Introduction

Trichinella spiralis is a tissue-dwelling parasitic nematode that infects humans and other mammals. Humans acquire the disease by ingesting raw or insufficiently cooked meat of pigs or other animals containing the infective larvae of the Trichinella parasite. Trichinellosis remains an important food-borne parasitic zoonosis with a cosmopolitan distribution. Of the 198 countries of the world, human trichinellosis has been documented in 55 (27.8%) countries around the world (Pozio 2007). In China, 17 outbreaks of human trichinellosis, with 828 cases and 11 deaths, were reported during 2000–2003 (Wang et al. 2006).

Unlike many parasites that demonstrate a high degree of host specificity, T. spiralis can be found in many species of carnivores and omnivores, but it requires only one host in its life cycle. After ingestion of contaminated meat, infective larvae are released from their capsules in the duodenum by the action of the host’s digestive enzymes. The infective larvae invade small intestinal epithelium where they undergo four molts to adulthood in 30–40 h (Ali Khan 1966; Kozek 1971). Worms copulate within a few hours after the final ecdysis, the gravid females produce newborn larvae that migrate through the blood and lymphatic circulation to muscular tissue, where they grow and encapsulate (Campbell 1983; Martinez and Rodriguez-Caabeiro 2005). As much as 99% of an oral dose of infective larvae is expelled from the intestinal epithelium in appropriately immunized adult rats and in neonatal rats whose dams are immune, this dramatic immune defense is called rapid expulsion (Bell and McGregor 1979; Appleton and McGregor 1984; Appleton et al. 1988; Bell et al. 1992; Ellis et al. 1994). Hence, whether the intestinal infective L1 larvae are expelled from intestinal or invade intestinal epithelium is the key step which the larvae infect the hosts. But, the mechanisms by which infective T. spiralis larvae recognize, invade, and migrate within the intestinal epithelium are unknown.

During their invasion and development, parasites encounter hostile host environment and immune defenses; therefore, they produce numerous molecules as survival strategies against these adverse changes (Bruschi 2002; Dzik 2006). Better understanding of these molecules would help elucidate the mechanism of parasite invasion and survival in the host and identify possible targets for vaccine development. The antigens of T. spiralis infective larvae can be classified according to their localization as surface, excretory–secretory (ES) and somatic antigens (Dea-Ayuela and Bolas-Fernandez 1999). Among them, ES antigens and surface antigens are directly exposed to the host’s immune system and are the main target antigens which induce the immune responses, so they may play an important role in the invasion and development process of Trichinella larvae (Bolás-Fernandez and Corral Bezara 2006; Robinson and Massie 2007).

Some studies have showed that when T. spiralis infective larvae are suspended in semisolid medium and inoculated onto monolayers of epithelial cells grown in vitro, the larvae invade cells, penetrate adjacent cells, and reside in the cytoplasm of the syncytia that they create (ManWarren et al. 1997; McVay et al. 2000; Gagliardo et al. 2002). Recently, we reported that infective larvae molt and ecdyse when they are inoculated into cultures of human colonic carcinoma cell line-8 (HCT-8) (Wang et al. 2010). However, the mechanism by which larvae recognize and invade the host’s intestinal epithelial cells at the cellular and molecular levels had not been determined yet. The aims of the present study were to further the protein change of HCT-8 cells induced in vitro by infective larvae of T. spiralis and identify the proteins related with invasion.

Materials and methods

Parasite

The isolate (ISS534) of T. spiralis used in the study were obtained from domestic pigs in Nanyang city of Henan Province, China. This isolate was maintained by serial passage in Kunming mice at 6–8-month intervals in our department (e.g., Department of Parasitology, Medical College, Zhengzhou University). T. spiralis muscle larvae were recovered from infected Kunming mice by digestion of carcasses with 0.33% pepsin (1:31,000) and 1% HCl (Li et al. 2010). These Kunming mice had been infected at least 42 days prior to larvae collection. For in vitro experiments, muscle larvae were activated by incubation in 5% bovine bile diluted with saline at 37°C in 5% CO2 for 2 h. Then, the larvae were washed four times in saline supplemented with 100 U penicillin/ml and 100 μg streptomycin/ml, and incubated in saline at 37°C in 5% CO2 for additional 1 h (ManWarren et al. 1997).

Experimental animals

BALB/c mice aged 6 weeks were purchased from the Experimental Animal Center of Henan province and bred in plastic micro-isolator cages were used for the study.

Cell culture

Human colonic carcinoma cell line HCT-8 was obtained from Cell Resource Center of Shanghai Institute for Biological Sciences of Chinese Academy of Sciences. Cells were cultured in RPMI-1640 with l-glutamine, nonessential amino acids and 10% fetal bovine serum (FBS). Monolayers were dispersed by trypsinization (0.25% trypsin, 0.02% ethylenediaminetetraacetic acid; EDTA).

Preparation of ES antigens and surface antigens

The ES antigens of T. spiralis muscle larvae were prepared as described previously (Mahannop et al. 1992; Kapel and Gamble 2000). In brief, after washing thoroughly in sterile saline, the larvae were again washed four times in serum-free RPMI-1640 medium supplemented with 100 U penicillin/ml and 100 μg streptomycin/ml. The larvae were incubated in the same medium at concentration of 5 000 worms/ml for 18 h at 37°C in 5% CO2. After incubation, the media contained the ES products were filtered through a 0.2 μm membrane into a 50-ml conical tube, then centrifuged at 4°C, 15,000×g for 30 min. The supernatant was dialyzed against deionized water at 4°C for 2 days.

Surface antigens of muscle larvae were prepared according to the method described by Pritchard et al. (1985). The larvae were culture in phosphate-buffered saline (PBS; pH 7.4, 1/15 mol/L) contained 0.25% hexadecyl trimethyl ammonium Bromide (Sigma, USA) and 2% sodium deoxycholate (Sigma, USA) at 37°C for 2.5 h. The supernatant was obtained by centrifugation at 4°C, 11,000×g for 20 min, and dialyzed against deionized water at 4°C for 2 days.

The supernatant contained ES or surface antigens were concentrated by a vacuum concentration and freeze drying (Heto Mxi-Dry-Lyo, Denmark), respectively. The protein concentration of ES antigens (1.26 mg/ml) and surface antigens (0.86 mg/ml) was determined by the method described by Bradford (1976). The ES and surface antigens were aliquoted and stored at −20°C before use.

Generation of mouse polyclonal antibodies to ES and surface antigens

Twenty male BALB/c mice were divided into two groups of ten mice each. Pre-immune sera were collected by tail bleeding 2 days prior to the first immunization. Two groups of BALB/c mice were subcutaneously immunized with 20 μg ES or surface antigens emulsified with complete Freund’s adjuvant (FCA), followed by three boosts with the same amount of protein emulsified with incomplete FCA at 2-week intervals. Seven days after the last boost, mice were bled and the sera were collected.

Antibody titer analysis of immune sera by ELISA

The specific IgG antibody titer of immune mouse sera were assayed by an indirect enzyme-linked immunosorbent assay (ELISA) using corresponding ES or surface antigens. The procedure was performed as previously described (Wang et al. 2006). In brief, 96-wells ELISA plates (Corning, USA) were coated with 2.5 μg proteins/well ES or surface antigens in 100 μl of bicarbonate buffer (pH 9.6) overnight at 4°C, were washed three times with 0.1% Tween-20 in PBS (PBS-T). After blocking with 3% skimmed milk in PBS-T, serial dilutions of immune sera were added. The plates were then washed and incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG (Sigma, USA) diluted at 1:5,000. The reactions were detected by addition of the substrate O-phenylendiamine dihydrochloride (Sigma, USA) plus H2O2 and stopped with 50 μl/well of 2M H2O2. Quantification of the reactions was determined by absorbance at 492 nm with an ELISA reader (TECAN, Austria). The IgG antibody titer of immune sera against ES and surface antigens was 1: 5 × 105 and 1: 1 × 105, respectively.

Immunofluorescencent test

Sterile coverslips (18 × 18 cm) were placed in 6-well culture plates (Costar, US) and the HCT-8 cells were grown to confluence on coverlips. The cell monolayer was cultured in 1640 medium with 20 μg ES antigens or surface antigens/ml medium at 37°C in 5% CO2 for 2 h, and then the coverslips bearing cell monolayer were fixed with cool acetone for 10 min after being washed three times with PBS. After blocking with 2% bovine serum albumin and 5% normal goat sera diluted in PBS, the monolayer was incubated in a moist chamber at 37°C for 40 min with 1:10 dilution of sera from mice infected with T. spiralis, mice immunized with ES or surface antigens, and normal mice. After washing three times in PBS, coverslips bearing the monolayer incubated with a 1:20 dilution of FITC-labeled goat anti-mouse IgG (Biotechnology Company of Zhongshan Goldenbridge, China), washed five times in PBS, and examined under fluorescent microscope (Olympus, Japan; Cui et al. 1999).

Protein preparation of HCT-8 cells cultured with T. spiralis larvae

The cell monolayer was overlaid by bile-activated larvae suspended in RPMI-1640 medium supplemented with l-glutamine, nonessential amino acids, 15 mM HEPES and 10% FBS. Following incubation for at 37°C, 5% CO2 for 18 h, larvae and medium were removed from the monolayer. Cells were harvested by trypsinization (0.5% trypsin, 0.02% EDTA), and then centrifuged at 4°C, 1,000 g for 5 min. The cells in 75 ml culture flask were mixed with 0.5 ml cell lysis (pH 7.2, 60 mM Tris–HCl; 2% sodium dodecyl sulfate; SDS). The cells were boiled at 100°C for 5 min, and then centrifuged at 4°C, 12,000×g for 5 min, and the supernatant was collected and stored at −20°C before used.

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis

Protein samples from the cell lysates were diluted with loading buffer (250 mM Tris–HCl pH 6.8, 50% glycerol, 10% SDS, 5% 2-mercaptoethanol, 0.5% bromophenol blue) up to a concentration of 15 µg/lane and denatured at 95°C for 10 min (Nöckler et al. 2009). After cooling, the proteins were separated by SDS-polyacrylamide gel electrophoresis (PAGE) on 12% acrylamide separating gel and 5% acrylamide stacking gels (83 × 73 × 1.0 mm) in a Mini-PROTEAN 3 Cell electrophoresis unit (Bio-Rad, USA) at 120 V for 2.5 h (Cui et al. 2003). The prestained protein markers were run in parallel. After electrophoresis, the gel was stained with 0.25% Coomassie brilliant blue R-250 for 4 h, and then bleached with the eluate (100 ml acetic acid, 50 ml ethanol, 850 ml dH2O). A second gel was prepared with the same proteins.

Western blotting analysis

After electrophoresis, proteins were transferred at 20 V for 25 min onto nitrocellulose membrane by blotting the gel in a trans-blot SD transfer cell with the electrode buffer recommended by the supplier (Bio-Rad, USA; Cui et al. 2003). After blotting, the membranes were stained with Ponceau S to verify transfer and to locate the protein marker and cut into strips. Each strip was blocked in 3% skimmed milk in TBST at 37°C for 2 h, incubated at 37°C for 1 h with 1:100 dilutions of different mouse sera. After washing, the strips were incubated at 37°C for 1 h with HRP-conjugated goat anti-mouse IgG (1:100 dilution), and finally with 3, 3′-diaminobenzidine tetrahydrochloride (Sigma, USA). The reaction was finally stopped by washing the strips with distilled water. Finally, the molecular sizes of the bands were evaluated by comparison with the molecular size ladder using a gel documentation system (GeneGenius, Syngene, USA) and related software (GeneSnap 6.08 and GeneTools, Syngene, USA).

Results

IFT of HCT-8 cells incubated with ES or surface antigens of T. spiralis muscle larvae

In this experiment, we wanted to identify the larval antigens binding onto the surface of HCT-8 cells. The IFT results were shown in Fig. 1. Normal HCT-8 cells had positive reaction with sera of mice immunized with ES or surface antigens and sera of mice infected with T. spiralis, but they had negative reaction with sera of normal mice. After incubated with ES or surface antigens, HCT-8 cells had stronger positive reaction with sera of mice immunized with ES or surface antigens and infected with T. spiralis, but they still had negative reaction with sera of normal mice.

Fig. 1
figure 1

Immunofluorescent staining of HCT-8 cells. Normal HCT-8 cells were recognized by sera of mice immunized with ES antigens (A1), sera of mice infected with T. spiralis (A2), normal mouse sera (A3), and PBS (A4); HCT-8 cells cultured with ES antigens were recognized by sera of mice immunized with ES antigens (B1), sera of the infected mice (B2), normal mouse sera (B3), and PBS (B4); HCT-8 cells cultured with surface antigens were recognized by sera of mice immunized with surface antigens (C1), sera of the infected mice (C2), normal mouse sera (C3), and PBS (C4); Normal HCT-8 cells were recognized by sera of mice immunized with surface antigens (C5)

Full size image

Change of protein components of HCT-8 cell induced in vitro by T. spiralis infective larvae

The lysates of HCT-8 cells were analyzed with SDS-PAGE and the results were shown in Fig. 2. Normal HCT-8 cells have 30 protein bands with a molecular weight of 131, 118, 108, 104, 96, 90, 85, 80, 74, 72, 68, 66, 61, 57, 53, 48, 46, 45, 43, 39, 37, 34, 32, 29, 28, 25, 23, 21, 20, and 15 kDa. However, we found that after HCT-8 cells were incubated with infective larvae, additional three protein bands with 38, 36, and 17 kDa were observed, and one protein band with 72 kDa was disappeared, compared with normal HCT-8 cells.

Fig. 2
figure 2

SDS-PAGE of the lysates of HCT-8 cells cultured with T. spiralis infective larvae. Proteins were separated in 12% polyacrylamide gels under reducing condition, the gel was stained with Coomassie brilliant blue R-250, and relative molecular weights (Mw) of standard markers are shown in kDa as described in “Materials and methods”. M protein marker with low molecular weights, 1 proteins from lysate of normal HCT-8 cells, 2 proteins from lysate of HCT-8 cells cultured with L1 larvae

Full size image

Western blot of HCT-8 cell lysates analyzed by immune sera were shown in Fig. 3. The results showed that 14 protein bands (131, 90, 68, 53, 48, 46, 45, 43, 39, 34, 32, 29, 25, and 23 kDa) of normal HCT-8 cells were recognized by sera of the infected mice. But after HCT-8 cells were incubated with infective larvae, 16 protein bands (131, 90, 68, 66, 61, 57, 53, 48, 46, 45, 39, 34, 32, 29, 25, and 17 kDa) were recognized by sera of the infected mice. Comparing with normal HCT-8 cell proteins, the additional four protein bands (66, 61, 57, and 17 kDa) of HCT-8 cells incubated with infective larvae were recognized by sera of the infected mice, but two protein bands (43 and 23 kDa) were not recognized.

Fig. 3
figure 3

Western blot analysis of the lysates of HCT-8 cells cultured with T. spiralis infective larvae (1, 3, 5, and 7) and normal HCT-8 cells (2, 4, 6, and 8). Proteins are recognized by different T. spiralis immune or infected sera. M protein marker with low molecular weights, 1 and 2 sera of the infected mice, 3 and 4 sera of mice immunized with surface antigens, 5 and 6 sera of mice immunized with ES antigens, 7 and 8 sera of normal mice

Full size image

When sera of mice immunized with surface antigens was used in Western blot, we found that six protein bands (48, 46, 34, 32, 29, and 23 kDa) of normal HCT-8 cells were recognized; after HCT-8 cells were incubated with infective larvae, six protein bands (48, 46, 45, 34, 32, and 29 kDa) were recognized. Comparing with normal HCT-8 cells, one additional 45-kDa protein was recognized, but the 23-kDa protein was not recognized after HCT-8 cells were incubated with infective larvae.

When sera of mice immunized with ES antigens were used, three protein bands (90, 48, and 43 kDa) of normal HCT-8 cells were recognized; after HCT-8 cells were incubated with infective larvae, six protein bands (90, 45, 43, 34, 21, and 17 kDa) were recognized. Comparing with normal HCT-8 cells, four additional protein bands (45, 34, 21, and 17 kDa) of HCT-8 cells were recognized, but the 48-kDa protein could not be recognized after HCT-8 cells were incubated with infective larvae.

Discussion

T. spiralis larvae cultured in bile or gut contents for 2–3 h invade and migrate through cells in a monolayer (ManWarren et al. 1997). In this paper, the proteins of HCT-8 cell cultured with activated larvae for 18 h were analyzed by SDS-PAGE, we found that normal HCT-8 cells have 30 protein bands, but after HCT-8 cells were incubated with infective larvae, additional three protein bands with 38, 36, and 17 kDa were observed, and one protein band with 72 kDa was disappeared. The results suggested that the additional three proteins possibly is the proteins released by the L1 larvae when the larvae contacted with HCT-8 cells, and the protein lost may be the release of cell-specific mediators that the infective larvae to invade cell (Butcher et al. 2000); however, the characteristic of cell-specific mediators and process of the mediators recognized by the larvae is poorly understood.

IFT and Western blot showed that both of normal HCT-8 cells and their proteins had positive reaction with sera of mice immunized with surface or ES antigens, and sera of the infected mice. The results suggested that HCT-8 cells may have the epitopes similar with Trichinella antigens. Additionally, IFA showed that after cultured with ES or surface antigens, HCT-8 cells had stronger positive reaction with above-mentioned sera than normal HCT-8 cells, suggesting that certain proteins from ES or surface antigens may be combined onto the cell surface or enter the cells.

Western blot also showed that after cultured with L1 larvae, seven additional protein bands (66, 61, 57, 45, 34, 21, and 17 kDa) of HCT-8 cells were recognized by sera of the infected or immunized mice. Among them, the 45 kDa protein was recognized by sera of mice immunized with both surface antigens and ES antigens; the 17 kDa protein was recognized by both sera of the infected mice and sera of mice immunized with ES antigens. On the other hand, after cultured with L1 larvae three protein bands (48, 43, and 23 kDa) of HCT-8 cells were not recognized by sera of the infected or immunized mice, comparing with normal HCT-8 cells. The 23-kDa protein could not be recognized by both sera of the infected mice and sera of mice immunized with surface antigens. The fact that the invasion of intestinal epithelia by T. spiralis can be inhibited with the antibodies to ES antigen and tyvelose that recognize several of the nematode’s ES antigens, suggests these ES antigens play a role in invasion (McVay et al. 1998; Inaba et al. 2003; Wang et al. 2010). Our observation suggested that after HCT-8 cells were incubated with L1 larvae, seven additional proteins (especially 45 and 17-kDa protein), which were recognized by both sera of the infected and immunized mice, may be related with invasion of intestinal epithelial cells by infective larvae. These proteins may proceed from the larval ES or surface antigens from the parasite that might mediate or facilitate entry into the intestinal epithelial cells. The 45-kDa protein is a tyvelose-bearing glycoprotein, which is an important antigen protein secreted by L1 larvae and is present in much greater amounts in the ES antigens, but its function remains unknown (Nagano et al. 2009). The 48, 43, and 23-kDa proteins (especially the 23-kDa protein), which were not recognized by sera of infected mice or immunized mice, may be the specific mediators or signal transduction protein released by HCT-8 cells when incubated with L1 larvae.

The infective larvae of T. spiralis do not invade all the epithelial cells. The larvae can invade some epithelial cell lines (human intestinal and colonic epithelial cells, rat intestinal epithelial cells and canine kidney epithelial cells, etc.), but cannot invade rat jejunum crypt cells IEC-6 (ManWarren et al. 1997). Both invasion-resistant and susceptible epithelial cells incubated with infective larvae undergo non-lethal wounding, although these injuries may not lead to the invasion of the larvae. When resistant IEC-6 cells were cultured with T. spiralis larvae, the larvae cannot invade cells, but ES antigens can be found within the cells. If ES antigens were added into culture medium, they cannot be uptaken by the cells, which suggested that ES antigens may not enter the cells through endocytosis but through the cell membrane damaged by the larvae into the cells (Butcher et al. 2000). The cell membrane wounding caused by larvae or entrance of ES antigens into cells is insufficient to allow entry of the parasite into resistant epithelial cells, suggesting that the invasion of larvae also requires the specific signals from host cell which permits or prohibits invasion.

The mechanism of the intestinal nematodes invade intestinal mucosa may be involved with both mechanical and chemical damage. For example, the hyaluronidase released by Ancyclostoma caninum and Anisakis sinpllex within the host small intestine may be related to degrade mucosa, invasion, and histolysis (Hotez et al. 1994). Trichuris, not only produce a pore-forming molecule but employs a buccal stylet to mechanically pierce cells (Drake et al. 1994). As for the infective larvae of T. spiralis, they do not possess oral appendices or a spike (Bruce 1970) and invade all the cells; the invasion of intestinal epithelial cells by infective larvae may be not only a result of mechanical penetration. The nematode relies on sensory mechanoreceptors and chemical receptors around its mouth to respond to these stimuli (Mclaren 1976). Additionally, the sera immunized with surface antigens or ES antigens can prevent the larvae invade the intestinal epithelial cells (Wang et al. 2010). Hence, it is proposed that the invasion of T. spiralis infective larvae relies on determined characteristics of intestinal epithelial cells and the parasite must receive chemical signals from the cells to initiate the invasive process (ManWarren et al. 1997; Romaris and Appleton 2001). Our results showed some proteins secreted by the larvae may cause the change of antigenic proteins on cell surface or within cells when larvae contact with the cells, meanwhile the larvae may induce the cells to release some proteins which possibly is the cell-specific mediators promoting the larvae to invade cells.