IgE in human resistance to S mansmi
Eur. J. Immunol. 1991. 21: 2679-2686
2679
Pascal Rihet, Christian E. Demeure., Main Bourgois, Aluizio Prata* and Main J. Dessein
Evidence for an association between human resistance to Schistosoma mansmi and high anti-larval IgE level@
Centre d'Immunologie
The anti-larval IgE antibody response of adolescents with high or low resistance to infection by Schistosoma mansoni was evaluated before parasitological cure with oxamniquine and over an extended post-treatment period during which the least resistant subjects regained high infections. IgE from most sera, taken at several bleeding times before and after treatment, reacted, on immunoblots,with a large number of antigens (Ag) in schistosomular tegument extract. A family of 120-165-kDa cross-reactingmolecules and a 85-kDa Ag were the most prominent Ag. Some of these determinants were shown to be located on the outer tegumental membrane and to be accessible to IgE on living larvae. The comparison of IgE between the two study groups showed that IgE levels were on average six- to eightfold higher (p CO.01) in the sera of the most resistant adolescents whereas there was no difference in patterns of Ag recognition between study groups. In contrast to IgE, anti-larval IgG and IgM levels were either similar in both groups or higher in the least resistant subjects when these exhibited high reinfection levels. IgG that competed for the binding of IgE to larval Ag were detected in most sera and their levels were higher in the Ieast resistant group after reinfection. FinaHy, the treatment had no observable long-lasting effects on the levels and on the specificity of the anti-larval IgE. Altogether, these observations can be taken as evidence supporting a role of I@ in human resistance to infection by S. mansoni.
INSERMXNRS de Marseille-Luminy and Facaldade de Medicha do lkiangulo Mineboa, Uberaba
1 Introduction
ent of animal [lo-131 and human [14-161 protective immunity against S. mamoni, we began to evaluate IgEdependent immunity in subjects with high resistance to infection. The objectives of the present work were: (a) to identify the principal schistowmula (sm) allergens; (b) to evaluate whether lower resistance levels were associated with qualitative and/or quantitative alterations of the anti-larval IgE response; and (c) to determine the effects of a parasitological cure with oxamniquine on this response.
Schistosomiasisdue to S. mansoni is endemic in large areas of subtropical countries where it still represents a serious health problem and a burden for the economy [l]. The discovery of potent and safe schistosomicids allowed a reduction of the morbidity associated with infection by this parasite. However, the control of schistosomiasiscannot be achieved by chemotherapy alone and potent, cheap and safe vaccines are needed [2-61. In endemic areas, certain individuals have no sign of active infection in spite of daily contacts with infected waters, indicating that these individuals exhibit a high level of immunity to S. mamoni [7-91. 2 Materials and methods Analysis of the immune response of individuals with different levels of resistance has been undertaken to 2.1 Study subjects understand the principal mechanisms of resistance and to identify their targets on the parasite. Since a number of Thirty-eight young individuals (age 10 to 19years) were observations indicated that IgE could be a critical compon- selected for similar and high water contacts among subjects of an area endemic for S. munsoni.The area, the subjects of this study and the evaluation of their water contact levels have been described in detail previously [9,17,18]. All 38 [I 94121 subjects were given oxamniquine 2 months after the beginning of this study and none of them excreted S. munsoni * This work received financial assistance from INSERM,CNRS, eggs for 3 months following treatment. lkenty of them the Rockefeller Foundation and the UNDPrWorld BanWWHO A, Table 1) were reinfected within 15 to 27 months (group Special Program for Research and Training inllopical diseases. This work was first published as an abstract in the proceedings of after treatment and progressively developed heavy infecthe 7th International Congress of Immunology in Berlin, tions; these individuals, who had also heavy infections 1989. before treatment, are referredto as subjectswith the lowest Supported by a fellowship from CNRS and the Provence Apes resistance to infection. In contrast, 18individuds (group B, a t e d'Azur Council. Bble 1) excreted no or few eggs for a 43-month period following treatment and had low infections before treatCorrespondence: Main J. Dessein, Centre d'Immunologie de ment; these subjects are referred to as individuals with the Marseille-Luminy, Campus de Luminy, Case 906, F-13288 Marhighest resistance to infection [9]. Since it has not been seille Cedex 09,France possible to bleed all 38 subjects at all study times, the sizes Abbreddons: m: Schistommula NRS: Normal rabbit se- of study groups at the different times are shown on Fig. 1. rum 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
0014-2980/91/1111-2679$3.50+ . Z / O
2680
Eur. J. Immunol. 1991.21: 2679-2686
I? Wet, C. E. Demeure, A. Bourgois et al.
2.2 Chemicals
2 6 Elution of Ab bound to blotted Ag
MEM, Hepes, penicillin and streptomycin were purchased from Gibco (Paisley, Scotland); n-octyl-glucoside(l-0n-octyl-fl-D-glucopyranoside) , pepstatin, PMSF, Chaps, a2-macroglobulin, leupeptin, 2-ME, acetyl-thiocholin iodide and dithionitrobenzoic acid were from BoehringerMannheim Biochemicals (Mannheim, FRG); aprotinin, protein A, bovine hemoglobin, BSA, Tween 20, Ponceau red stain, Mr markers and p-nitrophenyl-phosphate were from Sigma Chemicals Co.(St. Louis, MO).The El24 and 6H10 anti-human (hu)IgE mouse mAb [191, acetylcholine esterase-coupled anti-huIgE mouse mAb, FITC-conjugated goat anti-rabbit IgG and FITC-conjugated goat anti-mouse Ig were from Immunotech (Marseille, France). Alkalinephosphatase-coupledanti-huIgG mouse mAb and anti-huIgM mouse mAb were purchased from ICN K & K Laboratories (Eschwege, FRG). Protein A-coupled Sepharose cG4B and cyanogen bromide-activated Sepharose were purchased from Pharmacia Fine Chemicals (Uppsala, Sweden).
Ag blotted on nitrocellulosepaper were reacted with a pool of human immune sera. Bands of interest were identified as in [9] and were cut out from the other part of the blot. Ab bound to these strips were eluted with a 0.2 M glycine/HCl buffer, pH 2.8, and eluates were adjusted to pH 7.4.
2.3 Schistowmula The Puerto-Rican strain of S. mansoni used in this study was maintained by passage through CBAIJ mice and Biomphalaria glabrata snails. Schistosomulawere prepared by the method of Ramalho-Pinto et al. [20], as modified by Lazdins et al. [21], and cultured (3 h, 37°C) in MEM supplemented with 10 m~ Hepes, pH 7.2,100 I U / d penicillin and 100 pg/ml streptomycin.
2.7 Ab titration Titrations were performed by ELISA [23] on plates coated with 100 pl of a 5 p g / d solution of sm tegument extract in 50 m~ sodium carbonate buffer, pH 9.6. Ag-coated plates were saturated with PBSBSA (30 g/l BSA) and reacted with serum dilutions (1 : 10 to 1:270 for IgE and 1:100 to 1:5000for IgG and IgM) for 16 h at 4 "C.Then, plates were washed three times with 0.05% Tween20 in PBS and reacted for 3 h at room temperature with either acetylcholine esterase-coupled anti-huIgE mAb or alkaline phosphatase-coupled anti-huIgG mAb. After three additional washes, enzymatic activities were revealed by addition of either 0.5 m~ acetylthiocholine iodide and 0.5 m~ dithionitrobenzoic acid in 50 mM phosphate buffer, pH 7.6 (for acetylcholine esterase) or 1mglml p-nitrophenyl-phosphate in 10% diethanolamine buffer, pH 9.6 (for alkaline phosphatase). The reactions were read at 405 nm. Positive and negative reference sera were used to draw a standard curve. Results were calculated by inference from this standard curve and expressed in U/ml. By definition, lOOU/ml is the amount of specific IgE, IgG or IgM measured in either a 1/10 (IgE) or a 1:lOO (IgG, IgM) dilution of the most reactive serum. 2.8 Evaluation of the competition effect
2.4 Tegument extraction Schistosomular tegument Ag were extracted by resuspending sm (l@/ml) in a 1% n-octylglucoside solution containing 10 mM sodium phosphate buffer, pH 7.1,4 mM MgC12, 140 m~ NaCl, to which the following protease inhibitors were added just before use: 50 p~ PMSF, 50 p g / d aprotinin, 1p~ pepstatin, 20 pg/ml leupeptin, and 10 pg/ml a2-macroglobulin. After 15 min shaking at 4°C in extraction buffer, sm bodies were pelleted (400 X g, 3 min, 4"C), and the SN were centrifuged at 100OOO x g for 35 min at 4°C. SN were stored aliquoted at -70°C until use.
Sera, diluted 1:2 in buffer A (0.2% n-ocytylglucoside, 0.1% Chaps in PBS), were centrifuged (1Omin; 100OOO x g) and the SN were reacted with protein ASepharose (90min, 4°C). Unbound material was collected in the first two washes of the beads with buffer A. Ig bound to protein A were eluted with a 0.1 M glycine buffer, pH 2.8 [24] and pH of the eluates were adjusted immediately to 7.4. Filtrates and eluates were then, dialyzed against PBS (4"C, 48 h) and anti-sm IgE Ab in these fractions were measured by ELISA. The competition exerted by IgG on the binding of IgE to sm Ag was expressed as the ratio of specific IgE levels (U/ml) in protein A-adsorbed fractions to IgE levels (U/ml) measured on the respective unfractionated sera.
2.5 Western blotting
Western blots were carried out following the method of Towbin et al. [22] modified as in 191 excepted that Ig bound to the blotted Ag were revealed with the El24 1251-labeled anti-huIgE mAb (2 x lo6 cpm/ml). This mAb binds to the heat-sensitive De2 IgE antigenic region [19]. Heating immune human sera (56"C, 3 h) before reacting them with the blots extinguished all bands (data not shown). M, were estimated by comparison to the following markers: myosin (205 m a ) , @-galactosidase(116 m a ) , phosphorylase b (97.4 kDa), BSA (68 m a ) , OVA (45 kDa) and carbonic anhydrase (29 kDa).
2.9 Preparation of anti-P165 rabbit sera
sm tegument extracts, containing 1mg of proteinlml,were heated (lOO"C, 3 min) in 50 mM Tris buffer, pH6.8, containing 1% SDS, 5% 2 - a , 1 mM EDTA, 10% (v/v) glycerol, and subjected to electrophoresis on a 7.5% SDS-polyacrylamidegel.The 165-kDaband was identified, cut off from the gel and its protein content was electroeluted as in [9]. Two rabbits were primed with 50 pg of the electroelutedP165preparation in CFA and boosted twice at 2-month intervals with 50 pg of the same material in F A .
Eur. J. Immunol. 1991.21: 2679-2686
IgE in human resistance to S.rnunsoni
2681
I
Sera were collected beginning 2 weeks after the last boost. 2.10 Immunopuri6cation of hnIgE The 6H10 [19] anti-huIgE mAb (1 mg) was coupled to 0.6 g of CNBr-activated Sepharose4B. Then, 200ml of the immunosorbent was reacted (2 h at room temperature) with 1.5 ml of a pool of sera from five infected subjects. After washing with 10 vol. of PBS, 1%BSA and 2 vol. of 0.1 M phosphate buffer, pH 9.2,1% BSA, IgE were eluted with a 0.1 M triethylamine solution, pH 11.5, containing 1% BSA.
10
-
0 0
;i
2.ll Immuno5uorescence studies Fresh sm washed in cold RPMI/FCS medium (WMI 1640 containing 10% FCS, 10 mM Hepes, pH 7.2, 100 IU/ml penicillin and 100 p g / d streptomycin),were reacted for 2 h at 4°C either with 1 / 1 0 dilution of the rabbit anti-P165 serum or 2 pg/ml of immunopurified IgE from a pool of sera from infected subjects. A 1/100 dilution of NRS and 2 pg/ml of myeloma huIgE were used as controls. Then, sm were washed three times in WMIiFCS and reacted (30 min, 4°C) either with FITC-conjugated goat anti-rabbit IgG or with a 100 p g / d solution of anti-huIgE mAb (E124) followed by three washes and an incubation (30 min, 4 "C) with FITC-conjugated goat anti-mouse Ig. Then, sm were washed three times in RPMIiFCS and examined under a fluorescence microscope at a X 100 magnification.
1-
0
o
m
0
0
0
-2 mo
+15 mo
+43 mo
pt0.005
pcO.005
ptO.O1
Figurel. Anti-sm IgE levels are higher in the most resistant adolescents as compared to the least resistant ones. Anti-sm IgE levels were assessed by ELISA on 1 :10 serum dilutions. Comparison of the two groups was performed using Student's t-test after logarithmic normalization of the data. Geometric means and SEM are drawn. Geometric means were 20.8 and 3.2Ulml at -2 months, 28.4 and 3.8 U/ml at +15 months and 26.7 and 4.4 U/ml at +43 months with sera from subjects with the highest (0)and lowest (0)resistance, respectively.
average, anti-sm IgE levels were higher in the sera of the most resistant subjects as compared to the least resistant ones at all study times, and there was a 6.5- to 7.6-fold 3.1 Effect of oxamniquine treatment on the anti-sm IgE difference in specific IgE levels between the two study response of study subjects groups. IgE levels in sera of group B were rather homogenous: all but one sera in this group exhibited a good Anti-sm IgE levels were measured by ELISA in sera taken reactivity in this assay; in contrast IgE levels in sera of from subjects 2 months before oxamniquine treatment as group A were heterogeneous with certain sera exhibiting well as 15 and 43months after. These measurements no or little reactivity, whereas others reacted well in the showed that no major changes in IgE levels occurred after assay. treatment. First, anti-sm IgE levels measured in sera taken 2 months before and 15 months after treatment as well as in sera taken 43 months and 15 months after treatment were 3.3 Anti-sm IgG and IgM levels are not increased in and T = 0.96, closely correlated (I = 0.95, p < subjects with the hig4est resistance to infection p< respectively); this rules out a selective effect of the treatment on IgE levels of certain subjects. Second, IgE To assess whether this observation of higher IgE levels in levels were comparable at various times of the study, as the most resistant subjects was sblective for IgE or common indicated by the slopes of the regression lines of IgE to other isotypes, anti-sm IgG and IgM were evaluated in (-2 months) to IgE (+15 months) and IgE (+43 months) the same sera. In contrast to the observations made on IgE, to IgE (+15 months) as these slopes are close to 1. This the levels of specific IgG and IgM were not significantly confirms that the increase of IgE levels observed in several different in the sera of the two study groups at -2 months studies, within few weeks after treatment [25,26], is a and +15 months; moreover, at +43 months, IgG and IgM transient phenomenon. were lower in the sera of the most resistant group (Table 1).
3 Results
3.2 Antiam IgE levels are higher m the sera of the most resistant sobjects We compared anti-sm IgE levels in the sera of subjects from the high- (B) and low-resistance groups (A). The titration was carried out by ELISA on whole sera (Fig. 1): on
3.4 IgG that compete with IgE for the binding to sm Ag are detected in most sera
Previous work [27-291 has shown that specific IgG can compete with specific IgE for the binding to Ag when IgE
2682
Eur. J. Immunol. 1991.21: 2679-2686
€? W e t , C. E. Demeure, A. Bourgois et al.
Table 1. Infection intensities, anti-sm IgM and IgG levels and competition indexes in study groups before and after oxamniquine
treatment Study
2 months before treatment
15 months after treatment
43 months after treatment
P U P S
Infection intensity (egg/g)
A
632a)
(443 -903)b)
34
(17 -68)
**W)
B A
87
(45 -166)
5
(3
(395 -922)
***
603
-
26
(8
10.4
(5.2-* 20.5)
***
- 83)
4.34
(2.3-
7.9)
4.2
9) (2.6- 6.8)
B
7.0
(2.9- 17.0)
2.9
(1.4- 5.8)
3.7
Anti-sm IgG (U/ml)
A
18.4
(10.3- 32.9)
15.4
(10.5-22.6)
18.7
(11.1- 31.6)
19.4
(10.2- 37.1)
12.1
(7.6-19.3)
6.9
(4.1- 11.8)
Competition index
B A
2.8=)
(1.8-
4.2)
5.1
(3.2- 8.1)
4.7
(2.2-* 10.2)
B
1.8
(1.2-
2.6)
5.4
(2.6-11.5)
1.8
(1
Anti-sm IgM (U/ml)
(2.0-
7.0)
**
-
3.1)
a) Infection intensities were the geometric mean of egg counts (eggdg) in three stool samples taken on different days. Since the lower detection value on one determination is 24 eggdg (8 eggs/g for 3 determinations), subjects with 3 negative stool examinations were assigned an egg count value between 0 to 8 that was determined randomly. b) Numbers in parentheses indicate the 95% confidence interval. c) Comparison between group A and group B by Student’s t-test: ***(, < 0.OOOl); **(, < 0.01); *O,< 0.05). d) Geometric mean. e ) Geometric mean of the ratio IgE level afterlIgE level before the adsorption on protein A for each individual serum.
are titrated in whole sera.The present study confirms these observations:IgE reactivity was increased in most sera after removal of 80% to 96% IgG by adsorption on protein ASepharose. Addition of the eluates from protein A-Sepharose inhibited IgE detection in adsorbed sera (data not shown). Furthermore, when eluates were incubated with graded amounts of sm extract, the anti-sm IgG reactivity and the inhibitory effect of the eluates were specifically reduced in the same proportion (r = 0.99,p < 10-4).Competing Ab could not, however, account for the difference in IgE levels recorded between study groups in sera taken 2 months before and 15 months after treatment: most sera contained competing IgG, the average levels of competing Ab were not significantly different between the two study groups (lhble 1) and moreover, the differences in anti-sm IgE levels were conserved (p < 0.01) whether titrations were carried out on whole sera or on sera adsorbed on protein A. As could be expected from the results of IgG titration, the competing effect in sera taken at +43 months was higher (p < 0.05, Bible 1) in sera of the leat resistant group vs. the most resistant one and partially accounted for the differences in IgE levels between the two study groups observed on titrations of unfractionated sera.
one of the two study groups. Similar observations were made on sera taken 2 months before and 15 months after treatment (data not shown) indicating that the treatment produced no major changes in the pattern of Ag recognition. Adsorption of sera on proteinA improved Ag detection by IgE, confirming the competition of IgG on the binding of IgE to sm Ag (Fig. 2). As observed in ELISA, the binding of IgE to Ag was more affected by IgG from the least resistant subjects (lanes 1 , 3 , 7 , 11, 12, 13) than from the most resistant ones (lanes2, 10, 14, 16, 17, 18) at +43 months. Several sera failed to show detectable IgE reactivity for sm Ag even after removal of IgG on protein A. Adsorption of the -2 months and +15 months sera on protein A also enhanced the detection of Ag by IgE and, as observed in ELISA, sera from both groups were equally affected by this effect (data not shown). As previously discussed [9], IgG from these sera reacted with a large number of Ag on immunoblots of sm tegument extracts including those Ag that reacted with IgE. It should be noticed that the Ag P-37,which is strongly recognized by IgG in sera from the most resistant subjects [9], was also revealed on IgE immunoblots but only by 15% of the sera and only after long exposure times (Fig. 3).
3.5 Identification of sm Ag bound by IgE sm Ag that bind IgE were identified by reacting Western blots of sm tegument extracts with +43 months sera from both groups of subjects; IgE bound to blotted Ag were revealed with a *251-labeledanti-huIgE mAb. Numerous sm membrane Ag reacted with IgE (Fig. 2); the high-M, proteins (202 to 120 kDa) and a 85-kDa protein were the most prominent bands and reacted with IgE from the greater number of sera (Fig. 3). None of these Ag showed any preferential reactivity with IgE from the sera of either
3.6 The major sm tegument allergens (165,145, l35 and 1u)kDa) are cross-reacting Ag Human sera that recognized the 165-kDa protein on immunoblots also reacted with the 120-kDa protein and, although more weakly,with the 135- and 145-kDa proteins (Fig. 2). This suggested that these proteins might be antigenically related. To evaluate this hypothesis, the 165, 145, 135- and 120-kDa bands and a control 70-kDa band were cut off from the immunoblots; Ab bound to the
Eur. J. Immunol. 1991. 21: 2679-2686 1
2
3
I Adsorption on I
k,otein A
4
5
6
-7
8
9
IgE in human resistance to S. mansoni 10
11
12
13
14
15
16
17
2683 18 19
MW 205
116
68
45
29
Figure 2. Identification of sm allergens on Western blot. A blot (made after electrophoresis on a 7.5% polyacrylamide gel) of sm tegument extract was reacted either with sera adsorbed on protein A (+) or with whole sera (-); then, IgE bound to the blot were revealed with a 1251-labeledanti-huIgE mouse mAb. Sera were taken at +43 months from subjects with the highest (open box) or the lowest (closed box) resistance to infection.
nitrocellulose strips were eluted and reacted separately with fresh blots of sm tegument extract; blots were then revealed with lzI-labeled protein A. Ab eluted from any of these four bands (165,145,135 and 120 kDa) bound to all four Ag (Fig. 4), whereas Ab eluted from the 70-kDa band only bound to this Ag (data not shown). This result indicates that these four major sm Ag are probably cross-reactingAg. In addition, rabbit sera raised againstthe 165-kDa molecule, revealed, on blots of sm tegument extract, not only the 165-kDaAg but also the 145-, 135-and 120-kDa molecules (data not shown).
ANTIGEN Mr (kDa)
Figure 3. Percentage of immune sera reacting with each sm Ag fraction. Percentages were evaluated from the Western blot presented in Fig. 2, but after three different exposure times.
NHS
IHS
////
3
4
Figure 4. Four major high M, sm allergens are cross-reactingAg. A serum (IHS) from an infected individual was reacted with a blot of sm tegument extract. Ab were eluted from bands 1to 4 (Ag M,165, 145,135 and 120 kDa) and were reacted separately with the strips of a new blot of sm tegument extract, then, Ab bound to the strips were revealed with lZ5I-labeledprotein A. NHS: normal human serum.
2684
I? W e t , C.E. Demeure, A. Bourgois et al.
Eur. J. Immunol. 1991.21: 2679-2686
Figure5 Anti-sm I@ and anti-P-165 Ab bind to the surface of living sm. Fresh sm were reacted with either anti-P-165 rabbit serum (A) or myeloma huIgE (B) or IgE purified from the sera of infected subjects (C). Surface-bound Ab were revealed either with FITC-conjugated goat anti-rabbit IgG (A) or with anti-huIgE mouse mAb and FITC-conjugated goat anti-mouse Ig (B and C). Controls incubated with NRS were comparable to those shown in (B). 9 mm = 20 pm.
3.7 IgE-binding determinants are accessible to Ab on living schistosomula To determine whether some of the tegumental Ag that bind IgE were on the outer surface of the parasite and accessible to IgE-mediated cell cytotoxicity, fresh living sm were reacted with either the anti-P-165 rabbit antiserum or with the IgE fraction from the sera of infected subjects. Unbound Ab were washed out and bound rabbit IgG or huIgE were revealed by immunofluorescence. Results presented in Fig. 5 show that a strong fluorescence signal was observed in both cases, while sm reacted with normal rabbit serum (NRS) or irrelevant huIgE were negative. Thus, IgE in immune human sera bind determinants on the outer part of the sm membrane, some of these determinants may be borne by the 120-165 cross-reactive family. This latter conclusion is further supported by immunofluorescence studies with mAb specific of this Ag family (Carvalho, D. et al., unpublished observations). However, the definitive demonstrationthat these Ag are on the larval surface requires biochemical studies that are being carried out.
induction of 1gE.Thisview is also consistent with the results of a study on Brugia malayi Ag [31] and with the demonstration that infection by the nematode N. brusiliensis stimulates the synthesis of IgE against irrelevant Ag administeredjust before the infection [32,33]. From recent studies in mice on Thl and Th2 lymphocytes and on the signals required for Ag presentation, it can be speculated that the strong stimulation of the IgE response caused by helminths might be related to the route of entry of the parasites which may select for certain APC, or to the induction by the parasite of co-stimulatory signals for a Th2-like response [34,35]. In this regard, the high-M, Ag may contribute to the induction of this response and may have specific properties that enable them to stimulate IgE production. Conversely, the high IgE response to the high-M, Ag may be the mere consequence of their strong immunogenicity. This high immunogenicity was first noticed on immunoblots reacted with sera from infected subjects and revealed with an IgG-specific probe [9].
Immunofluorescence studies have shown that IgE from infected subjects bound to the outer part of the tegumental membrane of the living parasites indicating that certain allergens can be targets of IgE-dependent cell-mediated cytotoxicity [lo-161. These Ag, together with excreted 4 Discussion products [36, 371, may also be responsible for the triggering This work has evaluated the anti-sm IgE response of young of mast cell degranulation allowing the early recruitment residents of an endemic area of S. mansoni whose suscep- and activation of helminthotoxiceffector cells at the site of tibility to infection had been carefully determined in penetration of the parasite. The multiplicity of the sm Ag previous studies [9, 301. The objectives were to identify the that bind IgE and the presence of certain of these Ag on the principal targets of IgE on sm, to determine whether the surface of the larvae are evidence for a role of IgE in human IgE responses of most resistant subjectswere different from protection against S. mansoni. This view is also supported those of the least resistant ones, and to evaluate the effects by the finding that IgE levels in the most resistant subjects of a parasitological cure with oxamniquine on these were, on average, sevenfold higher than those of the least resistant ones. This observation is consistent with a recent responses. report showing that subjectswith high levels of anti-sm IgE Immunoblots of sm tegument extracts showed that sera are more resistant to reinfection by S. haematobium [38]. from most study subjects contained IgE that bound to Buttenvorth et al. [8] analyzing the IgE Ab response of various sm Ag. The pattern of Ag recognized by IgE subjects exposed to infection by S. mansoni observed a differed from one subject to the other and altogether, IgE trend, although not significant, for higher IgE levels in the from the different sera reacted with a large number of sm most resistant subjects as compared to the least resistant Ag. This makes it unlikely that all these Ag have unusual ones; differences in the sensitivity of the assays and biochemical or structural properties accounting for the differences in the criterions used to define resistant and
Eur. J. Immunol. 1991. 21: 2679-2686
IgE in human resistance to S. rnansoni
2685
susceptible groups may account for the discrepancy the membrane, indicating putative targets of IgE-depenbetween these and our results. Nevertheless, in our study, dent cell-mediated cytotoxicity reactions on the invading one subject with high resistance had low IgE levels and larvae. Third, IgE levels were higher in the most resistant conversely, several subjects among the least resistant, had subjects than in the least resistant ones. These observations medium to high IgE levels.This indicates that a strong IgE altogether are consistent with the view that IgE may play a response is probably neither always sufficient nor, in some significant role in human immune defenses against S.mancases, necessary to allow high protection. Such a result was soni infections. expected since human immune defences against helminths are thought to involve a number of protective mechanisms which differ in relative importance for protection from one We wish to thank Dr. M . Delaage (Imrnunotech, Marseille, France) for the gift ofthe anti-human IgE mouse m A b and the acetylcholine individual to another. esterase anti-human IgE mAb.
In contrast to IgE, anti-sm IgM and anti-sm IgG levels were not significantly different between study groups 2 months before and 15 months after treatment and even increased with reinfection intensities 43 months after treatment. Consistentwith our findings,lower levels of IgM Ab against a p38 larval Ag were observed in subjects with the highest resistance to infection in the Kenyan study [39]. Since IgE and IgG were able to bind to the same Ag, and probably to the same or spatially related epitopes, it was expected that IgG would compete with IgE for the binding to Ag in ELISA and on Western blots. High levels of competing Ab were detected in most subjects regardless of their levels of resistance, and competing Ab levels were not, on average, different between the two study groups before treatment and 15 months later. However, studies on subjects with atopy [40,41] or with helminth infections [42,43], have suggested that anaphylactic reactions may be regulated by such inhibiting Ab. Competing Ab in sera taken 43 months after treatment were on average higher in the least resistant group than in the most resistant one, although unsufficiently to account for the differencesin IgE titers between the two study groups.This result is corroborated by the higher titers of specific IgG in the least resistant group. In a more recent study (Demeure, C. et al., to be published) we have obtained evidence indicating that the competing Ab are IgG4. IgG4 were also measured in the Gambian study and higher IgG4 levels were recorded in the least resistant subjects [38]. Our data show that from +15 to t-43 months, a significant difference in IgG competing Ab takes place between the two groups as the most resistant group resists reinfection, while the least resistant one develops high infections.Whether these competing Ab modulate the anti-sm IgE-dependent immunity in vivo remains to be evaluated. This study does not reveal a major effect of oxamniquine treatment on either the levels or specificitiesof anti-sm IgE in patients’ sera. This observation does not rule out a transient stimulation of IgE within few months after treatment as reported in severalstudies [25,26] .This lack of a major effect of the treatment on IgE levels and specificities is consistent with our observation that treatment does not modify susceptibility to infection of these subjects [91. In conclusion,this work reveals several interestingfeatures of the anti-sm IgE response. First, IgE levels remained relatively steady during the whole study while subjects experienced large variations in infection intensities, indicating that aborted or low infections are sufficient to maintain high IgE levels. Second, IgE binds to a large variety of sm Ag, some of them being on the outer part of
Received March 27, 1991; in revised form June 12, 1991.
5 References 1 UNDFlWorld BanWHO, in WHO (Ed.), Tropical Disease Research, Seventh Programm Report, Barthelemy, A . , 1983, p. 313. 2 Butterworth, A. E.,Wilkins, H. A., Capron, A . and Sher, A., Parasitol. Today 1987. 3: 1. 3 Fenwick, A., Parasitol. Today 1987. 3: 70. 4 Klumpp, R. K. and Chu, K.Y.,Parasitol. Today 1987. 3: 74. 5 Prata, A., Bailiere’s Clinic. Trop. Med. Comrn. Dis.1987. 2: 349. 6 Wilkins, H. A., Parasitol. Today 1989. 5: 83. 7 Katz, N., reported by Von Lichtenberg,F., Am. J. Trop. Med. Hyg. 1985. 34: 78. 8 Butterworth, A. E., Capron, M., Cordingley, J. S., Dalton, F! R., Dunne, D. W., Kariuki, H. C., Kimani, G., Koech, D., Mugambi, M., Ouma, J. H., Prentice, M. A., Richardson, B. A., Arap Siongok, T. K., Sturrock, R. F. and Taylor, W. D., Trans. R. SOC. Trop. Med. Hyg. 1985. 78: 393. 9 Dessein, A. J., Begley, M., Demeure, C., Caillol, D., Fueri, J., Galvao dos Reis, M.,Andrade, Z . A., Prata, A. and Bina, J., J. Irnrnunol. 1988.140: 2727. 10 Capron, A., Dessaint, J. €?, Capron, M. andBazin, H., Nature 1975.253: 474. 11 Capron, A., Dessaint, J. I?, Joseph, M., Rousseaux, R., Capron, M. and Bazin, A., Eur. J. Immunol. 1977. 7: 315. 12 Capron, M., Bazin, H., Joseph, M. and Capron, A., J. Immunol. 1981. 126: 1764. 13 Kigoni, E. F!,Elsas, F! X., Lenzi, H. L. and Dessein, A. J., Eur. J. Irnmunol. 1986. 16: 589. 14 Joseph, M., Capron, A . , Butterworth, A . E., Sturrock, R. F. and Houba,V., Clin. Exp. Irnrnunol. 1978. 33: 48. 15 Joseph, M., Auriault, C., Capron, A.,Vorng, H. and Viens, F!, Nature 1983. 303: 810. 16 Capron, M., Spiegelberg, H. L., Prin, L., Bennich, H., Butterworth, A. E., Pierce, R. J., Ouaissi, M. A. and Capron, A., J. Immunol. 1984.132: 462. 17 Bina, J. C. and Prata, A., in Azvedo, E. S., Rocha, H. and Andrade, Z . , (Eds.), Aspectos peculiares da infecgao por Schistosoma mansoni. Cedre Publication, Bahia 1984, p. 13. 18 Dessein, A. J., Couissinier, E , Rihet, F!, Demeure, C. E., Oatara, M. B.,Carvalho, D., Kohlstiidt, S., Goudot-Crozel,V., Bourgois, A. and Abel, L., Clin. Immunol. Allergy Rev. 1991, in press. 19 Anfosso, F., Demeure, C., Delaage, M., Cheballah,R., Bellot, F. and Bourgois, A., Mol. Immunol. 1987. 24: 1129. 20 Ramalho-Pinto, F. S., Gazinelli, G., Howells, G. E., MotaSantos,T. A., Figueredo,E. A. and Pellegrino,Exp. Parasitol. 1974. 36: 360. 21 Lazdins, J. K., Stein, M. J., David, J. R. and Sher, A . , Exp. Parasitol. 1982. 53: 39. 22 Towbin, H., Staehlin,T. and Gordon, J. C., Proc. Natl. Acad. Sci. USA 1979. 76: 4350. 23 Engvall, E. and Perlmann, F!, J. Imrnunol. 1972. 109: 129.
2686
F! Rihet, C. E. Demeure, A. Bourgois et al.
24 Lindmark, R.,Yhoren-Tolling, K. and Sjijsvist, J., J. Immunol. Methods 1983. 62: 1. 25 Mendes, E. and Mendes, N. F., Trop. Geogr. Med. 1982. 34: 133. 26 EvengArd, B., Hammarstrom, L., Smith, C. I. E., Johansson, S. G. 0. and Linder, A., Clin. Exp. Immunol. 1988. 73: 383. 27 Cooke, R.A., Barnard, J. H., Hebald, S. and Stull,A., J. Exp. Med. 1935. 62: 733. 28 Lichtenstein, L. M., Holtzman, N. A. and Burnett, L. S., J. Immunol. 1968. 101: 317. 29 Gleich, G. J., Zimmermann, E. M., Gleich, M. C. and Yunginger, J. W., J. Immunol. 1981. 126: 575. 30 Abel, L., Demenais, F., Prata, A., Souza, A. E. and Dessein, A. J., Am. J. Hum. Genet. 1991. 48: 959. 31 Hussain, R. and Ottessen, E. A., J. Immunol. 1985. 135: 1415. 32 Orr,T. S. C. and Blair, A. M. J. N., Life Sci. 1969. 8: 1073. 33 Jarrett, E. E. E., Immunology 1972.22: 1099. 34 Janeway, C., Immunol. Today 1989. 10: 283. 35 Weaver, C. T. and Unanue, E. R., Immunol. Today 1990. 11: 49.
Eur. J. Immunol. 1991. 21: 2679-2686 36 Auriault, C., Darnonneville, M., Verwaede, C., Pierce, R., Joseph, M., Capron, M. and Capron, A., Eur. J. Immunol. 1984.14: 132. 37 Velge-Roussel, F. ,Verwaede, C., Grzych, S. M., Auriault, A. and Capron, A., J. Immunol. 1989.142: 2527. 38 Hagan, P.,Blumenthal, U. J., Dunn, D., Simpson, A. J. G. and Wilkins, H. A., Nature 1991. 349: 243. 39 Khalife, J., Capron, M., Capron, A., Gnych, J.-M., Butterworth, A. E., Dunne, D. W. and Ouma, J. H., J. Exp. Med. 1986. 164: 1626. 40 Lessof, M. H., Kagey-Sobokta, A. and Lichtenstein, A. M., John Hopkins Med. 1. 1978. 142: 1. 41 Hunt, K. J.,Valentine, M. D., Kagey-Sobokta, A., Benton, A. W., Amodio, F. J. and Lichtenstein, L. M., N. Engl. J. Med. 1978. 299: 157. 42 Ottessen, E. A., Kumaraswami,V.,Paranjade, R., Poindexter, R. W. and Tirpathy, S. I?, J. Immunol. 1981. 127: 2014. 43 Hoffstetter, M., Poindexter, R. W., Ruiz-Tiben, E. and Ottessen, E. A., Immunology 1982. 46: 777.
Eur. J. Immunol. 1991. 21: 2679-2686
2679
Pascal Rihet, Christian E. Demeure., Main Bourgois, Aluizio Prata* and Main J. Dessein
Evidence for an association between human resistance to Schistosoma mansmi and high anti-larval IgE level@
Centre d'Immunologie
The anti-larval IgE antibody response of adolescents with high or low resistance to infection by Schistosoma mansoni was evaluated before parasitological cure with oxamniquine and over an extended post-treatment period during which the least resistant subjects regained high infections. IgE from most sera, taken at several bleeding times before and after treatment, reacted, on immunoblots,with a large number of antigens (Ag) in schistosomular tegument extract. A family of 120-165-kDa cross-reactingmolecules and a 85-kDa Ag were the most prominent Ag. Some of these determinants were shown to be located on the outer tegumental membrane and to be accessible to IgE on living larvae. The comparison of IgE between the two study groups showed that IgE levels were on average six- to eightfold higher (p CO.01) in the sera of the most resistant adolescents whereas there was no difference in patterns of Ag recognition between study groups. In contrast to IgE, anti-larval IgG and IgM levels were either similar in both groups or higher in the least resistant subjects when these exhibited high reinfection levels. IgG that competed for the binding of IgE to larval Ag were detected in most sera and their levels were higher in the Ieast resistant group after reinfection. FinaHy, the treatment had no observable long-lasting effects on the levels and on the specificity of the anti-larval IgE. Altogether, these observations can be taken as evidence supporting a role of I@ in human resistance to infection by S. mansoni.
INSERMXNRS de Marseille-Luminy and Facaldade de Medicha do lkiangulo Mineboa, Uberaba
1 Introduction
ent of animal [lo-131 and human [14-161 protective immunity against S. mamoni, we began to evaluate IgEdependent immunity in subjects with high resistance to infection. The objectives of the present work were: (a) to identify the principal schistowmula (sm) allergens; (b) to evaluate whether lower resistance levels were associated with qualitative and/or quantitative alterations of the anti-larval IgE response; and (c) to determine the effects of a parasitological cure with oxamniquine on this response.
Schistosomiasisdue to S. mansoni is endemic in large areas of subtropical countries where it still represents a serious health problem and a burden for the economy [l]. The discovery of potent and safe schistosomicids allowed a reduction of the morbidity associated with infection by this parasite. However, the control of schistosomiasiscannot be achieved by chemotherapy alone and potent, cheap and safe vaccines are needed [2-61. In endemic areas, certain individuals have no sign of active infection in spite of daily contacts with infected waters, indicating that these individuals exhibit a high level of immunity to S. mamoni [7-91. 2 Materials and methods Analysis of the immune response of individuals with different levels of resistance has been undertaken to 2.1 Study subjects understand the principal mechanisms of resistance and to identify their targets on the parasite. Since a number of Thirty-eight young individuals (age 10 to 19years) were observations indicated that IgE could be a critical compon- selected for similar and high water contacts among subjects of an area endemic for S. munsoni.The area, the subjects of this study and the evaluation of their water contact levels have been described in detail previously [9,17,18]. All 38 [I 94121 subjects were given oxamniquine 2 months after the beginning of this study and none of them excreted S. munsoni * This work received financial assistance from INSERM,CNRS, eggs for 3 months following treatment. lkenty of them the Rockefeller Foundation and the UNDPrWorld BanWWHO A, Table 1) were reinfected within 15 to 27 months (group Special Program for Research and Training inllopical diseases. This work was first published as an abstract in the proceedings of after treatment and progressively developed heavy infecthe 7th International Congress of Immunology in Berlin, tions; these individuals, who had also heavy infections 1989. before treatment, are referredto as subjectswith the lowest Supported by a fellowship from CNRS and the Provence Apes resistance to infection. In contrast, 18individuds (group B, a t e d'Azur Council. Bble 1) excreted no or few eggs for a 43-month period following treatment and had low infections before treatCorrespondence: Main J. Dessein, Centre d'Immunologie de ment; these subjects are referred to as individuals with the Marseille-Luminy, Campus de Luminy, Case 906, F-13288 Marhighest resistance to infection [9]. Since it has not been seille Cedex 09,France possible to bleed all 38 subjects at all study times, the sizes Abbreddons: m: Schistommula NRS: Normal rabbit se- of study groups at the different times are shown on Fig. 1. rum 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991
0014-2980/91/1111-2679$3.50+ . Z / O
2680
Eur. J. Immunol. 1991.21: 2679-2686
I? Wet, C. E. Demeure, A. Bourgois et al.
2.2 Chemicals
2 6 Elution of Ab bound to blotted Ag
MEM, Hepes, penicillin and streptomycin were purchased from Gibco (Paisley, Scotland); n-octyl-glucoside(l-0n-octyl-fl-D-glucopyranoside) , pepstatin, PMSF, Chaps, a2-macroglobulin, leupeptin, 2-ME, acetyl-thiocholin iodide and dithionitrobenzoic acid were from BoehringerMannheim Biochemicals (Mannheim, FRG); aprotinin, protein A, bovine hemoglobin, BSA, Tween 20, Ponceau red stain, Mr markers and p-nitrophenyl-phosphate were from Sigma Chemicals Co.(St. Louis, MO).The El24 and 6H10 anti-human (hu)IgE mouse mAb [191, acetylcholine esterase-coupled anti-huIgE mouse mAb, FITC-conjugated goat anti-rabbit IgG and FITC-conjugated goat anti-mouse Ig were from Immunotech (Marseille, France). Alkalinephosphatase-coupledanti-huIgG mouse mAb and anti-huIgM mouse mAb were purchased from ICN K & K Laboratories (Eschwege, FRG). Protein A-coupled Sepharose cG4B and cyanogen bromide-activated Sepharose were purchased from Pharmacia Fine Chemicals (Uppsala, Sweden).
Ag blotted on nitrocellulosepaper were reacted with a pool of human immune sera. Bands of interest were identified as in [9] and were cut out from the other part of the blot. Ab bound to these strips were eluted with a 0.2 M glycine/HCl buffer, pH 2.8, and eluates were adjusted to pH 7.4.
2.3 Schistowmula The Puerto-Rican strain of S. mansoni used in this study was maintained by passage through CBAIJ mice and Biomphalaria glabrata snails. Schistosomulawere prepared by the method of Ramalho-Pinto et al. [20], as modified by Lazdins et al. [21], and cultured (3 h, 37°C) in MEM supplemented with 10 m~ Hepes, pH 7.2,100 I U / d penicillin and 100 pg/ml streptomycin.
2.7 Ab titration Titrations were performed by ELISA [23] on plates coated with 100 pl of a 5 p g / d solution of sm tegument extract in 50 m~ sodium carbonate buffer, pH 9.6. Ag-coated plates were saturated with PBSBSA (30 g/l BSA) and reacted with serum dilutions (1 : 10 to 1:270 for IgE and 1:100 to 1:5000for IgG and IgM) for 16 h at 4 "C.Then, plates were washed three times with 0.05% Tween20 in PBS and reacted for 3 h at room temperature with either acetylcholine esterase-coupled anti-huIgE mAb or alkaline phosphatase-coupled anti-huIgG mAb. After three additional washes, enzymatic activities were revealed by addition of either 0.5 m~ acetylthiocholine iodide and 0.5 m~ dithionitrobenzoic acid in 50 mM phosphate buffer, pH 7.6 (for acetylcholine esterase) or 1mglml p-nitrophenyl-phosphate in 10% diethanolamine buffer, pH 9.6 (for alkaline phosphatase). The reactions were read at 405 nm. Positive and negative reference sera were used to draw a standard curve. Results were calculated by inference from this standard curve and expressed in U/ml. By definition, lOOU/ml is the amount of specific IgE, IgG or IgM measured in either a 1/10 (IgE) or a 1:lOO (IgG, IgM) dilution of the most reactive serum. 2.8 Evaluation of the competition effect
2.4 Tegument extraction Schistosomular tegument Ag were extracted by resuspending sm (l@/ml) in a 1% n-octylglucoside solution containing 10 mM sodium phosphate buffer, pH 7.1,4 mM MgC12, 140 m~ NaCl, to which the following protease inhibitors were added just before use: 50 p~ PMSF, 50 p g / d aprotinin, 1p~ pepstatin, 20 pg/ml leupeptin, and 10 pg/ml a2-macroglobulin. After 15 min shaking at 4°C in extraction buffer, sm bodies were pelleted (400 X g, 3 min, 4"C), and the SN were centrifuged at 100OOO x g for 35 min at 4°C. SN were stored aliquoted at -70°C until use.
Sera, diluted 1:2 in buffer A (0.2% n-ocytylglucoside, 0.1% Chaps in PBS), were centrifuged (1Omin; 100OOO x g) and the SN were reacted with protein ASepharose (90min, 4°C). Unbound material was collected in the first two washes of the beads with buffer A. Ig bound to protein A were eluted with a 0.1 M glycine buffer, pH 2.8 [24] and pH of the eluates were adjusted immediately to 7.4. Filtrates and eluates were then, dialyzed against PBS (4"C, 48 h) and anti-sm IgE Ab in these fractions were measured by ELISA. The competition exerted by IgG on the binding of IgE to sm Ag was expressed as the ratio of specific IgE levels (U/ml) in protein A-adsorbed fractions to IgE levels (U/ml) measured on the respective unfractionated sera.
2.5 Western blotting
Western blots were carried out following the method of Towbin et al. [22] modified as in 191 excepted that Ig bound to the blotted Ag were revealed with the El24 1251-labeled anti-huIgE mAb (2 x lo6 cpm/ml). This mAb binds to the heat-sensitive De2 IgE antigenic region [19]. Heating immune human sera (56"C, 3 h) before reacting them with the blots extinguished all bands (data not shown). M, were estimated by comparison to the following markers: myosin (205 m a ) , @-galactosidase(116 m a ) , phosphorylase b (97.4 kDa), BSA (68 m a ) , OVA (45 kDa) and carbonic anhydrase (29 kDa).
2.9 Preparation of anti-P165 rabbit sera
sm tegument extracts, containing 1mg of proteinlml,were heated (lOO"C, 3 min) in 50 mM Tris buffer, pH6.8, containing 1% SDS, 5% 2 - a , 1 mM EDTA, 10% (v/v) glycerol, and subjected to electrophoresis on a 7.5% SDS-polyacrylamidegel.The 165-kDaband was identified, cut off from the gel and its protein content was electroeluted as in [9]. Two rabbits were primed with 50 pg of the electroelutedP165preparation in CFA and boosted twice at 2-month intervals with 50 pg of the same material in F A .
Eur. J. Immunol. 1991.21: 2679-2686
IgE in human resistance to S.rnunsoni
2681
I
Sera were collected beginning 2 weeks after the last boost. 2.10 Immunopuri6cation of hnIgE The 6H10 [19] anti-huIgE mAb (1 mg) was coupled to 0.6 g of CNBr-activated Sepharose4B. Then, 200ml of the immunosorbent was reacted (2 h at room temperature) with 1.5 ml of a pool of sera from five infected subjects. After washing with 10 vol. of PBS, 1%BSA and 2 vol. of 0.1 M phosphate buffer, pH 9.2,1% BSA, IgE were eluted with a 0.1 M triethylamine solution, pH 11.5, containing 1% BSA.
10
-
0 0
;i
2.ll Immuno5uorescence studies Fresh sm washed in cold RPMI/FCS medium (WMI 1640 containing 10% FCS, 10 mM Hepes, pH 7.2, 100 IU/ml penicillin and 100 p g / d streptomycin),were reacted for 2 h at 4°C either with 1 / 1 0 dilution of the rabbit anti-P165 serum or 2 pg/ml of immunopurified IgE from a pool of sera from infected subjects. A 1/100 dilution of NRS and 2 pg/ml of myeloma huIgE were used as controls. Then, sm were washed three times in WMIiFCS and reacted (30 min, 4°C) either with FITC-conjugated goat anti-rabbit IgG or with a 100 p g / d solution of anti-huIgE mAb (E124) followed by three washes and an incubation (30 min, 4 "C) with FITC-conjugated goat anti-mouse Ig. Then, sm were washed three times in RPMIiFCS and examined under a fluorescence microscope at a X 100 magnification.
1-
0
o
m
0
0
0
-2 mo
+15 mo
+43 mo
pt0.005
pcO.005
ptO.O1
Figurel. Anti-sm IgE levels are higher in the most resistant adolescents as compared to the least resistant ones. Anti-sm IgE levels were assessed by ELISA on 1 :10 serum dilutions. Comparison of the two groups was performed using Student's t-test after logarithmic normalization of the data. Geometric means and SEM are drawn. Geometric means were 20.8 and 3.2Ulml at -2 months, 28.4 and 3.8 U/ml at +15 months and 26.7 and 4.4 U/ml at +43 months with sera from subjects with the highest (0)and lowest (0)resistance, respectively.
average, anti-sm IgE levels were higher in the sera of the most resistant subjects as compared to the least resistant ones at all study times, and there was a 6.5- to 7.6-fold 3.1 Effect of oxamniquine treatment on the anti-sm IgE difference in specific IgE levels between the two study response of study subjects groups. IgE levels in sera of group B were rather homogenous: all but one sera in this group exhibited a good Anti-sm IgE levels were measured by ELISA in sera taken reactivity in this assay; in contrast IgE levels in sera of from subjects 2 months before oxamniquine treatment as group A were heterogeneous with certain sera exhibiting well as 15 and 43months after. These measurements no or little reactivity, whereas others reacted well in the showed that no major changes in IgE levels occurred after assay. treatment. First, anti-sm IgE levels measured in sera taken 2 months before and 15 months after treatment as well as in sera taken 43 months and 15 months after treatment were 3.3 Anti-sm IgG and IgM levels are not increased in and T = 0.96, closely correlated (I = 0.95, p < subjects with the hig4est resistance to infection p< respectively); this rules out a selective effect of the treatment on IgE levels of certain subjects. Second, IgE To assess whether this observation of higher IgE levels in levels were comparable at various times of the study, as the most resistant subjects was sblective for IgE or common indicated by the slopes of the regression lines of IgE to other isotypes, anti-sm IgG and IgM were evaluated in (-2 months) to IgE (+15 months) and IgE (+43 months) the same sera. In contrast to the observations made on IgE, to IgE (+15 months) as these slopes are close to 1. This the levels of specific IgG and IgM were not significantly confirms that the increase of IgE levels observed in several different in the sera of the two study groups at -2 months studies, within few weeks after treatment [25,26], is a and +15 months; moreover, at +43 months, IgG and IgM transient phenomenon. were lower in the sera of the most resistant group (Table 1).
3 Results
3.2 Antiam IgE levels are higher m the sera of the most resistant sobjects We compared anti-sm IgE levels in the sera of subjects from the high- (B) and low-resistance groups (A). The titration was carried out by ELISA on whole sera (Fig. 1): on
3.4 IgG that compete with IgE for the binding to sm Ag are detected in most sera
Previous work [27-291 has shown that specific IgG can compete with specific IgE for the binding to Ag when IgE
2682
Eur. J. Immunol. 1991.21: 2679-2686
€? W e t , C. E. Demeure, A. Bourgois et al.
Table 1. Infection intensities, anti-sm IgM and IgG levels and competition indexes in study groups before and after oxamniquine
treatment Study
2 months before treatment
15 months after treatment
43 months after treatment
P U P S
Infection intensity (egg/g)
A
632a)
(443 -903)b)
34
(17 -68)
**W)
B A
87
(45 -166)
5
(3
(395 -922)
***
603
-
26
(8
10.4
(5.2-* 20.5)
***
- 83)
4.34
(2.3-
7.9)
4.2
9) (2.6- 6.8)
B
7.0
(2.9- 17.0)
2.9
(1.4- 5.8)
3.7
Anti-sm IgG (U/ml)
A
18.4
(10.3- 32.9)
15.4
(10.5-22.6)
18.7
(11.1- 31.6)
19.4
(10.2- 37.1)
12.1
(7.6-19.3)
6.9
(4.1- 11.8)
Competition index
B A
2.8=)
(1.8-
4.2)
5.1
(3.2- 8.1)
4.7
(2.2-* 10.2)
B
1.8
(1.2-
2.6)
5.4
(2.6-11.5)
1.8
(1
Anti-sm IgM (U/ml)
(2.0-
7.0)
**
-
3.1)
a) Infection intensities were the geometric mean of egg counts (eggdg) in three stool samples taken on different days. Since the lower detection value on one determination is 24 eggdg (8 eggs/g for 3 determinations), subjects with 3 negative stool examinations were assigned an egg count value between 0 to 8 that was determined randomly. b) Numbers in parentheses indicate the 95% confidence interval. c) Comparison between group A and group B by Student’s t-test: ***(, < 0.OOOl); **(, < 0.01); *O,< 0.05). d) Geometric mean. e ) Geometric mean of the ratio IgE level afterlIgE level before the adsorption on protein A for each individual serum.
are titrated in whole sera.The present study confirms these observations:IgE reactivity was increased in most sera after removal of 80% to 96% IgG by adsorption on protein ASepharose. Addition of the eluates from protein A-Sepharose inhibited IgE detection in adsorbed sera (data not shown). Furthermore, when eluates were incubated with graded amounts of sm extract, the anti-sm IgG reactivity and the inhibitory effect of the eluates were specifically reduced in the same proportion (r = 0.99,p < 10-4).Competing Ab could not, however, account for the difference in IgE levels recorded between study groups in sera taken 2 months before and 15 months after treatment: most sera contained competing IgG, the average levels of competing Ab were not significantly different between the two study groups (lhble 1) and moreover, the differences in anti-sm IgE levels were conserved (p < 0.01) whether titrations were carried out on whole sera or on sera adsorbed on protein A. As could be expected from the results of IgG titration, the competing effect in sera taken at +43 months was higher (p < 0.05, Bible 1) in sera of the leat resistant group vs. the most resistant one and partially accounted for the differences in IgE levels between the two study groups observed on titrations of unfractionated sera.
one of the two study groups. Similar observations were made on sera taken 2 months before and 15 months after treatment (data not shown) indicating that the treatment produced no major changes in the pattern of Ag recognition. Adsorption of sera on proteinA improved Ag detection by IgE, confirming the competition of IgG on the binding of IgE to sm Ag (Fig. 2). As observed in ELISA, the binding of IgE to Ag was more affected by IgG from the least resistant subjects (lanes 1 , 3 , 7 , 11, 12, 13) than from the most resistant ones (lanes2, 10, 14, 16, 17, 18) at +43 months. Several sera failed to show detectable IgE reactivity for sm Ag even after removal of IgG on protein A. Adsorption of the -2 months and +15 months sera on protein A also enhanced the detection of Ag by IgE and, as observed in ELISA, sera from both groups were equally affected by this effect (data not shown). As previously discussed [9], IgG from these sera reacted with a large number of Ag on immunoblots of sm tegument extracts including those Ag that reacted with IgE. It should be noticed that the Ag P-37,which is strongly recognized by IgG in sera from the most resistant subjects [9], was also revealed on IgE immunoblots but only by 15% of the sera and only after long exposure times (Fig. 3).
3.5 Identification of sm Ag bound by IgE sm Ag that bind IgE were identified by reacting Western blots of sm tegument extracts with +43 months sera from both groups of subjects; IgE bound to blotted Ag were revealed with a *251-labeledanti-huIgE mAb. Numerous sm membrane Ag reacted with IgE (Fig. 2); the high-M, proteins (202 to 120 kDa) and a 85-kDa protein were the most prominent bands and reacted with IgE from the greater number of sera (Fig. 3). None of these Ag showed any preferential reactivity with IgE from the sera of either
3.6 The major sm tegument allergens (165,145, l35 and 1u)kDa) are cross-reacting Ag Human sera that recognized the 165-kDa protein on immunoblots also reacted with the 120-kDa protein and, although more weakly,with the 135- and 145-kDa proteins (Fig. 2). This suggested that these proteins might be antigenically related. To evaluate this hypothesis, the 165, 145, 135- and 120-kDa bands and a control 70-kDa band were cut off from the immunoblots; Ab bound to the
Eur. J. Immunol. 1991. 21: 2679-2686 1
2
3
I Adsorption on I
k,otein A
4
5
6
-7
8
9
IgE in human resistance to S. mansoni 10
11
12
13
14
15
16
17
2683 18 19
MW 205
116
68
45
29
Figure 2. Identification of sm allergens on Western blot. A blot (made after electrophoresis on a 7.5% polyacrylamide gel) of sm tegument extract was reacted either with sera adsorbed on protein A (+) or with whole sera (-); then, IgE bound to the blot were revealed with a 1251-labeledanti-huIgE mouse mAb. Sera were taken at +43 months from subjects with the highest (open box) or the lowest (closed box) resistance to infection.
nitrocellulose strips were eluted and reacted separately with fresh blots of sm tegument extract; blots were then revealed with lzI-labeled protein A. Ab eluted from any of these four bands (165,145,135 and 120 kDa) bound to all four Ag (Fig. 4), whereas Ab eluted from the 70-kDa band only bound to this Ag (data not shown). This result indicates that these four major sm Ag are probably cross-reactingAg. In addition, rabbit sera raised againstthe 165-kDa molecule, revealed, on blots of sm tegument extract, not only the 165-kDaAg but also the 145-, 135-and 120-kDa molecules (data not shown).
ANTIGEN Mr (kDa)
Figure 3. Percentage of immune sera reacting with each sm Ag fraction. Percentages were evaluated from the Western blot presented in Fig. 2, but after three different exposure times.
NHS
IHS
////
3
4
Figure 4. Four major high M, sm allergens are cross-reactingAg. A serum (IHS) from an infected individual was reacted with a blot of sm tegument extract. Ab were eluted from bands 1to 4 (Ag M,165, 145,135 and 120 kDa) and were reacted separately with the strips of a new blot of sm tegument extract, then, Ab bound to the strips were revealed with lZ5I-labeledprotein A. NHS: normal human serum.
2684
I? W e t , C.E. Demeure, A. Bourgois et al.
Eur. J. Immunol. 1991.21: 2679-2686
Figure5 Anti-sm I@ and anti-P-165 Ab bind to the surface of living sm. Fresh sm were reacted with either anti-P-165 rabbit serum (A) or myeloma huIgE (B) or IgE purified from the sera of infected subjects (C). Surface-bound Ab were revealed either with FITC-conjugated goat anti-rabbit IgG (A) or with anti-huIgE mouse mAb and FITC-conjugated goat anti-mouse Ig (B and C). Controls incubated with NRS were comparable to those shown in (B). 9 mm = 20 pm.
3.7 IgE-binding determinants are accessible to Ab on living schistosomula To determine whether some of the tegumental Ag that bind IgE were on the outer surface of the parasite and accessible to IgE-mediated cell cytotoxicity, fresh living sm were reacted with either the anti-P-165 rabbit antiserum or with the IgE fraction from the sera of infected subjects. Unbound Ab were washed out and bound rabbit IgG or huIgE were revealed by immunofluorescence. Results presented in Fig. 5 show that a strong fluorescence signal was observed in both cases, while sm reacted with normal rabbit serum (NRS) or irrelevant huIgE were negative. Thus, IgE in immune human sera bind determinants on the outer part of the sm membrane, some of these determinants may be borne by the 120-165 cross-reactive family. This latter conclusion is further supported by immunofluorescence studies with mAb specific of this Ag family (Carvalho, D. et al., unpublished observations). However, the definitive demonstrationthat these Ag are on the larval surface requires biochemical studies that are being carried out.
induction of 1gE.Thisview is also consistent with the results of a study on Brugia malayi Ag [31] and with the demonstration that infection by the nematode N. brusiliensis stimulates the synthesis of IgE against irrelevant Ag administeredjust before the infection [32,33]. From recent studies in mice on Thl and Th2 lymphocytes and on the signals required for Ag presentation, it can be speculated that the strong stimulation of the IgE response caused by helminths might be related to the route of entry of the parasites which may select for certain APC, or to the induction by the parasite of co-stimulatory signals for a Th2-like response [34,35]. In this regard, the high-M, Ag may contribute to the induction of this response and may have specific properties that enable them to stimulate IgE production. Conversely, the high IgE response to the high-M, Ag may be the mere consequence of their strong immunogenicity. This high immunogenicity was first noticed on immunoblots reacted with sera from infected subjects and revealed with an IgG-specific probe [9].
Immunofluorescence studies have shown that IgE from infected subjects bound to the outer part of the tegumental membrane of the living parasites indicating that certain allergens can be targets of IgE-dependent cell-mediated cytotoxicity [lo-161. These Ag, together with excreted 4 Discussion products [36, 371, may also be responsible for the triggering This work has evaluated the anti-sm IgE response of young of mast cell degranulation allowing the early recruitment residents of an endemic area of S. mansoni whose suscep- and activation of helminthotoxiceffector cells at the site of tibility to infection had been carefully determined in penetration of the parasite. The multiplicity of the sm Ag previous studies [9, 301. The objectives were to identify the that bind IgE and the presence of certain of these Ag on the principal targets of IgE on sm, to determine whether the surface of the larvae are evidence for a role of IgE in human IgE responses of most resistant subjectswere different from protection against S. mansoni. This view is also supported those of the least resistant ones, and to evaluate the effects by the finding that IgE levels in the most resistant subjects of a parasitological cure with oxamniquine on these were, on average, sevenfold higher than those of the least resistant ones. This observation is consistent with a recent responses. report showing that subjectswith high levels of anti-sm IgE Immunoblots of sm tegument extracts showed that sera are more resistant to reinfection by S. haematobium [38]. from most study subjects contained IgE that bound to Buttenvorth et al. [8] analyzing the IgE Ab response of various sm Ag. The pattern of Ag recognized by IgE subjects exposed to infection by S. mansoni observed a differed from one subject to the other and altogether, IgE trend, although not significant, for higher IgE levels in the from the different sera reacted with a large number of sm most resistant subjects as compared to the least resistant Ag. This makes it unlikely that all these Ag have unusual ones; differences in the sensitivity of the assays and biochemical or structural properties accounting for the differences in the criterions used to define resistant and
Eur. J. Immunol. 1991. 21: 2679-2686
IgE in human resistance to S. rnansoni
2685
susceptible groups may account for the discrepancy the membrane, indicating putative targets of IgE-depenbetween these and our results. Nevertheless, in our study, dent cell-mediated cytotoxicity reactions on the invading one subject with high resistance had low IgE levels and larvae. Third, IgE levels were higher in the most resistant conversely, several subjects among the least resistant, had subjects than in the least resistant ones. These observations medium to high IgE levels.This indicates that a strong IgE altogether are consistent with the view that IgE may play a response is probably neither always sufficient nor, in some significant role in human immune defenses against S.mancases, necessary to allow high protection. Such a result was soni infections. expected since human immune defences against helminths are thought to involve a number of protective mechanisms which differ in relative importance for protection from one We wish to thank Dr. M . Delaage (Imrnunotech, Marseille, France) for the gift ofthe anti-human IgE mouse m A b and the acetylcholine individual to another. esterase anti-human IgE mAb.
In contrast to IgE, anti-sm IgM and anti-sm IgG levels were not significantly different between study groups 2 months before and 15 months after treatment and even increased with reinfection intensities 43 months after treatment. Consistentwith our findings,lower levels of IgM Ab against a p38 larval Ag were observed in subjects with the highest resistance to infection in the Kenyan study [39]. Since IgE and IgG were able to bind to the same Ag, and probably to the same or spatially related epitopes, it was expected that IgG would compete with IgE for the binding to Ag in ELISA and on Western blots. High levels of competing Ab were detected in most subjects regardless of their levels of resistance, and competing Ab levels were not, on average, different between the two study groups before treatment and 15 months later. However, studies on subjects with atopy [40,41] or with helminth infections [42,43], have suggested that anaphylactic reactions may be regulated by such inhibiting Ab. Competing Ab in sera taken 43 months after treatment were on average higher in the least resistant group than in the most resistant one, although unsufficiently to account for the differencesin IgE titers between the two study groups.This result is corroborated by the higher titers of specific IgG in the least resistant group. In a more recent study (Demeure, C. et al., to be published) we have obtained evidence indicating that the competing Ab are IgG4. IgG4 were also measured in the Gambian study and higher IgG4 levels were recorded in the least resistant subjects [38]. Our data show that from +15 to t-43 months, a significant difference in IgG competing Ab takes place between the two groups as the most resistant group resists reinfection, while the least resistant one develops high infections.Whether these competing Ab modulate the anti-sm IgE-dependent immunity in vivo remains to be evaluated. This study does not reveal a major effect of oxamniquine treatment on either the levels or specificitiesof anti-sm IgE in patients’ sera. This observation does not rule out a transient stimulation of IgE within few months after treatment as reported in severalstudies [25,26] .This lack of a major effect of the treatment on IgE levels and specificities is consistent with our observation that treatment does not modify susceptibility to infection of these subjects [91. In conclusion,this work reveals several interestingfeatures of the anti-sm IgE response. First, IgE levels remained relatively steady during the whole study while subjects experienced large variations in infection intensities, indicating that aborted or low infections are sufficient to maintain high IgE levels. Second, IgE binds to a large variety of sm Ag, some of them being on the outer part of
Received March 27, 1991; in revised form June 12, 1991.
5 References 1 UNDFlWorld BanWHO, in WHO (Ed.), Tropical Disease Research, Seventh Programm Report, Barthelemy, A . , 1983, p. 313. 2 Butterworth, A. E.,Wilkins, H. A., Capron, A . and Sher, A., Parasitol. Today 1987. 3: 1. 3 Fenwick, A., Parasitol. Today 1987. 3: 70. 4 Klumpp, R. K. and Chu, K.Y.,Parasitol. Today 1987. 3: 74. 5 Prata, A., Bailiere’s Clinic. Trop. Med. Comrn. Dis.1987. 2: 349. 6 Wilkins, H. A., Parasitol. Today 1989. 5: 83. 7 Katz, N., reported by Von Lichtenberg,F., Am. J. Trop. Med. Hyg. 1985. 34: 78. 8 Butterworth, A. E., Capron, M., Cordingley, J. S., Dalton, F! R., Dunne, D. W., Kariuki, H. C., Kimani, G., Koech, D., Mugambi, M., Ouma, J. H., Prentice, M. A., Richardson, B. A., Arap Siongok, T. K., Sturrock, R. F. and Taylor, W. D., Trans. R. SOC. Trop. Med. Hyg. 1985. 78: 393. 9 Dessein, A. J., Begley, M., Demeure, C., Caillol, D., Fueri, J., Galvao dos Reis, M.,Andrade, Z . A., Prata, A. and Bina, J., J. Irnrnunol. 1988.140: 2727. 10 Capron, A., Dessaint, J. €?, Capron, M. andBazin, H., Nature 1975.253: 474. 11 Capron, A., Dessaint, J. I?, Joseph, M., Rousseaux, R., Capron, M. and Bazin, A., Eur. J. Immunol. 1977. 7: 315. 12 Capron, M., Bazin, H., Joseph, M. and Capron, A., J. Immunol. 1981. 126: 1764. 13 Kigoni, E. F!,Elsas, F! X., Lenzi, H. L. and Dessein, A. J., Eur. J. Irnmunol. 1986. 16: 589. 14 Joseph, M., Capron, A . , Butterworth, A . E., Sturrock, R. F. and Houba,V., Clin. Exp. Irnrnunol. 1978. 33: 48. 15 Joseph, M., Auriault, C., Capron, A.,Vorng, H. and Viens, F!, Nature 1983. 303: 810. 16 Capron, M., Spiegelberg, H. L., Prin, L., Bennich, H., Butterworth, A. E., Pierce, R. J., Ouaissi, M. A. and Capron, A., J. Immunol. 1984.132: 462. 17 Bina, J. C. and Prata, A., in Azvedo, E. S., Rocha, H. and Andrade, Z . , (Eds.), Aspectos peculiares da infecgao por Schistosoma mansoni. Cedre Publication, Bahia 1984, p. 13. 18 Dessein, A. J., Couissinier, E , Rihet, F!, Demeure, C. E., Oatara, M. B.,Carvalho, D., Kohlstiidt, S., Goudot-Crozel,V., Bourgois, A. and Abel, L., Clin. Immunol. Allergy Rev. 1991, in press. 19 Anfosso, F., Demeure, C., Delaage, M., Cheballah,R., Bellot, F. and Bourgois, A., Mol. Immunol. 1987. 24: 1129. 20 Ramalho-Pinto, F. S., Gazinelli, G., Howells, G. E., MotaSantos,T. A., Figueredo,E. A. and Pellegrino,Exp. Parasitol. 1974. 36: 360. 21 Lazdins, J. K., Stein, M. J., David, J. R. and Sher, A . , Exp. Parasitol. 1982. 53: 39. 22 Towbin, H., Staehlin,T. and Gordon, J. C., Proc. Natl. Acad. Sci. USA 1979. 76: 4350. 23 Engvall, E. and Perlmann, F!, J. Imrnunol. 1972. 109: 129.
2686
F! Rihet, C. E. Demeure, A. Bourgois et al.
24 Lindmark, R.,Yhoren-Tolling, K. and Sjijsvist, J., J. Immunol. Methods 1983. 62: 1. 25 Mendes, E. and Mendes, N. F., Trop. Geogr. Med. 1982. 34: 133. 26 EvengArd, B., Hammarstrom, L., Smith, C. I. E., Johansson, S. G. 0. and Linder, A., Clin. Exp. Immunol. 1988. 73: 383. 27 Cooke, R.A., Barnard, J. H., Hebald, S. and Stull,A., J. Exp. Med. 1935. 62: 733. 28 Lichtenstein, L. M., Holtzman, N. A. and Burnett, L. S., J. Immunol. 1968. 101: 317. 29 Gleich, G. J., Zimmermann, E. M., Gleich, M. C. and Yunginger, J. W., J. Immunol. 1981. 126: 575. 30 Abel, L., Demenais, F., Prata, A., Souza, A. E. and Dessein, A. J., Am. J. Hum. Genet. 1991. 48: 959. 31 Hussain, R. and Ottessen, E. A., J. Immunol. 1985. 135: 1415. 32 Orr,T. S. C. and Blair, A. M. J. N., Life Sci. 1969. 8: 1073. 33 Jarrett, E. E. E., Immunology 1972.22: 1099. 34 Janeway, C., Immunol. Today 1989. 10: 283. 35 Weaver, C. T. and Unanue, E. R., Immunol. Today 1990. 11: 49.
Eur. J. Immunol. 1991. 21: 2679-2686 36 Auriault, C., Darnonneville, M., Verwaede, C., Pierce, R., Joseph, M., Capron, M. and Capron, A., Eur. J. Immunol. 1984.14: 132. 37 Velge-Roussel, F. ,Verwaede, C., Grzych, S. M., Auriault, A. and Capron, A., J. Immunol. 1989.142: 2527. 38 Hagan, P.,Blumenthal, U. J., Dunn, D., Simpson, A. J. G. and Wilkins, H. A., Nature 1991. 349: 243. 39 Khalife, J., Capron, M., Capron, A., Gnych, J.-M., Butterworth, A. E., Dunne, D. W. and Ouma, J. H., J. Exp. Med. 1986. 164: 1626. 40 Lessof, M. H., Kagey-Sobokta, A. and Lichtenstein, A. M., John Hopkins Med. 1. 1978. 142: 1. 41 Hunt, K. J.,Valentine, M. D., Kagey-Sobokta, A., Benton, A. W., Amodio, F. J. and Lichtenstein, L. M., N. Engl. J. Med. 1978. 299: 157. 42 Ottessen, E. A., Kumaraswami,V.,Paranjade, R., Poindexter, R. W. and Tirpathy, S. I?, J. Immunol. 1981. 127: 2014. 43 Hoffstetter, M., Poindexter, R. W., Ruiz-Tiben, E. and Ottessen, E. A., Immunology 1982. 46: 777.
