• View in gallery

    Ultrasonography demonstrating the degree of pathology in the bladder (A and B) or kidney (C and D). A, Fully extended bladder with no apparent pathology. B, Bladder with masses extending into the lumen (arrows) (classified as severe). C, Kidney with no abnormalities. D, Kidney with severe dilation (arrows) and < 2 cm parenchyma.

  • View in gallery

    Number of parasite clusters in relation to urinary egg counts for those with four or more miracidial isolates. Bars show the mean ± SD.

  • 1

    Dabo A, Durand P, Morand S, Diakite M, Langand J, Imbert-Establet D, Doumbo O, Jourdane J, 1997. Distribution and genetic diversity of Schistosoma haematobium within its bulinid intermediate hosts in Mali. Acta Trop 66 :15–26.

    • Search Google Scholar
    • Export Citation
  • 2

    Shiff C, Brouwer KC, Clow L, 2000. Schistosoma haematobium: population genetics of S. haematobium by direct measurement of parasite diversity using RAPD-PCR. Exp Parasitol 96 :47–51.

    • Search Google Scholar
    • Export Citation
  • 3

    Davies CM, Webster JP, Kruger O, Munatsi A, Ndamba J, Woolhouse ME, 1999. Host-parasite population genetics: a cross-sectional comparison of Bulinus globosus and Schistosoma haematobium.Parasitology 119 :295–302.

    • Search Google Scholar
    • Export Citation
  • 4

    Brouwer KC, Ndhlovu P, Munatsi A, Shiff CJ, 2001. Genetic diversity of a population of Schistosoma haematobium derived from schoolchildren in east central Zimbabwe. J Parasitol 87 :762–769.

    • Search Google Scholar
    • Export Citation
  • 5

    Brouwer KC, Ndhlovu P, Wagatsuma Y, Munatsi A, Shiff CJ, 2003. Epidemiological assessment of Schistosoma haematobium induced kidney and bladder pathology in rural Zimbabwe. Acta Trop 85 :339–347.

    • Search Google Scholar
    • Export Citation
  • 6

    Taylor P, Makura O, 1985. Prevalence and distribution of schistosomiasis in Zimbabwe. Ann Trop Med Parasitol 79 :287–299.

  • 7

    Ndhlovu P, Cadman H, Gundersen S, Vennervald BJ, Friis H, Christensen NO, Mutasa G, Kaondera K, Mandaza G, Deelder AM, 1996. Circulating anodic antigen (CAA) levels in different age groups in a Zimbabwean rural community endemic for Schistosoma haematobium determined using the magnetic beads antigen-capture enzyme-linked immunoassay. Am J Trop Med Hyg 54 :537–542.

    • Search Google Scholar
    • Export Citation
  • 8

    Katz N, Chaves A, Pellegrino J, 1972. A simple device for quantitative stool thick-smear technique in Schistosomiasis mansoni. Rev Inst Med Trop Sao Paulo 14 :397–400.

    • Search Google Scholar
    • Export Citation
  • 9

    Mott KE, Baltes R, Bambagha J, Baldassini B, 1982. Field studies of a reusable polyamide filter for detection of Schistosoma haematobium eggs by urine filtration. Trop Med Parasitol 33 :227–228.

    • Search Google Scholar
    • Export Citation
  • 10

    Weber MD, Blair DM, Clarke V deV, 1967. The pattern of schistosome egg distribution in a micturition flow. Cent Afr J Med 13 :75–88.

  • 11

    WHO/TDR, 1996. Ultrasound in Schistosomiasis. International Workshop on the Use of Ultrasonography in Relation to Schistosomiasis. Niamey, Niger: CERMES.

  • 12

    Swofford DL, Selander RB, Black WC, 1997. BIOSYS-2: A Computer Program for the Analysis of Allelic Variation in Genetics. Urbana, IL: University of Illinois and Fort Collins, CO: Colorado State University.

  • 13

    Black WC, 1998. FORTRAN Programs for the Analysis of RAPD-PCR Markers in Populations. Fort Collins, CO: Colorado State University.

  • 14

    Excoffier L, Smouse PE, Quattro JM, 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131 :479–491.

    • Search Google Scholar
    • Export Citation
  • 15

    Apostol BL, Black WC, Miller BR, Reiter P, Beaty BJ, 1993. Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti.Theor Appl Genet 86 :991–1000.

    • Search Google Scholar
    • Export Citation
  • 16

    Subramanian AK, Mungai P, Ouma JH, Magak P, King CH, Mahmoud AA, King CL, 1999. Long-term suppression of adult bladder morbidity and severe hydronephrosis following selective population chemotherapy for Schistosoma haematobium.Am J Trop Med Hyg 61 :476–481.

    • Search Google Scholar
    • Export Citation
  • 17

    Wagatsuma Y, Aryeetey ME, Sack DA, Morrow RH, Hatz C, Kojima S, 1999. Resolution and resurgence of Schistosoma haematobium-induced pathology after community-based chemotherapy in Ghana, as detected by ultrasound. J Infect Dis 179 :1515–1522.

    • Search Google Scholar
    • Export Citation
  • 18

    Mostafa MH, Sheweita SA, O’Connor PJ, 1999. Relationship between schistosomiasis and bladder cancer. Clin Microbiol Rev 12 :97–111.

  • 19

    Vizcaino AP, Parkin DM, Boffetta P, Skinner ME, 1994. Bladder cancer: epidemiology and risk factors in Bulawayo, Zimbabwe. Cancer Causes Control 5 :517–522.

    • Search Google Scholar
    • Export Citation
  • 20

    Thomas JE, Bassett MT, Sigola LB, Taylor P, 1990. Relationship between bladder cancer incidence, Schistosoma haematobium infection, and geographical region in Zimbabwe. Trans R Soc Trop Med Hyg 84 :551–553.

    • Search Google Scholar
    • Export Citation
  • 21

    Pugh RN, Bell DR, Gilles HM, 1980. Malumfashi Endemic Diseases Research Project, XV. The potential medical importance of bilharzia in northern Nigeria: a suggested rapid, cheap and effective solution for control of Schistosoma haematobium infection. Ann Trop Med Parasitol 74 :597–613.

    • Search Google Scholar
    • Export Citation
  • 22

    Elem B, Vandal MT, 1981. Bilharziasis of the urinary tract in Zambia. (Observation on 100 consecutive cases). Med J Zambia 15 :48–51.

  • 23

    Chisty MM, Nargis M, Sato H, Inaba T, Takahashi G, Kamiya H, 2002. Schistosoma mansoni: kinetics of glomerulonephritis in Mongolian gerbils and its correlation with intensity and duration of infection. Parasite 9 :143–151.

    • Search Google Scholar
    • Export Citation
  • 24

    Hurst MH, Willingham AL, Lindberg R, 2002. Experimental schistosomiasis japonica in the pig: immunohistology of the hepatic egg granuloma. Parasite Immunol 24 :151–159.

    • Search Google Scholar
    • Export Citation
  • 25

    Qian BZ, Bogh HO, Johansen MV, Wang PP, 2000. Congenital transmission of Schistosoma japonicum in the rabbit. J Helminthol 74 :267–270.

  • 26

    Houba V, 1979. Experimental renal disease due to schistosomiasis. Kidney Int 16 :30–43.

  • 27

    Cheever AW, Kuntz RE, Moore JA, Bryan GT, Brown RR, 1976. Animal model of human disease: carcinoma of the urinary bladder in Schistosoma haematobium infection. Am J Pathol 84 :673–676.

    • Search Google Scholar
    • Export Citation
  • 28

    Sobh MA, Moustafa FE, Ramzy RM, Deelder AM, Ghoneim MA, 1991. Schistosoma haematobium-induced glomerular disease: an experimental study in the golden hamster. Nephron 57 :216–224.

    • Search Google Scholar
    • Export Citation
  • 29

    Smith T, Beck HP, Kitua A, Mwankusye S, Felger I, Fraser-Hurt N, Irion A, Alonso P, Teuscher T, Tanner M, 1999. Age dependence of the multiplicity of Plasmodium falciparum infections and of other malariological indices in an area of high endemicity. Trans R Soc Trop Med Hyg 93 :15–20.

    • Search Google Scholar
    • Export Citation
  • 30

    Roper C, Richardson W, Elhassan IM, Giha H, Hviid L, Satti GM, Theander TG, Arnot DE, 1998. Seasonal changes in the Plasmodium falciparum population in individuals and their relationship to clinical malaria: a longitudinal study in a Sudanese village. Parasitology 116 :501–510.

    • Search Google Scholar
    • Export Citation
  • 31

    Webbe G, James C, 1971. A comparison of two geographical strains of Schistosoma haematobium.J Helminthol 45 :271–284.

  • 32

    Thiongo FW, Madsen H, Ouma JH, Andreassen J, Christensen NO, 1997. Host-parasite relationships in infections with two Kenyan isolates of Schistosoma mansoni in NMRI mice. J Parasitol 83 :330–332.

    • Search Google Scholar
    • Export Citation
  • 33

    Fulford AJ, Mbugua GG, Ouma JH, Kariuki HC, Sturrock RF, Butterworth AE, 1991. Differences in the rate of hepatosplenomegaly due to Schistosoma mansoni infection between two areas in Machakos District, Kenya. Trans R Soc Trop Med Hyg 85 :481–488.

    • Search Google Scholar
    • Export Citation

 

 

 

 

URINARY TRACT PATHOLOGY ATTRIBUTED TO SCHISTOSOMA HAEMATOBIUM: DOES PARASITE GENETICS PLAY A ROLE?

View More View Less
  • 1 The W. Harry Feinstone Department of Molecular Microbiology & Immunology, and Department of International Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland; Blair Research Laboratory, Ministry of Health and Child Welfare, Harare, Zimbabwe

Disease outcome in persons infected with Schistosoma haematobium varies dramatically, ranging from mild symptoms to severe damage of the kidneys and/or bladder. We used ultrasonography to characterize the extent of urinary tract pathology of infected children in Zimbabwe, and random genetic markers to examine the relationship between genetic diversity of S. haematobium and clinical outcome. One hundred thirty-three parasite isolates from 12 students with mild lesions and 13 with severe lesions were compared. Using four randomly amplified polymorphic DNA (RAPD) markers, we scored parasite allelic frequencies at 53 loci. Although parasite heterogeneity did not differ, allelic frequencies at eight loci differed significantly between the mild and severe groups. Parasite isolates were analyzed further using a modified cluster analysis that segregated the population into 13 clusters of associated genotypes. Three clusters were significantly over-represented in children with severe lesions. Our findings, although preliminary, suggest that parasite genetic associations may be important in clinical outcome.

INTRODUCTION

Genetic variability among parasite populations is an important factor in their potential for producing harmful effects on the human populations they infect. Since damage from schistosome infections is so closely linked to the immune reaction to parasite eggs deposited in tissue, diversity of this infection may play an important role in development of pathology with heterogeneous versus homogeneous infections resulting in different clinical outcomes. Genetic differences may also lead to some strains being innately more immunogenic or fecund than others.

As techniques become available to assess genotypic associations at the population level, considerable variation has been noted in natural populations of schistosomes.1,2 Recent work in Zimbabwe demonstrated that schistosomes derived from snails along non-connected river systems showed substantial genetic diversity, with genetic distance increasing with geographic separation.3 Brouwer and others4 have shown that schistosomes found among Zimbabwean school children living in the same general area can be clustered into a series of related groups that share similar genotypes.

Few data exist on the genetic variability of the parasite Schistosoma haematobium in human populations and the potential existence of associated pathology. Therefore, the present study was undertaken to assess both genetic variability of the parasite and the level of pathology in students from a hyperendemic area of Zimbabwe by using ultrasound examination of urinary tract organs. Parasite samples were taken from a subset of students to determine the level of parasite genetic diversity in the area. Genetic markers were then compared with clinical outcome. By examining these issues, we endeavored to identify and characterize S. haematobium diversity within its definitive host and explore its possible implication on disease outcome.

MATERIALS AND METHODS

Study area.

The study covered a roughly 200 km2 section of the Chikwaka Communal Land in Zimbabwe (17°03′02″–17°13′26″S and 30°05′39″–31°10′10″E). This territory is approximately 80 km northeast of Harare in the Mashonaland Central province and has been mapped and described in detail elsewhere.5 Settlements are scattered among two main river systems. The site was chosen for its high prevalence of S. haematobium and low prevalence of S. mansoni.6,7 For more than a decade, no organized treatment campaign had been conducted in the area.

Population.

The population studied was composed of children 9–16 years of age from three primary schools. Criteria for inclusion of individuals in the study were 1) informed consent for ultrasound examination and consent to provide a series of fecal and urine specimens, 2) a minimum age of nine years, and 3) no obvious indicator of ill health or other confounding aspect of health status. Between June 1998 and May 1999, 551 children participated in the survey.

Patient identifiers on all information were coded so as to maintain privacy. All infected participants were given a single dose of 40 mg/kg of praziquantel (Biltricide®; Bayer, Ltd., Leverkussen, Germany) at the conclusion of the investigation at their school. Informed consent was obtained from adult participants and parents or legal guardians of minors involved in the study. The study was approved by the Medical Research Council of Zimbabwe and the Institutional Review Board of the Johns Hopkins Bloomberg School of Public Health.

Parasitologic measures.

From each participant, fecal specimens were screened for S. mansoni and geohelminths.8 Additionally, a total of three urine specimens per student were collected on alternating days to survey the prevalence of S. haematobium.9 Collections were made between 10:00 am and 2:00 pm to ensure maximum yield.10 Measurements were expressed as the arithmetic mean number of eggs passed per 10 mL of urine. To determine the presence and extent of hematuria and proteinuria, dipsticks (Hemastix®; Bayer Diagnostics, Fernwald, Germany) were also used on all urine specimens.

Ultrasound examinations.

To avoid confounding with the S. haematobium pathologic assessment, students with S. mansoni- or geohelminth-positive stool specimens (n = 73) and those from whom we did not receive a fecal specimen (n = 22) were excluded from the pathologic assessment. Of the remaining 456 students, 256 were infected with S. haematobium. Since the purpose of the study was to look for epidemiologic or genetic differences among those infected with urinary schistosomiasis, only 17% (33 of 190) of those uninfected were given ultrasound examinations compared with 71% (189 of 266) of those students infected with S. haematobium. The S. haematobium-infected students who did not undergo ultrasound examination, either because they did not show up on the assessment days or were lacking other parasitologic data, did not differ significantly in intensity of infection or major demographic characteristics from those who were given ultrasound examinations and those who had dual S. mansoni and S. haematobium infections.

Ultimately, ultrasonographic assessments were performed on 222 of the children with a portable ultrasound device (UF-5800A; Fukuda Denshi Co., Ltd., Tokyo, Japan) with a 3.5 MHz standard size convex probe. Bladders were examined when full and the ultrasonographer was unaware of patient infection status. A transverse measurement of the bladder was taken and pathology was classified according to thickening, presence of polyps, or the existence of masses protruding into the lumen. Such pathology was divided into the following categories based on World Health Organization:11 0 = no pathology: wall <5 mm; no masses, polyps, or thickenings (Figure 1A); 1 = mild: wall <5 mm; focal thickenings, no masses or polyps; 2 = severe: wall ≥5 mm; and/or masses or polyps (Figure 1B).

Kidneys and ureters were examined post-voiding. Kidney dilation was initially classified as follows:11 0 = no dilation (Figure 1C); 1 = <5 mm; 2 = 5–10 mm or 10–15 mm with >2 cm parenchyma; 3 = >15 mm with ≤2 cm parenchyma or end-stage with absence of parenchyma (Figure 1D). Because one kidney can compensate for the other, we further categorized the data according whether damage was unilateral or bilateral. These groupings were 0 = no pathology; 1 = mild: one kidney with grade 1 pathology and the other with grades 0 or 1; 2 = moderate: one kidney with grade 2 pathology and the other with grades 0 or 1; 3 = severe: both kidneys with grade 2 pathology or at least one kidney with grade 3. It is these latter categories that were used in construction of tables and calculation of statistics for kidney pathology.

Questionnaire.

Participants were questioned in regards to type and duration of water contact activities. A sense of the length of the current infection was determined by past or present occurrence of perceived symptoms as well as treatment history. Travel, age, weight, sex, and education information were also collected. All data were transferred from the collection sheets into Epi-Info 6.04b (Centers for Disease Control and Prevention, Atlanta, GA). Additionally, a worker from the Ministry of Health who had lived in the area interviewed students in regards to the approximate position of their residence and type and location of each water contact activity.

Parasite genetics.

Based on damage to the kidney or bladder, adequate urinary egg counts (for infecting laboratory-bred snails), and access to complete parasitologic information, 13 students with severe pathology and 12 with mild pathology were randomly selected from the two pathology groups for genetic characterization of their infections. Mild pathology was defined as bladder and kidney pathology each of grades 1 or less. A person was considered to have severe pathology if they had the highest grade of bladder pathology and/or the highest grade of kidney pathology. Two or three additional urine specimens were taken from each student for monomiracidial exposures of 48–96 snails per student.2,4 Snails were maintained through patency and cercariae were collected periodically. In all, 133 parasite isolates were obtained by this method. A polymerase chain reaction (PCR) was done with four random amplified polymorphic DNA (RAPD) primers on all cercarial material from patent infections.2,4 For each monomiracidial isolate, the presence or absence of bands at 53 different loci were entered into a Microsoft (Bellevue, WA) Excel® database.2

Analytical methods.

Genetic analyses were done as detailed by Shiff and others2 and Brouwer and others.4 Allele frequencies at each locus and overall heterozygosity were compared between isolates from severely and mildly infected patients using RAPDBIOS followed by BIOSYS-2.12,13 An analysis of molecular variance was calculated to find the contribution of each population to the total genetic variance.14 This is essentially an analysis of variance based on the genotype banding patterns produced by RAPD-PCR and is used to compare intra-population and inter-population variances.

We have introduced an additional procedure that is helpful in providing a picture of the genetic relationships among individuals within panmictic populations. It is based on a cluster analysis derived from matched pairs of alleles detected by RAPD-PCR. It does not represent permanent associations between related individuals, but gives an estimate of related groupings of the population at a particular time. As such, it is useful to examine the interrelationships of individual parasites and their hosts. We have termed these relationships clusters of associated genotypes.4 Genotypic clusters of the parasites were determined by analysis of loci with prevalence higher than 0.1 and less than 0.6, assuming Hardy-Weinberg proportions in the overall population. These values were chosen based on a previous study showing that loci with these frequencies optimally estimate the number of genetically related families in a panmictic population.15 In our field population, we do not know the parentage of isolates yet the groupings, whether representing true sibling families or simply closely related groups of schistosomes, are useful in comparing parasite genetics. Differences in the number and type of parasite clusters represented by the populations of parasites derived from mild and severe patients were compared. Essentially Apostol and others15 showed with the mosquito Aedes aegypti that it was feasible to use DNA fingerprinting (RAPD-PCR) to estimate the number of full sibling families in individuals arising from oviposition sites. The similarity of pairs of individuals are measured by examining the shared absence and presence of bands to estimate the fraction of matches (M) using the formula M = NAB/NT, where NAB is the total number of matches in individuals A and B (i.e., both bands absent or present) and NT is the total number of loci scored in the overall study. An M value of 1 indicates that two individuals have identical patterns and a value of 0 means they are completely different.

Values of M are calculated among all n(n - 1)/2 pairs of n individuals to develop an estimate of a value for 1 - M that will separate clusters of full siblings.15 The discriminating value averaging over all loci from the entire population is calculated using the program MCALC.13 This then is applied using those polymorphic loci to estimate the number of families (or clusters of associated genotypes when true parentage cannot be confirmed) represented by groups of related genotypes using the program FINGERS.13

Questionnaire and summarized genetic data were further analyzed in SPSS 10.0 for Windows (SPSS, Inc., Chicago IL). An odds ratio (OR), Pearson chi-square, or chi-square for trend was used, where appropriate, to measure associations among ordinal data. An analysis of variance was performed when the response variable was numeric. Correlation of variables was determined before fitting a multivariate logistic regression model using variables found significant in the univariate analysis. P values ≤0.05 using a two-sided test were considered significant.

RESULTS

Ultrasound examinations.

Results of ultrasound examination of the 222 children originally screened in this study have been detailed elsewhere.5 Briefly, of those with S. haematobium infection, 28% had no pathology, 22% had mild pathology of the bladder and/or kidneys, 15% had severe pathology of the bladder but no or mild pathology to the kidneys, 24% had moderate to severe kidney pathology but no or mild bladder pathology, and 11% had moderate to severe pathology of both the bladder and kidneys. Most of the 25 patients selected for genetic characterization of their infection belonged to either the first (n = 9, no pathology) or last (n = 10, moderate to severe pathology of both the bladder and kidneys) category. More specific details on each patient are included in the Clusters and pathology section.

Genetic frequencies.

Cercarial material derived from 73 miracidial isolates from those with severe infections were compared with 60 from those with little to no urinary system damage. Four RAPD primers produced 53 unambiguous loci, of which 22 (41.5%) were polymorphic. Details of banding patterns of these primers, including gel electrophoretic patterns, have been published elsewhere.2,4 When looking at the presence or absence of dominant bands, distributions at eight of the polymorphic loci differed appreciably between those derived from patients with severe or mild infections (χ2 = 35.4, P < 0.01) (Table 1). Allelic frequencies, after taking into account the effect of dominance and Hardy Weinberg proportions,12,13 yielded similar relationships.

Heterozygosity and pathology.

Although certain primers amplified more heterogeneous areas of the parasite genome than others, there was no difference in the average heterozygosity for the pathologic groups (Table 2). An analysis of molecular variance also showed no difference in the distribution of variance among the two populations.

Characterization of genetic clusters.

Using the program MCALC, alleles at 10 loci were found to have frequencies between 0.1 and 0.6. Grouping of polymorphic loci into genetically related clusters indicated that at least 13 clusters were circulating in the area at the time of the study (Table 3).4 Each cluster was present in anywhere from 2 to 12 persons among the 25 individuals examined.

The number of clusters per infected patient ranged from 2 to 9 (mean = 5.6). Males tended to have more clusters represented per number of isolates than females (0.75 for males; 0.54 for females; population mean ± SD = 0.68 ± 0.27). However, this trend was not statistically significant and disappeared entirely upon stratification by mean egg count. There was no apparent relationship between the distribution of clusters and the age of participants, nor was there any indication of geographic isolation because most of these associated groups could be found even in participants from opposite ends of the study area. Likewise, treatment and travel history of the students did not obviously influence the number and type of parasites making up individual infections.

Clusters and pathology.

The arithmetic mean parasite egg count was 217 eggs/10 mL of urine (median = 140) for those with severe urinary tract pathology compared with 76 eggs/10 mL of urine (median = 62) for those with mild pathology (F = 5.15, P = 0.03). Persons with higher egg counts tended to have more parasite clusters represented in their infection (χ2 for trend = 3.10, P = 0.08 (Figure 2). Those with more diverse infections (clusters/number of isolates) were also more likely to have proteinuria (mean ± SD clusters/total isolates = 0.76 ± 0.23) compared with those without proteinuria (0.41 ± 0.17; F = 9.32, P < 0.01). However, there was no significant difference in number of parasite clusters per infection for severe versus mild pathology as assessed by ultrasound.

One interesting phenomenon was the distribution of certain clusters of associated genotypes among severely and mildly affected individuals. Three parasite genetic clusters (1, 5, and 8) were present almost exclusively in individuals with severe lesions, and one (cluster 7) was over-represented in milder infections (Tables 3 and 4). As mentioned earlier, 10 polymorphic loci were used to distinguish the clusters of associated genotypes and banding patterns at five of these loci differed significantly between parasites from severely affected and mildly affected hosts (those marked with an † in Table 1). In fact, the banding patterns at these five loci were completely inverted when comparing members of cluster 1 (which occurred exclusively in severely affected patients) versus cluster 7 (found mainly in mildly affected persons) (Table 4). In each case, the banding patterns followed the trend expected for severe and mild groups as detailed in Table 1. A dendrogram showing the relationship between representatives of each of the 13 clusters is published elsewhere.4

Previous work from our laboratory showed that a number of other variables (risk factors or symptoms), in addition to parasite genetics, were significantly associated with clinical outcome of S. haematobium pathology in this population.5 From the questionnaire, information on the distribution of these variables in those with or without clusters 1, 5, or 8 were compared (Table 5). Although none of the relationships were statistically significant, those with clusters 1, 5, or 8 were more likely to have pain upon urination, spend time fishing in rivers in the area, and have a history of previous S. haematobium infection. To control for the effect of possible confounders of pathology and better measure the effect of parasite genetics on clinical outcome, we performed a multivariate logistic regression controlling for demographic (age), parasitologic (number of isolates analyzed per person), exposure (fishing, playing), and historical (infected before) factors. This regression further demonstrated that children with parasites from clusters 1, 5, or 8 had an increased chance of severe urinary tract pathology compared with those without parasites from these clusters (adjusted OR = 3.95, 95% confidence interval = 1.22–12.72, P = 0.021) (Table 6).

DISCUSSION

The association between schistosomiasis haematobia and urinary tract pathology has been well established. From longitudinal studies, bladder wall pathology and, to some extent, hydronephrosis of the kidneys have been found to regress upon treatment of S. haematobium infection.16,17 The association of this parasite with squamous cell carcinoma, the only form of bladder cancer with a parasitologic etiology, has been verified through hospital and laboratory-based studies.18–20 Such bladder or kidney pathology does not develop uniformly in those infected. Differences in extent of urinary tract pathology associated with schistosome infection have been noted for quite some time among regions of Africa and even within communities.21,22 Whether differences in severity of disease are influenced by or are due to parasite genetic variance has yet to be determined.

With new methods to evaluate the clinical outcome of schistosomiasis in humans, such as portable ultrasound machines, and ever improving techniques for molecular characterization of the parasite, we embarked upon a study of S. haematobium genetic diversity in humans. The current study, although limited in size, indicates that the extent of S. haematobium diversity in humans reflects that seen in the intermediate snail host and that there are significant associations between certain genetic clusters of the parasite and clinical outcome.

While a number of animal models have been developed to explore pathology caused by S. mansoni and S. japonicum,23–25 developing an animal model for S. haematobium has been complicated by the fact that it is primarily an anthroponosis. The laboratory animal models created thus far have failed to duplicate the type of bladder wall pathology and kidney pathology (due to obstruction of the ureters rather than immune complexes) seen in humans.26–28 We were able to avoid this issue by exploring genetic diversity of S. haematobium, and its possible relationship to pathology, directly in school children from a single communal land in Zimbabwe. Although limited by the fact that we cannot control host factors, such as differences in exposure and host genetics, we have attempted to minimize such confounding by controlling for variables from our questionnaire that were related to pathology. From miracidia obtained from 25 students, 133 distinct parasite isolates were amplified in snails and subsequently characterized by RAPD-PCR. Banding patterns at 53 loci showed moderate diversity of the local parasite population, with a mean heterozygosity of 0.116. This figure is similar to that seen with different RAPD primers in a survey of schistosome genotypes in Bulinids.1

For Plasmodium falciparum, it has been suggested that clinical malaria is possibly related to the human immunologic response when encountering new strains of the parasite.29,30 Similarly, with the immune response to parasite eggs responsible for most schistosomiasis-related pathology, increased diversity of schistosome genotypes could lead to repeated activation of the immune system with more or larger granulomas developing. When we assessed whether diversity of S. haematobium infection was greater for those with as compared with those without severe pathology, our findings showed no relationship between either heterozygosity or number of parasite clusters per infection and urinary tract pathology (Tables 2 and 3). There was, however, a trend of increasing diversity as the intensity of infection (as estimated by urinary egg counts) increased (Figure 2). This may be related to the behaviors that led to development of such intense infection in the first place. If intensely infected students participated in repeated and extended water contact activities, the chance for exposure to a number of parasite genotypes would likely increase. This was in accordance with the observation that males, who had more water contact and more intense infections than females, also tended to have greater numbers of parasite clusters per number of isolates obtained. Those with more diverse infections (clusters/number of isolates) were also more likely to have proteinuria.

Virulence of particular parasite strains, rather than diversity of an infection, is another possible contributor to disease development. Genetic differences may lead to some strains being innately more immunogenic or fecund than others. In exploring this hypothesis, it was found that at a number of loci there were significant differences in frequencies of dominant bands between samples derived from mildly or severely infected persons (Table 1). Upon dividing the population into clusters of associated genotypes using a technique developed by Apostol and others,15 it was found that three clusters occurred almost exclusively in severe infections and one was over-represented in those with mild infections (Tables 3 and 4). Inspection of allelic distributions for each of these four clusters revealed that cluster 1 (severe) and cluster 7 (mainly mild) had inverse genotypes at loci significantly different between pathologic groups (Table 1, loci marked by an †). These findings support the notion that particular parasite strains or genetic factors may be associated with clinical outcome. The actual reasons behind these pathologic differences may be many. Antigenic differences may lead to more or less intense immune reactions to the infection. Another possibility is that certain parasite strains may be more fecund than others. Mean egg counts for those harboring clusters over-represented in severe infections were considerably higher (although not significantly so) than the mean egg count for cluster 7, which was over-represented in mild infections (Table 4). Increased fecundity of parasites in these clusters could be the reason behind the increased pathology in hosts harboring them.

The possible implications of these genetic differences between isolates derived from mild or severe infections warrant further investigation. Laboratory studies have shown that S. haematobium infections may differ under controlled circumstances, where factors such as exposure to parasites, infection intensity, nutrition, and host genetics are kept as similar as possible. Comparisons between Nigerian and Iranian strains showed differential mortality and worm recovery in hamsters. They also differed in degree of snail infectivity.31 More recently Thiongo and others32 investigated S. mansoni strains from two districts in Kenya known for their disparate rates of hepatomegaly and splenomegaly despite comparable fecal egg output.33 When the two strains were passaged through mice, significant differences were found in the egg excretion/ egg retention ratio for mice (P < 0.001), with the strain that had less retention corresponding to the area with less human pathology. Although our study in human subjects did not allow for strict control of host factors, as can be done in animal studies, we were able to control statistically for a number of other variables associated with pathology (Table 6). In our multivariate model, parasite genetics remained a significant contributor to clinical outcome (OR = 3.95, P = 0.021) (Table 6).

To our knowledge the present investigation is the first study to look at genetic diversity of S. haematobium within schoolchildren. It is also the first to compare the distribution of S. haematobium genotypes in the definitive host with clinical outcome. While limited in size and scope, the study provides evidence of parasite factors significantly associated with disease outcome. Further characterization of the parasite clusters associated with severe clinical outcome and comparison to the one cluster with over-representation in mild infections may reveal a possible genetic mechanism for parasite virulence. Since one cannot rule out the possibility that pathology develops over time and that the parasites that make up the current infection are not necessarily responsible for all pathology seen, an expanded, longitudinal study would be required to confirm associations between particular parasite clusters and morbidity. Given the disparity in clinical outcomes and the possible importance of genetic variability when considering drug or vaccine targets, clearly there is great need to learn more about the genetics of this parasite.

Table 1

Allelic distribution at eight loci (percentage of miracidial isolates with a band present at that locus) by pathologic status*

G11-350G11-360†G11-410G11-430†G11-760†Z08-670Z08-1070†Z08-1090†
* Statistics refer to the strength of the difference between the severe and mild categories.
† Locus used in calculation of genetic clusters.
Severe (n = 73)9734864986217754
Mild (n = 60)855897287274971
χ26.607.914.315.644.045.1810.84.15
P0.01<0.010.040.020.050.02<0.010.04
Table 2

Mean heterozygosity (standard error) for each primer and pathologic group

PrimerMildSevere
A020.043 (0.027)0.038 (0.027)
A120.103 (0.067)0.103 (0.068)
G110.206 (0.048)0.229 (0.043)
Z080.108 (0.046)0.115 (0.046)
Overall0.120 (0.044)0.104 (0.056)
Table 3

Distribution of parasite clusters in relation to disease severity

Cluster
Patient no.Pathology* status12345678910111213Bladder pathology†Kidney pathology‡Mean egg count§
* S = severe; M = mild.
† 0 = none; 1 = mild; 2 = severe.
‡ 0 = none; 1 = mild; 2 = moderate; 3 = severe.
§ Arithmetic mean per 10 ml of urine.
122SX2235.3
179SXXXX2240.7
193SXXXXX23290.0
230SXXXX23354.7
234SX22489.0
237SX2021.3
276SX2139.0
354SXXXXXXX23544.7
379SXXXXXXX13232.0
507SXXXXXXXXX2391.3
560SXXX22140.0
564SX2276.0
588SXXXXX22466.3
178MXXX007.3
182MXX1073.7
187MXX004.0
216MX00367.3
218MX006.0
227MX0059.7
265MXXX0025.0
283MX002.0
329MXXXXXX0090.0
441MXXXXX1064.7
448MXXXX0080.7
511MXXXXXX01125.7
Table 4

Clusters overrepresented in pathologic groups*

Cluster 1Cluster 5Cluster 7Cluster 8
* Percent severe refers to the number with severe pathology divided by the total number of persons with that cluster. Arithmetic mean egg count is per 10 ml of urine.
% Severe100 (7/7)88 (7/8)17 (1/6)86 (6/7)
χ28.975.943.954.43
P<0.010.020.060.05
Arithmetic mean egg count (median)231 (232)211 (116)82 (49)253 (290)
Table 5

Characteristics of children with and without parasites from severe clusters (1, 5, and 8)*

%
Persons with clusters 1, 5, or 8 (n = 13)All others (n = 12)P
* Unless otherwise indicated, values are percentages.
Sex
    Male30.841.70.57
Water contact
    Fishing61.533.30.16
    Bathing61.566.70.79
    Playing30.841.70.57
School
    Mauhudzi53.866.70.57
    Chipangura38.533.3
    Nyagui7.70
Disease history
    Infected before84.658.30.14
    Treated before45.555.60.65
Age (years)
    9–1038.533.30.97
    11–1246.250.0
    ≥1315.416.7
Parasite isolates
    Isolates/person ± SE5.85 ± 1.494.75 ± 1.380.60
Parasite egg count
    <5030.841.70.37
    50–30038.550.0
    >30030.88.3
Symptoms
    Dysuria66.741.70.22
    Macrohematuria83.375.00.62
    Microhematuria84.691.70.59
    Proteinuria84.666.70.29
Table 6

Multivariate analysis of the effect of parasite genotype cluster on clinical outcome

GenotypeSevere pathology n = 13 (%)Mild pathology n = 12 (%)P*Adjusted OR for odds of HIV transmission† (95% CI)Adjusted P
* By chi-square test.
† Derived from multivariate logistic regression, controlling for the confounding effect of age, water contact (fishing, playing), number of parasite isolates analyzed per person, and infection history (infected before). OR = odds ratio; HIV = human immunodeficiency virus; CI = confidence interval.
Clusters 1, 5, or 811 (84.6)2 (15.4)0.0013.95 (1.22–12.72)0.021
Other clusters2 (16.7)10 (83.3)
Figure 1.
Figure 1.

Ultrasonography demonstrating the degree of pathology in the bladder (A and B) or kidney (C and D). A, Fully extended bladder with no apparent pathology. B, Bladder with masses extending into the lumen (arrows) (classified as severe). C, Kidney with no abnormalities. D, Kidney with severe dilation (arrows) and < 2 cm parenchyma.

Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 68, 4; 10.4269/ajtmh.2003.68.456

Figure 2.
Figure 2.

Number of parasite clusters in relation to urinary egg counts for those with four or more miracidial isolates. Bars show the mean ± SD.

Citation: The American Journal of Tropical Medicine and Hygiene Am J Trop Med Hyg 68, 4; 10.4269/ajtmh.2003.68.456

Authors’ addresses: Kimberly C. Brouwer and Clive J. Shiff, The W. Harry Feinstone Department of Molecular Microbiology & Immunology, Bloomberg School of Public Health, 615 N. Wolfe Street, Johns Hopkins University, Baltimore, MD 21205, Telephone: 410-955-1263, Fax: 410-955-0105, E-mail: cshiff@jhsph.edu. Patricia D. Ndhlovu and Anderson Munatsi, Blair Research Laboratory, Ministry of Health and Child Welfare, Harare, Zimbabwe. Yukiko Wagatsuma, Department of International Health, Bloomberg School Public Health, Johns Hopkins University, Baltimore, MD 21205.

Acknowledgments: We thank the field staff of the Blair Research Laboratory who facilitated collection and processing of parasitologic samples. We are very grateful to the Chikwaka District Health Office, the staff of the Bosha Rural Health Centre, and the Nyagui, Chipangura, and Mavhudzi schools for their cooperation with our surveys. Thanks are due to the Japan International Cooperation Agency, which provided us with the ultrasound machine for examinations. Special thanks are given to Dr. Fred Lewis, Dr. Yung-San Liang, and Francis Barnes (Biomedical Research Institute, Rockville, MD) for providing snails and parasite material for the study.

Financial support: This work received financial support from the National Institutes of Health (grant 1 RO3 DK53207-01 and training grant T32 AI-07417) and the J. William Fulbright Fellowship program.

REFERENCES

  • 1

    Dabo A, Durand P, Morand S, Diakite M, Langand J, Imbert-Establet D, Doumbo O, Jourdane J, 1997. Distribution and genetic diversity of Schistosoma haematobium within its bulinid intermediate hosts in Mali. Acta Trop 66 :15–26.

    • Search Google Scholar
    • Export Citation
  • 2

    Shiff C, Brouwer KC, Clow L, 2000. Schistosoma haematobium: population genetics of S. haematobium by direct measurement of parasite diversity using RAPD-PCR. Exp Parasitol 96 :47–51.

    • Search Google Scholar
    • Export Citation
  • 3

    Davies CM, Webster JP, Kruger O, Munatsi A, Ndamba J, Woolhouse ME, 1999. Host-parasite population genetics: a cross-sectional comparison of Bulinus globosus and Schistosoma haematobium.Parasitology 119 :295–302.

    • Search Google Scholar
    • Export Citation
  • 4

    Brouwer KC, Ndhlovu P, Munatsi A, Shiff CJ, 2001. Genetic diversity of a population of Schistosoma haematobium derived from schoolchildren in east central Zimbabwe. J Parasitol 87 :762–769.

    • Search Google Scholar
    • Export Citation
  • 5

    Brouwer KC, Ndhlovu P, Wagatsuma Y, Munatsi A, Shiff CJ, 2003. Epidemiological assessment of Schistosoma haematobium induced kidney and bladder pathology in rural Zimbabwe. Acta Trop 85 :339–347.

    • Search Google Scholar
    • Export Citation
  • 6

    Taylor P, Makura O, 1985. Prevalence and distribution of schistosomiasis in Zimbabwe. Ann Trop Med Parasitol 79 :287–299.

  • 7

    Ndhlovu P, Cadman H, Gundersen S, Vennervald BJ, Friis H, Christensen NO, Mutasa G, Kaondera K, Mandaza G, Deelder AM, 1996. Circulating anodic antigen (CAA) levels in different age groups in a Zimbabwean rural community endemic for Schistosoma haematobium determined using the magnetic beads antigen-capture enzyme-linked immunoassay. Am J Trop Med Hyg 54 :537–542.

    • Search Google Scholar
    • Export Citation
  • 8

    Katz N, Chaves A, Pellegrino J, 1972. A simple device for quantitative stool thick-smear technique in Schistosomiasis mansoni. Rev Inst Med Trop Sao Paulo 14 :397–400.

    • Search Google Scholar
    • Export Citation
  • 9

    Mott KE, Baltes R, Bambagha J, Baldassini B, 1982. Field studies of a reusable polyamide filter for detection of Schistosoma haematobium eggs by urine filtration. Trop Med Parasitol 33 :227–228.

    • Search Google Scholar
    • Export Citation
  • 10

    Weber MD, Blair DM, Clarke V deV, 1967. The pattern of schistosome egg distribution in a micturition flow. Cent Afr J Med 13 :75–88.

  • 11

    WHO/TDR, 1996. Ultrasound in Schistosomiasis. International Workshop on the Use of Ultrasonography in Relation to Schistosomiasis. Niamey, Niger: CERMES.

  • 12

    Swofford DL, Selander RB, Black WC, 1997. BIOSYS-2: A Computer Program for the Analysis of Allelic Variation in Genetics. Urbana, IL: University of Illinois and Fort Collins, CO: Colorado State University.

  • 13

    Black WC, 1998. FORTRAN Programs for the Analysis of RAPD-PCR Markers in Populations. Fort Collins, CO: Colorado State University.

  • 14

    Excoffier L, Smouse PE, Quattro JM, 1992. Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics 131 :479–491.

    • Search Google Scholar
    • Export Citation
  • 15

    Apostol BL, Black WC, Miller BR, Reiter P, Beaty BJ, 1993. Estimation of the number of full sibling families at an oviposition site using RAPD-PCR markers: applications to the mosquito Aedes aegypti.Theor Appl Genet 86 :991–1000.

    • Search Google Scholar
    • Export Citation
  • 16

    Subramanian AK, Mungai P, Ouma JH, Magak P, King CH, Mahmoud AA, King CL, 1999. Long-term suppression of adult bladder morbidity and severe hydronephrosis following selective population chemotherapy for Schistosoma haematobium.Am J Trop Med Hyg 61 :476–481.

    • Search Google Scholar
    • Export Citation
  • 17

    Wagatsuma Y, Aryeetey ME, Sack DA, Morrow RH, Hatz C, Kojima S, 1999. Resolution and resurgence of Schistosoma haematobium-induced pathology after community-based chemotherapy in Ghana, as detected by ultrasound. J Infect Dis 179 :1515–1522.

    • Search Google Scholar
    • Export Citation
  • 18

    Mostafa MH, Sheweita SA, O’Connor PJ, 1999. Relationship between schistosomiasis and bladder cancer. Clin Microbiol Rev 12 :97–111.

  • 19

    Vizcaino AP, Parkin DM, Boffetta P, Skinner ME, 1994. Bladder cancer: epidemiology and risk factors in Bulawayo, Zimbabwe. Cancer Causes Control 5 :517–522.

    • Search Google Scholar
    • Export Citation
  • 20

    Thomas JE, Bassett MT, Sigola LB, Taylor P, 1990. Relationship between bladder cancer incidence, Schistosoma haematobium infection, and geographical region in Zimbabwe. Trans R Soc Trop Med Hyg 84 :551–553.

    • Search Google Scholar
    • Export Citation
  • 21

    Pugh RN, Bell DR, Gilles HM, 1980. Malumfashi Endemic Diseases Research Project, XV. The potential medical importance of bilharzia in northern Nigeria: a suggested rapid, cheap and effective solution for control of Schistosoma haematobium infection. Ann Trop Med Parasitol 74 :597–613.

    • Search Google Scholar
    • Export Citation
  • 22

    Elem B, Vandal MT, 1981. Bilharziasis of the urinary tract in Zambia. (Observation on 100 consecutive cases). Med J Zambia 15 :48–51.

  • 23

    Chisty MM, Nargis M, Sato H, Inaba T, Takahashi G, Kamiya H, 2002. Schistosoma mansoni: kinetics of glomerulonephritis in Mongolian gerbils and its correlation with intensity and duration of infection. Parasite 9 :143–151.

    • Search Google Scholar
    • Export Citation
  • 24

    Hurst MH, Willingham AL, Lindberg R, 2002. Experimental schistosomiasis japonica in the pig: immunohistology of the hepatic egg granuloma. Parasite Immunol 24 :151–159.

    • Search Google Scholar
    • Export Citation
  • 25

    Qian BZ, Bogh HO, Johansen MV, Wang PP, 2000. Congenital transmission of Schistosoma japonicum in the rabbit. J Helminthol 74 :267–270.

  • 26

    Houba V, 1979. Experimental renal disease due to schistosomiasis. Kidney Int 16 :30–43.

  • 27

    Cheever AW, Kuntz RE, Moore JA, Bryan GT, Brown RR, 1976. Animal model of human disease: carcinoma of the urinary bladder in Schistosoma haematobium infection. Am J Pathol 84 :673–676.

    • Search Google Scholar
    • Export Citation
  • 28

    Sobh MA, Moustafa FE, Ramzy RM, Deelder AM, Ghoneim MA, 1991. Schistosoma haematobium-induced glomerular disease: an experimental study in the golden hamster. Nephron 57 :216–224.

    • Search Google Scholar
    • Export Citation
  • 29

    Smith T, Beck HP, Kitua A, Mwankusye S, Felger I, Fraser-Hurt N, Irion A, Alonso P, Teuscher T, Tanner M, 1999. Age dependence of the multiplicity of Plasmodium falciparum infections and of other malariological indices in an area of high endemicity. Trans R Soc Trop Med Hyg 93 :15–20.

    • Search Google Scholar
    • Export Citation
  • 30

    Roper C, Richardson W, Elhassan IM, Giha H, Hviid L, Satti GM, Theander TG, Arnot DE, 1998. Seasonal changes in the Plasmodium falciparum population in individuals and their relationship to clinical malaria: a longitudinal study in a Sudanese village. Parasitology 116 :501–510.

    • Search Google Scholar
    • Export Citation
  • 31

    Webbe G, James C, 1971. A comparison of two geographical strains of Schistosoma haematobium.J Helminthol 45 :271–284.

  • 32

    Thiongo FW, Madsen H, Ouma JH, Andreassen J, Christensen NO, 1997. Host-parasite relationships in infections with two Kenyan isolates of Schistosoma mansoni in NMRI mice. J Parasitol 83 :330–332.

    • Search Google Scholar
    • Export Citation
  • 33

    Fulford AJ, Mbugua GG, Ouma JH, Kariuki HC, Sturrock RF, Butterworth AE, 1991. Differences in the rate of hepatosplenomegaly due to Schistosoma mansoni infection between two areas in Machakos District, Kenya. Trans R Soc Trop Med Hyg 85 :481–488.

    • Search Google Scholar
    • Export Citation
Save