xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001681133
007 cr mnu|||uuuuu
008 060109s2004 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0000612
Detection, cloning, and analysis of a u32 collagenase in streptococcus mutans gs-5
h [electronic resource] /
by Marios Ioannides.
[Tampa, Fla.] :
b University of South Florida,
Thesis (MSci)--University of South Florida, 2004.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 57 pages.
ABSTRACT: Streptococcus mutans is a recognized principal etiologic agent in coronal caries. Although S. mutans has the ability to bind collagen and degrade FALGPA, a synthetic peptide mimicking collagen substrate, its role in dental root caries has not yet been fully elucidated. Degradation of collagen fibrils in dentin was attributed to S. mutans, but a collagenase enzyme has not yet been isolated from this organism. Considering the increased incidence of dental root decay among the elderly, an understanding of the role of the pathogenic factors is necessary to the development of preventive measures. The present study has focused on the cloning and analysis of S. mutans collagenase enzyme. Toward this goal, a putative collagenase gene was identified in S. mutans UA159 by genomic analysis and a primer set was designed and used to amplify the corresponding gene in S. mutans GS-5 used as a model organism.The PCR product was cloned into the vector pCR 2.1 TOPO-TA, and the gene sequenced and analyzed. Alignment of the S. mutans GS-5 and UA159 putative collagenase genes showed 99% homology. The gene was next cloned in frame into the inducible expression vector pET100/D TOPO. Induction and expression of recombinant protein in E. coli were confirmed by SDS-PAGE and Western immunoblotting, while biochemical analysis indicated that it was a calcium- dependent metalloproteinase. Enzyme analysis of the recombinant enzyme showed both gelatinolytic and collagenolytic activity. Further analysis of the GS-5 gene using databases such as ExPASy, Pfam, and SMART indicated that it was highly homologous to the U32 peptidase family, which includes the PrtC collagenase of Porphyromonas gingivalis, a bacterium causing periodontitis. The present study was the first to unequivocally demonstrate the existence of a collagenase gene in S.
Adviser: Dao, My Lien.
Dental root decay.
t USF Electronic Theses and Dissertations.
Detection, Cloning, and Anal ysis of a U32 Collagenase in Streptococcus mutans GS-5 by Marios Ioannides A thesis submitted in partial fulfillment of the requirement s for the degree of Master of Science Department of Biology College of Arts and Sciences University of South Florida Major Professor: My Lien Dao, Ph.D. Daniel Lim, Ph.D. Andrew Cannons, Ph.D. Date of Approval: July 2, 2004 Keywords: gelatinase, metalloproteinase, collagen, U32 peptidase, dental root decay Copyright 2004, Marios Ioannides
AKNOWLEDGMENTS I would like to express my gratitude to my advisor, Dr. My Lien Dao, for her encouragement, enthusia stic spirit, and guidanc e during my graduate studies, especially for helping me discove r my strength and deve lop my research skills. I would also like to thank the me mbers of my committ ee, Dr. Daniel Lim and Dr. Andrew Cannons, for their va luable advice and words of encouragement. I would like to thank my colleagues, Thomas Han, Ross Myers, Valerie Carson, and Justin Federico for their help, positive interaction, and especially for making the laboratory a friendly, ener getic, and comfortable environment for productive research. I would like to thank my grandparents and especially my parents for their love, confidence in me, and for their eno rmous emotional and fi nancial sacrifice in the last seven years in order to send me abroad and offer me a better education. My gratitude is endless and I am blessed to have them in my life. Last but not least, I would like to thank my girlfriend, Juliana Saavedra, for her friendship, encouragement, and under standing during the last four years of my studies. She has been a source of insp iration to me, and with her everything seems to be so much easier and within reach.
i TABLE OF CONTENTS LIST OF FIGURES iii LIST OF TABLES iv ABSTRACT v INTRODUCTION 1 MATERIALS AND METHODS 12 Chemicals and Reagents 12 Bacterial Strains and Growth Conditions 12 Plasmid Isolation 13 Isolation of Genomic DNA from S. mutans GS-5 13 PCR Amplification of Genomic DNA 14 Cloning of the 1.2 Kb PCR Product into pCR 2.1 -TOPO TA Vector 15 Subcloning of the smcol Gene Into the Expression Vector pET100/DTOPO 16 Expression of smcol -pET 17 SDS-PAGE 17 Purification of Recombinant S. mutans Collagenase 18 Western Blot Analysis 19 Sample Preparation for Collagenase Assay 20 Collagenase Assay 20 Gelatin Hydrolysis 21 Collagenase Inhibition Assay 22 Protein Concentration 22 DNA Sequencing and Sequence Analysis 23 RESULTS 24 PCR Amplification of S. mutans Collagenase Gene and Cloning into pCR 2.1 TOPO-TA Vector 24 Sequence Analysis of smcol -TA 26 Cloning of the smcol -pET Gene 27 Sequence Analysis of smcol -pET 29 SDS PAGE of the IPTG Induced smcol -pET 33 SDS and Western Blot Analysis of Purified Recombinant Protein 34 Gelatin Hydrolysis 35 Collagen Assay 36 Inhibition Assay 38
ii DISCUSSION 41 REFERENCES 45
iii LIST OF FIGURES Figure 1 PCR amplification of the putative collagenase gene of S. mutans GS-5 25 Figure 2 Insertion of PCR product into pCR 2.1-TOPO 26 Figure 3 Alignment of the smcol sequence 27 Figure 4 PCR product obtained from genomic DNA 29 Figure 5 Map of pET 100/D-TOPO with collagenase insert 31 Figure 6 Confirmation of successful insertion of the collagenase gene into pET 100/D-TOPO. 32 Figure 7 SDS analysis of the induced E. coli BL 21 33 Figure 8 SDS-PAGE of purified recombinant protein 35 Figure 9 Western immunoblots of the E. coli cell lysates containing the His-Collagenas e fusion protein and the purified collagenase 36 Figure 10 Gelatin zymogram of recombinant E. coli cells and purified recombinant protein 38 Figure 11 Degradation of fluorescein-c onjugated Type I collagen 39
iv LIST OF TABLES Table 1 PCR primer pairs used for confirmation of successful ligation of smcol in pET100 32 Table 2 Inhibition assays for collagenase activity 40
v Detection, Cloning, and Anal ysis of a U32 Collagenase in Streptococcus mutans GS-5 Marios Ioannides ABSTRACT Streptococcus mutans is a recognized principal et iologic agent in coronal caries. Although S. mutans has the ability to bind collagen and degrade FALGPA, a synthetic peptide mi micking collagen substrate, it s role in dental root caries has not yet been fully elucidated. Degradation of collagen fibrils in dentin was attributed to S. mutans but a collagenase enzym e has not yet been isolated from this organism. Consi dering the increased incidenc e of dental root decay among the elderly, an understanding of the role of the pathogenic factors is necessary to the development of prevent ive measures. The present study has focused on the clon ing and analysis of S. mutans collagenase enzyme. Toward this goal, a putative colla genase gene was identified in S. mutans UA159 by genomic analysis and a pr imer set was designed and used to amplify the corresponding gene in S. mutans GS-5 used as a model organism. The PCR product was cloned into the vector pCR 2.1 TOPO-TA, and the gene sequenced and analyzed. Alignment of the S. mutans GS-5 and UA159 putative collagenase genes showed 99% homology. The gene was next cloned in frame into the inducible expression vector pET1 00/D TOPO. Induction and expression of recombinant protein in E. coli were confirmed by SDS-PAGE and Western
vi immunoblotting, while biochem ical analysis indicated that it was a calciumdependent metalloproteinase. Enzyme anal ysis of the recombinant enzyme showed both gelatinolytic and collagenolytic activity. Further analysis of the GS5 gene using databases such as ExPASy, Pfam, and SMART indicated that it was highly homologous to the U32 peptidase family, which includes the PrtC collagenase of Porphyromonas gingivalis, a bacterium causing periodontitis The present study was the first to unequivoca lly demonstrate the existence of a collagenase gene in S. mutans and to identify it as a member of the U32 peptidase family. The obtaining of the S. mutans collagenase gene should help in further investigation of the role of this enzyme in dental root decay and its potential use as a dental root caries vaccine.
1 INTRODUCTION Dental root decay is an in fectious disease, with hi gh prevalence in older individuals worldwide (19). It has been s hown that nearly two thirds of tested elderly people experienced root caries, with 23% having untreated root caries (48). Nowadays, with the high technologic al advances, the exce llent health care, and dental health awar eness, there is a need for the populations to retain their teeth much longer than ever before ( 38, 46). Thus, an adequat e understanding of the pathology of oral diseases is esse ntial in order to devise preventive measures and to insure the qualit y of life in aging populations. Dental root caries is initiat ed on root surfaces that have been exposed to the oral environment due to gum recession (19). The etiology of the gingival recession varies from aging as the leading factor, to trauma. It has been shown that gingival recession is a fr equent phenomenon in people aged 30 years and older with a 2.8fold incr ease in root caries in ol d individuals (8,37). Other factors associated with gum recession inclu de anatomical factors or deficiencies, such as loss of alveolar bone, t ooth shape, and abnormal tooth position. Furthermore, trauma resulting from vi gorous toothbrushing and operating procedures could often be related to gum recession (37). The now exposed root is susceptible to microbial attack (42), whic h, if not taken care by professionals, can result in the infection of root canal and ultimately in the potential loss of the tooth (48). The factors invo lving the etiology and process of dental root decay,
2 even though similar to coronal caries, s eems to be more complicated because of higher susceptibility of the exposed dentin the main structural component of tooth root (14). Dentin, wh ich lacks the highly mineral ized enamel present in the crown, is composed of an inorganic ma terial (hydroxyapatite) (14) and an organic matrix 90% of which is type I collagen and 10% non-collagenous proteins (14,33). Furthermore, the susceptib le root caries involve two additional factors: the presence of bacteria and fermentable carbohydrates (8). In accordance to a number of st udies, the microorgani sms isolated from root caries lesion s include predominately Streptococcus spp., Actinomyces spp., Lactobacillus spp. and yeast (14, 42). While bot h root and coronal caries involve demineralization of the dentin and enamel respectively, the rate of this process differs substantially. In root caries loss of dentin minerals occurs twice as rapidly as compared to the enamel demineralizat ion at higher pH, since the crown enamel has almost twice as much mineral s (8). As mentioned before, due to gingival recession, the dentin is exposed to the oral microbial environment. Dung et al. (14), in their review about the molecular pathogenesis of root caries, proposed a possible mechanism of the dis ease process. Production of acid, after fermentation of certai n carbohydrates by oral bacteria, such as sucrose, results in the demineralizati on of the exposed dentin. Th is is followed by another oral bacterial invasion involving prot eolytic degradation of dentinÂ’s collagen, mainly by collagenases. Furthermore according to the same author, collagenolytic activity of acid-deminera lized dentin is enhanced by the addition of trypsin. It was also s uggested that acid pretreatm ent may cause a degradation
3 of collagen by trypsin and other non-specific proteases. Another in vitro investigation (29) involving a pH cycling model, which closely mimics a demineralization/reminerilaztion process in the oral environment, suggested that the organic matrix is attacked during re mineralization and pr edominately after demineralization of the dentin. Studies by Dung et al. (15) and van Strijp et al. ( 47), showed that proteases, such as collagenases or gelatinases present in saliva, tissue, and dentin may be activat ed either by the presence of bacteria or pH fluctuation in the oral environment. These host derived proteases may also play an important role in root dentin collagen degradation. Even though all studies are in agreement t hat demineralization of d entin and degradation of its organic matrix is a pre requi site for the pathogenesis of root decay, the exact mechanism of this multifunctional di sease is yet to be investigated. Collagen is the major fibrous component of animal extracellular connective tissues: skin, tendon, blood vessels, bone, and dentin. To date at least 19 types of collagen have been characterized (21). D entin collagen -type Iconsists of two identical 1 and a different 2 chain (49), with the repeating sequence Gly-X-Y present in all three, with X and Y often being pr oline and hydroxyproline respectively. These 100kDa fibrous prot eins are laterally aggregated and twist around a common axis to form the triple helic al structure of colla gen. Stability of this structure is conferred to the Â“steric repulsion of the proline and hydroxyproline residuesÂ” (43) as well as the covalent cross-linkages, mineral coating, and interaction of collagen with non-collagen pr oteins (14). Disruption of interand intramolecular linkages of co llagen molecule results in the dispersal
4 of the polypeptide chains and in the production of gelatin (21). In this state, the strands are vulnerable to cleavage by a wide variety of nonspecific proteases. Collagen degradation is a result of a group of enzymes called collagenases. These enzymes di gest the peptide bonds of fibrous collagen in the triple helix region, under physiologic al conditions and pH (21). Mammalian collagenases produced by tissue resident cells, macrophages, or neutrophils, aid in tissue regeneration and rem odeling as well as in immune responses against invasion of host tissue pathogens (12, 40). As they are responsible for degradation of extracellular matrix, they belong to the family of matrix metalloproteinases (MMP). Activity and st ability of these proteases depend on the presence of Zn2+ and Ca2+ (9, 12). Moreover, invading microorganisms may also have the ability to produce collagenases. Unlike mammalian collagenases that attack their substrate at specific regions producing two distinct fragments (21), bacterial collagenases possess broad substrate specificities and can degrade both native and denatured collagens. Over the last few years many microorganisms have been reported to prod uce enzymes that degrade collagen. Among them are some that are associated with human di seases, thus identifying the presence of the enzyme as a possible virulence factor. Much of the knowledge of bacterial collagenases has come from studies involving the enzyme from Clostridium histolyticum. This Gram-positive, spore forming, anaerobic organism has been associated with progression of myonecrosis (54). It has been show n to produce up to seven different collagenases/gelatinases, ranging from 68kD a to 130kDa (34). Investigations
5 though, have been concentrated mainly on the 116kDa co llagenase (ColH). Initially, cloning and expression of the colH gene revealed that the 116kDa collagenase successfully degrad ed Pz-peptide (54), a synt hetic peptide analog to collagen. Interestingly, zymography of the partially purified collagenase revealed the presence of two gelatinases (116kDa 98kDa). Sequence analysis of the two indicated that the 98kDa gelatinase is a result of the C-terminal cleavage of ColH, finding which suggests that the C te rminal of the enzyme is responsible for binding or accessing the native collagen (54). In acco rdance of many research groups (27, 54), ColH was found to be a zinc metalloproteinase containing the consensus sequence HEXXH that corresponds to the zinc-binding site and the catalytic site (27). Mutations of the codo ns at the catalytic site (histidine residues) resulted in the abolishment of zinc binding as well as catalytic activity of the enzyme, result that c onfirmed that zinc is require d for enzyme activity (27). Matsushita et al. (35), in an effort to char acterize a 78kDa gelatinase in C. histolyticum, showed that there is another gene, colG, which also confers collagenolytic activity. The 116kDa enzyme, initially produced as an inactive proenzyme, was isolated from the culture supernatant and able to hydrolyze the Pzpeptide, insoluble collagen, as well as gelatin (34, 35). Results from hybridization experiments and sequence analyses of the two genes suggested that colH and colG are a single copy genes; thus, the presence of the seven different collagenases/gelatinases are due the proteolytic cleavage of the Cterminus of either one the genes resulting in t he multiplicity of collagenases (35).
6 Clostridium perfringens is another tissue invasive pathogen, able to degrade collagen (2, 36). The 120kDa enzyme (ColA) was able to degrade the Pz-peptide, as well as type I fibrils. A nalysis of the deduced aminoacid sequence of the colA gene revealed great similari ties with the products of colH and colG genes found in C. histolyticum having a greater degree of resemblance with the colG (35,36,54). A Pz-peptide sequence as we ll as the zinc-binding motif were also identified, with the former pr obably being a signal sequence for self processing of the inactive pro-enzyme ( 36). Notable is the fact that all three collagenases ColA, ColH, and ColG, showed to have high homology with the Vibrio alginolyticus collagenase (34,35,36, 54). Alignment of the aminoacid sequences indicated the presence of the H EXXH motif. Also, sim ilarities of the proposed segmental enzyme structures sugges ted that the three organisms have a common ancestral collagenase gene, which was then diverged and duplicated in each descendant (34, 35). The PrtC collagenase, produced by the periodontal pathogen Porphyromonas gingivalis is another well studied proteolytic enzyme. Unlike the previously mentioned homologues, th is enzyme does not possess the zincbinding motif. Instead, it is a member of U32-peptidas e family, with a conserved catalytic domain yet to be characterized. Kato et al (28) successfully cloned and characterized the prtC gene. SDS-PAGE showed a 35kDa enzyme, but in the active form the enzyme acts as a dime r as shown by gel filtration assays. According to the same research group, t he enzyme has no significant similarities with its clostridial counterparts. It neit her hydrolyzed gelatin nor the Pz-peptide,
7 while it degraded azocoll, heat denatured type I collagen, as well as fibrillar reconstituted type I collagen (28). The presence of Zn2+ and Fe3+ inhibited the enzymeÂ’s activity. The metal chelator ED TA also inhibited collagen degradation, but there was no effect on the activi ty when the enzyme was incubated with PMSF (28). On the other hand, Ca2+ stimulated the activity of the purified protein. Furthermore, Houle et al. (23) used three different assays using fluoresceinlabeled Type I collagen and two cysteine proteinases mutants (Argand Lysgingipapains) as substrates, in order to test for the collagenolytic activity of P. gingivalis Their results demonstrated that the abi lity of the organism to hydrolyze native collagen resides mainly in the presence of Arggingipapain. One possible reason for the contradictory results of Kato and Houle is the heterogeneity of the prtC gene. Indeed, PCR and sequence analysis performed on six clinical isolates by Wittstock et al. (51) showed that none of the sequences agreed with that of the prtC gene reported by Kato, and each one had distinct differences as compared to each other. Further clarific ation regarding the function and role of prtC gene is necessary. Vibrio parahemolyticus is a human pathogen that produces collagenase (PrtV). Cloning and expressi on of the gene showed activi ty against type I, II, III, and IV (55, 56) collagen. T he 62 kDa protein is highly hydrophobic and active at temperature as high as 40o C. EDTA and 1,10-phenanthroli ne inhibit the activity of the enzyme (55). This is also a Zn-metalloproteinase, but double immunodiffusion with antibodies against PrtV showed no similarities between the other two zinc-metalloproteinases of Vibrio alginolyticus and ColH (56).
8 Collagenolytic activity was also det ected in Group B streptococci. This group of organisms has been associated with neonatal diseas es and premature rupture of the amniochorionic memb rane, probably due to the presence of collagenases (25). Jackson et al. showed that the cell-associated enzyme was able to degrade the synthetic peptide analogue to colla gen, FALGPA, as well as to migrate through the placental tissue in vitro Inhibition assays also revealed that EDTA and 1,10-p henanthroline inhibited the enzyme activity suggesting that GBS collagenase also needs a metal co-factor for its activity (25). Additionally, cross reactivity with anti-clostridial ant ibodies as well as the presence of gelatinolytic activity of the enzyme further suppor ted the notion that GBS collagenase is closely related to ColH (25). Streptococcus mutans has been established by many research groups as an oral pathogen, especially in relation to dental caries (5, 29). In particular, S. mutans serotype c strains are most frequent ly isolated from the human oral cavity (29). This acidouric organism has the ability to produce lactic acid while attached tightly on tooth surfaces (29, 53). A number of pr oteins and enzymes including glucosyltranferases, glucan binding proteins, and wall associated proteins, have been extensivel y studied and their role in the colonization of the tooth and the induct ion of oral diseases has been established. Directed mutagenesis by Qian and Dao (39) demonstr ated that the wallassociated protein A (WapA) is important in sucros e depended aggregation an d binding of the organism to smooth surfaces. Another im portant factor that promotes the accumulation of S. mutans on the tooth surface is t he production of glucan from
9 sucrose via glucosyltranferases (GTFs) (2 9, 45). There are three different types of GTFs based on their ability to produc e water-soluble or water insoluble glucans (29, 45). GTFs, besides producin g glucans, also have the ability to bind these glucose polymers (45) and consequently promote co lonization of the tooth and built up of dental pl aque. In the same line, S. mutans produces three nonenzymatic glucan binding proteins, G BPA, GBPB, and GBPC (29). Once the organism establishes itself on the toot h surface, the toot h enamel and the exposed dentin in the root are deminera lized by acid production. This gives the organism even more opportunities to coloni ze and to further invade the tooth, through binding and degradation of collagen. It is notewo rthy that immunization with a single antigen only provides a decrease in caries score, while inactivation of a single gene in S. mutans can only cause a decrease in virulence. These results indicate that dental caries in volves multiple virulence factors. Furthermore, S. mutans is considered to be one of the major pathogenic microorganisms isolated from root caries (8, 14, 29), but its role in the pathogenesis of the disease is not yet clear. Studies on the microbiology of root caries by Shen et al. (42) demonstrated that S. mutans was isolated in high numbers, implying the importance of th is pathogen to the disease. Another research group suggested that the high prevalence of S. mutans from patients who previously received periodontal tr eatment may be of importance in the development of root caries in these patients (5). Seve ral reports in the literature (5, 32) have proved that the cell surfac e antigen I/II is necessary for the initial adherence of the organism on collag en type I and dentin tubules. Isogenic
10 mutants showed reduction in binding of cells, suggesting that antigen I/II may be important in the process of dental or root caries. In favor of this, two S. mutans isolates were shown to have significantly high specific affinity ratio to type I collagen (33). This was in suppor t with the finding of Switalski et al (44) who demonstrated that S. mutans isolated from adult patient s had the ability to bind collagen type I independently of the environm ental conditions and changes in pH. In addition to collagen-binding properties S. mutans also has the ability to invade host tissues (31). Yet, S. mutans collagenolytic activity is still in the early stages of investigation. Initially, human isol ates of the bacterium have been shown to cause extensive loss of bone and the br eakdown of periodontal ligament in gnotobiotic rats (21). Harrington et al (22) also demonstrat ed that the organism possesses collagenolytic activity by descr ibing two extracellular proteases (50 and 52kDa) that are capable of hydrol yzing PZ-PLGPA and azocoll. Later, Jackson et al (24) discovered that S. mutans also expressed a cell-associated collagenolytic enzyme that degraded the synthetic co llagen peptide FALGPA, and that the organism could mi grate through placental tissue in vitro In the same work it was demonstrated t hat the collagenolytic activity of S. mutans was inhibited by EDTA, and 1, 10-phenanthroline. Gelatin zymogram analysis showed the ability of the organism to hydrolyze gelatin (22,24), while immunoblot analysis of S. mutans proteins indicated that it cros s reacted with polyclonal rabbit antiC. histolyticum collagenase antiserum. Previously in Dr. DaoÂ’s laborator y, a 4020bp contig sequence was obtained from the Okl ahoma Genome Center S. mutans sequencing project
11 database. BLAST search showed sequence similarities with other putative collagenases. Submission of the contig sequence to the O pen Reading Frame Finder (OPR Finder) at the NCBI, rev ealed an open reading frame of 1287 bp that encoded for 428 amino acid protein. PCR amplification, using a set of primers that flanked the OR F of the gene, produced a 1. 5 kb fragment but failed to show expression and activity of t he enzyme when cloned into pCR2.1-TOPO. Additionally, immunoscreening of S. mutans GS-5 library with a rabbit anti C. histolyticum collagenase gave false pos itive results due to the cross reaction of antibodies with epitopes pres ent on unrelated proteins (Seijo, Biology M.S. Thesis, USF, Tampa, FL). In the present study, the ORF of S. mutans GS-5 collagenase ( smcol ) was amplified by PCR and the gene was first cloned into pCR2.1-TOPO for sequence analysis. Primers were then designed to amplify smcol from the S. mutans genomic DNA directly in order to clone the gene in-frame into the inducible pET100/D-TOPO expression ve ctor. Expression of collagenase and gelatinase activity in the recombinant E. coli was demonstrated.
12 MATERIALS AND METHODS Chemicals and Reagents Restriction endonucleases and P CR reagents were purchased from Promega Inc. (Madison, WI) and used accord ing to the manufacturerÂ’s protocol. Cloning vectors pCR2.1-TOPO TA and pET 100/D TOPO were from Invitrogen (Carlsbad, CA). PCR primers were obt ained from Qiagen (Valencia, CA). All other chemicals and reagents were purc hased from Sigma Al drich Chemical Company (St. Louis, MO) or Fisher Sci entific (Pittsburg, PA) unless otherwise specified. Bacterial Strains and Growth Conditions S. mutans GS-5 serotype c, originally obtained from J. J. Ferretti (University of Oklahoma Health Scienc es Center, Oklahoma City, OK), was cultured in brain heart infu sion broth (BHI obtained fr om Difco, Detroit, MI) at 37oC with 5% CO2. Competent E. coli Top 10 was obtained from Invitrogen and used for the cloning of recombinant plasmids. Competent E. coli BL21 Star (DE3) (Invitrogen) was used for the cloning and expression of the recombinant S. mutans collagenase ( smcol ) gene. LB medium contai ning Ampicillin added at 100 g per ml (LB-amp) was used for the selection of transformants and culture of recombinant E. coli Plates were prepared by adding agar at 1.5%.
13 Plasmid Isolation Plasmid isolation was performed by th e alkaline lysis procedure using the QIAprep MiniprepTM kit and protocol from Qiagen Inc (Valencia, CA). Briefly, 1ml of an overnight culture was centri fuged at 10,000 X g and the pellet was resuspended in 250l of 50mM Tris.Cl (p H 8) buffer containing 10 mM EDTA and 0.1mg/ml of RNase A. Following a s equential addition of 250l of 200mM NaOH containing 1% SDS solution and 350l of 3M Potassium Acetate, the sample was mixed gently and centrifuged for 10 minut es. The supernatant was carefully applied to the QIAprep column (Qiagen) and centrifuged briefly. The column was washed with 0.5ml of G uanidine and isopropanol solution followed by 0.75ml of 80% ethanol. Finally, the pl asmid DNA was eluted by the addition of 50l of 10mM Tris.Cl (pH 8.5), and centrifuged for 1 minute at maximum speed in a microcentrifuge. The plasmid DNA obtain ed was measured in a SmartSpec Plus spectrophotometer (Bio-Rad, Richmond, CA) in order to determine concentration and purity (OD260/280 > 1.7), and analyzed by electro phoresis in a 1% agarose gel containing ethidium bromide. Isolation of Genomic DNA from S. mutans GS-5 Genomic DNA isolation was performed using the WizardTM Genomic Isolation Kit obtained from Promega and according to the manufacturerÂ’s protocol. Briefly, a 5 ml overnight culture of S. mutans GS-5 was aliquoted in 1.5 ml centrifuge tubes and centrifuged at 13,000 X g for 2 minutes. The supernatant was discarded and the pellet was resuspended in 480l of EDTA
14 and 100 l of 10mg/ml lysozyme. The sample was centrifuged for 2 minutes at 13,000 X g. The pellet was resuspended in 600l of in Nuclei Lysis Solution and incubated for 60 minutes at 80oC. Three l of RNase solution were added and the sample was mixed and in cubated for 15 minute at 37oC. Next, the sample was treated with 200l of Pr otein Precipitat ion Solution and vortex-mixed at maximum speed in a microcentrifuge for 20 seconds. After incubation on ice for 5 minutes and subsequent centrifugation for 3 minutes the supernatant was transferred to a new microcentrifuge tube containing 600l of room temperature isopropanol. The sample was then mixe d and centrifuged for 2 minutes at 13,000 X g. The pellet was washed with 70% ethanol and centri fuged again for 2 minutes. Finally, the pellet was air-dried and resuspended in 100l of autoclavesterilized distilled water. Concentrati on of DNA was determined by measurement in the SmartSpec Plus spectrophot ometer (Bio Rad) and the OD 260/280 of 1.7 or higher was indicative of purity. PCR Amplification of Genomic DNA Direct PCR amplification was performed on S. mutans genomic DNA for subsequent cloning and expression of the smcol gene. For TOPO TA cloning, 2X PCR Master Mix (Promega) was used. Taq polymerase and dNTPs were used at a final concentration of 1.25 U and 200M respectively. The 1.2kb fragment ( smcol ) was amplified using 0.2M of the primers F1: ATGACAAAACAATTAAAACG and R1: T AAGTTCTAACAGTAAGGC under the following amplification profile : Initial denaturation at 95oC for 1 minute, 30 cycles
15 each consisting of a denaturation step at 95 oC for 1 minute, an annealing step at 50 oC for 1 minute, and extension step at 72 oC for 1 minute and a final extension at 72 oC for 7 minutes. The PCR product fl anked by dA overhangs was cloned into the TOPO TA cloning vector and used to transform E. coli Top 10 as described below. The plasmid containing the smcol gene inserted into the TOPO TA vector was isolated from the recombinant E. coli and designated smcol -TA. The PCR reaction mixture fo r amplification of the smcol gene for cloning into the pET100/DTOPO vector ( smcol -pET), as described below, consisted of the following: 2.5 U of Pfu Turbo Cx hotstart DNA polymerase (Stratagene, La Jolla, CA), 25ng of DNA temp late, 0.2M of each of the following primers : F2:CACCATGACAAAACAATTAAAACG and R1:TTAAGTTCTAACAGTAAGGC 200M dNTPs, and 1X reaction buffer (Strat agene). An initial denaturation step was performed at 95oC for 2 minutes. Next, the sa mple was subjected to 30 cycles of the following : Denaturation at 95oC for 30 seconds-Annealing at 50oC for 30 seconds-Extension at 72oC for 30 seconds. After the last cycle the products were extended for another 10 minutes at 72oC. The final PCR products were analyzed by electrophoresis in 1% agarose gel, and DNA was stained with ethidium bromide for observ ation under UV light. Cloning of the 1.2 Kb PCR Product into pCR 2.1 -TOPO TA Vector The smcol fragment obtained by PCR amplif ication was cloned into the pCR2.1-TOPO vector. Briefly, 3l of fr esh PCR products were mixed with 1l of TOPO vector and 2l of ster ile distilled water. 2l of the above cloning reaction
16 were used to transform 250l of E. coli Top 10 cells, using the heat shock method. The cells were mixed with the insert, incubated on ice for 30 minutes and then heat-shocked for 30 seconds at 42oC. Then, 250l of room temperature of SOC medium (Invitr ogen) was added to the t ube and incubated at 37oC with shaking for an additional 1 hour. Finally, 25l and 50l of the transformed cells were spread on pre-warmed LB-amp agar plates. Due to the presence of the Lac Z a fragment in the vector for ea sy blue/white colony screening, the identification of recombinant E. coli was performed by addi ng 40l of 40mg/ml Xgal on the LB plate and white colonies we re selected. Recombinant plasmid was isolated and analyzed by PCR amplific ation (using forward M13 primer (Invitrogen) and reverse R1 prime r), and restriction digestion by Eco RI Results were analyzed by agarose gel electr ophoresis. The gene was also sequenced and the resulting sequence was used as a query for investigation using databases on the Internet. Subcloning of the smcol Gene Into the Expressi on Vector pET100/D-TOPO The smcol fragment obtained by PCR wa s cloned into pET100/D-TOPO vector and used to transform competent E. coli BL21 (DE3) by the heat shock method as described previous ly with some modification s: the fresh PCR product was first diluted 1:50 and 4l of the dilu ted PCR product were added to 1l of the vector and 1l of distilled water. Transformation of E. coli was performed a described below. Plasmid was isolated from recombinant E. coli analyzed by agarose gel electrophoresis, and amplified by PCR usi ng a combination of the
17 following primers: T7 forward and re verse with F2 and R1 primers. Gene sequencing was also performed to make sure that the smcol gene was ligated in frame with the vect or and in the corre ct orientation. Expression of smcol -pET Recombinant smcol -pET plasmid (10ng in a 5l volume) was used to transform 250l of competent E. coli BL21 Star (DE3) by the heat shock method. After 1 hour of incubation at 37oC the transformation reaction was added to 10ml of LB-amp and grown for 16 hours at 37 oC with shaking. Then, 500 l or 2.5ml of the overnight culture was added to 10ml or 50ml of LB-amp medium and grown for another 2 hours or until OD600 reached 0.5. The culture was then divided into 2 tubes, one of which was i nduced by the addition of 1mM isopropyl -D thiogalactoside (IPTG), while the ot her was used as a non-induced control. Competent cells transformed wit h empty vector were used as a negative control. From each culture 500l were removed every hour for 4 hours and analyzed by SDS PAGE. SDS-PAGE Samples to be analyzed (Recombinant E. coli cells, cell extract or pure enzyme) were separated on a 12% SDS polyacrylamide gel (0.75mm) using a Mini-PROTEAN II electrophoresis cell (B io-Rad). Separation was performed at 150V for 1h and the gel was stained with Coomassie brilliant blue R-100
18 overnight. The stain was then removed by incubating the gels for 30 minutes in a destaining solution (Methanol: Acetic acid: H2O at a ratio of 4:1:5 respectively). Purification of Recombinant S. mutans Collagenase Smcol was cloned into the pET vector in order to obtain a fusion protein with 6XHis-tag at the Nterminal in order to facilitate its purification by affinity column chromatography using ProBondTM (Invitrogen) resin precharged with Ni2+. Purification was performed under native an d hybrid conditions as described by the manufacturer. The pellet from an ov ernight 50ml culture was harvested and resuspended in 8ml a buffer containing 50mM NaPO4-0.5M NaCl (pH 8). The solution was sonicated on ic e using six 10-second bursts at high intensity with a 10-second cooling period between each bur st. After 15-minute centrifugation at 3000 X g the supernatant was removed and purified under native conditions. In order to recover the insoluble proteins the pellet was pur ified under hybrid conditions. Briefly, the pellet was resus pended in 8ml of Guanidinium Lysis buffer (6M Guanidine HCl, 20mM NaPO4, 500mM Na Cl, pH 7.8), sonicated on ice with three 5-second pulses at high intensit y and centrifuged for 15 minutes. The supernatant was then used fo r column purification. Columns were prepared according to th e respective sample (soluble or insoluble). Initially, the resin was washed with sterile water followed by a Native Binding Buffer wash (50mM NaPO4, 0.5M NaCl, 10mM imidazole, pH 8) or Denaturing Binding Buffer wash (8M Ur ea, 20mM NaPO4, 500mM NaCl, pH 7.8). The supernatant containing the soluble pr oteins was applied onto the column and
19 allowed to bind for 1 hour on a rota ting wheel. The supernatant was then aspirated and the resin was washed thr ee times with Native Binding Buffer (50mM NaPO4, 0.5M NaCl) and used to pack a colu mn. The soluble protein was collected in 1ml fractions after the addition of 8ml of Native Elution Buffer. Sample containing insoluble prot eins was added to the column and allowed to bind for 30 minutes. Unbound proteins were removed and the resin was washed sequentially with a Denaturi ng Binding Buffer (pH 7.8) and twice with a Denaturing Wash Buffer (pH 6). The resin was then resuspended four times in Native Wash Buffer and used to pa ck a column. Elution and collection of the protein was as abov e. Purified lysate of E. coli BL21 transformed with an empty vector served as the negative co ntrol. The samples obtained were analyzed by SDS-PAGE. Western Blot Analysis Western immunoblot was performed on cell lysate samples containing collagenase and on the purified recombi nant protein obtai ned under hybrid conditions. Proteins were separated by SDS-PAGE as previously described. Transfer of protein bands to nitrocellulo se was carried out overnight at 30V in a Mini Trans-Blot Electrophoretic Transfer Cell (Bio-Rad). The membrane was blocked for 1 hour in 10ml blocking buffe r (10% Dry milk in PBST) and washed twice with PBST (0.05% Tween-20 in PBS) The membrane was then incubated for 1 hour with anti-HisG (Invitrogen) di luted 1: 5,000 in blocking buffer and washed twice with PBST at room tem perature. Finally, the membrane was
20 incubated for 1 hour with rabbit anti-mous e IgG (Sigma) diluted 1:30,000 in blocking buffer at room temperature. Excess conjugate wa s removed by two washes with PBST. The band developmen t was performed according to DaoÂ’s method (11) by incubating the membrane with sodium borate buffer (60mM Sodium Tetraborate, 10mM M agnesium Sulfate, pH 9.7) containing 0.025% of odianisidine and 0.025% of -naphthyl acid phosphate. Sample Preparation for Collagenase Assay Cell lysates used for collagen assay were prepared as fo llowed: A pellet from centrifugation (8000 X g for 10 min) of 10ml culture was resuspended in lysis buffer and freeze-thawed 4 times on dry ice and 42oC water bath respectively. The lysates were then centrifuged at 8000 X g for 20 minutes to remove intact cells, cellular debris, and cell membranes. 100l of the lysate was used for collagenase assay. To dete rmine collagenase activity in whole recombinant E. coli cells, 10ml culture were centrifuged at 800 X g for 10 minutes and the pellet obtained was resuspended in 2ml collag en assay reaction buffer. 100l of the sample was used for the assay. Before readings were taken, the reaction mixture was centrifuged briefly for 10 seconds. Collagenase Assay Collagenase activity was measured using the EnzChekTM Collagenase Assay Kit (Molecular Probes Inc., Eugene, OR) as described by the manufacturerÂ’s protocol. Each 200l-r eaction mixture cont ained 80l of 1X
21 Reaction Buffer (0.05M Tris -HCl, 0.15M NaCl, 5mM CaCl2, 0.2mM NaN, pH 7.6) and 20l of DQ co llagen in a final concentrat ion of 100g/ml. Various concentrations of the purified recombinant enzyme were used for the assay (10, 20, 30, or 40 g of the recombinant enzyme diluted in 1X Reaction Buffer) as well as cell lysates and whole cells. Clostridium histolyticum collagenase IV (0.4 U/ml, Molecular Probes Inc.) was used as a posi tive control 100l of reaction buffer were used as a negative blank, and whole cells and cell lysate from E. coli BL21 transformed with empty vector were used as negative controls. 10g of trypsin were used to assess the specificity of t he assay. The samples were incubated at room 37oC for 24h and the fluorescence from the digestion products was measured using a fluoromete r at excitation and emission wavelengths of 490 and 520nm respectively. Then, the highest conc entration of the purified enzyme was incubated for additional 24h in order to se e the rate of collagen degradation as compared to the positive cont rol. All experiments were performed in triplicate. Gelatin Hydrolysis Determination of gelatinolytic activity was performed initially by gelatin zymography. Samples were incubated at ro om temperature for 2 hours with 2X Sample Buffer (1.25M Tris-HCl, pH 6.8, Glycerol, 2. 5% SDS, 0.005% Bromophenol Blue), without he ating, and under non-reduci ng condition (without the addition of -mercaptoethanol) and then app lied in ready made 10% polyacrylamide gels contai ning 0.1% gelatin (Bio-Rad ). The electrophoresis was performed at 125V for 1 hour and the gels were incubated with 2.5% Triton X-
22 100 (Renaturing Buffer) for 30 minutes at 25oC, and then overnight (16h) at 37oC with a buffer containing 50mM Tr is-HCl, 0.2M NaCl, 5mM CaCl2, and 0.02% Brij 35. Finally, the gels were stained with 0.5% Coomassie Blue R-250 and destained with the de staining solution. Areas of gelatinase activity appeared as clear bands against a blue background. The samples tested included purified recombinant collagenase, IPTG induced E. coli BL21 whole cells containing the smcol -pET, C. histolyticum collagenase as a positive control, and E. coli BL21 transformed with empty vector as a negative control. Additionally, 40g of the purified enzyme was in cubated for 24h at 37oC using 20 l/ml DQ gelatin. The same controls used in the collagen assa y were also tested in parallel. All experiments were performed in triplicate. Collagenase Inhibition Assay Collagenase activity was assayed as described above but in the presence of the following known protease inhi bitors: 10mM of ED TA, 10mM of 1,10phenanthroline, or 2mM PMSF. Incubation was at 37oC for 48h. Release of fluorescence was recorded and compared to that obtained in the absence of protease inhibitor. All experim ents were performed in triplicate Protein Concentration Protein concentration was determined by the method of Bradford (6). A serial dilution of BSA was prepared and assayed in parallel with the test samples. After addition of the Dye Reage nt Concentrate (Bio Rad) to the samples at the
23 concentration specified by the m anufacturer the absorption at OD595 was recorded. Values obtained with the BSA were used to establish a standard curve from which the protein concentration of the test samples was derived. DNA Sequencing and Sequence Analysis DNA sequencing of recombinant plasmids was performed by the Molecular Biology Core Facility at the H. Lee Moffitt Cancer Center & Research Institute (Tampa, FL). The sequence of smcol -TA and smcol -pET were used as a query for various databases on the inter net. The sequence obtained from cloning of the gene in pCR2.1-TOPO wa s initially compared to the S. mutans UA 159, (Oklahoma Genome Center) using BLAST2seq database (1). Additionally, other databases such as SWISSPROT, PROTOMAP, DAS (10,16,17) gave information about conserved sequenc es of the deduced aminoacid sequence expected molecular weight and pot ential transmembrane regions.
24 RESULTS PCR Amplification of S. mutans Collagenase Gene and Cloning into pCR 2.1 TOPO-TA Vector Following genomic isolation of S. mutan s GS-5, the gene of interest ( smcol ) was amplified by PCR. Custom primers were designed based on the DNA sequence of putative S. mutans UA159 collagenase (Oklahoma Genome Center). The aim was to amplify the gene from the start c odon (ATG) through the stop codon (TAA). Once the 1287bp am plicon was obtained and confirmed by agarose gel electrophoresis (Fig. 1), the gene was cloned into pCR2.1-TOPO TA vector for sequence analysis. TOPO TA vectors (Invitrogen) employ an easy and efficient way to clone PCR products. The single deoxyribose adenosine (dA) overhangs added by Taq polymerase to the 3Â’ end of PCR products, are complementary to the 3Â’ deoxyribose thym idine (dT) overhangs of the linearized vector, thus binding spontaneously the PC R product and the vector. Additionally, the vector is conjugated with Topoisome rase I, an enzyme that cleaves the phosphodiester bonds at specific sites on the vector backbone. This creates a high energy covalent bond between the phosphate resi due of the cleaved DNA and the enzyme (tyrosine molecule). This energy is released when the newly inserted DNA attacks the bond, thus reve rsing the reaction by ligating the inserted gene into the vector and releasing the Topoisomerase.
25 Positive clones were selected on t he basis of blue/white screening. White color colonies were an indication of posit ive clones, because of interruption of the LacZ gene by the inserted gene. On the other hand, blue colonies suggested that the gene was not present in the specific clone. 10 white colonies were selected and analyzed for the presence of the coll agenase gene. The cloning efficiency was as high as 80%. Only one positive clone was used for further analyses. After plasmid isolation, the recombinant plasmid ( smcol -TA) was subjected to restriction enzyme digestion by Eco RI and PCR analysis (Fig. 2). As anticipated, restriction enzyme digestion gave two distinct bands around 3.9 kb and 1.2 kb (Fig. 2A) corresponding to the linearized vector and the smcol insert, respectively. PCR amplification using 20 ng of the smcol -TA as a template and M13-complementary forwar d primer (vector) and c ol -complementary R1 reverse primer confirmed these results as a singl e discrete band of a pproximately 1.3 kb was obtained (Fig. 2B) Figure 1. PCR amplification of the putative collagenase gene of S. mutans GS-5. Agarose gel electrophoresis of the PCR product showed a DNA band of approximately 1.2 kb.
26 Sequence Analysis of smcol -TA The sequence of smcolTA was obtained and anal yzed. Alignment of smcol sequence with that of the correspondi ng gene in UA 159 strain showed 99% homology (Fig. 3). There were a total of 4 nucleotide variat ions that resulted in three amino acid differences and one silent mutation. Furthermore, the deduced amino acid sequence of the gene wa s used as a query for sequence analyses. Alignment using the TBLASTX database, showed high similarity of the protein with other proteases, predominatel y collagenases, including the PrtC collagenase from t he periodontopathogen P. gingivalis These preliminary results open the way for further investigation of the S. mutans GS-5 collagenase gene. Figure 2. Insertion of P CR product into pCR 2.1-TOPO was confirmed by restriction enzyme digestion (A ) and PCR amplification (B). A: Lane 1: -Hind III MW marker Lane 2: Supercoil ed recombinant plasmid Lane 3: Eco RI digest of the recombinant vector produced 2 distinct bands corresponding to the linearized vector (3.9Kb) and the smcol insert (1.2kb) B: Lane 1: 100bp ladder. Lane 2: PCR amplific ation of the inserted gene produc ed the expected size product.
27 GS-5 UA 159 Figure 3. Alignment of the smcol sequence of strain GS-5 with the corresponding sequence of UA 159 strain Alignment (BLAST2seq) revealed 99% similarity between the two genes. Th e red circle represents 1 of the 4 nucleotides where the two s equences showed discrepancy. Cloning of the smcol-p ET Gene To clone and express S. mutans collagenase the pET1 00/D-TOPO vector (Invitrogen) was used as it offered m any advantages over pr evious TOPO TA cloning vectors which require the use of Taq polymerase to generate PCR product with dA overhangs. Like TOPO TA vectors the pET100/D-TOPO vector was constructed with topoisomera se properties, but instead of Taq polymerase it requires the use of proofr eading polymerase such as Pfu hotstart polymerase that removes the overhangs. Hence, t he gene of interest can be cloned in frame with the genes encoded by the vector Another requirement for in-frame as well as directional ligation is the presence of 5Â’-CACC-3Â’ upstr eam of the desired gene to be cloned. This ensures efficient li gation of the gene due to the presence of 3Â’ complementary sequence on the linear ized vector. Thus, a forward primer incorporating the 4 additi onal nucleotides was used (F2 primer). As an expression vector, pET-100 contains an IPTG-inducible promoter, as well as a
28 binding site for lac repressor to reduce ba sal expression of the protein. It also consists of an N-terminal histidine fusion tag that allows purif ication and detection of the recombinant protein. The 1.2 kb PCR product (Fig. 4) obtained was diluted 1:50 and used for subcloning into the pET100/D-TOP O. Since no blue/white screening was possible, 10 randomly selected colonies were chosen for analysis of positives clones. The cloning efficiency was only 40 %. Only one positive clone was used in subsequent studies. As previously observ ed, restriction enzyme analysis of the plasmid isolated from the recombinant clone showed two discrete bands corresponding to the linearized vector (5.7 kb) and the smcol insert (1.2 kb) (Fig. 6A). Confirmation of t he cloned gene came from P CR amplificat ion using smcol pET as a template and combinations of 2 sets of primers (Table 1, Figure 5): T7 forward and reverse primers hybridizi ng to vector sequence and F2 and R1 primers hybridizing to the smcol gene. As shown in Figure 6B different sizes of PCR products ranging from 1.2 Â– 1.5 kb were obtai ned. The 1.2 kb reflects the actual size of the smcol gene (Lane 3, F2 a nd R1 primers), while the rest (Lanes 2 and 4, T7 Forward and R1, F2 and T7 revers e respectively) indicate that the 5Â’ and 3Â’ of the collagenase g ene was successfully ligated. The larger 1.5 kb (Lane 5, T7 forward and reverse) confirmed that there was no deletion in the vector in the proximity of the insertion site.
29 Figure 4. PCR product obtained from genomic DNA Lane 1: -Hind III MW marker Lane 2: 1287bp PCR product Sequence Analysis of smcolpET The cloned smcolpET was subjected to sequence analysis in order to detect any possible mutations that had pot entially taken place during cloning or PCR amplification as well as to confi rm that the gene was ligated in-frame with the histidine residues (Figure 5) BLAST2 alignment (1) of the smcolpET gave 100% similarity with the respective sequence of the S. mutans UA 159. A more extensive analysis of the deduced amin o acid sequence was also performed using various databases to examine impor tant motifs and conserved regions of the suspected collagenase. Databases such as ExPASy, Pfam, and SMART (3,16,41), gave more information about t he protein in search. The results indicated high homology with U32 fa mily peptidases. Even though these enzymes are of unknown catalytic orig in, they are represented by well
30 characterized species, P. gingivalis The consensus sequence present in all U32 peptidases is E-x-F-x(2)-G-[ SA]-[LIVM]-C-x(4)-G-x-C-x-[ LIVM]-S and is used as a signature pattern for these families. Furthermore, the sequence was run as a query using the ProtoMap databas e. ProtoMap (17) classi fies all proteins in SWISSPROT and TrEMBL databases into gr oups of related proteins. The query was identified as a member of cluste r #1872. There are 24 members that belong to this cluster, 13 of which belong to the U32 family peptidases. Notable is the fact at 50% of the member s are collagenases, including P. gingivalis prtC, while the rest are putative bacter ial proteases. In agreement with the results mentioned above, PSI-BLAST (1) placed the consensus sequence in U32 Peptidase family as well. The European Molecular Bi ology Laboratory (EMBL) database also placed the enzyme into E.C. 3.4, a designat ed broad family of hydrolases that act on peptide bonds; members of this family include collagenases such as the well characterized zinc-metalloproteinases of C. histolyticum and V. alginolyticus In order to get a more conclusive pictur e of the protein, another annotation was employed. Investigations regarding the presence of any phylogenetic pattern between our sequence and t he Cluster of Orthologous Group of Protein (COG) database (1) were also performed. This pr ogram consists of carefully annotated sets of likely orthologs and assigns new pr oteins to COGs by comparing them to the sequences in the COG dat abase. When three or more genome-specific best hits (BeTS) are detected then the quer y sequence is considered a likely new COG member (4). Submission of our qu ery gave BeTS to 6 clades of COG0826 designated as collagenase and related proteases.
31 Another important information was al so obtained, which added to our knowledge of the enzyme in search. Dense Alignment Surface (DAS) method uses a single sequence as a query and predicts possible transmembrane segments in prokaryotic membrane protei ns (10). Using the deduced aminoacid sequence as a query it was suggested t hat the enzyme potentially spans the membrane at position 96 through 108. Figure 5. Map of pET 100/ D-TOPO with collagenase inse rt (blue). The 5Â’-CACC3Â’ in the 5Â’ end of the insert provides directional and in-frame cloning of the gene with the 6x His fusion tag. Ampicillin resi stant positive clones were selected and the recombinant plasmid was subjected to sequence analysis.T7 promoter was IPTG induced for high expression of the gene.
32 Table 1. PCR primer pairs used for c onfirmation of successful ligation of smcol in pET100 Forward Primer Reverse Primer DNA Size (bp) Lane (Figure 6) T7 Forward R1 1489 2 F2 R1 1291 3 F2 T7 Reverse 1373 4 T7 Forward T7 Reverse 1563 5 Figure 6. Confirmation of successful in sertion of the colla genase gene into pET 100/D-TOPO. A. Lanes 1 and 2: Molecular weight standard Lane 3: Undigeste d recombinant plasmid Lane 4: Recombi nant plasmid digested with Eco RI:Two bands were obtained corresponding to lin earized vector (5.7 kb) and the smcol insert (1.2kb) B.PCR products generated by different primer sets (T able 1) confirmed successful ligation and correct orientation of the insert.
33 SDS PAGE of the IPTG Induced smcol -pET 1mM of IPTG was used to induce E. coli BL21 harboring the smcol -pET recombinant plasmid. The induction was performed over a 4-hour period with samples taken every hour after the addition of IPTG. Figure 7 shows the results obtained when samples were analyzed by SDS PAGE. Induction was already apparent at 1 hour post-induction as compared to at time 0, and the expression continued to be induced over the 4 hour period. One strong un ique band of 50kDa was observed to increase in size wit h time. The size of the induced band was larger than the anticipat ed collagenase size of 47kDa to account for the presence of the 3 kDa His-tag. Figure 7. SDS analysis of the induced E. coli BL 21 cells transformed with the recombinant plasmid smcol -pET. Equal volumes of cell lysates were loaded into each lane. Induction by IPTG was followed over 4Â–hour time period and samples were collected at hourly intervals. Lane 1: E. coli BL 21 transformed with empty vector (negative control); Lanes 2-6: Expression of a 50kDa band was evident at 1 hour post-induction and increased ther eafter. MW: SDS molecular weight marker
34 SDS and Western Blot analysis of Purified Recombinant Protein High affinity purification column wa s used in order to purify the recombinant protein that was tagged with si x tandem histidine residues at the Nterminal end. The samples applied onto th e column consisted of protein purified under native and hybrid conditions. Althou gh there was no prot ein eluted from the soluble supernatant under native condi tions, the insoluble pellet that was recovered after lysis, showed one strong 50kDa band after SDS analysis (Fig. 8). This sample, in parallel with a non-purifi ed sample as well as a negative control, was subjected to western blot analysis in order to confirm that the 50kDa was the anticipated fused protein. As shown in Figure 9A, the non-purif ied sample (Lanes 2, 3) showed many bands wit h higher intensity being the 50kDa fused protein. In agreement with this, the purified protein also showed high intensity band at the same area confirming that the enzyme was successfully expressed and purified (Fig. 9B). Smaller bands with lower size and intensity were also present in the fraction tubes.
35 Figure 8. SDS-PAGE of purif ied recombinant protein. Purity was confirmed by the obtaining of a single 50kDa band corre sponding to the expression of a fusion protein formed by the 47kDa S. mutans collagenase and the 3kDa his-tag. Lane 1: SDS MW marker; Lane 2: Purifi ed recombinant fusion protein. Gelatin Hydrolysis Gelatin zymogram was performed in order to investigate the ability of the enzyme to degrade gelatin, a denatured fo rm of collagen. After 16 hours of incubation at 37oC with collagenase sample buffer, E. coli whole cells indicated the presence of gelatinolyt ic activity as a clear band of approximately 90-100 kDa over a blue background (Fig. 10). No activity was observed with the negative control or with the pure fusion enzyme pr otein. In order to see if a longer incubation time would have any effect on t he gelatinolytic activity of the enzyme, the purified collagenase-his tag fusion protein was incubat ed for 24h at 37oC with fluorescein conjugated gelatin Increase in fluorescence wo uld be an indication of gelatin hydrolysis. As in the gelatin zymography ex periment, no significant fluorescent change was observed at 16h inc ubation, but at 24 hours a significant
36 increase in the fluorescence was observ ed. The result suggested that the purified fusion protei n had weak activity. The specificity of the assay was demonstrated by the strong activity observed with C. histolyticum collagenase, whereas the negative control had no effect on the substrate. Figure 9. Western immunoblots of E. coli cell lysates containing the HisCollagenase fusion protein (A ) and the purified collagenase (B). Expression of the fusion protein was test ed by western blot analysis using a mouse specific antibody to his-tag as a primary antibody and an alkaline phosphatase conjugated rabbit as a secondary antibody. A purple coloration was indicative of a positive reaction A. Lane 1: E. coli transformed with empty vector(-) Lane 2: E. coli transformed with recombinant vector 1h after induction Lane 3: E. coli transformed with recombinant vector 4h after induction B. Purified recombinant protein. Collagen Assay True collagenases are capable of hydr olyzing collagen in its native form. The Enzchek collagenase assay kit (Molec ular Probes) provides a sensitive and rapid way to assay for collagen hydrolysis. Fluorescein-conjugated Type I collagen was used as a subs trate, and any significant increase in fluorescence
37 would be an indication of co llagen hydrolysis. Initially, as showed in figure 11, only the 30 and 40g of the purifie d enzyme demonstrated significant collagenolytic activity 24 hour s after incubation at 37oC (Fig. 11A). A sample containing 40g of purified fusion protein was next in cubated for 48h before measurement of collagenolyti c activity (Fig. 11B). Collagen degradation was comparable to the levels obser ved with that of the positive C. histolyticum collagenase (positive control). As anticipated, no activity was observed with the negative control. It is notewor thy that whole recombinant E. coli cells degraded collagen while the cell lysa te supernatant did not, sugg esting that the enzyme may be expressed as bound to the cell me mbrane (Fig. 11A). The specificity of the assay was confirmed by the failure of trypsin, a non-specific protease, to degrade native collagen. A small backgr ound of fluorescenc e was observed and attributable due to the degradation of nonhelical components of collagen or to the presence of small amounts of denatured collagen spontaneously generated during the commercial preparat ion of the substrate.
38 Figure10. Gelatin zymogram of recombinant E. coli cells and purified recombinant protein. The samples were separated in a gelati n-containing gel under non-reducing conditions, and incubated for 16h at 37oC. A clear band in the gel indicated gel atin degradation. Lane 1: Purified recombinant fusion protein Lane 2: Recombinant E. coli whole cells expressing the recombinant protein as a 95kDa band corresponding to a dimer of the 50kDa observed in SDS-PAGE under reducing conditions Inhibition Assay Three different inhibitors were used to further characterize the properties of the S. mutans GS-5 collagenase. As shown in Table 2, only EDTA reduced the enzyme activity. No inhibition took pl ace with either the metalloproteinase inhibitor 1,10-phenathroline or the serine protease inhibitor PMSF.
39 A. Collagen Degradation 24h After Incubation0 100 200 300 400 500 600 700 800Fluorescenc e E. coli BL 21 Trypsin 10g SMC 20g SMC 30g SMC 40g SMC Whole Cell Cell lysate B. Collagen Degradation of Purified Enzyme 0 500 1000 1500 2000 2500 3000 05101520253035404550Time (hours)Fluorescenc e C. histolyticum collagenase S. mutans collagenase Figure 11. Degradation of fluorescein -conjugated Type I collagen by purified fusion protein. Increase in fluorescenc e indicated collagen degradation. A. E. coli cells transformed with empty vector and tr ypsin were used as negative controls. E. coli cell lysate did not s how significant degradation after 24h incubation, while whole cells degraded a signi ficant amount of collagen. Different amounts of purified enzyme were tested (10-40g) to show dose-dependent enzyme activity. SMC: Streptococcus mutans collagenase. B. Consider able amounts of collagen degradation was evident 48h after incubat ion. Activity of 0.4U/ml of C. histolyticum collagenase was tested a posit ive control for the assay.
40 Table2. Inhibition assays for collagenase activity Organism Fluorescence w/o Inhibitor Fluorescence w/t inhibitor EDTA 1,10-PHEN PMSF 10mM 10mM 2mM C. histolyticum (0.4U/ml) 2524.88 452.15 393.2 3349.87 S. mutans GS-5 (40g) 2618.2 1040.15 2276.55 2542.45 Three common protease inhibitors were used to further characterize the S. mutans collagenase enzyme (EDTA, 1,10phenant hroline, PMSF). Activity of S. mutans collagenase was reduced by the pr esence of EDTA, but not by 1,10phenanthroline nor PMSF, whereas C. histolyticum collagenase was drastically inhibited by EDTA and 1,10 phenanthroline. These results we re consistent with the difference in the nat ure of the two enzymes.
41 DISCUSSION The present study aimed at the cl oning of a collagenase gene of S. mutans and its expression as an ac tive enzyme in recombinant E. coli. Although S. mutans has been implicated in dental root decay and found in dental root in association with collagen fibril degradation, there has been no proof of the production of collagenase enzyme by S. mutans A two step cloning approach was tak en. First, the pCR2.1-TOPO was used in the cloning of PCR amplified smcol gene for sequence analysis. Once the sequence was establis hed a strategy was devised in order to clone the smcol gene into the expression vector pET100/DTOPO in frame and directionally in order to insure its expression. Primer s were designed direct ly on the start and stop codon so that only t he open reading frame was PCR amplified. Sequence analysis of the initial cloned gene in the pCR2.1-TOPO TA vector showed 99% similarity (Fig. 3) with S. mutans UA 159, but 100% homology was observed with the gene cloned in the pET100/D-TOPO. This difference in results was due to the error prone nature of Taq polymerase as compared to the proof reading Pfu hotstart polymerase used in the smcol gene amplification for cloning into pET100/D-TOPO. This enzyme is characte rized by a 3Â’-5Â’ exonuclease activity that minimizes the possibility of inserti ng mutations during PCR amplification. The cloned gene was successfully expre ssed as a 50 kDa protein at high levels after induction with IPTG (Fig. 7) The recombinant protein appeared to be
42 associated with the insoluble memb rane fraction of the recombinant E. coli and consequently was difficult to recover fr om the insoluble pel let. Nevertheless, western blot analysis of the pur ified protein confirmed that the expressed, purified protein was the desired enzym e. As depicted in Figure 9, there was a major 50kDa recombinant protein band corresponding to the 47kDa S. mutans collagenase fused with a 3kDa His-Tag. The presence of some smaller bands in the western blot was probably attri butable to proteolytic degradation, a phenomenon commonly observed with the expression of S. mutans antigens in E. coli as reported by Yoder et al. for the S. mutans wall-associate protein (53). Jackson et al. (24) showed that the S. mutans cell-associated enzyme was responsible for the collagenolytic ac tivity of the organism. In the present study it is suggested t hat the collagenase enzyme spans the membrane from aminoacid 96 through 108 (10). This was co nfirmed by the observed collagenase activity in recombinant E. coli whole cells (Fig. 11A) and lack of activity in the soluble cell extract. Inhibition of S. mutans collagenase by EDTA wa s in agreement with prior works (22, 24), thus the enzyme can be considered a metalloproteinase. Unlike bacterial zinc metalloproteinases that are inhibited by bot h EDTA and 1,10phenanthroline, the purified S. mutans fusion enzyme activity was reduced only by the addition of EDTA, and not by 1,10 phenanthroline. These data are in partial agreement with Jackson et al. (24) that showed inhibition of S. mutans cell-associated collagenolytic activity by both EDTA and 1, 10phenanthroline. This apparent discrepancy may be attributed to either a difference in assay
43 conditions, testing of whole S. mutans GS-5 cells in the study by Jackson et al. (1997) versus a purified S. mutans collagenase-fusion protein in the present study, or to the possibility of more than one collagenase enzyme being produced by S. mutans The possibility of such an occurrence was supported by the presence of two antigen bands in pr evious western immunoblot of S. mutans (24). Further purification of the S. mutans enzyme without the His-tag and activity testing will dete rmine whether or not there wa s any interference caused by the extra His-tag region. Sequence analysis of the deduced amino acid sequence gave high similarities with other bacterial prot eases, mainly collagenases. Specifically, S. mutans collagenase belongs to the U32 fa mily peptidase, a broad family of enzymes with unknown catalytic domain. A shared characteristic is the presence of two cysteine residues within the consensus sequence E-x-F-x(2)-G[SA]-[LIVM]-C-x(4)-G-x-C-x -[LIVM]-S. Surprisingly, PrtC is one of the well characterized members of the U 32 peptidases. Alignment of the S. mutans collagenase with the PrtC of P. gingivalis deduced amino acid sequences only showed 48% homology. This reflect ed the heterogeneity in the U32 enzyme family. Indeed, heterogeneity of the prt C gene was also observed in different clinical isolates (51). Nevertheless, the two enzym es shared two important characteristics. Firstly, Kato et al. showed by gel filtration that PrtC was a dimer in its active form. Similarly, in t he present work, gelatin zymogram analysis showed an active band of 95kDa under non-reducing conditions. Secondly, both
44 enzymes, PrtC and S. mutans collagenase were inhibited by EDTA suggesting a requirement for Ca2+, but not by 1, 10-phenanthroline. Dental root decay is a disease implicating many organisms and different virulence factors, making it hard to investig ate a single factor individually. Despite this difficulty, we successfully cloned and expressed S. mutans collagenase in E. coli and demonstrated that it had bot h collagenase and gelatinase activity. Experimental results (Fig. 11B) indicated th at the activity was relatively weaker than that of the C. histolyticum collagenase. This observation was, however, in agreement with the slow progres sion of dental root decay in infected individuals, and the in vitro studies of S. mutans collagenolytic activity (24). Undoubtedly, more work is needed to fully elucidate the nature and function of S. mutans collagenase in dental root dec ay. From the data presented above, it is already clear that S. mutans collagenase is a novel enzyme, sharing some characteristics with P. gingivalis PrtC but not all. For instance, PrtC did not have gelatinolytic activity. The present study has provided the tool necessary to further our understanding of S. mutans collagenase. Not only have we conclusively demonstrated for the first ti me the existence of a collagenase gene in S. mutans GS-5, but with the obtai ning of a recombinant E. coli clone expressing the S. mutans enzyme, further analysis will be greatly facilitated.
45 REFERENCES 1. Altschul, S.F., T.L. Madden, A.A. Sch ffer, J. Zhang, Z. Zhang, W. Miller, and D.J. Lipman.1997. Gapped BLAST and PSI-BLAST: a new generation of protein dat abase search programs." Nucleic Acids Res. 25:3389-3402. 2. Awad,M.M.,D.M. Ellemor, A.E. Bryant, O. Matsushita, R.L. Boyd, D.L.Stevens, J.J. Emmins, and J.I Rood. 2000. Construction and virulence testing of a collagenase mutant of Clostridium perfringens Microb. Pathog. 28, 107-17. 3. Bateman A, E. Birney, L. Cerruti R. Durbin, L. Etwille r, S.R. Eddy, S. Griffiths-Jones, K.L. Howe, M. Mars hall, E.L Sonnhammer. The Pfam Protein Families Database. 2002. Nucleic Acids Research 30(1):276-280 4. Baxevanis, A.D. and B.F. F. Ouellette. Bioinformati cs. A practical guide to the analysis of genes and proteins. 2nd Ed. Wiley Interscience. New York. 2001. 5. Beg, A.M, M.N. Jones, T. M. To rbert, and R.G. Holt. 2002. Binding of Streptococcus mutans to extracellular matrix molecules and fibrinogen. Bioch. Biophys. Res. Communic. 298: 75-79. 6. Bradford M.M. A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of pr otein-dye binding. 1976. Anal. Biochem. 72: 248-54. 7. Brady, J.J., L. Marloes, A. Til burg, C.E. Alford, and W.P. McArt hur. 2000. Monoclonal antibody-mediated modul ation of the humoral immune response against mucosally applied Streptococcus mutans Infec. Immun. 68:1796-1805. 8. Burgess, O.B., J.R. Gallo. 2002. Tr eating Root Surface Caries. Dent. Clin. N. Am. 46: 385-404. 9. Chang, Y.C., S.F. Yang, J.Y. Liu, and Y.S. Hs ieb. 2002. Regulation of matrix metalloprotease production by cytokines, pharmacological agents, and periodontal agents in human perio dontal ligament and fibroblast cultures. J. Period. Res. 37: 129-203.
46 10. Cserzo, M., E. Wallin, I. Simon, G. von Heijne and A. Elofsson.1997. Prediction of transmembrane alpha-he lices in prokaryotic membrane proteins: the Dense Alignment Surf ace method. Prot. Eng.10(6):673-76. 11. Dao, M.L. 1985. An improved method of antigen detection in nitrocellulose: In situ staining of alkaline phosp hatase conjugated antibody. J. Immunol. Methods. 82: 225-231. 12. De Souza, A.P., S.R.P. Line. 2002. Th e biology of matrix metalloproteases. Rev. FOB. 10(1): 1-6. 13. Dumas J, N Hurion, R. We ill, and B.Keil. 1991. Collagenase in mineralized tissues of human teeth. Arch Oral Biol.36(4):299-304 14. Dung, S.Z., A. H. Liu. 1999. Molecular pathogenesis of root dentin caries. Review. Oral Diseases. 5: 92-99. 15. Dung, S.Z., R.L Grecory, and G.K St ookey. Effect of lactic acid and proteolytic enzymes on t he release of organic matrix components from human root dentine. 1995. Ca ries Res. 26 (6): 483-89. 16. Gasteiger E., A. Gattiker, C. Hoogl and, I. Ivanyi, R.D. Appel, and A. Bairoch. ExPASy: the proteomics serv er for in-depth protein knowledge and analysis. Nucleic Acids Res. 31:3784-3788. 17. Golan, Y., Linial, N., and Lini al M. 2000. ProtoMap: Automatic classification of protein sequences and hierarchy of protein families. Nucleic Acids Research, 28: 49-55. 18. Hahn, C.L., A.M. Best, and J.G. Tew. 2000. Cytokine induction by Streptococcus mutans and pulpal pathogenesis. Infec. Immun. 68(12):6785-89. 19. Hahn, P., D. Reinhardt, H.G. Schaller, and E. Hellwig. 1999. Root lesions in a group of 50Â–60 year-old Germans rela ted to clinical and social factors. Clin. Oral Invest. 3:168Â–74 20. Han, T.K., S. Yoder, C. Cao, K.E. Ugen, and M.L. Dao. 2001. Expression of Streptococcus mutans wall-associated prot ein A gene in Chinese hamster ovary cells: Prospect for dent al caries DNA vaccine. DNA and Cell. 20(9): 595-601. 21. Harrington, D.J. 1996.Bacterial collagenases and collagen-degrading enzymes and their potential role in hum an disease. Infec. Immun. 64(6): 1885-91.
47 22. Harrington, D.J and R.R.B. Russel. I dentification and characterization of two extracellular proteases of S. mutans 1994. FEMS Microb. Letters. 121: 237-42. 23. Houle, M.A., D. Grenier, P. Plamondon, and K. Mikayama. 2003. The collagenolytic activity of Porphyromonas gingivalis is due to Arg-gingipain. FEMS Letters. 181-85. 24. Jackson, R.J., D. V. Lim, and M.L. Dao. 1997. Identific ation and analysis of a collagenolytic activity in Streptococcus mutans Cur. Microb. 34:4954. 25. Jackson, R.J., M.L. Dao, and D.V. Lim. 1994. Cell-associated collagenolytic activity by group B str eptococci. Infec. Immun. 62(12): 564751. 26. Juarez, Z.E., and M.W. Stinson. 1999. An extracellular protease of Streptococcus gordonii hydrolyzes Type IV collagen and collagen analogues. Infec I mmun. 67:271-278. 27. Jung, C.M., O. Matsushita, S. Katagam a, J. Minami, J. Sakurai, and A. O. Kabe. 1999. Identification of metal ligands in Clostridium histolyticum collagenase. J. Bact. 181(9): 2816-22. 28. Kato T., N. Takahashi, and H. K. Kuramitsu. 1992. Sequence analysis and characterization of the Porphyromonas gingivalis prtC gene, which expresses a novel collagenase activi ty. J. Bact. 174(12):3889-3895 29. Kawasaki, K. and J.D.B. Feathers tone. Effects of collagenase on root demineralization. 1997. 76(1): 588-95. 30. Koga T, Oho T, Sh imazaki Y, Nakano Y. 2002. Immunization against dental caries. Vaccine 1-18. 31. Love R.M. 2002. The effect of tiss ue molecules on bacterial invasion of dentin. Oral Microb. Immun. 17:32-37. 32. Love, R.M., M. McMillan, and I.F. Jenkinson. 1997. Inva sion of dentinal tubules by oral streptococci is associated with collagen mediated by Antigen I/II family peptides. In fec. Immun. 65(12): 5157-64. 33. Makihira, S., H. Nikawa, M. Tamgami, T. Hamda, H. Nishimura, K. Ishida, and H. Yamashiro. 2002. Bacterial and Candida adhesion to intact and denatured collagen in vitro Mycoses. 45: 389-92.
48 34. Matsushita, O. and A. Okabe. 2001. Clostridial hydrolytic enzymes degrading extracellular com ponents. Toxicon. 39: 1769-80. 35. Matsushita, O., C.M. Jung, S. Kataya ma, J. Minami, Y. Takahashi, and A. Okabe. 1999. Gene dublicat ion and multiplicity of collagenases in Clostridium histolyticum J. Bact. 181(3):923-933 36. Matsushita, O., K. Yoshihara, S. Katayama, J. Minami, and A. Okabe. 1994. Purification and characterization of a Clostridium perfringens 120kilodalton collagenase and nucleotide sequence of the corresponding gene. J. Bact. 176(1): 149-156. 37. Moawia, M.K. and R.E. Cohen. 2003. The etiology and prevalence of gingival recession. JADA. 134: 220-25. 38. Oral Health 2000: Facts and Figures U.S. Department of Health and Human Services.May2000. h ttp://www.cdc.gov/OralHealth /factsheets/sgr2000-fs1.htm 39. Qian H and M.L. Dao. 1993. Inactivation of the Streptococcus mutans wall-associated protein A gene (wapA) results in a decrease in sucrosedependent adherence and aggreg ation. Infec. Immun. 61(12):5021-8. 40. Romanelli, R., S. Mancini, C. La schinger, C.M. Overall, J. Sodek, and C.A.G. McCulloh. 1999. Activation of neutrophils co llagenase in periodontidis. Infec. Immun. 67(5): 2319-26. 41. Schultz J., F. Milpetz, Bork P, an d C.P. Ponting. 1998. SMART, a simple modular architecture research tool: identification of signaling domains. Proc. Natl Acad. Sci U S A. 95(11):5857-64 42. Shen, S., L.P. Samar anayake, H.K. Yip, and J. E. Dyson. 2002. Bacterial and yeast flora of root surface caries in elderly, ethnic Chinese. Oral Microbiol. 8: 207-217. 43. Stryer, L.1995. Biochemistry .4th Edition, W. H. Freeman and Company, NewYork. 44. Switalski, L.M., W.G. Buthcer, P.C. Caufield, and M. S. Lantz. 1993. Collagen mediates adhesion of S treptococcus mutans to human dentin. Infec. Immun. 61:4119-4125. 45. Taubman, M.A., D.J. Smith, C.J. Holberg, and J.W. Eastcott. 2000. Coimmunization with complementary gl ycosyltransferases peptides results in enhances immunogenicity and protection against dental caries. Infec. Immun. 68:2698-2703.
49 46. Van der Reijden,W. A., N. Dellemijn-Kippuw A. M. Stijne-van Nes J. J. de Soet and J. van Winkelhoff 2001. Mutans streptoc occi in subgingival plaque of treated and unt reated patients with periodontitis. J Clini Periodontol. 28: 686-91. 47. Van Strijp, A.J.P., D.C. Jansen, J. DeGroot, J.M. ten Cate, and V. Everts.2003. Host-derived prot einases and degradation of dentine collagen in situ. Caries Res. 37: 58-65. 48. Vargas, M. C, E.A. Kramarow, and J. A. Yellowitz. 2001.T he Oral Health of Older American. CDC. 49. Vitagliano, L, G. Zagari, and H.A. Scheraga. 1995. Structure of type I collagen. Molecule based on confo rmational energy. Computating the triple stranded helix and the N terminal telopeptide. J. Mol. Biol. 247: 6880. 50. Warren, J.J., H.J. Cowen, C.M. Watkins, and H.S. Hand. 2000. Dental caries prevalence and dental utilization among very old. J Am Dent Assoc. 131(11)1571-19. 51. Wittstock M., T.F. Flemming, H. Schmidt, and H. Karch. 2000. Heterogeneity of prtC gene of Porphyromonas gingivalis Oral Microbiol. Immun. 15: 33-39. 52. Wittstock M., H. Schmidt, T.F. Flemmi ng, R. Mutters, and H. Karch. 1996. Serodiagnosis of Porphyromonas gingivalis infection by immunoblot analysis with recombinant collagenase. J. Clin. Microb. 34(10): 2411-13. 53. Yoder, S., C. Cao, K. E. Ugen, and M.L. Dao. 2000. High-level expression of truncated wall associated prot ein A from the dental cariogenic Streptococcus mutans DNA and Cell. 19:401-408. 54. Yoshihara, K., O. Matsushita, J. Minami, and A. Okabe 1994. Cloning and nucleotide analysis of the colH gene from Clostridium histolyticum encoding a collagenase and a gelatinase. J. Bact. 176(21): 6489-649. 55. Yu, M.S. and C.Y. Lee. 1999. Expr ession and characterization of the prtV gene encoding a collagenase from Vibrio parahemolyticus in Escherichia coli Microbiology. 145: 143-150. 56. Yu, M.S., M.N. Yap, and C. Y. Lee. 2000. Metal content and biochemical analyses of a recombinant collagenase prtV from Vibrio parahemolyticus. Micriobiol. Immunol. 44(10): 805-13.