Monocytes Activate Eosinophils for Enhanced Helminthotoxicity And

MONOCYTES ACTIVATE EOSINOPHILS FOR ENHANCED HELMINTHOTOXICITY

AND INCREASED GENERATION OF LEUKOTRIENE C4

by P. Eisas (1), T.H. Lee (2), H.L. Lenzi (3), and A.J. Dessein (1)

(1) Centre d'Immunologie INSERM-CNRS de Marseille-Luminy, Case 906, 13288 Marseille Cedex 9 (France) (2) Guy's Hospital, London SEI 9RT (UK), and (3) Depto de Patologia, Fiocruz, Rio de Janeiro (Brazil)

SUMMARY

Recent observations have shown that eosinophils are activated in certain clinical conditions and that activation may enhance the role of eosinophils in immune protection against helminth parasites and in the pathogenesis of certain diseases associated with high eosinophilia. Our laboratory has attempted to identify the immunological mechanisms causing such an activation. The data summarized here show that eosinophils can be activated in vitro with supernatants of resting or stimulated monocytes. The supernatants enhance eosinophil helminthotoxicity by increasing cell degranulation; they also enhance the generation of leukotrienes in eosinophils by exerting a per• missive effect on an early step of arachidonic acid metabolism. Biochemical analysis of the enhancing activities suggests that they are car• ried by a unique molecule or a unique set of molecules whose biochemical and functional properties are different from those of previously described monokines such as IL-l, IFN-o:,~, CSF and TNF. Studies on individuals with chronic schistosomiasis suggest that such regulatory interactions between eosinophils and macrophages may take place in the hepatic granulomatous reactions in patients with hepatosplenic schistosomiasis.

KEY-WORDS: Monocyte, Eosinophilia, Leukotriene, Allergy, Helmin• thotoxicity; Immunoregulation, Schistosomiasis.

Received November 24, 1986. 98 P. ELSAS AND COLL.

Introduction.

The eosinophil leukocyte is located in the connective tissues under the epithelial layer of the skin, bronchi, gastrointestinal tract and the wall of the uterus [1]. In healthy individuals, only a small number of eosinophils, estimated at 11400 of the total eosinophil pool of the organism, circulate in the blood, where they remain for a short (3 to 8 h) period before entering the tissues. Most infections cause an increase in neutrophil number without increasing blood and tissue eosinophilia. Some bacterial infections cause eosinopenia (reviewed in [2]). Blood and tissue eosinophilia are markedly augmented in individuals with allergic diseases or with helminth infections [3]. In certain of these subjects, eosinophiIs represent up to 90 010 of the blood white cells and may be observed in large numbers in certain organs, i.e. in the bronchi of certain individuals undergoing allergic crisis or in parasitized tissue [2, 4]. The association between atopic diseases, helminthiasis and high eosinophilia have suggested that eosinophils take part in the regulation of allergic reac• tions, in particular in hypersensitivity type I reactions and in immune protec• tion against helminth parasites [2]. Eosinophils secrete a variety of mediators which might affect the development of the inflammation. The most impor• tant of these mediators are probably the 5-lipoxygenase metabolites of arachidonic acid: 5-hydroxyeicosatetraenoic acid (5-HETE) and leukotriene C4 (LTC4) [5-9]. 5-HETE is a potent chemotactic factor for neutrophils, LTC4 is a smooth muscle constrictor and a vasospastic and venopermeabili• ty factor [10-12]; LTC4, with its cysteinylglycyl and cysteinyl analogues LTD4 and LTE4, constitutes the activity previously termed slow-reacting substance of anaphylaxis [13]. A critical role of eosinophils in protective immunity against parasites is suggested by numerous studies: eosinophils are strongly helminthotoxic in vitro [14] and can be found attached and degranulating on disintegrated helminth larvae in tissues of immune-infected hosts [15]. Several eosinophil degranulation products such as the major basic protein (MBP), eosinophil cationic protein (ECP) and eosinophil peroxidase are highly toxic for the parasite, either because they dismember parasite tegument [16] or because they generate oxygen radicals [17].

AA arachidonic acid. HPLC high-performance liquid chroma• Ab antibody. tography. ADCC antibody-dependent cellular cytotoxicity. IFN interferon. C complement. IL-l interleukin-l. CSF colony-stimulating factor. LT leukotriene. EAF eosinophil-activating factor. MBP major basic protein. ECEF eosinophil cytotoxicity-enhancing MEM minimal essential medium. factor. MK monokine. ECP eosinophil cationic protein. NDGA nordihydroguaiaretic acid. EPO eosinophil peroxidase. PBMC peripheral blood mononuclear cell. HES hypereosinophilic syndrome. RP-HPLC reverse phase HPLC. HETE 5-hydroxyeicosatetraenoic acid. TNF tumour necrosis factor. MONOCYTES, EOSINOPHILS, HELMINTHS AND LTC4 99

Since eosinophils appear to be a major component of the host's response to helminth infections, factors modulating their helminthotoxic capacity have received much attention. Lymphokines secreted by cells from Schistosoma mansoni egg granulomas [18] and mast-cell-derived mediators, such as the eosinophil-chemotactic factor of anaphylaxis [19, 20, 21], were shown to poten• tiate the helminthotoxicity of eosinophils in vitro. Early reports have shown that eosinophils, in certain hypereosinophilic states such as in the hypereosinophilic syndrome (HES) or in tropical eosinophilia, appear sometimes degranulated and vacuolated [22, 23]. The proteins, normally stored in eosinophil granules, were detected in abnormal quantities in the blood of certain of these patients [24]. Since these substances are toxic for mammalian cells, it has been suggested that abnormal degranula• tion of the eosinophils in hypereosinophilic states may cause or aggravate organ damage associated with certain hypereosinophilia. More recently, eosinophils have also been implicated in the pathogenesis of certain asthmatic conditions. The mechanism(s) causing degranulation of eosinophils in hypereosino• philia is unknown. However, recent observations suggest that it may result from the activation of the cells: eosinophils from individuals with eosinophilia from various aetiological origins exhibit enhanced helminthotoxicity [25], low surface charge, augmented metabolism and high lysosomal enzymatic activities [26]. Finally, eosinophils from hypereosinophilic subjects could be fractionated on density gradients into two fractions, eosinophils in the less dense fraction demonstrated enhanced helminthotoxicity and altered oxidative metabolism, and they may represent an activated cell population [27]. The association between eosinophilia and eosinophil activation has sug• gested that eosinopoietic factors may have the dual function of stimulating stem cells and activating mature blood eosinophils. This hypothesis has received some support from the demonstration that semi-purified colony-stimulating factor (CFS-ex) of placental origin markedly enhanced the helminthotoxicity of blood eosinophils, probably by enhancing cell degranulation [28]. These observations have been recently confirmed and extended in experiments per• formed with cloned recombinant CSF-ex. During the course of the CSF studies, we observed that supernatants of resting PBMC also markedly enhanced eosinophil cytotoxicity. Further analysis of this phenomenon showed that it was caused by a monokine which exerts profound regulatory effects on several eosinophil functions. These studies and their possible biological implications are discussed here.

Resting human blood mononuclear cells produce, in culture, factors which enhance ADCC by human blood eosinophils.

Human peripheral blood mononuclear cells (PBMC) purified by centrifuga• tion of heparinized blood on «Ficoll Hypaque» were cultured (2 x 106 cells/ml) in serum-free defined medium; the supernatants were collected 24 h 100 P. ELSAS AND COLL. later and added to human eosinophils (5 III of conditioned medium for 106 eosinophils) which had been purified (> 85 0,10 pure; contaminating cells were neutrophils) from the blood of individuals with 2 to 15 % eosinophilia. After a short incubation period (5 to 60 min) in the conditioned medium, eosinophils were tested in an ADCC reaction on S. mansoni larvae in the presence of heat-inactivated human antischistosomular antibodies. At low Ab concen• trations, control medium-treated eosinophils only adhered in small numbers to the larvae (fig. 1) and killed a small fraction (2 to 10 %) of schistosomula (fig. 2). In contrast, eosinophils treated with the conditioned media adhered in large numbers to schistosomula and demonstrated high helminthotoxicity (40 to 80 %). PBMC supernatants were not toxic for the larvae, as assessed by microscopic examination and maturation of worms when injected in

A B

o o

o e

FIG. 1. - Enhancement oj eosinophil adherence to Ab-coated schistosomula by blood mononuclear eel/-conditioned medium. Eosinophils purified on metrizamide gradients were incubated for 30 min with a 1/5 dilution of control medium (A) or mononuclear cell-conditioned medium (B) and added to Ab• coated schistosomula. Pictures were taken 90 min later. The conditioned medium was prepared by culturing 2 x 106 human blood mononuclear cells/ml of serum-free medium for 24 h. MONOCYTES, EOSINOPHILS, HELMINTHS AND LTC4 101 mice [29, 30]. The factor(s) causing enhancement of eosinophil cytotoxicity was termed ECEF (eosinophil cytotoxicity-enhancing factor). Control eosinophils neither adhered to nor killed schistosomula in the absence of antibodies. However, eosinophils treated with ECEF adhered to the larvae without Ab [31]. The number of adhering cells was low, however, as com• pared to cells reacted with Ab-coated larvae; nevertheless, this observation suggested that ECEF changed eosinophil surface properties, thus modifying their adherence capabilities. The effect of ECEF on eosinophils from normal individuals and from eosinophilic individuals with atopy, helminth infections, tropical or idiopathic eosinophilia is shown in [31]. Eosinophils were collected on the 23-24 % metrizamide gradient layer. Less dense eosinophils which sedimented on the 22 and 21 070 metrizamide gradient layer were contaminated with neutrophils and/or mononuclear cells and were not included in the study. Mononuclear cell supernatants enhanced the cytotoxicity of eosinophils from almost all normal and eosinophilic individuals. Though few preparations failed to demonstrate enhanced cytotoxicity after incubation in the conditioned medium, this property could not be ascribed to any particular clinical group of individuals.

ECEF production in chronic schistosomiasis.

Chronic infections by Schistosoma mansoni present a spectrum of clinical manifestations which are related to the intensity of infection. On one extreme, light infections have no severe clinical manifestations despite chronic egg

CmdltlOned medium 80 -

FIG. 2. - Enhancement of the Ab-dependent eosinophil-mediated killing of schistosomula by blood mononuclear-cell-conditioned medium. Same legend as in figure I. Larval mortality was scored after 24 h of incubation of the larvae with the eosinophils. 102 P. ELSAS AND COLL. elimination in the faeces. On the other hand, heavily infected individuals develop progressive disease with involvement of the liver and spleen (hepatosplenomegaly), characterized by severe inflammatory reactions of the periportal areas leading to scarring, fibrosis and portal hypertension. These hepatic inflammatory reactions are infiltrated by large numbers of macrophages and eosinophils and usually undergo modulation as the infec• tion becomes more chronic. Modulation is mediated by suppressor T lym• phocytes and is accompanied by a diminution of cell-mediated immunity to schistosome antigen (reviewed in [32]). It has been suggested that the pro• gressive tissue destruction characteristic of hepatosplenomegaly may result from a failure to modulate the hepatic inflammatory reactions caused by S. mansoni eggs and to limit the harmful side-effects of products released by activated macrophages and eosinophils [32]. Initial studies in a hyperendemic area for S. mansoni showed decreased production of ECEF in PBMC cultures from most S. mansoni-infected in• dividuals relative to healthy controls and to subjects harbouring other helminth parasites. Only 27 and 23 070 of PBMC cultures from S. mansoni individuals in two independent studies produced detectable ECEF activity, whereas PBMC cultures of 84 % of healthy controls produced ECEF. Low ECEF produc• tion in the infected subjects could not be attributed to the culture conditions or to the presence of an ECEF inhibitor or a substance interfering with eosinophil activation. ECEF production was studied in patients with different clinical status and egg excretion: PBMC from individuals with light or medium infections pro• duced less ECEF than those of healthy controls; in contrast, PBMC from patients with heavy infections (> 500 eggs/gm of faeces), some of them with developing hepatosplenomegaly, produced ECEF at high levels. This suggested that regulatory mechanisms limited the production of ECEF in light infections and that these controls were no longer operative in individuals with severe hepatic inflammation. These studies also revealed a possible association between ECEF and the inflammatory process, which is consis• tent with the effect of ECEF on the generation of 5-HETE and LTC4 by eosinophils, as analysed below.

Origin and biochemical properties of ECEF.

Conditioned media prepared with plastic adherent and plastic non-adherent blood mononuclear cells contained enhancing activity [33, 34]. The activity was also produced by non-adherent mononuclear cells depleted of T lym• phocytes by rosetting with sheep erythrocytes. Finally, PBMC and plastic adherent cells depleted of T and B lymphocytes by treatment with anti-Leu-l and anti-Ig Ab + C conserved their ability to produce the enhancing activity [33]. These results suggested that ECEF was produced by monocytes. This conclusion was supported by the observation that the human macrophage- MONOCYTES, EOSINOPHILS, HELMINTHS AND LTC4 103 like histyocytic lymphoma U937 and the promyelocytic leukaemia bipoten• tial line HL-60 produced ECEF-like factor(s) [33]. The enhancing activity in PBMC and adherent mononuclear cell super• natants was shown to be destroyed by trypsin treatment and to be stable at 70°C for 1 h. Sephadex-G100 gel chromatography and isoelectrofocussing of the mononuclear cell supernatants showed that the enhancing activity was associated with molecules having an apparent MW of 14,000 to 60,000 daltons and PI of 3.8, 4.2, 4.5 and 4.8.

Mechanism of the enhancing effect.

The bridging of eosinophil Fc receptors by antibodies bound to schisto• somula causes eosinophils to secrete the contents of their granules, including the basic proteins MBP and ECP and EPa on the parasite tegument. Tegu• mental damage is caused by oxygen radicals, by MBP and probably ECP. ECEF has no effect on eosinophil helminthotoxicity in anaerobic conditions [34]. This shows that oxidative metabolism is required for the enhancement of eosinophil cytotoxicity and led us to test the effect of ECEF on the pro• duction of oxygen radicals by eosinophils [34]; ECEF neither enhanced superoxide anion production nor increased iodination by resting or stimulated eosinophils. In contrast, CSF-cx, which also enhanced eosinophil helmintho• toxicity [28], caused a marked increase in the generation of superoxide anions and iodination by eosinophils [34]. These observations suggested that the enhancement of eosinophil cytotoxicity by mononuclear cells was not primarily due to a stimulation of eosinophil oxidative burst and to the generation of reactive oxygen radicals. The observation that ECEF-treated eosinophils, unlike control cells, were capable of adhering to schistosomula in the absence of antibodies [31,35] suggested that ECEF may enhance cell degranulation and that ECEF-treated eosinophils adhered to degranulation products on the larvae. These products have been shown to act as a « glue» between the eosinophil plasma membrane and the larval tegument [36]. To assess the effect of ECEF on cell degranulation, the extent of eosinophil discharge on Ab-coated schistosomula was measured by light microscopy after labelling EPa pro• ducts, or by electron microscopy, by estimating the larval surface area covered by electron-dense material of eosinophil origin. Both measurements indicated an increased release of MBP and EPa on the larval tegument by ECF-treated eosinophils, suggesting that the enhanced cytotoxicity is the consequence of an enhanced ability to degranulate [35]. These results, however, did not fully explain the strict oxygen dependency of the enhancement of cytotoxicity caused by ECEF [34]. Since cell degranulation involves important remodeling of the cell membrane and rapid turnover of membrane lipids, and since previous studies have also shown that oxygenation products of arachidonic acid stimulated degranulation of neutrophils [37, 38], we decided it would be useful to evaluate the possible role of arachidonic acid oxygenation metabolites in the activation of eosinophils by ECEF. We tested the effect of inhibitors 104 P. ELSAS AND COLL. of 5-lipoxygenase (NDGA) or of cyclooxygenase (Indomethacin) on the enhancement of cytotoxicity by monocyte supernatants (fig. 3). Indomethacin had no detectable effect on the enhancement of cytotoxicity. NDGA prevented it when added to eosinophils at the beginning of incubation with ECEF. NDGA, however, did not reduce the helminthotoxicity of control eosinophils, showing that it was not toxic to the cells and that it did not inhibit the killing mechanism itself. These observations, although they should be inter• preted with caution since NDGA could affect other cellular oxygenation reactions, suggest that the enhancement of eosinophil cytotoxicity caused by monokines requires normal functioning of the 5-lipoxygenase pathway. This suggested that the monokine may regulate the 5-lipoxygenase pathway in eosinophils.

~ ~ 580 EO.s+MK o~ + Indomethacin G60 ol.I.. ~ 40 w EOS+ MK +NDGA L t3 20 z « I 0 zw ~ 3 6 125 25 50 DRUGS (}JM)

FIG. 3. - Effect of NGDA and indomethacin on the enhancement of eosinophil helminthotoxicity by blood mononuclear-cell-conditioned medium.

Effect of eosinophil-activating monokines on the generation of arachidonic acid metabolites.

Arachidonic acid, released from membrane phospholipids after cell activa• tion, is metabolized by the cyclooxygenase pathway to prostaglandins and thromboxane and/or by the 5-lipoxygenase pathway to leukotrienes, accor• ding to the cell type. 5-lipoxygenation of arachidonic acid, which occurs MONOCYTES, EOSINOPHILS, HELMINTHS AND LTC4 105 in each type of human granulocyte, produces 5(S)-hydroperoxy-6-trans• 8, 11, 14-cis-eicosatetraenoic acid; a portion of this compound is reduced to 5-HETE and the remainder is enzymatically converted to an unstable in• termediate, 5,6-oxido-7,9-trans-ll ,14-cis-eicosatetraenoic acid (leukotriene A4 , LTA4). LTA4 is processed in eosinophils into one major product, 5(S)• hydroxy-6(R)-S-glutathionyl-7,9-trans-ll,14-cis- eicosatetraenoic acid (LTC4) by a glutathionyl-S-transferase termed LTC4 synthetase. In the neutrophil, LTA4 is converted predominantly to 5(S),12(R)-dihydroxy-6,14-cis-8,10-trans• eicosatetraenoic acid (LTB4) by an epoxide hydrolase [5-9].

The effect of PBMC culture supernatants on the release of LTC4 by eosinophils activated with the calcium ionophore is shown in figure 4. When the supernatants and the calcium ionophore were added simultaneously to eosinophils, no effect of the supernatants on LTC4 production was observed. In contrast, preincubation of eosinophils with the supernatants for a few minutes was sufficient to cause a marked increase in LTC4 generation by CA2 + ionophore-activated eosinophils. Maximal enhancement was observed with eosinophils pre-incubated in the PBMC supernatants for 3 to 5 min. This suggested that no protein synthesis was required for enhancement. The kinetics of LTC4 generation by control and PBMC supernatant-treated eosinophils 2 which were activated with the Ca + ionophore is shown in figure 5: LTC4 production by supernatant-treated cells was accelerated as compared to con-

z 0 ~ u 2000 ~ -/r- 0 0 B: 1000 u~ ~ t 200 "6 ~ z w 100 T ~ w u z 50 ~

FIG. 4. - Time-dependent enhancing effects ofECEF-containing supernatants on subsequent LTC4 production by Ca ++ -ionophore-activated eosinophils in 3 separate experiments.

Symbols represent different eosinophil donors. The data represent (1,10 enhancement of LTC4 generation relative to controls incubated in culture medium containing LPS for the indica• ted periods of time. 106 P. ELSAS AND COLL. troIs and reached higher maximal levels. The enhancing effect was observed at all Ca2 + ionophore concentrations [39] ; it was of larger magnitude when the cells were activated with concentrations (1 to 2.5 flM) of calcium ionophore which were suboptimal for the generation of LTC4 by untreated eosinophils.

Enhancement of LTC4 levels in eosinophil supernatants resulted from increased synthesis and not from enhanced release of LTC4 from the cells, since cell-associated LTC4 (measured in methanol extracts of experimental and control eosinophil pellets) represented a minor fraction of total LTC4 (in pellets + supernatants) and did not significantly differ between experimen• tal and control groups.

PBMCsup • + o -

(/) -' I Q. ~ 10

(/) o w '"12 g- 50

~ U f• -'

o 20 40 60 TI ME AFTER IONOPHORE ADDIT ION

FIG. 5. - Effect ofpreincubation of eosinophils for 2.5 min with a 1/3 dilution of ECEF-containing PBMC supernatants (e) or control medium containing LPS (0) on the time course ofleukotriene generation after addition ofCa+ + ionophore (2.5 IJ-M final). The data are from one representative experiment. Each point represents the mean of duplicate determinations on a single experimental sample.

To identify the source of the enhancing activity, PBMC were fractionated according to their adherence properties and surface membrane markers (Ig and Leu-l antigen). The fractionated cells were tested for the production of the enhancing activity in vitro (fig. 6). The enhancing moieties were produc• ed by resting or LPS-stimulated adherent mononuclear cells which were negative for surface immunoglobulins and for the Leu-l antigen. These MONOCYTES, EOSINOPHILS, HELMINTHS AND LTC4 107

N 0• 0• x x W W

12,5

~ 1) I + I I + I + I ~. « I + I ---l S!' 15 Il U~~I u u $

FIG. 6. - Identification of the cell source of the activity enhancing LTC4 production in eosinophils. Total or fractionated PBMC from 2 normal individuals were treated with (+) or without (-) anti-Ig and anti-Leu-l antibodies (Ab) and complement (C) and were subsequently cultured at a density of 2 x 106 cells/ml in MEM (Eagle's) with 5 IJ.g/ml LPS for 24 h. Super• natants from these cultures were tested on eosinophils from 2 separate donors for enhance• ment of LTC4 production after ionophore stimulation. Dashed lines indicate LTC4 generation by control eosinophils treated with medium + LPS. Supernatants from PBMC cultured in the absence of LPS did not significantly increase LTC4 production above this level.

findings strongly suggested that the enhancing activity was produced by monocytes.

The enhancing effect of various monokine (MK) preparations on LTC4 generation by eosinophils purified from the blood of 11 different donors is shown in figure 7. MK enhanced LTC4 production in all but one eosinophil preparation, and the magnitude of the enhancement was inversely correlated 108 P. ELSAS AND COLL.

z Q A ~ 8 UJz 120 t!5 3000 UJz 80 UJ Vi' a: r= -0.79 ~ p< 0.01 ~~ 40 0 ~ LL 01 20 c 0 1000 I- ~ z u 15 UJ 800 I- ~ • ---.J W 600 10 uz • « I 400 5 z UJ 200 0 ;f? _MK +MK 0 10 20 6 30 LTC (ng'10cells) 4

FIG. 7. - Effect on LTC4 generation of pre-incubation of eosinophils from 11 different donors with monocyte culture supernatants (e) and control medium containing LPS (0).

Panel A, control (-MK) and enhanced (+ MK) LTC4 production for each one of the 11 individual experiments. Panel B, correlation of the percent enhancement of LTC4 production by MK preincubation with LTCi production by eosinophils pre-incubated in control medium before stimulation by ionophore. The data in panel B are replotted from the experiments depicted in panel A. Where obtained, the percent enhancement of the responses by eosinophils from the same donor to more than one monocyte supernatant were averaged and depicted as a single point.

(r = - 0.79, P < 0.01) with the amounts of LTC4 released by eosinophils in the absence of monokine. It was suggested that the MK might have less effect on eosinophils which had been exposed to activating stimuli in vivo, i.e. during an inflammatory process in certain eosinophilic donors. To further analyse the effects of the monokine on AA metabolism, eosinophils were labelled with 3H-arachidonic acid prior to treatment with MK or control medium and stimulation by Ca2+ ionophore. RP-HPLC analysis of the released products showed that MK treatment increased counts at the retention times of LTC4 (119 and 137 070), 5-HETE (313 and 35 %) and free arachidonic acid (10 and 45 %) in 2 independent experiments, one of which is depicted in figure 8. These results suggested that MK exerts a positive effect on fatty acid hydrolase activity. The enhancing activity was not reduced by heating at 56°C for 30 min and was only partially decreased by heating at 90°C for 30 min. TSK-250 gel filtra- MONOCYTES, EOSINOPHILS, HELMINTHS AND LTC4 109

LTC LTB 5-HETE AA 4 4 1200 ~ ! ! !

1000 9 w I f-- I => I Z - 800 ~ a: w 600 0..

If) f-- ~ 400 I; ..

820011 I:! ~~ I ~~tD:I.. -.0 t 4 "L o 10 20 30 40 50 60 70 RETENTION TIME (min)

FIG. 8. - Reverse-phase HPLC resolution of arachidonic acid metabolites released from 3H-arachidonic-acid-labelled eosinophils which were incubated in supernatants of LPS-stimulated monocytes (e) or in control medium containing LPS (0) and activated with ionophore. The retention times of authentic standards are indicated.

tion of the monocyte supernatants yielded four peaks of enhancing activity (150,50,20 and 5 Kd). Isoelectrofocussing of the monocyte supernatant reveal• ed that the enhancing activity was associated with molecules of PI 4.2, 4.5 and 4.9 (fig. 9). This isoelectrofocussing profile showed close similarities with that of ECEF. All peaks of LT-generation-enhancing activity correspond to peaks of eosinophil-cytotoxicity-enhancing activity (fig. 9). Charge heterogenei• ty has been reported for other cytokines and may be accounted for by postranslational modifications, by limited differences in amino-acid sequences or by several unrelated molecules [40·43]. In summary, monocytes release factors(s) which enhance two biologically important eosinophil functions: helminthotoxicity and generation of AA-5-lipoxygenation metabolites. Although the data do not demonstrate that these effects are caused by a unique molecule or a family of related molecules, several observations point to this suggestion: the two enhancing activities are 110 P. ELSAS AND COLL. produced by monocytes in the same culture conditions; they are associated with molecules having comparable heat stability, size and charge heterogeneity. In addition, the data suggest that the enhancement of eosinophil helmin• thot6xicity depends on the normal functioning ofthe 5-lipoxygenase pathway. Since others have reported that 5-HETE enhanced neutrophil degranulation [37, 38] and that LTC4 and LTB4 increased eosinophil helminthotoxicity [44], it is possible that the enhancement of cytotoxicity is a consequence of the increased 5-lipoxygenase metabolism of AA. We were unable, however, to

!1 4.5 8-85 I 4.9 I

....c::'" 0.... 6 60 0c:: Vi 0 42 ~ ~'" ", ;: S2 .' 4 40 ;" ' (\J- "- C"> c:: u'

~ , i 0 01. 0 3 4 PH

FIG. 9. - Isoelectric focussing of LPS-stimulated adherent cell culture supernatants.

A. Enhancement of LTC4 production: eosinophils from 2 different donors were preincubated (2 separate experiments) with the fractions of isoelectric focussing of supernatants prepared with 2 different mononuclear cell preparations; then eosinophils were stimulated and assayed for LTC4 production. LTC4 production in samples incubated in control medium with LPS was 12 ng/ml (.A.) and 0.6 ng/ml (.)/2 x 10 eosinophils in the respective experiments. B. Enhancement of eosinophil cytotoxicity: eosinophils from 2 different donors were incubated with schistosomula, antischistosomular antibody and fractions from isoelectric focussing. Cytotoxicity was scored after 18 h. No enhancing activity was found outside the pI range 3.5-5 (data not shown). MONOCYTES, EOSINOPHILS, HELMINTHS AND LTC4 111 confirm the effects of LTC4 and LTB4 on eosinophil cytotoxicity and/or to activate the 5-lipoxygenase pathway with Ab-coated larvae. Several findings show that the MK which enhances eosinophil cytotoxici• ty and LT generation is distinct from previously described monokines: CSF-CI. and interleukin-1 (lL-1), interferons CI. and ~ (IFN-Cl.-and~) and tumour necrosis factor-CI. (TNF), which are also secreted by human monocytes in short-term cultures. IL-1 has been reported to increase the metabolism of exogenously added arachidonic acid in murine thymoma and to stimulate thromboxane synthesis in endothelial cells [45]. However, neither IL-1 nor the interferons increase eosinophil cytotoxicity [46]. Furthermore, IL-1 activity is heat-labile and its activity is destroyed after heating at 70°C for 30 min., unlike the activites described here [47]. IL-1 was also separated from the eosinophil• cytotoxicity-enhancing activity in RP-HPLC [48]. The two species of in• terferons secreted by monocytes, IFN-CI. and IFN-~, are destroyed by heating at 95°C for 30 min. This treatment only slightly affected the enhancing activities described here; furthermore, LPS does not induce large IFN secre• tion by monocytes (M. Criscuollo, personal communication). Recent reports indicate that TNF-CI. increased eosinophil cytotoxicity [49] and leukotriene production by neutrophils and eosinophils (R. Roubin, P. Elsas, W. Fiers and A. Dessein, manuscript submitted). However, TNF-CI. was not as potent as ECEF for the enhancement of eosinophil cytotoxicity and, unlike ECEF, TNF-CI. enhanced more arachidonic acid metabolism in neutrophils than in eosinophils. Moreover, TNF-CI. enhanced eosinophil respiratory burst, whereas ECEF did not. Finally, high eosinophil cytotoxicity and LT-generation-enhancing activity is released by U937 cells, which are poor producers of TNF-Cl.. A monokine with properties very similar to those of ECEF, termed eosinophil-activating factor (EAF) has been reported independently by But• terworth and collaborators [46,50-52]. EAF is a 40-Kd protein of pI 4.4 pro• duced by monocytes from individuals with moderate eosinophilia. It was shown to increase eosinophil cytotoxicity against schistosomula and against antibody• coated virus-infected mammalian cells. It also increased eosinophil degranula• tion, superoxide production and hydrogen peroxide production in response to immunological stimulants. It is heterogeneous in molecular weight, with a major component of about 40 Kd and a minor fraction of less than 10 Kd. It is also heterogeneous in charge, as assessed by TSK-DEAE chromatography. Recently, EAF was shown to be different from both CSF-CI. and TNF-CI. in TSK-DEAE-5PW chromatography [46]. Since EAF was shown to account for most of the eosinophil cytotoxicity-enhancing activity in monocyte super• natants, it is likely that EAF and ECEF are the same molecules. It is possible that other activities in these supernatants, i.e. CSF, account for the stimula• tion of eosinophil respiratory burst by EAF. The understanding of the biological role of these factors in vivo and their structural and functional relationships to other cytokines requires purificaton to homogeneity of the substances. This work is made difficult by the size and 112 P. ELSAS AND COLL. charge heterogeneity of the active moieties and by the small quantity of these factors in monocyte culture supernatants. The demonstration that U937 cells produce EAF which exhibit functional and biochemical properties iden• tical to those of the blood monocyte product enables us to undertake these studies [48].

RESUME

AUGMENTATION DE L'HELMINTHOTOXICITE ET DE LA PRODUCTION DE LEUKOTRIENE C4 PAR DES EOSINOPHILES ACTIVES PAR DES MONOCYTES

Les eosinophiles sont normalement presents en faible nombre dans Ie sang et les tissus des individus sains; leur nombre est par contre tres augmente dans certaines allergies et dans les helminthiases. Quoique les fonctions biologi• ques des eosinophiles ne soient pas totalement comprises, Ie role de ceux-ci dans la destruction des helminthes et dans la regulation des reactions d'hyper• sensibilite immediate est suggere par de nombreuses observations in vivo et in vitro. De recents travaux montrent que les eosinophiles isoles de certains patients sont dans un etat active, et plusieurs groupes de chercheurs, dont Ie notre, se sont interesses aux mecanismes immunologiques responsabIes de cette activation. Nous resumons ici, les resultats de nos etudes concernant Ie role possible des monocytes dans l'activation des eosinophiles humains par des facteurs proteiques solubles. II est rapporte que les surnageants de culture de monocytes humains augmentent tres fortement les capacites helminthotoxiques des eosinophiles ; ceux-ci apres incubation, durant quelques minutes, avec les surnageants sont capables de detruire les larves de Schistosoma mansoni en presence de quan• tites d'anticorps mille fois inferieures a celles requises par les eosinophiles temoins. Cette augmentation de la cytotoxicite est la consequence d'un effet «potentiateur» exerce par les produits des monocytes sur la degranulation de l'eosinophile. Les surnageants de monocytes augmentent egalement la production de leu• kotrienes par les eosinophiles; ceux-ci apres traitement durant quelques minutes avec les surnageants sont capables de produire jusqu'a 100 fois plus de leukotrienes que les temoins. Cette augmentation est la consequence d'un effet permissif exerce par les surnageants sur une des premieres etapes du metabo• lisme de l'acide arachidonique, probablement celles impliquant phospholi• pases ou 5-lipoxygenase. La ou les molecules produites par les monocytes et capables d'activer les eosinophiles ont ete caracterisees biochimiquement. Cette analyse montre que les differents effets activateurs des surnageants sont tres probablement causes par une seule molecule ou par une famille de molecules et que cette ou ces molecules sont distinctes de IL-l, CSF, INF-

MOTS-CLES: Monocyte, Eosinophilie, Helminthotoxicite, Allergie, Leuko• triene; Immunoregulation.

ACKNOWLEDGEMENTS

We would like to thank Dr Irmeli Penttila for reading the manuscript and Drs 10hn David and K. Frank Austen for helpful advice and discussion during the course of these studies.

REFERENCES

[1] RYTOMAA, T., Organ distribution and histochemical properties of eosinophil granulocytes in the rat. Actapath. microbiol. scand., 1960,50, (supp!. 140), 1-118. [2] DESSEIN, A.l. & DAVID, 1.R., The eosinophil in parasitic diseases. Adv. Host DeL Mech., (J.I. Gallin & A.S. Fauci) Vol. 1, (p. 243). Raven Press, New York, 1982. [3] EHRLICH, P. & LAZARUS, A., Die Aniimie, in «Specielle Pathologie und Therapie» H. Nothnagel. 1898, Vienna, Holder. Translated in English in «Histology of the blood: normal and pathological» (N. Myers), Cambridge University Press, Cambridge, New York, 1900. [4] BROWN, T., Studies on trichinosis, with special reference to the increase of the eosinophilic cells in the blood and muscle, the origin of these cells and their diagnostic importance J. expo Med., 1898, 3, 315. [5] SAMUELSSON, B., Leukotrienes: mediators of immediate hypersensitivity reactions and inflammation. Science, 1983, 220, 568. [6] LEWIS, R.A & AUSTEN, K.F., The biologically active leukotrienes: biosynthesis, metabolism, receptors, functions, and pharmacology. J. din. Invest., 1984, 73, 889. [7] 10RG, A., HENDERSON, W.R., MURPHY, R.C. & KLEBANOFF, S.l., Leukotriene generation by eosinophils. J. expo Med., 1982, 155, 390. [8] WELLER, P.F., LEE, C.W., FOSTER, D.W., COREY, E.l., AUSTEN, K.F. & LEWIS, R.A., Generation and metabolism of 5-lipoxygenase pathway leukotrienes by human eosinophils: predominant production of leukotriene C4 • Proc. nat. A cad. Sci. (Wash.), 1984, 80, 7626. [9] HENDERSON, W.R., HARLEY, 1.B. & FAUCI, AS., Arachidonic acid metabolism in normal and hypereosinophilic syndrome human eosinophils : generation of leukotrienes B4 , C4, D4 and 15-lipoxygenase products. Immunology, 1984, 51, 679. [10] DRAZEN, 1.M., AUSTEN, K.F., LEWIS, R.A., CLARK, D.A., GOTO, G., MARFAT, A. & COREY, E.1., Comparative airway and vascular activities of leukotrienes C and D in vivo and in vitro. Proc. nat. A cad. Sci. (Wash.), 1980, 77, 4354. 114 P. ELSAS AND COLL.

[11] PECK, M.J., PIPER, P.J. & WILLIAMS, T.J., The effect of leukotrienes C4 and D4 on the microvasculature of guinea-pig skin. Prostaglandins, 1981, 21, 315. [12] DAHLEN, S.E., BJORK, J., HEDQVIST, P., ARFORS, K.E., HAMMARSTROM, S., LiNDGREN, J.A. & SAMUELSSON, B., Leukotrienes promote plasma leakage and leukocyte adhesion in post-capillary venules: in vivo effects with relevance to the acute inflammatory response. Proc. nat. Acad. Sci. (Wash.), 1981, 78, 3887. [13] LEWIS, R.A., DRAZEN, J.M., AUSTEEN, K.F., CLARK, D.A. & COREY, E.J., Identification of the C(6)-S-conjugate of leukotriene A with cysteine as a naturally occuring slow-reacting substance of anaphylaxis (SRS-A). Impor• tance of the 11-cis-geometry for biological activity. Biochem. biophys. Res. Commun., 1980, 96, 271. [14] BUTTERWORTH, A.E., STURROCK, R.F., HOUDA, V. & REES, P.M., Antibody• dependent cell-mediated damage to schistosomula in vitro. Nature (Lond.), 1974, 252, 503. [15] Hsu, S.Y.L., Hsu, M.F., PENICK, G.D., HANSON, H.O., SCHILLER, H.J. & CHENG, H.F., Immunoglobulin E, mast cells and eosinophils in the skin of rhesus monkeys immunized with X-irradiated cercariae of S. Japonicum. Int. Arch. Allergy, 1979, 59, 383. [16] BUTTERWORTH, A.E., WASSOM, D.L., GLEICH, G.J., LOEGERMG, D.A. & DAVID, J .R., Damage to schistosomula of S. mansoni induced directly by eosinophil major basic protein. J. Immunol., 1979, 122, 221. [17] JONG, E.L., MAHMOUD, A.A.F. & KLEBANOFF, S.J., Peroxidase-mediated toxicity to schistosomula of S. mansoni. J. Immunol., 1980, 126, 468. [18] JAMES, S.L. & COLLEY, D.G., Eosinophil-mediated destruction of Schistosoma mansoni eggs. - III. Lymphokine involvement in the induction of eosinophil functional abilities. Cell. Immunol., 1978, 38, 48. [19] ANWAR, A.R.E., McKEAN, J.R., SMITHERS, S.R. & KAY, A.B., Human eosinophil-and neutrophil-mediated killing of schistosomula of Schistosoma mansoni in vitro. - I. Enhancement of complement-dependent damage by mast-cell-derived mediators and formyl-methionyl peptides. J. Immunol., 1980, 124, 1122. [20] CAPRON, M., CAPRON, A., GOETZL, E.J. & AUSTEN, K.F., Tetrapeptides of the eosinophil chemotactic factor of anaphylaxis (ECF-A) enhance eosinophil Fe receptors. Nature (Lond.), 1981, 289, 71. [21] CAPRON, M., ROUSSEAUX, J., MAZINGUE, C., BAZIN, H. & CAPRON, A., Rat mast cell eosinophil interaction in antibody-dependent eosinophil cyto• toxicity to Schistosoma mansoni schistosomula. J. Immunol., 1978, 121, 2518. [22] SPRY, C.J.F. & TAl, P.C., Studies on blood eosinophils. - II. Patients with Laffler's cardiomyopathy. Clin. expo Immunol., 1976, 24, 423. [23] SPRY, C.J.F., Alteration in blood eosinophil morphology, binding capacity for complexed IgG and kinetics in patients with tropical (filarial) eosinophilia. Parasite Immunol., 1981, 3, 1. [24] WASSOM, D.L., FAUCI, A.S., GLEICH, G.S., LOEGERMG, D.A., MOORE, S.G., SCHOOLEY, R.T. & SOLLEY, G.O., Elevated serum levels of the eosinophil granule major basic protein in patients with eosinophilia. J. din. Invest., 1981, 67, 651. [25] DAVID, J.R., ANDRADE, Z.A., BINA, S.C., BUTTERWORTH, A.E., CARVALHO, E.M., DAVID, R.A., BRITO, P.A. & VADAS, M.A., Enhanced helminthotoxic capacity of eosinophils from patients with eosinophilia. New Engl. J. Med., 1980, 303, 1147. [26] BASS, D.A., GORVER, W.H., LEWIS, J.C., SZEJDA, P., DECHATELET, C.R. & MCCALL, C.E., Comparison of human eosinophils from normals and patients with eosinophilia. J. din. Invest., 1980, 66, 1265. MONOCYTES, EOSINOPHILS, HELMINTHS AND LTC4 115

[27] PRIN, L., CAPRON, M., TONNEL, A.B., BLETRY, O. & CAPRON, A., Heterogeneity of human peripheral blood eosinophils: variability in cell density and cytotoxic ability in relation to the level and the origin of eosinophilia. Int. Arch. Allergy, 1983, 72, 376. [28] DESSEIN, A.J., VADAS, M.A., NICOLA, N.A., METCALF, D. & DAVID, J.R., Enhancement of human blood eosinophil cytotoxicity by semi-purified eosinophil colony-stimulating factor(s). J. expo Med., 1982, 156, 90. [29] DESSEIN, A.J., LENZI, H.L., VADAS, M.A. & DAVID, J.R., A new class of eosinophil activators that enhance eosinophil helminthotoxicity, in «Im• munobiology of the eosinophil» (T. Yoshida & M. Torisu) (p. 369). Elsevier, Amsterdam, 1983. [30] DESSEIN, A.J., LENZI, H.L. & DAVID, J.R., Modulation of the cytotoxicity of human blood eosinophils by factors secreted by monocytes and T lym• phocytes. Monogr. Allergy, 1983, 18, 45. [31] DESSEIN, A.J., LENZI, H.L., BINA, J.C., CARVALHO, E.M., WEISER, W.Y., ANDRADE, Z.A. & DAVID, J.R., Modulation of eosinophil cytotoxicity by blood mononuclear cells from healthy subjects and patients with chronic Schistosomiasis mansoni. Cell. Immunol., 1984, 85, 100. [32] SMITHERS, S.R. & DOENHOFF, M.J., Schistosomiasis, in «Immunology of parasitic infections» (S. Cohen and K.S. Warren) (pp. 527-607). Blackwell Scientific Pub!., Oxford, 1982. [33] LENZI, H.L., MEDNIS, A.D. & DESSEIN, A.J., Activation of human eosinophils by monokines and lymphokines: source and biochemical characteristics of the eosinophil cytotoxicity-enhancing activity produced by blood mononuclear cells. Cell. Immunol., 1985, 94, 333. [34] PINCUS, S.H., DESSEIN, A.J., LENZI, H.L., VADAS, M.A. & DAVID, J.R., Eosinophil-mediated killing of Schistosomula of S. mansoni: oxidative requirement for enhancement by eosinophil colony-stimulating factor (CSF) and supernatants with eosinophil cytotoxicity-enhancing activity (ECEA). Cell. Immunol., 1984, 87, 424. [35] CAUFIELD, J.P., LENZI, H.L., ELSAS, P. & DESSEIN, A.J., Ultrastructure of the attack of eosinophils stimulated by blood mononuclear cells products on Schistosomula of S. mansoni. Amer. J. Path., 1985, 120, 380. [36] CAUFIELD, J.P., KORMAN, G., BUTTERWORTH, A.E., HOGAN, M. & DAVID, J.R., Partial and complete detachment of neutrophils and eosinophils from schistosomula: evidence for the establishment of continuity between fused and normal parasite membranes. J. Cell. BioI., 1980, 86, 64. [37] STENSON, W.F. & PARKER, C.W., Monohydroxyeicosatetraenoic acids (HETEs) induce degranulation of human eosinophils. J. Immunol., 1980, 124, 2100. [38] BOKOCH, G.M. & REED, P.W., Effect of various lipoxygenase metabolites of arachidonic acid on degranulation of polymorphonuclear leukocytes. J. bioI. Chem., 1981,256, 5317. [39] DESSEIN, A.J., LEE, T.H., ELSAS, P., RAVALESE, J., SILBERSTEIN, D., DAVID, J.R., AUSTEN, K.F. & LEWIS, R.A., Enhancement by monokines of leukotriene generation by human eosinophils and neutrophils stimulated with calcium ionophore A23187. J. Immunol., 1986, 136, 3829. [40] MARCH, C.J., HOSLEY, B., LARSEN, A., CERRETTI, D.P., BARADT, G., PRICE, V., GILLIS, S., HENNEY, C.S., KRONHEIM, S.R., GROBSTEIN, K., CONLOSE, P.J., Hopp, T.P. & COSMAN, D., Cloning, sequence and expression of two distinct human interleukin-l complementary DNAs. Nature (Lond.), 1985, 315, 641. [41] CAMERON, P., LIMJUCO, G., RODKEY, J., BENNETT, C. & SCHMIDT, T.A., Amino acid sequence analysis of human interleukin-l (IL-l). Evidence for biochemically distinct forms of IL-l. J. expo Med., 1985, 162, 790. [42] STRANDER, M., Interferons (IFNs). Advanc. Cancer Res., 1986, 46, 1-9. 116 P. ELSAS AND COLL.

[43] WEISER, M.Y., GREINEDER, D.K., REMOLD, H.G. & DAVID, J.R., Studies on human migration inhibitory factor: characterization of three molecular species. J. Immuno/., 1981, 126, 1918. [44] MOQBEL, R., SAss-KuHN, S.P., GOETZL, E.J. & KAY, A.B., Enhancement of neutrophil-and eosinophil-mediated complement-dependent killing of Schistosomula of Schistosoma mansoni in vitro by leukotriene B4. CUn. expo Immuno/., 1983, 52, 519. [45] FARRAR, W.L. & HUMES, J.L., The role of arachidonic acid metabolism in the activities of interleukin-l and-2. J. Immuno/., 1985, 135, 1153. [46] THORNE, K.J., RICHARDSON, B.A., TAVESNER, J., WILLIAMSON, D.J., VADAS, M.A. & BUTTERWORTH, A., A comparison of eosinophil activating factor with other monokine and lymphokines. Europ. J.lmmuno/., 1986,46, 1143. [47] ROSENWASSER, L.J. & DINARELLO, C.A., Ability of human leukocytic pyrogen to enhance phytohemagglutinin-induced murine thymocyte proliferation. Cell. Immuno/., 1981, 63, 134. [48] SILBERSTEIN, D.S., DESSEIN, A.J., ELSAS, P.P., FONTAINE, B. & DAVID, J.R., Characterization of a factor from the U937 cell line that enhances the toxicity of human eosinophils to Schistosoma mansoni larvae. J. Immuno/., (in press). [49] SILBERSTEIN, D.S. & DAVID, J .R., Tumor necrosis factor enhances eosinophil toxicity to Schistosoma mansoni. Proc. nat. Acad. Sci. (Wash.), 1986,83, 1055. [50] VEITH, M.C., BUTTERWORTH, A.E. & BOYLSTON, A.W., The enhancement of eosinophil-mediated killing of schistosomula of Schistosoma mansoni by mononuclear cell products, in «Immunobiology of the eosinophil» (by T. Yoshida & M. Torisu) (pp. 305). Elsevier, Amsterdam, 1983. [51] VEITH, M.C. & BUTTERWORTH, A.E., Enhancement of human eosinophil• mediated killing of Schistosoma mansoni larvae by mononuclear cell products in vitro. J. expo Med., 1983, 157, 1828. [52] VEITH, M., TAYLOR, D.W., THORNE, K., RICHARDSON, B.A. & BUTTERWORTH, A.E., Studies on the enhancement of human eosinophil function by mononuclear cell products in vitro. CUn. expo Immuno/., 1984, 58, 603.