabolism along the differentiation process.

Whatever the precise function of caspases in the process, the present study indicates that the differentiation of monocytes into macrophages is specifically associated with an activation of these enzymes. This activation must be tightly regulated to prevent premature death of cells committed to differentiate. Further studies will determine how caspases contribute to specific lineage commitment during normal hematopoiesis. These studies could also provide new insights in the mechanisms of caspase activation and abnormal differentiation associated with early phases of myelodysplastic syndromes.47 48 

The authors thank Jacqueline Bréard and Françoise Sainteny for providing cell lines, Anne-Helene Lak-Hal for her help during the course of the study, and the Laboratory of Hematology, Centre Hospitalier Universitaire (CHU) Dijon, for surface marker expression analyses.