INTRODUCTION
Trichinellosis is an intestinal and tissular parasitosis caused by the nematode Trichinella spp. that has a worldwide distribution.
In humans, the early diagnosis of the disease is currently achieved on the basis of signs and typical clinical symptoms (gastrointestinal disorders, myalgia, periorbital oedema and eosinophil counts >500/mm3) (Pawlowski, Reference Pawlowski and Campbell1983; Murrell and Bruschi, Reference Murrell, Bruschi and Tsieh Sun1994). Later on, diagnosis can be confirmed by serological tests such as the enzyme-linked immunosorbent assay (ELISA) or indirect immunofluorescence (IIF) which employ, among others, the excretory-secretory products of the muscle larva (ML-ESP) and cryostat sections of free ML respectively (Murrell and Bruschi, Reference Murrell, Bruschi and Tsieh Sun1994; Costantino et al. Reference Costantino, Malmassari, Dalla Fontana, Diamante and Venturiello2001; Dupouy-Camet et al. Reference Dupouy-Camet, Kociecka, Bruschi, Bolas-Fernandez and Pozio2002).
An early diagnosis of the disease, that is, during the intestinal phase, is essential to limit the infection because anti-helminthic drugs are more effective during this phase of the infection (Soulé and Dupouy-Camet, Reference Soulé, Dupouy-Camet, Soulé and Dupouy-Camet1991; Dupouy-Camet et al. Reference Dupouy-Camet, Kociecka, Bruschi, Bolas-Fernandez and Pozio2002).
Despite trichinellosis being an intestinal parasitosis, parasitological tests performed on stool samples is not routinely employed because both the migratory stage (newborn larvae, NBL) and the adult worms (AW) are hardly found in faeces.
Based on the detection of the immune response of the host's intestinal mucosa during a parasite infection or the detection of parasite antigens in stool samples, many tests have been developed as useful tools to diagnose human enteric parasite diseases. These immunodiagnostic tests have proven to be more sensitive than parasitological tests (Espino et al. Reference Espino, Diaz, Perez and Finlay1998; Jamornthanyawat, Reference Jamornthanyawat2002; Allan et al. Reference Allan, Wilkins, Tsang and Craig2003; Abdel Baki et al. Reference Abdel Baki, Younis, Habib, Ramadan, Metwally, Ismail and Saber2004).
In a previous study we developed ELISAs for the detection of coproantigens (CAgs), coproantibodies (CAbs) and faecal immune complexes (FIC) in rats experimentally infected with Trichinella spiralis and we were able to detect the infection as early as days 2–4 p.i., before serum antibodies could be detected (10 days p.i.) (Nuñez et al. Reference Nuñez, Costantino and Venturiello2003).
At present there are no published reports on the detection of the immunological parameters of the intestinal phase in human trichinellosis. In the present work we report the detection by ELISA of specific CAbs against T. spiralis and specific FIC, but not CAgs, in stool samples of individuals undergoing the first 3 months of infection (acute phase). Moreover the isotypes of immunoglobulins present were evaluated.
MATERIALS AND METHODS
Faecal samples
Faecal samples from 18 individuals belonging to different outbreaks of trichinellosis that arose in Argentina were obtained and kept at −20°C until used. All patients included in this work had positive serology for T. spiralis as determined by ELISA, IIF and immunoelectrotransfer blot assay (IETB), and had the typical signs and symptoms of the disease. Seventeen out of the 18 patients were treated with anti-parasite drugs (benzimidazole derivates) and anti-inflammatory drugs prior to or during the collection of samples. Individual No. 6 (Table 1) did not receive any treatment. Times ranged from 20 days to 3 months p.i. Seven healthy individuals from endemic areas were included as controls. Conventional parasitological tests were performed in all samples rendering negative results.
Detection of anti-ML-ESP antibodies (CAbs) and immune complexes (FIC) by ELISA in stool samples from individuals undergoing the acute phase of trichinellosis
(Igs, immunoglobulins. Values in parenthesis represent OD values rendered by each sample. Samples rendering OD values greater than 2-fold that of the mean OD of normal faecal extracts were considered positive (+). ND, Not done.)
Preparation of faecal extracts
Faecal extracts (FE) were prepared following a modified protocol employed by Espino and Finlay (Reference Espino and Finlay1994). Briefly, samples were weighed and thoroughly homogenized with 1·5 vols of cold phosphate-buffered saline (PBS) containing 0·05% Tween 20 and a mixture of protease inhibitors: 2 mm phenylmethylsulphonyl fluoride (Sigma, St Louis, MO, USA), 2 mm pepstatin A (Sigma) and 2 mm EDTA. Homogenates were then centrifuged for 15 min at 900 g at 4°C. Supernatants were collected, aliquoted and kept at −20°C until used. FE from healthy individuals were also prepared.
Detection of IgG, IgA, IgM, IgE and total immunoglobulins against muscle larva- and adult worm-excretory-secretory proteins
Total CAbs and isotypes against T. spiralis were detected by indirect ELISA, employing as antigen ML-ESP, obtained as previously described (Gamble et al. Reference Gamble, Anderson, Graham and Murrell1983). Briefly, flat-bottomed-PolySorp polyvinyl microtitre plates (Nunc, Roskilde, Denmark) were coated with ML-ESP at 5 μg/ml in carbonate buffer, pH 9·6, during 1 h at 37°C and overnight at 4°C. After coating, and between each step, plates were washed 3 times with PBS containing 0·05% Tween 20 (PBST). Plates were blocked using 3% non-fat dried milk in PBS for 1 h at 37°C. FE were then added undiluted and incubated for 1 h at 37°C. For detection of total specific immunoglobulins, a biotinylated goat anti-human IgG (H+L) serum (Vector, Burlingame, CA, USA) followed by a macromolecular complex of avidin and biotinylated peroxidase (Vector) were added according to the manufacturer's directions. For detection of IgE, goat anti-human IgE (Vector) was employed, followed by peroxidase-conjugated rabbit anti-goat serum (Bethyl, TX, USA) adequately diluted. For detection of IgA, rabbit anti-human IgA (Dako Corporation, CA, USA) was used and followed by peroxidase-conjugated goat anti-rabbit serum (Sigma) suitably diluted. For detection of IgM, rabbit anti-human IgM (Dako) was employed, followed by peroxidase-conjugated goat anti-rabbit serum (Sigma). For detection of IgG, goat anti-human IgG (Sigma) was used followed by peroxidase-conjugated rabbit anti-goat serum (Bethyl) adequately diluted. The colour reaction was developed using 0·004% (w/v) o-phenylendiamine (Sigma) and 0·004% (v/v) H2O2 in citrate buffer pH 5. Reaction was stopped by addition of 4 N H2SO4 and the OD read at 490 nm in an ELISA reader (Metertech Σ960). Those samples rendering negative results for ML-ESP were tested for total immunoglobulins against AW-ESP obtained as described by Nuñez et al. (Reference Nuñez, Costantino and Venturiello2003) employing the avidin-biotin system as described above. Samples were run in duplicate. Samples rendering OD values greater than 2-fold that of the mean OD of normal FE were considered positive.
The specificity of the antibodies detected was tested by IETB according to the method described by Nuñez et al. (Reference Nuñez, Malmassari, Costantino and Venturiello2000).
Detection of coproantigens by ELISA
CAgs were detected by capture ELISA. The mAb PG6B1, with reactivity against the 45–55 kDa components of the ML-ESP and kindly provided by Dr Bruschi, was used at 5 μg/ml, or an ammonium sulphate-precipitated rat anti-T. spiralis gammaglobulin fraction was used at 160 μg/ml in carbonate buffer pH 9·6, to coat microtitre plates. Plates were blocked as above and all incubations were carried out at 37°C with 6 washes with PBST in between. FE were added undiluted. After incubation with the samples, a human anti-T. spiralis serum, suitably diluted in PBST, was added and incubated for 1 h at 37°C. A biotinylated anti-human IgG (H+L) (Vector) plus 10% normal mouse serum was added followed by a macromolecular complex of avidin and biotinylated peroxidase (Vector) according to the manufacturer's directions. The colour reaction was developed and stopped as above. A standard curve of ML-ESP diluted in normal FE ranging from 40 ng/ml to 1250 ng/ml was run in each plate. All samples were run in duplicate.
Faecal immune complexes
Precipitation
One vol. of FE was mixed with 1 vol. of 6% polyethylene glycol 8000 (PEG, ICN, CA, USA) in borate buffer pH 8·5. Mixtures were then incubated overnight at 4°C and then centrifuged for 1 h at 900 g at 4°C. Precipitates were then washed 3 times with 3% PEG and the final pellet resuspended in PBS to a final volume equivalent to 1/10 of the initial volume. No free antigen was precipitated under these experimental conditions.
Detection
FIC were detected by capture ELISA. An ammonium sulphate-precipitated rat anti-T. spiralis gammaglobulin fraction was used to coat microtitre plates at a concentration of 160 μg/ml in carbonate buffer, pH 9·6 (1 h at 37°C and overnight at 4°C). Plates were then blocked using 10% non-fat dry milk in PBST for 1 h at 37°C. All steps were followed by 3 washes with PBST. Precipitated FIC were added and incubated for 1 h at 37°C and overnight at 4°C. Secondary and conjugated antibodies and conditions for the colour reaction were the same as those used for detection of free IgG, IgA, IgM and IgE. As positive control, a FIC preparation was made by mixing ML-ESP with an anti-T. spiralis human gammaglobulin fraction diluted in normal FE at a previously determined ratio. All samples were run in duplicate. Samples rendering OD values greater than 2-fold that of the mean OD of normal FE were considered positive.
RESULTS
In 89% of the samples analysed either free (CAbs) or complexed (FIC) anti-ML-ESP antibodies were detected in the following isotypes of immunoglobulins: IgA (56%), IgG (56%), IgE (28%) and IgM (22%).
Detection of coproantibodies
Anti-ML-ESP CAbs were detected in 12 out of the 18 faecal samples analysed (67%) regardless of the p.i. time. All samples from individuals undergoing the first 30 days of infection had total immunoglobulins and/or isotypes against ML-ESP. CAbs detected as total immunoglobulins appeared in 11 out of the 18 individuals studied (61%). All specific isotypes were detected by the methodologies developed, finding CAbs IgA in 50% of the individuals, IgG in a 33%, IgE in a 28% and IgM in a 17% (Table 1).
On analysing the frequency of appearance of specific isotypes in those samples presenting anti-ML-ESP antibodies it was observed that the IgA was present in 75% of the individuals, the IgG in 50%, the IgE in 42% and the IgM in 25% (Table 1).
Those IETB-positive samples showed reactivity for the components of ∼45, ∼55, ∼66 and/or 97 kDa of the ML-ESP.
All the samples that were positive or negative for total immunoglobulins against ML-ESP were positive for the AW-ESP.
Detection of coproantigens
Free CAgs were not detected in any of the individuals by employing either the mAb PG6B1 or the ammonium sulphate-precipitated rat anti-T. spiralis gammaglobulin fraction as capture antibodies.
Detection of faecal immune complexes
FIC detected with an anti-total immunoglobulins conjugate appeared in 11 out of the 14 studied samples (79%). FIC bearing all the specific isotypes of immunoglobulins were detected, finding FIC IgG in 38% of the individuals, FIC IgA in 13% and FIC IgE and FIC IgM in a 6% (Table 1).
It was observed that IgG was involved in the formation of FIC in 55% of FIC-positive individuals, IgA was involved in 18% and both IgM and IgE were involved in the formation of FIC in 9% (Table 1).
DISCUSSION
The analysis of stool samples by the ELISAs developed herein allowed determination, as in the rat model (Nuñez et al. Reference Nuñez, Costantino and Venturiello2003), of specific antibodies in either the free or the complexed form in an elevated percentage of individuals (89%) undergoing the acute phase of trichinellosis. The specificity of these CAbs was demonstrated by the IETB assay to find that they reacted with the components of ∼45 and ∼55 kDa of the ML-ESP. These glycoproteins are specific of Trichinella spp. (Appleton et al. Reference Appleton, Bell, Homan and van Knapen1991; Nuñez et al. Reference Nuñez, Malmassari, Costantino and Venturiello2000). As a rule, only those samples with OD greater than 1·000 in the ELISA gave positive results in the IETB, probably due to the presence of specific FIC.
The failure to detect free CAgs is probably due to a low amount of antigens present in the intestine that cannot be detected, even when the system can detect concentrations of about 40 ng/ml, or due to immune complex formation.
The detection of anti-ML-ESP CAbs and/or FIC was possible in 100% of the analysed samples during the first 30 days p.i. when all or some isotypes were present. The latter result is of great importance since this is the period during which both the epidemiological and the clinical scenarios call for a confirmatory diagnostic test.
As it can be observed in Table 1, no specific pattern of antibody isotypes could be found at any time p.i. This finding could be attributed to the individual immune response generated, mainly due to the host's genetic background and/or the different parasite load ingested (Pawlowski, Reference Pawlowski and Campbell1983). This possible explanation would also apply to the lack of anti-ML-ESP CAbs and FIC obtained with some of the samples (Table 1, individuals Nos 7 and 8). Moreover, those samples that were negative for anti-ML-ESP total immunoglobulins were still positive in the corresponding ELISA employing AW-ESP. This finding is in line with those obtained in the rat model where, by means of the kinetics of appearance of CAbs for both antigens, we demonstrated that the response to AW-ESP is more sustained over time. The latter finding is in agreement with literature data indicating the existence of stage-specific antigenic epitopes (Appleton et al. Reference Appleton, Bell, Homan and van Knapen1991). Taking these facts into account, the performance of the AW antigens in the immunodiagnosis should be reconsidered.
The results obtained in this work demonstrate for the first time that in human trichinellosis the immune response of the intestinal mucosa against T. spiralis is also featured by the synthesis and secretion of specific IgG which was as frequent as IgA, a typical immunoglobulin of the mucosal immune system.
The finding of anti-Trichinella IgE is in line with previous studies that demonstrated in experimentally infected rodents the presence of this immunoglobulin in intestinal fluid and stool (Negrão-Corrêa et al. Reference Negrão-Corrêa, Adams and Bell1996, Reference Negrão-Corrêa, Adams and Bell1999; Nuñez et al. Reference Nuñez, Costantino and Venturiello2003). The presence of this immunoglobulin was expected since this isotype has been proven to play a role in the process of immune expulsion of the nematode from the gut (Ahmad et al. Reference Ahmad, Wang and Bell1991; Bell, Reference Bell1998).
Finally, the results obtained in this work, together with those of the rat model (Nuñez et al. Reference Nuñez, Costantino and Venturiello2003) allow us to propose that the detection of CAbs, in either the free or complexed form, could be applied to confirm early human T. spiralis infection, the time at which anti-parasite drugs are more efficient.
The present work was supported by grants B073 from the University of Buenos Aires and PIP No. 02858 from the National Research Council of Argentina (CONICET).
