INTRODUCTION
Invasive larvae (cercariae) of schistosomatid trematodes infect their hosts by actively penetrating the skin. Trichobilharzia regenti is a bird (family Anatidae) schistosome with a unique route of migration; its larvae enter peripheral nerves and the spinal cord to reach the brain and ultimately the nasal cavity (Hrádková and Horák, Reference Hrádková and Horák2002) evoking severe pathology (Horák et al. Reference Horák, Dvořák, Kolářová and Trefil1999; Kolářová et al. Reference Kolářová, Horák and Čada2001). Although ducks are fully permissive to T. regenti, cercariae of this species are also able to invade mammals (including man), and can survive for a limited period of time causing severe dermatological and neurological pathologies related to host immune status (Kouřilová et al. Reference Kouřilová, Hogg, Kolářová and Mountford2004a, Reference Kouřilová, Syrůček and Kolářováb; Horák and Kolářová, Reference Horák and Kolářová2005). Repeated invasion can stimulate an allergic reaction in man, which is manifested as cercarial dermatitis (Horák and Kolářová, Reference Horák and Kolářová2001; Horák et al. Reference Horák, Kolářová and Adema2002). This disease is becoming an emerging public health problem in Europe (e.g. Bayssade-Dufour et al. Reference Bayssade-Dufour, Vuong, Rene, Martin-Loehr and Martins2002).
The mechanism of skin penetration is partially understood in some schistosome species (e.g. Schistosoma mansoni and S. japonicum) and it is agreed that it involves the release of specific proteolytic enzymes (peptidases) (Dalton and Brindley, Reference Dalton, Brindley, Fried and Graczyk1997; McKerrow and Salter, Reference McKerrow and Salter2002; Whitfield et al. Reference Whitfield, Bartlett, Marc, Brown and Marriott2003; Ruppel et al. Reference Ruppel, Chlichlia and Bahgat2004; He et al. Reference He, Salafsky and Ramaswamy2005). These peptidases are present in 2 groups of large penetration glands (post-acetabular and circumacetabular) filling almost two thirds of the cercarial body. After close attachment of the cercarial to host skin, the contents of these glands are released and the enzymes facilitate disruption of surface proteins and underlying tissues (for reviews see Horák et al. Reference Horák, Kolářová and Adema2002; McKerrow et al. Reference McKerrow, Caffrey, Kelly, Loke and Sajid2006).
Several cercarial serine peptidases have been described in S. mansoni, the best characterized of which is the 28 kDa or 30 kDa cercarial elastase (SmCE) (Landsperger et al. Reference Landsperger, Stirewalt and Dresden1982; McKerrow et al. Reference McKerrow, Pino-Heiss, Linquist and Werb1985; Marikovsky et al. Reference Marikovsky, Arnon and Fishelson1988, Reference Marikovsky, Arnon and Fishelson1990; Chavez-Olortegui et al. Reference Chavez-Olortegui, Resende and Tavares1992; Salter et al. Reference Salter, Choe, Albrecht, Franklin, Lim, Craik and McKerrow2002). This chymotrypsin-like peptidase is localized in the circumacetabular glands and can cleave human skin elastin (McKerrow et al. Reference McKerrow, Pino-Heiss, Linquist and Werb1985; Salter et al. Reference Salter, Lim, Hansell, Hsieh and McKerrow2000). However, the identity and function of cercarial cysteine peptidases is not well known but the presence of cathepsins L1 and B1 in S. mansoni has been reported (Dalton et al. Reference Dalton, Clough, Jones and Brindley1996, Reference Dalton, Clough, Jones and Brindley1997; Brady et al. Reference Brady, Brindley, Dowd and Dalton2000; Skelly and Shoemaker, Reference Skelly and Shoemaker2001) these enzymes may be involved in the disruption of the outer keratinized layer of skin (Dalton et al. Reference Dalton, Clough, Jones and Brindley1996, Reference Dalton, Clough, Jones and Brindley1997). In bird schistosomes, 6 isoforms of cathepsin B1 (TrCB1) were reported from schistosomula of T. regenti (Dvořák et al. Reference Dvořák, Delcroix, Andrea Rossi, Vopálenský, Pospíšek, Šedinová, Mikeš, Sajid, Sali, McKerrow, Horák and Caffrey2005). In this stage, these peptidases are localized in the gut and probably serve as digestive enzymes. Cathepsin B1 has also been recently identified in a cDNA library from sporocysts/cercariae of T. regenti, and its sequence is 100% identical to the schistosomular isoform TrCB1.1 (Dolečková et al. Reference Dolečková, Kašný, Mikeš, Mutapi, Stack, Mountford and Horák2007).
In the present study, we show that the predominant peptidase activity in extracts of T. regenti cercariae is of the cysteine class, and has a similar substrate specificity, pH optimum and molecular size to a cysteine peptidase found in extracts of S. mansoni cercariae. The peptidases of both parasites are capable of degrading native keratin and collagens (types II, IV), and we show that they are cathepsin B-like enzymes.
MATERIALS AND METHODS
Parasites and reagents
Cercariae of T. regenti (laboratory strain) from intermediate host snails (Radix peregra s. lat.) were collected, washed twice in ice-cold 0·1 m phosphate-buffered saline (PBS) pH 7, and concentrated by centrifugation prior to storage at −80°C. Live S. mansoni cercariae were obtained from 3 infected Biomphalaria glabrata snails donated by Dr Libuše Kolářová (Institute for Postgraduate Medical Education, Prague, CZ). Lyophilized S. mansoni cercariae were kindly provided by Professor Michael Doenhoff (School of Biological Sciences, University of Wales, UK). Eight-day-old schistosomula of T. regenti from duckling spinal cords were obtained as described by Dvořák et al. (Reference Dvořák, Delcroix, Andrea Rossi, Vopálenský, Pospíšek, Šedinová, Mikeš, Sajid, Sali, McKerrow, Horák and Caffrey2005). Extracts of cercariae (CE) and schistosomula (SE) were prepared by 2 cycles of sonication (7W, 30 s each, Vibracell-72405 100-W ultrasonicator, Bioblock Scientific, France) in 0·1 m PBS followed by centrifugation for 20 min at 16 000 g (4°C). Protein concentration in supernatants was measured using the Bicinchoninic Acid Protein Assay (BCA-1, Sigma-Aldrich).
The DCG-04 probe for cysteine peptidase detection was a gift from Dr C. Caffrey and Dr D. Greenbaum, Sandler Center for Basic Research in Parasitic Diseases, UCSF, USA. All chemicals used were purchased from Sigma-Aldrich unless otherwise noted.
Fractionation of parasite extracts by chromatography
Anion-exchange chromatography was performed with the Bio-Logic system using a Mono Q column (Bio-Rad). Soluble T. regenti CE was first filtered using 0·22 μm Ultrafree-MC Sterile filtration device (Millipore) and then loaded (1·5 mg) onto a column equilibrated with 20 mm Bis-Tris-HCl buffer (Bis[2-hydroxyethyl]amino-tris[hydroxymethyl]methane) pH 6·8. Elution was performed using a linear gradient of ionic strength (0–1 m NaCl in the same buffer) at the flow rate 1 ml/min. Fractions (300 μl) containing peptidase activity were further fractionated by gel permeation chromatography (GPC) using a Superdex 200 column (Amersham Pharmacia Biotech, USA).
T. regenti and S. mansoni CE (4 mg in 1 ml of 0·1 m PBS pH 7) were fractionated using a Superdex 75 column (Amersham Pharmacia Biotech, USA). Eluted proteins were collected in 80 fractions (300 μl/fraction) at a flow rate of 0·5 ml/min.
Fluorometric assays of enzyme activity
Peptidase activity was measured using fluorogenic (aminomethylcoumarin – AMC) peptide substrates (Bachem, Switzerland) each designed to assay specific peptidase activites: Z-Phe-Arg-AMC (FR; to assay cathepsins B and L), Z-Arg-Arg-AMC (RR; cathepsin B), Boc-Val-Leu-Lys-AMC (VLK; cathepsins B and L), Z-Arg-AMC (R; cathepsin H, aminopeptidase B, C), Z-Gly-Pro-Arg-AMC (GPR; trypsin-like), Z-Pro-Arg-AMC (PR; thrombin-like), Suc-Ala-Ala-Pro-Phe-AMC (AAPF; chymotrypsin-like), Boc-Leu-Gly-Arg-AMC (LGR; trypsin-like), Boc-Val-Pro-Arg-AMC (VPR; thrombin, trypsin) and Boc-Ala-Gly-Pro-Arg-AMC (AGPR; trypsin-like). Assays were carried out in 96-well black plates (Nunc, Denmark). Released free AMC was measured at excitation and emission wavelengths of 340–360 and 440–460 nm respectively in a Bio-Tek, Synergy HT fluorometer (Bio-Tek, USA) or Spex Fluoromax 3 (Jobin Yvon Horiba, France) continually for 120 min at 37°C or RT. Activity against the fluorogenic substrates was screened in a broad pH range – in 0·1 m citrate-phosphate buffer pH 3–8 (for cysteine peptidase detection containing 5 mm DTT, or 5 mm L-cysteine) and in 0·1 m glycine buffer, pH 8·5–10·5. Monitoring of peptidase activity was started by addition of 200 μl/well of appropriate buffer with 50 μm peptide-AMC substrate to 2 μg of protein (CE), or 1 μg of protein (particular fraction).
The specificity of peptidase activity was investigated using a spectrum of differentiating inhibitors added to incubation buffers. Inhibitors of cysteine peptidases: 10 μm E-64 [N-trans-(epoxysuccinyl) -L-leucine 4-guanidinobutylamide], irreversible broad spectrum; 10 μm CA-074 [N-(L-3-trans-propylcarbamoyloxirane-2-carbonyl) -Ile-Pro-OH], irreversible selective inhibitor of cathepsin B (Towatari et al. Reference Towatari, Nikawa, Murata, Yokoo, Tamai, Hanada and Katunuma1991); 10 μm Z-Phe-Ala-CHN2, irreversible inhibitor of cathepsins B and L (Dalton and Brindley, Reference Dalton, Brindley, Fried and Graczyk1997); 10 μm calpain II inhibitor [Ac-Leu-Leu-Met-aldehyde], irreversible selective inhibitor of cathepsins L and B (Donkor, Reference Donkor2000). Inhibitors of serine peptidases: 10 μm PMSF [phenylmethylsulfonyl fluoride], irreversible broad spectrum; elastatinal [N-(Na-Carbonyl-Cpd-Gln-Ala-al) -Leu], an irreversible specific inhibitor of neutrophil and pancreatic elastase but not other serine peptidases like trypsin or chymotrypsin (Bieth, Reference Bieth, Barrett, Rawlings and Woessner2004); 1·5 μm aprotinin from bovine lung, reversible broad spectrum.
Peptidase activity was also examined in the penetration gland secretory products (GSP) of live T. regenti and S. mansoni cercariae. Stimulation of penetration gland secretion was performed by praziquantel (Mikeš et al. Reference Mikeš, Zídková, Kašný, Dvořák and Horák2005). Suspensions of live cercariae were placed onto microscope slides and incubated for 30 min with praziquantel (0·1 μg/ml in water; from 10 000× stock solution in pure ethanol). Slides were then incubated with either 100 μm FR-AMC alone (30 min), or with 10 μm E-64 (15 min), followed by a mixture of 100 μm FR-AMC and 10 μm E-64 (30 min). The activity of released GSP was monitored by fluorescence microscopy (Olympus BX51).
Electrophoresis and zymography
The CE (5–20 μg of protein) and chromatographic fractions of T. regenti and S. mansoni were separated by SDS-PAGE (MiniProtean-3 apparatus, Bio-Rad) in 10% and 12% gels, or in 4–20% gradient gels. Peptidase activities were assayed by zymography in gels co-polymerized with 0·1% gelatin. Samples were mixed with a non-reducing sample buffer, or reducing buffer (10 mm DTT), and allowed to stand at room temperature for 10 min prior to loading. Following electrophoresis, zymographic gels were washed 2×10 min in either 0·1 m citrate-phosphate buffer, pH 3·5 to 8, or glycine buffer pH 9 and 10 (both with/without 10 μm E-64, or 1·5 μm aprotinin) containing 2·5% Triton X-100 and then 1×10 min in an appropriate Triton-free buffer. Overnight incubation was carried out in the same buffers (in the case of citrate-phosphate buffer, 10 mm L-cysteine was added). All gels were stained with Coomassie Brilliant Blue R-250, or using a Silver Stain Kit (Bio-Rad).
Hydrolysis of macromolecular substrates
Individual fractions (0·75 μg of protein in 5 μl) were mixed with collagen type II from bovine nasal septum, type IV from human placenta, or keratin from human epidermis (all 10 μg in 10 μl of 0·1 m PBS pH 7). Collagenase from Clostridium histolyticum (0·5 μg in 5 μl) was used as a positive control. The mixtures were incubated for 6 h at 37°C and then separated by SDS-PAGE in 12% polyacrylamide gels for detection of digestion products.
Ligand blotting with DCG-04
DCG-04 is a biotinylated analogue of the irreversible Clan CA cysteine peptidase inhibitor E-64 which covalently binds to the active site of cysteine peptidases (Greenbaum et al. Reference Greenbaum, Medzihradszky, Burlingame and Bogyo2000). T. regenti and S. mansoni CE, T. regenti fraction 7′ that has the highest cysteine peptidase activity, and T. regenti SE (2 μg of total protein each) were incubated for 1 h with 5 μm DCG-04 in 0·1 mm citrate-phosphate buffer, pH 6·0, containing 5 mm DTT. Controls were pre-incubated with 100 μm cysteine peptidase inhibitors E-64, or CA-074. After SDS-PAGE, proteins were transblotted onto nitrocellulose membrane (1 h, 1·5 mA/cm2) and blocked for 1 h in 5% non-fat milk (Bio-Rad Blotting Grade Blocker) in 20 mm Tris-buffered 0·15 m saline, pH 7·8 (TBS; Tris[hydroxymethyl]aminomethane) containing 0·05% Tween-20 (TBS-T). The membranes were washed 3×5 min in TBS-T and then incubated with streptavidin-HRP (2 μg/ml in 1% non-fat milk in TBS-T) for 30 min and washed again 3×5 min in TBS-T (Mikeš and Man, Reference Mikeš and Man2003). The membrane was developed using the Opti-4CN™ Substrate Kit (Bio-Rad).
RESULTS
Fluorometric enzyme assays with T. regenti and S. mansoni cercarial extracts, their fractions and gland secretion products
Soluble CE were screened for peptidase activities, and particularly for the presence of cysteine peptidases (e.g. cathepsins B and L). The optima for both T. regenti and S. mansoni cysteine peptidase activities in the presence of FR substrate were between pH 4·5 and 5 (Fig. 1A). The level of activity was 3·4 times lower for T. regenti compared to S. mansoni. Peptidolytic activities of CEs with other substrates were minor at this pH (Fig. 2; pH 4·5).
The pH profile of cysteine peptidase-like and serine peptidase-like activities in CE of Trichobilharzia regenti and Schistosoma mansoni. Cysteine peptidase-like activity was assayed using the fluorogenic peptide substrate FR (A). Serine peptidase-like activity was assayed using GPR substrate (B). Activity was measured in 0·1 m citrate-phosphate buffer, pH 3–8, and in 0·1 m glycine buffer, pH 8·5–10·5, for 120 min at 37°C.
Peptidase activity of CE from Trichobilharzia regenti (Tr) and Schistosoma mansoni (Sm). Assays performed using fluorogenic substrates in 0·1 m citrate-phosphate, buffer pH 4·5, in the presence of 5 mm L-cysteine (A), or 0·1 m glycine buffer, pH 10 (B). The activity on each substrate is expressed as nmoles AMC released/min/mg protein. Activity was measured for 120 min at 37°C.: FR, Z-Phe-Arg-AMC; R, Z-Arg-AMC; RR, Z-Arg-Arg-AMC; PR, Z-Pro-Arg-AMC; VPR, Boc-Val-Pro-Arg-AMC; VLK, Boc-Val-Leu-Lys-AMC; GPR, Z-Gly-Pro-Arg-AMC; AGPR, Boc-Ala-Gly-Pro-Arg-AMC; AAPF, Suc-Ala-Ala-Pro-Phe-AMC; LGR, Boc-Leu-Gly-Arg-AMC.
Trypsin-like serine peptidase activity was also noted in S. mansoni CE (rank order GPR>VPR>LGR>AGPR) (Fig. 2; pH 10), but only negligible activity was detected for T. regenti CE at the strongly alkaline values of the pH optimum 10–10·5 (Fig. 2; pH 10 and Fig. 1B). Chymotrypsin-like activity was demonstrated by slight cleavage of AAPF substrate in S. mansoni CE only (Fig. 2; pH 10). In both species, cysteine peptidase activities predominated over serine peptidase activities.
Both T. regenti and S. mansoni CE were fractionated by GPC on the Superdex 70 column into 80 fractions, yielding 0·15–0·19 mg/ml protein per 300 μl fraction (Fig. 3). The position of T. regenti fractions with the greatest cysteine peptidase activities (Ct1, 2, 3; peak of activity in fraction Ct2; Fig. 3A) corresponded to the positions of active fractions of S. mansoni CE elution profiles (Cs1, 2, 3; peak of activity in fraction Cs2; Fig. 3B). The fractions Ct2 and Cs2 were able to cleave 4 fluorogenic peptide substrates (rank order FR>>VLK>RR>R) at pH 4·5. The activity against FR substrate in T. regenti Ct2 fraction increased 2·5 times compared to CE. This implies a successful partial purification of a cysteine peptidase.
Elution profile from gel permeation chromatography (Superdex 75 column) of Trichobilharzia regenti (A) and Schistosoma mansoni (B) CE. Numbers above the peaks mark the elution time. The bars under the peaks indicate the greatest cysteine peptidolytic activity of fractions against FR (T. regenti Ct1, 2, 3 and S. mansoni Cs1, 2, 3). The most active fractions Ct2 and Cs2 are in bold.
Fractions of T. regenti CE obtained after anion-exchange FPLC were also screened for cysteine peptidase activities. Fraction 18 showed the highest preference for the cathepsin B and L substrate FR (Fig. 4A). This fraction was further fractionated by GPC (Superdex 200 column). Fractions 6′ and 7′ from GPC expressed activity with the FR substrate (Fig. 4B). In later experiments only fraction 7′ was used because of its much higher peptidase activity.
Elution profile from ion-exchange FPLC (A) and subsequent gel permeation chromatography (B) of Trichobilharzia regenti CE. SDS-PAGE (12% gel) of fractions with the highest cysteine peptidase activity (f18, f7′). Shaded arrows show the fractions 18 (A) and 7′ (B). Dashed line, panel (A), conductivity. Dotted line, panel (A), theoretical gradient of salt. Dotted line, panel (B), conductivity. Lanes M, molecular weight marker. Lane f18, fraction 18 eluted in anion-exchange FPLC. Lane f7′, fraction 7′ eluted in gel permeation chromatography. Lanes a, b and a′, b′, zymographic analysis of fractions 18 and f7′, respectively. Lanes b, b′, overnight incubation of the gelatin gels in 0·1 m citrate-phosphate buffer pH 4·5 without inhibitor. Lanes a, a′, overnight incubation in 0·1 m citrate-phosphate buffer, pH 4·5, with 10 μm E-64. Both gels were stained with Coomassie Brilliant Blue R.
Using FR substrate, peptidase activity was also detected in T. regenti and S. mansoni praziquantel-stimulated GSP (Fig. 5B). This was visible as a fluorescent cloud adjacent to penetration gland openings in front of the head organ. The activity was inhibited by 10 μm E-64 (Fig. 5D).
Cysteine peptidase activity in the penetration gland secretions of Trichobilharzia regenti cercariae. Penetration gland secretion was induced by adding praziquantel to a suspension of cercariae. Released excretory/secretory products were incubated with FR substrate (B) or with E-64 inhibitor followed by FR substrate (D). Released products in front of cercarial head organ, observed under bright field conditions (A, C black arrows). Degradation of peptidyl substrate by gland peptidases revealed by fluorescence microscopy (B, white arrow). Inhibition of this activity was recorded in control containing E-64 inhibitor (D, white arrow). Analogous results were obtained with S. mansoni cercariae under the same experimental conditions (not shown).
Inhibition assays
The peptidase activity of the chromatographic fractions and whole soluble CE was screened using a panel of differentiating peptidase inhibitors (Table 1). Z-Phe-Ala-CHN2 was the most potent inhibitor (>96% inhibition) for all samples (Ct2, Cs2, TrCE, and SmCE) as tested by the FR substrate at pH 4·5. Other cysteine peptidase-specific inhibitors were also highly effective and caused inhibition of at least 85%. On the other hand, the serine peptidase-specific inhibitors aprotinin and PMSF only inhibited peptidase activity by 20%, although elastatinal caused 70% inhibition.
The effect of inhibitors on peptidolytic activity of Trichobilharzia regenti and Schistosoma mansoni cercarial extracts and fractions using Z-Phe-Arg-AMC substrate
(Activity was measured in 0·1 m citrate-phosphate buffer pH 4·5, in the presence of 5 mm L-cysteine. Values are means of 3 independent triplicate assays with standard deviations (±sd).)
a, c Cercarial protein extracts of T. regenti and S. mansoni.
b, d T. regenti and S. mansoni chromatographic fractions with the highest activity against Z-Phe-Arg-AMC substrate.
Gelatinolytic activity of T. regenti and S. mansoni samples
Gelatin gels of T. regenti and S. mansoni CE and fractions (Ct1, 2, 3; Cs1, 2, 3) at pH 4·5 showed lysis in the regions 25–37 kDa (Fig. 6). This agrees with the result showing that fractions 18 and 7′ contained molecules with peptidase activity around 30 kDa (see Fig. 4). Cysteine peptidase activities of the different parasite preparations were significantly inhibited by incubating the gelatin gels overnight with 10 μm E-64 at pH 4·5 (Fig. 4 and Fig. 6).
Analysis of proteolytic activity in cercarial protein extracts and fractions of Trichobilharzia regenti and Schistosoma mansoni by zymography (12% polyacrylamide gel co-polymerized with 0·1% gelatin). Lanes A, CEs of T. regenti (10 μg of protein per lane). Lanes B, CEs of Schistosoma mansoni (10 μg of protein per lane). Ct1, 2, 3 and Cs1, 2, 3, fractions of T. regenti and S. mansoni with the highest cysteine-like peptidase activity (1 μg of protein per lane; see Fig. 3). Control, the gels after SDS-PAGE incubated overnight in 0·1 citrate-phosphate buffer (pH 4·5) and 0·1 m glycine buffer (pH 10; 37°C). Aprotinin and E64, gels incubated overnight in 0·1 m citrate-phosphate buffer (pH 4·5) and 0·1 m glycine buffer (pH 10; 37°C) with the respective inhibitor.
Conversely at pH 10, gelatinolytic activity of CE from both schistosome species was localized in the area ⩾45 kDa (Fig. 6). Aprotinin significantly inhibited the serine peptidase activity of CE in this area (Fig. 6).
Degradation of keratin and collagen substrates
The native substrates, keratin and collagen (types II, IV), were degraded by particular fractions Ct2, Cs2 of T. regenti and S. mansoni, after overnight incubation at neutral pH (Fig. 7). The pattern of hydrolysis products differed between the two species.
Degradation of keratin, collagen II and collagen IV by Trichobilharzia regenti Ct2 fraction (A) and Schistosoma mansoni Cs2 fraction (B). Samples were separated by SDS-PAGE in 12% polyacrylamide gel for detection of digestion products. Lane M, molecular weight marker. Lanes c1, c2, c3, control collagenase (0·5 μg in 5 μl) with keratin, collagen II, collagen IV. Lanes k1, k2, k3; keratin, collagen II, collagen IV alone (10 μg in 10 μl). Lanes 1, 2, 3; fractions Ct2 or Cs2 (0·5 μg of protein in 5 μl) incubated with keratin, collagen II, collagen IV. Collagenase and fractions were incubated with keratin, collagen II, collagen IV (10 μg in 10 μl) in 0·1 m PBS, pH 7, for 6 h at 37°C, prior to electrophoresis.
Cysteine peptidase active site-labelling by DCG-04
For T. regenti, incubation with the DCG-04 probe led to the detection of a prominent 33 kDa band in CE, as well as in chromatographic fraction 7′ (with the highest cysteine peptidase activity) (Fig. 8). A similar 33/34 kDa band doublet was recorded in T. regenti SE. In S. mansoni CE, the band migrated at aproximately 33–34 kDa. Controls without DCG-04 and with E-64 or CA-074 pre-incubated sample showed no reaction in these regions.
Binding of DCG-04, a cysteine peptidase-specific probe, to the peptidases in CE and SE extracts of Trichobilharzia regenti, chromatographic fraction 7′ of T. regenti and CE of Schistosoma mansoni. Lane M, molecular weight marker. Lanes 1–4, T. regenti CE, lanes 5–8, S. mansoni CE, lanes 9–11, T. regenti chromatographic fraction 7′, lanes 12–14, T. regenti SE. Lanes 1, 5, 9 and 12, positive reaction with DCG-04. Lanes 2, 6, 10 and 13, no binding of DCG-04, reaction blocked by cysteine peptidase inhibitor E-64. Lanes 3 and 7, no binding of DCG-04, reaction blocked by selective cathepsin B inhibitor CA-074. Lanes 4, 8, 11 and 14, controls of non-specific avidin-Px binding without DCG-04; 2 μg of protein per lane, 5 μm DCG-04. Arrows show the detected cathepsin B in T. regenti CE and fraction 7′ (33 kDa, lanes 1 and 9), in T. regenti SE (33/34 kDa, lane 12) and in S. mansoni CE (33–34 kDa, lane 5).
DISCUSSION
The study described here constitutes a comparative analysis of the peptidases released by T. regenti and S. mansoni focussing upon cercarial cysteine peptidases. S. mansoni was used in order to evaluate the level of activity in T. regenti. Although we would be able to collect appropriate amounts of penetration gland products of T. regenti cercariae (Mikeš et al. Reference Mikeš, Zídková, Kašný, Dvořák and Horák2005), we did not dispose of enough live S. mansoni cercariae. Therefore, cercarial protein extracts of both parasites were employed in this study. We were aware of the possible appearance of non-gland peptidases in the samples, however, this cannot be excluded even when working with cercarial penetration gland secretions (Knudsen et al. Reference Knudsen, Medzihradszky, Lim, Hansell and McKerrow2005; Mikeš et al. Reference Mikeš, Zídková, Kašný, Dvořák and Horák2005).
General screening for peptidase activity initially compared selected fluorogenic substrates known to reveal the presence of various trematode peptidases, especially cathepsins B and L and S. mansoni elastase. Whole cercarial extracts (CE) were tested for their activity against specific substrates and at different pH optima. The preference for cysteine peptidase substrate FR was similar in CE of both T. regenti and S. mansoni, although the level of peptidase activity at the optimum of pH 4·5 was much greater for the latter species. It is not clear whether this was due to subtle differences in affinity for FR substrate or because of lower cysteine peptidase content in the case of T. regenti CE.
Trematode cysteine peptidases generally exhibit their activity between pH 4 and 10 (Dalton and Brindley, Reference Dalton, Brindley, Fried and Graczyk1997; Caffrey et al. Reference Caffrey, Salter, Lucas, Khiem, Hsieh, Lim, Ruppel, McKerrow and Sajid2002; Sajid and McKerrow, Reference Sajid and McKerrow2002) and it is known that in vitro the pH optima of S. mansoni cysteine peptidases are shifted to acid pH (e.g. SmCB1 – pH 6·0; SmCB2 – pH 5·0–5·5; SmCL1 – pH 6·5; SmCL2 – pH 5·35; for reviews see Caffrey and McKerrow (Reference Caffrey, McKerrow, Barrett, Rawlings and Woessner2004) and Dalton et al. (Reference Dalton, McKerrow, Brindley, Barrett, Rawlings and Woessner2004)). Similarly, schistosomular cathepsin B1 of T. regenti expressed optimal activity against FR substrate at pH 4·5–5·5 (Dvořák et al. Reference Dvořák, Delcroix, Andrea Rossi, Vopálenský, Pospíšek, Šedinová, Mikeš, Sajid, Sali, McKerrow, Horák and Caffrey2005). Serine peptidase optima, on the other hand, generally occur at basic pH values (e.g. S. mansoni cercarial elastinolytic protease – pH 8–10; McKerrow et al. (Reference McKerrow, Pino-Heiss, Linquist and Werb1985); S. mansoni cercarial elastase – pH>9; Salter et al. (Reference Salter, Lim, Hansell, Hsieh and McKerrow2000); for review see Dalton and Brindley (Reference Dalton, Brindley, Fried and Graczyk1997)). Our findings fully correspond with these results, although it should be noted that we were unable to detect significant quantities of serine peptidases in CE of T. regenti.
Inhibition studies showed that the general cysteine peptidase inhibitor E-64 and the cathepsin B and L inhibitor Z-Phe-Ala-CHN2 had significant effects on peptidase activity against the FR substrate at pH 4·5. The cathepsin B-selective inhibitor CA-074 also resulted in a comparable level of inhibition. These results imply that the major peptidase activities in T. regenti CE are of cysteine peptidase origin – most likely cathepsin B and, to a certain degree, cathepsin L. The relatively high inhibitory effect of elastatinal on cysteine peptidase activity with FR substrate was unexpected and its action is questionable. Elastatinal is usually regarded as a specific inhibitor of pancreatic and neutrophil elastases (serine peptidases). Considering its structure [N-(Na-Carbonyl-Cpd-Gln-Ala-al) -Leu] and the fact that Z-Phe-Ala-diazomethylketone was the best inhibitor, it is likely that the aldehyde on Ala2 of elastatinal inhibits the cysteine peptidase activity when situated in P1 position. The data presented here with FR substrate and the inhibitors corroborate previously reported results on cathepsins B or L. These enzymes are ubiquitous in somatic extracts or excretory/secretory products of trematodes including S. mansoni larvae and adults (Dalton et al. Reference Dalton, Clough, Jones and Brindley1996; Caffrey et al. Reference Caffrey, Rheinberg, Mone, Jourdane, Li and Ruppel1997; Dalton et al. Reference Dalton, Clough, Jones and Brindley1997; Brady et al. Reference Brady, Brindley, Dowd and Dalton2000; Sajid et al. Reference Sajid, McKerrow, Hansell, Mathieu, Lucas, Hsieh, Greenbaum, Bogyo, Salter, Lim, Franklin, Kim and Caffrey2003) or T. regenti schistosomula (Dvořák et al. Reference Dvořák, Delcroix, Andrea Rossi, Vopálenský, Pospíšek, Šedinová, Mikeš, Sajid, Sali, McKerrow, Horák and Caffrey2005).
Screening for cysteine peptidase activity in fractionated CE (fractions Ct2, Cs2 and 7′) of T. regenti revealed results consistent with those obtained with S. mansoni. The 2 protein bands (17 and 28 kDa) detected in T. regenti cysteine peptidase active fraction 7′ were characterized by mass spectrometry methods (MALDI-TOF MS). Their tryptic peptides were de novo sequenced (LC MS/MS, ion trap). However, none of the obtained peptide sequences (from the 17 and 28 kDa bands) aligned with known peptidase sequences in databases (not shown). This implies that the cysteine peptidase activity was produced by a minute amount of a highly active enzyme (not detectable in the polyacrylamide gel). This is consistent with the proteomic surveys of S. mansoni cercariae performed by Knudsen et al. (Reference Knudsen, Medzihradszky, Lim, Hansell and McKerrow2005) and Curwen et al. (Reference Curwen, Ashton, Sundaralingam and Wilson2006), who did not find cysteine peptidases among penetration gland-secreted proteins. Nevertheless, cysteine peptidase activity was clearly demonstrated in cercarial GSP of both species in our experiments.
The cleavage of the GPR substrate was negligible in the case of T. regenti compared to S. mansoni. Hydrolysis of this substrate is usually related to serine peptidase activity (Zimmerman et al. Reference Zimmerman, Ashe, Yurewicz and Patel1977; Dalton et al. Reference Dalton, Clough, Jones and Brindley1997; Bahgat and Ruppel, Reference Bahgat and Ruppel2002) specifically at alkaline pH values 8–10 (Dalton and Brindley, Reference Dalton, Brindley, Fried and Graczyk1997). However, it is known that trypsin-like peptidases preferentially cleave this substrate compared to chymotrypsin-like peptidases including cercarial elastase from S. mansoni (Salter et al. Reference Salter, Lim, Hansell, Hsieh and McKerrow2000, Reference Salter, Choe, Albrecht, Franklin, Lim, Craik and McKerrow2002). Therefore, our measurements may result from contamination of S. mansoni CE by trypsin-like peptidases of snail origin (Salter et al. Reference Salter, Lim, Hansell, Hsieh and McKerrow2000). Nevertheless, the presence of peptidases active against the AAPF substrate confirmed the presence of S. mansoni cercarial elastase in CE. The virtual absence of a serine peptidase activity in T. regenti CE indicates that an elastase orthologue is not present in T. regenti. This reinforces the failure to reveal an elastase-like enzyme in this species using anti-S. mansoni elastase antibodies (Mikeš et al. Reference Mikeš, Zídková, Kašný, Dvořák and Horák2005) or molecular techniques (Dolečková et al. Reference Dolečková, Kašný, Mikeš, Mutapi, Stack, Mountford and Horák2007). At this point, the T. regenti penetration mechanism is reminiscent of that in the human schistosome S. japonicum, where no cercarial elastase was reliably identified yet (e.g. Fan et al. Reference Fan, Minchella, Day, McManus, Tiu and Brindley1998; Chlichlia et al. Reference Chlichlia, Schauwienold, Kirsten, Doenhoff, Fishelson and Ruppel2005).
In zymographs of fraction 7′, the ∼30 kDa lytic band is possibly the same as that detected at pH 4·5 in T. regenti CEs and in the Ct2 fraction. This lytic band apparently did not match the sequenced 28 kDa band in fractions 7′ and 18. The ∼30 kDa band within fraction 7′ most likely corresponds to the ∼33 kDa band detected after incubation of T. regenti CE, fraction 7′ and schistosomular extract with DCG-04. The difference of ∼3 kDa in size could be caused by changed mobility of the protein caused either by the covalently bound probe in case of ligand blotting, or by the gelatin content in the case of zymographic gels.
Although many other researchers have reported analogous gelatinolytic patterns caused by peptidase activities in S. mansoni CE (McKerrow et al. Reference McKerrow, Pino-Heiss, Linquist and Werb1985; Marikovsky et al. Reference Marikovsky, Arnon and Fishelson1988; Chavez-Olortegui et al. Reference Chavez-Olortegui, Resende and Tavares1992; Dalton et al. Reference Dalton, Clough, Jones and Brindley1997; Bahgat et al. Reference Bahgat, Doenhoff, Kirschfink and Ruppel2002), only one group has analysed the proteolytic activity of T. regenti enzymes (Dvořák et al. Reference Dvořák, Delcroix, Andrea Rossi, Vopálenský, Pospíšek, Šedinová, Mikeš, Sajid, Sali, McKerrow, Horák and Caffrey2005; Mikeš et al. Reference Mikeš, Zídková, Kašný, Dvořák and Horák2005). There is also limited information on the role of cysteine peptidases in cercarial penetration, although it is speculated that they may aid the entry of schistosome cercariae through the outer keratinized layer of the skin (Dalton et al. Reference Dalton, Clough, Jones and Brindley1997). Our results illustrating degradation of keratin and collagen (type II and IV) by cercarial cysteine peptidases support this theory. Indeed, the positive reaction of DCG-04 with T. regenti and S. mansoni proteins at ∼33 kDa and 33–34 kDa, respectively, proved the presence of cysteine peptidases. The inhibition of these reactions by CA-074 confirmed that they are cathepsins B. In schistosomula of T. regenti, the 33/34 band doublet corresponding to the ∼33 kDa band of cercariae identified by DCG-04, has previously been determined as cathepsin B1 (TrCB1 – Dvořák et al. Reference Dvořák, Delcroix, Andrea Rossi, Vopálenský, Pospíšek, Šedinová, Mikeš, Sajid, Sali, McKerrow, Horák and Caffrey2005). The occurence of TrCB1 in cercariae of this species is also supported by the identification of its sequence in the cDNA library of cercarial germ balls (Dolečková et al. Reference Dolečková, Kašný, Mikeš, Mutapi, Stack, Mountford and Horák2007). Thus, we have demonstrated peptidases of cathepsin B-type in the penetration glands of cercariae of 2 schistosome species. The ability of these cysteine peptidases to hydrolyse skin proteins supports their role in host invasion.
This work was supported by the grants of the Czech Science Foundation no. 524/04/P082 and 206/06/0777, Grant Agency of the Charles University no. 263/2004/B/Bio/PrF, the Wellcome Trust Collaborative Research Initiative Grant no. 072255/Z/03/Z and grants of the Ministry of Education of the Czech Republic MSM 0021620828 and MSM LC06009. M. K. was a holder of a Mobility Fund travel grant of the Charles University in Prague, Czech Literary Fund travel grant and a travel grant of the Foundation „Nadání Josefa, Marie a Zdeňky Hlávkových“. This research has been facilitated by the Institute for the Biotechnology of Infectious Diseases (IBID), University of Technology Sydney (John Dalton is a recipient of the NSW BioFirst Award in Biotechnology) and access to the Australian Proteome Analysis Facility established under the Australian Government's Major National Research Facilities Programme. We wish to thank Professor M. Doenhoff, University of Wales, Bangor, UK and Dr L. Kolá
ová (Institute for Postgraduate Medical Education, Prague, CZ) for provision of S. mansoni cercariae. We are also grateful to Dr C. Caffrey and Dr D. C. Greenbaum, Tropical Disease Research Unit, UCSF, USA, for supply of the DCG-04 cysteine peptidase probe and to Petr Man, MSc., Institute of Organic Chemistry and Biochemistry, Academy of Sciences of the Czech Republic, Prague, CZ, for mass spectrometry analyses.
