Effect of CC chemokines on mediator release from human skin mast cells and basophils

Role of CCL7 in Type I Hypersensitivity Reactions in Murine Experimental Allergic Conjunctivitis

Molecules that are necessary for ocular hypersensitivity reactions include the receptors CCR1 and CCR3; CCL7 is a ligand for these receptors. Therefore, we explored the role of CCL7 in mast cell activity and motility in vitro and investigated the requirement for CCL7 in a murine model of IgE-mediated allergic conjunctivitis. For mast cells treated with IgE and Ag, the presence of CCL7 synergistically enhanced degranulation and calcium influx. CCL7 also induced chemotaxis in mast cells. CCL7-deficient bone marrow-derived mast cells showed decreased degranulation following IgE and Ag treatment compared with wild-type bone marrow-derived mast cells, but there was no difference in degranulation when cells were activated via an IgE-independent pathway. In vivo, CCL7 was upregulated in conjunctival tissue during an OVA-induced allergic response. Notably, the early-phase clinical symptoms in the conjunctiva after OVA challenge were significantly higher in OVA-sensitized wild-type mice than in control challenged wild-type mice; the increase was suppressed in CCL7-deficient mice. In the OVA-induced allergic response, the numbers of conjunctival mast cells were lower in CCL7-deficient mice than in wild-type mice. Our results demonstrate that CCL7 is required for maximal OVA-induced ocular anaphylaxis, mast cell recruitment in vivo, and maximal FcεRI-mediated mast cell activation in vitro. A better understanding of the role of CCL7 in mediating ocular hypersensitivity reactions will provide insights into mast cell function and novel treatments for allergic ocular diseases.

Anti-Fas/CD95 and tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) differentially regulate apoptosis in normal and neoplastic human basophils

Basophilia is associated with allergic and parasitic diseases and advanced chronic myeloid leukemia. In the present study, we characterized the expression and function of the death receptors Fas/CD95 and tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) receptors in basophils from healthy donors compared to neoplastic basophils. Peripheral blood basophils obtained from healthy donors (HD-PBB) and from patients with chronic myeloid leukemia (CML-PBB) were found to express high levels of Fas/CD95 and low levels of TRAIL-R2, whereas the basophil-like chronic myeloid leukemia cell line KU-812 expressed significant levels of TRAIL-R1 and TRAIL-R2. HD-PBB underwent apoptosis in response to anti-Fas/CD95, but showed resistance to TRAIL, unless they were co-treated with actinomycin D. Interestingly, CML-PBB and KU-812 cells exhibited the opposite response pattern with resistance to anti-Fas/CD95, but significant susceptibility to TRAIL-induced apoptosis. Our data show that anti-Fas/CD95 and TRAIL differentially regulate apoptosis of normal and neoplastic human basophils, which may direct the development of novel therapeutic strategies.

Murine and human Langerhans cells express a functional histamine H4 receptor: modulation of cell migration and function

BACKGROUND

Histamine is an important mediator of allergic reactions, and recent studies indicated that the function of different types of antigen presenting cells (APC) can be modulated by histamine, in particular via the newly described histamine H(4) receptor (H(4)R). Therefore, we investigated possible interactions of histamine via the H(4)R on Langerhans cells (LC), which represent the professional APC in the skin and therefore have an important role in the initiation and maintenance of allergic skin diseases.

METHODS

The expression of the H(4)R was evaluated by real-time PCR, flow cytometry and immunofluorescence staining. The function of the H(4)R was determined by intracellular flow cytometric measurement of chemokine production and LC migration assays.

RESULTS

Here, we show H(4)R expression on in vitro generated monocyte-derived LC (mRNA and protein) and on primary LC from murine and human skin samples (protein). The immunofluorescence staining in murine and human skin samples clearly proved that LC express the H(4)R in situ. Stimulation with histamine or a H(4)R agonist downregulated the chemokine (C-C motif) ligand 2 (CCL2) in human monocyte-derived LC and primary LC. Prestimulation with a selective H(4)R antagonist abolished this effect. Moreover, migration of LC from the epidermis was increased after H(4)R agonist stimulation in ex vivo migration assays using human epidermis and murine in vivo assays.

CONCLUSION

Our findings show that LC express a functional H(4)R and point towards a possible pathogenic relevance of the H(4)R in inflammatory and allergic diseases.

Recent advances in laboratory assays for nonhemolytic transfusion reactions

Nonhemolytic transfusion reactions (NHTRs) are the most common transfusion reactions and include transfusion-related acute lung injury (TRALI) and allergic and febrile reactions. White blood cell (WBC) antibodies (Abs) against human leukocyte antigen (HLA) and human neutrophil antigen (HNA) in blood components are frequently implicated in NHTRs, especially in TRALI. Recently, we established a five-cell-lineage immunofluorescence test, a modified granulocyte immunofluorescence test, and a panel of cell lines stably expressing HNAs for efficient detection of Abs against well-known HNA-1 to HNA-5, and also Abs against neutrophil antigens other than HNA-1 to HNA-5. Using these techniques, we found that most of the non-HLA WBC Abs detected in NHTR cases were against antigens other than HNA-1 to HNA-5. In addition, using our newly established neutrophil activation test, which assesses neutrophil activation elicited by immunologic stimuli such as WBC Abs and immune complex, we found that heparin-binding protein is a potential final effector molecule that induces NHTRs, including TRALI. More recently, we reported that the basophil activation test, which was originally developed to identify allergens in the field of allergic diseases, might be useful in transfusion medicine. In this review, we summarize these new techniques and other related tests.

CCR1 expression and signal transduction by murine BMMC results in secretion of TNF-alpha, TGFbeta-1 and IL-6

Chemokine receptors (CCRs) are important co-stimulatory molecules found on many blood cells and associated with various diseases. The expression and function of CCRs on mast cells has been quite controversial. In this study, we report for the first time that murine bone marrow-derived mast cells (BMMC) express messenger RNA and protein for CCR1. BMMC cultured in the presence of murine recombinant stem cell factor and murine IL-3 expressed CCR1 after 5-6 weeks. We also report for the first time that mBMMC(CCR1+) cells endogenously express neurokinin receptor-1 and intercellular adhesion molecule-1. To examine the activity of CCR1 on these BMMC, we simultaneously stimulated two receptors: CCR1 by its ligand macrophage inflammatory protein-1alpha and the IgE receptor FcepsilonRI by antigen cross-linking. We found that co-stimulation enhanced BMMC degranulation compared with FcepsilonRI stimulation alone, as assessed by beta-hexosaminidase activity (85 versus 54%, P < 0.0001) and Ca(2+) influx (223 versus 183 nM, P < 0.05). We also observed significant increases in mast cell secretion of key growth factors, cytokines and chemokine mediators upon CCR1-FcepsilonRI co-stimulation. These factors include transforming growth factor beta-1, tumor necrosis factor-alpha and the cytokine IL-6. Taken together, our data indicate that CCR1 plays a key role in BMMC function. These findings contribute to our understanding of mechanisms for immune cell trafficking during inflammation.

Skin reactivity and local cell recruitment in human atopic and nonatopic subjects by CCL2/MCP-1 and CCL3/MIP-1alpha

BACKGROUND

Monocyte chemotactic protein (MCP-1/CCL2), the ligand for CCR2 and CCR5, and macrophage inflammatory protein-1alpha (MIP-1alpha/CCL3), the ligand for CCR1 and CCR5, are potent chemo-attractants in vitro and produce lesions in experimental animals, which resemble immediate and delayed-type hypersensitivity (DTH) reactions. CCL3 induces mononuclear cell and granulocyte infiltration in human atopic and nonatopic skin. Whether CCL2 (MCP-1) has comparable activity in man is uncertain as is the capacity of both the chemokines to elicit immediate- and DTH-like reactions in humans.

METHODS

Inflammatory cells were counted by immunohistochemistry in 24 and 48-h skin biopsies from atopics and nonatopics after intradermal injection of CCL2 and CCL3. Immediate (15 min) wheals-and-flares and delayed (24 and 48 h) indurations were also recorded.

RESULTS

Both chemokines induced immediate- (15 min) and delayed (24 and 48 h) reactions, which were associated with significant infiltrations of CD68+ macrophages, CD3+, CD4+ (but not CD8+) T cells, neutrophils, and eosinophils in biopsies from injection sites. CCL2, but not CCL3, also induced infiltration of basophils. Neither chemokine produced significant changes in the numbers of tryptase+ cutaneous mast cells. There were no differences in the pattern of skin reactivity or the numbers of infiltrating leukocytes in response to CCL2 and CCL3 between atopic and nonatopic subjects. In general, maximal infiltration of inflammatory cells was observed at the 24-h, rather than the 48-h, time point.

CONCLUSIONS

CCL2 and CCL3 induce both immediate and delayed skin reactions in atopics and nonatopics, and evoke a similar profile of local T cell/macrophage and granulocyte recruitment which, in general, confirm previous in vitro findings and in vivo experimental animal data.

CCL17 and CCL22 attenuate CCL5-induced mast cell migration

BACKGROUND

Mast cells (MCs) accumulate at sites of allergic mucosal inflammation where they act as central effectors and regulatory cells. Chemokines are believed to be crucial for the recruitment of MCs to sites of inflammation. We recently reported that human umbilical cord blood MCs (CBMCs) expresses the CC chemokine receptors, CCR1 and CCR4. We found a unique response profile to ligands of the respective receptors in which, of all tested ligands, only CCL5/RANTES-induced migration.

OBJECTIVE

To further investigate the function of CCR4 in MCs.

METHODS

CBMCs were used for competition binding experiments, migration, and intracellular calcium mobilization and release response studies.

RESULTS

The natural ligands for CCR4, CCL17/TARC and CCL22/MDC could both compete for binding with radiolabelled CCL5. Further, both CCL17 and CCL22 act as CCR4 antagonists by inhibiting CCL5-induced migration. Although both CCL17 and CCL22 caused mobilization of intracellular calcium, none of them induced migration or histamine release.

CONCLUSIONS

These results suggest that CCL5-induced migration of MCs via CCR4 can be regulated by the natural agonists CCL17 and CCL22, which are up-regulated at sites of allergic inflammation.

Biologic activity of RANTES in apheresis PLT concentrates and its involvement in nonhemolytic transfusion reactions

BACKGROUND

RANTES, one of the PLT-derived biologic response modifiers, accumulates in PLT concentrates (PCs) during storage and may play a causative role in nonhemolytic transfusion reactions (NHTRs) after PC transfusion.

STUDY DESIGN AND METHODS

To investigate the association of RANTES with NHTRs, the biologic activity of RANTES in the supernatant of stored PC at the intravascular concentration expected after PC transfusion was assessed by examining chemotaxis and histamine release in human basophils. In addition, the levels of RANTES in PCs involved in NHTRs were compared with those in PCs causing no transfusion reactions.

RESULTS

The supernatant of PC diluted to contain 1 nM RANTES significantly increased the migration of and release of histamine from basophils. Neutralizing antibody to RANTES suppressed the PC-triggered migration, but not histamine release. The levels of RANTES in PCs involved in NHTRs after PC transfusion were comparable to those in PCs that did not cause any transfusion reactions.

CONCLUSION

RANTES that accumulated in PCs during storage was biologically active in a basophil chemotaxis assay at the intravascular concentration expected after PC transfusion. However, the NHTRs after PC transfusion were not simply related to the RANTES level in PCs.

Aromatic components of food as novel eliciting factors of pseudoallergic reactions in chronic urticaria

BACKGROUND

Pseudoallergic reactions (PARs) against both additives and natural foods have been reported to elicit chronic urticaria, but in natural food the responsible ingredients are largely unknown.

OBJECTIVE

The study was aimed at identifying novel pseudoallergens in food and focused on evaluating tomatoes, white wine, and herbs as frequently reported food items eliciting wheal responses in urticaria.

METHODS

In 33 patients with chronic urticaria and PARs to food (proved by means of elimination diet and subsequent re-exposure with provocation meals), oral provocation tests were performed with field-grown tomatoes, organically grown white wine (whole food, steam distillates, and residues), oily extracts from herbs, and food additives. In addition, skin biopsy specimens from patients were studied for in vitro mast-cell histamine release with tomato distillate alone or on subsequent stimulation with anti-IgE, substance P, and C5a.

RESULTS

Seventy-six percent of patients reacted to whole tomato (steam distillate, 45%; residue, 15%), 50% to food additives, 47% to herbs, and 44% to whole wine (extract, 27%; residue, 0%). Histamine, protein, and high levels of salicylate were only found in residues. The tomato distillate was further analyzed by means of mass spectroscopy, identifying low molecular-weight aldehydes, ketones, and alcohol as major ingredients. In vitro histamine release was not caused by tomato extract itself but was enhanced by means of subsequent stimulation with substance P and C5a but not by anti-IgE.

CONCLUSION

Aromatic volatile ingredients in food are novel agents eliciting PARs in chronic urticaria. Histamine, salicylate, and a direct mast-cell histamine release are not involved in this reactivity to naturally occurring pseudoallergens.

Macrophage inflammatory protein-1alpha and C-C chemokine receptor-1 in allergen-induced skin late-phase reactions: relationship to macrophages, neutrophils, basophils, eosinophils and T lymphocytes

BACKGROUND

Macrophage inflammatory protein (MIP)-1alpha binds to C-C chemokine receptor (CCR)-1 with high affinity. CCR-1 is expressed on neutrophils, eosinophils, monocytes, T lymphocytes and basophils; cells characteristic of atopic allergic inflammation. In vitro, MIP-1alpha is chemotactic for monocytes, T cells and basophils and is also a potent histamine-releasing factor for basophils and mast cells. Although increased levels of MIP-1alpha were shown in atopic allergic disorders, the kinetics of expression of these CC chemokines in vivo is largely unknown.

OBJECTIVE

To investigate the kinetics of expression of MIP-1alpha and receptor CCR-1 and the relationships between the expression and infiltration of inflammatory cells in allergen-induced cutaneous late-phase reactions in atopic subjects.

METHODS

Cryostat sections, obtained from skin biopsies from 10 human atopic subjects at 6, 24, 48, 72 h and 7 days after allergen challenge, were processed for immunohistochemistry and in situ hybridization using 35S-labelled riboprobes.

RESULTS

The peak expression of allergen-induced mRNA for MIP-1alpha and CCR-1 was 6 h. This was maintained at 24 h, and gradually returned to base line at 7 days. At 6 h, the number of cells expressing MIP-1alpha mRNA significantly correlated with elastase+ neutrophils and BB-1+ basophils. At 24 h, the MIP-1alpha mRNA+ cells significantly correlated with CD68+ macrophages. There were significant inverse correlations between the numbers of MIP-1alpha mRNA cells and the numbers of Tryptase+ mast cells at 6 and 24 h after allergen challenge.

CONCLUSION

Allergen-induced cutaneous late-phase reactions in humans were associated with increased expression of MIP-1alpha and CCR-1. This may be relevant to the infiltration of neutrophils, eosinophils, basophils and macrophages.

Evaluation of chronic urticaria in patients with Hashimoto thyroiditis

The coincidence of Hashimoto thyroiditis (HT) and chronic idiopathic urticaria (CIU) is a commonly observed phenomenon in western New York. Previous literature suggested that there may be a direct relationship between them. We undertook these studies to determine whether humoral or cell-mediated mechanisms might link HT and CIU. Skin biopsies from patients with CIU, with or without HT, were indistinguishable by light microscopy. No immune complex deposition was observed, although only the skin from patients with CIU and HT contained perivascular fibrin deposits. Similarly, immunohistochemical studies evaluating cellular expression of CD3, CD4, CD8, CD20, and CD68 failed to differentiate between CIU with or without HT. Analysis of Vbeta restriction in thyroid tissue of patients with HT and the skin of patients with CIU and HT by in situ polymerase chain reaction failed to reveal any oligoclonal T-lymphocyte subpopulations. In contrast, only patients with CIU and HT had anti-FcepsilonRI antibodies in their sera that could induce degranulation of normal basophils. Some sera from patients with CIU and HT caused degranulation of normal basophils in the absence of anti-FceRI. The factor causing basophil degranulation in these sera was not determined. Patients with CIU and HT failed to improve clinically with thyroid replacement therapy. All CIU patients were equally well managed with symptomatic therapies. In conclusion, HT likely represents a marker of other autoimmunity, rather than being a direct causative agent in CIU. Management of CIU, with or without HT and with or without anti-FceRI antibodies, should be the same. Future studies will have to examine whether cell-mediated responses participate in CIU, especially in association with HT.

Antigen-specific T cell tolerance down-regulates mast cell responses in vivo

Fel d I is the major cat allergen that induces asthma and allergic rhinitis in humans. To investigate the mechanism of allergic responses to this allergen, a mouse model was developed. Mice sensitized to chain 1 of Fel d I exhibited T cell responses, B cell responses, and mast cell responses when challenged with the protein. Subcutaneous injections of peptides containing the dominant T cell epitopes of the allergen induced T cell tolerance in presensitized mice. When challenged with the allergen intratracheally, these tolerized mice produced a decreased amount of histamine in vivo. The decrease in histamine release was not solely dependent on the reduction of allergen-specific IgE. These data show that mast cell activity in mice with an ongoing sensitivity to allergen can be regulated through peptide-induced T cell tolerance.

Human mast cells transmigrate through human umbilical vein endothelial monolayers and selectively produce IL-8 in response to stromal cell-derived factor-1 alpha

Mature mast cells are generally considered to be less mobile cells residing within tissue sites. However, mast cell numbers are known to increase in the context of inflammation, and mast cells are recognized to be important in regulating local neutrophil infiltration. CXC chemokines may play a critical role in this process. In this study two human mast cell-like lines, HMC-1 and KU812, and human cord blood-derived primary cultured mast cells were employed to examine role of stromal cell-derived factor-1 (SDF-1) in regulating mast cell migration and mediator production. It was demonstrated that human mast cells constitutively express mRNA and protein for CXCR4. Stimulation of human mast cells with SDF-1, the only known ligand for CXCR4, induced a significant increase in intracellular calcium levels. In vitro, SDF-1 alpha mediated dose-dependent migration of human cord blood-derived mast cells and HMC-1 cells across HUVEC monolayers. Although SDF-1 alpha did not induce mast cell degranulation, it selectively stimulated production of the neutrophil chemoattractant IL-8 without affecting TNF-alpha, IL-1beta, IL-6, GM-CSF, IFN-gamma, or RANTES production, providing further evidence of the selective modulation of mast cell function by this chemokine. These findings provide a novel, SDF-1-dependent mechanism for mast cell transendothelial migration and functional regulation, which may have important implications for the local regulation of mast cells in disease.

Effects of filtration and gamma radiation on the accumulation of RANTES and transforming growth factor-beta1 in apheresis platelet concentrates during storage

BACKGROUND

Platelet-derived biologic response modifiers (BRMs) including RANTES and transforming growth factor (TGF)-beta1 accumulate in platelet components during storage because of platelet activation, and they may play a causative role in nonhemolytic febrile transfusion reactions. The majority of PCs with high unit values are provided by single donor apheresis in Japan.

STUDY DESIGN AND METHODS

RANTES and TGF-beta1 levels in platelet units prepared from single-donor apheresis platelet concentrates (apheresis PCs) and units from whole blood (buffy coat PCs) were investigated. The effects of prestorage and poststorage filtration and gamma radiation on the levels of RANTES and TGF-beta1 in the supernatant of apheresis PCs during storage were also examined.

RESULTS

The levels of RANTES and TGF-beta1 increased during storage from Day 0 to Day 5. The levels of RANTES and of TGF-beta1 correlated with the platelet concentration (p<0.01), but not with the residual white cell concentration in apheresis PCs that were not white cell reduced by filtration (p>0.05). In addition, there was a correlation between RANTES and TGF-beta1 levels (p<0.01). In white cell-reduced apheresis PCs using negatively charged filters as well as in gamma-radiated apheresis PCs, the levels of these two BRMs-did not differ at any storage time from those of untreated apheresis PCs. Filtration of apheresis PCs with negatively charged filters after 3 days of storage significantly (p<0.05) reduced the levels of RANTES, but not of TGF-beta1. There was no reduction in the levels of RANTES and TGF-beta1 levels by positively charged filters. The RANTES levels in buffy coat PCs were slightly higher than but not significantly different from those of apheresis PCs during storage, except for the level on Day 1. There were no differences in the TGF-beta1 levels in apheresis and buffy coat PCs during storage.

CONCLUSION

Prestorage filtration and gamma radiation had neither preventive effects on the accumulation of RANTES and TGF-beta1 nor adverse effects on platelet activation. Negatively charged filters might be useful for the reducing the levels of RANTES in stored apheresis PCs.

Expression and Functional Activity of the IL-8 Receptor Type CXCR1 and CXCR2 on Human Mast Cells

To further elucidate mechanisms involved in mast cell accumulation at sites of cutaneous inflammation, we have studied the ability of human leukemic mast cells (HMC-1 cells) to express functionally active IL-8 receptors. Expression of mRNA for both types of IL-8 receptors (CXCR1 and CXCR2) was demonstrated by PCR and of both proteins by flow cytometry. Binding and competition studies with 125I-labeled IL-8 and its homologue melanoma growth stimulating activity (125I-labeled MGSA) revealed two specific binding sites for IL-8, K1 = 1.1 × 1011 M−1 and K2 = 5 × 107 M−1; and for MGSA, K1 = 2.8 × 1010 M−1 and K2 = 5 × 107 M−1. This finding was supported by a dose-dependent rise of cytosolic free calcium concentration ([Ca2+]i) induced by both chemokines and to a lesser extent by the homologue neutrophil-activating peptide-2 (NAP-2). A significant migratory response of human leukemic mast cells (HMC-1) was observed with all three chemokines at a range from 10−8 M to 10−9 M. Moreover, the formation of cellular F-actin was induced in a rapid, dose-dependent fashion, with a maximally 1.7-fold increase at 10−7 M. Using postembedding immunoelectron microscopy, we could show the expression of CXCR1 on the cytoplasmatic membrane of isolated human skin mast cells whereas CXCR2 was located in mast cell-specific granules. These findings demonstrate for the first time the functional expression of both types of IL-8 receptors on human mast cells, suggesting a role for their ligands during mast cell activation and recruitment.

C3a and C5a Stimulate Chemotaxis of Human Mast Cells

The factors that control migration of mast cells to sites of inflammation and tissue repair remain largely undefined. Whereas several recent studies have described chemotactic factors that induce migration of murine mast cells, only stem cell factor (SCF ) is known to induce migration of human mast cells. We report here that the anaphylatoxins C3a and C5a are chemotactic factors for the human mast cell line HMC-1, human cord blood-derived mast cells (CBMC) and cutaneous mast cells in vitro. The presence of an extracellular matrix protein, laminin, was required for chemotaxis in response to complement peptides. Migration of mast cells towards C3a and C5a was dose-dependent, peaking at 1 μg/mL (100 nmol/L), and was inhibited by specific antibodies. Pretreatment with pertussis toxin inhibited the anaphylatoxin-mediated migration of HMC-1 cells, indicating that Gi proteins are involved in complement-activated signal transduction pathways in human mast cells. Both C3a and C5a also induced a rapid and transient mobilization of intracellular free calcium ([Ca2+]i ) in HMC-1 cells. Besides SCF, other chemotactic factors tested, such as interleukin-3, nerve growth factor, transforming growth factor β, RANTES (regulated upon activation, normal T cell expressed and secreted), monocyte chemotactic protein-1 (MCP-1), MCP-2, MCP-3, macrophage inflammatory protein-1α (MIP-1α), and MIP-1β, failed to stimulate migration of human mast cells. In summary, these findings indicate that C3a and C5a serve as chemotaxins for human mast cells. Anaphylatoxin-mediated recruitment of mast cells might play an important role in hypersensitivity and inflammatory processes.

Histamine-releasing factors

There is increasing evidence that human basophils accumulate at sites of chronic inflammation, and, in particular, in allergic asthma. Investigators have, therefore, become very interested in identifying proteins that activate these cells. Recently, the gene encoding a candidate for this function, a novel molecule, the IgE-dependent histamine-releasing factor, was cloned.

The inhibitory effect of pentamidine on the production of chemotactic cytokines by in vitro stimulated human blood cells

Pentamidine is an antiprotozoal drug with additional antiinflammatory activities that are not well understood. We now report that pentamidine inhibited the human whole blood production of the chemotactic cytokines (chemokines) interleukin (IL)-8, growth related gene alpha (GRO alpha) and monocyte chemotactic protein-1 (MCP-1). The title compound dose-dependently suppressed the lipopolysaccharide (LPS)- and phytohemagglutinin (PHA)-stimulated whole blood generation of these chemokines with IC50-values of 2.1 and 2.2 microM (IL-8), 2.4 and 1.8 microM (GRO alpha) and 2.8 and 2.4 microM (MCP-1). The inhibition was specific: when tested at 10 microM, pentamidine had no significant inhibitory effect on the PHA-induced generation of the non-chemotactic cytokines tumor necrosis factor-alpha (TNF-alpha), IL-1 beta, IL-2, IL-4, IL-5, IL-10 and interferon-gamma (IFN-gamma), except for a partial inhibition on IL-6. Time course experiments indicated that pentamidine (10 microM) retained its ability to inhibit PHA-stimulated IL-8 production even when its addition was delayed for up to 24h after mitogen stimulation. Furthermore, reverse transcription PCR studies showed that pentamidine had no effect on IL-8 mRNA expression. These findings indicate that pentamidine is a post-transcription acting inhibitor of human chemokine production. This activity may contribute to the anti-inflammatory action ascribed to the title compound.