| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Veterinary and Biomedical Sciences1, Center for Biotechnology2 and Department of Biochemistry3, University of Nebraska-Lincoln, Lincoln, NE 68583-0905, USA
Correspondence Gerald E. Duhamel gduhamel1{at}unl.edu, Gautam Sarath gsarath1{at}unl.edu
Received September 19, 2003
Accepted December 3, 2003
A membrane-associated, subtilisin-like, serine protease activity was found in both pathogenic and non-pathogenic strains of Brachyspira species in a previous study, but the biochemical properties of the enzyme were not investigated. The purpose of the present study was to characterize further the biochemical properties, including substrate specificity, of the membrane-associated protease of Brachyspira pilosicoli isolated from humans with intestinal disorders. Protease activity of detergent-enriched membrane protein extracts of B. pilosicoli was assessed using fluorescent dye-labelled synthetic peptides as substrates and determination of electrophoretic mobility of cleavage products in agarose gels. Each activity was further confirmed with class-specific protease inhibitors and thermal denaturation. The presence of a hydrophilic membrane-associated thermolabile serine endopeptidase with specificity for Leu was confirmed. Two additional hydrophilic membrane-associated thermostable proteolytic activities were identified, one with a putative Ala specificity, and one a carboxypeptidase. Taken together, these data suggest that, in addition to a previously described membrane-associated subtilisin-like serine protease, the membrane of B. pilosicoli contains proteins with at least two other proteolytic activities.
Abbreviation: ODG, n-octyl ß-D-glucopyranoside.
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Proteases have been found in spirochaetes and a role in damage to host cells and the mucosal barrier has been proposed (Ellen et al., 2000; Grenier et al., 1990). Among spirochaetes, trypsin- and chymotrypsin-like proteases of Treponema denticola, an oral spirochaete associated with periodontitis, have been characterized (Ellen et al., 2000; Grenier et al., 1990). A role for T. denticola proteases in periodontal tissue invasion by degradation of host cell proteins and inactivation of bioactive peptides has been suggested (Grenier et al., 1990; Reijntjens et al., 1986). The purpose of the present study was to characterize the biochemical properties of the membrane-associated protease activity present in human strains of B. pilosicoli. The substrate specificities of the proteases were identified using synthetic peptides representative of all predicted target amino acid residues designed on the basis of cleavage patterns obtained in a previous study (Muniappa & Duhamel, 1997).
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Membrane protein extraction.
Membrane-associated proteins of spirochaetes were extracted using the non-ionic detergents n-octyl ß-D-glucopyranoside (ODG; Sigma), as described by Comstock et al. (1993) with modifications, and 1 % (v/v) Triton X-114 (Sigma), as described by Bordier (1981) with modifications (Brusca & Radolf, 1994). Spirochaete cells were harvested by centrifugation at 800 g for 5 min, washed in Tris buffer pH 7.2 (50 mM Tris-base, 27 mM KCl, 145 mM NaCl; Sigma) and approximately 1010 spirochaetes in 968 µl Tris buffer were incubated with 32 µl of 25 % (w/v) ODG stock solution (8 mg final concentration) at room temperature for 15 min. The supernatant containing ODG-enriched membrane proteins was separated from protoplasmic cylinders by centrifugation at 10 000 g for 5 min at 4 °C. Both Triton X-114-aqueous and detergent phases were washed in Tris buffer and examined for the presence of protease activity.
Peptide synthesis.
Synthetic peptides were assembled manually using fmoc-amino acid-pentafluorphenyl esters as described previously (Sigal & Wylie, 1996). All chemicals for peptide synthesis were from a commercial supplier (Applied Biosystems), unless indicated. Briefly, approximately 10 µmol equivalent of peptides were synthesized as peptide amides. Amino acids were added (8-fold excess) as their respective fmoc-aminoacyl-pentafluorphenyl esters dissolved in 0.1 M 1-hydroxybenzotriazole in dimethylformamide. After coupling of the final amino acid, the free peptide N terminus was coupled for 12 h to 5-(and-6)-carboxyfluorescein using 5-(and-6)-carboxyfluorescein succinimidyl ester (4-fold excess, Molecular Probes), dissolved in dimethylformamide containing 0.1 M diisopropylethylamine. The resin was washed extensively with dimethylformamide, followed by dichloromethane, dried and cleaved using trifluoroacetic acid/water/triisopropylsilane/thioanisole (94 : 2 : 2 : 2, by vol.). Crude peptide was precipitated into cold methyl t-butyl ether, collected by centrifugation at 500 g for 10 min, and lyophilized twice from water. Peptides were used directly or after purification by reversed phase chromatography.
Protease assay.
The proteolytic activities of detergent-enriched membrane protein extracts of spirochaetes were assessed using fluorescent dye-labelled synthetic peptides as substrates and the electrophoretic mobility of cleavage products with altered size and charge in agarose gels. Assay mixtures containing 1 µl synthetic peptide (
1 mg ml-1), 8 µl 0.25 M Tris buffer, pH 7.2 (Sigma) and 21 µl of either detergent-enriched membrane proteins, 0.2 µg sequencing grade trypsin (Promega) or 2 µg alkaline protease (Promega) were incubated for 3 h at 37 °C. A 10 µl aliquot of the reaction mixture was mixed with an equal volume of glycerol (Sigma) and subjected to electrophoresis in 1 % (v/v) agarose gels (FMC Bio Products) prepared with Tris-boric acid buffer, pH 8.5 at 7 V cm-1 for 45–60 min. Peptide cleavage products were visualized and photographed under UV illumination.
The optimal pH for protease activity was determined by incubating ODG-enriched membrane proteins of strain SP16 with peptide N (Table 1) dissolved in Tris buffer or HEPES buffer adjusted to either pH 5.0, 6.0, 6.2, 6.4, 6.6, 6.8, 7.0, 7.2, 7.4, 7.6, 7.8, 8.0 or 9.0 for 180 min at 37 °C. In subsequent experiments, the optimal time required for complete digestion of peptide N dissolved in Tris buffer, pH 7.2 at 37 °C from 15, 30, 45, 60, 90, 120, 150 and 180 min was examined. Because the maximum cleavage of peptide was observed when samples were incubated for 180 min, this incubation time was used in all subsequent experiments.
|
Inhibition of B. pilosicoli membrane proteases.
To better define the class of protease present in detergent-enriched membrane protein extracts of strain SP16, the chelating agent EDTA (Sigma), at final concentrations of 1, 5 and 25 mM, the cysteine protease inhibitor trans-epoxysuccinyl-L-leucylamido-(4-guanidino)butane (Sigma) at final concentrations of 1, 10 and 100 µM and the serine protease inhibitor 3,4-dichloroisocoumarin (Sigma) at final concentrations of 0.01, 0.1, 0.5 and 1.0 mM (Powers & Harper, 1986) were added in a volume of 3 µl to 21 µl ODG-enriched membrane proteins and 5 µl Tris buffer, pH 7.2. After incubation at 37 °C for 1 h, enzymic activities were determined by adding 1 µl peptide N followed by a further incubation for 3 h at 37 °C and gel electrophoresis.
Stability of B. pilosicoli membrane proteases.
The thermal stability of ODG-enriched membrane proteases of strain SP16 was investigated by incubating extracts at either 37, 50, 75 or 95 °C for 10 min, prior to assay for protease activity. Similarly, the stability of the membrane proteases was determined at regular intervals after storage of ODG-enriched membrane proteins for a period of up to 12 days at 4 °C or 5 months at -80 °C. The stability of ODG-enriched membrane proteases in 0.01, 0.1 and 1.0 % (v/v) of the non-ionic detergents Triton X-100 and Triton X-114 was also examined.
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
To confirm the amino acid specificity of the membrane-associated protease further, four additional dye-labelled tripeptides were synthesized (Table 1). The presence of the non-polar aromatic amino acid Phe next to the Leu residue did not affect the enzymic cleavage of peptide A (Fig. 1). Similarly, enzymic cleavage of peptide B at the Leu residue was not affected by the presence of a negatively charged Asp residue (Fig. 1). Peptide B also confirmed the absence of a trypsin-like activity with specificity for Lys in the detergent-enriched membrane proteins of B. pilosicoli. Peptide C was designed to verify Ala specificity (Table 1). Cleavage at the Leu–Ala bond would release the dye-Leu residue, whereas cleavage at the Ala–Lys bond would release dye-Leu-Ala. Since both products have the same net charge of -2 (fluorescein dye has a charge of -1), although with slightly altered molecular masses, these potential products could not be resolved. Therefore, peptide D was designed to demonstrate the presence of a potential Ala-specific protease activity in the membrane protein extract. Cleavage at either the Ala–Ala or the Ala–Lys bond would suggest Ala-specific protease activity. Cleavage of peptide D by the membrane protein extract confirmed that Ala is a cleavage site (Fig. 1); however, the cleavage products could also have been attributable to carboxypeptidase-type activity. To differentiate between these possibilities, a larger dye-labelled peptide, designated peptide N, that contained multiple potential cleavage sites was synthesized (Table 1). On the basis of predicted cleavage sites and corresponding net charges, it was anticipated that peptide N would discriminate among the proteolytic activities when used in conjunction with class-specific protease inhibitors. To enhance electrophoretic separation of cleavage products, the locations of Leu and Ala residues were designed to produce a net charge of -2.
|
Incubation of peptide N with membrane protein extracts of human and the porcine reference B. pilosicoli strains revealed four cleavage products (Fig. 2a). Based on our earlier observations with the dye-labelled tripeptides A–D, it was concluded that the band exhibiting maximal mobility, designated N4, was the N-terminal dye-labelled Ala. Band N3 was consistent with peptide products of Leu cleavage at position 7 (Leu7), and possibly some peptide cleaved at Ala5, since both of these products had calculated net charges of -2 and should exhibit similar mobilities in our assay system (Table 1). Additionally, the two products at N3 were likely to serve as substrates for the enzyme that released the N-terminal dye-Ala product at N4. On the basis of gel mobility, the peptide products N1 and N2 appeared to have larger molecular masses than N3 with a more negative charge than the parent peptide N (Fig. 2a). The presence of these products may be attributable to deamidation of the C-terminal Arg11 and/or removal of residues close to the C terminus via the action of a carboxypeptidase-type activity.
|
Incubation of peptide N with the alkaline protease control generated two bands that corresponded to the four expected cleavage products, one product following cleavage at Ile4, migrating alone, and products following cleavage at Trp6, Leu7 and Val8 migrating closely together (data not shown). Furthermore, incubation of peptide N with trypsin generated the expected dye-Ala-Lys product with a net charge of -1 (Fig. 2b, lane 3), confirming that the B. pilosicoli membrane proteins do not contain trypsin-like activity (Muniappa & Duhamel, 1997). Taken together, these observations suggested that the B. pilosicoli membrane contains three types of proteolytic activities, a protease with specificity for Leu and possibly Ala, another with a specificity for Ala, and a carboxypeptidase-type enzyme. In order to characterize each activity further, additional assays with class-specific protease inhibitors and thermal denaturation experiments were conducted.
The membrane-associated proteases of strain SP16 were not inhibited by the chelating agent EDTA. This observation is consistent with a previous report (Muniappa & Duhamel, 1997) indicating that the membrane-associated proteases of Brachyspira species are not metalloenzymes and do not require divalent cations as co-factors for activation. Furthermore, a lack of inhibition by the cysteine-specific inhibitor, trans-epoxysuccinyl-L-leucylamido-(4-guanidino)butane, at any concentration, suggested the absence of cysteine proteases in the membrane of strain SP16. Absence of the N3 products after incubation of the membrane extract with increasing concentrations of the serine-specific inhibitor 3,4-dichloroisocoumarin confirmed the presence of a serine endopeptidase activity with specificity for Leu (Table 2 and Fig. 3a). The lack of inhibition by this compound of N1, N2 and N4 cleavage product generation provided further evidence that the N3 product is predominantly the Leu7 fragment and not the Ala5, as suggested earlier. Furthermore, inhibition of the serine endopeptidase caused apparent accumulation of the other cleavage products N1, N2 and N4, suggesting that the N1 and N2 fragments are substrates for the Leu-endopeptidase, and thus are larger than the Leu7 fragment seen following cleavage at N3. The results confirmed the presence of a serine endopeptidase with Leu specificity and further suggested that additional protease activities were present in the membrane proteins of B. pilosicoli. Because these other enzyme activities were not inhibited by the class-specific inhibitors examined in the present study, the catalytic classes of these proteases remain unknown.
|
|
In our preliminary experiments, samples incubated in HEPES buffer produced low yields of peptide N cleavage products, whereas cleavage of the same peptide was complete when detergent-enriched membrane extracts were incubated in Tris buffer at either pH 6.8 or 7.2, but not at any other pH examined. Because of the qualitative nature of band intensity estimation, it was not possible to correlate substrate preference and optimal pH; all four products N1, N2, N3 and N4 were equally intense. However, these observations further suggested that more than one protease activity, with different pH optima, were present in the membrane proteins of B. pilosicoli.
Absence of the N3 products after incubation of peptide N with detergent-enriched membrane proteins of strain SP16 at 75 and 95 °C for 10 min (Fig. 3b) confirmed previous observations (Muniappa & Duhamel, 1997), indicating thermal instability of the serine endopeptidase. Conversely, enhanced intensity of N1, N2 and N4 products at higher incubation temperatures suggested thermal stability of the Ala-specific (N4) protease and carboxypeptidase activities. These observations further supported the conclusion that two or more proteases are present in the membrane proteins of B. pilosicoli.
The ODG-enriched membrane proteases of strain SP16 were stable after storage for 12 days at 4 °C and 5 months at -80 °C. Proteolytic cleavage of peptide N by the membrane proteins of strain SP16 was also not affected by the presence of Triton X-100 or Triton X-114 at any of the concentrations examined, indicating that the proteases are stable in non-ionic detergents. Because integral membrane proteins are anchored to the hydrophobic core of the lipid bilayer by hydrophobic domains or by amphiphilic groups covalently linked to the protein, it is possible to separate them by phase partitioning in the non-ionic detergent Triton X-114 into hydrophobic and amphiphilic detergent phase and hydrophilic aqueous phase (Bordier, 1981; Brusca & Radolf, 1994). Protease activity was limited to the aqueous phase of the Triton X-114 extract, suggesting that the proteases of B. pilosicoli are hydrophilic or low-hydrophobic membrane-associated proteins rather than integral membrane proteins.
Identification of a serine endopeptidase with specificity for Leu, and at least two other protease activities in the membrane proteins of B. pilosicoli provides a basis for examining the diversity of proteolytic activities among intestinal spirochaetes in general and differential expression in relation to virulence of pathogenic intestinal spirochaetes in particular. Further work to define the genetic basis of proteolytic activity and role in virulence of intestinal spirochaetes is needed.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. P. Dassanayake, G. Sarath, and G. E. Duhamel Penicillin-Binding Proteins in the Pathogenic Intestinal Spirochete Brachyspira pilosicoli Antimicrob. Agents Chemother., April 1, 2005; 49(4): 1561 - 1563. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
