JOURNAL OF INTERFERON AND CYTOKINE RESEARCH 17:211-217 (1997)Mary Ann Liebert, Inc.

Relationship Among Circulating Interferon, Tumor NecrosisFactor-a, and Interleukin-6 and Serologie Reaction Against

Parasitic Antigen in Human HydatidosisCHAPIA TOUIL-BOUKOFFA,1 JOSIANE SANCÉAU,2 BOUCHENTOUF TAYEBI,1 and

JUANA WIETZERBIN2

ABSTRACT

Human hydatidosis is a parasitic disease vectored by the larval stage cestode Echinoccocus granulosus. It con-stitutes a major health problem in North Africa. We investigated the production of circulating interferon(IFN), tumor necrosis factor-« (TNF-a), and interleukin-6 (IL-6) in Algerian patients with liver, lung, or oc-ular hydatidosis. In all, 101 serum samples from these patients were analyzed. Immunoreactivity and cytokineactivities were undetectable in sera from ocular hydatidosis patients. However, we observed the presence ofIFN (a mixture of IFN-a, IFN-/3, and IFN-y, range 32-500 U/ml), TNF-a (range 32-100 U/ml), and IL-6(range 32-500 U/ml) in all patients who had liver or lung cysts ór both and displayed immunoreactivity againstparasitic antigen (antigen 5). After surgical removal of the cysts, serum cytokine levels declined rapidly andwere undetectable at 30 days. IFN and IL-6 activity was undetectable in sera from two liver hydatidosis pa-tients who relapsed and did not display any immune response against parasitic antigen. These results suggestthat in liver and lung hydatidosis, cytokine production contributes to the host defense mechanism against theextracellular parasite.

INTRODUCTION

HYDATIDOSIS CONSTITUTES A MAJOR HEALTH PROBLEM inNorth Africa and is common in some regions of the

Mediterranean and Latin America/1' It is a form of parasitosisvectored in humans by the larval stage cestode Echinococcusgranulosus, which induces the formation of cysts in various or-

gans. The major cyst locations are the liver (60%) and lung(30%) and, less frequently, the kidney, brain, eye, and bone.The clinical evolution of these cysts is silent for several months,and the symptoms are not specific. Diagnosis is difficult andoften belated, and surgery constitutes the only therapy. Adult-stage larvae are transmitted by the dog, a major source of eggsthat contaminate sheep and humans, leading to the evolutivecycle of hydatidosis. The hydatid fluid that fills the cysts con-

tains mineral salts, lipids, and proteins, one of which, antigen5 (also called fraction 5), is a specific antigen used for im-

munodiagnosis of hydatidosis.*2,3' Antigen 5 has been identi-fied by several groups as the major specific antigen ofEchinococcus granulosus.,(4,5)

Cytokines are extracellular signaling molecules often pro-duced during parasitic infections and involved in host de-fense.(6-7) Interferon (IFN) has been shown to be produced inthe course of experimental and human trypanosomiasis(8-9) andhuman malaria/10-1'' Other cytokines, such as tumor necrosisfactor (TNF) and interleukin-6 (IL-6), have been shown to bepresent in serum from patients with leishmaniasis(12-13) and inhuman and mutine malaria.'14-15' However, to our knowledge,no information is available about the production of cytokinesin human patients carrying a macroparasitic infection, such as

hydatidosis. We, therefore, thought it would be interesting toinvestigate the presence of cytokines in sera from patients in-fected with E. granulosus. We report here a study carried outon 101 Algerian hydatid patients. We investigated their sero-

'Laboratoire de Biochimie, ISN-USTHB, Université Bab-Ezzouar, Algiers, Algeria.2Unité 365 INSERM "Interférons et Cytokines," Institut Curie, Section de Recherche, Paris, France.

211

212 TOUIL-BOUKOFFA ET AL.

logic immunoreactivity against the specific antigen 5 and thepresence of circulating IFN, TNF-a, and IL-6 during the de-velopment of hydatidosis and after surgical removal of the cysts.

MATERIALS AND METHODS

Antigen preparationCrude E. granulosus cystic fluid was obtained from a fertile

human hepatic hydatid cyst. It was filtered to eliminate scolex,centrifuged at 600g and 4°C, and lyophilized. The resulting prepa-ration was processed by gel filtration chromatography usingSephadex G-200 equilibrated in 0.1 M Tris HC1, 1 M NaCl, pH8. Elution was carried out at a flow rate of 7 ml/cm2/h and was

maintained by a peristaltic pump. Eluted peaks were detected byUV recording at 280 nm and were analyzed against specific rab-bit anti-antigen 5 hyperimmune serum by immunoelectrophore-sis, as previously reported.'1,2) Antigen 5 was localized in a peakcorresponding to a molecular weight of around 65 kDa. This frac-tion was pooled, lyophilized, and used for sérologie testing.

SeraA total of 101 sera from three groups of Algerian patients

were analyzed. These patients had been hospitalized in the Ser-vice de Chirurgie Générale in Algiers and the Clinique Ophtal-mologique at Annaba in Algeria for hepatic, pulmonary, and oc-

ular hydatidosis. Their ages ranged from 5 to 65 years. Theaverage age of the subgroups for each location affected were 39,32, 42, and 12 years for the liver, lung, liver plus lung, and eye,respectively. Patients with ocular hydatidosis exhibited a singlecyst in the left eye. Two thirds of the patients were male, and 2were in the relapsing stage. None of the patients had previouslybeen transfused, and no surgical or pharmacologie treatment was

given before blood sampling. No patient displayed another par-asitic or viral infection or any bacterial pathology at the time ofblood sample collection. Hydatid patients were surgically con-

firmed cases. Their sérologie reaction against antigen 5 was

tested by immunoelectrophoresis. The three groups of patientsstudied were not tested at the same time because they belong todifferent periods in their hospitalization and clinical evolution.We did not stock the sera to test all of them together.Cells

Wish and L929 cells were maintained, respectively, in MEMand DMEM medium supplemented with 10% heat-inactivatedfetal bovine serum (FBS) (Gibco, Parsley, Scotland), 2 mM glu-tamine, and gentamicin (How laboratories, Rockwell, MD) ina 5% humidified atmosphere at 37°C. The 7TD1 hybridomacell line (a gift from Dr. J. Van Snick) was grown at 37°C in7% C02 in DMEM medium supplemented with 10% FBS, 1.5mM L-glutamine, 0.24 mM L-asparagine, 0.55 mM L-arginine,50 mM thymidine (Gibco), and 20 U/ml of rIL-6 (produced inINSERM Unit 365, Paris).

IFN assay and neutralization test

IFN activity was assayed by a cytopathogenic inhibition teston Wish cells with vesicular stomatitis virus (VSV) as a chal-

lenge. Sixteen thousand cells per well were seeded in 96-wellmicroplates (Nunclon) in 100 pi of medium. After 24 h, theplates were drained, and 100 pfu of VSV was added in 100 piof medium containing 5% FBS. International NIH referencesamples of IFN-a, IFN-/3, and IFN-y were included in each as-

say. Titers were expressed in international reference units as

the reciprocal of the dilution giving 50% of the cytopathogeniceffect/16' The lower limit of detection was 1 IU/ml.

Neutralization assays were carried out by incubating the sam-

ples for 2 h at 37°C with an excess of neutralizing anti-IFNantibodies and assaying them for residual IFN activity, as de-scribed. Rabbit polyclonal antibodies against human recombi-nant IFN-a and human recombinant IFN-y were prepared inINSERM Unit 365 and had a neutralization titer of 40,000against 10 U of the corresponding IFN. Antihuman IFN-/3 was

a gift from Dr. Paula Pitha and had a neutralization titer of100,000 against 10 U of IFN-/3.

77VF-a test

TNF-a activity was tested by the cytotoxicity assay on L929cells in the presence of 1 pg/ml of actinomycin D (Sigma, Saint-Louis, MO), as previously reported/17' A sample of pure re-

combinant TNF-a (2000 U/ml, sp. act. 5 X 107 U/mg, Genen-tech, provided by Boehringer-Ingelheim, Vienna, Austria) was

used as reference. The limit of detection of the biologic test was

1 TJ/ml. A rabbit antiserum against recombinant TNF-a (neu-tralization titer 30,000 against 10 U/ml of TNF-a) prepared inINSERM unit 365 was used for characterization.

IL-6 test

IL-6 activity was quantified on 7TD1 hybridoma cells as a

function of its growth factor activity, as described by Van Snicket al/18' A sample of pure rIL-6 (2000 U/ml, sp. act. 2 X 108U/mg) was used as reference in each assay. The limit of de-tection was 1 U/ml. A monoclonal antibody against recombi-nant IL-6 (neutralization titer 10,000 against 10 U/ml of IL-6)(gift of Dr. J. Wijdenes, Besançon, France) was used for char-acterization.

RESULTS AND DISCUSSION

Correlation of circulating IFN with immunoreactivityagainst parasite-specific antigen and cystic location inhydatid patients

We assayed IFN activity in a series of serum samples frompatients with hydatid cysts in the liver, lung, liver plus lung,and eye. As shown in Table 1, IFN activity was detected in 31of 33 patients with cysts in the liver, lung, or both. In contrast,no significant IFN activity was detected in patients with ocularcyst or in the 2 patients who relapsed after radical surgery fora liver cyst. We noted with interest that IFN production corre-

lated with immunoreactivity against parasitic antigen in all pa-tients with lung, liver, or lung plus liver cysts, suggesting thatimmunostimulation by the parasitic antigen is involved in theimmune response leading to IFN production. The patients in re-

CYTOKINE PRODUCTION IN HYDATIDOSIS 213

Table 1. IFN Activity, Immunoreactivity Versus Parasitic Antigen(Antigen 5) and Cystic Location in Hydatid Patients

Cysticlocation

Serain)

Surgicaltreatment

IFNrange (U/ml)

Medianrange

Serologie reactionagainst antigen 5

LungLiverLiver

Liver+ lung

EyeHealthy

donors

172

833

NoneNonePatients

in relapse2None

None

64-50018-5000-4

96-256

2-242-16

192128

2

128

164

++

aPatients who relapsed 18 months after surgical removal of cysts.

Table 2. Antigenic Characterization of Circulating IFN by Anti-IFN Antibodies

Sample None*+ Anti-IFN-a

+ Anti-IFN-ß

+ Anti-IFN-y

+ Anti-lFN-a, ß, ycombined

Serum 1Serum 2Serum 3Serum 4IFN-aIFN-/3IFN-y

2001286464

200200200

1003232162

200200

3216

88

2004

200

64321616

200200

0

42-4

2400o

aNot treated with anti-IFN antibodies.

lapse and those with ocular hydatidosis exhibited no im-munoreactivity against antigen 5.

Table 2 shows the effects of antibodies against IFN-a, IFN-ß, or IFN-y on the antiviral activities in the sera of hydatid pa-tients. Partial neutralization of IFN activities was observed us-

ing anti-a, anti-/3, or anti-y neutralizing antibodies, andcomplete inhibition was obtained in the presence of a combi-nation of the three antisera. These results suggest that the an-

tiviral activity present in patient sera was a mixture of IFN-a,IFN-/3, and IFN-y.

Effect of surgical cyst removal on circulating IFNlevels in patients with liver hydatidosis

Because we observed a correlation between the immunore-activity against the parasitic antigen and circulating IFN levels,

Table 3. Serum IFN Levels Before and After SurgicalCyst Removal in Patients with Liver Hydatidosis

Surgical IFN Serologie reactionSera treatment range (U/ml) Median against antigen 5

n = 40n = 40n = 40Healthy

donorsn = 19

32-2008-644-162-16

10032

84

++

"Period before surgical removal covers 1 week.bPeriod following surgery covers 72 hb and 30 days.c

214 TOUIL-BOUKOFFA ET AL.

Table 4. IFN, TNF-a, and IL-6 Serum Levels Before and After Surgical Removal of Hydatid Cysts from Liver

Sera

IFN TNF-a IL-6

Surgicaltreatment

Range(U/ml) Median

Range(U/ml) Median

Range(U/ml) Median

Serologiereactionagainst

antigen 5

n = 20n = 20n = 20n = 2

Healthydonorsn = 8

Patientsin relapse0

32-50032-1008-160-4

2-16

1003216

32-10016-328-16

64

4-16

6432

32-50064-5000-32

0

0-8

100250

8

++

"Before surgery.b72 h after surgery.c30 days after surgery.dPatients who relapsed 18 months after surgery.

n = 20 n = 20 n = 20 n = 8

500

»200

*100

I3D 64

32

164-8-14

Patients before Patients after Surgical removal Healthysurgical removal 72 hours 30 days donors

FIG. 1. Serum IFN levels before and after surgical removal of cysts in patients with liver hydatidosis. Each point representsthe mean of triplicate assay values for the same patient during the period indicated. Dashed line indicates the upper limit of allvalues found in sera of healthy donors and of patients 30 days after surgery. The period before surgery covered 1 week.

CYTOKINE PRODUCTION IN HYDATIDOSIS 215

we decided to explore the effect of surgical removal of the cyston these levels. For this purpose, we tested another group of 40patients before and after cyst removal. Parallel tests were con-

ducted on the sera of 19 healthy donors from the same region.As shown in Table 3, IFN activity was detected in all patientsbefore surgery (IFN titers ranged from 32 to 200 U/ml; median100 U/ml). Interestingly, a marked drop in IFN levels was al-ready observed 72 h after surgery (range 8-64, median 32U/ml). Thirty days thereafter, circulating IFN levels were verylow and almost comparable to the levels detected in healthydonors. The reduction in IFN activity correlated with the lossof immunoreactivity against antigen 5, suggesting that this anti-gen was involved in triggering the host response. These resultsindicate that the reduction in IFN production and the loss ofimmunoreactivity were due to the fact that surgical removal ofthe cysts drastically reduced the availability of soluble parasiticantigen, especially the major antigen 5.

TNF-a and IL-6 circulating levels before and aftersurgical removal of cysts

Since parasitic infections have been shown to be associatedwith cytokine production, especially of TNF and IL-6, we com-

pared the production of these cytokines with IFN production insera from a third group of 20 hydatid patients. TNF, IL-6, and

IFN activity were detected in the serum of all 20 patients be-fore surgery (Table 4 and Figs 1, 2, and 3). After surgery, TNFand IFN activity declined very rapidly, reaching a very lowlevel after 30 days. We noted with interest that the patients whorelapsed 18 months after surgery exhibited sustained levels ofTNF activity (64 U/ml) in contrast to their IFN and IL-6 lev-els, which were almost undetectable. As these patients displayedno immunoreactivity against antigen 5, this observation impliedthat their TNF production was not triggered by antigen 5. Highlevels of circulating IL-6 were detected in all the 20 patientstested before surgery. Unlike TNF and IFN levels, IL-6 levelsdid not decline during the 72 h after surgery, probably becausedifferent factors contribute to IL-6 induction. Nevertheless, 30days after surgery, IL-6 levels were very low.

In conclusion, the results reported here show, for the firsttime as far as we know, that circulating IFN, TNF, and IL-6are produced during human hydatidosis. Our observations are

in line with those of other recent studies showing that after stim-ulation by parasitic antigen, peripheral blood mononuclear cellsproduce IFN-y in v/iro/19' Our results indicate that cytokineproduction might be implicated in the host defense mechanismagainst the parasitic infection and that IFN and IL-6 productioncorrelated with the presence of antibodies against antigen 5.This correlation suggests that antigen 5 is at least partly re-

sponsible for IFN and IL-6 production, unlike that of TNF,

n = 20 n = 20 n = 20 n = 8

100 4••

c

i64

32 >•

16

Patients beforesurgical removal

Patients after72 hours

Surgical removal30 days Healthy

donors

FIG. 2. TNF-a serum levels before and after surgical removal of cysts in patients with liver hydatidosis. Each point representsthe mean of triplicate assay values for the same patient during the period indicated. Dashed line indicates the upper limit of allvalues found in sera of healthy donors and of patients 30 days after surgery. The period before surgery covered 1 week.

216 TOUIL-BOUKOFFA ET AL.

n = 20 n = 20 n = 20

500

»

250

200

»100 >•••••••

I Xcp 64

32 i

164-

> »Patients before

surgical removalPatients after

72 hoursSurgical removal

30 daysHealthydonors

FIG. 3. IL-6 seruniJevels before and after surgical removal of cysts in patients with liver hydatidosis. Each point representsthe level of IL-6 observed in each individual. Dashed line indicates the upper limit of all values found in patients 30 days aftersurgery. The period before surgery covered 1 week.

which persisted in the relapsing cases. Consequently, it may bepostulated that after surgical removal of hydatid cysts, mea-

surement of serum cytokine levels, especially the TNF level,might allow early detection of relapses. Further investigationsare necessary to demonstrate the usefulness of these measure-

ments for diagnosis.

ACKNOWLEDGMENTS

The authors thank Dr. B. Hamrioui (Service de Parasitolo-gie, Institut Pasteur, Alger), the clinicians of the Service deChirurgie (CHU, Alger), and Dr. W. Boukoffa (Service d'Oph-talmologie, CHU, Annaba, Algeria) for providing serum sam-

ples. We are grateful to Mrs Agnès Birot for expert secretarialand iconographie assistance, Dr. Johan Van Weyenbergh forcritical reading of the manuscript, Dr. Brigitte Bauvois for help-ful discussion, and Mrs C. Silvestri for technical assistance.

This work was supported by the French Ministry for ForeignAffairs (93 Murs 28 program) and the INSERM.

REFERENCES

1. CESBRON, J.Y., CAPRON, M., and CAPRON, A. (1986). Le di-agnostic immunologique de l'hydatidose humaine. Gastroenterol.Clin. Biol. 10,415-418.

2. BOUT, D., FRUIT, J., and CAPRON, A. (1974). Purification d'unantigène spécifique du liquide hydatique. Ann. Immunol. 125,775-788.

3. FARAG, H., BOUT, D., and CAPRON, A. (1975). Specific im-munodiagnosis of human hydatidosis by the enzyme linked im-munosorbent assay (ELISA). Biomedicine 23, 276-278.

4. DI FELICE, G., PINI, C, AFFERNI, C, and VICARI, G. (1986).Purification and partial characterization of major antigen (antigen5) with monoclonal antibodies. Mol. Biochem. Parasitol. 20,133-137.

CYTOKINE PRODUCTION IN HYDATIDOSIS

5. SIRACUSANO-ATEGGI, A., QUIENTERI, F., DEROSA, F.N.,and VICARI, G. (1988). Cellular immune responses of hydatid pa-tients to Echinococcus granulosus antigen. Clin. Exp. Immunol.72, 400-405.

6. COX, F.E.G., and LIEW, Y. (1992). Cell subsets and cytokines inparasitic infections. Parasitol. Today 8, 371-374.

7. BARCINSKI, M.A., and COSTA-MOREIRA, M.E. (1994). Cel-lular response of protozoan parasites to host-derived cytokines. Par-asitol. Today 10, 352-355.

8. KIERSZENBAUM, F., and SONNENFELD, G. (1982). Charac-terization of the antiviral activity produced during Trypanosomacruzi infection and protective effects of exogenous interferonagainst experimental Chagas. J. Parasitol. 68, 194—198.

9. BRANCFORT, G.I., SUTTON, C.J., MORRIS, A.G, andASKONAS, A.G. (1986). Production of interferon during experi-mental African trypanosomiasis. Immunology 52, 135-143.

10. OJO-AMAIZE, E.A., LEKAN, S.S., WILLIAMS, A.I.O., AKIN-WOLERE, O.A.O., SHABO, R., ALM, G., and WIGZEL, H.(1981). Positive correlation between degree of parasitemia, inter-feron titers, and natural killer cell activity in Plasmodium falci-parum infected children. J. Immunol. 127, 2296-2300.

11. RHODES-FEUILLETTE, A., BELLAS-QUARDO, M., DRUILHE,P., BALLET, J., CHOUSTERMAN, J., CANTVET, M„ and PERTES,J. (1985). The interferon comportment of the immune response inhuman malaria: presence of the serum interferon gamma follow-ing the acute attack. J. Interferon Res. 5, 169-178.

12. CACERES-DITTMAR, G, TAPIA, F.J., SANCHEZ, M.A., YA-MAMURA, M., UYEMURA, K., MODLIN, R.L., BLOOM, B.R.,and CONVIT, J. (1993). Determination of the cytokine profile inAmerican cutaneous leishmaniasis using the polymerase chain re-

action. Clin. Exp. Immunol. 91, 500-505.13. MELBY, P.C., ANDRADE-NARVAEZ, F.J., DARNELL, B.J.,

VALENCIA-PACHECO, G., TRYAN, W., and PALOMO-CETINA, A. (1994). Increased expression of proinflammatory cy-tokines in chronic lesions of human cutaneous leishmaniasis. In-fect. Immunol. 62, 837-842.

14. KERN, P., HEMMER, C.J., VAN DAMME, J., GRUSS, HJ., and

217

DIETRICH, M. (1989). Elevated tumor necrosis factor alpha andinterleukin-6 as markers for complicated Plasmodium falciparummalaria. Am. J. Med. 87, 139-143.

15. PRADA, J., MALINOWSKI, J., MÜLLER, S., BIENZLE, U, andKREMSNER, P.G (1995). Hemozoin differentially modulates theproduction of interleukin-6 and tumor necrosis factor in murinemalaria. Eur. Cytokine Netw. 6, 109-111.

16. WEITZERBIN, J., KOLB, J.P., SENIK, A., DER STEPANI, L.,ANDREW, G, FALCOFF, E., and FALCOFF, R. (1984). Studiesof purification of human gamma interferon: Chromatographie be-havior of accompanying IL-2 and B cell helper activity. J. Inter-feron Res. 4, 141-151.

17. SUGARMAN, B., AGGARWAL, B., HARS, P.E., FIGARI, I.S.,PALLADINO, M.A., and SHEPARD, M.S. (1985). Recombinanttumor necrosis factor: effect on proliferation of normal and trans-formed cells in vitro. Science 230, 141-151.

18. VAN SNICK, J., CAYPHAS, S., VINK, A., UYTENHOVE, C,COULIE, P., and SIMPSON, R.J. (1986). Purification and NH2-terminal amino acid sequence of a T-cell derived lymphokine withgrowth factor activity for B cell hybridoma. Proc. Nati. Acad. Sei.USA 83, 9676-9679.

19. RIGANO, R., PROFUMO, E., DIFELICE, G, ORTONA, E.,TEGGI, A., and SIRACUSANO, A. (1995) In vitro production ofcytokines by peripheral blood mononuclear cells from hydatid pa-tients. Clin. Exp. Immunol. 99, 433^139.

Address reprint requests to:Dr. Juana Wietzerbin

Institute Curie, Unité 365Section de Recherche

Pavillion Pasteur26, rue d'Ulm

75231 Paris cedex 05France

Received 27 lune 1996/Accepted 24 December 1996