in animal models is surprisingly limited [11]. mechanisms of this acquired resistance remain poorly undereosinophilia

Transcription

1 512 Increased Interleukin-4 and Interleukin-5 Production in Response to Schistosoma haematobium Adult Worm Antigens Correlates with Lack of Reinfection after Treatment Ahmed Medhat, Magda Shehata, Kim Bucci, Shoari Mohamed, Ahmed Diab Enas Dief, Saad Badary, Hesham Galal, Mohamed Nafeh, and Christopher L. King Department of Tropical Medicine, Assiut University, Assiut, Egypt; Division of Geographic Medicine, Department of Medicine, Case Western Reserve University, Cleveland, Ohio Acquired immunity to human schistosomiasis correlates with increased serum levels of schistosome antigen specific IgE. Since interleukin (IL)-4 stimulates IgE production, the hypothesis that Th2-associated cell-mediated immunity participates in protection to reinfection was studied in a cohort of adolescent boys months after chemotherapeutic cure in Upper Egypt. Initial Schistosoma haematobium prevalence was 51% and posttreatment incidence was 44%. Water contact was similar between putatively resistant and susceptible patients. Resistant persons had a 3.5- to 14- fold greater frequency of schistosome adult worm antigen (SWAP) specific lymphocytes secreting IL-5 or IL-4 (by ELISPOT) and IL-5 or IL-4 production in peripheral blood lymphocyte culture supernatants (P õ.05 to õ.001, n Å 48) versus susceptible subjects (n Å 38). In contrast, SWAPinduced interferon-g and IL-10 production and lymphocyte proliferation were similar between the 2 groups. Schistosome egg antigen and streptolysin O each stimulated similar cytokine production in susceptible and resistant persons. Thus, enhanced SWAP-driven IL-4 and IL-5 production correlates with immunity to reinfection in adolescents exposed to urinary schistosomiasis. Epidemiologic studies of human schistosomiasis suggest that that IgE participates in acquired immunity to helminth infections acquired immunity gradually develops with age [1, 2]. The in animal models is surprisingly limited [11]. mechanisms of this acquired resistance remain poorly undereosinophilia It is possible that elevated levels of IgE and peripheral blood stood and could prove useful in identifying immune responses are markers of anti-parasite cell-mediated immu- needed for an effective vaccine. What is known is that persons nity. Since interleukin (IL)-4 and IL-5 participate in the inducwith increased resistance to reinfection after chemotherapeutic tion and regulation of IgE and eosinophilia, increased produc- cure have elevated serum levels of schistosome adult worm tion of these cytokines may participate in the development of antigen (SWAP) specific IgE [3 5], IgA directed to the 28- acquired immunity. Several studies indicate such an association kda schistosome antigen glutathione-s-transferase [6], and peeral may exist in Schistosoma mansoni infection. In Kenya, periph- ripheral blood eosinophilia [7, 8]. Conversely, increased serum blood mononuclear cells (PBMC) from putatively resistant levels of IgG4, IgG2, and IgM directed against adult worm persons produced IL-5; however, this association was con- antigens are associated with an increased susceptibility to reinnot founded by age, and cytokine production was directed to egg, fection [3, 9, 10]. Antigen-specific IgG4 and IgG2 have been adult worm antigens [12]. postulated to block IgE-mediated killing of the parasite and T cell clones generated in response to adult worm antigens thus enhance susceptibility to infection [10]. These studies, from putatively immune persons produced an increased ratio of however, are only correlative, and direct experimental evidence IL-4 to interferon (IFN)-g release compared with a susceptible individual; however, the study examined only 4 subjects [13]. Recently, a locus on the human chromosome 5q31-q33 has been linked to controlling the intensity of infection by S. mansoni infection [14]. This same region of chromosome 5 also Received 30 October 1997; revised 25 February Presented in part: annual meeting of the American Society of Tropical contains genes that regulate IL-4, IL-5, and IL-13 levels [15], Medicine and Hygiene, Orlando, FL, December 1997 (abstract 187). which further supports an association between increased IL-4 Informed consent was obtained from all subjects and their parents. The and IL-5 production with resistance. No studies, to our knowlstudies were approved by the appropriate authorities of the Government of Egypt and the Human Studies Committee of University Hospitals of Cleveland, edge, have examined correlates of cell-mediated immunity to Case Western Reserve University, Cleveland. resistance in Schistosoma haematobium infections. Grant support: Schistosome Research Project ( from the US Agency Most studies examining the mechanisms of acquired immufor International Development and the Egyptian Government) and an NIH Research Career Development Award (AI to C.L.K.). nity have investigated cohorts of subjects with a broad range Reprints or correspondence: Dr. Christopher L. King, Division of Geographic of age [9, 12, 16 18]. Because age and exposure to contami- Medicine, Case Western Reserve University School of Medicine, nated water affect the intensity of infection, they can act as Room W137, 2109 Adelbert Rd., Cleveland, OH (cxk21@po.cwru.edu). confounding variables in assessing the role of acquired immu- The Journal of Infectious Diseases 1998;178:512 9 nity and must be controlled for in population-based studies by the Infectious Diseases Society of America. All rights reserved /98/ $02.00 Previous studies of acquired immunity also examined popula-

2 JID 1998;178 (August) Acquired Immunity to Urinary Schistosomiasis 513 tions with moderate to heavy parasite exposure that may obtact mained ova-negative, and had documentation of direct water con- scure development of partial immunity. at potential transmission sites at least twice a month during In such treatment and reinfection studies, resistance was the observation period. often defined as persons with low (resistant) compared with Antigen and mitogens. SWAP and soluble egg antigen (SEA) were prepared as a saline extract of adult-stage parasites or from high (susceptible) egg excretion. This poses a problem for freshly harvested ova [23]. Endotoxin in these preparations was studying urinary schistosomiasis, in which egg excretion is õ0.5 ng/ml; 5- to 50-fold less than that required for lipopolysachighly variable and correlates poorly with overall intensity of charide stimulation of cytokines from human lymphocytes [24]. infection [19]. To address these issues, we examined a cohort Isolation of PBMC and culture conditions for in vitro cytokine of adolescent boys in Upper Egypt, where the incidence of production. All studies were performed on fresh PBMC separeinfection remains high but the intensities of urinary schistoso- rated by density-gradient centrifugation on ficoll-hypaque from miasis are low compared with those in East Africa, for example heparinized cord and venous blood and resuspended in C-RPMI: [20]. Infected persons were treated and reinfection was moni- RPMI 1640 supplemented with 10% fetal calf serum (FCS), 4 mm tored over the following 18 months. Prior water contact studies L-glutamine, 25 mm HEPES, and 40 mg/ml gentamicin (BioWhit- in Upper Egypt show that adolescent boys consistently have taker, Walkersville, MD). PBMC were cultured at cells/ the highest and most uniform exposure [21]. The present study ml in C-RPMI in a total volume of 1 ml. To duplicate cultures was added medium alone, SWAP (50 mg/ml), SEA (10 mg/ml), examines the cell-mediated immunologic correlates in putastreptolysin O (SLO) (10 mg/ml), or phytohemagglutinin (PHA, tively resistant versus susceptible subjects as determined by 5 mg/ml). Cells were incubated at 37 C in5%co 2 ; supernatants whether they are reinfected or not after chemotherapeutic cure. were collected at 36 h (for IL-4) and 72 h (IL-5, IL-10, and IFNg) and immediately frozen at 070 C for subsequent determination of cytokine production. Materials and Methods Cytokine production by PBMC subsets. To assess cytokine production by lymphocyte subpopulations, PBMC were washed Study design. Study subjects resided in a village on the East once in cold RPMI with 2% fetal bovine serum. CD4 cell depletion Bank of the Nile River 30 km from Assiut, a city in Upper Egypt. was performed using magnetic beads directly conjugated to anti- The village population was Ç4000, and the principal occupation CD4 antibodies (Dynal, Lake Success, NY). For control data, imis farming. In this village, a cohort of 145 male children aged 9 15 munomagnetic depletion was accomplished according to the manuyears were selected on the basis of their willingness to participate. facturer s instructions with whole PBMC from 4 healthy North Midday urine was obtained on 2 consecutive days at the beginning American adults, using the identical reagents and protocols used of the study and at 3, 6, and months later. Any individual for the study subjects; ú88% of CD4 cells were removed. The found infected with schistosomiasis was treated with 40 mg/kg efficiency of CD4 cell depletion was not checked on individual praziquantel. Between 12 and 18 months after treatment, Ç20 ml samples in the study. of peripheral venous blood was obtained to prepare PBMC. Cytokine assays. Cytokine levels in cell supernatants were Exclusion criteria from the final analysis comprised persons who measured by ELISA and expressed in picograms per milliliter by (1) were infected before enrollment in the study but did not have interpolation from standard curves based on recombinant lymphoa urine sample collected 3 or 6 months after treatment to determine kines by using antibodies and methods previously described [25]. the efficacy of therapy or (2) remained ova-positive at 3 months Antibody pairs for capture and detection (all biotinylated) for the after treatment, indicating probable treatment failure. cytokines studied were as follows: for IL-5, TRFK5 and 5D10 Water contact studies. Transmission sites were identified on (PharMingen, San Diego); for IL-4, 8D4 and 25D2 (PharMingen); the basis of where people frequented the canals and the presence for IFN-g, M-700 and M-701 (Endogen, Cambridge, MA); and of S. haematobium infected snails, Bulinus alexandria. The fre- for IL-10, 18551D and 18652D (PharMingen). The limits of detecquency of, duration of, and amount of the body in contact with tion for ELISA for each cytokine were 18 pg/ml for IL-5, 16 pg/ the water (e.g., whole body [swimming], to waist, just arms and ml for IL-4, 10 pg/ml for IFN-g, and 16 pg/ml for IL-10. legs) were recorded for study subjects by trained local observers The ELISPOT assays utilized T-spot plates (Athersys, Clevethroughout the course of the transmission season (April through land), which were coated with capture antibodies (the same as September) based on previously established protocols [22]. used in the ELISAs) in sterile PBS overnight at 4 C and blocked Criteria for classification of susceptible versus putatively resis- with C-RPMI. Plates were then washed three times with sterile tant subjects. Susceptible persons were (1) ova-positive at the PBS. Single cell suspensions were prepared, counted, transferred beginning of the study, cured with treatment (ova-negative at 3 to the plate at 200 ml/well, and incubated at 37 C with 5% CO 2 months), and then ova-positive at 6 and/or 12 months after treat- for 24 h (IL-4 and IFN-g) and 48 h (IL-5 and IL-10). For antigenment or (2) ova-negative at the onset of the study and at 3 months specific cytokine production, cells were added per and then ova-positive at 2 of the follow-up times between 6 and well. For mitogen-driven cytokine production, this was reduced to 18 months after the study began. Water contact behavior was not cells/well. The wells were run in duplicate if there used to exclude these subjects from the study, since the fact that were enough cells. After incubation, plates were washed three they became infected demonstrated exposure. times with PBS followed by three washes with PBS-Tween Putatively resistant subjects were (1) ova-positive at the study (0.05%). Detecting antibodies (the same as used in the ELISAs) beginning and became ova-negative after treatment and remained were added and incubated overnight at 4 C. The resulting spots so throughout the study or (2) ova-negative at study onset, re- were initially enumerated with a dissecting microscope (15 20),

3 514 Medhat et al. JID 1998;178 (August) Table 1. Population of adolescent boys studied for schistosomiasis. detection of cytokine production in culture supernatants by ELISA. Susceptible Resistant Relationship of water contact behavior with susceptibility (n Å 43) (n Å 58) or resistance to infection. All the water contact data were Median age, years (range) 11 (10 15) 12 (9 15) condensed into two variables for each individual: the average % ova-positive duration of water contact in minutes and the frequency of Geometric mean ova count (range) 2.6 (0 31) 2.7 (0 46) contact. The amount of body-water contact showed little vari- Water contact ability in adolescent boys. Most boys either swam or exposed Frequency, per week* 2.7 (0 13) 2.2 (1 12) Average duration (range) more than half of their body in the water. Therefore, this vari- 65 (3 120) 57 (1 120) able was not considered further in the analysis. From the five * Mean (range) no. of water contacts per week based on 8hofdaily observa- major water contact sites, 6050 Bulinus truncatus snails were tion 6 days per week. Direct water contact observations were on alternating collected, of which 76 (1.3%) had S. haematobium infections. weeks during time of transmission (April to September). Minutes of water contact per week. Although the number and proportion of infected snails varied considerably, infected snails were found at all sites. However, most water contact by study subjects occurred at two sites. Most of the infected snails were recovered from these two sites and numbers were verified by an analyzer (T Spot Image Analyzer; (combined): ú4500 snails, 1.6% of which were infected with Athersys) that is designed to detect the immunoplaques using predetermined criteria based on size, shape, and colorimetric density. S. haematobium. Statistics. Results are expressed as mean { SD, and experito be similar between these two sites, and no different weight Therefore, we considered that the risk of infection was likely mental conditions were compared using Student s t test of logwas transformed data. given for water contact at the various sites in the final analysis. We found no correlation between the frequency and duration of water contact with the intensity of infection or Results with levels of antigen-driven cytokine production. The mean Study population. In the initial cohort of 145 persons frequency and duration of water contact were equivalent between susceptible and resistant subjects (table 1). within the study villages, 101 fulfilled the criteria as either Increased frequency of SWAP-induced IL-5 and IL-4 by putatively resistant or susceptible (table 1). Thirty persons were PBMC from putatively resistant subjects. The frequency of excluded because they did not provide urine samples in one or cytokine-secreting cells in response to adult worm antigens more of the follow-up periods, particularly at 3 months to was examined using ELISPOT. Only some subjects (n Å 27) verify treatment efficacy, or at 12 months. An additional 14 were examined using this procedure because of the large number subjects remained ova-positive at 3 months after treatment and of lymphocytes required for the assay. The frequency of were excluded because of treatment failure. Of those remaining SWAP-induced IL-4 and IL-5 secreting cells from putatively (table 1), susceptible and putatively resistant subjects were of similar age and similar prevalence and intensity of infection at study entry. Susceptible subjects had similar intensity and prevalence of infection compared with resistant subjects prior to infection. The geometric mean (GM) egg count among susceptible subjects obtained Ç12 months after treatment (or 6 months if the 12-month specimen was unavailable) was 4.2/ 10 ml of urine. Age profoundly affects the susceptibility to infection, and years of age is perhaps the most critical time in which an individual begins to shift from susceptible to resistant. To examine whether the age range of susceptible versus resistant subjects differed over this critical period of age, the distribution of subjects was compared (figure 1). Most subjects in the present study were years old, with similar age distribution between the 2 groups. Thus, age is unlikely to account for apparent resistance observed in some subjects in the present cohort. Of the 101 persons classified as susceptible or resistant, PBMC were successfully obtained from 86 for cytokine studies, Figure 1. Age distribution of susceptible and putatively resistant either by ELISPOT, if sufficient cells were available, or by subjects in study cohort.

4 JID 1998;178 (August) Acquired Immunity to Urinary Schistosomiasis 515 Figure 2. Frequency of cytokinesecreting cells in peripheral blood mononuclear cells (PBMC) from susceptible (Sus) and putatively resistant (Res) subjects. PBMC ( /well) were stimulated with antigen for 24 or 48 h, depending on cytokine, as described in Materials and Methods. Each dot represents mean value of duplicate cultures from 1 patient months after treatment. Unstimulated cultures usually showed rare or undetectable cytokine-secreting cells. Bars represent geometric mean for each group. P values were calculated by Student s t test using log-transformed data. IFN, interferon. immune subjects was significantly higher than that observed SWAP, susceptible subjects, mean stimulation index [SI] Å for susceptible subjects (figure 2). No difference in the frequency 2.2 [n Å 25] vs. resistant subjects, mean SI Å 1.9 [n Å 22]; of SWAP-specific IFN-g secreting cells was observed for SEA, susceptible subjects, mean SI Å 2.3 vs. resistant between the 2 groups. Of note, the frequencies of detectable subjects, mean SI Å 2.7). Mitogen-driven lymphocyte prolifer- IL-5 secreting cells was 5- to 10-fold greater than that ob- ation was also equivalent between susceptible (mean SI Å served for IL-4 and IFN-g secreting cells. There was no 68.5) and resistant subjects (mean SI Å 109.6). Although SEA- correlation between the frequency of IL-5 and IL-4 secreting induced IL-5 production did not correlate with SWAP-driven cells among either resistant or susceptible subjects (data not cytokine production, levels tended to be higher among puta- shown). There were fewer subjects with ELISPOT data for IL- tively resistant (GM Å 103 pg/ml; range, ) versus 4 and IFN-g than for IL-5 because of a lack of sufficient susceptible subjects (GM Å 20 pg/ml; range, ; P Å lymphocytes for some assays..052). No significant difference was observed for SEA-induced The increased frequency of IL-4 and IL-5 secreting cells IL-4 between putatively resistant (GM Å 4.6 pg/ml; range, by PBMC from resistant compared with susceptible subjects 0 234) and susceptible subjects (GM Å 3.4 pg/ml; range, 0 was restricted to SWAP. The frequencies of SEA-reactive or SLO (the nonparasite antigen)-reactive PBMC were similar between the 2 groups (table 2). Numbers of mitogen (PHA)- Table 2. Schistosome egg antigen (SEA), streptolysin O (SLO), induced IL-5, IL-4, and IFN-g secreting cells were also and mitogen (phytohemagglutinin, PHA) induced cytokine produc- equivalent between susceptible and resistant subjects (table 2). tion in susceptible versus resistant subjects. In some conditions, the number of PHA-driven cytokine-secre- Infection status ting cells were fewer than was observed for SWAP-stimulated Antigen/mitogen cytokine production because insufficient cells were available. cytokine n Resistant n Susceptible Increased levels of SWAP-induced IL-5 and IL-4 production in PBMC culture supernatants from putatively resistant versus SEA IL (4 110) (0 800) susceptible subjects. Because there were insufficient cells IL (0 53) 8 9 (0 14) from most study subjects for ELISPOT, PBMC cultures were IFN-g 14 8 (0 176) (0 300) done instead. Similar to the observations for cytokine-secreting SLO cells, SWAP-induced IL-5 and IL-4 levels were significantly IL (0 105) (0 38) greater among putatively resistant subjects (figure 3). Mean IL (0 48) (0 25) IFN-g (0 1000) (0 600) IL-5 release was 14-fold greater (GM Å 142 pg/ml) in resistant PHA subjects than in susceptible subjects (GM Å 10 pg/ml). A IL ( ) ( ) similar, but less striking difference was observed for IL-4 pro- IL ( ) ( ) duction. No difference in the levels of SWAP-induced IFN-g IFN-g (300 30,000) (412 17,000) and IL-10 was observed between the 2 groups (figure 3). Simi- NOTE. Data are geometric mean (range) no. of cytokine secreting cells larly, both SWAP- and SEA-induced lymphocyte proliferation per 10 6 (SEA and SLO) or 10 5 (PHA) peripheral blood mononuclear cells. IL, did not differ between susceptible and resistant subjects (for interleukin; IFN, interferon. No differences were statistically significant.

5 516 Medhat et al. JID 1998;178 (August) Figure 3. Cytokine production by culture supernatants between susceptible (Sus) and putatively resistant (Res) subjects. Values represent net schistosome adult worm antigen (SWAP) induced cytokine production in culture supernatants of peripheral blood mononuclear cells/ml in 0.5-mL cultures harvested 72 h (IL-5, IL-10, and interferon [IFN]-g) or 24 h (IL-4) after addition of antigen. Each dot represents mean value of single or duplicate cultures from 1 patient. P values were calculated by Student s t test using log-transformed data. 250). Similarly, there was no significant difference in the levels is likely to be directed to antigens expressed by developing of SEA-induced IFN-g (GM for resistant Å 7 pg/ml vs. susceptible larvae and adults, which share many of the same molecules Å 11 pg/ml, P Å.45) or IL-10 (GM for resistant Å [26]. We observed no difference in antigen-induced lympho- 97 pg/ml vs. susceptible Å 171 pg/ml, P Å.22). cyte proliferation or IFN-g or IL-10 production between the 2 CD4 lymphocytes in PBMC are the major source of helminth groups, indicating that the protective effect was restricted to a antigen driven cytokine production. To determine whether Th2-associated cytokine response. One application of these CD4 cells are the primary source of antigen-driven cytokine findings suggests that adjuvants in a schistosome vaccine that production, lymphocytes from subjects from whom sufficient augment antigen-specific IL-4 and IL-5 production may en- samples were obtained underwent CD4 cell depletion. The hance its efficacy. ELISPOT test was performed because of its potentially greater Study subjects were classified as resistant or susceptible on sensitivity in detecting IL-4 release. Figure 4 shows 3 represen- the basis of documented frequent exposure and whether they tatives of 15 patients in whom SWAP induced cytokine production became infected or not over the year after chemotherapeutic and whose lymphocytes were subjected to CD4 depletion. cure with praziquantel. This epidemiologic design was based Depletion of CD4 cells significantly reduced or abolished detectable on previous studies examining the mechanisms of acquired cytokine-secreting cells in PBMC from most subjects resistance to human schistosome infections [9, 12, 16 18]. studied, indicating CD4 cells were the predominant cell type However, most of those studies and others found that the risk for antigen-induced cytokine production. Depletion of IL-4 re- of infection was strongly associated with age [2, 27 29]. Although lease by removing CD4 cells was less pronounced than that age may represent cumulative exposure to the parasite observed for SWAP-induced IL-5 and IFN-g, indicating other and thus development of immunity, the consistent drop in the cell subpopulations are important for IL-4 production. intensity and frequency of infection among young adults in schistosomiasis-endemic areas, irrespective of intensity of exposure, Discussion suggests a component of age-related innate immunity [30]. In the present study, many putatively resistant persons had To reduce the potentially confounding effect of age, the an increased frequency of adult worm antigen specific T cells present study examined a cohort of adolescents who were 9 that primarily secreted IL-4 and IL-5 compared with susceptible 15 years old. However, even among adolescents, age-related persons. Only adult worm antigens, not egg antigens or SLO, differences in immunity may develop. First we found the same induced cytokine production that correlated with protection. age distribution between resistant and susceptible subjects in This is not surprising, since protection against schistosomiasis our cohort. Since most in the study cohort were years

6 JID 1998;178 (August) Acquired Immunity to Urinary Schistosomiasis 517 Figure 4. Effect of CD4 cell depletion on frequency of schistosome adult worm antigen (SWAP) induced cytokine-secreting cells by peripheral blood mononuclear cells (PBMC) from 1 putatively immune patient (Pt 34) and 2 susceptible patients (Pt 80 and Pt 120). PBMC were incubated as described in figure 2. In parallel cultures, PBMC were initially depleted of CD4 cells by immunomagnetic negative selection and then cultured under identical conditions as whole PBMC. These depletion experiments are representative of data from 15 patients. IFN, interferon. adult worm antigens that may participate in this resistance. Whether these same molecules or others participate in cell- mediated acquired resistance has not been addressed and is of potential importance in the development of a schistosome vaccine. Some of the putatively resistant subjects had no detectable IL-4 and/or IL-5 production. This could reflect an insensitivity of the assay to detect these cytokines, or schistosome-reactive lymphocytes may not be adequately represented in a single sample if they do not consistently appear in the intravascular compartment. Indeed, we obtained multiple samples from some persons, and the patterns of immune response differed consider- ably from one time to the next in a few subjects. An alternative conclusion might be that IL-4 and IL-5 may not be critical for protection in all persons and that other effector mechanisms may participate in resistance. More likely, however, putatively resistant subjects may have been misclassified and are really susceptible subjects that escaped adequate exposure despite their documented water contact. The mechanisms by which increased IL-4 and IL-5 produc- tion participates in elimination of the parasite remains uncertain. The enhanced Th2-associated cytokine response to adult worm antigens is consistent with previous studies indicating that human resistance is associated with IgE-mediated immunity to adult worm antigens [3 5, 9] and increased peripheral eosinophilia [7, 8]. In vivo studies show that IgE in rats [35] and eosinophils in mice [36] play critical roles in protection against reinfection in these animal models, and activated human eosinophils and platelets or macrophages armed by IgE kill schistosomula in vitro [37]. In addition, natural immunity to S. mansoni infection is reduced in IgE-deficient mice [11]. This old, children may have been too young for age to have a large influence on susceptibility.to examine this possibility further, subjects were stratified into 9 11 and years old and compared for differences in reinfection rates or levels of antigen-induced cell-mediated immunity. No differences were observed between the 2 age groups, demonstrating that the enhanced SWAP-induced IL-4 and IL-5 production in putatively resistant subjects occurred independent of age in our study population. A danger of classifying persons as susceptible or resistant may reflect inaccuracies in determining exposure [31]. For example, putatively resistant persons may have shown high levels of water contact but fortuitously avoided exposure to cercariae, while a susceptible individual may have had relatively little water contact but had a brief exposure to water containing a large number of cercariae. To reduce this potential variation of exposure, our cohort included only boys. Previous detailed studies of water contact behavior in adolescent boys from similar villages in Upper Egypt show a uniformly high exposure in this group [21]. Moreover, we monitored water contact in our study population and excluded those in the putatively resistant group who did not have documented exposure to potential transmission sites. Our use of crude schistosome antigen preparations has the advantage of maximizing possible immune responses between susceptible and immune persons. However, it also has the possible drawback that a mixture of antigens may produce a complex array of cytokines in vitro, with cross-modulatory properties that may obscure important responses to individual antigens. Immune sera [6, 32, 33] or T cell clones [34] from putatively resistant persons have been used to identify specific

7 518 Medhat et al. JID 1998;178 (August) is also consistent with the role of parasite-specific IgG1 in the potential suppressive effect that active infection may have protection in mice that is elicited by multiple immunizations the cytokine response. with irradiated cercariae [38]. However, protection in mice Other studies, however, found no association of schistosome immunized once with irradiated cercariae has been shown to antigen induced lymphocyte proliferation and resistance, simi- be IFN-g dependent [39]. This suggests that multiple mecha- lar to findings observed here. Instead, they observed that egg nisms may generate protection; however, repeated exposure to antigen induced IL-5 production was elevated among subjects infective stages of the parasite, as is typical of natural infection, who were putatively resistant to S. mansoni infection [12]. favors a more Th2-type protective response. However this association was confounded by age, which could IL-4 and IL-5 may play a role in resistance apart from stimu- imply that the cumulative exposure with age to egg antigens lating a protective IgE response. For example, in humans, elevated results in enhanced Th2-associated responses and not protective serum levels of parasite-specific IgG4 occur with active immunity. In the present study, no association was found be- infection and are associated with increased susceptibility to tween IL-2, IFN-g, IL-4, or TNF production and resistance. reinfection [3, 9]. It has been postulated that this results from Our findings indicate that Th2-mediated immunity participates blocking the protective effect of IgE [3, 9]. Since IL-4 stimulates in development of acquired immunity to human schistoblocking both IgE and IgG4 production in human helminth infec- somiasis, although multiple effector mechanisms are likely involved. tions [40], this suggests that the role of IL-4 in protection may To advance the study of acquired immunity to human be more complex. IL-4 has an important role in controlling schistosomiasis, additional studies of immune responses to selected Heligmosomoides polygyrus and Trichuris muris and can contribute antigens will be needed in larger population-based stud- to the protection against Nippostrongylus brasiliensis ies that carefully control for age and exposure. In this way, and Trichinella spiralis infections in mice independent of IgE vaccine candidate molecules can be administered in a context (reviewed in [41]). IL-4 also stimulates T cell growth and that might enhance their efficacy. However, the complex nature enhances expression of VCAM-1 (the endothelial cell receptor of immunity that must develop in human schistosomiasis will for the integrin VLA-4, which is involved in the migration of likely require vaccines with several antigens delivered in a macrophages, lymphocytes, and eosinophils) [42]. In addition, fashion to evoke multiple effector responses. IL-4 stimulates nitric oxide (NO) synthase, the enzyme that catalyzes macrophage production of NO [43], which has been hypothesized to participate in killing schistosomula in mice immunized with irradiated cercariae [44]. References Population-based studies that have examined cell-mediated 1. Butterworth A. Immunity in human schistosomiasis. Acta Trop Suppl immunity in putatively immune subjects are surprisingly lim- 1987;12: Wilkins H, Blumenthal U, Hagan P, Hayes R, Tulloch S. Resistance to ited, and none, to our knowledge, has been previously reported reinfection after treatment of urinary schistosomiasis. Trans R Soc Trop for S. haematobium infections. Studies of the association of Med Hyg 1987;81: cell-mediated immunity with resistance have revealed inconsis- 3. Hagan P, Blumenthal U, Dunn D, Simpson A, Wilkins H. Human IgE, tent findings. Antibody or mononuclear cell supernatants from IgG4 and resistance to reinfection with Schistosoma haematobium. Na- persons putatively resistant to S. mansoni infection showed no ture 1991;349: Dunne DW, Butterworth AE, Fulford AJC, et al. Immunity after treatment enhancement of eosinophil-dependent killing of schistosomula of human schistosomiasis: association between IgE antibodies to adult in vitro compared with susceptible subjects in Kenya [31]. worm antigens and resistance to reinfection. Eur J Immunol 1992;22: Another study found increased lymphocyte proliferation re sponses to S. mansoni cercarial and egg antigens, but not adult 5. Rihet P, Demeure CE, Bourgois A, Prata A, Dessein AJ. Evidence for an worm antigens, among putatively resistant versus susceptible association between human resistance to Schistosoma mansoni and high anti-larval IgE levels. Eur J Immunol 1991;21: subjects in Egypt [45]. However, the study was small (18 sub- 6. Grzych JM, Grezel D, Xu CB, et al. IgA antibodies to a protective antigen jects) and the degree of water contact was not evaluated. A in human schistosomiasis mansoni. J Immunol 1993;150: larger study of S. mansoni infection in Brazil found that resis- 7. Sturrock R, Kimani R, Cottrell B, et al. Observations on possible immunity tant persons with frequent water contact had significantly to reinfection among Kenyan schoolchildren after treatment for Schisto- higher lymphocyte proliferative responses to schistosomula and soma mansoni. Trans R Soc Trop Med Hyg 1983;77: Hagan P, Wilkins H, Blumenthal U, Hayes R, Greenwood B. Eosinophilia adult worm antigens than infected persons [46]. In this latter and resistance to Schistosoma haematobium in man. Parasite Immunol study, they only examined IFN-g production, which was unde- 1985;7: tectable in most subjects. These observations were supported by 9. Demeure C, Rihet P, Abel L, Ouattara M, Bourgois A, Dessein A. Resistance another study of S. mansoni in Brazil [47]. Putatively resistant to Schistosoma mansoni in humans: influence of the IgE/IgG4 (uninfected) compared with susceptible (infected) subjects balance and IgG2 in immunity to reinfection after chemotherapy. J Infect Dis 1993;168: showed significantly increased schistosome antigen induced 10. Dunne DW, Grabowska AM, Fulford AJ, et al. Human antibody responses lymphocyte proliferation and, in this study, IFN-g production. to Schistosoma mansoni: the influence of epitopes shared between dif- However, the study of the putatively resistant group or en- ferent life-cycle stages on the response to the schistosomulum. Eur J demic normals failed to consider the effect of exposure or Immunol 1988;18:

8 JID 1998;178 (August) Acquired Immunity to Urinary Schistosomiasis King CL, Xianli J, Malhotra I, Liu S, Mahmoud A, Oettgen H. Mice with 30. Chandiwana S, Woolhouse M, Bradley M. Factors affecting the intensity a targeted deletion of the IgE gene have increased worm burdens and of reinfection with Schistosoma haematobium following treatment the reduced granulomatous inflammation following primary infection with praziquantel. Parasitology 1991;102: Schistosoma mansoni. J Immunol 1997;158: Butterworth AE, Capron M, Cordingley JS, et al. Immunity after treatment 12. Roberts M, Butterworth A, Kimani G, et al. Immunity after treatment of human schistosomiasis mansoni. II. Identification of resistant individof human schistosomiasis: association between cellular responses and uals, and analysis of their immune responses. Trans R Soc Trop Med resistance to reinfection. Infect Immun 1993;61: Hyg 1985;79: Couissinier-Paris P, Dessein A. Schistosoma-specific helper T cell clones 32. Dessein A, Begley M, Demeure C, et al. Human resistance to Schistosoma from subjects resistant to infection by Schistosoma mansoni are Th0/2. mansoni is associated with IgG reactivity to a 37-kDa larval surface Eur J Immunol 1995;25: antigen. J Immunol 1988;140: Marquet S, Abel L, Hillaire D, et al. Genetic localization of a locus 33. Webster M, Fulford A, Braun G, et al. Human immunoglobulin E recontrolling the intensity of infection by Schistosoma mansoni on chromansoni sponses to a recombinant 22.6-kilodalton antigen from Schistosoma mosome 5q31-q33. Nat Genet 1996;14: adult worms are associated with low intensities of reinfection 15. Marsh D. Linkage analysis of IL-4 and other chromosome 5q31.1 markers after treatment. Infect Immun 1996;64: and total serum immunoglobulin E concentrations. Science 1994;264: 34. Couissinier-Paris P, Bourgois A, Dessein H, et al. Identification of a major T cell immunogen in the anti-schistosome response of adult residents 16. Butterworth AE, Dalton PR, Dunne DW, et al. Immunity after treatment in an area endemic for Schistosoma mansoni. Eur J Immunol 1995;25: of human schistosomiasis mansoni. I. Study design, pretreatment obser vations and the results of treatment. Trans R Soc Trop Med Hyg 1984; 35. Pierce RJ, Verwaerde C, Damonneville M, LePresle T, Capron A. Aller- 78: gens of Schistosoma mansoni: I. Comparison of the IgE response in 17. Hagan P, Blumenthal U, Chaudri M, et al. Resistance to reinfection with human and experimental infections towards characterized allergens from Schistosoma haematobium in Gambian children: analysis of their im- adult worm products. Int Arch Allergy Appl Immunol 1983;72:219 mune responses. Trans R Soc Trop Med Hyg 1987;81: Wilkins H, Goll P, Marshall T, Moore P. Dynamics of Schistosoma haemaacquired 36. Mahmoud AAF, Warren KS, Peters PA. A role for the eosinophils in tobium infection in a Gambian community. I. The pattern of human resistance to Schistosoma mansoni infection as determined by infection in the study area. Trans R Soc Trop Med Hyg 1984;78:216 antieosinophil serum. J Exp Med 1975;142: Capron M, Capron A. Immunoglobulin E and effector cells in schistosomi- 19. Smith JH, Elwi A, Kamel IA, von Lichtenberg F. A quantitative post asis. Science 1994;264: mortem analysis of urinary schistosomiasis in Egypt. II. Evolution and 38. Delgado B, McLaren DJ. Evidence for enhancement of IgG1 subclass epidemiology. Am J Trop Med Hyg 1975;24: expression in mice polyvaccinated with radiation attenuated cercariae 20. King C, Keating C, Muruka J, et al. Urinary tract morbidity in schistosomiimmunity. of Schistosoma mansoni and the role of this isotype in serum transferred asis haematobia: associations with age and intensity of infection in an Parasite Immunol 1990;12: endemic area of Coast Province, Kenya. Am J Trop Med Hyg 1988; 39. Smythies LE, Coulson PS, Wilson RA. Monoclonal antibody to IFN-g 39: modifies pulmonary inflammatory responses and abrogates immunity 21. Kloos H, Higashi G, Schlinski V, Mansour N, Murrell K, Miller F. Water to Schistosoma mansoni in mice vaccinated with attenuated cercariae. J Immunol 1992;149: contact and Schistosoma haematobium infection: a case study from an 40. King CL, Nutman TB. IgE and IgG subclass regulation by interleukin-4 Upper Egyptian Village. Int J Epidemiol 1990;19: and interferon-g in human helminth infections: assessment by B cell 22. Fulford A, Ouma J, Kariuki H, et al. Water contact observations in Kenyan precursor frequencies. J Immunol 1993;151: communities endemic for schistosomiasis: methodology and patterns of 41. Finkelman F, Shea-Donohue T, Goldhill J, et al. Cytokine regulation of behaviour. Parasitology 1996;113: host defense against parasitic gastrointestinal nematodes: lessons from 23. Boros DV, Warren KS. Delayed hypersensitivity-type granuloma formastudies with rodent models. Ann Rev Immunol 1997;15: tion and dermal reaction induced and elicited by a soluble factor isolated 42. Schleimer RP, Sterbinsky SA, Kaiser J, et al. IL-4 induces adherence of from Schistosoma mansoni eggs. J Exp Med 1970;132: human eosinophils and basophils but not neutrophils to endothelium. 24. de Waal Malefyt R, Haanen J, Spits H, et al. Interleukin 10 (IL-10) and Association with the expression of VCAM-1. J Immunol 1992;148: viral IL-10 strongly reduce antigen-specific human T cell proliferation by diminishing the antigen-presenting capacity of monocytes via down- 43. Guo F, Uetani K, Haque S, et al. Interferon g and interleukin 4 stimulate regulation of class II major histocompatibility complex expression. J prolonged expression of inducible nitric oxide synthase in human airway Exp Med 1991;174: epithelium through synthesis of soluble mediators. J Clin Invest 1997; 25. Mahanty S, Abrams JS, King CL, Limaye AP, Nutman TB. Parallel regula- 100: tion of IL-4 and IL-5 in human helminth infections. J Immunol 1992; 44. James S. Role of nitric oxide in parasite infections. Microbiol Rev 1995; 148: : Lukacs NW, Boros DL. Identification of larval cross-reactive and egg- 45. Colley DG, Barsoum IS, Dahawi HSS, Gamil F, Habib M, Alamy ME. specific antigens involved in granuloma formation in murine schistoso- Immune responses and immunoregulation in relation to human schistomiasis mansoni. Infect Immun 1991;59: somiasis in Egypt. III. Immunity and longitudinal studies of in vitro 27. Butterworth AE, Dunne DW, Fulford AJ, et al. Human immunity to Schis- responsiveness after treatment. Trans R Soc Trop Med Hyg 1986;80: tosoma mansoni: Observations on mechanisms, and implications for control. Immunol Invest 1992;21: Ribeiro de Jesus A, Almeida R, Bacellar O, et al. Correlation between 28. Clarke VD. The influence of acquired resistance in the epidemiology of cell-mediated immunity and degree of infection subjects living in an bilharziasis. Cent Afr J Med 1966;12:1 30. endemic area of schistosomiasis. Eur J Immunol 1993;23: Bradley D, McCullough F. Egg output stability and the epidemiology of 47. Viana I, Sher A, Carvalho O, et al. Interferon-g production by peripheral Schistosoma haematobium. Part II. An analysis of the epidemiology of blood mononuclear cells from residents of an area endemic for Schistosoma S. haematobium. Trans R Soc Trop Med Hyg 1973;67: mansoni. Trans R Soc Trop Med Hyg 1994;88: