Lyophilized soluble S. haematobium adult worm antigen (SWAP) was obtained from the Theodor Bilharz Institute (Egypt) and reconstituted as previously described elsewhere 11. The parasite strain is one used for previous immuno-epidemiology studies 15. Crude P. falciparum schizont antigen preparations were a gift of Dr. P. Druilhe, Institute Pasteur, Paris. The merozoite surface protein antigens used were prepared as recombinant proteins in Escherichia coli. Two antigens derived from merozoite surface protein 1 (MSP-1), DPKMWR and MSP-1₁₉ antigens also known as p190, gp195, 16 were used. MSP-1 can be divided into 17 distinct blocks, based on its conserved, semi-conserved and variable regions. Block 2 is highly polymorphic, with over 50 sequences described. However these sequences fall into one of 3 categories or serotypes: K1; Mad20; and RO33 17 and DPKMWR is a recombinant protein composed of Block 2 sequences from all 3 block 2 serotypes. MSP-1₁₉ is the conserved C terminal of MSP-1, and is the only part to remain on the surface of the merozoite, with the rest of the protein being shed upon erythrocyte invasion. Two full-length recombinant MSP-2 antigens were also used, namely CH150/9 (5/6) and Dd2 (13/14). MSP-2 has 2 serotypes: CH150/9 5/6 is taken from serotype A (3D7-like), and Dd2 belongs to serotype B (FC27-like).

The study was conducted in the Mashonaland East Province of Zimbabwe (31°30'E; 17°45'S) where S. haematobium and P. falciparum are endemic. The study area is described in detail elsewhere 18. Briefly the area is rural with the majority of the people living in thatched houses made up of clay. The main activity in these villages is subsistence farming and human water contact is frequent with at least 4 contacts/person/week due to insufficient safe drinking water and sanitation facilities (see 18 for studies in neighboring villages). Drinking water is collected from open wells while bathing and washing is conducted in two main rivers in the villages. Most families maintain a garden located near the river where water is collected for watering the crops. The villages were selected because health surveys regularly conducted in the region by the Provincial Medical Director showed little or no infection with other helminths and a low S. mansoni prevalence (<5%). Permission to conduct the work in this province was obtained from the Provincial Medical Director and ethical approval received from the Medical Research Council of Zimbabwe http://www.afronets.org/guidelines_for_researchers_and_ethics_committees.pdf. Informed consent (following explanation of the study aims and procedures) was obtained before the collection of parasitology and blood samples. The selected villages had not been included in the National Schistosome Control Program and there were no active malaria vector control programmes in the area. Therefore participants had not received treatment for schistosomiasis or other helminth infections meaning that we could study natural immune responses in the absence of drug-altered schistosome responses 919.

Plasmodium falciparum is the predominant species of malaria in Zimbabwe 20 where malaria transmission is largely unstable in nature. Approximately 5.5 million people out of a total population of 12.7 million live in malarious areas 21. Out of the 56 districts in Zimbabwe, malaria transmission occurs in 42. In 2002, a revised stratification based on a national parasite prevalence survey, HMIS data, entomological data and expert opinion was prepared which classified our study area under the sporadic transmission regions with low transmission 212223 meaning that this is a mesoendemic area for malaria 24. The peak P. falciparum transmission occurs from February to May 24 and our sampling times were in March and then 6 weeks later in April. The schools surveyed (a secondary school and its feeder primary school (i.e. where the majority of the primary school children come from) Goromonzi and Shangure Schools in Village, and Chindenga and Nyambanje Schools in Mutoko Village) were all in close proximity to rivers. Stool and urine specimens were assayed for S. haematobium, S. mansoni and geo-helminths using standard procedures 2526 and the intensities of S. haematobium calculated from at least 2 urine samples (maximum 3) collected on consecutive days, and those of S. mansoni calculated from mean intensities of at least 4 Kato-Katz slides (maximum 6), 2 from each of the 2–3 stool samples collected. Anti-helminth treatment was administered to all compliant people after the first survey as well as 6 weeks later at the end of the study. In order to be included in the cohort, participants had to meet all the following criteria at the two sampling points (pretreatment and 6 weeks post-treatment): 1) have provided at least 2 urine and 2 stool samples on 3 consecutive days which are used for S. haematobium and intestinal helminths detection including S. mansoni; 2) be negative for intestinal helminths including S. mansoni (no one was excluded on this criteria as everyone was negative for these infections), 3) be negative for schistosome infection at the 6 week survey if they had received ant-helminth treatment, 4) have given at least 10 mls of blood for serological assays and preparation of a thick smear slide for microscopic detection of Plasmodium parasites. 117 people (53 male, 64 female) aged between 6–18 years old met these criteria and were included in the study. Of these, 89 people received praziquantel treatment while 28 people were not treated either because they did not take western medicine for religious reasons (23) or were not present on treatment days. The 28 people agreed to take part in the study and effectively became untreated controls. Thus the population was stratified by its schistosome and malaria staus as well as schistosome treatment history as shown in table 1. People presenting with malaria were treated according to the treatment regime prescribed by the Ministry of Health in Zimbabwe.

The antibody subclasses associated with protection against the MSP antigens are IgG1 and IgG3 1214. Studies in S. haematobium have also associated IgG1 and IgG3 subclasses with protection against schistosomiasis 927. Therefore this study focused on these two responses directed against crude schistosome worm antigen (WWH), crude malaria schizont and MSP-1/2 antigens. Indirect enzyme linked immunosorbent assays (ELISA) already published for S. haematobium 15 and for the P. falciparum antigens 282930 were used for the assays. Briefly, ELISAs were conducted in duplicate for each sample using ELISA plates (Nunc-Immulon, Denmark) that were coated with 50 ul/well of 50 ng/ml antigen for recombinant antigens and 1 ug/ml for both crude antigens in 60 mM carbonate-bicarbonate buffer (pH 9.6) and incubated overnight at 4°C. Plates were blocked with 200 μl/well of skimmed milk (5% milk in phosphate buffered saline (PBS)/0.03% Tween 20) for 1 hr and washed three times in PBS/Tween 20, which was used for all washes. 100 μl of serum was added to each well at 1:100 dilution for all assays except for MSP1₁₉ which was 1:200; plates were incubated overnight at 4°C and then washed three times. 100 μl of subclass-specific monoclonal antibody was added at 1:1000 dilution for the detection of IgG1 and IgG3 (The Binding Site, UK). Plates were incubated overnight at 4°C, washed six times and 100 μl of ABTS substrate solution (KPL, Canada) was added, before the absorbance was read at 405 nm. Three negative controls from European volunteers who had never traveled to malaria or schistosome endemic areas were used on each plate and people were considered reactive if they had an absorbance value greater than the mean plus 2 standard deviations of the negative controls.

Statistical analyses were conducted using the statistical package SPSS and were used to test three distinct hypotheses, 1) P. falciparum status is not associated with schistosome-specific responses or schistosome infection level, 2) schistosome infection intensity is not related to plasmodia-specific responses or P. falciparum status, 3) treatment of schistosome-infected people with praziquantel does not affect plasmodia-specific responses.

For all statistical analyses, participants were divided into three age groups, 6–11 years old, 12–13 years, 14–18 years old reflecting the epidemiological groupings where schistosome infection levels are rising, peaking and declining as well as give large enough sample sizes for the statistical analyses. The initial statistical procedure was conducted to reduce the number of variables used in hypothesis testing. Partial correlation analysis was used to explore correlations between antibody levels directed against schistosome and plasmodia antigens. This was followed by factor analysis using principal components (PCA) 31. PCA is a standard technique for reducing multivariate data down to its main independent features 31. Components are extracted according to the amount of variation in the data they explain so the first component explains the most variation and each subsequent component is included if it explains a significant amount of variation present with the data 32. This procedure was used in this instance for two reasons (1) to avoid type I and type II errors in multiple tests and using correlated independent variables 3334 and (2) to determine if anti-plasmodial and anti-schistosome responses clustered separately. Following the factor analysis principal components were analyzed further with logistic regression to determine the risk factors associated with being infected with malaria to specifically address hypothesis 1. The PCA were also used in an analysis of variance (ANOVA) to address hypothesis 2. For this analysis the following variables were categorical; sex (male, female); age (3 categories already described); malaria status (positive, negative) and the PCA factors 1 and 2 as continuous variables. Schistosome infection intensity (log (x+1) transformed) as the dependent variable.

To test the effects of treatment on immune responses it was necessary to first establish if there had been a significant change in antibody responses between the two time points. A non parametric 2-tailed Wilcoxon test for related samples was used for this analysis since antibody levels were not normally distributed. This was followed by an analysis determining if the observed differences were associated with praziquantel treatment, comparing the change in treated vs. untreated people after allowing for the effects of host sex, age, pre-treatment schistosome infection intensity and pre-treatment malaria infection status using repeated measures ANOVA. All statistical analyses were performed using the statistical software SPSS, and type III sums of squares were used to calculate the F-value where appropriate and the p value was set at 0.05 (see 33). Unless otherwise stated, all statistical procedures used the following variables as categorical; sex (male, female); age (3 categories already described); malaria status (positive, negative); and schistosome treatment (treated, untreated). The following variables were continuous; mean schistosome infection intensity (log₁₀ (x+1) transformed), antibody level (absorbance at 405 nm) and the principle components.

Parasitological examination gave an overall schistosome infection prevalence of 59% and a mean infection intensity of 39 eggs/10 ml of urine (SE = 11 with a range between 0 and 1000 eggs) before praziquantel treatment. Infection prevalence and intensity followed a convex age-infection profile with infection levels rising to peak in children 12–13 years old and infection intensity declining faster than infection prevalence thereafter (Figure 1). Eighty-nine people received praziquantel treatment and 28 were not treated. The prevalence of schistosome infection in this group did not differ significantly (57% and 60% respectively), but infection intensity was higher in the treated group (45 vs. 19 eggs/10 ml urine). This difference was accounted for by allowing for the effects of infection intensity before testing for the effects of treatment on the change in antibody level in the statistical analyses. All participants except 1 showed a detectable anti-schistosome antibody response.

Malaria infection as determined by microscopic examination of thick smear slides was P. falciparum and had a prevalence of 66%. Parasite densities were not determined. P. falciparum prevalence peaked earlier (in 6–11 year olds) than schistosome prevalence (Figure 2) and did not differ significantly between people who received anti-helminth treatment and those who remained untreated (65% and 68% respectively). P. falciparum positive participants were parasitemic but asymptomatic and none of the participants suffered malaria attacks at the time of sample collection or were treated for malaria in between sampling times. There was no significant change in P. falciparum prevalence between the two sampling time points (66% vs. 62%) but a few people moved between infection status; 36 people remained negative, 68 positive while 9 people changed status from positive to negative and 4 from negative to positive. All participants except 3 had a detectable IgG3 response to P. falciparum schizont antigen a marker for relative exposure levels to P. falciparum infection 35 and these 3 did have a detectable IgG1 response against this antigen. For all antigens (plasmodia and schistosome specific), the most prevalent response was IgG3 and this showed different patterns with age to the IgG1 (Figures 3 and 4).

The participants mounted immune responses to both plasmodia and schistosome antigens as shown in Figures 3 and 4. There were no significant differences in responses between the two groups (those to receive treatment and those to remain untreated) at the beginning of the study before anti-helminth treatment (Figures 5 and 6). When measuring the levels of antibody responses, those directed against the schistosome whole worm crude antigen were predominantly IgG1 while the responses to malaria schizont antigen were mixed both before treatment and 6 weeks later. Responses to the MSP-1 antigens were predominantly IgG1 and while those directed against the MSP-2 antigens were mixed.

Initial partial correlation analyses controlling for sex, age, malaria status and pre-treatment schistosome infection intensity showed no significant correlations between anti-plasmodial and anti-schistosome responses. This was confirmed by factor analysis which reduced the 12 antibody responses (IgG1 and IgG3 directed against the 6 antigens; WWH, schizont, MSP-1 antigens (DPKMWR, MSP-1₁₉), MSP-2 antigens (CH150 and Dd2) into two components partitioned by parasite species with all the anti-plasmodial responses falling in one component (PCA1) and the two anti-schistosome responses falling in a separate component (PCA2) as shown in Table 2. The next step was to determine if these two variables representing anti-schistosome and anti-plasmodial responses affected the risk of a person being positive for malaria infection. This was conducted using a logistic regression which included host age, sex and schistosome infection intensity. This analysis showed that host sex, age and of level of anti-plasmodial/schistosome antibodies did not affect the risk of P. falciparum infection. However, schistosome infection intensity was a significant risk factor with the likelihood of being parasitemic for P. falciparum increasing with increasing schistosome infection intensity (B = 0.589, SE = 0.270, Wald = 4.77, df = 1, p = 0.029). Conversely schistosome infection intensity was significantly higher in people who were parasitemic for P. falciparum (F = 5, 67, df = 1, 117, p = 0.019) and was positively associated with PCA2 (F = 21.5, df = 1,117, p < 0.001). The later means that schistosome infection intensity increased with increasing levels of anti-WWH IgG1 and IgG3. There was no significant association between anti-plasmodial responses (PCA1) and schistosome infection intensity (F= 0.002, df = 1, 117, p = 0.969).

Six weeks after treatment of schistosome infections with the drug praziquantel, anti-schistosome and anti-plasmodial responses were assessed again to determine the factors associated with the change in the levels of these responses. There were significant changes in levels of IgG1 directed against WWH, MSP-2 (Dd2), MSP-1₁₉ and schizont antigen (Table 3) and in IgG3 directed against MSP-1₁₉. Where significant changes occurred, for them to be associated with praziquantel treatment, they had to differ significantly between praziquantel-treated people and the untreated controls after allowing for all other confounding variables (sex, age, P. falciparum status and pre-treatment schistosome infection intensity). Only changes in anti-schistosome IgG1 differed significantly between treated and untreated people (Table 3), declining in untreated people, while increasing in treated people (Figures 7 and 8).