Early serodiagnosis of trichinellosis by ELISA using excretory–secretory antigens of Trichinella spiralis adult worms

The excretory–secretory (ES) antigens of Trichinella spiralis muscle larvae (ML) are the most commonly used diagnostic antigens for trichinellosis. Their main disadvantage for the detection of anti-Trichinella IgG is false-negative results during the early stage of infection. Additionally, there is an obvious window between clinical symptoms and positive serology. ELISA with adult worm (AW) ES antigens was used to detect anti-Trichinella IgG in the sera of experimentally infected mice and patients with trichinellosis. The sensitivity and specificity were compared with ELISAs with AW crude antigens and ML ES antigens. In mice infected with 100 ML, anti-Trichinella IgG were first detected by ELISA with the AW ES antigens, crude antigens and ML ES antigens 8, 12 and 12 days post-infection (dpi), respectively. In mice infected with 500 ML, specific antibodies were first detected by ELISA with the three antigen preparations at 10, 8 and 10 dpi, respectively. The sensitivity of the ELISA with the three antigen preparations for the detection of sera from patients with trichinellosis at 35 dpi was 100 %. However, when the patients’ sera were collected at 19 dpi, the sensitivities of the ELISAs with the three antigen preparations were 100 % (20/20), 100 % (20/20) and 75 % (15/20), respectively (P < 0.05). The specificities of the ELISAs with the three antigen preparations were 98.11, 95.60 and 89.31 %, respectively (P < 0.05). The sensitivity and specificity of the T. spiralis AW ES antigens were superior to those of the AW crude antigens and ML ES antigens. Thus, the AW ES antigens might serve as potential antigens for the early and specific serodiagnosis of trichinellosis.

. At present, a definitive diagnosis of human trichinellosis can be made by detecting larvae in a muscle biopsy sample or specific anti-Trichinella IgG antibodies [8], but the parasitological examination of biopsy samples is not sufficiently sensitive to detect Trichinella larvae in light infections and during the early stage of infection. Because T. spiralis muscle larvae (ML) are easily collected from experimentally infected animals, their excretory-secretory (ES) antigens can be prepared by the in vitro cultivation of isolated muscle larvae, resulting in a preparation with high specificity [9]. Thus, an ELISA with the ML ES antigens is the most commonly used serological method for the diagnosis of trichinellosis and is recommended by the International Commission on Trichinellosis (ICT) [10,11]. The main disadvantage of the detection of anti-Trichinella IgG against ML ES antigens is the occurrence of a high rate of false negative results during the early stage of infection [12]. These false negatives may occur because the majority of T. spiralis ML ES antigens are stage-specific; thus, the ML ES antigens are not recognized by antibodies induced by the parasites during the intestinal phase [13,14]. Several studies demonstrated that the maximum detection rate of 100 % of anti-Trichinella IgG was not reached until at least 1-3 months after human infection with the parasite [10]. There is an obvious time lag (window period) between the time when patients develop clinical symptoms and positive serology during the acute stage of trichinellosis [6,15]. Additionally, the ML ES antigens also have cross-reactivity with the sera of patients with other helminthiases (e.g., paragonimiasis, schistosomiasis, clornorchiasis, cysticercosis, and anisakiasis) [15][16][17]. Hence, there is an urgent need to develop new specific early diagnostic antigens for acute trichinellosis.
Trichinella spiralis ML develop to AW in the small intestine 31 h after infection, live in the intestinal mucosa and persist for 10-20 days in mice and rats or 4-6 weeks in humans [18]. During the intestinal stage of trichinellosis, the ES antigens produced by the AW result in early exposure to the immune system and elicit the production of specific anti-Trichinella antibodies by the host. Previous studies showed that AW crude antigens were recognized by infected sera from mice or swine 7 days post-infection (dpi) [19]. However, to the best of our knowledge there has been no report on the serodiagnosis of trichinellosis using T. spiralis AW ES antigens. The aim of this study was to evaluate the potential of T. spiralis AW ES antigens for the early serodiagnosis of acute trichinellosis.

Ethics statement
All procedures of animal experiments and the use of the patients' serum samples in this study were approved by the Life Science Ethics Committee of Zhengzhou University (no. 2011-016). Before the investigation, oral informed consent was obtained from all individuals.

Parasites and experimental animals
The isolate (ISS534) of T. spiralis used in this study was obtained from domestic pigs in Nanyang, Henan Province, China. The reference Trichinella isolates used in this study were T. nativa (T2, ISS10), T. britovi (T3, ISS100), T. pseudospiralis (T4, ISS13) and T. nelsoni (T7, ISS29); these isolates were obtained from the International Trichinella Reference Centre (ITRC; Rome, Italy). The Trichinella isolates were maintained by serial passage in BALB/c mice every 6-8 months in our laboratory. Specific pathogen-free (SPF) female BALB/c mice aged 6 weeks were purchased from the Experimental Animal Centre of Henan Province.

Serum samples
T. spiralis was used to infect the BALB/c mice. Twenty mice were randomly divided into two groups (ten mice per group): mice infected with 500 ML and mice infected with 100 ML. Approximately 100 μl of tail blood was collected from the infected mice on alternate days from 2-28 dpi to isolate sera [12]. Serum samples from normal mice were used as the negative control.
Sera were obtained from BALB/c mice infected with 300 ML of T. spiralis, T. nativa, T. britovi, T. pseudospiralis and T. nelsoni 42 dpi. Serum samples of mice infected with three spargana of Spirometra mansoni were collected 30 dpi in our department. Serum samples of mice infected with Toxoplasma gondii, Schistosoma japanicum and Angiostrongylus cantonensis were gifted by Prof. GR Yin (Shanxi Medical University), Dr. JH Lei (Tongji Medical College of Huazhong University of Science and Technology), and Prof. ZY Lv (Department of Parasitology, Zhongshan School of Medicine, Sun Yat-Sen University), respectively. All of the serum samples were stored at −80°C prior to use.
Forty-two serum samples from patients with trichinellosis were used to identify the sensitivity of the T. spiralis AW ES antigens. These serum samples were collected from the patients during two outbreaks with trichinellosis that occurred in the Yunnan province of southwestern China in 2003 and 2013 [20,21]. These patients with trichinellosis presented with a clinical suspicion of trichinellosis. The diagnosis of trichinellosis was confirmed for all of these patients by the presence of a high fever associated with periorbital or facial oedema, myalgia, a high level of eosinophilia, and a history of ingestion of raw or undercooked meat. All of these patients had positive ELISA or IIF results for trichinellosis, and two patients had a Trichinella-positive muscle biopsy. Out of 42 patients with trichinellosis, 22 serum samples were collected 35 dpi and 20 were collected 19 dpi. Sera used to assess cross-reactivity were selected from 109 patients with other parasitic diseases confirmed by parasitological examination of faeces, histopathological examination of biopsy samples or positive specific serological test results: schistosomiasis (36 patients with Schistosoma japonicum infection), paragonimiasis (20 patients with Paragonimus skrjabini infection), clonorchiasis (7 patients with Clonorchis sinensis), cystic echinococcosis (21 patients), Taenia solium cysticercosis (18 patients), and Spirometra erinaceieuropaei sparganosis (7 patients). These patients were diagnosed and serum samples were collected in our department [22,23]. Serum samples from 50 presumably healthy persons who tested negative for the above-mentioned parasitic diseases were also included in the study. All of the serum samples were stored at −80°C prior to use.

Collection of worms and preparation of ES antigens
One hundred mice were orally infected with 5000 ML of T. spiralis and euthanized 3 dpi. The small intestine was collected, cut along its entire length, and washed in pre-warmed PBS. Then, the small intestine was cut into pieces and incubated in PBS at 37°C for 1.5 h on a 300 μm sieve. The released AW were separated from the intestinal debris by filtration through a 200 μm sieve and differential sedimentation for 30 min. After several washes in PBS supplemented with 100 U penicillin/ml and 100 μg streptomycin/ml, the worms were centrifuged at 600 × g for 10 min and collected [19,24,25]. The average number of AW recovered from the infected mice was approximately 1000 worms per mouse. AW ES antigens were prepared as previously described [9,26]. Briefly, after thorough washing in sterile saline and serum-free RPMI-1640 medium supplemented with 100 U penicillin/ml and 100 μg streptomycin/ml, the AW were incubated in the same medium at a density of 2000 worms/ml at 37°C in 5 % CO 2 for 18 h. After incubation, the media containing the ES proteins were poured into 50-ml conical tubes, and the worms were allowed to settle for 20 min. The supernatant containing the ES products was filtered through a 0.2 μm membrane. The ES products were dialyzed and then lyophilized by vacuum concentration and freeze-drying (Heto-Holten A/S, Denmark). The protein concentration of the ES antigens of the adult worms was determined by the Bradford assay [27].
T. spiralis ML were recovered from the infected mice 42 dpi by artificial digestion [28,29]. The ML ES antigens were prepared as described above.
The crude (somatic) antigens of T. spiralis AW were prepared as previously described [30,31]. Briefly, adults were resuspended in deionized water. The suspension was subjected to 5 freeze-thaw cycles. The worms were homogenized on ice in a glass tissue grinder. Then, the larval fragments were further homogenized with ultrasonication (99 3-s cycles, 100 W, 0°C). The supernatant was collected after centrifugation at 15,000 × g for 1 h at 4°C. The protein concentration of the AW crude antigens was determined by the Bradford assay [27].

Enzyme-linked immunosorbent assay
The optional dilutions of various reagents were determined using checkerboard titration. ELISA was performed as previously described [15,32]. Briefly, 96-well ELISA plates (Corning, USA) were coated with AW ES antigens (2.0 μg/ ml), ML ES antigens (2.5 μg/ml) or AW crude antigens (1.5 μg/ml) in 100 μl of bicarbonate buffer (pH 9.6) overnight at 4°C. After blocking with PBS-0.1 % Tween 20 (PBST) containing 5 % skimmed milk at 37°C for 2 h, the following reagents were sequentially added and incubated at 37°C for 1 h: (1) human or mouse sera diluted 1:200 or 1:100, respectively, in PBST, and (2)  The cut-off value of the ELISA was evaluated for the three antigen preparations based on a 2.1-fold increase over the average OD value of the negative samples. Ratios <2.1 of the samples to be tested/negative sample (OD values of the samples to be tested divided by OD of the negative, S/N < 2.1) were regarded as negative, whereas S/N ≥2.1 was regarded as positive [15,23,33]. A test was considered valid when: (1) the OD values of the negative controls were higher than the OD values of the blank controls; and (2) the OD values of the negative controls were lower than the cut-off values. The sensitivity and specificity of the ELISA were assessed according to the following formulas: Sensitivity = no. of true positives/(no. of true positives + no. of false negatives) × 100 and Specificity = no. of true negatives/(no. of false positives + no. of true negatives) × 100 [34].

Statistical analysis
All statistical analyses were performed with SPSS for Windows, version 17.0 (SPSS Inc., Chicago, IL, USA). The sensitivity and specificity of the AW ES, crude antigens and ML ES antigens were compared with the Chi-square test. Repeated-measures analysis of variance was used to determine the difference between antibody levels at various time points post-infection and with different infecting doses. P < 0.05 was considered statistically significant.

The ELISA cut-off values
The cut-off values of the ELISA with the AW ES antigens, crude antigens and ML ES antigens for the detection of experimentally infected mice were 0.19, 0.23 and 0.20, respectively. The cut-off values of the ELISA with the above three antigen preparations for the detection of patients with trichinellosis were 0.41, 0.47 and 0.45, respectively.

Detection of anti-Trichinella IgG in mice infected with Trichinella and other parasites
The levels of anti-Trichinella IgG in the serum samples of mice infected with T. spiralis and other parasites were determined; the results are shown in Table 1. The sensitivity of the ELISA with the AW ES antigens, crude antigens and ML ES antigens for the detection of anti-Trichinella IgG in the sera of mice infected with 300 T. spiralis ML collected at 42 dpi was 100.00 % (35/35). No cross-reactivity of the ELISAs with the three antigen preparations was observed with the sera of mice infected with Angiostrongylus cantonensis, Spirometra mansoni, Schistosoma japonicum, and T. gondii and normal mice.
The levels of anti-Trichinella IgG in mice infected with other Trichinella species were assessed by ELISA; the results are shown in Table 2. The detection rates of anti-Trichinella IgG by ELISA with the AW and ML ES antigens were 100 % in mice infected with T. nativa, T. britovi and T. nelsoni. However, the antibody detection rate (90.91 %) of mice infected with T. pseudospiralis by ELISA with the ML ES antigens was 9.09 % higher than the ELISA with the AW ES antigens (χ 2 = 29.455, P < 0.05). When the sera of mice infected with T. nativa, T. britovi, T. nelson or T. pseudospiralis were assessed by ELISA with the AW crude antigens, the antibody detection rates were significantly lower compared to the ELISA with the ML ES antigens (χ 2 T2 = 29.939, χ 2 T3 = 17.368, χ 2 T4 = 36.667, and χ 2 T7 = 14.545, respectively, P < 0.05.

Serum anti-Trichinella IgG dynamics in mice experimentally infected with different numbers of parasites
The levels of anti-Trichinella IgG in the sera from the 2 groups of infected mice at different time points after infection were determined by ELISA with the three antigens preparations (Fig. 1). The results showed that the serum anti-Trichinella IgG levels in mice infected with 100 and 500 larvae began to increase rapidly 8 dpi and continued to elevate throughout the experimental period. There were significant differences in the anti-Trichinella IgG levels at different time points post-infection in the 2 groups of infected mice (F WA ES = 408.307, F AW crude = 204.921, and F ML ES = 165.990, P < 0.05). When the AW ES antigens were used, the anti-Trichinella IgG levels in the mice infected with 100 larvae were higher than the mice infected with 500 larvae (F = 68.023, P < 0.05). However, when the AW crude antigens and ML ES antigens were used, the anti-Trichinella IgG levels in the mice infected with 500 larvae were obviously higher compared to the mice infected with 100 larvae (F AW crude = 38.118 and F ML ES = 5.999, P < 0.05). In the mice infected with 100 ML, anti-Trichinella IgG was first detected by ELISA with the AW ES antigens,    2a and 3). In the mice infected with 500 ML, anti-Trichinella IgG was first detected by ELISA with the three antigens on 10, 8 and 10 dpi with detection rates of 60, 20 and 50 %, respectively; the antibody positivity rates reached 100 % on 12, 10 and 16 dpi, respectively (Fig. 2b).

Detection of anti-Trichinella IgG in serum samples of patients with trichinellosis and other parasitoses
The sensitivity of ELISA with the AW ES antigens, crude antigens and ML ES antigens for the detection of anti-Trichinella IgG in the sera of patients with trichinellosis was 100 % on 35 dpi (22/22) (

Discussion
In the present study, the AW ES antigens of T. spiralis were used as diagnostic antigens for the detection of anti-Trichinella IgG in the sera of experimentally infected mice. The sensitivity and specificity of the ELISA with the AW ES antigens were 100 % (35/35) and 100 % (106/106) in the infected mice, respectively. Moreover, there was no significant difference in the antibody detection rates in the sera of mice infected with T. nativa, T. britovi, and T. nelsoni by ELISA with the AW ES antigens and ML ES antigens (P > 0.05). The results suggested that there were common AW ES antigens among the encapsulated Trichinella species (i.e., T. spiralis, T. nativa, T. britovi, and T. nelsoni), but not in the non-encapsulated Trichinella species (T. pseudospiralis) [35,36]. Moreover, most of the epitopes of the AW ES antigens recognized by sera from infected mice were common to the encapsulated Trichinella species [37]. Our results indicated that the AW ES antigens of T. spiralis could also be used for the serodiagnosis of trichinellosis caused by other encapsulated Trichinella species.  Our ELISA results showed that the antigenic epitopes of the AW ES antigens were recognized by sera from infected mice 8-12 dpi, suggesting that the AW ES antigens might be secreted into the peripheral blood circulation of the host during the intestinal stage of Trichinella infection and induce an early antibody response that continues into the muscle larval stage [15]. Importantly, the anti-Trichinella IgG in 50 % of the mice infected with 100 larvae was detected by ELISA with the AW ES antigens as soon as 8 dpi, whereas the ELISA with the ML ES antigens did not permit detection of the infected mice before 12 dpi. In contrast, anti-Trichinella IgG in the mice infected with 500 larvae was first detected by ELISA with the AW ES antigens on 10 dpi. The mechanism underlying the earlier detection of anti-Trichinella IgG by the AW ES antigens in mice infected with 100 larvae compared to mice infected with 500 larvae is unclear, but may be due to the suppression of the host's immune response to the AW ES antigens by the higher infecting doses during the early stage of Trichinella infection [38,39]. Other studies showed that the recombinant proteins from the early adult worms 20 h post-infection were recognized by Western blotting as soon as 15-20 dpi in sera from pigs experimentally infected with 20,000 T. spiralis muscle larvae [40]. These results suggested that the AW ES antigens were more sensitive than the recombinant AW antigens for the serodiagnosis of trichinellosis. Thus, the AW ES antigens of T. spiralis exhibit potential for the early diagnosis of trichinellosis.
The sensitivity of the ELISAs with the AW ES antigens and ML ES antigens towards early patient serum samples collected on 19 dpi were 100 and 75 % (P < 0.05), respectively, suggesting that the AW ES antigens had early serodiagnostic value for human trichinellosis.
Cross-reactive antigens from different species of helminths have been known since the initial studies of Capron et al. [41]. Previous studies showed that the ELISA with the ML crude antigens exhibited obvious cross-reactivity with the sera of patients infected with other helminths [42]. Although the use of ML ES antigens could increase the sensitivity of the ELISA, 39.2 % of the false-positive results were observed with sera from patients with other parasitic diseases (cysticercosis, hydatidosis, schistosomiasis, fascioliasis, strongyloidosis, toxocariasis, anisakiasis or filariasis) [17]. When synthetic tyvelose antigens of T. spiralis ML were used for the serodiagnosis of trichinellosis, cross-reactivity occurred with cases of anisakiasis [16]. Recently, recombinant antigens of T. spiralis ML have been developed in some laboratories; however, the recombinant Ts21 or Ts 43 antigens still had a high occurrence of cross-reactivity with the sera of patients with   [43,44]. Western blotting of crude or ES antigens of T. spiralis ML is the serological method for the confirmatory test for trichinellosis recommended by the ICT. When specific IgG antibodies against the ML antigen components (40-70 kDa) are detected, the diagnosis of Trichinella infection is confirmed [10]. Western blotting with the ML crude or ES antigens is mainly used to confirm ELISA-positive sera from animals and humans. The sensitivity and specificity of this technique range from 96 to 99 % and from 95.6 to 99.6 %, respectively [45,46]. Hence, Western blotting has been widely applied to the serodiagnosis and seroepidemiological investigation of Trichinella infection in humans and animals [47][48][49][50]. However, the 40-70 kDa protein bands of the T. spiralis ML also exhibited cross-reactivity with the sera of patients with other helminthiases [17,51]. Moreover, shared antigens have been identified for T. spiralis ML and several other parasitic worms (Trichuris spp., Fasciola hepatica, Schistosoma mansoni, Schistosoma japonicum, Paragonimus westermani and Clornorcis sinensis) [52][53][54][55]. Anti-gp50 monoclonal antibodies against schistosomes and anti-gp50-positive sera from patients with schistosomiasis cross-reacted with the hypodermis and stichocytes of the T. spiralis ML [56]. These results demonstrated that different kinds of antigens from the T. spiralis ML had partial cross-reactivity with the sera of patients with other helminthiases. Hence, new sources of diagnostic antigens from other stages of T. spiralis should be developed to improve the sensitivity and specificity of the serodiagnosis of trichinellosis.
Our results showed that the specificity of the T. spiralis AW ES antigens was significantly higher than the ML ES antigens. No cross-reactions of the AW ES antigens were observed with the sera of patients with schistosomiasis, clonorchiosis, echinococcosis, or sparganosis or healthy persons. The AW ES antigens were sensitive and specific for the detection of anti-Trichinella IgG in serum samples of patients in the early stage of trichinellosis. The application of AW ES antigens for the serodiagnosis of trichinellosis could remarkably shorten the window when the detection of anti-Trichinella IgG is not possible by the ELISA with ML ES antigens. The AW ES antigens are a good alternative to the ML ES antigens and should be considered as a potential specific diagnostic antigen preparation for early trichinellosis. However, cross-reactivity of the ELISA with the AW ES antigens was observed with 5.56 % (1/18) of patients with cysticercosis and 10 % (2/20) of patients with paragonimiosis. Hence, a large-scale validation of the specificity of the AW ES antigens needs to be performed with the sera of the patients with other helminthiases. Moreover, the Trichinella-specific protein bands in the AW ES antigens should be identified and validated by Western blotting, and specific recombinant AW antigens should be developed in the future.

Conclusions
This study demonstrated that the AW ES antigens of T. spiralis were sensitive and specific for the early diagnosis of trichinellosis. These antigens provided a new source of diagnostic antigens and might serve as potential antigens for the early and specific serodiagnosis of trichinellosis. However, these results need to be further evaluated with a large-scale trial using the sera of patients with trichinellosis and other helminthiases.