Introduction

Opisthorchiasis caused by Opisthorchis viverrini remains a major public health problem in many parts of Southeast Asia including Thailand, Lao People’s Democratic Republic, Vietnam, and Cambodia (IARC 1994). The infection is associated with a number of hepatobiliary diseases, including cholangitis, obstructive jaundice, hepatomegaly, cholecystitis, and cholelithiasis (Harinasuta et al. 1984; Sripa 2003). Moreover, both experimental and epidemiological evidence strongly implicate liver fluke infection as the major risk factor in cholangiocarcinoma, cancer of the bile ducts (Thamavit et al. 1978; IARC 1994; Sripa et al. 2007).

Proteases (peptide hydrolases) of parasites, aside from known catabolic functions and protein processing, play diverse roles in parasites, including excystment–encystment, immunoevasion, digestion of host tissue, and activation of inflammation leading to pathology (Sajid and McKerrow 2002; Williamson et al. 2003; Donnelly et al. 2006; McKerrow et al. 2006; Knox 2007). Proteases have been reported in a number of liver flukes (e.g., Cordova et al. 1999; Smooker et al. 2000; Park et al. 2001; Dalton et al. 2003). Cysteine proteases have been purified from the oriental liver fluke of humans and close relative of O. viverrini, Clonochis sinensis (Park et al. 1995, 2001; Chung et al. 2000), and a 24-kDa cysteine protease from C. sinensis exhibits cytotoxic effects on Chinese hamster ovary cells (Park et al. 1995). In addition, cysteine proteases are promising antigens for immunodiagnosis (Kim et al. 2001) and vaccination against (Lee et al. 2006) clonorchiasis. However, to our knowledge, protease activity has not yet been reported from O. viverrini. In this study, we showed that the major catalytic activity detected in O. viverrini extracts and secreted proteins belongs to the cysteine protease class, specifically ~30-kDa cathepsin L-like enzymes.

Materials and methods

Parasite preparations

O. viverrini metacercariae were obtained from naturally infected cyprinoid fish captured from an endemic area of Khon Kaen province, Thailand. The fish were digested with pepsin–HCl at 37°C for 2 h. After several washes in normal saline and sedimentations, the metacercariae were collected and identified under a dissecting microscope. Moving viable metacercariae were used for infecting hamsters. Adult O. viverrini worms obtained from the livers and bile ducts of infected hamsters were washed several times in cold normal saline containing penicillin (200 U/ml) and streptomycin (200 μg/ml) to remove any debris and residual blood. After washing, viable worms were used for collection of excretory–secretory (Ov-ES) products (see below), and inactive worms were snap frozen in liquid nitrogen and stored at −80°C. For somatic extracts (Ov-SE), frozen worms were crushed and ground in liquid nitrogen. The ground powder was then solubilized in sterile phosphate-buffered saline (PBS), sonicated on ice, centrifuged at 10,000 rpm for 30 min at 4°C and the supernatant stored at −80°C. Ov-ES products were prepared by in vitro culture of viable flukes in Roswell Park Memorial Institute medium containing penicillin (100 U/ml) and streptomycin (100 kg/ml) and the spent medium was collected every 6 h for up to 5 days. Dead worms were periodically removed. The culture media were centrifuged and the supernatant and pellets were stored at−80°C for preparing Ov-ES and egg soluble extract (Ov-EG), respectively. For Ov-ES, the pooled supernatant was concentrated by membrane ultrafiltration (Amicon Ultra-15, Millipore, MA, USA), dialyzed in PBS several times, and then aliquoted at −80°C. For Ov-EG, the eggs were washed several times in normal saline to remove any residual Ov-ES products. After centrifugation, the egg pellets were crushed in liquid nitrogen as above and centrifuged and the supernatant was aliquoted at −80°C. Protein concentrations of all O. viverrini preparations were determined by the Bradford method (Bio-Rad, Hercules, CA, USA).

Gelatinolytic activity

The X-ray film method described by Cheung et al. (1991) was initially used to detect gelatinolytic activity. The technique relies on the presence of a gelatin emulsion which coats commercial X-ray film; the gelatin in this emulsion can serve as the substrate to assay for protease activity. Briefly, Ov-ES or Ov-SE (10 μg protein/10 μl) were twofold serially diluted in an equal volume of PBS (pH 6.0) and then 2 μl of the extracts were spotted onto unprocessed X-ray film (Kodak). The experiment was conducted in a moisture chamber at 37°C overnight. The X-ray film was developed in Kodak D-19 Developer (Sigma P-5670) and fixed in Kodak Fixer (Sigma, P-6557) before washing in tap water. The gelatinolytic activity was visualized as a zone of clearance, using a dissecting microscope. Papain (1 μg/ml, Sigma) and PBS served as the positive and negative controls, respectively.

Zymography (Gelatin–SDS–PAGE)

O. viverrini protein preps diluted 1:1 in non-denaturing sample buffer were applied to a sodium dodecyl sulfate (SDS)–polyacrylamide gel containing (PAGE) 0.1% porcine gelatin (Sigma) in the separating gel and electrophoresis was performed to resolve the proteins. To activate the proteases by regeneration of the enzyme in situ, the gel was incubated in 0.1 M sodium phosphate, pH 6.0, 0.1% Triton X-100 and 5 mM dithiothreitol for 1 h to remove SDS. Subsequently, the gel was stained with Coomasie Brilliant Blue and destained according to standard procedures. Activity of proteases was detected as bands of clearance of the gelatin against the blue background representing Coomasie-stained (undigested) gelatin.

Fluorometric assay for protease activity

Protease activity(ies) in O. viverrini metacercariae soluble extract (Ov-MC), Ov-SE, Ov-EG and Ov-ES were measured fluorometrically using peptide-7-amino-4-methylcoumarin substrates (Hill and Sakanari 1997). The following buffers were used for pH activity profiles: 0.1 M sodium acetate (pH 4.0–5.0), 0.1 M sodium phosphate (pH 5.5–7.0), 0.1 M Tris-HCl (pH 7.5–8.0), and 0.1 M glycine (pH 8.5–11.0). Class-specific substrates assessed included Z–Arg–AMC (cysteine protease, cathepsin H), Z–Arg–Arg–AMC (cysteine protease, cathepsin B), Z–Phe–Arg–AMC (cysteine protease, cathepsin L), o-aminobenzoyl–Ile–Glu–Phe–nPhe–Arg–Leu–NH2 (aspartic protease), Z–Gly–Gly–Arg–AMC (serine protease). Protease inhibitors tested were Z–Ala–CHN2 (cysteine proteases), E-64 (cysteine proteases), pepstatin (aspartic protease), and soy bean trypsin (serine protease). The substrates and inhibitors were purchased from Sigma. The assay conditions used were as follows: 50 μl of appropriate AMC substrate (10-μM final concentration) was incubated with 50 μl of Ov-ES, Ov-SE, Ov-EG, or Ov-MC (1-μg of final amount–well) for 60 min at 37°C in 96-well black clear-bottom plates (Corning-Costar, NY, USA) in appropriate buffer. Inhibition assays were carried out by incubating class-specific protease inhibitors with parasite extracts for 10 min prior to the addition of substrate. Cleavage of AMC was detected using a fluorescence plate reader (Fluorostar Optima, BMG) with excitation at 360 and emission at 460. The amount of AMC released was determined from a standard curve generated using free AMC (Sigma), and one unit (U) of enzyme activity was defined as the amount that catalyzed the release of 1 nM of AMC per minute per milligram of protein at 37°C.

Purification of O. viverrini cysteine protease

Proteases from Ov-SE and Ov-ES were purified by affinity chromatography using thiol-sepharose (Knox et al. 1999). Briefly, Ov-SE or Ov-ES were dialyzed into 0.1-M sodium phosphate buffer, pH 6.0. Thiol-sepharose 4B resin (Sigma) was equilibrated with the same buffer, then 2.0 mg of each protein extract in a final volume of 2 ml was loaded onto the thiol-sepharose column equipped with Bio-Rad Econo chromatographic system at a flow rate of 1 ml/min. After thoroughly washing with the same buffer, bound material was eluted from the column by washing with equilibration buffer containing 25 mM cysteine. The peak fractions were pooled and the cysteine removed by passage through a Sephadex-G25 column which was equilibrated in 10 mM Tris, 0·1% reduced Triton X-100, 0·1% sodium azide, pH 7·4. The protein peak fractions were pooled and then concentrated using Amicon Ultra-15 (Millipore, MA, USA) before protein determination by using the Bradford method. The thiol-sepharose-purified protein was analyzed by SDS–PAGE and assayed for protease activity as described above.

Results

Ov-ES and Ov-SE digested gelatin impregnated in X-ray film, visualized as a clear zone similar to that generated when the gelatin was digested with papain. The PBS negative control did not digest gelatin. Proteolytic activity was detected in Ov-ES at a titer as low as 1:64 (~0.0156 μg), whereas activity in Ov-SE was detected at a titer as low as 1:32 (~0.0313 μg; Fig. 1a). O. viverrini protease activity was then assessed by gelatin acrylamide gel electrophoresis (zymography) to determine the molecular mass of the proteases responsible for this activity. Ten micrograms of Ov-SE and Ov-ES showed detectable protease activity with clear zones on the gel corresponding to molecular mass of approximately 30 kDa for both extracts (Fig. 1b).

Fig. 1
figure 1

Gelatinolytic activity of Opisthorchis viverrini protease(s) as detected using X-ray film emulsion (a) and in gelatin SDS–PAGE gels (zymography; b). Both the somatic extract (Ov-SE) and excretory–secretory products (Ov-ES) exhibited protease activity as revealed by clear zones of gelatin digestion up to a titer of 1:64 for a serially diluted Ov-ES (a). At pH 6.0 in the presence of the reducing agent DTT, gelatinolytic activity ascribable to a protease of ~30 kDa in size was readily apparent (b)

Full size image

The mechanistic classes of enzymes responsible for this activity were further explored using specific fluorogenic peptide substrates. Ov-SE showed highest protease activity per milligram of protein when Z–Phe–Arg–AMC was used as a substrate. The protease activity of cysteine, serine, and aspartic protease was 59.83, 5.70 and 6.8 U/μg protein, respectively. Class-specific cysteine protease activity was shown to be mainly cathepsin L (59.83 U) but not cathepsin H (6.55 U) or cathepsin B (7.43 U; Table 1). The cathepsin L activity was inhibited with E-64 and Z–Ala–CHN2 while weaker activity of the other classes of enzyme inhibited by their class-specific inhibitors (Table 2).

Table 1 Protease activities in O. viverrini somatic extract (Ov-SE) against the synthetic peptides Z–Arg–AMC (diagnostic for cathepsin H), Z–Arg–Arg–AMC (cathepsin B), Z–Phe–Arg–AMC (cathepsin L), AB–NH2 [o-aminobenzoyl–Ile–Glu–Phe–nPhe–Arg–Leu–NH2 (aspartic protease)] and Z–Gly–Gly–Arg–AMC (serine protease)
Full size table
Table 2 Inhibition of protease activities of Ov-SE, pOv-CP-ES, and pOv-CP-SE
Full size table

The greatest cysteine protease activity was found in metacercariae extract followed by excretory–secretory products, egg extract, and adult worm extract. The protease activities using Z–Phe–Arg–AMC of Ov-MC, Ov-ES, Ov-EG, and Ov-SE were 97.54, 81.17, 77.37 and 64.19 U, respectively. Cleavage of Z–Phe–Arg–AMC by Ov-SE was minimal at pH 4.0, was increased sharply at pH 5.0, was optimal at pH 6.0, and was dropped off at pH 6.5. Cleavage of the substrate was inhibited by 98% with E-64 and 86% with Z–Ala–CHN2; pepstatin-A and soybean trypsin inhibitor did not inhibit cleavage (Table 2).

O. viverrini cysteine protease(s) was purified from Ov-SE and Ov-ES using thiol-sepharose, and purified eluate proteins are referred to as pOv-CP-SE and pOv-CP-ES, respectively. Both eluates migrated with approximate molecular mass of 30 kDa (Fig. 2). Both purified proteases digested gelatin X-ray film at a dilution of 1:256 (Fig. 3a), indicating that the specific activity had been enhanced compared with Ov-SE (1:32) and Ov-ES (1:64) by affinity purification. The pH profiles of cleavage of Z–Phe–Arg–AMC by pOv-CP-ES and pOv-CP-SE was the same as that observed with the crude extract, Ov-SE (Fig. 3b), and the activity was inhibited by the cysteine protease inhibitors, E-64 and Z–Ala–CHN2, but not by inhibitors for serine, aspartic proteases, and metalloproteases (Table 2). Details of the protease activities in different Opisthorchis protein extracts and thiol-purified products are shown in Table 3.

Fig. 2
figure 2

SDS–PAGE–Coomassie-stained gel analysis of thiol-sepharose-4B-purified O. viverrini proteins from excretory–secretory product (pOv-CP-ES; lane 1) and somatic extract (pOv-CP-SE) (lane 2). Molecular masses in kilodaltons are shown at the left

Full size image
Fig. 3
figure 3

Gelatinolytic activity of thiol-sepharose purified O. viverrini protease(s) from excretory–secretory product (pOv-CP-ES) and somatic extract (pOv-CP-SE): a activity detected on X-ray film substrate; b pH optima for enzyme activity of the two parasite extracts against Z–Phe–Arg–AMC. In b, enzyme activities of 1-μg Opisthorchis extract per reaction were determined in duplicate and mean values are presented. One unit of enzyme activity catalyzes the release of 1 nM of AMC per minute per milligram of protein

Full size image
Table 3 Purification by affinity chromatography of cysteine protease activity from extracts and secretions of Opisthorchis viverrini
Full size table

Discussion

Proteases of a number of parasites have been investigated to elucidate their roles in parasitism, pathogenesis, and pathology (Auriault et al. 1982; Tamashiro et al. 1987; McKerrow 1989; Sakanari et al. 1989; Carmona et al. 1993). Accordingly, cysteine proteases of parasites are known to play roles in diverse developmental processes including egg hatching and subsequent stage transitions, invasion, and migration through host tissues. They also participate in nutrient acquisition and immunological modulation (Chung et al. 1995; Michel et al. 1995; Ward et al. 1997; Kong et al. 1998; Mottram et al. 1998; Syfrig et al. 1998). These enzymes might also be developed as targets for vaccines (Jankovic et al. 1996; Piacenza et al. 1999) and chemotherapy (Coombs et al. 1997; Engel et al. 1998; Abdulla et al. 2007). Whereas numerous reports of trematode proteases are available, none has focused on the proteolytic enzymes of the Oriental liver fluke, O. viverrini. In the present study, we have demonstrated and characterized protease activities in the developmental stages of O. viverrini. A major band of protease of ~30 kDa was enriched from both Ov-ES and Ov-SE. On activated thiol-sepharose 4B, class-specific substrate and inhibition profiles demonstrated this thiol-sepharose-enriched enzyme to be an O. viverrini cathepsin L. Cathepsin L is a papain-like cysteine protease, an endopeptidase belonging to clan CA of the cysteine proteases according to the MEROPS classification, ID C01.032 (http://merops.sanger.ac.uk/cgi-bin/make_frame_file?id=C01.032). The enzymological profile of the cathepsin L activity from O. viverrini is similar to cathepsin-L-like proteases reported from related fluke species. In particular, the O. viverrini cysteine protease displayed optimal catalytic activity at pH 6.0 similar to that of Fasciola gigantica protease which has optimal activity at pH 5.5 to 7.5 (Mohamed et al. 2005). The 27-kDa cysteine protease of Paragonimus westermani exhibited endodipeptidolytic activity at pH 5–8.5 and remained active and stable at neutral pH for 3 days (Yamakami and Hamajima 1987). P. westermani cysteine protease is inhibited by the cysteine protease inhibitor E-64 (Chung et al. 1995). Moreover, the proteases efficiently hydrolyzed collagen, fibronectin, and myosin at pH 8 (Chung et al. 1997).

Furthermore, the O. viverrini activity profile was in general similar to the protease activity profiles of other trematode parasites reported to date, including for C. sinensis, P. westermani, Fasciola hepatica and the human schistosomes (e.g., Song et al. 1990; Chung et al. 1995; Thorsell et al. 1965; Tort et al. 1999; Bogitsh et al. 2001; Delcroix et al. 2006). Minor activities ascribable to serine and aspartic proteases also were apparent in the O. viverrini extracts. Metacercariae of O. viverrini displayed highest activity against Z–Phe–Arg–AMC, followed by excretory–secretory product, egg, and adult worm, respectively. This is a similar developmental expression of protease activity to C. sinensis (Song and Rege 1991). Protease activities are also present in P. westermani excretory–secretory products of newly excysted metacercariae, and in a similar fashion to O. viverrini, mature adult P. westermani lung flukes show less specific activity of cysteine protease than do the metacercariae (Chung et al. 1995). In P. westermani metacercariae, much of the cysteine protease is localized in the excretory bladder, in excretory granules, and in newly excysted juvenile worms (Chung et al. 1995). Early release of the proteases from the excretory bladder of the encysted larva is thought to accelerate excystation of P. westermani metacercariae (Chung et al. 2005). A pioneering investigation of hydrolases in the cyst wall of metacercariae of the western liver fluke F. hepatica identified cysteine activity (Thorsell et al. 1965); these earlier findings along with the findings presented here implicate cathepsin L or other papain-like cysteine proteases in the escape of the immature adult fluke from the metacercarial cyst. Cathepsin-L-like cysteine protease activities have been well characterized in Schistosoma mansoni and Schistosoma japonicum where they participate in the parasite gut in digestion of ingested blood (e.g., Bogitsh et al. 2001; Delcroix et al. 2006). Schistosome cathepsin L activity has been characterized in the penetration glands of the parasite eggs and miracidia, where it likely functions in penetration of the host snail (Sung and Dresden 1986; Yoshino et al. 1993) and in the preacetabular glands of the cercariae (Dalton et al. 1997).

In conclusion, this report provides the first biochemical characterization of protease activities of O. viverrini and its developmental stages. More specifically, the present findings demonstrated that O. viverrini has a cathepsin-L-like cysteine protease(s), and its elevated developmental expression in metacercariae suggested that hydrolase might participate in larval excystation during mammalian infection. Given the remarkable link between infection with O. viverrini and induction of cholangiocarcinoma (see Parkin 2006; Sripa et al. 2007), an enhanced understanding of the enzymes and other proteins secreted by O. viverrini can be expected to enhance our understanding of the pathogenesis of liver-fluke-induced liver cancer.