Abstract
Vascular endothelial growth factor (VEGF) is a pleiotropic polypeptide that mediates endothelial-cell-specific responses such as induction of proliferation and vascular leakage. We examined the expression of VEGF messenger RNA (mRNA) and protein by human eosinophils in response to granulocyte–macrophage colony-stimulating factor (GM-CSF) and interleukin-5 (IL-5). Immunoreactive VEGF protein was detected in freshly isolated eosinophils by immunocytochemistry. Eosinophils spontaneously released VEGF protein in culture medium, and this release was upregulated by GM-CSF or IL-5. Freshly isolated eosinophils constitutively expressed VEGF mRNA. Although incubation of eosinophils in culture medium reduced steady-state VEGF mRNA levels, eosinophil VEGF mRNA levels were enhanced by GM-CSF and IL-5, and this enhancement was blocked by the transcription inhibitor actinomycin D. Analysis of alternatively spliced mRNA species revealed that eosinophils contained transcripts mainly encoding for the 121- and 165-amino-acid forms of VEGF. VEGF mRNA expression and VEGF release in cytokine-stimulated eosinophils were significantly reduced by treatment with a glucocorticosteroid, a protein-tyrosine kinase inhibitor, or a protein kinase C inhibitor. Cytokine-activated eosinophils may be an important source of a vascular permeability factor, namely VEGF, thus contributing to tissue edema formation at sites of allergic inflammation.
Eosinophils are blood leukocytes associated with helminthic infections and allergic diseases, especially bronchial asthma (1, 2). Recent studies have shown that eosinophils synthesize a number of cytokines (3) that may contribute to the pathogenesis of asthma. Vascular endothelial growth factor (VEGF), also known as vascular permeability factor, is a potent, multifunctional cytokine that exerts several important actions on the vascular endothelium (4, 5). VEGF was purified on the basis of its ability to induce transient vascular leakage and endothelial cell mitogenesis. It is a heparin-binding glycoprotein occurring in four molecular forms, generated by alternative splicing, that contain 121 (VEGF121), 165 (VEGF165), 189 (VEGF189), or 206 (VEGF206) amino acids (6, 7). VEGF is produced by many cell types, including pituitary cells (8), smooth-muscle cells (9), cardiac myocytes (10), keratinocytes (11), mesangial cells, and peripheral blood mononuclear cells (12).
Plasma exudation from the bronchial microvasculature and the formation of tissue edema are characteristic features of asthma that may be mediated by active substances such as platelet-activating factor (PAF), leukotrienes, histamine, and VEGF (13). Although eosinophils, prominent inflammatory cells found in asthmatic airway tissues, are well known to release PAF and leukotrienes (14, 15), the capacity of eosinophils to produce VEGF has not been described. Production by eosinophils of VEGF, which enhances vascular permeability, might contribute to the tissue edema and airway hyperresponsiveness in asthma associated with eosinophilia (16).
Eosinophilopoietic cytokines, such as granulocyte– macrophage colony-stimulating factor (GM-CSF) and interleukin-5 (IL-5), stimulate various eosinophil functions and prevent eosinophil apoptosis (17, 18). Expression of GM-CSF and IL-5 by inflammatory cells has been demonstrated in bronchial biopsy material from asthmatic individuals (19). Studies indicate that these eosinophilopoietic cytokines have the ability to enhance the generation of several cytokines by eosinophils (20-24). In this study, we evaluated whether human eosinophils produce VEGF, and whether such eosinophil VEGF production is effected by stimulation with GM-CSF or IL-5, or inhibited by dexamethasone (DEX), a widely used antiinflammatory glucocorticosteroid known to induce eosinophil apoptosis (25).
Recombinant human (rh) GM-CSF and rhIL-5 (R&D Systems, Minneapolis, MN) and actinomycin-D, DEX, genistein, and chelerythrine (Sigma Chemical Co., St. Louis, MO) were purchased from their manufacturers. Actinomycin D, genistein, and chelerythrine were dissolved in dimethylsulfoxide (DMSO) at 5 mg/ml, 10 mg/ml, and 10 mM, respectively.
Citrate-anticoagulated peripheral blood was obtained from 18 healthy volunteers. After dextran (McGaw, Irvine, CA) sedimentation of erythrocytes, the granulocyte pellet was obtained by density-gradient centrifugation on Ficoll-Paque (Pharmacia, Uppsala, Sweden). Mononuclear cells were prepared from the interface. Eosinophils were purified by negative immunomagnetic selection, using a magnetically activated cell separator system (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) as previously described (26). To eliminate mononuclear cells, anti-CD3-, anti-CD14-, and anti-CD19-coated micromagnetic beads were added to the anti-CD16/granulocyte mixture. By negative selection, highly purified eosinophils (> 99.5%), depleted of neutrophils (CD16+) and any contaminating mononuclear cells (CD3+/CD14+/CD19+), were routinely obtained. Eosinophils were cultured for the indicated times in RPMI medium supplemented with 10% fetal calf serum (FCS), 100 U/ml penicillin, and 100 μg/ml streptomycin in the absence or presence of 20 ng/ml GM-CSF or 20 ng/ml IL-5. Eosinophil viability after culture was examined by trypan blue exclusion. In experiments in which the effects of inhibitors were tested, inhibitors were added 15 min before cytokine stimulation and remained present throughout the incubation. These inhibitors, at the concentrations used in this study, did not diminish eosinophil viability. In addition, the frequency of eosinophil apoptosis was evaluated morphologically in cytocentrifuge preparations of eosinophils fixed in methanol and stained with Diff-Quick (Baxter Laboratories, McGaw Park, IL). Two hundred cells/slide were examined microscopically to determine the number of cells with apoptotic morphology (27). The morphologic changes reflecting apoptosis in eosinophils consisted of cell shrinkage, nuclear coalescence (monolobed nucleus), nuclear condensation, and the anucleate state.
Antihuman VEGF antibody, which was raised in a rabbit with the amino-terminal epitope (residues 1 to 20) of human VEGF as the immunogen, and thus recognizes the four splice variants depicted in Figure 1, was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Immunocytochemistry was performed on cytospin preparations of freshly isolated peripheral blood eosinophils as described previously (26). After fixation in 4% paraformaldehyde (Sigma) for 10 min and permeabilization in 0.05% saponin (Sigma), slides were incubated with rabbit antihuman VEGF polyclonal antibody (Santa Cruz Biotechnology) or a control nonspecific rabbit IgG (Jackson Immunoresearch, West Grove, PA) as primary antibody for 1 h. Following incubation with biotin-conjugated secondary antibody and avidin–glucose oxidase enzyme complex (Vector, Burlingame, CA), slides were developed with nitroblue tetrazolium (Vector) as the glucose oxidase substrate.
Fig. 1. The four possible splice variants of VEGF. Positions of oligonucleotide primers (arrows) and probes (bars) are indicated for VEGF206.
[More] [Minimize]Eosinophils (2 × 106 cells/ml in complete RPMI medium, at 250 μl/well in 96-well tissue culture plates) were cultured in the presence or absence of the indicated cytokines and drugs for up to 15 h. VEGF contents in supernatants were measured by a specific enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems). The minimum detectable concentration of VEGF (sensitivity) with this kit is 5.0 pg/ml.
Eosinophils (2 × 106 cells/ml in supplemented RPMI medium, 1 ml/tube in tissue culture tubes) were cultured in the presence or absence of the indicated cytokines and drugs for 6 h. Total cellular RNA was extracted from freshly isolated eosinophils or cultured eosinophils, using RNAsol (Biotek, Houston, TX) with the addition of 10 μg glycogen (Boehringer Mannheim, Indianapolis, IN). First-strand complementary DNA (cDNA) was generated by reverse transcription (RT) of RNA with an oligo-dT primer (Promega, Madison, WI) and avian myeloblastosis virus reverse transcriptase (Promega), and was used for amplification of VEGF or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA through the polymerase chain reaction (PCR). After 35 cycles of amplification, PCR products were run on a 1.2% agarose gel and stained with ethidium bromide. The amplified GAPDH band was used to normalize the cDNA among the samples. VEGF PCR products run on an agarose gel were transferred to nylon filters by capillary blotting, and the filters were hybridized with an oligonucleotide probe labeled with digoxigenin (DIG; Boehringer Mannheim). Filters were treated with anti-DIG–alkaline phosphatase and Lumi-Phos 530 (Boehringer Mannheim) as a chemiluminescent substrate, and were subsequently exposed to X-ray film.
In experiments in which modulation of VEGF mRNA levels by cytokines or inhibitors was tested, the sense primer 5′-GCCTCGCCTTGCTGCTCTACC-3′ (exon 1) and the antisense primer 5′-CACACTCCAGGCCCTCGT- CATTG-3′ (exon 3) were used to generate 252-bp PCR products, which were detected by the DIG-labeled oligonucleotide probe 5′-ATGGCAGAAGGAGGAGGGCAGAATC-3′ (exon 2). In experiments in which alternative splicing of VEGF mRNA was studied, the sense primer 5′-GCGGGGGCTGCTGCAATGACGA-3′ (exon 3) and the antisense primer 5′-CGCCCTCCGGACCCAAAGTGCTCT-3′ (exon 8), encompassing alternatively spliced exons 6 and 7, were used, and PCR products were detected with the DIG-labeled oligonucleotide probes 5′-TCACCAAGGCCAGCACATAGGAGAG-3′ (exon 4), 5′-CGAGGAAAGGGAAAGGGGCAAAAACGAA-3′ (exon 6), and 5′-CAAGATCCGCAGACGTGTAAATGTTCCT-3′ (exon 7). The expected sizes of the PCR products are 394 bp, 526 bp, 598 bp, and 649 bp for VEGF121, VEGF165, VEGF189, and VEGF206, respectively. Figure 1 illustrates the localizations of primers and probes within the VEGF cDNA. For GAPDH cDNA amplification, the sense primer 5′-GGGGAGCCAAAAGGGTCATCATCT-3′ and antisense primer 5′-GACGCCTGCTTCACCACCTTCTTG-3′ were used.
Competitive PCR was performed with the PCR MIMIC kit (Clontech, Palo Alto, CA), as described elsewhere (28), in order to examine VEGF mRNA expression in eosinophils. Total cellular RNA was extracted from eosinophils and reverse-transcribed as described earlier. The PCR for 252-bp VEGF cDNA amplification was spiked with serial dilutions of a 340-bp competitor DNA (internal standard) fragment synthesized with this kit. After 35 cycles of amplification, the 252-bp VEGF target and 340-bp internal standard PCR products were stained with ethidium bromide in an agarose gel and analyzed by densitometry. Ratios of internal standard/VEGF densitometric values were plotted against the input amounts of internal standard. The amount of VEGF cDNA was calculated from the equivalence point (ratio = 1) by regression analysis, assuming that it was equal to the amount of internal standard at this dilution. Competitive PCR for the housekeeping gene GAPDH mRNA was performed in a similar manner, to normalize VEGF mRNA expression for variations due to RNA extraction and RT (29).
The statistical significance of differences between various treatment groups was assessed with the paired Student's t test.
To determine whether resting human eosinophils contain preformed VEGF, cytocentrifuged preparations of freshly isolated peripheral blood eosinophils were permeabilized with saponin and incubated with anti-VEGF antibody. With anti-VEGF antibody, eosinophils exhibited positive intracellular staining in a punctate, granule-associated pattern, whereas no staining was observed with control antibody (Figure 2). Eosinophils from all four donors were likewise positive, with more than 75% of cells exhibiting positive staining for VEGF. No alteration in VEGF staining was noted in eosinophils cultured in GM-CSF or IL-5 (not shown).
Fig. 2. Detection of preformed VEGF protein in eosinophils through immunocytochemistry. Results of a representative experiment showing the presence of immunoreactive VEGF (A) within freshly isolated eosinophils, compared with a negative IgG control (B). VEGF appears to be associated with cytoplasmic granules. Original magnification: ×600.
[More] [Minimize]The capacity of human peripheral blood eosinophils to release VEGF was evaluated with eosinophils cultured for up to 15 h without and with 20 ng/ml rhGM-CSF or 20 ng/ml rhIL-5. Under each of these conditions, eosinophil viability after 15 h of culture was routinely > 90%. Without cytokines, 10.4 ± 1.9% of eosinophils exhibited morphologic changes of apoptosis, whereas apoptosis was significantly less common (P < 0.05, n = 5, for each) among eosinophils cultured in GM-CSF (5.8 ± 1.4%) or IL-5 (5.4 ± 1.5%). Increasing amounts of immunoreactive VEGF were released from eosinophils, in an effect that was enhanced by the addition of GM-CSF or IL-5 (Figure 3A). Whereas eosinophils cultured in medium alone released 11.6 pg of VEGF/2 × 106 cells after 15 h, eosinophil VEGF release in the presence of 20 ng/ml of rhGM-CSF or rhIL-5, increased significantly to 28.7 pg/2 × 106 cells and 22.7 pg/2 × 106 cells, respectively. For comparison, peripheral blood mononuclear cells spontaneously released approximately 2-fold more VEGF under the same culture conditions, whereas stimulation with GM-CSF or IL-5 for 15 h did not increase VEGF release from these cells (data not shown). Dose–response analysis showed that GM-CSF or IL-5 at concentrations as low as 0.2 ng/ml caused significant enhancement of eosinophil VEGF release, whereas optimum release was obtained at higher concentrations of up to 20 ng/ml (Figure 3B).
Fig. 3. Effect of GM-CSF or IL-5 on VEGF release from eosinophils. (A) Eosinophils were incubated for up to 15 h in the absence or presence of either 20 ng/ml GM-CSF or 20 ng/ml IL-5. After the indicated incubation times, culture supernatants were analyzed with an ELISA for VEGF protein. The means + SEM of four experiments are shown. *P < 0.05, paired t test, compared with conditions without cytokine supplementation. (B) Eosinophils were cultured for 15 h in the presence of various concentrations of GM-CSF or IL-5. Incubation media were analyzed with an ELISA for VEGF protein. The means + SEM of four experiments are shown. *P < 0.05, paired t test, compared with conditions without cytokine supplementation.
[More] [Minimize]Constitutive expression of VEGF mRNA transcripts was readily detectable in freshly isolated eosinophils (12 of 12 donors) with RT-PCR followed by Southern hybridization (data not shown). To investigate the regulation of VEGF mRNA, we stimulated eosinophils with GM-CSF or IL-5 for 6 h in the presence or absence of 2.5 μg/ml actinomycin D. Total cellular RNA obtained from the eosinophils after culture was reverse-transcribed, and the resulting cDNA was subjected to competitive PCR analysis to quantitate the amount of VEGF cDNA (Figure 4A). Whereas incubation in culture medium alone caused downregulation of VEGF mRNA transcripts as compared with freshly isolated eosinophils, eosinophils stimulated with rhGM-CSF or rhIL-5 exhibited higher levels of steady-state VEGF mRNA (Figure 4B). Addition of actinomycin D almost completely abolished VEGF mRNA (Figure 4B), indicating that the increased quantity of VEGF mRNA detected in eosinophils after culture with GM-CSF or IL-5 was dependent on de novo RNA synthesis.
Fig. 4. Expression of VEGF mRNA in eosinophils. (A) Competitive PCR quantitation of eosinophil VEGF cDNA. Eosinophil cDNA was subjected to PCR amplification of VEGF cDNA in the presence of internal standard (lanes from left to right contain 2.5, 5, 10, 20, 40, 80, 160, and 320 fg of cDNA). PCR products were stained with ethidium bromide in an agarose gel and analyzed by densitometry. Ratios of internal standard/VEGF densitometric values were plotted against the input amounts of internal standard. The amount of eosinophil VEGF cDNA was calculated from the equivalence point. A representative analysis is demonstrated. (B) Modulation of VEGF mRNA levels in eosinophils. Eosinophils were cultured for 6 h with medium alone or with 20 ng/ml of GM-CSF or IL-5 in the presence or absence of 2.5 μg/ml actinomycin D. RT was performed on RNA from freshly isolated eosinophils or eosinophils after culture. Eosinophil VEGF cDNA and GAPDH cDNA were analyzed by competitive PCR. The amount of VEGF cDNA was calculated and corrected for GAPDH cDNA. Values obtained from freshly isolated eosinophils were arbitrarily set to 100. The means + SEM of four experiments are shown. *P < 0.05, paired t test, compared with medium control; **P < 0.05 paired t test, compared with the same condition without actinomycin D.
[More] [Minimize]The VEGF gene contains eight exons that are expressed in different combinations by alternative splicing mechanisms (Figure 1). We performed RT-PCR with oligonucleotide primers flanking alternatively spliced regions of VEGF mRNA (Figure 5). Although amplification of eosinophil cDNA resulted in two bands, of 394 bp and 526 bp, respectively, amplification of mononuclear cell cDNA yielded an additional 598-bp band, all of which hybridized with an oligonucleotide positioned in exon 4 of the VEGF gene. When we rehybridized the same blots with oligonucleotide probes that recognize exons 6 and 7, which are deleted in the known splice variants of VEGF, we found that the major 394-bp band lacked both exons, as expected for VEGF121, whereas the 526-bp band contained exon 7 but not exon 6, which is characteristic of VEGF165. The presence of both exons identified the 598-bp band found in PCR products from mononuclear cells, but not in those from eosinophils, as cDNA corresponding with VEGF189.
Fig. 5. Characterization of mRNA splice variants by RT-PCR and Southern analysis. RT-PCR was performed on RNA isolated from mononuclear cells (MN) and from eosinophils (Eo). Amplification products were hybridized with VEGF-specific oligonucleotide probes located in exon 4, exon 7, and exon 6. Numbers at the left are sizes of the DNA in base pairs. This figure is representative of three experiments that yielded similar results.
[More] [Minimize]To investigate whether a glucocorticosteroid could modulate VEGF production by eosinophils, we examined the effect of DEX treatment on cytokine-induced VEGF secretion from eosinophils. At concentrations of 10 nM or more, DEX significantly reduced VEGF release from eosinophils stimulated for 15 h with GM-CSF or IL-5 (Figure 6). Spontaneous VEGF release from eosinophils was moderately reduced by DEX. Morphologic analysis was performed to examine the effect of DEX treatment on eosinophil apoptosis after 15 h of culture with or without cytokine supplements. For eosinophils cultured without cytokine supplements, DEX (1 μM), as expected, induced increased numbers of apoptotic eosinophils (10.4 ± 1.9% versus 14.8 ± 2.7%, P < 0.05, n = 5). In contrast, there were no differences in apoptosis for eosinophils cultured without and with DEX, respectively, in the presence of 20 ng/ml GM-CSF (5.8 ± 1.4% versus 6.2 ± 1.5%, n = 5) or 20 ng/ml IL-5 (5.4 ± 1.5% versus 4.6 ± 0.9%, n = 5). Thus, DEX suppression of cytokine-stimulated VEGF release was not due to apoptosis.
Fig. 6. Effect of DEX on VEGF release from activated eosinophils. Eosinophils were cultured in medium alone or were stimulated with either 20 ng/ml GM-CSF or 20 ng/ml IL-5 for 15 h in the presence of various concentrations of DEX. Incubation media were analyzed with an ELISA for VEGF protein. The means + SEM of four experiments are shown. *P < 0.05, paired t test, compared with control culture without DEX.
[More] [Minimize]To investigate the roles of protein tyrosine kinase (PTK) and protein kinase C (PKC) in VEGF production induced by GM-CSF and IL-5, the effects of genistein, a selective PTK inhibitor, and chelerythrine, a selective PKC inhibitor, were examined (Figure 7). Genistein had no effect on basal VEGF release from eosinophils, but it significantly reduced GM-CSF- and IL-5-induced VEGF release. Chelerythrine almost completely abolished basal and cytokine-induced VEGF release from eosinophils.
Fig. 7. Effect of genistein and chelerythrine on VEGF release from eosinophils. Eosinophils were cultured in medium alone or with 20 ng/ml of either GM-CSF or IL-5 for 15 h in the presence or absence of 20 μg/ml genistein or 5 μM chelerythrine. Incubation media were analyzed with an ELISA for VEGF protein. The means + SEM of four experiments are shown. *P < 0.05, paired t test, compared with control culture without inhibitors.
[More] [Minimize]We also examined the effect of treatment with DEX, genistein, or chelerythrine on VEGF mRNA expression by eosinophils stimulated with GM-CSF or IL-5, using the competitive PCR technique. As shown in Figure 8, addition of DEX, genistein, or chelerythrine significantly suppressed steady-state levels of VEGF mRNA in cytokine-stimulated eosinophils.
Fig. 8. Effect of DEX, genistein, or chelerythrine on VEGF mRNA expression by cytokine-stimulated eosinophils. RT was performed on RNA obtained from eosinophils stimulated with GM-CSF or IL-5 in the presence or absence of 1 μM DEX, 20 μg/ ml genistein, or 5 μM chelerythrine for 6 h. Eosinophil VEGF cDNA and GAPDH cDNA were analyzed by competitive PCR. The amount of VEGF cDNA was calculated and corrected for GAPDH cDNA. Values obtained from GM-CSF- or IL-5-stimulated eosinophils in the absence of drugs were arbitrarily set to 100. The means + SEM of four experiments are shown. *P < 0.05, paired t test, compared with culture without drugs.
[More] [Minimize]In this study, we demonstrate that human eosinophils have the capacity to generate VEGF, a vascular-endothelial-cell-specific cytokine that mediates angiogenesis and vascular permeability. Spontaneous release of immunoreactive VEGF protein by peripheral blood eosinophils in culture supernatants was readily detectable with a sensitive ELISA. VEGF transcripts were found in highly purified eosinophils with RT-PCR, and eosinophils were shown to contain VEGF protein through immunocytochemistry. These findings extend the number of growth factors synthesized by eosinophils, beyond transforming growth factor-α (TGF-α) (30), transforming growth factor-β1 (TGF-β1) (31), heparin-binding epidermal growth factor-like growth factor (32), and platelet-derived growth factor (PDGF) (33).
The development and function of eosinophils are regulated by a number of cytokines, including GM-CSF and IL-5. GM-CSF stimulates the development of eosinophils as well as other leukocytes, whereas IL-5 is specific in promoting the development and terminal differentiation of eosinophils (34, 35). GM-CSF and IL-5 also support the survival of mature eosinophils and modulate their functional activities (17). We studied the effects of GM-CSF and IL-5 on VEGF protein and mRNA expression by eosinophils. Stimulation of eosinophils with GM-CSF or IL-5 resulted in increased VEGF release in culture supernatant. The immunocytochemical localization of VEGF in eosinophils demonstrated that eosinophils contain preformed quantities of VEGF, which are likely to be stored in cytoplasmic granules, as shown for other eosinophil-derived cytokines including tumor necrosis factor-α (TNF-α) (36), GM-CSF (37), and IL-4 (38). Since IL-5 and GM-CSF can stimulate degranulation of eosinophils (1), the enhanced release of VEGF protein elicited by these two cytokines may in part be due to release from granule stores.
GM-CSF and IL-5 also stimulated VEGF gene transcription. Steady-state VEGF mRNA levels in eosinophils declined when these cells were incubated in culture medium for 6 h. Moreover, addition of the transcription inhibitor actinomycin D almost completely abolished VEGF mRNA, suggesting that VEGF mRNA is short-lived and continuously synthesized in cultured eosinophils. In the presence of GM-CSF or IL-5, VEGF mRNA levels in cultured eosinophils were enhanced. Since actinomycin D blocked the enhancing effects of GM-CSF and IL-5, it is likely that these two cytokines were acting to enhance VEGF gene expression. Thus, GM-CSF and IL-5 appeared to stimulate de novo synthesis of VEGF mRNA.
Human cells may express four different VEGF transcripts, encoding polypetides of 121, 165, 189, and 206 amino acids, through the alternative splicing of a single gene whose coding region is divided among eight exons. These different isoforms apparently express identical biologic activities. However, VEGF121, and to a large extent VEGF165, are secreted in soluble form, whereas the two larger isoforms (VEGF189, VEGF206) remain cell-associated, perhaps because of their greater affinity for cell-surface proteoglycans (7, 39). We examined the expression of mRNA for splice variants by human peripheral blood mononuclear cells and eosinophils, using RT-PCR and Southern hybridization. Whereas RT-PCR of a mononuclear cell population demonstrated three alternatively spliced forms (VEGF121, VEGF165, VEGF189), as previously reported (12), analysis of eosinophil-derived VEGF mRNA by RT-PCR yielded two amplification products, which were shown by Southern hybridization to correspond to VEGF121 and VEGF165. Two other, alternatively spliced forms of VEGF (VEGF189, VEGF206) were not detected, indicating that these forms are not efficiently produced in eosinophils.
DEX both suppressed VEGF product release from eosinophils and diminished steady-state levels of eosinophil VEGF mRNA. Inhibition of cytokine mRNA transcription by glucocorticoids can result from interaction of the activated glucocorticoid receptor with transcription factors, such as activator protein-1 (AP-1) (40), whose response element is located in the VEGF gene promoter (6). Thus, it is conceivable that inactivation of transcription factors by DEX treatment resulted in downregulation of VEGF mRNA and contributed to decreased VEGF release from eosinophils. Although DEX can stimulate eosinophil apoptosis, eosinophil apoptosis was not increased by DEX in the presence of GM-CSF and IL-5, as noted previously (40, 41). Therefore, decreased VEGF release from GM-CSF- or IL-5-stimulated eosinophils in the presence of DEX was not related to the ability of DEX to induce eosinophil apoptosis. Inhibition by glucocorticoids of VEGF production by activated eosinophils may contribute to the effects of these drugs in alleviating symptoms of asthma.
Recent articles have reported that ligand binding to GM-CSF and IL-5 receptors activates nonreceptor PTK and several other protein kinases (42, 43). The present study evaluated the roles of PTK and PKC in the induction of VEGF expression by eosinophils stimulated with GM-CSF or IL-5. The results show that signal-transduction pathways involving PTK and PKC are active in mediating actions of GM-CSF and IL-5 on VEGF mRNA and protein expression. Genistein did not inhibit basal VEGF secretion, but this specific PTK inhibitor did inhibit enhancement of VEGF secretion by GM-CSF and IL-5, indicating a role for PTK activation in GM-CSF- and IL-5-stimulated VEGF release. On the other hand, chelerythrine, a specific PKC inhibitor, nearly abolished both basal and cytokine-induced VEGF secretion. Thus, PKC activity is essential for VEGF expression by eosinophils. Genistein and chelerythrine appeared to prevent VEGF transcription, since both substances inhibited the induction of VEGF mRNA by GM-CSF and IL-5. Thus, PTK and PKC activation may be involved in the expression of VEGF derived from activated eosinophils.
In summary, human eosinophils constitutively express mRNA encoding VEGF, and also store VEGF, probably in cytoplasmic granules. Moreover, eosinophils release VEGF protein product following stimulation with GM-CSF or IL-5, a process sensitive to inhibition by a glucocorticoid. PTK and PKC have essential roles in the intracellular signal pathways coupling eosinophil stimulation with VEGF expression. Increased numbers of blood and tissue eosinophils are regularly observed in patients with bronchial asthma. VEGF production by activated eosinophils at sites of allergic inflammation may contribute to vascular permeability and subsequent tissue edema, which are characteristic pathologic findings in these patients.
This work was supported by National Institutes of Health grants AI20241, DE08680, HL46563, and HL56386.
| 1. | Martin L. B., Kita H., Leiferman K. M., Gleich G. J.Eosinophils in allergy: roles in disease, degranulation, and cytokines. Int. Arch. Allergy Immunol.1091996207215 |
| 2. | Weller P. F.The immunology of eosinophils. N. Engl. J. Med.324199111101118 |
| 3. | Moqbel R., Levi-Schaffer F., Kay A. B.Cytokine generation by eosinophils. J. Allergy Clin. Immunol.94199411831188 |
| 4. | Dvorak H. F., Brown L. F., Detmar M., Dvorak A. M.Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am. J. Pathol.146199510291039 |
| 5. | Thomas K. A.Vascular endothelial growth factor, a potent and selective angiogenic agent. J. Biol. Chem.2711996603606 |
| 6. | Tischer E., Mitchell R., Hartman T., Silva M., Gospodarowicz D., Fiddes J. C., Abraham J. A.The human gene for vascular endothelial growth factor. Multiple protein forms are encoded through alternative exon splicing. J. Biol. Chem.26619911194711954 |
| 7. | Park J., Keller G.-A., Ferrara N.The vascular endothelial growth factor (VEGF) isoforms: differential deposition into the subepithelial extracellular matrix and bioactivity of extracellular matrix-bound VEGF. Mol. Biol. Cell4199313171326 |
| 8. | Leung D. W., Cachianes G., Kuang W.-J., Goeddel D. V., Ferrara N.Vascular endothelial growth factor is a secreted angiogenic mitogen. Science246198913061309 |
| 9. | Li J., Perrella M. A., Tsai J.-C., Yet S.-F., Hsieh C.-M., Yoshizumi M., Patterson C., Endege W. O., Zhou F., Lee M.-E.Induction of vascular endothelial growth factor gene expression by interleukin-1β in rat aortic smooth muscle cells. J. Biol. Chem.2701995308312 |
| 10. | Levy A. P., Levy N. S., Loscalzo J., Calderone A., Takahashi N., Yeo K.-T., Koren G., Colucci W. S., Goldberg M. A.Regulation of vascular endothelial growth factor in cardiac myocytes. Circ. Res.761995758766 |
| 11. | Frank S., Hübner G., Breier G., Longaker M. T., Greenhalgh D. G., Werner S.Regulation of vascular endothelial growth factor expression in cultured keratinocytes. J. Biol. Chem.27019951260712613 |
| 12. | Iijima K., Yoshikawa N., Connolly D. T., Nakamura H.Human mesangial cells and peripheral blood mononuclear cells produce vascular permeability factor. Kidney Int.441993959966 |
| 13. | Roger D. F., Evans T. W.Plasma exudation and oedema in asthma. Br. Med. Bull.481991120134 |
| 14. | Lee T.-C., Lenihan D. J., Malone B., Roddy L. L., Wasserman S. I.Increased biosynthesis of platelet-activating factor in activated human eosinophils. J. Biol. Chem.259198455265530 |
| 15. | Weller P. F., Lee C. W., Foster D. W., Corey E. J., Austen K. F., Lewis R. A.Generation and metabolism of 5-lipoxygenase pathway leukotrienes by human eosinophils: predominant production of leukotriene C4. Proc. Natl. Acad. Sci. USA80198376267630 |
| 16. | Smith H.Asthma, inflammation, eosinophils and bronchial hyperresponsiveness. Clin. Exp. Allergy221992187197 |
| 17. | Weller P. F.Cytokine regulation of eosinophil function. Clin. Immunol. Immunopathol.621992S55S59 |
| 18. | Simon H. U., Blaser K.Inhibition of programmed eosinophil death: a key pathologic event for eosinophilia? Immunol. Today1619955355 |
| 19. | Bentley A. M., Meng Q., Robinson D. S., Hamid Q., Kay A. B., Durham S. R.Increases in activated T lymphocytes, eosinophils, and cytokine mRNA expression for interleukin-5 and granulocyte/macrophage colony-stimulating factor in bronchial biopsies after allergen inhalation challenge in atopic asthmatics. Am. J. Respir. Cell Mol. Biol.819933542 |
| 20. | Walz T. M., Nishikawa B. K., Malm C., Briheim K., Wasteson A.Transforming growth factor alpha expression in normal human blood eosinophils: differential regulation by granulocyte-macrophage colony-stimulating factor and interleukin-3. Leukemia81994612619 |
| 21. | Brach M. A., Sott C., Kiehntopf M., Herrmann F.Expression of the transforming growth factor-alpha gene by human eosinophils is regulated by interleukin-3, interleukin-5, and granulocyte-macrophage colony-stimulating factor. Eur. J. Immunol.241994646650 |
| 22. | Moqbel R., Hamid Q., Ying S., Barkans J., Tsicopoulos A., Wardlaw A. J., Kay A. B.Expression of mRNA and immunoreactivity for the granulocyte/macrophage colony-stimulating factor in activated human eosinophils. J. Exp. Med.1741991749752 |
| 23. | Alam R., Forsythe P., Stafford S., Fukuda Y.Transforming growth factor β abrogates the effects of hematopoietins on eosinophils and induces their apoptosis. J. Exp. Med.179199410411045 |
| 24. | Nakajima H., Gleich G. J., Kita H.Constitutive production of IL-4 and IL-10 and stimulated production of IL-8 by normal peripheral blood eosinophils. J. Immunol.156199648594866 |
| 25. | Meagher L. C., Cousin J. M., Seckl J. R., Haslett C.Opposing effects of glucocorticoids on the rate of apoptosis in neutrophilic and eosinophilic granulocytes. J. Immunol.156199644224428 |
| 26. | Lim K. G., Wan H.-C., Bozza P. T., Resnick M. B., Wong D. T. W., Cruikshank W. W., Kornfeld H., Center D. M., Weller P. F.Human eosinophils elaborate the lymphocyte chemoattractants: IL-16 (lymphocyte chemoattractant factor) and RANTES. J. Immunol.156199625662570 |
| 27. | Stern M., Meagher L., Savill J., Haslett C.Apoptosis in human eosinophils: programmed cell death in the eosinophil leads to phagocytosis by macrophages and is modulated by IL-5. J. Immunol.148199235433549 |
| 28. | Siebert P. D., Larrick J. W.PCR MIMICS: competitive DNA fragments for use as internal standards in quantitative PCR. BioTechniques141993244249 |
| 29. | Zhao J., Araki N., Nishimoto S. K.Quantitation of matrix Gla protein mRNA by competitive polymerase chain reaction using glyceraldehyde-3-phosphate dehydrogenase as an internal control. Gene1551995159165 |
| 30. | Wong D. T. W., Weller P. F., Galli S. J., Elovic A., Rand T. H., Gallagher G. T., Chiang T., Chou M. Y., Matossian K., McBride J., Todd R.Human eosinophils express transforming growth factor-α. J. Exp. Med.1721990673681 |
| 31. | Wong D. T. W., Elovic A., Matossian K., Nagura N., McBride J., Chou M. Y., Gordon J. R., Rand T. H., Galli S. J., Weller P. F.Eosinophils from patients with blood eosinophilia express transforming growth factor β1. Blood78199127022707 |
| 32. | Powell P. P., Klagsburn M., Abraham J. A., Jones R. C.Eosinophils expressing heparin-binding EGF-like growth factor mRNA localize around lung microvessels in pulmonary hypertension. Am. J. Pathol.1431993784793 |
| 33. | Ohno I., Nitta Y., Yamauchi K., Hoshi H., Honma M., Woolley K., O'Byrne P., Dolovich J., Jordana M., Taura G., Tanno Y., Shirato K.Eosinophils as a potential source of platelet-derived growth factor B-chain (PDGF-B) in nasal polyposis and bronchial asthma. Am. J. Respir. Cell Mol. Biol.131995639647 |
| 34. | Clutterbuck E. J., Hirst E. M., Sanderson C. J.Human interleukin-5 (IL-5) regulates the production of eosinophils in human bone marrow cultures: comparison and interaction with IL-1, IL-3, IL-6, and GM-CSF. Blood73198915041512 |
| 35. | Sonoda Y., Arai N., Ogawa M.Humoral regulation of eosinophilopoiesis in vitro: analysis of the targets of interleukin-3, granulocyte/ macrophage colony-stimulating factor (GM-CSF), and interleukin-5. Leukemia319891418 |
| 36. | Beil W. J., Weller P. F., Tzizik D. M., Galli S. J., Dvorak A. M.Ultrastructural immunogold localization of tumor necrosis factor-α to the matrix compartment of eosinophil secondary granules in patients with idiopathic hypereosinophilic syndrome. J. Histochem. Cytochem.41199316111615 |
| 37. | Levi-Schaffer F., Lacy P., Severs N. J., Newman T. M., North J., Gompers B., Kay A. B., Moqbel R.Association of granulocyte-macrophage colony-stimulating factor with the crystalloid granules of human eosinophils. Blood85199525792586 |
| 38. | Moqbel R., Ying S., Barkans J., Newman T. M., Kimmitt P., Wakelin M., Taborda-Barata L., Meng Q., Corrigan C. J., Durham S. R., Kay A. B.Identification of messenger RNA for IL-4 in human eosinophils with granule localization and release of the translated product. J. Immunol155199549394947 |
| 39. | Houck K. A., Leung D. W., Rowland A. M., Winer J., Ferrara N.Dual regulation of vascular endothelial growth factor bioactivity by genetic and proteolytic mechanism. J. Biol. Chem.26719922603126037 |
| 40. | Schleimer R. P., Bochner B. S.The effects of glucocorticoids on human eosinophils. J. Allergy Clin. Immunol.94199412021213 |
| 41. | Wallen N., Kita H., Weiler D., Gleich G. J.Glucocorticoids inhibit cytokine-mediated eosinophil survival. J. Immunol.147199134903495 |
| 42. | Taniguchi T.Cytokine signaling through nonreceptor protein tyrosine kinases. Science2681995251255 |
| 43. | Bates M. E., Bertics P. J., Busse W. W.IL-5 activates a 45-kilodalton mitogen-activated protein (MAP) kinase and Jak-2 tyrosine kinase in human eosinophils. J. Immunol.1561996711718 |
