Introduction

Fasciolosis is a zoonotic disease commonly transmitted by metacercariae-infected herbs or vegetables. The causative agents, Fasciola hepatica and Fasciola gigantica, infect both livestock and humans (Mas-Coma et al. 2009). F. hepatica, a well-known liver fluke, is a concern worldwide, and according to estimates provided by the World Health Organization (WHO 2008), 2.4 million people are infected with F. hepatica and nearly 17 million are at risk for infection (WHO 2008; Mas-Coma et al. 2009). The human infection statistics corelate closely with animal infections. Thus, economic consequences will exceed diminished animal products, e.g., milk, wool, and meat (Mezo et al. 2011; Mateus Fracasso et al. 2017). Diagnosis of sheep fasciolosis is usually performed by observation of parasite eggs in the feces approximately from 7 weeks post-infection and in some cases may require multiple stool examinations; however, this method has low reliability and sensitivity during the prepatent period of the infection. For this reason, more potent diagnostic methods should be developed (Martínez-Valladares et al. 2010; Martínez-Pérez et al. 2013).

Development of sensitive and specific antibody detection methods, and rapid diagnosis of infection, may prevent the negative impact of fasciolosis on the productivity of ovine herds and prevent economic loss (Guobadia and Fagbemi 1995; Sriveny et al. 2006). The desirable properties of ELISA and other immunoassays make them the methods of choice for detection of Fasciola-specific antibodies in sheep sera (Sriveny et al. 2006; Anuracpreeda et al. 2009).

Recently, immunologic studies focused on characterization and production of immunogenic parasite molecules, as well as application of appropriate ones in serodiagnosis and vaccinology (Velusamy et al. 2004; Turner et al. 2016; Mokhtarian et al. 2016a, b). A combination of immunodiagnostic methods with clinical findings could likely aid in early diagnosis of ovine fasciolosis.

More recently, purified native and recombinant F. hepatica antigens have become available; however, evaluations of the sensitivities and specificities of those proteins in serodiagnosis are necessary. Previous studies showed that cathepsin L1 (CL1), fatty acid-binding proteins (FABPs), and glutathione-S-transferase (GST) are the major immunogenic proteins of F. hepatica and could be used for diagnostic purposes (Anuracpreeda et al. 2016; Mokhtarian et al. 2016a, b). The aim of this study was to compare the main immunogenic recombinant and native proteins from the F. hepatica excretory/secretory antigens (E/S Ags) for serodiagnosis of sheep fasciolosis.

Materials and methods

Sheep sera

All experiments were approved by the Ethics Committee of the Faculty of Iran University of Medical Science. F. hepatica-positive serum samples were collected from 30 F. hepatica-infected sheep slaughtered at Kahrizak slaughterhouse, Tehran, Iran. Samples were diagnosed by observation of adult Fasciola flukes in the sheep livers and bile ducts in chronic phase of the infection. The infection-negative serum samples were collected from 40 healthy sheep whose stool exams were negative and showed no evidence of intestinal or liver parasite infections. The positive and negative sera aided in immunoassay development. Subsequently, antibody cross-reactivity was determined using 43 serum samples from sheep with other parasitic infections including hydatidosis (N = 10), cysticercosis (N = 7), coenurosis (N = 8), dicrocoeliasis (N = 8), and brucellosis (10), all of which were purchased from the Research Center of Veterinary, Tehran University, Tehran, Iran. Clinical diagnosis of the mentioned infections was performed by observation of parasites in special organs of the sheep and also molecular confirmatory methods. Since these common ovine parasites may show cross-reactivity with Fasciola and interfere with results of the immunoassays, we applied them in this study.

Preparation of F. hepatica E/S Ags

F. hepatica were collected from infected sheep livers from the Kahrizak abattoir in Tehran, Iran. Adult flukes were removed from the bile ducts and washed several times with 150 mM phosphate-buffered saline (PBS), pH 7.4, to remove blood and bile. Finally, F. hepatica E/S Ags were prepared as previously described (Mokhtarian et al. 2016a, b). The E/S products were centrifuged at 10,000×g for 60 min at 4 °C. The clear supernatant was then concentrated by ultrafiltration using 4 kDa membrane filters (Eppendorf, Germany). The protein concentration was measured by the Bradford method and the product stored at − 80 °C (Bradford 1976).

Purification of F. hepatica native fatty acid-binding protein and native glutathione-S-transferase

Native FABP of E/S Ags was separated by precipitation with 70% saturated ammonium sulfate solution. The supernatant containing the desired proteins was dialyzed against 20 mM Tris-HCl, pH 8, at 4 °C overnight and subjected to anion exchange chromatography (DEAE sepharose 6B, Amersham, Sweden) and size exclusion gel chromatography columns, respectively, as previously described (Timanova-Atanasova et al. 2004; Mokhtarian et al. 2016a, b).

Native GST was purified from F. hepatica E/S Ags on a glutathione-affinity agarose matrix (Sigma-Aldrich, USA) using a wash-batch method, as previously described by Brophy et al. (1995), and the purified protein was stored at − 80 °C.

Molecular cloning and expression of recombinant proteins

The genes encoding CL1, FABP, and GST were cloned and their proteins expressed previously (Mokhtarian et al. 2016a, b). In brief, the cDNA encoding FhCL-1 (GenBank accession no. U62288.2) was cloned into pET-21 plasmid (Invitrogen, USA), and the cDNAs encoding FhFABP and FhGST (GenBank accession nos. AJ250098.1 and HM584608.1, respectively) were cloned into pET-28 and expressed in Escherichia coli BL21 (DE3) cells (Invitrogen, USA). In this study, the recombinant polypeptides were purified using nickel affinity chromatography with a His Trap column (Ni-IDA Resin, Parstous, Iran).

The purified fractions were analyzed by SDS-PAGE and Coomassie Blue staining, and the protein concentrations were determined by the Bradford assay.

Enzyme-linked immunosorbent assay

The immunological interactions of the native and recombinant proteins with the infected and healthy ovine sera were determined by indirect ELISA as described by Espino et al., with minor modifications (Espino and Hillyer 2004). Following determination of the optimal concentrations of the antigens by a preliminary checkerboard titration test, polysorb microtitration plates (Nunc, Denmark) were coated with 100 μl of a 10 μg/ml solution of each native or recombinant protein in 0.1 M bicarbonate buffer, pH 9.6, at 4 °C overnight. After washing with PBS containing 0.05% Tween 20 (PBS-T), wells were blocked with 2% BSA for 2 h at 37 °C. After a second wash, 100 μl of 1:500 diluted sheep sera (from healthy sheep or sheep with fasciolosis or other parasitic infections) was added and the plates were incubated for 2 h at 37 °C. Following another wash, 100 μl of 1:5000 diluted horseradish peroxidase-labeled polyclonal rabbit anti-sheep IgG (Sigma-Aldrich, Germany) was added and the plates incubated for 1 h at 37 °C. After an extensive wash, 100 μl of tetramethylbenzidine/H2O2 mixture was added as chromogen/substrate solution. Following a 15-min incubation in the dark, the enzymatic reaction was stopped by adding 100 μl of 2 N H2SO4 and the optical density was measured at 450/630 nm.

Data analysis

The results of the immunoreactivity assays were compared between groups using the Tukey’s or Dunn’s multiple comparison tests. The optimal cutoff value for each ELISA method, with a 95% confidence interval (CI), was established by receiver-operating characteristic (ROC) curve analysis using Tukey’s multiple comparison tests. All statistics were analyzed and graphs plotted using GraphPad Prism version 6.0 (GraphPad Software, La Jolla, CA, USA). P values < 0.05 were considered statistically significant. According to arbitrary guidelines, the area under the curve (AUC) values were considered as follows: AUC < 0.5, non-informative; 0.5 < AUC < 0.7, low accuracy; 0.7 < AUC < 0.9, moderate accuracy; and 0.9 < AUC < 1, high accuracy (Swets 1988).

In the ROC analysis, cutoff values for the following conditions were derived according to cutoff for the maximum specificity (SP) where the sensitivity (SE) was still 100%. Cutoff for the maximum value for Youden’s J was measured by the formula (SE + SP − 1).

Results

Distribution of absorbance values (at 450 nm)

As expected, the distribution of absorbance values varied significantly between positive and negative samples; however, it also varied between different assays and different optimal cutoff values for all five ELISA tests (Fig. 1; see the “ROC analyses” section).

Fig. 1
figure 1

Box plots showing the absorbance values (450 nm) of ELISA results for discrimination of the immunoreactivity of native and recombinant Fasciola hepatica antigens in Fasciola-infected sheep versus healthy and other parasite-infected sheep (a–e). Absorbance values of ELISA results for native and recombinant F. hepatica antigens (f). nFABP native fatty acid-binding protein, nGST native glutathione-S-transferase, rCL1 recombinant cathepsin L1, rFABP recombinant fatty acid-binding protein, rGST recombinant glutathione-S-transferase

Full size image

ROC analyses

The immunoreactivities of the recombinant and native forms of F. hepatica antigens were compared with each other. The optimal cutoff value for each ELISA result was derived from four different conditions according to the following: (a) maximum specificity at which the sensitivity was still 100%, (b) maximum sensitivity at which the specificity was still 100%, (c) maximum value for Youden’s index J (SE + SP-1), and (d) maximum accuracy (maximum true positive and negative results) (Table 1). Receiver-operating characteristic analyses determined that the AUC values of native fatty acid-binding protein (nFABP), recombinant fatty acid-binding protein (rFABP), and recombinant cathepsin L1 (rCL1) were 0.96, 0.97, and 0.99, respectively. The AUC values for recombinant glutathione-S-transferase (rGST) and native glutathione-S-transferase (nGST) were 0.96 and 0.90, respectively (Fig. 2; Table 1).

Table 1 Specificity and sensitivity of immunoreactivity of native and recombinant Fasciola hepatica proteins based on cutoff values and Youden’s J results obtained from Fasciola-infected sheep sera. The table includes also the ROC AUC for the five different Fasciola-specific ELISAs
Full size table
Fig. 2
figure 2

Receiver-operator characteristic (ROC) curve for five Fasciola hepatica antigens. The values for the area under the ROC curve (accuracy) were 0.96 for the nFABP, 0.97 for the nFABP, 0.99 for the rCL1, 0.90 for the nGST, and 0.96 for the rGST in ELISA

Full size image

The ROC-optimized cutoff values for nFABP and rFABP were 0.7 and 0.72 in ELISA, respectively. Sensitivities for nFABP and rFABP were both 97%, and specificities were 92 and 96% in ELISA, respectively.

The ROC-optimized cutoff values for nGST and rGST were 0.72 and 0.71, respectively, and their sensitivities were 75 and 86%, which were the lowest sensitivities of the proteins studied. Their specificity values were 91 and 97%, respectively.

Based on the established cutoff value of 0.6 for rCL1, no seropositive samples were found in the true-negative population. Specificity and sensitivity for rCL1 were 100 and 97%, respectively.

According to the arbitrary guidelines, the AUC values of 0.9–1.0 indicated high accuracy for discrimination between positive and negative results.

Cross-reactivity

Cross-reactivity, as the discrimination of positive and negative results, was measured based on the designated cutoff value and presented according to the parasitic disease investigated (Table 2). Sera from hydatidosis cases displayed considerable cross-reactivity in all five ELISA reactions, whereas all other cross-reactions were individually scattered among individual ELISAs. The highest specificity score of 97% was obtained by the rCL1-based ELISA, with two instances of cross-reactivity as described above (Table 2). nFABP cross-reacted with six sera out of 43 samples with parasitic diseases with 86.04% specificity.

Table 2 Potential cross-reactivity of Fasciola-infected sheep sera with other parasites
Full size table

Discussion

In this study, we focused on improving the diagnosis of fasciolosis using native and recombinant proteins and a panel of sheep positive and negative control serum samples. A critical point for the evaluation of a new immunodiagnostic test is the establishment of cutoff points. We determined these points with control sera. Using ROC curve and Youden’s J analyses, we selected the cutoff values and found the best balance of sensitivity and specificity for the ELISA (Zweig and Campbell 1993).

In ELISA, the highest sensitivities were obtained with rCL1 and rFABP, and the lowest with nGST. High antibody titers during active sheep infections indicate that these molecules are repeatedly and effectively exposed to the host immune system.

Our findings support previous studies indicating high sensitivity of rCL1 as a prominent secretory enzyme of adult liver flukes for serodiagnosis of fasciolosis (Cordova et al. 1999; Tantrawatpan et al. 2007; Valero et al. 2012; Gonzales Santana et al. 2013; Kuerpick et al. 2013; Anuracpreeda et al. 2016). In the present study, similar to the study of Mezo et al., distinct cutoff values were used to discriminate between the sera of fasciolosis cattle and healthy ones with 100% sensitivity and 97% specificity (Mezo et al. 2011). Cornelissen et al. and Kuerpick et al. used rCL1 for determination of cattle fasciolosis and achieved 100% sensitivity and 88.6–94.6% specificity (Cornelissen et al. 1992; Kuerpick et al. 2013).

Anuracpreeda et al. used an anti-CL1 monoclonal antibody-based immunochromatography method to identify the acute and chronic phases of F. gigantica-caused fasciolosis in cattle and achieved 96.7% sensitivity with 100% specificity (Anuracpreeda et al. 2016). Our results were similar.

Interestingly, more than 95% sensitivity was achieved with both rFABP and nFABP, with cutoff values of 0.7, indicating their applicability in differentiating healthy from infected ovine sera.

Notably, a 14.7-kDa FABP was previously recognized in F. hepatica and its recombinant analogue has been studied due to its ability to stimulate an anti-parasite host immune response (Hillyer et al. 1987; Muro et al. 1997). Several studies suggest that FABP can induce protection against F. hepatica in small and large laboratory animal models and could be a candidate for immunodiagnosis of fasciolosis (Sobhorn et al. 1998; Dalton et al. 2003; Wedrychowicz and Wisniewski 2003). Fasciola-infected ovine sera have specific antibodies against this immunogenic protein. Overall, our findings agree with studies in cattle and sheep (Casanueva et al. 2001; Timanova-Atanasova et al. 2004; Rabia et al. 2007; Allam et al. 2012; Teofanova et al. 2012; Anuracpreeda et al. 2016) and other animal species, but disagree with results achieved by (Aly et al. 2014), which could be due to specific characteristics of the studied antigens, the genotypic diversity of various F. hepatica isolated, status and phase of the infection, and the applied diagnostic techniques.

Our results confirmed that both CL1 and FABP have similar AUCs, thus to choose the most suitable protein for differential diagnosis of fasciolosis we used Youden’s J index. Similar to our previous finding in human fasciolosis (Mokhtarian et al. 2016a, b), in this study, CL1, with Youden’s J = 0.97, was determined to be the preferred immunogenic protein for serodiagnosis of sheep fasciolosis.

Another immunogenic protein we studied was GST. Both rGST and nGST showed low specificity and sensitivity with cutoff values of 0.7. Youden’s J was calculated as 0.83 and 0.66, respectively, for rGST and nGST; thus, both forms showed low immunological affinity to positive control sera. Glutathione-S-transferases belong to a well-known family of cytosolic enzymes involved in detoxification. Glutathione-S-transferases in adult flukes are related with gut lamellae and may play a role in absorptive function of adult gut. They are therefore tempting targets for both chemotherapy and vaccine development (Meyer and Thomas 1995; Rossjohn et al. 1997).

Fasciolosis is endemic in Iran; therefore, it is possible that a dose of infectious metacercariae could affect the development of a detectable antibody response to particular antigens (Cornelissen et al. 1999). Due to availability of naturally F. hepatica infected sheep, we used their sera to determine immunogenic proteins of this parasite. It means that the results are generalizable to the rest of the sheep population in the region.

In the present study, we found significant cross-reactivity between fasciolosis and echinococcosis, especially with GST and FABP. This finding is consistent with those of several other studies (Intapan et al. 2003; Rabia et al. 2007; Allam et al. 2012; Aly et al. 2014). This cross-reactivity may be due to common antigenic lipoprotein components or epitopes of Echinococcus multilocularis, Echinococcus granulosus, and Fasciola species (Yamano et al. 2009). Previous reports demonstrated cross-reactivity of F. hepatica Ags with schistosomiasis, paragonimiasis, and clonorchiasis. Cornelissen et al. showed cross-reactivity between infected ovine sera with Fasciola and Eimeria species, as well as E. granulosus; however, they also reported 100% sensitivity and 96% specificity. Interestingly, infections with Schistosoma spp., Paragonimus spp., and Clonorchis spp. have not been reported in some countries, including Iran, which is commonly due to the absence of the specific carrier in those areas. Thus, it remains necessary to design a larger confirmatory study with more positive serum samples during both acute and chronic phases of fasciolosis and determine the cross-reactivity with endemic helminths, particularly Taenia, Echinococcus, and Strongyloides stercoralis.

In conclusion, in this study, we developed sensitive and specific ELISA, employing rCL1 and rFABP, to detect F. hepatica-specific antibodies in ovine sera. Application of recombinant proteins will allow scaling up of the procedure for mass screening. Also, sufficient antigen quantities are needed for applications in acute, chronic, and post-treatment studies to gain more consistent results and adjust the established ELISA methods for commercialization and further justification in areas where the disease is endemic.