A novel metalloprotease from Bacillus cereus for protein fibre processing

A novel protease produced by Bacillus cereus grown on wool as carbon and nitrogen source was purified. B. cereus protease is a neutral metalloprotease with a molecular mass of 45.6 kDa. The optimum activity was at 45 °C and pH 7.0. The substrate specificity was assessed using oxidized insulin B-chain and synthetic peptide substrates. The cleavage of the insulin B-chain was determined to be Asn³, Leu⁶, His¹⁰-Leu¹¹, Ala¹⁴, Glu²¹, after 12 h incubation. Among the peptide substrates, the enzyme did not exhibit activity towards ester substrates; with p-nitroanilide, the kinetic data indicate that aliphatic and aromatic amino acids were the preferred residues at the P₁ position. For furylacryloyl peptides substrates, which are typical substrates for thermolysin, the enzyme exhibited high hydrolytic activity with a K_m values of 0.858 and 2.363 mM for N-(3-[2-Furyl]acryloyl)-Ala-Phe amide and N-(3-[2-Furyl]acryloyl)-Gly-Leu amide, respectively. The purified protease hydrolysed proteins substrates such as azocasein, azocoll, keratin azure and wool.     

Expression in Escherichia coli and simple purification of human Fhit protein

The fragile histidine triad (Fhit) protein is a homodimeric protein with diadenosine 5',5"'-P(1),P(3)-triphosphate (Ap(3)A) asymmetrical hydrolase activity. We have cloned the human cDNA Fhit in the pPROEX-1 vector and expressed with high yield in Escherichia coli with the sequence Met-Gly-His(6)-Asp-Tyr-Asp-Ile-Pro-Thr-Thr followed by a rTEV protease cleavage site, denoted as "H6TV," fused to the N-terminus of Fhit. Expression of H6TV-Fhit in BL21(DE3) cells for 3 h at 37 degrees C produced 30 mg of H6TV-Fhit from 1 L of cell culture ( approximately 4 g of cells). The H6TV-Fhit protein was purified to homogeneity in a single step, with a yield of 80%, using nickel-nitrilotriacetate resin and imidazole buffer as eluting agent. Incubation of H6TV-Fhit with rTEV protease at 4 degrees C for 24 h resulted in complete cleavage of the H6TV peptide. There were no unspecific cleavage products. The purified Fhit protein could be stored for 3 weeks at 4 degrees C without loss of activity. The pure protein was stable at -20 degrees C for at least 18 months when stored in buffer containing 25% glycerol. Purified Fhit was highly active, with a K(m) value for Ap(3)A of 0.9 microM and a k(cat)(monomer) value of 7.2 +/- 1.6 s(-1) (n = 5). The catalytic properties of unconjugated Fhit protein and the H6TV-Fhit fusion protein were essentially identical. This indicates that the 24-amino-acid peptide containing the six histidines fused to the N-terminus of Fhit does not interfere in forming the active homodimers or in the binding of Ap(3)A.     

Autoprocessing of the HIV-1 protease using purified wild-type and mutated fusion proteins expressed at high levels in Escherichia coli

Various constructs of the human immunodeficiency virus, type 1 (HIV-1) protease containing flanking Pol region sequences were expressed as fusion proteins with the maltose-binding protein of the malE gene of Escherichia coli. The full-length fusion proteins did not exhibit self-processing in E. coli, thereby allowing rapid purification by affinity chromatography on cross-linked amylose columns. Denaturation of the fusion protein in 5 M urea, followed by renaturation, resulted in efficient site-specific autoprocessing to release the 11-kDa protease. Rapid purification involving two column steps gave an HIV-1 protease preparations of greater than 95% purity (specific activity approximately 8500 pmol.min-1.micrograms protease-1) with an overall yield of about 1 mg/l culture. Incubation of an inactive mutant protease fusion protein with the purified wild-type protease resulted in specific trans cleavage and release of the mutant protease. Analysis of products of the HIV-1 fusion proteins containing mutations at either the N- or the C-terminal protease cleavage sites indicated that blocking one of the cleavage sites influences the cleavage at the non-mutated site. Such mutated full-length and truncated protease fusion proteins possess very low levels of proteolytic activity (approximately 5 pmol.min-1.micrograms protein-1).     

Expression of catalytic subunit of bovine enterokinase in the filamentous fungus Aspergillus niger

The cDNA encoding for catalytic subunit of bovine enterokinase (EK(L)), to which the sequence for Kex2 protease cleavage site was inserted, was expressed in the protease deficient filamentous fungus Aspergillus niger AB1.13. Fungal transformants were obtained in which expression of the glucoamylase fusion gene resulted in secretion of the protein into growth medium. Fusion polypeptide was processed to mature EK(L) by endogenous Kex-2 like protease cleavage during secretory pathway. The highest quantity of EK(L), up to 5 mg l(-1), was obtained in soya milk medium. The secreted EK(L) was easily purified from other proteins found in A. niger culture supernatant, using ion exchange and affinity chromatography. The yield of the purified and highly active EK(L) was 1.9 mg l(-1) of culture.     

Substrate requirements of human rhinovirus 3C protease for peptide cleavage in vitro

A series of synthetic peptides representing authentic proteolytic cleavage sites of human rhinovirus type 14 were assayed as substrates for purified 3C protease. Competition cleavage assays were employed to determine the relative specificity constants (Kcat/Km) for substrates with sequences related to the viral 2C-3A cleavage site. Variable length peptides representing the 2C-3A cleavage site were cleaved with comparable efficiency. These studies defined a minimum substrate of 6 amino acids (TLFQ/GP), although retention of the residue at position P5 (ETLFQ/GP) resulted in a better substrate by an order of magnitude. Amino acid substitutions at position P5, P4, P1', or P2' indicated that the identity of the residue at position P5 was not critical, whereas substitutions at position P4, P1' or P2' resulted in substrates with Kcat/Km values varying over 2 orders of magnitude. In contrast to the 2C-3A cleavage site, small peptide derivatives representative of the 3A-3B cleavage site were relatively poor substrates, which suggested that residues flanking the minimum core sequence may influence susceptibility to cleavage. The 3C protease of rhinovirus type 14 was also capable of cleaving peptides representing comparable cleavage sites predicted for coxsackie B virus and poliovirus.     

In vitro determination of dengue virus type 2 NS2B-NS3 protease activity with fluorescent peptide substrates

The NS2B-NS3(pro) polyprotein segment from the dengue virus serotype 2 strain 16681 was purified from overexpressing E. coli by metal chelate affinity chromatography and gel filtration. Enzymatic activity of the refolded NS2B-NS3(pro) protease complex was determined in vitro with dansyl-labeled peptide substrates, based upon native dengue virus type 2 cleavage sites. The 12mer substrate peptides and the cleavage products could be separated by reversed-phase HPLC, and were identified by UV and fluorescence detection. All of the peptide substrates (representing the DEN polyprotein junction sequences at the NS2A/NS2B, NS2B/NS3, NS3/NS4A and NS4B/NS5 sites) were cleaved by the recombinant protease NS2B-NS3(pro). No cleavage was observed with an enzymatically inactive S135A mutant of the NS3 protein, or with a modified substrate peptide of the NS3/NS4A polyprotein site that contained a K2093A substitution. Enzymatic activity was dependent on the salt concentration. A 50% decrease of activity was observed in the presence of 0.1 M sodium chloride. Our results show that the NS3 protease activity of the refolded NS2BNS3(pro) protein can be assayed in vitro with high specificity by using cleavage-junction derived peptide substrates.     

Identification of the sequence on NS4A required for enhanced cleavage of the NS5A/5B site by hepatitis C virus NS3 protease.

In addition to NS3 protease, the NS4A protein is required for efficient cleavage of the nonstructural protein region of the hepatitis C virus polyprotein. To investigate the function and the sequence of NS4A required for the enhancement of NS3 protease activity, we developed an in vitro NS3 protease assay system consisting of three purified viral elements: (i) a recombinant NS3 protease which was expressed in Escherichia coli as a maltose-binding protein-NS3 fusion protein (MBP-NS3), (ii) synthetic NS4A fragments, and (iii) a synthetic peptide substrate which mimics the NS5A/5B junction. We showed that the NS3 protease activity of MBP-NS3 was enhanced in a dose-dependent manner by 4A18-40, which is a peptide composed of amino acid residues 18 to 40 of NS4A. The optimal activity was observed at a 10-fold molar excess of 4A18-40 over MBP-NS3. The coefficient for proteolytic efficiency, kcat/Km, of NS3 protease was increased by about 40 times by the addition of a 10-fold molar excess of 4A18-40. Using a series of truncations of 4A18-40, we estimated that amino acid residues 22 to 31 in NS4A (SVVIVGRIIL) constituted the core sequence for the effector activity. Single-substitution experiments with 4A21-34, a peptide composed of amino acid residues 21 to 34 of NS4A, suggested the importance of several residues (Val-23, Ile-25, Gly-27, Arg-28, Ile-29, and Leu-31) for its activity. In addition, we found that some single-amino-acid substitutions in 4A21-34 were able to inhibit the enhancement of NS3 protease activity by 4A18-40. This approach has potential as a novel strategy for inhibiting the NS3 protease activity important for hepatitis C virus proliferation.     

Hirudin as a novel fusion tag for efficient production of lunasin in Escherichia coli

Fusion expression provides an effective means for the biosynthesis of longer peptides in Escherichia coli. However, the commonly used fusion tags are primarily suitable for laboratory scale applications due to the high cost of commercial affinity resins. Herein, a novel approach exploiting hirudin as a multipurpose fusion tag in combination with tobacco etch virus (TEV) protease cleavage has been developed for the efficient and cost-effective production of a 43-amino acid model peptide lunasin in E. coli at preparative scale. A fusion gene which allows for lunasin to be N-terminally fused to the C-terminus of hirudin through a flexible linker comprising a TEV protease cleavage site was designed and cloned in a secretion vector pTASH. By cultivation in a 7-L bioreactor, the fusion protein was excreted into the culture medium at a high yield of ~380?mg/L, which was conveniently recovered and purified by inexpensive HP20 hydrophobic chromatography at a recovery rate of ~80%. After polishing and cleavage with TEV protease, the finally purified lunasin was obtained with ≥95% purity and yield of ~86?mg/L culture medium. Conclusively, this hirudin tagging strategy is powerful in the production of lunasin and could be applicable for the production of other peptides at preparative scale.     

Cleavage specificity of purified recombinant hepatitis A virus 3C proteinase on natural substrates.

Hepatitis A virus (HAV) 3C proteinase expressed in Escherichia coli was purified to homogeneity, and its cleavage specificity towards various parts of the viral polyprotein was analyzed. Intermolecular cleavage of the P2-P3 domain of the HAV polyprotein gave rise to proteins 2A, 2B, 2C, 3ABC, and 3D, suggesting that in addition to the primary cleavage site, all secondary sites within P2 as well as the 3C/3D junction are cleaved by 3C. 3C-mediated processing of the P1-P2 precursor liberated 2A and 2BC, in addition to the structural proteins VP0, VP3, and VP1-2A and the respective intermediate products. A clear dependence on proteinase concentration was found for most cleavage sites, possibly reflecting the cleavage site preference of 3C. The most efficient cleavage occurred at the 2A/2B and 2C/3A junctions. The electrophoretic mobility of processing product 2B, as well as cleavage of the synthetic peptide KGLFSQ*AKISLFYT, suggests that the 2A/2B junction is located at amino acid position 836/837 of the HAV polyprotein. Furthermore, using suitable substrates we obtained evidence that sites VP3/VP1 and VP1/2A are alternatively processed by 3C, leading to either VP1-2A or to P1 and 2A. The results with regard to intermolecular cleavage by purified 3C were confirmed by the product pattern derived from cell-free expression and intramolecular processing of the entire polyprotein. We therefore propose that polyprotein processing of HAV relies on 3C as the single proteinase, possibly assisted by as-yet-undetermined viral or host cell factors and presumably controlled in a concentration-dependent fashion.     

Nicking of single chain Clostridium botulinum type A neurotoxin by an endogenous protease

Botulinum neurotoxin (NT) serotype A isolated from cells from young cultures (approximately 8 h) of Clostridium botulinum type A is a approximately 150 kDa single chain protein. Supernatant from older cultures (96 h) yields approximately 150 kDa dichain NT composed approximately 50 and approximately 100 kDa subunits, that remain associated by disulfide and noncovalent bonds. This had led to the assumption that an endogenous protease cleaves a peptide bond at 1/3rd the distance from the N- or C-terminals of the single chain protein. An endogenous protease that causes such a cleavage (nicking) has now been purified greater than 1,000-fold from C. botulinum type A (Hall strain) culture; this culture also produces the single chain NT and eventually yields the dichain NT. The purified protease nicked the pure preparation of single chain type A NT, in vitro at pH 5.6, into a dichain form that was indistinguishable from the dichain NT normally isolated from 96 h cultures. The protease appears specific for nicking serotype A NT because it did not nick single chain serotype B and E NT nor did it enhance toxicity of serotype A, B and E NT.     

Purification and characterization of poliovirus polypeptide 3CD, a proteinase and a precursor for RNA polymerase.

A cDNA clone encoding the 3CD proteinase (3CDpro) of poliovirus type 2 (Sabin), the precursor to proteinase 3Cpro and RNA polymerase 3Dpol, was expressed in bacteria by using a T7 expression system. Site-specific mutagenesis of the 3C/3D cleavage site was performed to generate active proteolytic precursors impaired in their ability to process themselves to 3Cpro and 3Dpol. Of these mutations, the exchange of the Thr residue at the P4 position of the 3C/3D cleavage site for a Lys residue (3CDpro T181K) resulted in a mutant polypeptide exhibiting the smallest amount of autoprocessing. This mutant was purified to 86% homogeneity and used for subsequent proteolytic studies. Purified 3CDproM (M designates the cleavage site mutant 3CDpro T181K) was capable of cleaving the P1 capsid precursor, a peptide representing the 2BC cleavage site, and the 2BC precursor polypeptide. Purified 3CDproM demonstrated the same detergent sensitivity in processing experiments with the capsid precursor as was observed by using P1 and crude extracts of poliovirus-infected HeLa cell lysates. Purified 3CDproM did not have any detectable RNA polymerase activity, whereas 3Dpol, separated from 3CDproM by gel filtration in the last step of purification, did. We conclude that 3CDproM can process both structural and nonstructural precursors of the poliovirus polyprotein and that it is active against a synthetic peptide substrate. Moreover, cleavage of 3CD to 3Dpol is needed to activate the 3D RNA polymerase.     

Properties of a cold-active protease from psychrotrophic Flavobacterium balustinum P104

Protease activity was detected in the culture medium of Flavobacterium balustinum P104 grown at 10 °C, which was isolated from salmon (Oncorhynchus keta) intestine. The enzyme, designated as CP-70 protease, was purified to homogeneity from the culture broth by ion exchange and gel filtration chromatographyies. The molecular mass of the protease was 70 kDa, and its isoelectric point was close to 3.5. Maximal activity toward azocasein was observed at 40 °C and from pH 7.0 to 9.0. The activity was strongly inhibited by phenylmethylsulfonyl fluoride, suggesting that the enzyme is a serine protease. The n-terminal amino acid sequence was Asp-Thr-Arg-Gln-Leu-Leu-Asn-Ala-Asn-Ser-Asp-Leu-Leu-Asn-Thr-Thr-Gly-Asn-Val-Thr-Gly-Leu-Thr-Gly-Ala-Phe-Asn-Gly-Glu-Asn. A search through the database for sequence homology yielded no significant match. The initial cleavage sites for oxidized insulin B-chain were found to be the Glu13-Ala14 and Phe24-Phe25 bonds. The result of the cleavage pattern of oxidized insulin B-chain suggests that CP-70 protease has a broader specificity than the other cold-active proteases against the peptide substrate. Received: 17 April 1998 / Received last revision: 17 June 1998 / Accepted: 10 July 1998     

Mechanism of the maturation process of SARS-CoV 3CL protease

Severe acute respiratory syndrome (SARS) is an emerging infectious disease caused by a novel human coronavirus. Viral maturation requires a main protease (3CL(pro)) to cleave the virus-encoded polyproteins. We report here that the 3CL(pro) containing additional N- and/or C-terminal segments of the polyprotein sequences undergoes autoprocessing and yields the mature protease in vitro. The dimeric three-dimensional structure of the C145A mutant protease shows that the active site of one protomer binds with the C-terminal six amino acids of the protomer from another asymmetric unit, mimicking the product-bound form and suggesting a possible mechanism for maturation. The P1 pocket of the active site binds the Gln side chain specifically, and the P2 and P4 sites are clustered together to accommodate large hydrophobic side chains. The tagged C145A mutant protein served as a substrate for the wild-type protease, and the N terminus was first digested (55-fold faster) at the Gln(-1)-Ser1 site followed by the C-terminal cleavage at the Gln306-Gly307 site. Analytical ultracentrifuge of the quaternary structures of the tagged and mature proteases reveals the remarkably tighter dimer formation for the mature enzyme (K(d) = 0.35 nm) than for the mutant (C145A) containing 10 extra N-terminal (K(d) = 17.2 nM) or C-terminal amino acids (K(d) = 5.6 nM). The data indicate that immature 3CL(pro) can form dimer enabling it to undergo autoprocessing to yield the mature enzyme, which further serves as a seed for facilitated maturation. Taken together, this study provides insights into the maturation process of the SARS 3CL(pro) from the polyprotein and design of new structure-based inhibitors.     

Human rhinovirus 3C protease: cloning and expression of an active form in Escherichia coli

A cDNA encoding the viral protease from the 3C region of human rhinovirus type 14 was expressed in Escherichia coli through the use of a periplasmic secretion vector. The recombinant protease contained an eight amino acid N-terminal extension that enabled its detection by a specific antibody. It was expressed at a level of approximately 1 mg/L of E. coli culture. Biological activity of the protease was assessed in vitro by using a chemically synthesized peptide consisting of a consensus picornavirus protease cleavage site, Arg-Ala-Glu-Leu-Gln-Gly-Pro-Tyr-Asp-Glu. The peptide was cleaved by the recombinant protease at the Gln-Gly bond, generating the product peptides Arg-Ala-Glu-Leu-Gln and Gly-Pro-Tyr-Asp-Glu, which could be separated from the substrate peptide by reversed-phase HPLC. An in vitro assay for the rhinovirus 3C protease was developed by observing the rate of disappearance of the substrate peak from chromatograms of the supernatants of digestion mixtures.     

Degradation of a signal peptide by protease IV and oligopeptidase A.

The degradation of the prolipoprotein signal peptide in vitro by membranes, cytoplasmic fraction, and two purified major signal peptide peptidases from Escherichia coli was followed by reverse-phase liquid chromatography (RPLC). The cytoplasmic fraction hydrolyzed the signal peptide completely into amino acids. In contrast, many peptide fragments accumulated as final products during the cleavage by a membrane fraction. Most of the peptides were similar to the peptides formed during the cleavage of the signal peptide by the purified membrane-bound signal peptide peptidase, protease IV. Peptide fragments generated during the cleavage of the signal peptide by protease IV and a cytoplasmic enzyme, oligopeptidase A, were identified from their amino acid compositions, their retention times during RPLC, and knowledge of the amino acid sequence of the signal peptide. Both enzymes were endopeptidases, as neither dipeptides nor free amino acids were formed during the cleavage reactions. Protease IV cleaved the signal peptide predominantly in the hydrophobic segment (residues 7 to 14). Protease IV required substrates with hydrophobic amino acids at the primary and the adjacent substrate-binding sites, with a minimum of three amino acids on either side of the scissile bond. Oligopeptidase A cleaved peptides (minimally five residues) that had either alanine or glycine at the P'1 (primary binding site) or at the P1 (preceding P'1) site of the substrate. These results support the hypothesis that protease IV is the major signal peptide peptidase in membranes that initiates the degradation of the signal peptide by making endoproteolytic cuts; oligopeptidase A and other cytoplasmic enzymes further degrade the partially degraded portions of the signal peptide that may be diffused or transported back into the cytoplasm from the membranes.     

Expression and purification of a recombinant LL-37 from Escherichia coli

Human cathelicidin-derived LL-37 is a 37-residue cationic, amphipathic alpha-helical peptide. It is an active component of mammalian innate immunity. LL-37 has several biological functions including a broad spectrum of antimicrobial activities and LPS-neutralizing activity. In order to determine the high-resolution three-dimensional structure of LL-37 using NMR spectroscopy, it is important to obtain the peptide with isotopic labels such as (15)N, (13)C and/or (2)H. Since it is less expensive to obtain such a peptide biologically, in this study, we report for the first time a method to express in E. coli and purify LL-37 using Glutathione S-transferase (GST) fusion system. LL-37 gene was inserted into vector pGEX-4T3 and expressed as a GST-LL-37 fusion protein in BL21(DE3) strain. The recombinant GST-LL-37 protein was purified with a yield of 8 mg/l by affinity chromatography and analyzed its biochemical and spectroscopic properties. Factor Xa was used to cleave a 4.5-kDa LL-37 from the GST-LL-37 fusion protein and the peptide was purified using a reverse-phase HPLC on a Vydac C(18) column with a final yield of 0.3 mg/l. The protein purified using reverse-phase HPLC was confirmed to be LL-37 by the analyses of Western blot and MALDI-TOF-Mass spectrometry. E. coli cells harboring the expression vector pGEX-4T3-LL-37 were grown in the presence of the (15)N-labeled M9 minimal medium and culture conditions were optimized to obtain uniform (15)N enrichment in the constitutively expressed LL-37 peptide. These results suggest that our production method will be useful in obtaining a large quantity of recombinant LL-37 peptide for NMR studies.     

Molecular characterization of the protease from Clostridium botulinum serotype C responsible for nicking in botulinum neurotoxin complex

A protease was purified from the culture medium of Clostridium botulinum serotype C strain Stockholm (C-St). The purified protease belonged to the cysteine protease family based on assays for enzyme inhibitors, activators and kinetic parameters. The protease formed a binary complex consisting of 41- and 17-kDa proteins held together non-covalently. The DNA sequence encoding the protease gene was shown to be a single open reading frame of 1593 nucleotides, predicting 530 amino acid residues including a signal peptide. The N-terminal region of the native enzyme underwent further proteolytic modification after processing by a signal peptidase. The protease introduced intermolecular cleavage into an intact single chain botulinum neurotoxin (BoNT) at a specific site. Homology modeling and docking simulation of C-St BoNT and C-St protease demonstrated that the specific nicking-site of the BoNT appears to fit into the deep pocket in the active site of the protease.     

Cleavage of small peptides in vitro by human rhinovirus 14 3C protease expressed in Escherichia coli.

The 3C region of human rhinovirus 14 was expressed in Escherichia coli. The microbially synthesized protease was functional, since the expressed precursor underwent autoproteolytic processing to generate mature molecules of the expected molecular weight and antigenicity. Mutation of the putative active-site Cys-146 residue to an alanine resulted in the synthesis of unprocessed precursor molecules. Large quantities of the 20-kilodalton protease were purified by a simple purification protocol, and the resulting molecule was shown to be biologically active in vitro against synthetic peptides corresponding to the 2C-3A cleavage site. This site was cleaved with high efficiency and fidelity and was used to generate kinetic data on the 3C protease. The protease exhibited sensitivity to Zn2+, was capable of cleaving five of seven rhinovirus cleavage site peptides tested with variable efficiency, and could distinguish authentic substrate peptides from control peptides containing the dipeptide cleavage sequence pair Gln-Gly.     

A novel fusion partner for enhanced secretion of recombinant proteins in <Emphasis Type="Italic">Saccharomyces cerevisiae</Emphasis>

Expressing proteins with fusion partners improves yield and simplifies the purification process. We developed a novel fusion partner to improve the secretion of heterologous proteins that are otherwise poorly excreted in yeast. The VOA1 (YGR106C) gene of Saccharomyces cerevisiae encodes a subunit of vacuolar ATPase. We found that C-terminally truncated Voa1p was highly secreted into the culture medium, even when fused with rarely secreted heterologous proteins such as human interleukin-2 (hIL-2). Deletion mapping of C-terminally truncated Voa1p, identified a hydrophilic 28-amino acid peptide (HL peptide) that was responsible for the enhanced secretion of target protein. A purification tag and a protease cleavage site were added to use HL peptide as a multi-purpose fusion partner. The utility of this system was tested via the expression and purification of various heterologous proteins. In many cases, the yield of target proteins fused with the peptide was significantly increased, and fusion proteins could be directly purified with affinity chromatography. The fusion partner was removed by in vitro processing, and intact proteins were purified by re-application of samples to affinity chromatography.     

Microbial reclamation of squid pen for the production of a novel extracellular serine protease by Lactobacillus paracasei subsp paracasei TKU012

A protease-producing bacterium, strain TKU012, was isolated from infant vomited milk and identified as Lactobacillus paracasei subsp paracasei. Strain TKU012 produced protease when it was grown in a medium containing squid pen powder of marine waste. An extracellular protease was purified from culture supernatant by DEAE-Sepharose and Sephacryl S-100 chromatography. A protease, purified 77-fold to homogeneity in an overall yield of 11%, has a molecular weight of about 49 kDa estimated by SDS-PAGE. The protease was maximally active at pH 10 and 60 degrees C and showed substrate specificity to casein and gelatin. The protease retains 21% and 91% activity in the presence of Tween 20 (2% w/v) and SDS (2mM), respectively. The enzyme activity was reduced in the presence of PMSF and showed 23% sequence coverage rate with metalloprotease of Serratia marcescens. This is the first report of extracellular proteases purified from lactobacilli.