| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
* IPF PharmaCeuticals, An-Institut of Hannover Medical School;
Department of Dermatology and Allergology, Hannover Medical School; and
Fraunhofer Institute for Toxicology und Experimental Medicine, Hannover, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A common feature of the eotaxins, the MCPs, and CCL5 is their ability to mediate chemotaxis via CCR3, which is shown to be expressed on eosinophils (14), mast cells (15), basophils (16), and Th2 cells (17, 18). The involvement of other chemokines was demonstrated in vivo, showing that different chemokines, mostly activators of CCR3, contribute to the complex pathophysiology of asthma (19). A major breakthrough, showing the impact of CCR3 in asthma, was achieved recently (20). Targeted disruption of CCR3 was successfully performed, demonstrating that eosinophils and other inflammatory cells are arrested in the subendothelial space of pulmonary vessels after bronchial allergen challenge in OVA-sensitized mice (20), suggesting that the local inflammatory response is abolished, targeting CCR3 already in the circulation. However, airway hyper-responsiveness (AHR) is enhanced in i.p. sensitized mice, which correlates to increased intraepithelial mast cells in the airways (20), whereas AHR is abrogated in CCR3-deficient mice sensitized by the epicutaneous route (21). Therefore, CCR3 appears to be particularly attractive as a drug target; its blockade has been proposed as a tool for the therapy of asthma (22).
Besides chemokines, cytokines such as IL-4, IL-5, and IL-13 are thought to be key molecules in the pathogenesis of allergic asthma (23). The first clinical trials to block IL-5 using mAbs showed a dramatic reduction of eosinophil counts in peripheral blood from patients suffering from allergic asthma (24, 25, 26). However, approximately half this cell type was still found in bone marrow and bronchial mucosa, which may contribute to the disappointing results of these studies regarding improvement of symptoms (24, 25, 26). With respect to these results, Kay and Menzies-Gow (27) speculate that a combination therapy to block the maturation of eosinophils using anti-IL-5 mAbs and to stop tissue accumulation using CCR3 antagonists may be more useful than just IL-5 blockade.
In this report we first describe the effects of NH2-terminal modifications of CCL14/hemofiltrate CC chemokine-1 (HCC-1) (28, 29) on its biological activity mediated via CCR3. Chemical modification of the most active form was performed with the intention to generate a CCR3 ligand leading to the inactivation of CCR3. Truncation of amino acid residues 1–8 of CCL14 and replacement of the amino acid in position 9 by nonanoic acid led to the identification of a potent CCR3 agonist, termed n-nonanoyl (NNY)-CCL14[10–74] (NNY-CCL14), with inactivating properties and that is resistant to cleavage by dipeptidyl peptidase IV (CD26/DPP IV; EC 3.4.14.5). CD26/DPP IV is an abundant peptidase found in blood plasma, tissue, and on the cell surface of various cell types. It is involved in the regulation of chemokine activity by hydrolysis of peptides with NH2-terminal Xaa-Pro and Xaa-Ala motifs (30). Most anti-inflammatory strategies are based on the mechanism that acts on target cells after they migrate to the site of inflammation (31). In contrast, we propose that cellular recruitment is prevented by receptor inactivation of inflammatory cells before they extravasate. We show that NNY-CCL14 impairs the migration of eosinophils into the airways of OVA-sensitized and Aspergillus fumigatus-sensitized mice. Our investigation presents a novel strategy for the treatment of CCR3-mediated diseases.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CCL11/eotaxin and CXCL12 were obtained from PeproTech (London, U.K.). C5a was obtained from Sigma-Aldrich (Taufkirchen, Germany). CCL14/HCC-1[1–74] was prepared as previously described (28).
Synthesis of CCL14 derivatives
CCL14/HCC-1[6/7/8/9/10/11/12–74] and NH2-terminally modified derivatives were prepared by F-moc-based, solid phase peptide synthesis as previously described (32). The synthesis of CCL14 peptides was conducted on a 433A peptide synthesizer (Applied Biosystems, Weiterstadt, Germany) at a scale of 0.1 mmol with a 10-fold excess of F-moc amino acid using 2-(1H-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium/1,1-hydroxybenzo-triazole activation. After peptide chain assembly, nonanoic acid was coupled as symmetrical anhydride (Sigma-Aldrich) in N-methylpyrrolidinone (65 equivalents) to the obtained polypeptide. The resulting peptides were cleaved and deprotected in the presence of trifluoroacetic acid/H2O/EDT/phenol (86/6/6/2, v/v/v/w; 15 ml/g), precipitated in cold tert-butylmethylether and purified chromatographically. The resulting chromatographically homogeneous peptides were analyzed by capillary zone electrophoresis and electrospray mass spectrometry. The purified derivatives were used for biological testing according to the net peptide content as determined by amino acid analysis.
Antibodies
The rat mAb against human CCR3 (clone 61828.111; IgG2a) and the murine mAb against human CCR1 (clone 53504.111; IgG2b) were obtained from R&D Systems (Wiesbaden, Germany). The rat IgG2a and mouse IgG2b isotype control Ab were obtained from Immunotech (Hamburg, Germany).
Eosinophil isolation
Eosinophils were purified from the venous blood of normal nonatopic healthy or atopic volunteers, all of whom had given their informed consent according to the local research ethics committee at Hannover Medical School, using Ficoll (Pharmacia, Erlangen, Germany) density gradient centrifugation as described previously (33). For further purification, granulocytes were resuspended in HEPES-buffered HBSS (Invitrogen, Karlsruhe, Germany), pH 7.4, containing 1 mg/ml BSA. Eosinophils were purified by negative selection using a MACS CD16 kit (Miltenyi Biotec, Auburn, CA) as described previously (33). The resulting eosinophil purity was >99% as determined by microscopic examination with Kimura staining.
Lucigenin-dependent chemiluminescence
The generation of reactive oxygen species (ROS) was measured by lucigenin-dependent chemiluminescence using a single-photon imaging system with a two-dimensional photon counting system that allows simultaneous measurement and analysis of 96 wells (MTP reader; Hamamatsu Photonics, Herrsching, Germany) as described previously (34). In brief, eosinophils were suspended at a density of 5 x 104 cells/ml in HEPES-buffered HBSS, pH 7.4, containing 1 mg/ml BSA with 200 µM lucigenin (Sigma-Aldrich). Aliquots (100 µl) containing 5 x 103 eosinophils were placed into flat-bottom, white microtiter plates (Microfluor; Dynatech, Denkendorf, Germany). Cells were stimulated with the indicated stimuli or with medium as a control. Measurements were performed in triplicate at 37°C. Data are expressed as the ratio between stimulus-induced intensity integral counts and medium-induced intensity integral counts (34).
Flow cytometric measurement of CCR3 internalization
These experiments were performed as previously described in detail (33, 34). For internalization of CCR3, the cells were preincubated for 30 min at 37°C with the indicated stimuli in a total volume of 100 µl of RPMI 1640 medium before staining. Thereafter, cells were fixed on ice and immediately washed with cold PBS. After the washing step, 105 eosinophils were stained by anti-CCR3 mAb as described below. The inhibition of CCR3 internalization was achieved by initial treatment of the cells for 5 min with 8 µM phenylarsine oxide (PAO; Sigma-Aldrich) at 37°C and alternatively by treatment of cells with the stimuli at 4°C. Both strategies are suitable to inhibit internalization of seven-transmembraneous receptors from the cell surfaces as described previously (33).
For flow cytometry (FACScan; BD Biosciences, Heidelberg, Germany) analysis, 105 eosinophils were incubated at 4°C for 30 min with anti-chemokine receptor mAb or isotype control at the concentrations recommended by the supplier. In a second step, the cells were stained by FITC-conjugated, goat anti-rat or goat anti-mouse Ab (Immunotech) and thereafter analyzed by flow cytometry. Data are expressed as an original plot (specific mAb vs isotype control) or as relative fluorescence intensity (percentage), which was calculated using the following equation: (median channel fluorescence [stimulus] – median channel fluorescence [isotype control])/(median channel fluorescence [medium] – median channel fluorescence [isotype control]) x 100% (33).
In vitro chemotaxis
Chemotaxis was assessed in 48-well chambers (NeuroProbe, Cabin John, MD) using polyvinylpyrrolidone-free polycarbonate membranes with 5-µm pores (Nucleopore, NeuroProbe) for 5 x 104 eosinophils/well as previously described (8). Cell suspension and chemokine dilution were conducted in RPMI 1640 containing 25 mM HEPES (pH 7.4) and 0.5% BSA. Migration was allowed to proceed for 60 min at 37°C in 5% CO2. The membrane was then removed, washed on the upper side with PBS, fixed, and stained. All assays were performed in triplicate, and the migrated cells were counted in five randomly selected fields at 1000-fold magnification. Spontaneous migration was determined in the absence of chemoattractant. For inhibition of CCL11/eotaxin-induced chemotaxis, eosinophils were preincubated for 15 min at room temperature with 10–7 M NNY-CCL14/HCC-1 and thereafter directly placed in the upper compartment of the chemotaxis chamber.
Measurement of intracellular calcium concentration ([Ca2+]i)
Eosinophils were loaded with 2 µM fura 2-AM (Molecular Probes, Eugene, OR), and [Ca2+]i was measured at 37°C in a Bowman series 2 spectrofluorometer (SLM-Aminco, Urbana, IL) as described previously (34). Autofluorescence represents 
10% of the total fluorescence of cells loaded with fura 2 and is slightly greater at 340 nm than at 380 nm. The fluorescence of unloaded cells was therefore subtracted from an equivalent density of cells loaded with fura 2 to correct the signal for autofluorescence. The autofluorescence of the cells was virtually unchanged upon addition of stimulus. Dual excitation spectra were collected at 340 and 380 nm, and emission was fixed at 510 nm as previously described (35). Receptor desensitization was tested by monitoring [Ca2+]i changes in response to sequential stimulation with chemokines as previously described (34).
Kinetics of CD26/DPP IV processing
To analyze the processing of the naturally occurring chemokine CCL14/HCC-1[9–74] (29) and the modified derivative NNY-CCL14, an in vitro kinetic study was performed by incubating 10–5 M chemokine with 6.6 x 10–4 U of porcine kidney DPP IV (lot 100K38002; Sigma-Aldrich) in Tris-HCl, pH 7.5, at 37°C. At specific time intervals, the reactions were stopped with 0.1% trifluoroacetic acid and placed on ice. For comparison, the DPP IV processing of the chemokines CCL11/eotaxin and CXCL12 (36) was examined in parallel. The composition of the reactions was determined offline with a MALDI mass spectrometer (Voyager DE-Pro; Applied Biosystems) in linear mode accumulating eight spectra of 100 shots each. 
-Cyano-4-hydroxycinnamic acid (Sigma-Aldrich) was used as the matrix. The instrument uses a high potential acceleration source (20 kV), and other parameters were optimized for the measurement of chemokines.
Animals
Female BALB/c mice, obtained from Charles River (Sulzfeld, Germany) at 8 wk of age and an average weight of 19 g, were used in the experiments as described previously (37). Mice were maintained on laboratory food and tap water ad libitum under pathogen-free conditions in a regular 12-h dark, 12-h light cycle at a temperature of 22°C and were allowed to become acclimated to their environment for a period of 7 days before the experiment.
Protocol of allergic sensitization with OVA or with A. fumigatus extract and NNY-CCL14 treatment
The animal experiment was approved by the Bezirksregierung Hannover and was conducted according to the guidelines of the institutional animal care and use committee. For OVA sensitization, animals were divided into two groups of four. Sensitization of the animals was conducted via the i.p. route on days 0, 14, and 21, each with 10 µg of OVA (chicken OVA grade VI; Sigma-Aldrich) together with 1 mg of Al(OH)3 (Alum Inject; Pierce, Rockford, IL) as adjuvant dissolved in sterile saline (37). To provoke an allergic airway response, aerosol challenge was performed using a Pari Master nebulization system (mass median aerodynamic diameter, 2.5 µm) and a 1% OVA solution. Animals were exposed to allergen on day 28 for 10 min, yielding a calculated airway allergen deposition of 10 µg of OVA/mouse. To examine the inhibitory effect of NNY-CCL14, four mice were treated with 10 nmol/kg NNY-CCL14/HCC-1 diluted in sterile saline (applied via the tail vein as a bolus) 30 min before and 3 and 8 h after the challenge. The other group (positive control) was injected with sterile saline at identical time points. For lung function measurement, the protocol was modified, and two allergen challenges on days 28 and 29 were performed. Subsequently, animals were treated four times with 3 nmol/kg NNY-CCL14 (30 min before and 6 h after each challenge). AHR in response to methacholine was assessed on day 30 by head-out plethysmography as described previously (38).
For A. fumigatus sensitization, animals were divided into three groups of eight to 12 mice. Allergic airway inflammation was induced using a modified allergy model described by Hogaboam et al. (39). Mice were sensitized s.c. and i.p. with an equal volume of 0.1 ml using a mixture of A. fumigatus extract (Greer Laboratories, Lenoir, NC; lot XPM3A3) in sterile saline emulsified with IFA. Sensitization was performed using an allergen dose of 5.4 µg of A. fumigatus/mouse. Control animals received sterile saline (negative control). Fourteen days later, animals (NNY-CCL14 group and positive controls) were challenged with aerosolized A. fumigatus extract using the Pari Master system and an A. fumigatus solution at a concentration of 5.4 mg/ml for 12 min, resulting in a final lung dose of 5 µg/mouse. The negative control animals received aerosolized saline. The second challenge was performed on day 21 exactly as the first challenge, except that the negative control animals received A. fumigatus aerosol as well. To examine the inhibitory effect of NNY-CCL14, one group of mice was treated on days 14 and 21 with 30 nmol/kg NNY-CCL14 30 min before and 6 h after the challenge (NNY-CCL14 group). The control groups were injected with sterile saline at identical time points.
To investigate inappropriate cellular activation by NNY-CCL14, nonallergen-sensitized/challenged mice were treated i.v. with 10 nmol/kg NNY-CCL14 or sterile saline (each group, n = 10). Five animals of each group were killed 30 min or 24 h after i.v. treatment and subjected to histological evaluation of the lung.
BALF and histological evaluation
Animals were killed 24 h after allergen challenge by injecting an overdose of sodium pentobarbital (Narcoren, Merial, Hallbergmoos, Germany) i.p. The trachea was catheterized, airways were lavaged with 0.8 ml of cold 0.9% NaCl, and lung dissection was performed. Total cell numbers in BALF were counted, and cytospins were evaluated (37). The left lungs of the mice were dissected and fixed in formalin for additional histological examination using H&E staining.
Statistical analysis
The number of experiments is given in the figures as n and represents different donors. Unless otherwise stated, the data in the text and figures were expressed as the mean ± SEM, as determined by SigmaStat (SPSS, Erkrath, Germany) analysis. Values of p 
0.05 were accepted as significant using Student’s t test.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To characterize the functional importance of the NH2-terminal domain of CCL14/HCC-1, the biological activities of 10 different CCL14 analogues were investigated. These included NNY-CCL14 (NNY-CCL14[10–74]), bis-NNY-CCL14[10–74], full-size CCL14[1–74], and seven NH2-terminally truncated variants. The NH2-terminal sequences of these derivatives and that of CCL11 are shown in Fig. 1.
|
To study the effects of the CCL14/HCC-1 derivatives on CCR3, we used freshly isolated human blood eosinophils, a natural cell population expressing high surface levels of CCR3. We screened 80 donors for CCR1 expression and found only two individuals with significant CCR1 surface expression (data not shown), who were excluded from the subsequent experiments.
The CCL14 analog NNY-CCL14 is a potent activator of the respiratory burst mediated by CCR3
First, we compared the effects of all CCL14/HCC-1 derivatives on the release of ROS using lucigenin-dependent chemiluminescence, which has been established as a sensitive method to study effector functions mediated by chemokine receptors on human eosinophils (34). Of all the derivatives studied, only CCL14[8–74], CCL14[9–74], CCL14[10–74], and NNY-CCL14 induced a significant release of ROS at concentrations up to 10–7 M (Fig. 2). These derivatives were compared at different doses with CCL11/eotaxin, which has been described as the most potent activator of the respiratory burst in eosinophils (40), and CCL14[1–74] as the naturally abundantly occurring form. As shown in Fig. 2, CCL14[9–74] and NNY-CCL14 were as efficacious as CCL11 in inducing the release of an identical amount of ROS at a concentration of 10–7 M, whereas CCL14[8–74] and [10–74] were less efficacious, and full-length CCL14[1–74] was virtually inactive. NNY-CCL14 was the most potent stimulus tested, being significantly active starting at concentrations of 10–9 M and almost reaching its maximal effect at 10–8 M (Fig. 2). The inactive analogues were further analyzed for antagonistic effects. Pretreatment of human eosinophils with 10–7 M CCL14[1–74], -[6–74], -[7–74], -[11–74], or -[12–74] or bis-NNY-CCL14[10–74] did not result in significant inhibition of ROS release after stimulation with CCL11 at identical concentrations (data not shown).
|
In the next set of experiments, human eosinophils were incubated for 30 min with the CCL14/HCC-1 derivatives or CCL11/eotaxin as positive control at a concentration of 10–7 M at 37°C. The cells were stained with anti-CCR3 mAb, and receptor expression was measured by flow cytometry. Preincubation of human eosinophils with CCL14[9–74], CCL14[10–74], and NNY-CCL14 led to a significant down-regulation of CCR3 (Fig. 3, A and B). The other derivatives, including CCL14[8–74], which is a weak inducer of ROS, did not affect CCR3 expression (Fig. 3A). At optimal doses, NNY-CCL14 and CCL11 both removed 80% of cell surface CCR3 and were significantly more effective than CCL14[9–74] (65%) and CCL14[10–74] (50%; Fig. 3A).
|
NNY-CCL14 is the most potent eosinophil chemoattractant among CCL14 derivatives and inhibits CCL11-induced chemotaxis
To evaluate whether the most active CCL14/HCC-1 derivatives induce chemotaxis of CCR3+ cells, migration assays were conducted in 48-well microchemotaxis chambers. As shown in Fig. 4A, CCL14 derivatives were effective chemoattractants for human eosinophils. The activity of the derivatives tested was similar in terms of efficacy, as indicated by the number of migrated cells. Maximal responses to CCL14[9–74] and CCL14[10–74] were reached at 3 x 10–7 and 10–6 M, respectively. NNY-CCL14 was more potent than the other derivatives, as its maximal effect was observed at 10–8 M, which is in the range of CCL11/eotaxin, as shown previously (8). When eosinophils were treated with 10–7 M NNY-CCL14 before transfer to the migration chamber, it almost abolished the migratory response to CCL11 (Fig. 4B).
|
CCL11/eotaxin-induced [Ca2+]i changes in human eosinophils are mediated exclusively via CCR3 (42). To study the potential of the active CCL14/HCC-1 derivatives to desensitize CCR3, we performed desensitization experiments with the most active CCL14 derivatives and CCL11. Stimulation of eosinophils with NNY-CCL14 or CCL14[9–74] at 10–7 M completely desensitized the cells to CCL11 at the same dose (Fig. 5, A and C). The less active form, CCL14[10–74], did not desensitize eotaxin at equal doses (Fig. 5E), in agreement with the results obtained for the release of ROS and the moderate CCR3 internalization after preincubation with this ligand. Stimulation of eosinophils with CCL11 completely desensitized the active CCL14 derivatives (Fig. 5, B, D, and F). These results demonstrate that NNY-CCL14 desensitizes CCR3 for CCL11 stimulation. The specificity for desensitization of CCR3 given as NNY-CCL14 is not able to desensitize the response of eosinophils toward C5a (Fig. 5A). The C5a peak after NNY-CCL14 stimulation had the same intensity as if the cells were stimulated with C5a alone (data not shown).
|
CD26/DPP IV processing of CCL11/eotaxin, CCL14[9–74], and NNY-CCL14/HCC-1 was analyzed in vitro, essentially as previously described (36). First the amount of enzyme applied was optimized to give a kinetic profile for CCL11 and CXCL12 comparable to that previously reported (Fig. 6A). CXCL12 was completely processed within 10 min, and CCL11[1–74] was fully converted into CCL11[3–74] after 1 h at 37°C. Using these conditions, the processing of CCL14[9–74] was performed similarly over a period of time. Although significantly slower than for CCL11 and CXCL12, a virtually complete conversion of CCL14[9–74] into CCL14[11–74] was achieved within 12 h (Fig. 6, B and C). In contrast to these chemokines, NNY-CCL14 remained completely resistant to CD26/DPP IV treatment after 24 h (Fig. 6, B and C) and even after 90 h of incubation (data not shown).
|
To test the hypothesis that NNY-CCL14/HCC-1 may influence the influx of eosinophils in vivo, we used two murine models of allergic airway inflammation. The i.v. application of 10 nmol/kg NNY-CCL14 30 min before and 3 and 8 h after OVA-aerosol provocation significantly reduced the infiltration of eosinophils into the airways of OVA-treated mice compared with the saline-treated positive control group. Eosinophil infiltration into the lung tissue was demonstrated by standard staining procedures and was clearly reduced in the NNY-CCL14-treated animals (Fig. 7). A reduced diapedesis and tissue infiltration of eosinophils was observed in the NNY-CCL14-treated animals, as shown in Fig. 7A. In the saline-treated group, a diffuse infiltrate consisting mainly of eosinophils was found (Fig. 7B). To exclude inappropriate cellular activation that might lead to trapping of eosinophils within the lung, nonallergen-sensitized/challenged mice were treated with NNY-CCL14 i.v. once at a dose of 10 nmol/kg vs NaCl i.v. NNY-CCL14-treated mice did not exhibit any trapping or sludging of cells in the vessels of the lung, as observed in those mice killed 30 min or 24 h after NNY-CCL14 treatment (Fig. 7, C and D). The inconspicuous image was identical in the saline-treated animals (data not shown). For quantification of the effect on inflammation, differential cell counts were performed on cytospins of BALF. Application of NNY-CCL14 significantly reduced the influx of eosinophils into the airways compared with the saline-treated group (0.2 vs 1.3 x 104 cells/ml BALF; p = 0.007; Fig. 8). To evaluate the effect of NNY-CCL14 treatment on AHR, lung function measurements were additionally performed. Application of NNY-CCL14 significantly reduced AHR, measured as the effective dose of methacholine to decrease midexpiratory flow to 50% (ED50). The AHR in saline-treated and OVA-challenged animals was 29.6 ± 7.6 µg of methacholine (positive control) and improved in NNY-CCL14-treated animals to 65.1 ± 13.8 (p = 0.016). For comparison, nonsensitized animals had an ED50 of 61.3 ± 15.1 µg of methacholine.
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recently, the importance of CCR3 in allergic asthma has been highlighted. This receptor is expressed constitutively or upon activation of cytokines on eosinophils, Th2 cells, basophils, and mast cells (49). All these cells contribute to the inflammatory infiltrate in allergic asthma (1). In contrast to eosinophils and basophils, the expression of CCR3 on a subpopulation of Th2 cells has been controversially discussed (17, 50). A recent study using CCR3 knockout mice demonstrated that the expression of the Th2 cytokine, IL-4, in OVA-sensitized skin was comparable in CCR3–/– mice and wild-type controls, suggesting that CCR3 does not play an important role in the recruitment of Th2 cells to sites of allergic inflammation (21). In contrast to mice, in skin lesions of patients with atopic dermatitis, there is evidence of a role for CCR3 in the recruitment of a subpopulation of T cells (51). However, the in vivo relevance of CCR3 for the attraction of T cells to the site of inflammation has to be evaluated in additional studies. The deletion of the CCR3 locus in the germline of mice gave new insight into the role of this receptor especially in the trafficking of eosinophil cells into the lung (20, 21). In a model of allergen-induced airway inflammation, these studies showed that allergen challenge results in subendothelial trapping of eosinophils in CCR3-deficient mice, whereas wild-type controls had an impressive infiltration of the lung accompanied by lymphocytes, neither of which was found in CCR3–/– mice. In addition, the CCR3-deficient mice are protected from allergen-induced AHR if the epicutaneous route is used for sensitization instead of the i.p. route, where an enhanced AHR was observed that correlated with increased intraepithelial mast cells, underlining the involvement of CCR3 in several phases of asthma (20, 21).
A previous study by our group demonstrated that the N-terminally truncated form of CCL14, CCL14[9–74], is an agonist of CCR3 (29). According to these data, we investigated a greater series of N-terminally truncated or modified peptides derived from the chemokine CCL14 to test their activity profile on CCR3 using human eosinophils expressing CCR3, but not CCR1. The derivatives CCL14[6–74/7–74/8–74/11–74/12–74] and bis-NNY-CCL14[10–74] show modest activity or are inactive on eosinophils expressing CCR3 and do not have antagonistic properties on this receptor. In contrast to these derivatives, CCL14[9–74], CCL14[10–74], and NNY-CCL14 evoke the release of ROS and chemotaxis of human eosinophils. The rank order of activity of these peptides was NNY-CCL14 > CCL14[9–74] > CCL14[10–74]. The activity of NNY-CCL14 on CCR3 was confirmed, because it induced mobilization of intracellular calcium in the nanomolar range in 300.19 cells transfected with human CCR3 (data not shown).
The inhibition of chemokine receptors through internalization is a reasonable therapeutic strategy being pursued for HIV and its interaction with CCR5. In this context, it has been shown that AOP-RANTES and NNY-RANTES induce a powerful internalization that improves effective inhibition of HIV entry in CCR5-transfected cell lines (52, 53). In this study we show that the active CCL14 derivatives are able to induce internalization of CCR3 with the same rank order as that for the release of ROS and chemotaxis. Moreover, of the active CCL14 derivatives, NNY-CCL14 was the most efficient to desensitize CCR3, as shown by its inhibition of CCL11-induced intracellular calcium release of human eosinophils. Additionally, pretreatment of human eosinophils with NNY-CCL14 inhibits CCL11/eotaxin-induced chemotaxis, overlapping with its capacity of CCR3 internalization. In this context, data on other modified chemokines strongly support the idea that receptor internalization plays a central role in chemokine-mediated inhibition of receptor function (34, 52, 54).
The enzyme CD26/DPP IV is an abundant peptidase found in blood plasma and tissue and on the cell surface of various cell types (30, 36). It is involved in the regulation of chemokine activity by hydrolysis of peptides with N-terminal Xaa-Pro and Xaa-Ala motifs (30). Truncation of chemokines by CD26/DPP IV, such as CCL5, CCL11, CXCL11, and CXCL12, leads to rapid inactivation and reduced binding capacity to their corresponding chemokine receptors (36). The most potent CCR3 peptide antagonist yet described, I-Tac/E0H1, contains the eight N-terminal amino acids of CXCL11 (I-TAC) (55). As CXCL11 is efficiently cleaved by CD26/DPP IV within 2 min (56), I-TAC/E0H1 is likely to share a similar fate. In this study we demonstrate that, in contrast to CCL11, the most potent of the CCL14 derivatives, NNY-CCL14 is resistant to degradation by CD26/DPP IV even after incubation for up to 90 h. Thus, substitution of the N-terminal Gly-Pro motif by the NNY-Pro motif protects NNY-CCL14 from degradation. This is a relevant difference in comparison with CCL11, which is processed rapidly into its inactive form, CCL11[3–74], resulting in a reduced interaction of CCL11 with CCR3 (57). In contrast, NNY-CCL14 remains in its active form to induce CCR3 internalization, desensitization, and inhibition of chemotaxis. In this context, latest experiments show that serum incubation of NNY-CCL14 for 48 h does not result in the loss of biological activity, with respect to CCR3 in vitro internalization, whereas CCL11’s activity is clearly affected (U. Forssmann, unpublished observations).
The therapeutic use of chemokines or their derivatives with antagonistic or agonistic properties, such as Met-RANTES, AOP-RANTES, or NNY-RANTES, has been intensively discussed in the literature, particularly for the inhibition of HIV infection (44). The treatment of asthma is one rationale for the development of CCR3 antagonists such as I-TAC-eotaxin hybrid-1 (55). Besides peptides, small m.w. antagonists are also considered potential drugs for the blockade of chemokine receptors. Because the latter compounds, derived from piperazine and piperidine, may exhibit unexpected side effects (particularly in the heart and CNS), peptide ligands inactivating chemokine receptors may represent an alternative due to better tolerance. There are already agonistic drugs, which are applied to induce inhibitory effects, such as luteinizing hormone-releasing hormone analogues, which are used to down-regulate hormone release (58). Hence, we propose the concept of an agonistic receptor inactivator mediating its effects through desensitization and internalization, thereby rendering target cells insensitive to further activation through the inactivated receptors.
A previous study focusing on the in vivo role of IL-8 in normal healthy NZW rabbits demonstrated that i.v. application of IL-8 blocked PMN extravasation (elicited 30 min after i.v. injection by s.c. IL-8 or FMLP application) during a 120-min interval. The authors of this study explained their results by adhesion inhibition rather than by desensitization (59). The transient fall of neutrophils during the first 30 min as described by Hechtman et al. (59 seems to be a feature of neutrophils, but not eosinophils. It has been shown that in guinea pigs, circulating eosinophil numbers increased after 5 min, with maximal levels at 30 min after i.v. eotaxin injection (60). Beyond the study by Hechtman et al. (59 , we used two murine models of allergic airway inflammation, which are relevant for allergic lung diseases, to prove whether NNY-CCL14 is able to block the infiltration of eosinophils into the lung tissue. Intravenous administration of NNY-CCL14 before and after allergen challenge was well tolerated, indicating that a potential intravascular activation of eosinophils was not harmful for the animals. Furthermore, the treatment of nonallergen-sensitized/challenged mice with NNY-CCL14 did not induce trapping of cells in lung vessels after 30 min, the time point at which such a phenomenon would be most likely expected. An identical observation was made after 24 h. In a separate experiment three nonsensitized mice were injected with a single dose of 1.000 nmol/kg NNY-CCL14, which was very well tolerated during an observation period of 1 wk (U. Forssmann and A. Heitland, unpublished observations). Remarkably, the number of eosinophils in BALF as well as lung tissue decreased dramatically in the NNY-CCL14-treated group. Both OVA and A. fumigatus are frequently used allergens for pharmacological efficacy studies (61, 62, 63). However, these allergens clearly differ in the picture of inflammation they induce. OVA induces a Th2 phenotype associated with an almost exclusive influx of eosinophils in this model (62). In contrast, A. fumigatus induces a more mixed immune response, with a proinflammatory and a Th2 component (61). According to this different phenotype of allergic inflammation, systemic treatment with NNY-CCL14 resulted in a reduction of eosinophil numbers in the OVA model and the reduction of eosinophil and lymphocyte numbers in the Aspergillus model. Moreover, AHR to methacholine was tested by head-out plethysmography in the OVA model. The reduced number of eosinophils in the NNY-CCL14 treatment group was associated with the prevention of AHR. Eosinophil numbers are well-established markers of allergic airway inflammation, but their contribution to AHR is controversially discussed (64, 65). Therefore, it remains open whether the effect of NNY-CCL14 on AHR is mediated by eosinophils only, or if other mechanisms are involved as well.
We suggest from the presented study that the blood represents an ideal compartment for application of such an agonistic receptor inactivator, as generated toxic substances such as reactive oxygen species accumulate much more slowly than in inflamed tissues. Therefore, peptides would be interesting candidates because they are mainly administered parenterally, are usually highly specific, and consequently have relatively low systemic toxicity (66). Finally, we speculate that CD26/DPP IV-resistant agonists, such as NNY-CCL14, which is able to efficiently internalize its receptors, can be given at significantly lower doses than antagonists such as Met-RANTES to inhibit extravasation of effector cells (19). Additional studies are necessary to demonstrate the proposed mechanism in vivo of NNY-CCL14 and to exclude an involvement of other receptors.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 This work was supported in part by a grant from the Deutsche Forschungsgemeinschaft (SFB 587, B2) and the Bundesministerium für Bildung und Forschung (FKZ 0311815). 
2 Address correspondence and reprint requests to Dr. Ulf Forssmann, IPF PharmaCeuticals, An-Institut of the Hannover Medical School, Feodor Lynen Strasse 31, 30625 Hannover, Germany. E-mail address: u.forssmann{at}ipf-pharma.de
3 Abbreviations used in this paper: BALF, bronchoalveolar lavage fluid; AHR, airway hyper-responsiveness; AOP, amino-oxypentane; bis-NNY, bis-n-nonanoyl; [Ca2+]i, intracellular calcium concentration; CD26/DPP IV, dipeptidylpeptidase IV; NNY, n-nonanoyl; PAO, phenylarsine oxide; ROS, reactive oxygen species; HCC-1, hemofiltrate CC chemokine-1.
Received for publication February 2, 2004. Accepted for publication June 11, 2004.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chemokine macrophage inflammatory protein 4, a property strongly enhanced by an amino-terminal alanine-methionine swap. J. Immunol. 164:1488.This article has been cited by other articles:
|
|
 |
), CXCL12/SDF-1
), and NNY-CCL14 (
; 10–5 M each) were incubated for the indicated times with porcine kidney CD26/DPP IV as described in Materials and Methods and were analyzed using mass spectrometry. Processing was calculated as the amount of full-length chemokine related to the total amount of the full-length and processed forms, as defined by the peak heights. C, Partial MALDI mass spectrometry spectra of chemokines after incubation for different periods. The relative molecular masses of unprocessed chemokines are indicated on the right of the peaks, and those of processed chemokines (minus two NH2-terminal amino acids) are given on the left.
