- db cAMP:
electrophoretic mobility shift assay
melanocyte growth stimulatory activity
nitroblue tetrazolium (test)
phenyl methyl sulphonyl fluoride
secretory alkaline phosphatase
TNF-related apoptosis-inducing ligand
Considering the role of interleukin-8 (IL-8) in a large number of acute and chronic inflammatory diseases, the regulation of IL-8-mediated biological responses is important. Alpha-melanocyte-stimulating hormone (α-MSH), a tridecapeptide, inhibits most forms of inflammation by an unknown mechanism. In the present study, we have found that α-MSH interacts predominantly with melanocortin-1 receptors and inhibits several IL-8-induced biological responses in macrophages and neutrophils. It down-regulated receptors for IL-8 but not for TNF, IL-4, IL-13 or TNF-related apoptosis-inducing ligand (TRAIL) in neutrophils. It down-regulated CXCR type 1 and 2 but not mRNA levels. α-MSH did not inhibit IL-8 binding in purified cell membrane or affinity-purified CXCR. IL-8 or anti-CXCR Ab protected against α-MSH-mediated inhibition of IL-8 binding. The level of neutrophil elastase, a specific serine protease, but not cathepsin G or proteinase 3 increased in α-MSH-treated cells, and restoration of CXCR by specific neutrophil elastase or serine protease inhibitors indicates the involvement of elastase in α-MSH-induced down-regulation of CXCR. These studies suggest that α-MSH inhibits IL-8-mediated biological responses by down-regulating CXCR through induction of serine protease and that α-MSH acts as a potent immunomodulator in neutrophil-driven inflammatory distress.
Defects in neutrophil migration are the best evidence of human neutrophil dysfunction in innate immunity 1. In contrast, during chronic inflammatory diseases such as rheumatoid arthritis, gout, asthma or inflammatory bowel disease, uncontrolled accumulation of neutrophils as well as residential macrophages at the site of infection liberate inflammatory molecules such as cytokines, reactive oxygen intermediates and proteolytic enzymes, which become major contributors to tissue damage 2, 3. Thus, regulation of neutrophil and macrophage recruitment into inflammatory sites and their clearance are critical processes assuring effective host defense without tissue injury. Extravasation of neutrophils from blood vessels requires adhesion to vascular endothelial cells and subsequent migration of these cells into the tissue 4. These events are mediated mostly by interleukin-8 (IL-8) 5, 6, a potent chemotactic agent for neutrophils 7; it triggers respiratory burst response, degranulation and neutrophil adhesion to the endothelial cells 8–10. IL-8 interacts with its receptor, IL-8R, also known as CXCR. The IL-8R are of two types, type A and type B. The type A receptor (IL-8R1 or CXCR1) binds IL-8 with high affinity but shows low affinity to melanocyte growth stimulatory activity (MGSA), whereas the type B receptor (IL-8R2 or CXCR2) binds to IL-8 and MGSA with high affinity 11. CXCR are rapidly internalized with their ligands, and the endocytosed receptors are recycled back to the cell surface. The cyclical process of endocytosis and subsequent recycling of receptors is intimately associated with the IL-8-mediated chemotactic response 12, 13. Since the IL-8-mediated response is related to the expression of functionally active CXCR, modulation of these receptors at the ligand interacting domain may be a viable strategy to regulate IL-8-induced inflammatory responses.
Nuclear transcription factor-kappa B (NF-κB) regulates the expression of various genes that play critical roles in inflammation, viral replication, tumorigenesis and apoptosis 14–16, and therefore this factor is an ideal target of pharmaceutical interest 17. IL-8 production from diverse cells is regulated by NF-κB, but information on IL-8-induced NF-κB activation is meager. Recently, we reported that IL-8 induces NF-κB through recruitment of TNF receptor-associated factor 6 (TRAF6) and IL-1 receptor-activating kinases (IRAK) 18. A possible strategy for ameliorating inflammatory distress may be the prevention of excessive neutrophil migration by reducing the interaction of neutrophils with the inflammatory cytokine or by regulating IL-8 production. Neurohormones have been found to modulate the immune system 19. Alpha-melanocyte-stimulating hormone (α-MSH) interacts with various cells of the immune system and down-regulates either the production or the action of pro-inflammatory cytokines such as IL-1, TNFα and IL-6 20–23, thus acting as an anti-inflammatory agent. Receptors for α-MSH have been detected on both monocytes/macrophages and neutrophils 21, 23. At the molecular level, the α-MSH-mediated regulation of inflammation induced by different stimuli is not understood. α-MSH down-regulates TNF-induced NF-κB activation and inhibits TNF-induced apoptosis by up-regulating cyclic AMP (cAMP) 24. Recently, we reported that α-MSH mediates anti-inflammatory responses by down-regulating CD14, the endotoxin receptor, from the surface of macrophages 25 and induces mast cell apoptosis 26. Activated neutrophils release several proteolytic enzymes that cleave protease-activated receptors (PAR) and activate downstream signaling 27. Therefore, involvement of proteases in down-regulation of the CXCR can't be ruled out.
In this report, we demonstrate for the first time that α-MSH inhibits IL-8-mediated biological responses by down-regulating CXCR from neutrophils and macrophages. Down-regulation of CXCR by α-MSH is mediated through activation of neutrophil elastase, which cleaves surface CXCR, and the CXCR are protected in the presence of N-methoxysuccinyl-Ala-Ala-Pro-Val-chloromethyl ketone (CMK), a specific neutrophil elastase inhibitor, and phenyl methyl sulphonyl fluoride (PMSF), a serine protease inhibitor. In this study, we aimed to understand the anti-inflammatory effect of α-MSH on neutrophils and macrophages isolated from human blood. Overall, our data suggest the down-regulation of CXCR by α-MSH through activation of serine protease, followed by inhibition of IL-8-mediated biological responses in macrophages and neutrophils. Modulation of this activity might be useful for therapy of neutrophil- and/or macrophage-driven inflammatory diseases.
α-MSH inhibits IL-8-induced biological responses
In this study, we examined the effect of α-MSH on IL-8-induced biological responses in macrophages and neutrophils. HL-60 cells were converted into neutrophils by incubation with DMSO for 2 days, and THP1 cells were induced to convert into macrophages through incubation with PMA for 16 h. These cells were used for most experiments. After conversion, neutrophils were cultured for up to 48 h without significant increase in cell death (cell viability was 99.02±2.24, 97.34±3.36, 94.78±3.24 and 85.46±5.52% at 0, 24, 36 and 48 h incubation respectively). IL-8 interacts with CXCR and induces biological responses in neutrophils. To detect the role of α-MSH in IL-8-mediated biological responses, neutrophils and macrophages were used to assay NF-κB activation, cell migration, proteolytic enzymes release and oxidative burst response.
α-MSH inhibits IL-8-induced NF-κB activation
To determine the role of α-MSH in IL-8-induced NF-κB activation in neutrophils and macrophages, cells were incubated with different concentrations of α-MSH for 24 h and then stimulated with IL-8 (100 ng/mL) for 2 h at 37°C. Nuclear extract (NE) was prepared and assayed for NF-κB by gel shift assay as described in the Materials and methods. IL-8 induced NF-κB in neutrophils (Fig. 1A) and macrophages (Fig. 1B). α-MSH alone did not activate NF-κB, but IL-8-induced NF-κB activation was inhibited in a dose-dependent manner, and at 100 nM concentration, α-MSH completely abrogated IL-8-induced NF-κB activation in both cell types. The activated band as shown by electrophoretic mobility shift assay (EMSA) was composed of p50 and p65 as shown by supershift using anti-p50 and/or anti-p65 Ab (data not shown). To ensure that the NF-κB activation was due to IL-8 and not endotoxin, cells were stimulated with IL-8 or IL-8/polymixin B sulphate mixture. The NE was then assayed for NF-κB. IL-8 pre-incubated with polymixin B sulphate activated NF-κB similarly to IL-8 alone, indicating that the IL-8-mediated NF-κB activation is not due to endotoxin (Fig. 1C).
α-MSH inhibits IL-8-induced myeloperoxidase, alkaline phosphatase and β-D-glucuronidase activity
To analyze the α-MSH-mediated effect on IL-8-induced proteolytic enzymes release, untreated and α-MSH-treated neutrophils were stimulated with different concentrations of IL-8 for 2 h, and culture supernatant was assayed for myeloperoxidase, alkaline phosphatase and β-D-glucuronidase (Fig. 1D). The absorbencies for all three enzymes were enhanced with increasing concentrations of IL-8, whereas neutrophils pre-treated with α-MSH showed no induction of enzymatic activity at any concentration of IL-8. α-MSH alone marginally increased the myeloperoxidase level in neutrophils. From these results it is clear that IL-8-mediated proteolytic enzyme release is inhibited by α-MSH.
α-MSH inhibits IL-8- but not formyl peptide (FMLP)-induced neutrophil migration
IL-8- and FMLP-directed neutrophil migration was assayed in Boyden chemotactic chambers 28. The number of IL-8- or FMLP-induced migrated cells (as shown by chemotactic index: migrated cell numbersinduced / migrated cell numbersuninduced) increased in a time-dependent manner. Cells pre-treated with α-MSH showed a 60–70% decrease in migration in response to IL-8 but not FMLP at all times of incubation (Fig. 2A). This result suggests that α-MSH inhibits IL-8- but not FMLP-induced neutrophil migration, which correlates with down-regulation of NF-κB activation and proteolytic enzyme release.
α-MSH inhibits IL-8- but not FMLP-induced oxidative burst response
IL-8- or FMLP-induced generation of reactive oxygen species (ROS) in neutrophils is well established 28. The generation of ROS is detected by the NBT (nitroblue tetrazolium) test. To address the issue of whether α-MSH modulates only IL-8-mediated biological responses or also affects other chemoattractant-mediated biological responses, neutrophils treated with different concentrations of α-MSH for 24 h were stimulated with IL-8 or FMLP for 2 h at 37°C in the presence of NBT solution (0.1%). NBT-positive cells were counted under the microscope, and the percentage of positive cells is shown (Fig. 2B). α-MSH did not increase the number of NBT-positive cells among unstimulated cells. IL-8 and FMLP stimulation induced 66% and 80% increases in NBT-positive cells, respectively. α-MSH treatment reduced the percentage of IL-8- but not FMLP-induced NBT-positive cells. These data indicate that α-MSH down-regulates the IL-8- but not FMLP-induced oxidative burst response, suggesting the specific activity of α-MSH.
α-MSH is unable to block FMLP-induced NF-κB activation
As α-MSH was unable to inhibit FMLP-induced neutrophil migration and oxidative burst response, FMLP-induced NF-κB activation was also assayed. α-MSH-treated cells did not block FMLP-induced NF-κB activation at any concentration (Fig. 2C), further suggesting specific action of α-MSH on IL-8-induced biological responses.
α-MSH inhibits IL-8- but not FMLP-induced NF-κB-dependent secretory alkaline phosphatase (SEAP) activation
To further prove α-MSH-mediated inhibition of NF-κB activation, HL-60 cells were transfected with an NF-κB reporter plasmid containing the SEAP gene and induced to differentiate into neutrophils as described above. The cells were then treated with α-MSH and stimulated with IL-8 or FMLP for 12 h, and the culture supernatant was assayed for SEAP activity. Both IL-8 (Fig. 2D1) and FMLP (Fig. 2D2) induced SEAP activity in a dose-dependent manner, but only IL-8-induced SEAP activity was completely blocked in cells pre-treated with α-MSH. The results suggest that α-MSH inhibits IL-8-induced NF-κB-dependent gene expression.
α-MSH inhibits IL-8 but not TNF, TNF-related apoptosis-inducing ligand (TRAIL), IL-4 or IL-13 binding
As α-MSH inhibits IL-8-induced biological responses and IL-8 exerts its effects through surface CXCR, the effect of α-MSH on expression of different receptors on neutrophils was analyzed. Neutrophils treated with different concentrations of α-MSH for 24 h in triplicate samples were incubated with 125I-labeled TNF, TRAIL, IL-4, IL-13 or IL-8 (5 × 104 cpm/tube), and binding was detected. 125I-IL-8 binding was decreased with increasing concentrations of α-MSH (79% at 1000 nM; Fig. 3A), while binding of TRAIL, IL-4 and IL-13 was unaffected. A marginal decrease in TNF binding was observed in the presence of 1000 nM α-MSH (16%). The results indicate that α-MSH inhibits binding of IL-8 but not TNF, TRAIL, IL-4 or IL-13 in neutrophils in a dose-dependent manner.
To determine the optimal incubation time for α-MSH treatment (resulting in maximal inhibition of IL-8 binding), neutrophils were treated with 1000 nM α-MSH for increasing lengths of time, and 125I-IL-8 binding was assayed at 4°C. As shown in Fig. 3B, IL-8 binding decreased gradually up to 24 h of treatment. From this result it is clear that the treatment time required for maximal inhibition of IL-8 binding by α-MSH is 24 h. Viable receptors were detected by accumulation of labeled ligand inside the cells (total binding). As shown in Fig. 3C, the accumulated 125I-IL-8 was decreased in α-MSH-treated cells by about 70–80% compared to untreated cells at any concentration of added 125I-IL-8, further suggesting down-regulation of CXCR by α-MSH.
α-MSH interacts predominantly with melanocortin-1 receptors (MC-1R) in neutrophils and macrophages
To determine the type of MCR used by α-MSH in neutrophils and macrophages to down-regulate CXCR, different cell types (THP-1, HL-60, differentiated neutrophils and macrophages, isolated neutrophils and macrophages from blood, A375 melanoma cells) were cultured in complete medium for 24 h at 37°C, and extracts were analyzed for MC-1R, MC-2R, MC-3R or MC-4R by Western blot. The level of MC-1R was increased in differentiated and isolated neutrophils and macrophages. All the MCR were expressed significantly in melanoma cells (Fig. 4A). To further detect the type of MCR used by α-MSH to down-regulate CXCR, cells were pre-incubated with with anti-MC-1R, -MC-2R, -MC-3R, -MC-4R or all these Ab for 2 h, treated with α-MSH for 24 h, and assayed for IL-8 binding. α-MSH inhibited 79% of IL-8 binding, and pre-incubation of α-MSH-treated cells with the anti-MCR Ab (MC-1R, MC-2R, MC-3R, MC-4R or all Ab) resulted in 25, 70, 67, 78 and 10% inhibition of IL-8 binding, respectively (Fig. 4B), suggesting that α-MSH exerts its signal mostly through MC-1R in neutrophils.
α-MSH does not compete with IL-8 binding
To evaluate the effect of α-MSH on the IL-8 binding site on neutrophils, cells were pre-incubated with 250 ng unlabeled (cold) IL-8 or α-MSH for 4 h at 4°C in triplicate, followed by incubation with different amounts of 125I-IL-8 for 2 h, and binding was assayed. Unlabeled IL-8 but not α-MSH almost completely suppressed binding of 125I-IL-8 (Fig. 4C), indicating that α-MSH does not compete for IL-8 binding sites on neutrophils.
α-MSH inhibits IL-8 binding
To estimate the number of CXCR expressed on α-MSH-treated neutrophils, Scatchard analysis was performed 28, 29. Cells, either untreated or treated with α-MSH (1000 nM) for 24 h, were incubated with different concentrations of 125I-IL-8 at 4°C for 2 h in presence or absence of 50-fold unlabeled IL-8. The specific binding of IL-8 is shown in Fig. 5A as the mean count ± SD of duplicate samples. From the specific count, bound/free (B/F) values were calculated and plotted against bound values (Fig. 5A, inset). From these data the total number of CXCR was calculated as 16 958 CXCR per unstimulated neutrophil (Kd =0.5 nM) versus 3390 CXCR (Kd =0.4 nM) per α-MSH-treated neutrophil. Thus, α-MSH down-regulated about 79% of CXCR as detected by Scatchard plot that is correlated with IL-8 binding data without changing the affinity towards IL-8.
To demonstrate α-MSH-mediated down-regulation of IL-8 binding, chemical coupling of 125I-IL-8 to receptors was performed at 4°C using the bifunctional cross-linker disuccinimidyl suberate (DSS). The intensity of the 67 and 75 kDa bands was decreased with increasing concentrations of α-MSH (Fig. 5B), indicating down-regulation of CXCR. To detect the type of CXCR down-regulated by α-MSH, cells were treated with α-MSH (1000 nM) for 24 h and then incubated with 250 ng IL-8 or MGSA for 2 h at 4°C, and 125I-IL-8 binding was assayed. The 50-fold unlabeled IL-8 suppressed almost 90% binding of labeled IL-8, whereas cold MGSA suppressed about 50% binding of labeled IL-8 (Fig. 5C), suggesting the down-regulation of both types of CXCR by α-MSH.
Down-regulation of CXCR was also detected by immunofluorescence. After α-MSH treatment, cells were fixed and incubated with anti-CXCR1 and anti-CXCR2 Ab, followed by staining with anti-rabbit IgG-Alexa-Fluor and DAPI. The cells were visualized under an immunofluorescence microscope. The Alexa-Fluor level was significantly decreased in α-MSH-treated cells (Fig. 5D), indicating the down-regulation of CXCR. α-MSH-mediated down-regulation of CXCR1 was also detected from cell extracts immunoprecipitated with anti-CXCR1 Ab followed by Western blot with anti-CXCR1 Ab (Fig. 5E).
α-MSH does not inhibit CXCR mRNA
As α-MSH down-regulates CXCR, the levels of CXCR1 and CXCR2 mRNA were determined by RT-PCR. Total RNA was isolated from neutrophils treated with α-MSH (1000 nM) or LPS (100 ng/mL) for 24 h. This RNA was used for RT-PCR, followed by PCR using CXCR1-, CXCR2- or actin-specific primers and analysis in a 1% agarose gel. The bands for CXCR1 and CXCR2 did not change in α-MSH-treated cells. LPS induced these bands, and actinomycin D inhibited LPS-induced CXCR1 and CXCR2 (Fig. 5F), suggesting that α-MSH-mediated down-regulation of CXCR is not at the mRNA level.
α-MSH does not inhibit IL-8 binding in isolated neutrophil membrane or purified CXCR
To analyze the effect of α-MSH on neutrophil membrane and purified CXCR, neutrophils, isolated membrane and affinity-purified CXCR were incubated with different concentrations of α-MSH for 24 h at 37°C, and binding of labeled IL-8 was assayed. The results shown in Fig. 6A indicate that 125I-IL-8 binding was decreased in neutrophils but not in neutrophil membrane or purified CXCR. From these results, it is clear that α-MSH interacts only with cells, not with isolated membrane or affinity-purified CXCR.
IL-8 or anti-CXCR Ab protects α-MSH-mediated inhibition of IL-8 binding
To determine whether α-MSH mediates CXCR modulation within the IL-8-binding domain of the receptor, a ligand protection experiment was performed. As monodansyl cadaverine (MDC) can protect endocytosis at 37°C 30, cells pre-treated with MDC were incubated with 500 ng/mL IL-8 for 1 h at 37°C. The cells were washed, treated with different concentrations of α-MSH for 24 h, and then washed with glycine-HCl (50 mM, pH 3). The binding of 125I-IL-8 was assayed at 4°C immediately after washing the cells. MDC-treated cells did not show any alteration of IL-8 binding. IL-8 binding was decreased in neutrophils following α-MSH treatment in a dose-dependent manner, but cells pre-incubated with IL-8 showed no inhibition of IL-8 binding (Fig. 6B), indicating that α-MSH-mediated down-regulation of CXCR is protected by IL-8.
To examine the protection of CXCR from the effect of α-MSH by anti-CXCR Ab, cells incubated with 1 μg each of anti-CXCR1 and anti-CXCR2 Ab/2 × 106 cells for 1 h at 37°C were incubated with different concentrations of α-MSH at 37°C in a CO2 incubator for 24 h. Cells were then washed with 0.5 M potassium thiocyanate (KSCN), which removes bound antibody from cell surface, for 10 s. After immediate washing, the mean binding of 125I-IL-8 (cpm) was assayed at 4°C. Anti-CXCR Ab protected the cells from the dose-dependent decrease in IL-8 binding resulting from α-MSH treatment (Fig. 6C).
α-MSH does not induce IL-8 production
A relevant question is whether α-MSH-mediated CXCR down-regulation occurs simply as a result of IL-8 production. To address this question, cells were treated with α-MSH and IL-8 levels analyzed. Neutrophils were treated with 100 or 1000 nM α-MSH or 100 ng/mL LPS for 24 h. Cell supernatants (10× concentrated) and extracts (from pellet) were assayed for IL-8 using an IL-8 assay kit. LPS induced a significant increase in the level of IL-8 in both cell supernatant and pellet, whereas the level of IL-8 expression was not enhanced in α-MSH-treated cells (Fig. 6D). These results indicate that α-MSH-mediated down-regulation of CXCR is not due to production of IL-8.
PMSF and CMK protect against α-MSH-mediated inhibition of IL-8 binding
As proteases are known to cleave different proteins from the cell surface, it is important to clarify the role of proteases in α-MSH-mediated down-regulation of CXCR. Neutrophils (1 × 106) were pre-incubated with different protease inhibitors for 2 h and subsequently treated with α-MSH (1000 nM) for 24 h. Then 125I-IL-8 binding was assayed. The results in Fig. 7A show that binding of IL-8 on neutrophils was not decreased by different protease inhibitors alone, but pre-treatment with PMSF or CMK (neutrophil elastase inhibitor) resulted in 80–90% protection, and leupeptin, TPCK or TLCK provided 40% protection of IL-8 binding on α-MSH-treated cells. Cell viability was not affected by the different protease inhibitors (data not shown). These results indicate that leupeptin, TPCK and TLCK partially protect and both PMSF and CMK almost completely protect CXCR from α-MSH-mediated down-regulation. CMK protected against α-MSH-mediated CXCR down-regulation in a dose-dependent manner (Fig. 7B).
To further analyze PMSF- and CMK-mediated protection of CXCR against down-regulation by α-MSH, IL-8-mediated NF-κB activation was assayed. Neutrophils were treated with α-MSH and pre-, co- or post- incubated with PMSF (100 μM) or CMK (1 μM). The cells were then incubated with IL-8 for 2 h, and NF-κB activity was assayed from NE. IL-8 induced NF-κB, which was down-regulated by α-MSH. PMSF (Fig. 7C) or CMK (Fig. 7D) alone did not interfere with IL-8-induced NF-κB activation. Pre-incubation of cells with CMK or PMSF protected against α-MSH-mediated down-regulation of NF-κB, but co- or post-incubation did not. These results clearly indicate that neutrophil elastase inhibitor (CMK) blocks α-MSH-mediated down-regulation of CXCR and thereby NF-κB activation by IL-8.
Anti-elastase but not anti-cathepsin G or anti-proteinase 3 Ab protects against α-MSH-mediated inhibition of IL-8 binding
To analyze the roles of different neutrophil proteases induced by α-MSH in the down-regulation of CXCR, neutrophils incubated with different Ab (anti-elastase, -cathepsin G or -proteinase 3) for 2 h were treated with α-MSH for different times at 37°C, and IL-8 binding was assayed. Fig. 8A shows that α-MSH decreased IL-8 binding in a dose-dependent manner, and anti-elastase but not anti-cathepsin G or anti-proteinase 3 Ab almost completely inhibited the α-MSH-mediated decrease in IL-8 binding. These data suggest that α-MSH induces elastase in neutrophils, which may down-regulate CXCR.
α-MSH increases neutrophil elastase level and activity
To examine the role of proteases secreted from α-MSH-treated neutrophils, cells were treated with α-MSH (1000 nM) for different times, and the culture supernatant was assayed for neutrophil elastase as described in the Materials and methods. The results show that elastase activity increases with time during α-MSH treatment (Fig. 8B). Culture supernatant from α-MSH-stimulated cells did not show elastase activity when incubated with elastase inhibitor CMK, indicating its specificity 31. The culture supernatant was concentrated, and 100 μg protein was analyzed to detect levels of elastase, cathepsin G and proteinase 3 by Western blot. The level of elastase but not proteinase 3 or cathepsin G was increased significantly with α-MSH treatment over time (Fig. 8C), indicating a possible role of elastase in α-MSH-mediated down-regulation of CXCR.
Neutrophil elastase down-regulates IL-8 binding to neutrophils, neutrophil membranes and affinity-purified CXCR
To study the role of elastase on CXCR, neutrophils, isolated membrane and affinity-purified CXCR were incubated with 200, 500 or 1000 ng recombinant human neutrophil elastase for 6 h. 125I-IL-8 or 125I-TNF binding to treated neutrophils and 125I-IL-8 binding to membrane and affinity-purified receptors was then determined. IL-8 but not TNF binding to neutrophils (Fig. 8D1) and IL-8 binding to both purified membrane and CXCR (Fig. 8D2) was decreased with increasing concentrations of elastase. These data suggest that elastase down-regulates CXCR.
α-MSH or neutrophil elastase cleaves the IL-8 receptor
To detect the possible cleavage of CXCR by α-MSH or elastase treatment, neutrophils were treated with 1000 nM α-MSH for 24 h or 500 ng neutrophil elastase for 6 h. Cell pellet extract (200 μg protein) and culture supernatant (100 μg) were used to detect CXCR1 by Western blot. The results show that both α-MSH treatment (Fig. 8E1) and elastase treatment (Fig. 8E2) decreased the level of CXCR1 in neutrophils, and the cleaved bands detected in culture supernatant further suggest the down-regulation of CXCR1. Both blots were re-probed with anti-proteinase 3 Ab.
α-MSH-mediated down-regulation of CXCR is independent of cAMP
As α-MSH-mediated cell signaling involves the generation of cAMP through activation of adenylate cyclase, neutrophils pre-treated with dideoxyadenosine (ddAdo), an adenylate cyclase inhibitor, were treated with α-MSH for 24 h or incubated with dibutyryl cAMP (db cAMP) for 4 h. The level of elastase was detected by Western blot and colorimetric assay (Fig. 8F, upper panels) from culture supernatant. The level of cAMP was examined in cell extract (Fig. 8F, lower middle panel), and IL-8 binding was determined on neutrophils (Fig. 8F, lower panel). The results show that α-MSH treatment increased elastase and cAMP levels but decreased IL-8 binding. ddADO alone did not influence the parameters (same as untreated). It also did not inhibit the α-MSH-mediated upregulation of elastase or decreased IL-8 binding, but it protected against the α-MSH-mediated increase in cAMP. Addition of db cAMP did not increase the elastase level or decrease IL-8 binding compared to untreated cells. These results suggest that although α-MSH increases cAMP levels, this increase is not responsible for its effects on elastase levels or IL-8 binding.
α-MSH effects on IL-8-responses in human blood-derived neutrophils and macrophages
So far the experiments described were performed in HL-60-differentiated neutrophils or PMA-differentiated macrophages. To determine whether α-MSH down-regulates CXCR - and thereby IL-8-mediated biological responses - in a more physiological setting, neutrophils and macrophages isolated from fresh human blood were examined. The isolated neutrophils and macrophages showed about 12 and 5.2% cell death, respectively, at 24 h of culture, and cell viability was not altered in presence of CMK. IL-8-induced NF-κB activation was inhibited by α-MSH in a dose-dependent manner in neutrophils (Fig. 9A) and macrophages (Fig. 9B). As shown in Fig. 9C, α-MSH inhibited 125I-IL-8 binding in a dose-dependent manner in both neutrophils and macrophages, and this inhibition of IL-8 binding was prevented by pre-incubation of neutrophils with anti-CXCR Ab (Fig. 9D). IL-8 induced myeloperoxidase (Fig. 9E) and β-D-glucuronidase (Fig. 9F) release in a dose-dependent manner, and α-MSH inhibited this release in human neutrophils. CMK protected against α-MSH-mediated down-regulation of CXCR in a dose-dependent manner in human neutrophils (Fig. 9G) to a similar extent as in HL-60-differentiated neutrophils (Fig. 7B), as detected by IL-8 binding assay.
Even though several studies have indicated that certain neuropeptides, such as α-MSH, have anti-inflammatory effects, the mechanism underlying this effect is not understood. The reports so far available are that α-MSH down-regulates TNF-induced NF-κB activation through generation of cAMP and activation of PKA 24, 26. As IL-8 is an important inflammatory cytokine and is involved in all forms of neutrophil- and macrophage-driven inflammation, it becomes interesting to study the effect of α-MSH on IL-8-mediated neutrophil and macrophage functions. The purpose of this study was to investigate the effect of α-MSH on IL-8-induced cell signaling. In our experiments, we found that α-MSH blocks IL-8-induced NF-κB activation, chemotaxis, oxidative burst response and enzyme release. IL-8-induced NF-κB activation is not due to endotoxin, as polymixin B sulphate-incubated IL-8 equally activated NF-κB (Fig. 1C). α-MSH decreased CXCR from the surface of neutrophils and macrophages without competing for the IL-8 binding site (Fig. 4C). In contrast, α-MSH had no effect on cell surface receptor levels of TNF, IL-1, IL-4, IL-13 or TRAIL. α-MSH reduced CXCR number without changing affinity and induced the secretion of elastase, a specific serine protease that may down-regulate CXCR. The serine protease inhibitors PMSF and CMK reversed all these effects of α-MSH.
Primarily cells in the brain and pituitary are responsible for production of α-MSH. α-MSH interacts through its receptors (MCR) on different cells, and it also stimulates melanocytes and other cell types. It was reported that both human and murine macrophage cell lines and human neutrophils express MCR 20, 21, 26. We have shown that MC-1R are the MCR predominantly expressed on neutrophils and macrophages (Fig. 4A, B). It has been reported that α-MSH mediates its effects through an increase in intracellular cAMP 24. In this study, we found that the blocking of cAMP generation by ddAdo (inhibitor of adenylate cyclase) in neutrophils did not alter the induction of neutrophil elastase or inhibition IL-8 binding resulting from α-MSH treatment (Fig. 8F). α-MSH down-regulated CXCR but not the death receptors IL-4R and IL-13R (Fig. 3A), which are structurally different from CXCR. However, down-regulation of CXCR reflects the inhibition of IL-8-induced biological responses in neutrophils as detected by NF-κB activation (Fig. 1A), NF-κB promoter-driven reporter gene expression (Fig. 2D1), release of proteolytic enzymes such as myeloperoxidase, alkaline phosphatase and β-D-glucuronidase (Fig. 1D), chemotaxis (Fig. 2A1) and oxidative burst response (Fig. 2B). FMLP is a strong inducer of neutrophil oxidative burst response, but α-MSH did not inhibit FMLP-induced oxidative burst response (Fig. 2B), chemotaxis (Fig. 2A2), NF-κB binding (Fig. 2C) or NF-κB-dependent reporter gene expression (Fig. 2D2), indicating its specific function. About 79% down-regulation of CXCR by α-MSH was detected by Scatchard analysis, and it was not due to changes in the affinity of receptors (Fig. 5A). Down-regulated receptors were also detected by chemical cross-linking of labeled ligand-receptor complex (Fig. 5B), immunofluorescent data (Fig. 5D) and immunoprecipitated CXCR (Fig. 5E). Down-regulation of CXCR1 was detected from cleavage fragments from culture supernatant of α-MSH-treated cells (Fig. 8E1). α-MSH down-regulated both types of CXCR as detected by cold competition with MGSA. MGSA binds to CXCR2 with high affinity. CXCR cross-linked with 125I-IL-8 showed two bands of equal intensity, further proving the down-regulation of both types of CXCR by α-MSH treatment. α-MSH had no effect on CXCR down-regulation in isolated neutrophil membrane or affinity-purified CXCR in vitro (Fig. 6A), indicating that α-MSH interacts with viable cells and down-regulates CXCR. IL-8 or anti-CXCR Ab protected against CXCR down-regulation (Fig. 6B, C), suggesting that they may be masking CXCR and protecting the IL-8 binding site from α-MSH-mediated modulation. Although MDC protected CXCR from internalization after binding with IL-8, the unprotected CXCR (stored inside the cells in different granules as the intracellular pool) might come out during the incubation with α-MSH at 37°C. As α-MSH down-regulated almost 80% of CXCR, these unprotected CXCR may be not accounted for.
PMSF is known to block trypsin- and chymotrypsin-like serine proteases 32. TPCK and TLCK are chymotrypsin-like and trypsin-like protease inhibitors, respectively. PMSF completely protected against α-MSH-mediated CXCR down-regulation, indicating the involvement of serine proteases in α-MSH-induced down-regulation of CXCR (Fig. 7A). CMK, a specific neutrophil elastase inhibitor, completely protected CXCR, suggesting involvement of elastase, a serine protease, in α-MSH-mediated down-regulation of CXCR (Fig. 7A, B). α-MSH activates cells through interaction with its receptor and signaling to induce or release different proteolytic enzymes from neutrophils, as these cells are involved in the first line of defense. Although release of neutrophil elastase from neutrophils was increased, the level of myeloperoxidase was not increased significantly by α-MSH treatment; as elastase and myeloperoxidase are both retained in the primary granules 32, the mechanism by which α-MSH specifically induces neutrophils to release elastase needs to be further studied. Several reports have suggested that neutrophil elastase enhances IL-8 levels 33, 34. Surprisingly, we did not detect any increase in IL-8 levels in α-MSH-treated cells (Fig. 6D), though it increased neutrophil elastase level and activity (Fig. 7B, C). The neutrophil elastase cleaves TNFR from cell surface, and about 6 μg/mL neutrophil elastase decreases TNF binding by 50% in neutrophils 35. We did not detect significant down-regulation of TNFR. The level of neutrophil elastase released in response to 1000 nM α-MSH treatment may have been too little to significantly affect TNFR. At 5 μg neutrophil elastase/2 × 106 neutrophils, 82% and 49% down-regulation of CXCR and TNFR was observed, respectively, as detected by ligand binding (data not shown). The minimal level of neutrophil elastase required to specifically down-regulate CXCR still needs to be determined. Anti-elastase Ab has been shown to protect against α-MSH-mediated down-regulation of CXCR (Fig. 8A). Neutrophil elastase down-regulated CXCR (cleaving CXCR1 into smaller fragments) in neutrophils, isolated membrane and affinity-purified CXCR in vitro. α-MSH-mediated down-regulation of CXCR was followed by IL-8-induced biological responses, as shown in isolated human neutrophils and macrophages (Fig. 9).
Migrated neutrophils and residential macrophages release proteases not only to clear microbes in the case of infection but also to destroy the surrounding tissues during inflammation. α-MSH induces the release of a specific protease from neutrophils that specifically down-regulates CXCR, thus regulating excessive migration of neutrophils into the inflamed tissues. In the case of neutrophil-driven inflammatory diseases, therapy with α-MSH may help to ameliorate the suffering of patients.
Materials and methods
RPMI 1640 medium, antibiotics-antimycin and fetal bovine serum (FBS) were obtained from Life Technologies Inc. (Grand Island, NY). Glycine, α-MSH, PMA, FMLP, bovine serum albumin, histopaque, dextran, ortho-phenylenediamine (OPD), p-nitrophenyl phosphate, p-nitrophenyl β-D-glucuronide, N-methoxysuccinyl-Ala-Ala-Pro-Met p-nitroanilide, monodansyl cadaverine (MDC), polymixin B sulphate, leupeptin, PMSF, bestatin, pepstatin, TPCK, TLCK and iodogen were obtained from Sigma (St. Louis, MO). Recombinant human IL-8, IL-13, IL-4, IL-1α, TRAIL and TNF were obtained from PeproTech Inc. (Rocky Hill, NJ). NF-κB oligonucleotide was synthesized by Gibco-BRL. Human neutrophil elastase, CMK (N-methoxysuccinyl-Ala-Ala-Pro-Val-chloromethyl ketone) and anti-neutrophil elastase Ab were obtained from Calbiochem (San Diego, CA). Ab against MC-1R, MC-2R, MC-3R, MC-4R, proteinase 3, cathepsin G, ICAM-1, IL-8R1 (CXCR1) and IL-8R2 (CXCR2) were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The human IL-8 assay kit was purchased from R&D Systems (Minneapolis, MN). The superfect transfection reagent was purchased from Qiagen (Hilden, Germany). The plasmid construct for NF-κB-SEAP was kindly supplied by Prof. B. B. Aggarwal of the University of Texas M. D. Anderson Cancer Center (Houston, TX).
HL-60 (human monocytic cells), THP1 (human monocytic macrophages) and A375 (human melanoma) cell lines were used in this study. These cells were obtained from the American Type Culture Collection (Manassas, VA) and cultured in complete medium (RPMI 1640 medium supplemented with 10% FBS, 100 U/mL penicillin and 100 µg/mL streptomycin). All cells were free from mycoplasma, as detected by Gen-Probe mycoplasma rapid detection kit (Fisher Scientific, Pittsburgh, PA).
Neutrophil and macrophage differentiation and isolation
HL-60 and THP1 cell lines were maintained in complete RPMI 1640 medium. HL-60 cells cultured in the presence of 1.3% DMSO for 2 days differentiated into neutrophils 36, 37. These differentiated cells were thoroughly characterized as neutrophils by determining the level of CXCR expression, morphology, IL-8-induced chemotaxis and phorbol myristate acetate (PMA)-induced oxidative burst response and enzyme release compared to HL-60 cells. We considered these cells as neutrophils in this study. THP1 cells were stimulated with 10 ng/mL PMA for 16 h, and adherent cells were used as macrophages 25.
Neutrophils and macrophages were separated from fresh blood collected from human peripheral vein by dextran T-500 sedimentation, followed by histopaque gradient centrifugation 28. The purity of the cells was examined by Giemsa staining, and cell viability was checked by trypan blue dye exclusion. The cell preparation contained 90–95% neutrophils, of which about 88% were viable until 24 h incubation.
Radiolabeling of IL-8, TNF, TRAIL, IL-4 and IL-13 and receptor binding assay
Human IL-8, IL-4, IL-13, TNF and TRAIL were iodinated with [125I]Na using the IODO-GEN method. Radiolabeled ligands were purified by G25 sepharose column. The specific activities of radiolabeled ligands were 0.5 × 107 to 1 × 107 cpm/μg protein. Cell surface receptors for different ligands were detected following the method as described previously 29.
NF-κB activation assays
To determine IL-8-induced NF-κB activation, EMSA was conducted essentially as described previously 38. Briefly, 8 µg NE was incubated with 32P end-labeled double-stranded NF-κB oligonucleotide from HIV-LTR (5′-TTGTTACAAGGGACTTTCCGCT GGGGACTTTC CAGGGAGGCGTGG-3′, bold indicates the NF-κB binding site) for 30 min at 37°C, and the DNA-protein complex was separated from free oligonucleotides on 6.6% native PAGE. The specificity of binding was also examined by competition with the unlabeled oligonucleotide. The radioactive bands from dried gel were quantitated by PhosphorImager (Fuji, Japan) using Image Reader software.
NF-κB-dependent reporter gene transcription
The effect of α-MSH on IL-8-induced NF-κB-dependent reporter gene transcription was measured as previously described 39. Briefly, HL-60 cells were transiently transfected using the Qiagen Superfect transfection reagent with 0.5 μg NF-κB promoter DNA linked to the heat-stable SEAP gene for 6 h, and cells were then cultured in the presence of 1.3% DMSO for 2 days. Cells were treated with α-MSH for 24 h, followed by stimulation with IL-8 for 12 h. The cell culture-conditioned medium (25 μL) was analyzed for SEAP activity as per the CLONTECH protocol (Palo Alto, CA). The mean (± SD) of relative fluorescent units for each transfection was determined and is reported as fold activation with respect to vector-transfected cells. This reporter system was specific, as IL-8-induced SEAP activity was inhibited by overexpression of IκBα mutants lacking either Ser32 or Ser3618, 39.
Enzyme release assay
Neutrophils (1 × 107 cells/mL) suspended in RPMI 1640 medium (phenol-red free) with 10% FBS (untreated or α-MSH pre-treated for 24 h) were stimulated with different concentrations of IL-8 for 2 h at 37°C. The supernatant was collected and used for assay of three different enzymes 28: myeloperoxidase, alkaline phosphatase and β-D-glucuronidase activity were measured using orthophenylenediamine, p-nitrophenyl phosphate and p-nitrophenyl β-D-glucuronide as substrate, respectively. Similarly, neutrophil elastase activity was measured using the specific chromogenic substrate N-methoxysuccinyl-Ala-Ala-Pro-Val p-nitroanilide, and the level of elastase was determined from a standard curve using known concentrations of elastase 31.
Detection of CXCR1 and CXCR2 by semiquantitative RT-PCR
After treatment, total RNA was extracted using TRIzol (Gibco BRL), and 1 μg total RNA was reverse-transcribed using poly-T oligonucleotide and M-MuLV reverse transcriptase (Invitrogen). The PCR was performed using primers for CXCR1 (5′-ACACCCTCATGAGGACCCAG-3′ and 5′-AGCATCCAGCCCTCATGAGG-3′) and CXCR2 (5′-CTATAGTGGCATCCTGCTAC-3′ and 5′-CCAAGAAGAACCAGTGGACA-3′). Following PCR, the amplicons were analyzed by gel electrophoresis with ethidium bromide staining. The expression of the investigated genes was determined by normalizing their expression against the expression of the actin gene.
For chemical cross-linking, neutrophils (1 × 107 cells/2 mL) subjected to 125I-IL-8 binding at 4°C for 2 h were pelleted, washed and suspended in 200 μL D-PBS. Disuccinimidyl suberate (DSS: 20 μL, from 10 mg/mL DMSO) was added slowly in 200 μL cell suspension and incubated for 1 h at 4°C. The cells were then washed, extracted and analyzed in 10% SDS-PAGE under reducing conditions. The gel was dried, exposed and scanned in a PhosphorImager (Fuji, Japan).
Neutrophil membrane was isolated from neutrophils (1 × 107 cells) with hypotonic lysis buffer, followed by sucrose gradient centrifugation as described earlier 40.
The Department of Biotechnology (DBT), Govt. of India, supported this work. We duly acknowledge the Council for Scientific and Industrial Research (CSIR), Govt. of India for providing fellowships (A.S. and Y.S.).