Eosinophils contain four predominant cationic molecules, including the eosinophil peroxidase (EPO), the major basic protein (MBP), the eosinophil cationic protein (ECP), and the eosinophil-derived neurotoxin (EDN). Eosinophil accumulation and the release of their granule protein contents in response to the proper mediators is a crucial factor in the pathophysiology of allergic inflammation.1-3

In the past decade or so, intensive research in the fields of neuropeptides and immune cells has resulted in accumulating evidence that supports the existence of a neuroimmune axis.4-6 However, the biological effects of vasoactive intestinal peptide (VIP) on human eosinophil are not fully characterized and it is not clear whether VIP possesses an anti-inflammatory or deleterious effect in hypersensitivity reactions.7, 8

The 28 amino acid VIP polypeptide, which exists in the parasympathetic nerves and to a lesser extent in the sensory fibers, is one of the most abundant of the neuropeptides found in the upper and lower airways9, 10 and has been detected in the fluids of hypersensitivity reactions.11 The biological effects of VIP on airway functions include regulation of vascular tone, airway caliber and mucus secretion.12-15 In addition to its direct physiologic effects on target tissues, VIP also has the potential to modulate T lymphocyte functions.16, 17 Although VIP affects T lymphocytes through its specific receptor, its effects on other immune and inflammatory cells is still obscure and there have not been any detailed reports of its effect on eosinophil chemotaxis or exocytosis.

Hemopoietic cytokines, such as IL-5, are known for being specific for human eosinophil differentiation and for prolonging cell survival in culture, and possess a strong priming effect on eosinophils, preparing the cell for an exaggerated inflammatory response to other inflammatory mediators.18, 19 Nonetheless, little is known about cytokineneuropeptide interactions in eosinophilic inflammation.

In the present study, we investigated the effects of VIP, in the absence or presence of interleukin (IL)-5, on human eosinophil chemotaxis and degranulation. Release of EDN was chosen as a degranulating marker for its relatively easy induction in vitro. Results are discussed in relation to the possible molecular pathway(s) involved.

Percoll solution and EDN radioimmunoassay (RIA) kits were purchased from Pharmacia (Uppsala, Sweden). NP-40 was from Iwai Kagaku Co., Ltd. (Tokyo, Japan). CD16 microbeads were from Miltenyi Biotec (Bergisch Gladbach, Germany). Cytochalasin B and VIP were from Sigma Chemical Co. (St Louis, MO, USA). The VIP receptor antagonist was from Peninsula Laboratories, Inc. Herbimycin A, staurosporine, BAPTA-AM and sodium orthovanadate were from Wako Pure Chemical Industries (Osaka, Japan). Recombinant human (rh)IL-5 was from Genzyme Corporation Chemicals (Cambridge, UK).

Eosinophils and lymphocytes from healthy volunteers were purified from venous blood anticoagulated with heparin by Percoll discontinuous gradient centrifugation followed by negative selection with CD16 immunomagnetic beads for high purification of eosinophils, as described elsewhere.20 Briefly, blood was sedimented with 6% dextran and the buffy coat was collected and washed twice with Hank's balanced salt solution (HBSS) containing 2% fetal calf serum (FCS) and cells were overlaid on a discontinuous Percoll gradient of 78%, 71% and 56% Percoll solution in 15 mL polystrene tubes. The tubes were then centrifuged at 400 g at 20°C for 30 min, resulting in the formation of three bands and a pellet. The pellet and the lowest band granulocytes were collected and sedimented red blood cells were removed by hypodense lysis. These lymphocyte- and monocyte-free granulocytes were incubated with CD16 immunomagnetic beads for 30 min at 6-12°C. CD16 is expressed on neutrophils, but not on normal resting eosinophils. Cells were applied to the top of a prewashed column MACS separator (Miltenyi Biotec). The CD16 negative cells were collected as 98% pure eosinophilic granulocytes as judged by Hinkelmann staining. Eosinophils showed a viability greater than 97%, which was confirmed by trypan blue dye exclusion. As for lymphocytes, cells were recovered from the upper band.

Chemotactic experiments were performed in 48-well microchemotaxis chambers as described precisely by us elsewhere.21 Results of the chemotaxis experiments were expressed as eosinophils/five selected high power fields (5 h.p.f.).

Eosinophil degranulation was induced by VIP or PAF using 96-well flat-bottom culture plates. Freshly purified eosinophils were washed twice in RPMI-1640 containing 10% FCS and were resuspended in the same culture medium at 5 × 10⁵ cells/mL. During preliminary studies, eosinophils adhering to culture plates released large amounts of EDN spontaneously, probably due to non-specific activation, which was inhibited by the addition of 5 µg/mL cytochalasin B to the assay system. Aliquots of cell suspension (100 µL) were added onto the culture plate and degranulation was initiated by adding 100 µL VIP or PAF diluted in the same culture medium as eosinophil and adjusted to the desired concentrations. After 1 h incubation at 37°C and 5% CO₂, supernatants from wells were collected carefully and stored at -20°C until assayed. Concentrations of EDN in the supernatants were analyzed by RIA using EDN RIA kits (Pharmacia). The total content of EDN was measured by lysing eosinophils with 1% Nonidet P-40.

Eosinophils were incubated with 1 ng/mL rhIL-5 for 30 min at 37°C. After this, cells were washed twice and resuspended in the culture medium and their chemotaxis and EDN degranulation by VIP were performed in the same way as described earlier.

Ca²⁺-depleted eosinophils were obtained by incubating 10⁷ cells/mL with 30 µmol/L of the calcium chelating agent BAPTA-AM in test medium (130 mmol/L NaCl, 5 mmol/L NaHCO₃, 4.6 mmol/L KCl, 5 mmol/L glucose, 2 mmol/L ethyleneglycol-bis(β-aminoethyl ether)-N,N, N',N'-tetraacetic acid (EGTA) and 20 mmol/L N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), for 30 min at 37°C.

Total cellular RNA of eosinophils and lymphocytes was extracted using the guanidium isothiocynate method; 1 µL total cellular RNA, 1 µL (500 ng) Oligo(dt) 12-18 primer supplied in diethyl pyrocarbonate (DEPC)-treated water and 3 µL sterile distilled water were heated to 70°C for 10 min followed by quick chilling on ice and were made up to a total volume of 19.3 µL containing first strand buffer, 10 mmol/L deoxyribonucleoside triphosphates (dNTP), 0.1 mol/L dithiothreitol (DTT), and superscript reverse transcriptase (GIBCO BRL). The samples were incubated for 10 min at 37°C and then 50 min at 42°C followed by 5 min at 95°C. Then 0.7 µL (2 U) RNase was added and further incubated for 20 min at 37°C.

Aliquots (1 µL) of reverse transcribed mixture were made up to 50 µL with reverse transcriptase (RT) buffer containing 12.5 pmol/µL specific primers and 1 U Taq DNA polymerase. The β-actin and VIP receptor 1 (R1) primers used were exactly similar as those reported by Eura et al. and Park et al., respectively.22, 23 The samples were overlaid with mineral oil and polymerase chain reaction (PCR) consisted of 1 min of denaturation at 94°C, 2 min of annealing at 60°C and 3 min of extension at 72°C for 40 cycles. The PCR products were analyzed on an ethidium bromide-stained 1.5% agarose gel.

Results were expressed as the mean ± SEM. Statistical significance was analyzed by paired Student's t-test. P < 0.05 was considered statistically significant. All experiments were performed in duplicate.

At a wide range of doses, VIP elicited eosinophil locomotion and EDN release as can be seen in Figure 1a,b. These effects were comparable with those induced by PAF (Fig. 1c,d). The EDN release showed a time-dependent mode and was not due to cytotoxicity, as judged by trypan blue dye exclusion, which was performed after each degranulation period to check cell viability (data not shown).

A checkerboard assay was performed to identify the nature of eosinophil locomotion in response to VIP. Whenever cells migrate in response to a concentration gradients of chemotactic molecules it is termed chemotaxis, whereas chemokinesis represent random migration without the requirement of a concentration gradient. Figure 2 shows that VIP activity was mainly chemotactic and to a lesser degree chemokinetic at lower concentrations; however, higher concentrations (10^-5 mol/L and above) demonstrated chemokinetic activity.

Figure 3 demonstrates the priming effect of 1 ng/mL rhIL-5 on eosinophil chemotaxis and EDN release induced by VIP. Recombinant human IL-5 at the above concentration did not possess any chemotactic or secretagogue activity against eosinophil (data not shown); however, when eosinophils were incubated first with rhIL-5, the cells showed significant sensitivity against VIP to chemotaxis and degranulation rather than chemokinetic activity.

Figure 4 illustrates the eosinophilotactic and secretagogue activities of VIP against Ca²⁺-depleted cells. The quantity of 10^-6 mol/L VIP was chosen because it is the common dose to induce significantly both functions.

Contrary to EDN release, which was almost completely blocked, eosinophil chemotaxis was not affected under depletion conditions.

Pre-exposure of eosinophils to 100 nmol/L staurosporine prior to VIP induction of eosinophil chemotaxis or degranulation resulted in a dose-dependent inhibition of EDN release, but did not affect cell migration (Table 1).

Recently, several reports assigned a role for tyrosine kinases in leukocyte migration, therefore we next studied their role on VIP-induced eosinophil chemotaxis using herbimycin A, the potent tyrosine kinase inhibitor.24, 25 Cells were incubated with various doses of the inhibitor for 1 h, washed and resuspended in the culture medium and then their chemotactic responses were tested against VIP in the same way as described in Methods. As shown in Figure 5a, pretreatment of eosinophils with herbimycin A resulted in a dose-dependent potentiation of VIP-induced eosinophilotactic response.

To further gain insight into the role of tyrosine phosphorylation in VIP-induced chemotaxis, eosinophils were treated with pervanadate, the tyrosine phosphatase inhibitor.26 Pervanadate treatment resulted in a dose-dependent inhibition of VIP-induced chemotaxis, as illustrated in Figure 5b. As a control, a comparative study on PAF was also performed; however, neither herbimycin A nor pervanadate modulated PAF eosinophilotactic activity (data not shown).

To test whether VIP interaction with human eosinophils is receptor mediated, eosinophils were pretreated with various concentrations of VIP receptor antagonist, [4 Cl-D-Phe⁶-Leu¹⁷] VIP,27 for 30 min before induction of chemotaxis and degranulation by VIP. None of the doses tested, from 10^-5 to 10^-7 mol/L, modified the activation of the cells by the corresponding agonist (Fig. 6).

Pertussis toxin (PTX) catalyzes ADP-ribophosphorylation of G_i proteins. To examine the role of G_i proteins in VIP-induced eosinophil chemotaxis, eosinophils were pre-incubated with various concentrations of PTX, which per se did not affect VIP-induced chemotaxis (Table 2).

The amplification products of VIP-R1 were identical to the predicted size for the mRNA template of 534 b.p. The RT-PCR products from lymphocytes, but not eosinophils, expressed signals for VIP-R1 (Fig. 7).

It seems certain that VIP induces eosinophil functional responses from the data presented in Figure 1, but the doseresponse curves are different from those induced by PAF, in which a specific cell-surface receptor has been identified. Although VIP modulates lymphocyte functions through specific lymphocyte receptors, which are typically seven transmembrane-coupled receptors, the VIP dose response of eosinophils is in favor of non-specific binding activation. This is supported by: (i) lack of modulation of VIP-induced chemotaxis and degranulation by VIP receptor antagonist (4 Cl- D-Phe⁶-Leu¹⁷) VIP; (ii) lack of blocking activity of PTX in VIP-induced eosinophil chemotaxis; and (iii) lack of VIP-R1 expression by human eosinophils when compared to human lymphocytes. It is possible that VIP non-specific binding to human eosinophils may be due to either a non-specific charge-related binding or a lipophilic property. It is of note that other peptides, such as melittin and substance P, have been suggested to activate guinea-pig eosinophils through non-specific peptidemembrane phospholipid interaction; moreover, the MBP stimulates several cell types through negatively charged phospholipid binding sites.28 Nonetheless, our failure to detect any signals for VIP-R1 mRNA in the gels does not solely exclude the possibility of expression of very low non-detectable amounts in the gel. In fact, only a few hundred protein molecules of G-protein-linked receptors are sufficient to mediate functional responses. In addition, the possibility of the existence of other receptors should be considered. The exact binding mechanism of VIP with human eosinophils remains of interest and requires further investigations.

Staurosporine and Ca²⁺ chelation did not modify VIP eosinophilotactic activity, which accordingly can exclude involvement of other pathways in VIP-induced chemotaxis, such as phospholipase A and phospholipase D, which are Ca²⁺ dependent.29, 30 However, modulation of tyrosine kinases and phosphatases by herbimycin A and pervanadate augmented and inhibited eosinophil chemotaxis, respectively. These interesting results suggest that the non-specific activation of human eosinophils by VIP seems to have two pathways: (i) a Ca²⁺-dependent pathway involving EDN release; and (ii) a Ca²⁺-independent tyrosine kinase and phosphatase pathway involving chemotaxis.

In vivo, allergic reactions result in the release of several mediators, many of which work in synergy, a phenomenon termed priming with respect to the in vitro studies. Therefore, we next investigated the priming activity of rhIL-5 and whether it cooperates with VIP in inducing significant eosinophil functional responses. Recombinant human IL-5 at 1 ng/mL did not induce any eosinophil chemotaxis or degranulation, but prepared the cell for an exaggerated response to subsequent stimulation with VIP, demonstrating a novel cytokineneuropeptide interaction in inducing eosinophil functional responses. It is possible that rhIL-5 causes a change in the signal transduction pathway(s).

In conclusion, our findings suggest that VIP released in allergic reactions may act as a pro-inflammatory mediator, which exacerbates eosinophilic inflammation, rather than an anti-inflammatory mediator as a result of increase in cAMP concentration in other inflammatory cells, and this effect cannot be modulated by a receptor-specific antagonist.

REFERENCES

1
Motojima S, Frigas E, Leogering DA, Gleich GJ. Toxicity of eosinophil cationic proteins for guinea pig tracheal epithelium in vitro. Am. Rev. Respir. Dis. 1989; 139: 801-805.
 

2
Ayras GH, Altman LC, MacManus MM et al. Injurious effect of the eosinophil peroxide-hydrogen peroxide-halide system and major basic protein on human nasal epithelium in vitro. Am. Rev. Respir. Dis. 1989; 140: 125-131.

3
Agosti JM, Altman LC, Ayras GH, Loegering DA, Gleich GJ, Klebanoff SJ. The injurious effect of eosinophil peroxide peroxidase, hydrogen peroxide, and halides on pneumocytes in vitro. J. Allergy Clin. Immunol. 1987; 79: 496-504.
 

4
El-Shazly AE, Masuyama K, Eura M, Ishikawa T. Immunoregulatory effect of substance P in human eosinophil migratory function. Immunol. Invest. 1996; 25: 191-201.
 

5
El-Shazly AE, Masuyama K, Ishikawa T. Mechanisms involved in activation of human eosinophil exocytosis by substance P: An in vitro model of sensory neuroimmunomodulation. Immunol. Invest. 1997; 26: 615-629.
 

6
Numao T, Agrawal DK. Neuropeptides modulate human eosinophil chemotaxis. J. Immunol. 1992; 149: 3309-3315.
 

7
Uden BJ, Dick EC, Buckner CK. Inhibition by vasoactive intestinal peptide of antigen-induced histamine release from guinea pig minced lung. Eur. J. Pharmacol. 1983; 88: 247-250.

8
Said SI. Neuropeptides as modulators of injury and inflammation. Life Sci. 1990; 47: PL19-PL21.
 

9
Ghatei MA, Sheppard M, O'Shaughnessy DJ et al. Regulatory peptides in the mammalian respiratory tract. Endocrinology 1982; 111: 1248-1254.
 

10
Baraniuk JN, Okayama M, Lundgren JD et al. Vasoactive intestinal peptide in human nasal mucosa. J. Clin. Invest. 1990; 86: 825-831.
 

11
Mosimann BL, White MV, Hohman RJ, Goldrich MS, Kaulbach HC, Kaliner MA. Substance P, calcitonin gene-related peptide, and vasoactive intestinal peptide increase in nasal secretions after allergen challenge in atopic patients. J. Allergy Clin. Immunol. 1993; 92: 95-104.
 

12
Said SI. Vasoactive peptides in the lung, with special reference to vasoactive intestinal peptide. Exp. Lung Res. 1982; 3: 343-348.
 

13
Palmer JBD, Cuss FMC, Barnes PJ. VIP and PHM and their role in nonadrenergic inhibitory responses in isolated human airway. J. Appl. Physiol. 1986; 61: 1322-1328.
 

14
Peatfield AC, Barnes PJ, Bratcher C, Nadel JA, Davis B. Vasoactive intestinal peptide stimulates tracheal submucosal gland secretion in ferret. Am. Rev. Respir. Dis. 1983; 128: 89-93.
 

15
Stjarne P, Lacroix JS, Anggard A, Lundberg JM. Compartment analysis of vascular effects of neuropeptides and capsaicin in the pig nasal mucosa. Acta. Physiol. Scand. 1991; 141: 335-342.
 

16
Delgado M, de la Fuente M, Martinez C, Gomariz RP. Pituitary abenylate cyclase-activating polypeptides inhibit the mobility of murine thymocytes and splenic lymphocytes: Comparison with and implication of cAMP. J. Neuroimmunol. 1995; 62: 137-146.
 

17
O'Dorisio MS, Shannaon BT, Fleshman DJ, Campolito LB. Identification of high affinity receptors for vasoactive intestinal peptide on human lymphocytes of B cell lineage. J. Immunol. 1989; 142: 3533-3536.
 

18
Lopez AF, Sanderson CJ, Gamble JR, Campell HD, Ypung IG, Vadas MA. Recombinant human interleukin 5 is a selective activator of human eosinophil function. J. Exp. Med. 1988; 167: 219-224.
 

19
Hakansson L, Vengue P. Priming of eosinophil and neutrophil migratory responses by interleukin 3 and interleukin 5. APMIS 1994; 102: 308-316.
 

20
Hansel TT, de Vries IJM, Iff T et al. An improved immunomagnetic procedure for the isolation of highly purified human blood eosinophils. J. Immunol. Methods 1991; 145: 105-110.
 

21
El-Shazly AE, Masuyama K, Samejima Y, Eura M, Ishikawa T. Inhibition of human eosinophil chemotaxis in vitro by the anti-allergic agent emedastine difumarate. Immunopharmacol. Immunotoxicol. 1996; 18: 587-595.
 

22
Eura M, Ogi K, Chikamatsu K et al. Expression of the mage gene family in human head and neck squamous cell carcinomas. Int. J. Cancer 1995; 64: 304-308.
 

23
Park SK, Olsen TA, Ercal N, Summers M, O'Dorisio MS. Characterization of vasoactive intestinal peptide receptors on human megakaryocytes and platelets. Blood 1996; 87: 4629-4635.
 

24
Zheng ZM, Specter S. Dynamic production of tumour necrosis factor-alpha messenger RNA, intracellular and extracellular TNF-alpha by murine macrophages and possible association with tyrosine phosphorylation of STAT 1 alpha and ERK2 as an early signal. Immunology 1996; 87: 544-550.
 

25
Yasui K, Yamazaki M, Miyabayashi M, Tsuno T, Komiyama A. Signal transduction pathway in human polymorphonuclear leukocytes for chemotaxis induced by a chemotactic factor. J. Immunol. 1994; 152: 5922-5929.
 

26
Heffetz D, Bushkin I, Dror R, Zick Y. The insulinomimetic agents H₂O₂ and vanadate stimulate protein tyrosine phosphorylation in intact cells. J. Biol. Chem. 1990; 265: 2896-2906.
 

27
Pandol SJ, Kiertisin D, Mari SS, Walie V, Jean R. Vasoactive intestinal peptide receptor antagonist (4 Cl-D-Phe-Leu) VIP. Am. J. Physiol. 1986; 250: G553-G557.
 

28
Kita H, Abu-Ghazaleh RI, Gleich GJ, Abraham RT. Role of pertussis toxin-sensitive G proteins in stimulus-dependent human eosinophil degranulation. J. Immunol. 1991; 147: 3466-3473.
 

29
Kessel RGC, Roos D, Verhoeven AJ. fMet-Leu-Phe-induced activation of phospholipase D in human neutrophils. J. Biol. Chem. 1991; 266: 23152-23156.

30
Clark JD, Lin LL, Kriz RW et al. A novel arachidonic acid-selective cytosolic PLA₂ contains a Ca²⁺-dependent translation domain with homology to PKC and GAP. Cell 1991; 65: 1043-1051.