Evaluation of three enzyme-linked immunosorbent assays (ELISAs) for the detection of serum antibodies in sheep infected with Echinococcus granulosus

Abstract

The aim of this study was to develop an immunological method for the identification of sheep infected with Echinococcus granulosus which would allow the monitoring of animals imported into countries free from hydatidosis and as an aid to countries where control schemes for the disease are in operation.

Three enzyme-linked immunosorbent assays (ELISAs) were developed and validated, using as antigen either a purified 8 kDa hydatid cyst fluid protein (8kDaELISA), a recombinant EG95 oncosphere protein (OncELISA) or a crude protoscolex preparation (ProtELISA). Sera used for the assay validations were obtained from 249 sheep infected either naturally or experimentally with E. granulosus and from 1012 non-infected sheep. The highest diagnostic sensitivity was obtained using the ProtELISA at 62.7 and 51.4%, depending on the cut-off. Assay sensitivities were lower for the 8kDaELISA and the OncELISA. Diagnostic specificities were high, ranging from 95.8 to 99.5%, depending on the ELISA type and cut-off level chosen. A few sera from 39 sheep infected with T. hydatigena and from 19 sheep infected with T. ovis were recorded as positive. Western immunoblot analysis revealed that the dominant antigenic components in the crude protoscolex antigen preparation were macromolecules of about 70–150 kDa, most likely representing polysaccharides.

This study demonstrated that the ProtELISA was the most effective immunological method of those assessed for detection of infection with E. granulosus in sheep. Because of its limited diagnostic sensitivity of about 50–60%, it should be useful for the detection of the presence of infected sheep on a flock basis and cannot be used for reliable identification of individual animals infected with E. granulosus.

Introduction

Cystic echinococcosis (CE) in humans and ruminants is caused by infection with the larval stage of the dog tapeworm, Echinococcus granulosus. The parasite has a wide geographical distribution, affecting many countries on all continents (Eckert et al., 2000). While CE persists in many parts of the world and is re-emerging in some regions (Brehm et al., 1999, Eckert et al., 2000, Jenkins, 1998), New Zealand has operated a national eradication campaign since 1959 and has successfully eradicated the disease (Anon., 1999, Heath, 1999, Kasper, 1990, Montgomery and van der Logt, 1994). However, live animals imported into the country will have the potential to re-establish the disease and need to be monitored (Mason and Orr, 1993). Such cases have recently occurred in imported cattle and were detected post-mortem at meat inspection (Anon., 2000, Bosson, 2000). The availability of immunological tests with the ability to detect the majority of ruminants infected with E. granulosus would be desirable for animal import monitoring and also in countries where control schemes for the disease are operating.

In order to develop useful screening tests for E. granulosus infections, one has to consider the host–parasite interactions that take place during the infections. Initially, immunological responses to E. granulosus in the intermediate host (humans, ruminants) will be directed against the invading oncospheres. Later, such responses may develop against components of the immature cysts and finally against components of the fertile metacestodes and the protoscoleces (Lightowlers and Gottstein, 1995). Immune responses can be humoral, leading to the formation of parasite-specific serum antibodies, or they may be cell-mediated, causing the proliferation of cytokine-producing T-cells. For comprehensive immunological assay development, antigenic macromolecules from every stage of the infectious agent may have to be applied. The readily available hydatid cyst fluid (HCF) has been used most frequently as a source for parasite antigens. Its components have been comprehensively investigated for their applicability in serological tests (Barbieri et al., 1998, Ibrahem et al., 1996, Ito et al., 1999, Leggatt et al., 1992, Leggatt and McManus, 1994, Lightowlers et al., 1984, Lightowlers et al., 1989, Lightowlers and Gottstein, 1995, Maddison et al., 1989, Moro et al., 1997, Young et al., 1984).

The so-called antigen 5 or antigen B was identified as a major antigenic component of the hydatids metacestode (Lightowlers and Gottstein, 1995). It comprises a regularly spaced group of molecules, of which the smallest subunit has a molecular weight of approximately 8 kDa, sometimes referred to as 12 kDa, and the other components appearing to be multimers of 16, 24 and 32 kDa (Gonzalez et al., 1996, Lightowlers et al., 1989). For humans infected with E. granulosus, it has been shown that the diagnostic sensitivity of assays using the 8 kDa subunit can be high, reaching up to 90%. Reports on the specificity of this antigen are contradictory and it seems that infections with other taenidae will give positive results as well (Barbieri et al., 1998, Ito et al., 1999, Leggatt et al., 1992, Leggatt and McManus, 1994, Lightowlers et al., 1989, Maddison et al., 1989). While published data on the use of the 8 kDa antigen in serological tests for ruminants are limited (Ibrahem et al., 1996, Moro et al., 1997), they indicate that the 8 kDa protein is potentially a useful antigen for the sero-diagnosis of E. granulosus infections in intermediate host animals.

Detailed investigations in the canine definite host have shown that proteins of the E. granulosus protoscoleces are antigenic. They have been successfully applied in the sero-diagnosis of dogs with high sensitivities and specificities (Ahmad and Nizami, 1998, Craig et al., 1995, Gasser et al., 1988, Gasser et al., 1993, Lightowlers and Gottstein, 1995). Studies are rare for the application of such antigens in the sero-diagnosis of sheep, despite reported diagnostic sensitivities of up to 80% (Ris et al., 1987).

Antigens of the E. granulosus oncospheres can be expected to cause immune responses in the early stages of infection in intermediate hosts. Purified oncosphere proteins caused strong antibody responses in experimentally immunised sheep (Heath and Lawrence, 1996) and a recombinant oncosphere protein EG95 conferred a high degree of protection to sheep, when used as a vaccine (Lightowlers et al., 1996, Lightowlers et al., 1999, Lightowlers et al., 2000, Woollard et al., 1998). Therefore, oncosphere protein antigens may be potentially useful in sero-diagnostic assays for the ruminant intermediate host of this parasite.

The cell-mediated immune response to E. granulosus infections has been investigated to a limited extend in humans (Hernandez-Pomi et al., 1997, Lightowlers and Gottstein, 1995) and mice (Dematteis et al., 1999, Rogan, 1998) but little in sheep. In humans, HCF antigen stimulated the proliferation of peripheral blood mononuclear cells (PMBC) in hydatid patients and increased levels of cytokines, such as interleukin-5 (IL-5) and gamma interferon (IFN-γ) were observed (Hernández-Pomi et al., 1997). Therefore, it could be worthwhile to investigate the antigen-dependent release of cytokines in blood from E. granulosus-infected ruminants as a potential immunological assay for the identification of infected animals.

In this study, we have investigated the usefulness of three different E. granulosus antigen preparations with samples from E. granulosus-infected or uninfected sheep for the detection of serum antibodies against E. granulosus and for the stimulation of antigen-dependent release of the cytokines IL-5 and IFN-γ from blood samples obtained from E. granulosus-infected and control sheep.