Am. J. Trop. Med. Hyg., 60(5), 1999, pp. 749±751 Copyright ᭧ 1999 by The American Society of Tropical Medicine and Hygiene

SHORT REPORT: IMMUNODIAGNOSIS OF HUMAN FASCIOLIASIS USING RECOMBINANT FASCIOLA HEPATICA CATHEPSIN L1 CYSTEINE PROTEINASE

SANDRA M. O'NEILL, MICHAEL PARKINSON, ANDREW J. DOWD, WILMA STRAUSS, RENE ANGLES, AND JOHN P. DALTON School of Biological Sciences, Dublin City University, Dublin, Ireland; Unidad de Parasitologia, Instituto Nacional de Laboratorios de Salud, Secretaria Nacional de Salud, La Paz, Bolivia

Abstract. Our laboratory recently developed a diagnostic test (ELISA) for human fascioliasis based on the detec- tion of serum IgG4 antibodies reactive with Fasciola hepatica cathepsin L1 (CL1). In the present study, we have used recombinant CL1, generated by functional expression of the cDNA in Saccharomyces cerevisiae, in this im- munodiagnostic test and compared its performance with native CL1. Sera obtained from 64 individuals living in Cutusuma village in the northern Altiplano of Bolivia, a region with a high prevalence of human fascioliasis, were analyzed by the IgG4-ELISA. A highly statistically signi®cant correlation (r2 ϭ 0.751, P Ͻ 0.001) was demonstrated between the absorbances obtained using the recombinant and native proteins. These assays showed that 38 (59%) of the individuals tested were seropositive for fascioliasis, whereas only 26 of them were coprologically positive for F. hepatica eggs. All seronegative patients were also coprologically negative. Serum from individuals infected with schistosomiasis mansoni, cysticercosis, hydatidosis, and Chagas disease did not contain antibodies reactive with the recombinant or native CL1. Therefore, recombinant CL1 shows excellent potential for the development of the ®rst standardized assay for the sensitive and speci®c diagnosis of human fascioliasis. Finally, our data supports earlier reports on the high prevalence of human fascioliasis in the Bolivian Altiplano, which collectively suggest that the disease has been endemic there for more than a decade.

Fascioliasis is caused by liver ¯ukes of the genus Fasci- Ϯ SD age of 15.23 Ϯ 12.46 years. Coprologic analysis for ola. Although the disease is predominantly one of domestic F. hepatica eggs was performed on fecal samples obtained animals such as sheep and cattle, it is now emerging as an from all individuals using the sedimentation method of Rich- important chronic disease of humans.1±3 Recent estimates ie.11 Serum samples obtained from patients infected with suggest that as many as 2.4 million people are infected with schistosomiasis mansoni (20), cysticercosis (15), hydatidosis the parasite worldwide and 2.4 million are at risk.3,4 The (15) and Chagas disease (15) were obtained from the Insti- disease is particularly prevalent among the Aymaran people tuto Nacional de Laboratorios de Salud (INLASA) serum of the northern Altiplano region of Bolivia, with incidences library and used to test for cross-reactivity in the test. Con- of up to 60% being reported.5±8 Our laboratory recently de- trol serum samples were provided by eight volunteers at veloped a simple ELISA for the detection of serum IgG4 Dublin City University. The study was approved by the Hu- antibodies reactive with F. hepatica cathepsin L1 (CL1) pro- man Ethics Committee of INLASA and the Department of teinase.7 The use of puri®ed CL1 antigen in this ELISA pro- Health of Bolivia. vided a more speci®c immunodiagnosis of human fascioli- The ELISAs were carried out as previously described by asis than tests that use parasite somatic or excretory/secre- O'Neill and others.7 Brie¯y, 100 ␮l of native or recombinant tory (ES) antigen. CL1 antigen (5 ␮g/ml) was incubated overnight at 37ЊCin We have also recently reported the expression of func- wells of microtiter plates (Nuclon, Kamstrup, Roskilde, Den- tionally active recombinant F. hepatica CL1 in the yeast mark). The wells were blocked with a solution of 2% bovine Saccharomyces cerevisiae.9 In the present study, we exam- serum albumin/0.1% Tween 20 in phosphate-buffered saline ined the potential of this recombinant protein as a diagnostic for 30 min at 37ЊC and human sera (1:125 dilution) was then reagent for human fascioliasis in the IgG4-ELISA and com- added into triplicate wells. Bound human antibodies were pared the performance of the antigen with native CL1 anti- detected using biotin-conjugated anti-human IgG4 (1:2,000 gen. Recombinant CL1 was puri®ed from yeast culture me- dilution), avidin-conjugated alkaline phosphatase, and the dium by gel ®ltration chromatography on Sephacryl S200 substrate 2,2-azino-bis (3-ethylbenzthiazoline-6-su¯onic HR,9 and native CL1 was puri®ed from ES products by a acid) (all from the Sigma Chemical Co., Poole, Dorset, Unit- combination of gel ®ltration chromatography on Sephacryl ed Kingdom). Plates were read after 15 min on a Titerteck S200 HR and ion exchange chromatography on QAE-Se- (Helsinki, Finland) multiscan ELISA plate reader at an ab- phadex.10 The concentration of each antigen preparation was sorbance of 405 nm. measured using a bicinchoninic acid protein assay kit (Pierce The absorbance readings obtained from the ELISAs using and Warner, Chester, United Kingdom). recombinant and native CL1 were plotted as scattergrams The nature of the study was explained to each individual, (Figure 1). Hierarchical agglomerative cluster analysis re- and prior to volunteering to provide blood and feces, each vealed that the scatter of points divided into two clusters, person signed a document of consent that also carried their one compact cluster of low absorbance readings and a dif- name and personal details. Blood samples were obtained in fuse cluster of high absorbance readings (Figure 1). Cut-off 1986 from 64 native Aymara living in the village of Cutu- points between the clusters were set at 3.09 standard devi- suma (approximately 50 km north of La Paz, Bolivia). The ations from the mean of the low absorbance cluster and were serum was separated and stored at Ϫ20ЊC. The age of the 0.117 and 0.121 for the recombinant and native proteins, volunteers ranged between seven and 76 years, with a mean respectively. Using the cut-off points, 38 individuals (59%) 749 750 O'NEILL AND OTHERS

TABLE 1 Summary of the serologic and coprologic data obtained from 64 individuals in the village of Cutusuma, Bolivia (1986)*

Group Recombinant CL1 Native CL1 Eggϩ Seropositive 26 26 Eggϩ Seronegative 0 0 EggϪ Seropositive 12 12 EggϪ Seronegative 26 26 Total 64 64 * CL1 ϭ cathespin L1.

cercosis, hydatidosis, and Chagas disease were negative by IgG4-ELISAs that used either recombinant or native CL1. Although schistosomes also express cathepsin L proteinas- es,12 we have shown that antibodies prepared against F. he- FIGURE 1. Scattergram of the data obtained with the IgG4- patica cathepsin L proteinases do not cross-react with these ELISA using recombinant cathespsin L1 (CL1) and native CL1 an- (Brady C, Dalton JP, unpublished data). tigens. Hierarchical agglomerative cluster analysis using the com- Human fascioliasis is becoming increasingly recognized bined data obtained for both proteins separated the population into as a serious public health problem, particularly in regions of low absorbance (seronegative) and high absorbance (seropositive) subpopulations. The vertical and horizontal dashed lines indicate the Bolivia and Peru. Regional climatic factors coupled with the calculated cut-off points for each antigen. The closed circles repre- rural lifestyle and dietary habits of the indigenous population sent individuals who were coprologically positive and the open cir- contribute to transmission of the disease year round.6,8,13 cles represent those who were coprologically negative. The diagonal Clinical manifestations of human fascioliasis include fever, line represents the regression of seropositive individuals for recom- binant CL1 against native CL1. right hypochondrial pain, anorexia, weight loss, persistent diarrhea, and vomiting. However, many individuals are asymptomatic or present vague symptoms rendering clinical were seropositive and 26 were seronegative by the IgG4- diagnosis problematic.1 Accordingly, the development of a ELISA using both the recombinant and native CL1 (Figure simple, sensitive, and cost-effective diagnostic assay is im- 1 and Table 1). Chi-square analysis on whether individuals perative. While coprologic analysis may ful®ll some of these were seronegative or seropositive showed a highly signi®- requirements, it is clear that it provides a de®nitive parasi- cant relationship between the two proteins. Regression and tologic diagnosis only when eggs are found in the patient's correlation analysis of the data for the seropositive individ- feces, and that early infections cannot be detected using this uals showed that there was a highly linear relationship be- method. tween the two antigens (r2 ϭ 0.751, P Ͻ 0.001) (Figure 1). Cathepsin L1 was considered a potential candidate for se- All patients who were coprologically positive (26) were rologic diagnosis of fascioliasis because it was secreted by also seropositive in the IgG4-ELISA using the recombinant all stages of the parasite that develop in the mammalian host or native proteins. Chi-square analysis showed that it was and was highly immunogenic in infected animals (antibodies highly unlikely (P Ͻ 0.001) that the coprologic and serolog- can be detected within two weeks after infection).10 O'Neill ically positive individuals would group together by chance and others7 demonstrated that this de®ned antigen discrimi- and thus reinforced the designation of the high absorbance nated between seropositive and seronegative individuals for cluster as the infected subpopulation. However, 12 copro- fascioliasis markedly better than crude antigens such as logically negative patients were identi®ed as seropositive. whole worm extracts of total ES products. Both the present These individuals may harbor early (prepatent) infections, study and that of O'Neill and others7 have shown that all since the appearance of eggs in the feces only follows the patients who were coprologically positive for F. hepatica maturation of the parasite in the bile duct, approximately 12 eggs were also serologically positive in the IgG4-ELISA that weeks after infection. An early diagnosis of fascioliasis is used CL1 as antigen. The fact that all seronegative patients important since it is the juvenile stages of the parasite that were also coprologically negative indicated that CL1 could cause extensive damage to the liver tissue (perforations and be used for the sensitive diagnosis of human fascioliasis, and hemorrhaging) as they migrate to the bile ducts. Another reassured us that those coprologically negative individuals possibility is that our coprologic analysis may have missed that were seropositive were indeed infected. some infected individuals since eggs are irregularly released Roche and others9 showed that recombinant CL1 pro- with the bile juices into the intestines. A de®nitive copro- duced in yeast was not only functionally active but also ex- logic analysis requires the examination of at least two fecal hibited similar physicochemical properties as the native samples taken at different times of the day. In our ®eld stud- CL1, sharing similar molecular size, pH pro®le of activity, ies, it was possible to obtain only a single fecal sample per stability, and substrate speci®city. Therefore, it is not overly patient. surprising that the recombinant antigen performed similarly The 26 individuals that were seronegative by the IgG4- (as shown by a regression line slope of 0.797, Figure 1) to ELISA were also coprologically negative, indicating that the the native antigen in the IgG-ELISAs described in this study. assay does not give false-positive results (Table 1). Sera However, the important development in this study was the from patients infected with schistosomiasis mansoni, cysti- demonstration that the possibility of producing suf®cient IMMUNODIAGNOSIS OF HUMAN FASCIOLIASIS 751 quantities of antigen to provide all diagnostic centers with a and aberrant forms of the disease. Medicine (Baltimore) 74: standardized serologic test for human fascioliasis exists. 13±23. 3. Hillyer GV, Apt W, 1997. Food borne trematode infections in In this study, we have con®rmed the high prevalence of the Americas. Parasitol Today 13: 87±88. human fascioliasis in the northern Altiplano region of Bo- 4. Anonymous, 1995. Control of foodborne trematode infections. livia. A large-scale regional survey is required to fully es- World Health Organ Tech Rep Ser 849. tablish the prevalence of the disease throughout the entire 5. Hillyer GV, De Galanes MS, Rodriguez-Perez J, De Lagrava MS, Bjorland J, de Lagrava MS, Ramirez Guzman S, Bryan region. Studies carried out to date have indicated a preva- RT, 1992. Use of the Falcon௢ assay screening test-enzyme- lence of human fascioliasis of approximately 59% in the linked immunosorbent assay (FAST-ELISA) and the enzyme- village of Cutusuma in 1986 (Table 1, this study), 49% in linked immunoelectrotransfer blot (ETIB) to determine the Calasaya in 1991,6,7 53% in Coropata in 1992,5 and 38.2% prevalence of human fascioliasis in the Bolivian Altiplano. Am J Trop Med Hyg 45: 603±609. 8 in Huacullani in 1993. Collectively, these reports indicate 6. Bjorland J, Byran RT, Strauss W, Hillyer GV, McAuley JB, that this disease has been endemic in the region for more 1995. An outbreak of acute fascioliasis among Aymara Indi- than a decade. ans in the Bolivian Altiplano. Clin Infect Dis 21: 1228±1233. 7. O'Neill SM, Parkinson M, Strauss W, Angles R, Dalton JP, 1998. Immunodiagnosis of Fasciola hepatica infection (fa- scioliasis) in a human population in the Bolivian Altiplano Financial support: This research was funded by Dublin City Uni- using puri®ed cathepsin L cysteine proteinase. Am J Trop Med versity, The Irish-American Partnership and Commission of the Eu- Hyg 58: 417±423. ropean Community Program in Life Sciences and Technologies for 8. Esteban JG, Angles R, Oviedo JA, Strauss W, Mas-Coma S, the Developing Countries (STD) (contract no. TS3-CT94±0294). 1997. The presence of a very high prevalence and intensity Andrew J. Dowd was a recipient of a post-doctoral fellowship fund- of infection with Fasciola hepatica among Aymara children ed by the Irish Higher Education Authority and a grant from the from the Bolivian Northern Altiplano. Acta Trop 66: 1±14. Higher Education Authority (Ireland). 9. Roche L, Dowd AJ, Tort J, McGonigle SM, McSweeney A, Curley P, Ryan T, Dalton JP, 1997. Functional expression of Authors' addresses: Sandra M. O'Neill, Michael Parkinson, Andrew Fasciola hepatica cathepsin L1 in Saccharomyces cerevisiae. J. Dowd, and John P. Dalton, School of Biological Sciences, Dublin Eur J Biochem 245: 373±380. City University, Dublin 9, Ireland. Wilma Strauss and Rene Angles, 10. Smith A, Dowd A, McGonigle SM, Keegan P, Brennan G, Unidad de Parasitologia, Insitituto Nacional de Laboratorios de Sal- Trudgett A, Dalton JP, 1993. Puri®cation of a cathepsin L-like ud, Secretaria Nacional de Salud, Pasaje Rafael Subieta No. 1889, proteinase secreted by adult Fasciola hepatica. Mol Biochem Mira¯ores, Casilla M-10019, La Paz, Bolivia. Parasitol 62: 1±8. 11. Richie LS, 1948. Another sedimentation technique for routine stool examination. Bull US Army Med Dept 8: 326. REFERENCES 12. Dalton JP, Clough KA, Jones MK, Brindley PJ, 1996. Charac- terization of the cathepsin-like cysteine proteinases of Schis- 1. Chen MG, Mott KE, 1990. Progress in morbidity due to Fas- tosoma mansoni. Infect Immun 64: 1328±1334. ciola hepatica infection. Trop Dis Bull 87: 1±37. 13. Knobloch J, 1985. Human fascioliosis in Cajamarca/Peru: hu- 2. Arjona R, Riancho J, Aguado J, Salesa R, Gonzalez-Macias J, moral antibody response and antigenaemia. Trop Med Par- 1995. Fascioliosis in developed countries: a review of classic asitol 36: 91±93.