The first two authors contributed equally to this work.
Macrophage colony-stimulating factor-dependent macrophage proliferation is mediated through a calcineurin-independent but immunophilin-dependent mechanism that mediates the activation of external regulated kinases
Article first published online: 13 OCT 2003
Copyright © 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
European Journal of Immunology
Volume 33, Issue 11, pages 3091–3100, November 2003
How to Cite
Comalada, M., Valledor, Annabel F., Sanchez-Tilló, E., Umbert, I., Xaus, J. and Celada, A. (2003), Macrophage colony-stimulating factor-dependent macrophage proliferation is mediated through a calcineurin-independent but immunophilin-dependent mechanism that mediates the activation of external regulated kinases. Eur. J. Immunol., 33: 3091–3100. doi: 10.1002/eji.200324074
- Issue published online: 13 OCT 2003
- Article first published online: 13 OCT 2003
- Manuscript Accepted: 8 SEP 2003
- Manuscript Revised: 22 AUG 2003
- Manuscript Received: 4 APR 2003
- Protein kinases/phosphatases;
- Signal transduction;
- Extracellular signal-regulated kinase
Calcineurin is constitutively expressed in bone marrow-derived macrophages. However, macrophage response to macrophage colony-stimulating factor (M-CSF) was not impaired by the use of either calcineurin inhibitors (W-13, chlorpromazine and trifluoperazine), calcium chelators (BAPTA-AM) or Ca2+ channel antagonists (verapamil, nifedipine and diltiazem). Inhibition of calcineurin expression by inhibitory antisense RNA treatment did not result in an inhibition of M-CSF-dependent proliferation. Only very high doses of cyclosporin A and FK506 inhibited macrophage proliferation induced by growth factors, such as M-CSF, granulocyte-macrophage (GM)-CSF or IL-3. This inhibitory action is mediated by the peptidylprolyl isomerase activity of the immunophilins, as demonstrated bythe use of specific inhibitors (rapamycin and sanglifehrin A). These isomerase inhibitors exerted a negative effect on a key element involved in macrophage proliferation, namely the M-CSF-dependent activation of the extracellular signal-regulated kinases (ERK). In summary, the data presented here provide new insights in the mechanism of macrophage proliferation, which may have relevant consequences. First, we showed that in M-CSF-dependent proliferation calcineurin is not involved, and second, that immunophilins play a key role and their activation blocks ERK activation.
Extracellular signal-regulated kinases
Inhibitory antisense RNA
Mitogen-activated protein kinase
Peptidyl-prolyl cis-trans isomerase
Macrophages have critical functions in the immune system. They behave as regulators of homeostasis and as effector cells in infection, wound healing and tumor growth 1. One of the features of the immune system is the production of a large number of any type of cells. Later, most unnecessary cells die through apoptosis. The small number of cells required to develop a functional activity survive by the presence of growth factors and cytokines as well as through interaction with the extracellular matrix 2, 3. Macrophages originatein the bone marrow, and through the expression of a series of transcription factors, in which PU.1 plays a crucial role, a differentiation process takes place where the macrophage colony-stimulating factor (M-CSF) receptors are expressed 4, 5. Macrophages proliferate in the presence of growth factors, such as M-CSF, granulocyte-macrophage (GM)-CSF or IL-3. In tissues, a small number of macrophages differentiate under the influence of cytokines and, depending on the tissue type, they may become osteoclasts (bone), Kupffer cells (liver), microglia (brain),etc.
Although phosphatases play a critical role in cellular physiology, little is known about the function of calcineurin in macrophages. Effects of calcineurin inhibitors on gene expression have been studied in a variety of systems, and in nearly all cases they have been found to be suppressive 6. In macrophages, conflicting results have been obtained. In some studies, calcineurin inhibitors caused extensive inhibition of bacteria-induced expression of IL-1α and TNF-β 7 or the inducible nitric oxide synthase 8 while in others, calcineurin inhibitors enhanced LPS-induced release of IL-10 9, IL-12 and TNF-α 10.
Calcineurin has been shown to be involved in cell cycle progression of several cell types 11–13. However, the role of this phosphatase in macrophage proliferation has not been investigated. Here we found that the macrophage response to M-CSF was not impaired by the use of either calcium chelators or calmodulin inhibitors. High doses of cyclosporin A (CsA) and FK506 inhibited macrophage proliferation induced by growth factors, such as M-CSF, GM-CSF or IL-3. We had shown that the activation of extracellular signal-regulated kinases (ERK)-1 and ERK-2, members of the mitogen-activated protein kinase (MAPK) superfamily, is required for the M-CSF-induced proliferation of macrophages 14, 15. Here we show for the first time that CsA and FK506 block the activation of ERK-1/2 in response to M-CSF, thus explaining their inhibitory effect on macrophage proliferation. However, this inhibitory action is not mediated by calcineurin but by the immunophilins, as demonstrated by the use of specific inhibitors.
2.1 Calcineurin expression in macrophages
The following studies were performed with bone marrow-derived macrophages because they constitute a homogeneous population of primary macrophages that become quiescent in the absence of M-CSF and proliferate again when growth factors are added 4. Although primary cultures are the best model to study the mechanisms involved in macrophage proliferation and survival, their transfection is highly inefficient 4. For this reason, in our experiments we used chemical inhibitors to assay the involvement of specific molecules in macrophage biology.
Phosphatases play a key role in the proliferation induced by growth factors and cytokines. We aimed to determine the role of calcineurin in macrophage proliferation. Previously, we described the critical role of MAPK phosphatase (MKP)-1 in the regulation of macrophage activities 14, 15. Northern blot analysis of macrophages stimulated with M-CSF for 30 min showed that MKP-1 expression is inducible by growth factors (Fig. 1), a finding that is consistent with those of other studies 16, 17. Using a probe for calcineurin, we observed a band of 3.6 kb, which corresponded to the size of the mRNA of this phosphatase 18, 19 (Fig. 1). In contrast to MKP-1, calcineurin showed a constitutive expression that was not modified after M-CSF treatment (Fig. 1). No modifications of calcineurin expression were observed after either LPS or IFN-γ treatment (data not shown).
2.2 ERK activation is necessary for macrophage proliferation
One of the earliest events in the signaling response to M-CSF is the induction of ERK activity 20, 21. To determine the time course of ERK activation in bone marrow macrophages stimulated with M-CSF, an in-gel kinase assay, using myelin basic protein as substrate, was used. ERK activation was detected as soon as 2–3 min after M-CSF treatment and decreased progressively after 20 min of stimulation (Fig. 2A). Some residual activity was observed at prolonged times of M-CSF treatment. Similar results were obtained when the phosphorylation state of ERK was measured either by a mobility shift assay or using an antibody that recognizes the phosphorylated form of ERK (data not shown). A similar pattern of ERK activation was obtained with GM-CSF or IL-3.
The next step was to find out whether this activation of ERK was necessary for the proliferative response of macrophages to M-CSF. Activation of the ERK pathway was blocked by incubating the cells with the MAPK/ERK kinase (MEK) inhibitor PD98059 (Fig. 2B). In the presence of M-CSF, bone marrow macrophages proliferated in a dose-dependent manner, as measured by thymidine incorporation (Fig. 2C). This method gives an efficient indication of macrophage proliferation as previously determined 4 and correlates with the counting of the number of cells. Macrophages pre-incubated with PD98059 were unable to proliferate in response to M-CSF (p<0.01), showing that activation of the MEK/ERK pathway is necessary for M-CSF-dependent macrophage proliferation (Fig. 2C). Similar results were obtained when we determined the proliferative activity of M-CSF, GM-CSF or IL-3 by cell counting.
The induction of apoptosis was analyzed using a commercial ELISA kit that measures DNA fragmentation caused by internucleosomal cleavage. Apoptosis was observed in cells deprived of M-CSF for 24 h (positive control), but not in cells growing in the presence of M-CSF or treated with PD98059 (Fig. 2D). Using cells stained with DAPI, a fluorescent dye that specifically binds to DNA, and measuring the cellular DNA content by flow cytometry, only a subdiploid peak representing apoptotic cells (Sub-G1) was observed in cells depleted of M-CSF for 24 h, but not in cells incubated with M-CSF or M-CSF and PD98059 (data non show). Thus, these results show that blockage of ERK activation inhibits M-CSF-induced proliferation without causing apoptosis.
2.3 Calcineurin is not involved in macrophage proliferation
Although calcineurin is constitutively expressed in macrophages, its activation requires the calcium/calmodulin complex. In a first attempt to determine the role of calcium/calmodulin on macrophage proliferation, we treated these cells with the specific inhibitor W-13, a sulfonamide that penetrates the membrane and is mainly concentrated in the cytosol 22, 23. The treatment of M-CSF-activated macrophages with this compound did affect proliferation, as measured by cell counting or thymidine incorporation (Fig. 3A), or the activation of ERK-1/2 (Fig. 3B) without inducing apoptosis. We also used two other calmodulin inhibitors, chlorpromazine and trifluoperazine, two phenothiazines that are structurally distinct from W-13. No effect on macrophage M-CSF-dependent proliferation was found when the phenothiazines were added to the medium (Fig. 3C). To assess the efficiency on calcineurin activity of the drugs used in this study, we looked at the levels of dephosphorylated, LPS-activated NF-AT by Western blot as described 10 (data not shown). Finally, we used the calcium chelator BAPTA-AM, which renders calmodulin inactive. Concentrations up to 1 μM of this compound did not inhibit proliferation in response to M-CSF (Fig. 3D).
Ca2+ channel antagonists are a group of drugs that consists of three main structurally unrelated types. These are the phenylalkylamines (verapamil), dihydropyridines (nifedipine) and the benzothiazipines (diltiazem) 24. These compounds bind to specific receptors of the Ca2+ channel located in the plasma membrane and thereby frustrate transmembrane calcium fluxes 25. In the presence of verapamil at concentrations up to 2 μg/ml, M-CSF-dependent proliferation was not affected (Fig. 4A). This drug was also unable to block M-CSF-dependent ERK activation. In addition, verapamil did not induce or protect from M-CSF withdrawal apoptosis (Fig. 4B). Similar results were found when we used the other two Ca2+ channel antagonists, nifedipine and diltiazem (data not shown). Finally, using the double-wavelength fluorimetry method 26 we were unable to detect mobilization of intracellular calcium during macrophage response to M-CSF (Fig. 4C).
2.4 CsA and FK506 at high doses inhibit macrophage proliferation
To corroborate these results, we blocked calmodulin activity using the inhibitors CsA and FK506. At 0.1 μg/ml of CsA or 0.1 μM of FK506, concentrations that inhibit T lymphocyte activation 13, neither inhibitor affected macrophage proliferation (Fig. 5A). To exclude any possible involvement of calcineurin in macrophage proliferation, we used these drugs at higher concentrations. Under these conditions thymidine incorporation was significantly inhibited (p<0.01; Fig. 5A) and the number of cells that grew in the presence of M-CSF was reduced (p<0.01; Fig. 5B). Similar results on proliferation were found when recombinant GM-CSF or IL-3 were used as growth factors (p<0.01; Fig. 6). When we tested the effect of CsA and FK506 on the levels of dephosphorylated, LPS-activated NF-AT, we found that concentrations of 0.1 μg/ml of CsA and 0.1 μM of FK506 were effective (data not shown).
Cell counting by Trypan blue exclusion (Fig. 5) already suggested that cell death or reduced viability was not extended in macrophages treated at high doses of these drugs. However, when assessing whether the inhibition of macrophage proliferation could be due to the induction of apoptosis, we found that CsA did not induce detectable levels of apoptosis in macrophages (Fig. 7A). In a dose-dependent manner, CsA partially protected from apoptosis induced by the absence of M-CSF (Fig. 7A). These results indicate that the inhibition of macrophage proliferation by CsA is not caused by an increase in cell death. At the highest concentration (10 μg/ml), FK506 induced low levels of apoptosis in bone marrow macrophages cultured in the presence of M-CSF (Fig. 7B). However, the extent of apoptosis induced by this drug was not sufficient to account for the global effect observed on macrophage proliferation at this concentration. Thus, although we used high doses of CsA and FK506, the viability of the cells was not compromised and our results are not attributable to the toxic effects of these drugs.
2.5 CsA and FK506 inhibit ERK activation
ERK are activated in response to M-CSF, GM-CSF or IL-3, and as we observed, the activation of ERK-1/2 is critical for the proliferation of bone marrow macrophages 15. For this reason, we studied the effect of CsA and FK506 on the activation of this pathway. M-CSF induced a transient and acute peak of ERK activation as analyzed by an in-gel kinase assay using myelin basic protein as substrate 14, 15 (Fig. 8A). CsA at a concentration that inhibits macrophage proliferation (10 μg/ml) blocked ERK-1/2 activation induced by M-CSF (Fig. 8A). Thus, the effect of this inhibitor on the activation of ERK correlates with its negative effect on macrophage proliferation. In addition, we examined whether CsA affects ERK activity or, in contrast, directly ERK expression. Western blot analysis showed that CsA treatment did not modify the levels of ERK expression in macrophages (Fig. 8B). FK506 at a concentration that blocks macrophage proliferation also inhibited M-CSF-induced ERK activity (Fig. 8C) and, similar to CsA, FK506 alone did not induce ERK activity. Again, the FK506 treatment did not modify the expression levels of ERK in macrophages (Fig. 8D).
We found that macrophage proliferation was only affected by CsA or FK506 at higher doses than those required to inhibit calcineurin function; therefore molecules other than calcineurin may account for the biological effects of CsA and FK506 on macrophages. To confirm the role of calcineurin in macrophage proliferation, cells were treated with two different inhibitory antisense RNA (iRNA) directed against the calcineurin coding sequence. iRNA treatment of macrophages resulted in a reduction of mRNA levels (Fig. 9A). [3H]Thymidine incorporation was studied with and without M-CSF-stimulation. Macrophage proliferation was not affected in iRNA-treated cells compared with control cells transfected with an unrelated oligonucleotide (Fig. 9B). Similar results were obtained by counting the cells, and the treatment with iRNA did not induce cell mortality. Therefore, calcineurin is not involved in M-CSF-dependent macrophage proliferation.
2.6 Rapamycin and sanglifehrin A inhibit ERK activation and macrophage proliferation
Peptidyl-prolyl cis-trans isomerase (PPIase), which is essential for the refolding of certain proteins, is identical to cyclophilin, the intracellular binding molecule for CsA 27. Since binding of CsA inhibits PPIase activity 27, CsA may also exerts its effect by inhibiting the isomerase-dependent refolding of proteins. To test this hypothesis we treated macrophages with rapamycin, which inhibits the isomerase activity but has no effect on calmodulin activity 28, 29. In a dose-dependent manner, rapamycin significantly inhibited M-CSF-dependent macrophage proliferation as determined by thymidine incorporation (p<0.01; Fig. 10A) or cell counting. The inhibition of cell proliferation is not due to an increase in apoptosis (Fig. 10B). Like CsA and FK506, rapamycin inhibits ERK activation (Fig. 10C). To confirm these data we used sanglifehrin A, which is a novel cyclophilin-binding immunosuppressant with a different mechanism of action from that of rapamycin 30–32. Again, in a dose-response manner, sanglifehrin A significantly inhibited M-CSF-dependent proliferation as measured by thymidine incorporation (p<0.01) or by counting the number of cells (data not shown) without a reduction in cell viability. As shown in the drug blocked the M-CSF-dependent ERK activation.
The phosphorylation-dephosphorylation of signal transduction proteins plays a critical role in cell response to distinct growth factors or cytokines. The dual-specificity tyrosine/serine MKP-1 plays a crucial role in the regulation of macrophages towards proliferation versus activation 14, 16. However, several other phosphatases may play critical physiological roles in macrophages. For instance, the serine/threonine phosphatase PP2A is a regulator of LPS-induced activation of c-Jun N-terminal kinase and cytokine expression 33. Calcineurin (protein phosphatase 2B) is the only serine/threonine phosphatase under the control of Ca2+/calmodulin and is an important mediator in signal transmission, connecting the Ca2+-dependent signaling to a wide variety of cellular responses 34, 35. We have found that calcineurin is constitutively expressed on bone marrow-derived macrophages; and to determine the role of this phosphatase on macrophage proliferation, we used several calmodulin inhibitors. Neither calcium chelators nor the inhibition of calmodulin-dependent pathways altered the macrophage response to M-CSF, suggesting that calcineurin is not involved in macrophage proliferation. This was confirmed using small inhibitory RNA to inhibit calcineurin expression.
Our results also show that the classical calmodulin inhibitors CsA and FK506 at concentrations sufficient to inhibit calcineurin 13 do not block macrophage proliferation. However, we observed, for the first time, that higher doses of these drugs inhibited not only M-CSF-, but also GM-CSF- or IL-3-dependent proliferation. The arrest of proliferation is not related to an increase of apoptosis induced, at least by CsA, and the slight increase in apoptosis induced by FK506 treatment was not sufficient to explain the effect on macrophage proliferation. Moreover, cell counts by Trypan blue exclusion confirmed the lack of toxicity of the drugs at the high doses used. Several studies have reported a variety of effects on macrophages by blocking calcineurin with CsA or FK506; however, the concentrations of these calcineurin inhibitors needed to block specific aspects of the biology of monocytic/macrophagic cells are higher than those required to inhibit the activities of other cells of the immune system, including T lymphocytes 36.
Although it is assumed that CsA and FK506 induce their biological effects through the inhibition of a common target, calcineurin, we have no evidence that the actions observed in macrophages are mediated through this mechanism. Although calmodulin is involved in the regulation of cell cycle in fibroblasts 37, this has not been confirmed in all cellular models 38. On the basis of our results we propose that, in macrophages, CsA and FK506 act on distinct targets independently of calcineurin. At high doses, these drugs are inhibitors of multidrug resistance proteins 39. However, inhibition with verapamil did not affect macrophage proliferation, thereby excluding a role of these proteins in the effect of these calcineurin inhibitors on macrophages 40, 41.
There are other targets for the effects of calcineurin inhibitors. PPIase, which is essential for the refolding of certain proteins, is identical to cyclophilin, the intracellular binding molecule for CsA 27. Since binding of CsA inhibits PPIase activity 27, CsA may also exerts its effect by inhibiting the isomerase-dependent refolding of proteins. Inhibition of PPIase activity may prevent certain proteins from achieving full activity. Rapamycin, which inhibits the isomerase activity but has no effect on calmodulin activity 28, 29, inhibits M-CSF-dependent macrophage proliferation without induction of apoptosis. Also, sanglifehrin A, which is a novel cyclophilin-binding immunosuppressant with a different mechanism of action from that of rapamycin 30–32, altered the macrophage response to M-CSF. All these data demonstrate that isomerase, a new component that has not been previously identified, plays a critical role in macrophage proliferation.
Moreover, although cyclophilin and FKBP12 are the best characterized immunophilins, other related members are thought to be expressed less abundantly in a number of cell types 42. Here we have demonstrated for the first time that isomerase inhibitors exert a negative effect on a key element involved in macrophage proliferation such as the ERK pathway. Although we couldnot present any data regarding the target that isomerase inhibitors use to block macrophage proliferation, our results indicate that their effect is mediated by disturbing the ERK pathway. In this regard, it has recently been reported that a novel immunophilin could bind and regulate Raf-1, a MAPK kinase upstream of the ERK pathway 43, 44. In summary, the data presented here provide new insights in the mechanism of macrophage proliferation, which may have relevant consequences. First, we showed that in M-CSF-dependent proliferation calcineurin is not involved, and second, that immunophilin plays a key role and its activation blocks ERK activation.
4 Materials and methods
4.1 Cells and reagents
Bone marrow-derived macrophages were obtained from BALB/c mice (Charles River Laboratories Inc., Wilmington, MA) as described 45. Cells were rendered quiescent by depriving them of L-cell-conditioned medium for 16–18 h. Analysis of intracellular Ca2+ by double-wavelength fluorimetry was performed as described 26. The use of animals was approved by the Comitè d'Experimentació Animal of the University of Barcelona with the number 2523.
CsA and sanglifehrin A were a kind gift from Novartis (Basel, Switzerland). FK506, BAPTA-AM, Verapamil, A23187 and W-13 were purchased from Calbiochem (La Jolla, CA). LPS, 2-chloro-10-(3-dimethylaminopropyl) phenothiazine hydrochloride (chlorpromazine), trifluoperazine dihydrochloride, diltiazem, nifedipine and rapamycin were obtained from Sigma Chemical Co. (St. Louis, MO). These drugs do not affect cell viability at the concentrations and for the time of incubation used in these studies. For some experiments we used recombinant M-CSF, GM-CSF and IL-3 (DNAX, Palo Alto, CA).
4.2 Proliferation assay
4.3 Detection of apoptosis by analysis of chromatin fragmentation
Fragmentation of DNA caused by internucleosomal cleavage was measured using a commercial ELISA kit (Cell Death Detection ELISA Kit plus; Boehringer Mannheim, Indianapolis, IN), as previously described 47. Each point was performed in triplicate and the results were expressed as the mean ± SD.
4.4 RNA extraction and Northern blot analysis
Total RNA was extracted by the acidic thiocyanate-phenol-chloroform method 48. Total RNA samples (15 μg) were separated on 1.2% agarose gels containing formaldehyde and transferred to nylon membranes (Genescreen; NEN Life Science Products, Boston, MA). For MKP-1 mRNA detection, we obtained the full-length cDNA fragment of MKP-1 following purification from a Hind III digestion of the plasmid pBSKS/MKP-1 (kindly provided by Dr. R. Bravo, Bristol-Myers Squibb, Princeton, NJ). The probe for calcineurin was obtained by reverse transcription (RT)-PCR using total RNA from macrophages and the following primers: forward, 5′-TATGACGCCTGTATGGATGCC-3′; and reverse, 5′-GGAGCCAGTACGGATGCGGGG-3′. As control we used an 18S rRNA probe 49. The probes were labeled with [α-32P]dCTP (ICN Pharmaceuticals, Costa Mesa, CA). After incubation in hybridization solution (20% formamide, 5× Denhart's, 5× SSC, 10 mM EDTA, 1% SDS, 25 mM Na2HPO4, 25 mM NaH2PO4 and 0.2 mg/ml salmon sperm DNA) at 65°C, the membranes were washed and exposed to Kodak X-AR films (Kodak Company, Rochester, NY).
4.5 Western blot analysis
Western blot analysis was performed as previously described 15. For analysis of ERK-1 expression we used a mouse monoclonal antibody against ERK-1 (Santa Cruz Biotechnology,Santa Cruz, CA). A mouse monoclonal antibody against β-actin (Sigma) was used as control. Peroxidase-conjugated anti-mouse IgG (Cappel-Organon Teknik, Durnham, NC) was used as secondary antibodies.
4.6 Measurement of ERK activity by in-gel kinase assay
This assay was performed as previously described 16. Total protein was separated by SDS-PAGE in the presence of myelin basic protein co-polymerized in the gel. To perform the phosphorylation assay, the gel was incubated in a solution containing [γ-32P]ATP.
4.7 Transfection of macrophages with small inhibitory antisense RNA
iRNA were prepared by Dharmacon (Lafayette, Colorado). The iRNA sequences were for CaN1: 5′-AACCUCGUGUGGAUAUCUUdTdT-3′; and for CaN2: 5′-AACAAGAUCCGAGCAAUAGdTdT-3′. Cells were transfected by electroporation 4. Cells (4×106) and 1.5 μM iRNA were resuspended in 400 μl and pulsed once at 350 V, 2,300 μF with a BTX ECM 600 electroporotor (BTX, San Diego, CA).
4.8 Statistical analysis
To calculate the statistic differences between the control and treated samples we used the Student's paired t-test. Values of p<0.05 were interpreted as significant.
This work was supported by grants from Ministerio de Ciencia y Tecnología (BMC2001–3040) and La Marató de TV3 (1010) to A.C. We thank Novartis (Basel, Switzerland) for kindly providing cyclosporin A and sanglifehrin A. We thank Tanya Yates for editing the English manuscript.
- 22Modulation of Ca2+-dependent regulatory systems by calmodulin antagonists and other agents. In Hidaka, H. and Hartshorne, D. (Eds.) Calmodulin antagonists and cellular physiology. Academic Press, New York 1985, pp 13–23.
- 24Calcium channel ligands. Annu. Rev. Pharmacol. Toxicol. 1987. 29: 349–369.
- 36Action of cyclosporin A on mononuclear phagocytes. Investig. Allergol. Clin. Immunol. 1996. 6: 222–231.
- 39Pharmacology of drugs that alter multidrug resistance in cancer. Pharmacol. Rev. 1990. 2: 155–199.